wandb: Currently logged in as: laurence_r. Use `wandb login --relogin` to force relogin
wandb: Tracking run with wandb version 0.16.3
wandb: Run data is saved locally in /home/ubuntu/torchtune-colorful-llama/colorful/wandb/run-20240218_171649-qsv24pfw
wandb: Run `wandb offline` to turn off syncing.
wandb: Syncing run lucky-envelope-8
wandb: ⭐️ View project at https://wandb.ai/laurence_r/colorful-llama
wandb: 🚀 View run at https://wandb.ai/laurence_r/colorful-llama/runs/qsv24pfw
Setting manual seed to local seed 42. Local seed is seed + rank = 42 + 0
Model is initialized. FSDP and Activation Checkpointing are enabled.
Compiling model using torch.compile. The first batch may take a few minutes while compilation occurs.
Tokenizer is initialized from file.
Optimizer is initialized.
Loss is initialized.
Dataset and Sampler are initialized.

  0%|          | 0/6470 [00:00<?, ?it/s]
1|1|Loss: 1.5560694932937622:   0%|          | 0/6470 [00:45<?, ?it/s]
1|1|Loss: 1.5560694932937622:   0%|          | 1/6470 [01:18<140:37:49, 78.26s/it]
1|2|Loss: 1.345805287361145:   0%|          | 1/6470 [02:54<140:37:49, 78.26s/it] 
1|2|Loss: 1.345805287361145:   0%|          | 2/6470 [02:55<160:50:09, 89.52s/it]
1|3|Loss: 2.25685715675354:   0%|          | 2/6470 [02:55<160:50:09, 89.52s/it] 
1|3|Loss: 2.25685715675354:   0%|          | 3/6470 [02:56<87:56:44, 48.96s/it] 
1|4|Loss: 9.205286026000977:   0%|          | 3/6470 [02:56<87:56:44, 48.96s/it]
1|4|Loss: 9.205286026000977:   0%|          | 4/6470 [02:57<53:46:36, 29.94s/it]
1|5|Loss: 9.649592399597168:   0%|          | 4/6470 [02:57<53:46:36, 29.94s/it]
1|5|Loss: 9.649592399597168:   0%|          | 5/6470 [02:57<34:44:15, 19.34s/it]
1|6|Loss: 7.697812080383301:   0%|          | 5/6470 [02:58<34:44:15, 19.34s/it]
1|6|Loss: 7.697812080383301:   0%|          | 6/6470 [02:58<23:16:54, 12.97s/it]
1|7|Loss: 8.006651878356934:   0%|          | 6/6470 [02:58<23:16:54, 12.97s/it]
1|7|Loss: 8.006651878356934:   0%|          | 7/6470 [02:58<16:05:11,  8.96s/it]
1|8|Loss: 7.790714740753174:   0%|          | 7/6470 [02:59<16:05:11,  8.96s/it]
1|8|Loss: 7.790714740753174:   0%|          | 8/6470 [02:59<11:26:01,  6.37s/it]
1|9|Loss: 8.520974159240723:   0%|          | 8/6470 [03:00<11:26:01,  6.37s/it]
1|9|Loss: 8.520974159240723:   0%|          | 9/6470 [03:00<8:17:14,  4.62s/it] 
1|10|Loss: 8.084063529968262:   0%|          | 9/6470 [03:00<8:17:14,  4.62s/it]
1|10|Loss: 8.084063529968262:   0%|          | 10/6470 [03:01<6:07:47,  3.42s/it]
1|11|Loss: 8.080673217773438:   0%|          | 10/6470 [03:01<6:07:47,  3.42s/it]
1|11|Loss: 8.080673217773438:   0%|          | 11/6470 [03:02<4:41:03,  2.61s/it]
1|12|Loss: 7.927273750305176:   0%|          | 11/6470 [03:02<4:41:03,  2.61s/it]
1|12|Loss: 7.927273750305176:   0%|          | 12/6470 [03:02<3:43:59,  2.08s/it]
1|13|Loss: 7.6634650230407715:   0%|          | 12/6470 [03:03<3:43:59,  2.08s/it]
1|13|Loss: 7.6634650230407715:   0%|          | 13/6470 [03:03<3:01:51,  1.69s/it]
1|14|Loss: 7.621446132659912:   0%|          | 13/6470 [03:04<3:01:51,  1.69s/it] 
1|14|Loss: 7.621446132659912:   0%|          | 14/6470 [03:04<2:25:21,  1.35s/it]
1|15|Loss: 7.575307369232178:   0%|          | 14/6470 [03:04<2:25:21,  1.35s/it]
1|15|Loss: 7.575307369232178:   0%|          | 15/6470 [03:04<2:02:22,  1.14s/it]
1|16|Loss: 7.750604629516602:   0%|          | 15/6470 [03:05<2:02:22,  1.14s/it]
1|16|Loss: 7.750604629516602:   0%|          | 16/6470 [03:05<1:49:40,  1.02s/it]
1|17|Loss: 7.60357666015625:   0%|          | 16/6470 [03:06<1:49:40,  1.02s/it] 
1|17|Loss: 7.60357666015625:   0%|          | 17/6470 [03:06<1:47:11,  1.00it/s]
1|18|Loss: 7.436343669891357:   0%|          | 17/6470 [03:07<1:47:11,  1.00it/s]
1|18|Loss: 7.436343669891357:   0%|          | 18/6470 [03:07<1:41:03,  1.06it/s]
1|19|Loss: 7.236898899078369:   0%|          | 18/6470 [03:07<1:41:03,  1.06it/s]
1|19|Loss: 7.236898899078369:   0%|          | 19/6470 [03:08<1:33:07,  1.15it/s]
1|20|Loss: 7.514071464538574:   0%|          | 19/6470 [03:08<1:33:07,  1.15it/s]
1|20|Loss: 7.514071464538574:   0%|          | 20/6470 [03:08<1:32:06,  1.17it/s]
1|21|Loss: 7.15755033493042:   0%|          | 20/6470 [03:09<1:32:06,  1.17it/s] 
1|21|Loss: 7.15755033493042:   0%|          | 21/6470 [03:09<1:29:26,  1.20it/s]
1|22|Loss: 7.447564601898193:   0%|          | 21/6470 [03:10<1:29:26,  1.20it/s]
1|22|Loss: 7.447564601898193:   0%|          | 22/6470 [03:10<1:24:31,  1.27it/s]
1|23|Loss: 7.391354560852051:   0%|          | 22/6470 [03:10<1:24:31,  1.27it/s]
1|23|Loss: 7.391354560852051:   0%|          | 23/6470 [03:11<1:22:10,  1.31it/s]
1|24|Loss: 7.190248489379883:   0%|          | 23/6470 [03:11<1:22:10,  1.31it/s]
1|24|Loss: 7.190248489379883:   0%|          | 24/6470 [03:11<1:20:40,  1.33it/s]
1|25|Loss: 7.189001083374023:   0%|          | 24/6470 [03:12<1:20:40,  1.33it/s]
1|25|Loss: 7.189001083374023:   0%|          | 25/6470 [03:12<1:23:01,  1.29it/s]
1|26|Loss: 7.167294502258301:   0%|          | 25/6470 [03:13<1:23:01,  1.29it/s]
1|26|Loss: 7.167294502258301:   0%|          | 26/6470 [03:13<1:22:36,  1.30it/s]
1|27|Loss: 6.962973594665527:   0%|          | 26/6470 [03:13<1:22:36,  1.30it/s]
1|27|Loss: 6.962973594665527:   0%|          | 27/6470 [03:14<1:22:00,  1.31it/s]
1|28|Loss: 7.197654724121094:   0%|          | 27/6470 [03:14<1:22:00,  1.31it/s]
1|28|Loss: 7.197654724121094:   0%|          | 28/6470 [03:14<1:20:47,  1.33it/s]
1|29|Loss: 7.1847639083862305:   0%|          | 28/6470 [03:15<1:20:47,  1.33it/s]
1|29|Loss: 7.1847639083862305:   0%|          | 29/6470 [03:15<1:16:06,  1.41it/s]
1|30|Loss: 7.174182891845703:   0%|          | 29/6470 [03:15<1:16:06,  1.41it/s] 
1|30|Loss: 7.174182891845703:   0%|          | 30/6470 [03:16<1:10:27,  1.52it/s]
1|31|Loss: 7.176393985748291:   0%|          | 30/6470 [03:16<1:10:27,  1.52it/s]
1|31|Loss: 7.176393985748291:   0%|          | 31/6470 [03:16<1:11:50,  1.49it/s]
1|32|Loss: 7.204705238342285:   0%|          | 31/6470 [03:17<1:11:50,  1.49it/s]
1|32|Loss: 7.204705238342285:   0%|          | 32/6470 [03:17<1:17:51,  1.38it/s]
1|33|Loss: 6.825334072113037:   0%|          | 32/6470 [03:18<1:17:51,  1.38it/s]
1|33|Loss: 6.825334072113037:   1%|          | 33/6470 [03:18<1:14:47,  1.43it/s]
1|34|Loss: 7.161884784698486:   1%|          | 33/6470 [03:18<1:14:47,  1.43it/s]
1|34|Loss: 7.161884784698486:   1%|          | 34/6470 [03:19<1:18:29,  1.37it/s]
1|35|Loss: 6.933444976806641:   1%|          | 34/6470 [03:19<1:18:29,  1.37it/s]
1|35|Loss: 6.933444976806641:   1%|          | 35/6470 [03:19<1:23:55,  1.28it/s]
1|36|Loss: 7.094783306121826:   1%|          | 35/6470 [03:20<1:23:55,  1.28it/s]
1|36|Loss: 7.094783306121826:   1%|          | 36/6470 [03:20<1:23:47,  1.28it/s]
1|37|Loss: 7.271488189697266:   1%|          | 36/6470 [03:21<1:23:47,  1.28it/s]
1|37|Loss: 7.271488189697266:   1%|          | 37/6470 [03:21<1:26:19,  1.24it/s]
1|38|Loss: 7.007736682891846:   1%|          | 37/6470 [03:22<1:26:19,  1.24it/s]
1|38|Loss: 7.007736682891846:   1%|          | 38/6470 [03:22<1:27:41,  1.22it/s]
1|39|Loss: 6.6992878913879395:   1%|          | 38/6470 [03:22<1:27:41,  1.22it/s]
1|39|Loss: 6.6992878913879395:   1%|          | 39/6470 [03:23<1:27:37,  1.22it/s]
1|40|Loss: 6.983700752258301:   1%|          | 39/6470 [03:23<1:27:37,  1.22it/s] 
1|40|Loss: 6.983700752258301:   1%|          | 40/6470 [03:24<1:29:06,  1.20it/s]
1|41|Loss: 6.9497904777526855:   1%|          | 40/6470 [03:24<1:29:06,  1.20it/s]
1|41|Loss: 6.9497904777526855:   1%|          | 41/6470 [03:24<1:25:20,  1.26it/s]
1|42|Loss: 6.86012077331543:   1%|          | 41/6470 [03:25<1:25:20,  1.26it/s]  
1|42|Loss: 6.86012077331543:   1%|          | 42/6470 [03:25<1:21:05,  1.32it/s]
1|43|Loss: 7.031021595001221:   1%|          | 42/6470 [03:25<1:21:05,  1.32it/s]
1|43|Loss: 7.031021595001221:   1%|          | 43/6470 [03:26<1:19:30,  1.35it/s]
1|44|Loss: 7.286438465118408:   1%|          | 43/6470 [03:26<1:19:30,  1.35it/s]
1|44|Loss: 7.286438465118408:   1%|          | 44/6470 [03:26<1:13:34,  1.46it/s]
1|45|Loss: 7.080848693847656:   1%|          | 44/6470 [03:27<1:13:34,  1.46it/s]
1|45|Loss: 7.080848693847656:   1%|          | 45/6470 [03:27<1:16:54,  1.39it/s]
1|46|Loss: 6.556846618652344:   1%|          | 45/6470 [03:27<1:16:54,  1.39it/s]
1|46|Loss: 6.556846618652344:   1%|          | 46/6470 [03:28<1:13:57,  1.45it/s]
1|47|Loss: 7.246750354766846:   1%|          | 46/6470 [03:28<1:13:57,  1.45it/s]
1|47|Loss: 7.246750354766846:   1%|          | 47/6470 [03:28<1:17:08,  1.39it/s]
1|48|Loss: 6.821289539337158:   1%|          | 47/6470 [03:29<1:17:08,  1.39it/s]
1|48|Loss: 6.821289539337158:   1%|          | 48/6470 [03:29<1:16:19,  1.40it/s]
1|49|Loss: 6.892756938934326:   1%|          | 48/6470 [03:30<1:16:19,  1.40it/s]
1|49|Loss: 6.892756938934326:   1%|          | 49/6470 [03:30<1:23:06,  1.29it/s]
1|50|Loss: 6.917089939117432:   1%|          | 49/6470 [03:31<1:23:06,  1.29it/s]
1|50|Loss: 6.917089939117432:   1%|          | 50/6470 [03:31<1:28:41,  1.21it/s]
1|51|Loss: 6.809078693389893:   1%|          | 50/6470 [03:32<1:28:41,  1.21it/s]
1|51|Loss: 6.809078693389893:   1%|          | 51/6470 [03:32<1:26:58,  1.23it/s]
1|52|Loss: 7.153294086456299:   1%|          | 51/6470 [03:32<1:26:58,  1.23it/s]
1|52|Loss: 7.153294086456299:   1%|          | 52/6470 [03:33<1:29:32,  1.19it/s]
1|53|Loss: 7.125644207000732:   1%|          | 52/6470 [03:33<1:29:32,  1.19it/s]
1|53|Loss: 7.125644207000732:   1%|          | 53/6470 [03:34<1:28:03,  1.21it/s]
1|54|Loss: 6.5913496017456055:   1%|          | 53/6470 [03:34<1:28:03,  1.21it/s]
1|54|Loss: 6.5913496017456055:   1%|          | 54/6470 [03:34<1:23:06,  1.29it/s]
1|55|Loss: 6.979879856109619:   1%|          | 54/6470 [03:35<1:23:06,  1.29it/s] 
1|55|Loss: 6.979879856109619:   1%|          | 55/6470 [03:35<1:21:37,  1.31it/s]
1|56|Loss: 6.853839874267578:   1%|          | 55/6470 [03:35<1:21:37,  1.31it/s]
1|56|Loss: 6.853839874267578:   1%|          | 56/6470 [03:36<1:24:18,  1.27it/s]
1|57|Loss: 6.6769232749938965:   1%|          | 56/6470 [03:36<1:24:18,  1.27it/s]
1|57|Loss: 6.6769232749938965:   1%|          | 57/6470 [03:36<1:19:11,  1.35it/s]
1|58|Loss: 6.627223014831543:   1%|          | 57/6470 [03:37<1:19:11,  1.35it/s] 
1|58|Loss: 6.627223014831543:   1%|          | 58/6470 [03:37<1:15:53,  1.41it/s]
1|59|Loss: 6.750972270965576:   1%|          | 58/6470 [03:37<1:15:53,  1.41it/s]
1|59|Loss: 6.750972270965576:   1%|          | 59/6470 [03:38<1:16:56,  1.39it/s]
1|60|Loss: 6.712663650512695:   1%|          | 59/6470 [03:38<1:16:56,  1.39it/s]
1|60|Loss: 6.712663650512695:   1%|          | 60/6470 [03:39<1:21:33,  1.31it/s]
1|61|Loss: 6.558324337005615:   1%|          | 60/6470 [03:39<1:21:33,  1.31it/s]
1|61|Loss: 6.558324337005615:   1%|          | 61/6470 [03:39<1:23:08,  1.28it/s]
1|62|Loss: 6.551995754241943:   1%|          | 61/6470 [03:40<1:23:08,  1.28it/s]
1|62|Loss: 6.551995754241943:   1%|          | 62/6470 [03:40<1:20:03,  1.33it/s]
1|63|Loss: 6.6519246101379395:   1%|          | 62/6470 [03:41<1:20:03,  1.33it/s]
1|63|Loss: 6.6519246101379395:   1%|          | 63/6470 [03:41<1:17:50,  1.37it/s]
1|64|Loss: 6.715220928192139:   1%|          | 63/6470 [03:41<1:17:50,  1.37it/s] 
1|64|Loss: 6.715220928192139:   1%|          | 64/6470 [03:42<1:22:33,  1.29it/s]
1|65|Loss: 6.968022346496582:   1%|          | 64/6470 [03:42<1:22:33,  1.29it/s]
1|65|Loss: 6.968022346496582:   1%|          | 65/6470 [03:43<1:25:33,  1.25it/s]
1|66|Loss: 6.706735134124756:   1%|          | 65/6470 [03:43<1:25:33,  1.25it/s]
1|66|Loss: 6.706735134124756:   1%|          | 66/6470 [03:43<1:23:45,  1.27it/s]
1|67|Loss: 6.586292266845703:   1%|          | 66/6470 [03:44<1:23:45,  1.27it/s]
1|67|Loss: 6.586292266845703:   1%|          | 67/6470 [03:44<1:21:33,  1.31it/s]
1|68|Loss: 6.945464611053467:   1%|          | 67/6470 [03:44<1:21:33,  1.31it/s]
1|68|Loss: 6.945464611053467:   1%|          | 68/6470 [03:45<1:24:35,  1.26it/s]
1|69|Loss: 6.6642632484436035:   1%|          | 68/6470 [03:45<1:24:35,  1.26it/s]
1|69|Loss: 6.6642632484436035:   1%|          | 69/6470 [03:46<1:21:45,  1.30it/s]
1|70|Loss: 6.603114128112793:   1%|          | 69/6470 [03:46<1:21:45,  1.30it/s] 
1|70|Loss: 6.603114128112793:   1%|          | 70/6470 [03:46<1:19:56,  1.33it/s]
1|71|Loss: 6.505764961242676:   1%|          | 70/6470 [03:47<1:19:56,  1.33it/s]
1|71|Loss: 6.505764961242676:   1%|          | 71/6470 [03:47<1:22:08,  1.30it/s]
1|72|Loss: 6.616352081298828:   1%|          | 71/6470 [03:48<1:22:08,  1.30it/s]
1|72|Loss: 6.616352081298828:   1%|          | 72/6470 [03:48<1:21:24,  1.31it/s]
1|73|Loss: 6.474825382232666:   1%|          | 72/6470 [03:48<1:21:24,  1.31it/s]
1|73|Loss: 6.474825382232666:   1%|          | 73/6470 [03:49<1:21:03,  1.32it/s]
1|74|Loss: 6.380812644958496:   1%|          | 73/6470 [03:49<1:21:03,  1.32it/s]
1|74|Loss: 6.380812644958496:   1%|          | 74/6470 [03:49<1:18:02,  1.37it/s]
1|75|Loss: 6.808317184448242:   1%|          | 74/6470 [03:50<1:18:02,  1.37it/s]
1|75|Loss: 6.808317184448242:   1%|          | 75/6470 [03:50<1:18:43,  1.35it/s]
1|76|Loss: 6.533715724945068:   1%|          | 75/6470 [03:51<1:18:43,  1.35it/s]
1|76|Loss: 6.533715724945068:   1%|          | 76/6470 [03:51<1:24:29,  1.26it/s]
1|77|Loss: 6.40634298324585:   1%|          | 76/6470 [03:51<1:24:29,  1.26it/s] 
1|77|Loss: 6.40634298324585:   1%|          | 77/6470 [03:52<1:20:33,  1.32it/s]
1|78|Loss: 6.382643222808838:   1%|          | 77/6470 [03:52<1:20:33,  1.32it/s]
1|78|Loss: 6.382643222808838:   1%|          | 78/6470 [03:52<1:18:38,  1.35it/s]
1|79|Loss: 6.386824607849121:   1%|          | 78/6470 [03:53<1:18:38,  1.35it/s]
1|79|Loss: 6.386824607849121:   1%|          | 79/6470 [03:53<1:16:37,  1.39it/s]
1|80|Loss: 6.564176559448242:   1%|          | 79/6470 [03:53<1:16:37,  1.39it/s]
1|80|Loss: 6.564176559448242:   1%|          | 80/6470 [03:54<1:16:36,  1.39it/s]
1|81|Loss: 6.291683673858643:   1%|          | 80/6470 [03:54<1:16:36,  1.39it/s]
1|81|Loss: 6.291683673858643:   1%|▏         | 81/6470 [03:54<1:12:48,  1.46it/s]
1|82|Loss: 6.465012073516846:   1%|▏         | 81/6470 [03:55<1:12:48,  1.46it/s]
1|82|Loss: 6.465012073516846:   1%|▏         | 82/6470 [03:55<1:10:14,  1.52it/s]
1|83|Loss: 6.484710693359375:   1%|▏         | 82/6470 [03:55<1:10:14,  1.52it/s]
1|83|Loss: 6.484710693359375:   1%|▏         | 83/6470 [03:56<1:09:30,  1.53it/s]
1|84|Loss: 6.62507963180542:   1%|▏         | 83/6470 [03:56<1:09:30,  1.53it/s] 
1|84|Loss: 6.62507963180542:   1%|▏         | 84/6470 [03:56<1:12:11,  1.47it/s]
1|85|Loss: 6.719769477844238:   1%|▏         | 84/6470 [03:57<1:12:11,  1.47it/s]
1|85|Loss: 6.719769477844238:   1%|▏         | 85/6470 [03:57<1:12:32,  1.47it/s]
1|86|Loss: 6.70286750793457:   1%|▏         | 85/6470 [03:57<1:12:32,  1.47it/s] 
1|86|Loss: 6.70286750793457:   1%|▏         | 86/6470 [03:58<1:14:02,  1.44it/s]
1|87|Loss: 6.482405185699463:   1%|▏         | 86/6470 [03:58<1:14:02,  1.44it/s]
1|87|Loss: 6.482405185699463:   1%|▏         | 87/6470 [03:58<1:12:23,  1.47it/s]
1|88|Loss: 6.416488170623779:   1%|▏         | 87/6470 [03:59<1:12:23,  1.47it/s]
1|88|Loss: 6.416488170623779:   1%|▏         | 88/6470 [03:59<1:11:24,  1.49it/s]
1|89|Loss: 6.270252704620361:   1%|▏         | 88/6470 [03:59<1:11:24,  1.49it/s]
1|89|Loss: 6.270252704620361:   1%|▏         | 89/6470 [04:00<1:13:38,  1.44it/s]
1|90|Loss: 6.398486614227295:   1%|▏         | 89/6470 [04:00<1:13:38,  1.44it/s]
1|90|Loss: 6.398486614227295:   1%|▏         | 90/6470 [04:00<1:13:38,  1.44it/s]
1|91|Loss: 6.965109825134277:   1%|▏         | 90/6470 [04:01<1:13:38,  1.44it/s]
1|91|Loss: 6.965109825134277:   1%|▏         | 91/6470 [04:01<1:20:41,  1.32it/s]
1|92|Loss: 6.359752655029297:   1%|▏         | 91/6470 [04:02<1:20:41,  1.32it/s]
1|92|Loss: 6.359752655029297:   1%|▏         | 92/6470 [04:02<1:25:30,  1.24it/s]
1|93|Loss: 6.416800498962402:   1%|▏         | 92/6470 [04:03<1:25:30,  1.24it/s]
1|93|Loss: 6.416800498962402:   1%|▏         | 93/6470 [04:03<1:25:30,  1.24it/s]
1|94|Loss: 6.249859809875488:   1%|▏         | 93/6470 [04:03<1:25:30,  1.24it/s]
1|94|Loss: 6.249859809875488:   1%|▏         | 94/6470 [04:04<1:19:04,  1.34it/s]
1|95|Loss: 6.430609226226807:   1%|▏         | 94/6470 [04:04<1:19:04,  1.34it/s]
1|95|Loss: 6.430609226226807:   1%|▏         | 95/6470 [04:04<1:14:14,  1.43it/s]
1|96|Loss: 6.11603307723999:   1%|▏         | 95/6470 [04:05<1:14:14,  1.43it/s] 
1|96|Loss: 6.11603307723999:   1%|▏         | 96/6470 [04:05<1:11:07,  1.49it/s]
1|97|Loss: 6.357972621917725:   1%|▏         | 96/6470 [04:05<1:11:07,  1.49it/s]
1|97|Loss: 6.357972621917725:   1%|▏         | 97/6470 [04:05<1:07:28,  1.57it/s]
1|98|Loss: 5.846924781799316:   1%|▏         | 97/6470 [04:06<1:07:28,  1.57it/s]
1|98|Loss: 5.846924781799316:   2%|▏         | 98/6470 [04:06<1:00:42,  1.75it/s]
1|99|Loss: 6.603648662567139:   2%|▏         | 98/6470 [04:06<1:00:42,  1.75it/s]
1|99|Loss: 6.603648662567139:   2%|▏         | 99/6470 [04:07<1:03:38,  1.67it/s]
1|100|Loss: 6.396233558654785:   2%|▏         | 99/6470 [04:07<1:03:38,  1.67it/s]
1|100|Loss: 6.396233558654785:   2%|▏         | 100/6470 [04:07<1:11:17,  1.49it/s]
1|101|Loss: 6.487743377685547:   2%|▏         | 100/6470 [04:08<1:11:17,  1.49it/s]
1|101|Loss: 6.487743377685547:   2%|▏         | 101/6470 [04:08<1:15:17,  1.41it/s]
1|102|Loss: 6.37262487411499:   2%|▏         | 101/6470 [04:09<1:15:17,  1.41it/s] 
1|102|Loss: 6.37262487411499:   2%|▏         | 102/6470 [04:09<1:21:13,  1.31it/s]
1|103|Loss: 6.136116027832031:   2%|▏         | 102/6470 [04:09<1:21:13,  1.31it/s]
1|103|Loss: 6.136116027832031:   2%|▏         | 103/6470 [04:10<1:18:00,  1.36it/s]
1|104|Loss: 6.3784966468811035:   2%|▏         | 103/6470 [04:10<1:18:00,  1.36it/s]
1|104|Loss: 6.3784966468811035:   2%|▏         | 104/6470 [04:11<1:19:32,  1.33it/s]
1|105|Loss: 6.379662036895752:   2%|▏         | 104/6470 [04:11<1:19:32,  1.33it/s] 
1|105|Loss: 6.379662036895752:   2%|▏         | 105/6470 [04:11<1:17:58,  1.36it/s]
1|106|Loss: 6.134491443634033:   2%|▏         | 105/6470 [04:12<1:17:58,  1.36it/s]
1|106|Loss: 6.134491443634033:   2%|▏         | 106/6470 [04:12<1:18:26,  1.35it/s]
1|107|Loss: 6.385514736175537:   2%|▏         | 106/6470 [04:12<1:18:26,  1.35it/s]
1|107|Loss: 6.385514736175537:   2%|▏         | 107/6470 [04:13<1:18:16,  1.35it/s]
1|108|Loss: 6.289677143096924:   2%|▏         | 107/6470 [04:13<1:18:16,  1.35it/s]
1|108|Loss: 6.289677143096924:   2%|▏         | 108/6470 [04:13<1:14:42,  1.42it/s]
1|109|Loss: 5.922205448150635:   2%|▏         | 108/6470 [04:14<1:14:42,  1.42it/s]
1|109|Loss: 5.922205448150635:   2%|▏         | 109/6470 [04:14<1:09:55,  1.52it/s]
1|110|Loss: 6.511906147003174:   2%|▏         | 109/6470 [04:14<1:09:55,  1.52it/s]
1|110|Loss: 6.511906147003174:   2%|▏         | 110/6470 [04:15<1:15:21,  1.41it/s]
1|111|Loss: 6.269786834716797:   2%|▏         | 110/6470 [04:15<1:15:21,  1.41it/s]
1|111|Loss: 6.269786834716797:   2%|▏         | 111/6470 [04:16<1:18:16,  1.35it/s]
1|112|Loss: 5.838456153869629:   2%|▏         | 111/6470 [04:16<1:18:16,  1.35it/s]
1|112|Loss: 5.838456153869629:   2%|▏         | 112/6470 [04:16<1:16:33,  1.38it/s]
1|113|Loss: 5.985480785369873:   2%|▏         | 112/6470 [04:17<1:16:33,  1.38it/s]
1|113|Loss: 5.985480785369873:   2%|▏         | 113/6470 [04:17<1:10:55,  1.49it/s]
1|114|Loss: 6.094171524047852:   2%|▏         | 113/6470 [04:17<1:10:55,  1.49it/s]
1|114|Loss: 6.094171524047852:   2%|▏         | 114/6470 [04:17<1:10:03,  1.51it/s]
1|115|Loss: 6.0838704109191895:   2%|▏         | 114/6470 [04:18<1:10:03,  1.51it/s]
1|115|Loss: 6.0838704109191895:   2%|▏         | 115/6470 [04:18<1:09:58,  1.51it/s]
1|116|Loss: 6.418234348297119:   2%|▏         | 115/6470 [04:18<1:09:58,  1.51it/s] 
1|116|Loss: 6.418234348297119:   2%|▏         | 116/6470 [04:19<1:09:47,  1.52it/s]
1|117|Loss: 6.306113243103027:   2%|▏         | 116/6470 [04:19<1:09:47,  1.52it/s]
1|117|Loss: 6.306113243103027:   2%|▏         | 117/6470 [04:20<1:15:50,  1.40it/s]
1|118|Loss: 6.337181091308594:   2%|▏         | 117/6470 [04:20<1:15:50,  1.40it/s]
1|118|Loss: 6.337181091308594:   2%|▏         | 118/6470 [04:20<1:17:26,  1.37it/s]
1|119|Loss: 6.270710468292236:   2%|▏         | 118/6470 [04:21<1:17:26,  1.37it/s]
1|119|Loss: 6.270710468292236:   2%|▏         | 119/6470 [04:21<1:17:35,  1.36it/s]
1|120|Loss: 6.40598201751709:   2%|▏         | 119/6470 [04:21<1:17:35,  1.36it/s] 
1|120|Loss: 6.40598201751709:   2%|▏         | 120/6470 [04:22<1:18:58,  1.34it/s]
1|121|Loss: 6.097846508026123:   2%|▏         | 120/6470 [04:22<1:18:58,  1.34it/s]
1|121|Loss: 6.097846508026123:   2%|▏         | 121/6470 [04:23<1:17:42,  1.36it/s]
1|122|Loss: 6.104159832000732:   2%|▏         | 121/6470 [04:23<1:17:42,  1.36it/s]
1|122|Loss: 6.104159832000732:   2%|▏         | 122/6470 [04:23<1:19:54,  1.32it/s]
1|123|Loss: 5.815454959869385:   2%|▏         | 122/6470 [04:24<1:19:54,  1.32it/s]
1|123|Loss: 5.815454959869385:   2%|▏         | 123/6470 [04:24<1:16:04,  1.39it/s]
1|124|Loss: 5.568240165710449:   2%|▏         | 123/6470 [04:24<1:16:04,  1.39it/s]
1|124|Loss: 5.568240165710449:   2%|▏         | 124/6470 [04:25<1:11:18,  1.48it/s]
1|125|Loss: 6.152309417724609:   2%|▏         | 124/6470 [04:25<1:11:18,  1.48it/s]
1|125|Loss: 6.152309417724609:   2%|▏         | 125/6470 [04:25<1:16:41,  1.38it/s]
1|126|Loss: 6.034735679626465:   2%|▏         | 125/6470 [04:26<1:16:41,  1.38it/s]
1|126|Loss: 6.034735679626465:   2%|▏         | 126/6470 [04:26<1:16:35,  1.38it/s]
1|127|Loss: 6.053035736083984:   2%|▏         | 126/6470 [04:26<1:16:35,  1.38it/s]
1|127|Loss: 6.053035736083984:   2%|▏         | 127/6470 [04:27<1:15:35,  1.40it/s]
1|128|Loss: 6.083605766296387:   2%|▏         | 127/6470 [04:27<1:15:35,  1.40it/s]
1|128|Loss: 6.083605766296387:   2%|▏         | 128/6470 [04:27<1:13:00,  1.45it/s]
1|129|Loss: 6.177404880523682:   2%|▏         | 128/6470 [04:28<1:13:00,  1.45it/s]
1|129|Loss: 6.177404880523682:   2%|▏         | 129/6470 [04:28<1:08:19,  1.55it/s]
1|130|Loss: 6.552560806274414:   2%|▏         | 129/6470 [04:28<1:08:19,  1.55it/s]
1|130|Loss: 6.552560806274414:   2%|▏         | 130/6470 [04:29<1:13:34,  1.44it/s]
1|131|Loss: 6.239373207092285:   2%|▏         | 130/6470 [04:29<1:13:34,  1.44it/s]
1|131|Loss: 6.239373207092285:   2%|▏         | 131/6470 [04:30<1:15:39,  1.40it/s]
1|132|Loss: 6.114636421203613:   2%|▏         | 131/6470 [04:30<1:15:39,  1.40it/s]
1|132|Loss: 6.114636421203613:   2%|▏         | 132/6470 [04:30<1:14:32,  1.42it/s]
1|133|Loss: 6.254937648773193:   2%|▏         | 132/6470 [04:31<1:14:32,  1.42it/s]
1|133|Loss: 6.254937648773193:   2%|▏         | 133/6470 [04:31<1:13:47,  1.43it/s]
1|134|Loss: 6.0200090408325195:   2%|▏         | 133/6470 [04:31<1:13:47,  1.43it/s]
1|134|Loss: 6.0200090408325195:   2%|▏         | 134/6470 [04:32<1:18:03,  1.35it/s]
1|135|Loss: 6.141507625579834:   2%|▏         | 134/6470 [04:32<1:18:03,  1.35it/s] 
1|135|Loss: 6.141507625579834:   2%|▏         | 135/6470 [04:32<1:17:40,  1.36it/s]
1|136|Loss: 5.667835235595703:   2%|▏         | 135/6470 [04:33<1:17:40,  1.36it/s]
1|136|Loss: 5.667835235595703:   2%|▏         | 136/6470 [04:33<1:15:06,  1.41it/s]
1|137|Loss: 6.203534126281738:   2%|▏         | 136/6470 [04:33<1:15:06,  1.41it/s]
1|137|Loss: 6.203534126281738:   2%|▏         | 137/6470 [04:34<1:12:32,  1.46it/s]
1|138|Loss: 6.296191215515137:   2%|▏         | 137/6470 [04:34<1:12:32,  1.46it/s]
1|138|Loss: 6.296191215515137:   2%|▏         | 138/6470 [04:35<1:14:20,  1.42it/s]
1|139|Loss: 5.912018775939941:   2%|▏         | 138/6470 [04:35<1:14:20,  1.42it/s]
1|139|Loss: 5.912018775939941:   2%|▏         | 139/6470 [04:35<1:14:10,  1.42it/s]
1|140|Loss: 6.214216709136963:   2%|▏         | 139/6470 [04:36<1:14:10,  1.42it/s]
1|140|Loss: 6.214216709136963:   2%|▏         | 140/6470 [04:36<1:16:15,  1.38it/s]
1|141|Loss: 5.807014465332031:   2%|▏         | 140/6470 [04:36<1:16:15,  1.38it/s]
1|141|Loss: 5.807014465332031:   2%|▏         | 141/6470 [04:37<1:13:59,  1.43it/s]
1|142|Loss: 6.6638078689575195:   2%|▏         | 141/6470 [04:37<1:13:59,  1.43it/s]
1|142|Loss: 6.6638078689575195:   2%|▏         | 142/6470 [04:37<1:15:41,  1.39it/s]
1|143|Loss: 6.253081321716309:   2%|▏         | 142/6470 [04:38<1:15:41,  1.39it/s] 
1|143|Loss: 6.253081321716309:   2%|▏         | 143/6470 [04:38<1:16:56,  1.37it/s]
1|144|Loss: 5.702915668487549:   2%|▏         | 143/6470 [04:39<1:16:56,  1.37it/s]
1|144|Loss: 5.702915668487549:   2%|▏         | 144/6470 [04:39<1:15:41,  1.39it/s]
1|145|Loss: 5.825263023376465:   2%|▏         | 144/6470 [04:39<1:15:41,  1.39it/s]
1|145|Loss: 5.825263023376465:   2%|▏         | 145/6470 [04:40<1:16:04,  1.39it/s]
1|146|Loss: 5.863857746124268:   2%|▏         | 145/6470 [04:40<1:16:04,  1.39it/s]
1|146|Loss: 5.863857746124268:   2%|▏         | 146/6470 [04:40<1:17:22,  1.36it/s]
1|147|Loss: 6.24310302734375:   2%|▏         | 146/6470 [04:41<1:17:22,  1.36it/s] 
1|147|Loss: 6.24310302734375:   2%|▏         | 147/6470 [04:41<1:25:01,  1.24it/s]
1|148|Loss: 6.140312671661377:   2%|▏         | 147/6470 [04:42<1:25:01,  1.24it/s]
1|148|Loss: 6.140312671661377:   2%|▏         | 148/6470 [04:42<1:25:42,  1.23it/s]
1|149|Loss: 5.707696437835693:   2%|▏         | 148/6470 [04:42<1:25:42,  1.23it/s]
1|149|Loss: 5.707696437835693:   2%|▏         | 149/6470 [04:43<1:19:05,  1.33it/s]
1|150|Loss: 5.906042575836182:   2%|▏         | 149/6470 [04:43<1:19:05,  1.33it/s]
1|150|Loss: 5.906042575836182:   2%|▏         | 150/6470 [04:44<1:19:56,  1.32it/s]
1|151|Loss: 5.604804039001465:   2%|▏         | 150/6470 [04:44<1:19:56,  1.32it/s]
1|151|Loss: 5.604804039001465:   2%|▏         | 151/6470 [04:44<1:16:34,  1.38it/s]
1|152|Loss: 5.604128360748291:   2%|▏         | 151/6470 [04:45<1:16:34,  1.38it/s]
1|152|Loss: 5.604128360748291:   2%|▏         | 152/6470 [04:45<1:13:43,  1.43it/s]
1|153|Loss: 6.422693729400635:   2%|▏         | 152/6470 [04:45<1:13:43,  1.43it/s]
1|153|Loss: 6.422693729400635:   2%|▏         | 153/6470 [04:45<1:13:18,  1.44it/s]
1|154|Loss: 6.431737899780273:   2%|▏         | 153/6470 [04:46<1:13:18,  1.44it/s]
1|154|Loss: 6.431737899780273:   2%|▏         | 154/6470 [04:46<1:18:29,  1.34it/s]
1|155|Loss: 6.002997398376465:   2%|▏         | 154/6470 [04:47<1:18:29,  1.34it/s]
1|155|Loss: 6.002997398376465:   2%|▏         | 155/6470 [04:47<1:22:00,  1.28it/s]
1|156|Loss: 5.572723865509033:   2%|▏         | 155/6470 [04:48<1:22:00,  1.28it/s]
1|156|Loss: 5.572723865509033:   2%|▏         | 156/6470 [04:48<1:18:59,  1.33it/s]
1|157|Loss: 5.354686737060547:   2%|▏         | 156/6470 [04:48<1:18:59,  1.33it/s]
1|157|Loss: 5.354686737060547:   2%|▏         | 157/6470 [04:48<1:12:36,  1.45it/s]
1|158|Loss: 6.166005611419678:   2%|▏         | 157/6470 [04:49<1:12:36,  1.45it/s]
1|158|Loss: 6.166005611419678:   2%|▏         | 158/6470 [04:49<1:14:02,  1.42it/s]
1|159|Loss: 5.645920753479004:   2%|▏         | 158/6470 [04:50<1:14:02,  1.42it/s]
1|159|Loss: 5.645920753479004:   2%|▏         | 159/6470 [04:50<1:15:26,  1.39it/s]
1|160|Loss: 5.541079044342041:   2%|▏         | 159/6470 [04:50<1:15:26,  1.39it/s]
1|160|Loss: 5.541079044342041:   2%|▏         | 160/6470 [04:51<1:12:37,  1.45it/s]
1|161|Loss: 6.00416374206543:   2%|▏         | 160/6470 [04:51<1:12:37,  1.45it/s] 
1|161|Loss: 6.00416374206543:   2%|▏         | 161/6470 [04:51<1:11:03,  1.48it/s]
1|162|Loss: 6.189484119415283:   2%|▏         | 161/6470 [04:52<1:11:03,  1.48it/s]
1|162|Loss: 6.189484119415283:   3%|▎         | 162/6470 [04:52<1:16:37,  1.37it/s]
1|163|Loss: 6.141447067260742:   3%|▎         | 162/6470 [04:53<1:16:37,  1.37it/s]
1|163|Loss: 6.141447067260742:   3%|▎         | 163/6470 [04:53<1:21:29,  1.29it/s]
1|164|Loss: 5.531610012054443:   3%|▎         | 163/6470 [04:53<1:21:29,  1.29it/s]
1|164|Loss: 5.531610012054443:   3%|▎         | 164/6470 [04:54<1:15:47,  1.39it/s]
1|165|Loss: 6.19791316986084:   3%|▎         | 164/6470 [04:54<1:15:47,  1.39it/s] 
1|165|Loss: 6.19791316986084:   3%|▎         | 165/6470 [04:54<1:18:25,  1.34it/s]
1|166|Loss: 5.847841739654541:   3%|▎         | 165/6470 [04:55<1:18:25,  1.34it/s]
1|166|Loss: 5.847841739654541:   3%|▎         | 166/6470 [04:55<1:20:05,  1.31it/s]
1|167|Loss: 6.085976600646973:   3%|▎         | 166/6470 [04:56<1:20:05,  1.31it/s]
1|167|Loss: 6.085976600646973:   3%|▎         | 167/6470 [04:56<1:20:02,  1.31it/s]
1|168|Loss: 6.153176784515381:   3%|▎         | 167/6470 [04:56<1:20:02,  1.31it/s]
1|168|Loss: 6.153176784515381:   3%|▎         | 168/6470 [04:57<1:20:47,  1.30it/s]
1|169|Loss: 6.203190326690674:   3%|▎         | 168/6470 [04:57<1:20:47,  1.30it/s]
1|169|Loss: 6.203190326690674:   3%|▎         | 169/6470 [04:58<1:24:42,  1.24it/s]
1|170|Loss: 5.906457424163818:   3%|▎         | 169/6470 [04:58<1:24:42,  1.24it/s]
1|170|Loss: 5.906457424163818:   3%|▎         | 170/6470 [04:58<1:24:38,  1.24it/s]
1|171|Loss: 6.0220537185668945:   3%|▎         | 170/6470 [04:59<1:24:38,  1.24it/s]
1|171|Loss: 6.0220537185668945:   3%|▎         | 171/6470 [04:59<1:26:32,  1.21it/s]
1|172|Loss: 5.698780059814453:   3%|▎         | 171/6470 [05:00<1:26:32,  1.21it/s] 
1|172|Loss: 5.698780059814453:   3%|▎         | 172/6470 [05:00<1:23:10,  1.26it/s]
1|173|Loss: 5.8715500831604:   3%|▎         | 172/6470 [05:00<1:23:10,  1.26it/s]  
1|173|Loss: 5.8715500831604:   3%|▎         | 173/6470 [05:01<1:26:16,  1.22it/s]
1|174|Loss: 5.8187713623046875:   3%|▎         | 173/6470 [05:01<1:26:16,  1.22it/s]
1|174|Loss: 5.8187713623046875:   3%|▎         | 174/6470 [05:02<1:25:08,  1.23it/s]
1|175|Loss: 5.476943492889404:   3%|▎         | 174/6470 [05:02<1:25:08,  1.23it/s] 
1|175|Loss: 5.476943492889404:   3%|▎         | 175/6470 [05:02<1:22:54,  1.27it/s]
1|176|Loss: 5.715050220489502:   3%|▎         | 175/6470 [05:03<1:22:54,  1.27it/s]
1|176|Loss: 5.715050220489502:   3%|▎         | 176/6470 [05:03<1:19:50,  1.31it/s]
1|177|Loss: 5.629769802093506:   3%|▎         | 176/6470 [05:03<1:19:50,  1.31it/s]
1|177|Loss: 5.629769802093506:   3%|▎         | 177/6470 [05:04<1:19:53,  1.31it/s]
1|178|Loss: 5.889807224273682:   3%|▎         | 177/6470 [05:04<1:19:53,  1.31it/s]
1|178|Loss: 5.889807224273682:   3%|▎         | 178/6470 [05:05<1:24:08,  1.25it/s]
1|179|Loss: 5.90416145324707:   3%|▎         | 178/6470 [05:05<1:24:08,  1.25it/s] 
1|179|Loss: 5.90416145324707:   3%|▎         | 179/6470 [05:06<1:29:26,  1.17it/s]
1|180|Loss: 5.936648845672607:   3%|▎         | 179/6470 [05:06<1:29:26,  1.17it/s]
1|180|Loss: 5.936648845672607:   3%|▎         | 180/6470 [05:07<1:30:16,  1.16it/s]
1|181|Loss: 5.679812431335449:   3%|▎         | 180/6470 [05:07<1:30:16,  1.16it/s]
1|181|Loss: 5.679812431335449:   3%|▎         | 181/6470 [05:07<1:29:55,  1.17it/s]
1|182|Loss: 6.145478248596191:   3%|▎         | 181/6470 [05:08<1:29:55,  1.17it/s]
1|182|Loss: 6.145478248596191:   3%|▎         | 182/6470 [05:08<1:25:45,  1.22it/s]
1|183|Loss: 6.330742835998535:   3%|▎         | 182/6470 [05:09<1:25:45,  1.22it/s]
1|183|Loss: 6.330742835998535:   3%|▎         | 183/6470 [05:09<1:24:33,  1.24it/s]
1|184|Loss: 5.915980815887451:   3%|▎         | 183/6470 [05:09<1:24:33,  1.24it/s]
1|184|Loss: 5.915980815887451:   3%|▎         | 184/6470 [05:10<1:25:31,  1.22it/s]
1|185|Loss: 5.7202348709106445:   3%|▎         | 184/6470 [05:10<1:25:31,  1.22it/s]
1|185|Loss: 5.7202348709106445:   3%|▎         | 185/6470 [05:11<1:26:44,  1.21it/s]
1|186|Loss: 5.765228748321533:   3%|▎         | 185/6470 [05:11<1:26:44,  1.21it/s] 
1|186|Loss: 5.765228748321533:   3%|▎         | 186/6470 [05:11<1:21:33,  1.28it/s]
1|187|Loss: 5.803440570831299:   3%|▎         | 186/6470 [05:12<1:21:33,  1.28it/s]
1|187|Loss: 5.803440570831299:   3%|▎         | 187/6470 [05:12<1:20:28,  1.30it/s]
1|188|Loss: 5.861421585083008:   3%|▎         | 187/6470 [05:12<1:20:28,  1.30it/s]
1|188|Loss: 5.861421585083008:   3%|▎         | 188/6470 [05:13<1:18:10,  1.34it/s]
1|189|Loss: 6.056919574737549:   3%|▎         | 188/6470 [05:13<1:18:10,  1.34it/s]
1|189|Loss: 6.056919574737549:   3%|▎         | 189/6470 [05:14<1:18:47,  1.33it/s]
1|190|Loss: 5.910367488861084:   3%|▎         | 189/6470 [05:14<1:18:47,  1.33it/s]
1|190|Loss: 5.910367488861084:   3%|▎         | 190/6470 [05:14<1:21:51,  1.28it/s]
1|191|Loss: 6.181035995483398:   3%|▎         | 190/6470 [05:15<1:21:51,  1.28it/s]
1|191|Loss: 6.181035995483398:   3%|▎         | 191/6470 [05:15<1:21:14,  1.29it/s]
1|192|Loss: 5.9643073081970215:   3%|▎         | 191/6470 [05:16<1:21:14,  1.29it/s]
1|192|Loss: 5.9643073081970215:   3%|▎         | 192/6470 [05:16<1:25:09,  1.23it/s]
1|193|Loss: 4.9475603103637695:   3%|▎         | 192/6470 [05:16<1:25:09,  1.23it/s]
1|193|Loss: 4.9475603103637695:   3%|▎         | 193/6470 [05:17<1:22:49,  1.26it/s]
1|194|Loss: 5.96120548248291:   3%|▎         | 193/6470 [05:17<1:22:49,  1.26it/s]  
1|194|Loss: 5.96120548248291:   3%|▎         | 194/6470 [05:18<1:23:53,  1.25it/s]
1|195|Loss: 6.067621231079102:   3%|▎         | 194/6470 [05:18<1:23:53,  1.25it/s]
1|195|Loss: 6.067621231079102:   3%|▎         | 195/6470 [05:18<1:19:34,  1.31it/s]
1|196|Loss: 5.613443374633789:   3%|▎         | 195/6470 [05:19<1:19:34,  1.31it/s]
1|196|Loss: 5.613443374633789:   3%|▎         | 196/6470 [05:19<1:19:06,  1.32it/s]
1|197|Loss: 5.954296588897705:   3%|▎         | 196/6470 [05:19<1:19:06,  1.32it/s]
1|197|Loss: 5.954296588897705:   3%|▎         | 197/6470 [05:20<1:22:39,  1.26it/s]
1|198|Loss: 5.014886379241943:   3%|▎         | 197/6470 [05:20<1:22:39,  1.26it/s]
1|198|Loss: 5.014886379241943:   3%|▎         | 198/6470 [05:21<1:23:03,  1.26it/s]
1|199|Loss: 5.902958393096924:   3%|▎         | 198/6470 [05:21<1:23:03,  1.26it/s]
1|199|Loss: 5.902958393096924:   3%|▎         | 199/6470 [05:21<1:20:42,  1.29it/s]
1|200|Loss: 5.824342727661133:   3%|▎         | 199/6470 [05:22<1:20:42,  1.29it/s]
1|200|Loss: 5.824342727661133:   3%|▎         | 200/6470 [05:22<1:18:05,  1.34it/s]
1|201|Loss: 6.132948875427246:   3%|▎         | 200/6470 [05:22<1:18:05,  1.34it/s]
1|201|Loss: 6.132948875427246:   3%|▎         | 201/6470 [05:23<1:15:13,  1.39it/s]
1|202|Loss: 6.058237552642822:   3%|▎         | 201/6470 [05:23<1:15:13,  1.39it/s]
1|202|Loss: 6.058237552642822:   3%|▎         | 202/6470 [05:24<1:16:39,  1.36it/s]
1|203|Loss: 5.3394551277160645:   3%|▎         | 202/6470 [05:24<1:16:39,  1.36it/s]
1|203|Loss: 5.3394551277160645:   3%|▎         | 203/6470 [05:24<1:13:58,  1.41it/s]
1|204|Loss: 5.964628219604492:   3%|▎         | 203/6470 [05:25<1:13:58,  1.41it/s] 
1|204|Loss: 5.964628219604492:   3%|▎         | 204/6470 [05:25<1:11:18,  1.46it/s]
1|205|Loss: 5.439516067504883:   3%|▎         | 204/6470 [05:25<1:11:18,  1.46it/s]
1|205|Loss: 5.439516067504883:   3%|▎         | 205/6470 [05:25<1:10:13,  1.49it/s]
1|206|Loss: 4.542027950286865:   3%|▎         | 205/6470 [05:26<1:10:13,  1.49it/s]
1|206|Loss: 4.542027950286865:   3%|▎         | 206/6470 [05:26<1:05:40,  1.59it/s]
1|207|Loss: 5.577425003051758:   3%|▎         | 206/6470 [05:26<1:05:40,  1.59it/s]
1|207|Loss: 5.577425003051758:   3%|▎         | 207/6470 [05:27<1:10:13,  1.49it/s]
1|208|Loss: 6.078001022338867:   3%|▎         | 207/6470 [05:27<1:10:13,  1.49it/s]
1|208|Loss: 6.078001022338867:   3%|▎         | 208/6470 [05:28<1:13:46,  1.41it/s]
1|209|Loss: 5.529018402099609:   3%|▎         | 208/6470 [05:28<1:13:46,  1.41it/s]
1|209|Loss: 5.529018402099609:   3%|▎         | 209/6470 [05:28<1:13:56,  1.41it/s]
1|210|Loss: 5.869032859802246:   3%|▎         | 209/6470 [05:29<1:13:56,  1.41it/s]
1|210|Loss: 5.869032859802246:   3%|▎         | 210/6470 [05:29<1:17:44,  1.34it/s]
1|211|Loss: 5.070560932159424:   3%|▎         | 210/6470 [05:29<1:17:44,  1.34it/s]
1|211|Loss: 5.070560932159424:   3%|▎         | 211/6470 [05:30<1:18:42,  1.33it/s]
1|212|Loss: 5.965181350708008:   3%|▎         | 211/6470 [05:30<1:18:42,  1.33it/s]
1|212|Loss: 5.965181350708008:   3%|▎         | 212/6470 [05:31<1:20:03,  1.30it/s]
1|213|Loss: 5.963892459869385:   3%|▎         | 212/6470 [05:31<1:20:03,  1.30it/s]
1|213|Loss: 5.963892459869385:   3%|▎         | 213/6470 [05:32<1:23:13,  1.25it/s]
1|214|Loss: 5.9310736656188965:   3%|▎         | 213/6470 [05:32<1:23:13,  1.25it/s]
1|214|Loss: 5.9310736656188965:   3%|▎         | 214/6470 [05:32<1:23:20,  1.25it/s]
1|215|Loss: 6.2674455642700195:   3%|▎         | 214/6470 [05:33<1:23:20,  1.25it/s]
1|215|Loss: 6.2674455642700195:   3%|▎         | 215/6470 [05:33<1:26:57,  1.20it/s]
1|216|Loss: 5.341044902801514:   3%|▎         | 215/6470 [05:34<1:26:57,  1.20it/s] 
1|216|Loss: 5.341044902801514:   3%|▎         | 216/6470 [05:34<1:25:53,  1.21it/s]
1|217|Loss: 5.705192565917969:   3%|▎         | 216/6470 [05:34<1:25:53,  1.21it/s]
1|217|Loss: 5.705192565917969:   3%|▎         | 217/6470 [05:35<1:25:50,  1.21it/s]
1|218|Loss: 5.807313442230225:   3%|▎         | 217/6470 [05:35<1:25:50,  1.21it/s]
1|218|Loss: 5.807313442230225:   3%|▎         | 218/6470 [05:36<1:27:33,  1.19it/s]
1|219|Loss: 5.238039016723633:   3%|▎         | 218/6470 [05:36<1:27:33,  1.19it/s]
1|219|Loss: 5.238039016723633:   3%|▎         | 219/6470 [05:37<1:30:24,  1.15it/s]
1|220|Loss: 5.968912601470947:   3%|▎         | 219/6470 [05:37<1:30:24,  1.15it/s]
1|220|Loss: 5.968912601470947:   3%|▎         | 220/6470 [05:38<1:32:21,  1.13it/s]
1|221|Loss: 5.717894554138184:   3%|▎         | 220/6470 [05:38<1:32:21,  1.13it/s]
1|221|Loss: 5.717894554138184:   3%|▎         | 221/6470 [05:38<1:26:29,  1.20it/s]
1|222|Loss: 5.06714391708374:   3%|▎         | 221/6470 [05:39<1:26:29,  1.20it/s] 
1|222|Loss: 5.06714391708374:   3%|▎         | 222/6470 [05:39<1:20:23,  1.30it/s]
1|223|Loss: 5.813944339752197:   3%|▎         | 222/6470 [05:39<1:20:23,  1.30it/s]
1|223|Loss: 5.813944339752197:   3%|▎         | 223/6470 [05:40<1:19:09,  1.32it/s]
1|224|Loss: 5.8749165534973145:   3%|▎         | 223/6470 [05:40<1:19:09,  1.32it/s]
1|224|Loss: 5.8749165534973145:   3%|▎         | 224/6470 [05:40<1:20:18,  1.30it/s]
1|225|Loss: 5.889381408691406:   3%|▎         | 224/6470 [05:41<1:20:18,  1.30it/s] 
1|225|Loss: 5.889381408691406:   3%|▎         | 225/6470 [05:41<1:21:37,  1.28it/s]
1|226|Loss: 5.5069098472595215:   3%|▎         | 225/6470 [05:42<1:21:37,  1.28it/s]
1|226|Loss: 5.5069098472595215:   3%|▎         | 226/6470 [05:42<1:21:56,  1.27it/s]
1|227|Loss: 5.482190132141113:   3%|▎         | 226/6470 [05:42<1:21:56,  1.27it/s] 
1|227|Loss: 5.482190132141113:   4%|▎         | 227/6470 [05:43<1:21:42,  1.27it/s]
1|228|Loss: 5.531434535980225:   4%|▎         | 227/6470 [05:43<1:21:42,  1.27it/s]
1|228|Loss: 5.531434535980225:   4%|▎         | 228/6470 [05:44<1:20:40,  1.29it/s]
1|229|Loss: 5.031965732574463:   4%|▎         | 228/6470 [05:44<1:20:40,  1.29it/s]
1|229|Loss: 5.031965732574463:   4%|▎         | 229/6470 [05:44<1:16:26,  1.36it/s]
1|230|Loss: 5.849062442779541:   4%|▎         | 229/6470 [05:45<1:16:26,  1.36it/s]
1|230|Loss: 5.849062442779541:   4%|▎         | 230/6470 [05:45<1:14:24,  1.40it/s]
1|231|Loss: 5.939398288726807:   4%|▎         | 230/6470 [05:45<1:14:24,  1.40it/s]
1|231|Loss: 5.939398288726807:   4%|▎         | 231/6470 [05:46<1:16:44,  1.35it/s]
1|232|Loss: 4.414419174194336:   4%|▎         | 231/6470 [05:46<1:16:44,  1.35it/s]
1|232|Loss: 4.414419174194336:   4%|▎         | 232/6470 [05:46<1:14:13,  1.40it/s]
1|233|Loss: 5.87341833114624:   4%|▎         | 232/6470 [05:47<1:14:13,  1.40it/s] 
1|233|Loss: 5.87341833114624:   4%|▎         | 233/6470 [05:47<1:10:57,  1.46it/s]
1|234|Loss: 5.5823211669921875:   4%|▎         | 233/6470 [05:47<1:10:57,  1.46it/s]
1|234|Loss: 5.5823211669921875:   4%|▎         | 234/6470 [05:48<1:09:01,  1.51it/s]
1|235|Loss: 5.268110275268555:   4%|▎         | 234/6470 [05:48<1:09:01,  1.51it/s] 
1|235|Loss: 5.268110275268555:   4%|▎         | 235/6470 [05:48<1:11:38,  1.45it/s]
1|236|Loss: 5.5823163986206055:   4%|▎         | 235/6470 [05:49<1:11:38,  1.45it/s]
1|236|Loss: 5.5823163986206055:   4%|▎         | 236/6470 [05:49<1:14:26,  1.40it/s]
1|237|Loss: 5.462166786193848:   4%|▎         | 236/6470 [05:50<1:14:26,  1.40it/s] 
1|237|Loss: 5.462166786193848:   4%|▎         | 237/6470 [05:50<1:15:43,  1.37it/s]
1|238|Loss: 5.439818382263184:   4%|▎         | 237/6470 [05:50<1:15:43,  1.37it/s]
1|238|Loss: 5.439818382263184:   4%|▎         | 238/6470 [05:51<1:14:01,  1.40it/s]
1|239|Loss: 5.328843593597412:   4%|▎         | 238/6470 [05:51<1:14:01,  1.40it/s]
1|239|Loss: 5.328843593597412:   4%|▎         | 239/6470 [05:51<1:09:18,  1.50it/s]
1|240|Loss: 5.559330940246582:   4%|▎         | 239/6470 [05:51<1:09:18,  1.50it/s]
1|240|Loss: 5.559330940246582:   4%|▎         | 240/6470 [05:52<1:11:14,  1.46it/s]
1|241|Loss: 5.252201557159424:   4%|▎         | 240/6470 [05:52<1:11:14,  1.46it/s]
1|241|Loss: 5.252201557159424:   4%|▎         | 241/6470 [05:53<1:14:13,  1.40it/s]
1|242|Loss: 5.776022911071777:   4%|▎         | 241/6470 [05:53<1:14:13,  1.40it/s]
1|242|Loss: 5.776022911071777:   4%|▎         | 242/6470 [05:53<1:19:21,  1.31it/s]
1|243|Loss: 5.698141574859619:   4%|▎         | 242/6470 [05:54<1:19:21,  1.31it/s]
1|243|Loss: 5.698141574859619:   4%|▍         | 243/6470 [05:54<1:19:49,  1.30it/s]
1|244|Loss: 5.58913516998291:   4%|▍         | 243/6470 [05:55<1:19:49,  1.30it/s] 
1|244|Loss: 5.58913516998291:   4%|▍         | 244/6470 [05:55<1:19:44,  1.30it/s]
1|245|Loss: 5.46005392074585:   4%|▍         | 244/6470 [05:55<1:19:44,  1.30it/s]
1|245|Loss: 5.46005392074585:   4%|▍         | 245/6470 [05:56<1:19:19,  1.31it/s]
1|246|Loss: 5.923584938049316:   4%|▍         | 245/6470 [05:56<1:19:19,  1.31it/s]
1|246|Loss: 5.923584938049316:   4%|▍         | 246/6470 [05:57<1:19:17,  1.31it/s]
1|247|Loss: 5.367226600646973:   4%|▍         | 246/6470 [05:57<1:19:17,  1.31it/s]
1|247|Loss: 5.367226600646973:   4%|▍         | 247/6470 [05:57<1:15:04,  1.38it/s]
1|248|Loss: 5.962573528289795:   4%|▍         | 247/6470 [05:58<1:15:04,  1.38it/s]
1|248|Loss: 5.962573528289795:   4%|▍         | 248/6470 [05:58<1:15:57,  1.37it/s]
1|249|Loss: 5.738149642944336:   4%|▍         | 248/6470 [05:58<1:15:57,  1.37it/s]
1|249|Loss: 5.738149642944336:   4%|▍         | 249/6470 [05:59<1:22:36,  1.26it/s]
1|250|Loss: 5.29693603515625:   4%|▍         | 249/6470 [05:59<1:22:36,  1.26it/s] 
1|250|Loss: 5.29693603515625:   4%|▍         | 250/6470 [06:00<1:18:48,  1.32it/s]
1|251|Loss: 4.685218811035156:   4%|▍         | 250/6470 [06:00<1:18:48,  1.32it/s]
1|251|Loss: 4.685218811035156:   4%|▍         | 251/6470 [06:00<1:11:59,  1.44it/s]
1|252|Loss: 5.875744819641113:   4%|▍         | 251/6470 [06:01<1:11:59,  1.44it/s]
1|252|Loss: 5.875744819641113:   4%|▍         | 252/6470 [06:01<1:18:26,  1.32it/s]
1|253|Loss: 5.657628536224365:   4%|▍         | 252/6470 [06:01<1:18:26,  1.32it/s]
1|253|Loss: 5.657628536224365:   4%|▍         | 253/6470 [06:02<1:18:39,  1.32it/s]
1|254|Loss: 5.598312854766846:   4%|▍         | 253/6470 [06:02<1:18:39,  1.32it/s]
1|254|Loss: 5.598312854766846:   4%|▍         | 254/6470 [06:02<1:12:48,  1.42it/s]
1|255|Loss: 5.546394348144531:   4%|▍         | 254/6470 [06:03<1:12:48,  1.42it/s]
1|255|Loss: 5.546394348144531:   4%|▍         | 255/6470 [06:03<1:11:36,  1.45it/s]
1|256|Loss: 5.711669445037842:   4%|▍         | 255/6470 [06:03<1:11:36,  1.45it/s]
1|256|Loss: 5.711669445037842:   4%|▍         | 256/6470 [06:04<1:12:09,  1.44it/s]
1|257|Loss: 5.672978401184082:   4%|▍         | 256/6470 [06:04<1:12:09,  1.44it/s]
1|257|Loss: 5.672978401184082:   4%|▍         | 257/6470 [06:04<1:12:12,  1.43it/s]
1|258|Loss: 5.60942268371582:   4%|▍         | 257/6470 [06:05<1:12:12,  1.43it/s] 
1|258|Loss: 5.60942268371582:   4%|▍         | 258/6470 [06:05<1:14:35,  1.39it/s]
1|259|Loss: 5.385684490203857:   4%|▍         | 258/6470 [06:06<1:14:35,  1.39it/s]
1|259|Loss: 5.385684490203857:   4%|▍         | 259/6470 [06:06<1:16:22,  1.36it/s]
1|260|Loss: 5.589463710784912:   4%|▍         | 259/6470 [06:06<1:16:22,  1.36it/s]
1|260|Loss: 5.589463710784912:   4%|▍         | 260/6470 [06:07<1:14:40,  1.39it/s]
1|261|Loss: 5.981697082519531:   4%|▍         | 260/6470 [06:07<1:14:40,  1.39it/s]
1|261|Loss: 5.981697082519531:   4%|▍         | 261/6470 [06:08<1:19:40,  1.30it/s]
1|262|Loss: 5.438195705413818:   4%|▍         | 261/6470 [06:08<1:19:40,  1.30it/s]
1|262|Loss: 5.438195705413818:   4%|▍         | 262/6470 [06:08<1:18:26,  1.32it/s]
1|263|Loss: 5.1568169593811035:   4%|▍         | 262/6470 [06:09<1:18:26,  1.32it/s]
1|263|Loss: 5.1568169593811035:   4%|▍         | 263/6470 [06:09<1:15:24,  1.37it/s]
1|264|Loss: 5.131386756896973:   4%|▍         | 263/6470 [06:09<1:15:24,  1.37it/s] 
1|264|Loss: 5.131386756896973:   4%|▍         | 264/6470 [06:09<1:10:17,  1.47it/s]
1|265|Loss: 4.73480224609375:   4%|▍         | 264/6470 [06:10<1:10:17,  1.47it/s] 
1|265|Loss: 4.73480224609375:   4%|▍         | 265/6470 [06:10<1:08:53,  1.50it/s]
1|266|Loss: 5.731576442718506:   4%|▍         | 265/6470 [06:10<1:08:53,  1.50it/s]
1|266|Loss: 5.731576442718506:   4%|▍         | 266/6470 [06:11<1:08:21,  1.51it/s]
1|267|Loss: 4.538343906402588:   4%|▍         | 266/6470 [06:11<1:08:21,  1.51it/s]
1|267|Loss: 4.538343906402588:   4%|▍         | 267/6470 [06:11<1:01:43,  1.67it/s]
1|268|Loss: 5.496281623840332:   4%|▍         | 267/6470 [06:12<1:01:43,  1.67it/s]
1|268|Loss: 5.496281623840332:   4%|▍         | 268/6470 [06:12<1:03:15,  1.63it/s]
1|269|Loss: 5.802313327789307:   4%|▍         | 268/6470 [06:12<1:03:15,  1.63it/s]
1|269|Loss: 5.802313327789307:   4%|▍         | 269/6470 [06:13<1:11:42,  1.44it/s]
1|270|Loss: 6.026726245880127:   4%|▍         | 269/6470 [06:13<1:11:42,  1.44it/s]
1|270|Loss: 6.026726245880127:   4%|▍         | 270/6470 [06:14<1:14:31,  1.39it/s]
1|271|Loss: 5.761280059814453:   4%|▍         | 270/6470 [06:14<1:14:31,  1.39it/s]
1|271|Loss: 5.761280059814453:   4%|▍         | 271/6470 [06:14<1:12:06,  1.43it/s]
1|272|Loss: 5.346465110778809:   4%|▍         | 271/6470 [06:15<1:12:06,  1.43it/s]
1|272|Loss: 5.346465110778809:   4%|▍         | 272/6470 [06:15<1:09:51,  1.48it/s]
1|273|Loss: 6.156787872314453:   4%|▍         | 272/6470 [06:15<1:09:51,  1.48it/s]
1|273|Loss: 6.156787872314453:   4%|▍         | 273/6470 [06:16<1:17:43,  1.33it/s]
1|274|Loss: 5.580544948577881:   4%|▍         | 273/6470 [06:16<1:17:43,  1.33it/s]
1|274|Loss: 5.580544948577881:   4%|▍         | 274/6470 [06:17<1:20:04,  1.29it/s]
1|275|Loss: 5.361913681030273:   4%|▍         | 274/6470 [06:17<1:20:04,  1.29it/s]
1|275|Loss: 5.361913681030273:   4%|▍         | 275/6470 [06:17<1:17:40,  1.33it/s]
1|276|Loss: 5.4039201736450195:   4%|▍         | 275/6470 [06:18<1:17:40,  1.33it/s]
1|276|Loss: 5.4039201736450195:   4%|▍         | 276/6470 [06:18<1:13:42,  1.40it/s]
1|277|Loss: 5.059382438659668:   4%|▍         | 276/6470 [06:18<1:13:42,  1.40it/s] 
1|277|Loss: 5.059382438659668:   4%|▍         | 277/6470 [06:18<1:08:11,  1.51it/s]
1|278|Loss: 5.759930610656738:   4%|▍         | 277/6470 [06:19<1:08:11,  1.51it/s]
1|278|Loss: 5.759930610656738:   4%|▍         | 278/6470 [06:19<1:10:00,  1.47it/s]
1|279|Loss: 5.746556282043457:   4%|▍         | 278/6470 [06:20<1:10:00,  1.47it/s]
1|279|Loss: 5.746556282043457:   4%|▍         | 279/6470 [06:20<1:15:23,  1.37it/s]
1|280|Loss: 5.63177490234375:   4%|▍         | 279/6470 [06:20<1:15:23,  1.37it/s] 
1|280|Loss: 5.63177490234375:   4%|▍         | 280/6470 [06:21<1:17:37,  1.33it/s]
1|281|Loss: 6.0308027267456055:   4%|▍         | 280/6470 [06:21<1:17:37,  1.33it/s]
1|281|Loss: 6.0308027267456055:   4%|▍         | 281/6470 [06:22<1:22:13,  1.25it/s]
1|282|Loss: 5.677985668182373:   4%|▍         | 281/6470 [06:22<1:22:13,  1.25it/s] 
1|282|Loss: 5.677985668182373:   4%|▍         | 282/6470 [06:23<1:27:22,  1.18it/s]
1|283|Loss: 4.714619159698486:   4%|▍         | 282/6470 [06:23<1:27:22,  1.18it/s]
1|283|Loss: 4.714619159698486:   4%|▍         | 283/6470 [06:23<1:22:57,  1.24it/s]
1|284|Loss: 5.643869400024414:   4%|▍         | 283/6470 [06:24<1:22:57,  1.24it/s]
1|284|Loss: 5.643869400024414:   4%|▍         | 284/6470 [06:24<1:24:06,  1.23it/s]
1|285|Loss: 5.637300491333008:   4%|▍         | 284/6470 [06:25<1:24:06,  1.23it/s]
1|285|Loss: 5.637300491333008:   4%|▍         | 285/6470 [06:25<1:26:35,  1.19it/s]
1|286|Loss: 5.6503987312316895:   4%|▍         | 285/6470 [06:26<1:26:35,  1.19it/s]
1|286|Loss: 5.6503987312316895:   4%|▍         | 286/6470 [06:26<1:23:21,  1.24it/s]
1|287|Loss: 5.498776435852051:   4%|▍         | 286/6470 [06:26<1:23:21,  1.24it/s] 
1|287|Loss: 5.498776435852051:   4%|▍         | 287/6470 [06:27<1:21:44,  1.26it/s]
1|288|Loss: 5.370388507843018:   4%|▍         | 287/6470 [06:27<1:21:44,  1.26it/s]
1|288|Loss: 5.370388507843018:   4%|▍         | 288/6470 [06:27<1:21:14,  1.27it/s]
1|289|Loss: 5.805822849273682:   4%|▍         | 288/6470 [06:28<1:21:14,  1.27it/s]
1|289|Loss: 5.805822849273682:   4%|▍         | 289/6470 [06:28<1:21:34,  1.26it/s]
1|290|Loss: 5.690608501434326:   4%|▍         | 289/6470 [06:29<1:21:34,  1.26it/s]
1|290|Loss: 5.690608501434326:   4%|▍         | 290/6470 [06:29<1:18:48,  1.31it/s]
1|291|Loss: 4.807567596435547:   4%|▍         | 290/6470 [06:29<1:18:48,  1.31it/s]
1|291|Loss: 4.807567596435547:   4%|▍         | 291/6470 [06:29<1:13:49,  1.39it/s]
1|292|Loss: 5.202611923217773:   4%|▍         | 291/6470 [06:30<1:13:49,  1.39it/s]
1|292|Loss: 5.202611923217773:   5%|▍         | 292/6470 [06:30<1:09:53,  1.47it/s]
1|293|Loss: 4.710308074951172:   5%|▍         | 292/6470 [06:30<1:09:53,  1.47it/s]
1|293|Loss: 4.710308074951172:   5%|▍         | 293/6470 [06:31<1:04:15,  1.60it/s]
1|294|Loss: 5.269035816192627:   5%|▍         | 293/6470 [06:31<1:04:15,  1.60it/s]
1|294|Loss: 5.269035816192627:   5%|▍         | 294/6470 [06:31<1:04:02,  1.61it/s]
1|295|Loss: 5.2309250831604:   5%|▍         | 294/6470 [06:32<1:04:02,  1.61it/s]  
1|295|Loss: 5.2309250831604:   5%|▍         | 295/6470 [06:32<1:05:27,  1.57it/s]
1|296|Loss: 5.517501354217529:   5%|▍         | 295/6470 [06:32<1:05:27,  1.57it/s]
1|296|Loss: 5.517501354217529:   5%|▍         | 296/6470 [06:32<1:01:49,  1.66it/s]
1|297|Loss: 4.877748966217041:   5%|▍         | 296/6470 [06:33<1:01:49,  1.66it/s]
1|297|Loss: 4.877748966217041:   5%|▍         | 297/6470 [06:33<58:04,  1.77it/s]  
1|298|Loss: 5.569310665130615:   5%|▍         | 297/6470 [06:33<58:04,  1.77it/s]
1|298|Loss: 5.569310665130615:   5%|▍         | 298/6470 [06:34<1:00:58,  1.69it/s]
1|299|Loss: 5.717474937438965:   5%|▍         | 298/6470 [06:34<1:00:58,  1.69it/s]
1|299|Loss: 5.717474937438965:   5%|▍         | 299/6470 [06:34<1:10:31,  1.46it/s]
1|300|Loss: 5.985641002655029:   5%|▍         | 299/6470 [06:35<1:10:31,  1.46it/s]
1|300|Loss: 5.985641002655029:   5%|▍         | 300/6470 [06:35<1:19:52,  1.29it/s]
1|301|Loss: 5.851487636566162:   5%|▍         | 300/6470 [06:36<1:19:52,  1.29it/s]
1|301|Loss: 5.851487636566162:   5%|▍         | 301/6470 [06:36<1:25:32,  1.20it/s]
1|302|Loss: 5.727380752563477:   5%|▍         | 301/6470 [06:37<1:25:32,  1.20it/s]
1|302|Loss: 5.727380752563477:   5%|▍         | 302/6470 [06:37<1:23:00,  1.24it/s]
1|303|Loss: 5.970484733581543:   5%|▍         | 302/6470 [06:38<1:23:00,  1.24it/s]
1|303|Loss: 5.970484733581543:   5%|▍         | 303/6470 [06:38<1:28:30,  1.16it/s]
1|304|Loss: 5.736577987670898:   5%|▍         | 303/6470 [06:39<1:28:30,  1.16it/s]
1|304|Loss: 5.736577987670898:   5%|▍         | 304/6470 [06:39<1:33:05,  1.10it/s]
1|305|Loss: 5.646793365478516:   5%|▍         | 304/6470 [06:40<1:33:05,  1.10it/s]
1|305|Loss: 5.646793365478516:   5%|▍         | 305/6470 [06:40<1:29:56,  1.14it/s]
1|306|Loss: 5.800378799438477:   5%|▍         | 305/6470 [06:40<1:29:56,  1.14it/s]
1|306|Loss: 5.800378799438477:   5%|▍         | 306/6470 [06:41<1:26:17,  1.19it/s]
1|307|Loss: 5.365697383880615:   5%|▍         | 306/6470 [06:41<1:26:17,  1.19it/s]
1|307|Loss: 5.365697383880615:   5%|▍         | 307/6470 [06:41<1:24:37,  1.21it/s]
1|308|Loss: 5.294834136962891:   5%|▍         | 307/6470 [06:42<1:24:37,  1.21it/s]
1|308|Loss: 5.294834136962891:   5%|▍         | 308/6470 [06:42<1:18:00,  1.32it/s]
1|309|Loss: 5.589354991912842:   5%|▍         | 308/6470 [06:42<1:18:00,  1.32it/s]
1|309|Loss: 5.589354991912842:   5%|▍         | 309/6470 [06:43<1:19:24,  1.29it/s]
1|310|Loss: 5.449570655822754:   5%|▍         | 309/6470 [06:43<1:19:24,  1.29it/s]
1|310|Loss: 5.449570655822754:   5%|▍         | 310/6470 [06:44<1:19:21,  1.29it/s]
1|311|Loss: 5.477928161621094:   5%|▍         | 310/6470 [06:44<1:19:21,  1.29it/s]
1|311|Loss: 5.477928161621094:   5%|▍         | 311/6470 [06:44<1:17:46,  1.32it/s]
1|312|Loss: 5.068150520324707:   5%|▍         | 311/6470 [06:45<1:17:46,  1.32it/s]
1|312|Loss: 5.068150520324707:   5%|▍         | 312/6470 [06:45<1:18:26,  1.31it/s]
1|313|Loss: 5.512414455413818:   5%|▍         | 312/6470 [06:46<1:18:26,  1.31it/s]
1|313|Loss: 5.512414455413818:   5%|▍         | 313/6470 [06:46<1:18:25,  1.31it/s]
1|314|Loss: 5.557101249694824:   5%|▍         | 313/6470 [06:46<1:18:25,  1.31it/s]
1|314|Loss: 5.557101249694824:   5%|▍         | 314/6470 [06:47<1:21:51,  1.25it/s]
1|315|Loss: 5.95080041885376:   5%|▍         | 314/6470 [06:47<1:21:51,  1.25it/s] 
1|315|Loss: 5.95080041885376:   5%|▍         | 315/6470 [06:48<1:22:29,  1.24it/s]
1|316|Loss: 5.242806911468506:   5%|▍         | 315/6470 [06:48<1:22:29,  1.24it/s]
1|316|Loss: 5.242806911468506:   5%|▍         | 316/6470 [06:48<1:18:23,  1.31it/s]
1|317|Loss: 5.518142223358154:   5%|▍         | 316/6470 [06:49<1:18:23,  1.31it/s]
1|317|Loss: 5.518142223358154:   5%|▍         | 317/6470 [06:49<1:13:42,  1.39it/s]
1|318|Loss: 5.706658840179443:   5%|▍         | 317/6470 [06:49<1:13:42,  1.39it/s]
1|318|Loss: 5.706658840179443:   5%|▍         | 318/6470 [06:50<1:17:14,  1.33it/s]
1|319|Loss: 5.165087699890137:   5%|▍         | 318/6470 [06:50<1:17:14,  1.33it/s]
1|319|Loss: 5.165087699890137:   5%|▍         | 319/6470 [06:50<1:14:59,  1.37it/s]
1|320|Loss: 5.1575422286987305:   5%|▍         | 319/6470 [06:51<1:14:59,  1.37it/s]
1|320|Loss: 5.1575422286987305:   5%|▍         | 320/6470 [06:51<1:10:30,  1.45it/s]
1|321|Loss: 4.864731788635254:   5%|▍         | 320/6470 [06:51<1:10:30,  1.45it/s] 
1|321|Loss: 4.864731788635254:   5%|▍         | 321/6470 [06:52<1:09:15,  1.48it/s]
1|322|Loss: 5.270750999450684:   5%|▍         | 321/6470 [06:52<1:09:15,  1.48it/s]
1|322|Loss: 5.270750999450684:   5%|▍         | 322/6470 [06:53<1:16:00,  1.35it/s]
1|323|Loss: 5.6283392906188965:   5%|▍         | 322/6470 [06:53<1:16:00,  1.35it/s]
1|323|Loss: 5.6283392906188965:   5%|▍         | 323/6470 [06:53<1:19:05,  1.30it/s]
1|324|Loss: 4.570036888122559:   5%|▍         | 323/6470 [06:54<1:19:05,  1.30it/s] 
1|324|Loss: 4.570036888122559:   5%|▌         | 324/6470 [06:54<1:09:52,  1.47it/s]
1|325|Loss: 5.696836948394775:   5%|▌         | 324/6470 [06:54<1:09:52,  1.47it/s]
1|325|Loss: 5.696836948394775:   5%|▌         | 325/6470 [06:55<1:12:28,  1.41it/s]
1|326|Loss: 5.599480152130127:   5%|▌         | 325/6470 [06:55<1:12:28,  1.41it/s]
1|326|Loss: 5.599480152130127:   5%|▌         | 326/6470 [06:55<1:13:12,  1.40it/s]
1|327|Loss: 5.815851211547852:   5%|▌         | 326/6470 [06:56<1:13:12,  1.40it/s]
1|327|Loss: 5.815851211547852:   5%|▌         | 327/6470 [06:56<1:17:15,  1.33it/s]
1|328|Loss: 4.692763805389404:   5%|▌         | 327/6470 [06:57<1:17:15,  1.33it/s]
1|328|Loss: 4.692763805389404:   5%|▌         | 328/6470 [06:57<1:15:40,  1.35it/s]
1|329|Loss: 6.061220645904541:   5%|▌         | 328/6470 [06:57<1:15:40,  1.35it/s]
1|329|Loss: 6.061220645904541:   5%|▌         | 329/6470 [06:58<1:23:56,  1.22it/s]
1|330|Loss: 5.562435626983643:   5%|▌         | 329/6470 [06:58<1:23:56,  1.22it/s]
1|330|Loss: 5.562435626983643:   5%|▌         | 330/6470 [06:59<1:26:59,  1.18it/s]
1|331|Loss: 5.089272499084473:   5%|▌         | 330/6470 [06:59<1:26:59,  1.18it/s]
1|331|Loss: 5.089272499084473:   5%|▌         | 331/6470 [07:00<1:21:40,  1.25it/s]
1|332|Loss: 5.306321620941162:   5%|▌         | 331/6470 [07:00<1:21:40,  1.25it/s]
1|332|Loss: 5.306321620941162:   5%|▌         | 332/6470 [07:00<1:20:34,  1.27it/s]
1|333|Loss: 5.289227485656738:   5%|▌         | 332/6470 [07:01<1:20:34,  1.27it/s]
1|333|Loss: 5.289227485656738:   5%|▌         | 333/6470 [07:01<1:18:00,  1.31it/s]
1|334|Loss: 5.50771427154541:   5%|▌         | 333/6470 [07:01<1:18:00,  1.31it/s] 
1|334|Loss: 5.50771427154541:   5%|▌         | 334/6470 [07:02<1:16:05,  1.34it/s]
1|335|Loss: 5.490447998046875:   5%|▌         | 334/6470 [07:02<1:16:05,  1.34it/s]
1|335|Loss: 5.490447998046875:   5%|▌         | 335/6470 [07:02<1:17:14,  1.32it/s]
1|336|Loss: 5.630566120147705:   5%|▌         | 335/6470 [07:03<1:17:14,  1.32it/s]
1|336|Loss: 5.630566120147705:   5%|▌         | 336/6470 [07:03<1:21:23,  1.26it/s]
1|337|Loss: 5.540143966674805:   5%|▌         | 336/6470 [07:04<1:21:23,  1.26it/s]
1|337|Loss: 5.540143966674805:   5%|▌         | 337/6470 [07:04<1:24:45,  1.21it/s]
1|338|Loss: 5.723693370819092:   5%|▌         | 337/6470 [07:05<1:24:45,  1.21it/s]
1|338|Loss: 5.723693370819092:   5%|▌         | 338/6470 [07:05<1:26:18,  1.18it/s]
1|339|Loss: 5.278192520141602:   5%|▌         | 338/6470 [07:06<1:26:18,  1.18it/s]
1|339|Loss: 5.278192520141602:   5%|▌         | 339/6470 [07:06<1:20:15,  1.27it/s]
1|340|Loss: 4.56523323059082:   5%|▌         | 339/6470 [07:06<1:20:15,  1.27it/s] 
1|340|Loss: 4.56523323059082:   5%|▌         | 340/6470 [07:06<1:13:37,  1.39it/s]
1|341|Loss: 4.987912654876709:   5%|▌         | 340/6470 [07:07<1:13:37,  1.39it/s]
1|341|Loss: 4.987912654876709:   5%|▌         | 341/6470 [07:07<1:07:57,  1.50it/s]
1|342|Loss: 5.335724353790283:   5%|▌         | 341/6470 [07:07<1:07:57,  1.50it/s]
1|342|Loss: 5.335724353790283:   5%|▌         | 342/6470 [07:07<1:05:57,  1.55it/s]
1|343|Loss: 5.438939094543457:   5%|▌         | 342/6470 [07:08<1:05:57,  1.55it/s]
1|343|Loss: 5.438939094543457:   5%|▌         | 343/6470 [07:08<1:12:32,  1.41it/s]
1|344|Loss: 4.871743679046631:   5%|▌         | 343/6470 [07:09<1:12:32,  1.41it/s]
1|344|Loss: 4.871743679046631:   5%|▌         | 344/6470 [07:09<1:12:49,  1.40it/s]
1|345|Loss: 4.832030773162842:   5%|▌         | 344/6470 [07:09<1:12:49,  1.40it/s]
1|345|Loss: 4.832030773162842:   5%|▌         | 345/6470 [07:10<1:06:43,  1.53it/s]
1|346|Loss: 5.557636737823486:   5%|▌         | 345/6470 [07:10<1:06:43,  1.53it/s]
1|346|Loss: 5.557636737823486:   5%|▌         | 346/6470 [07:10<1:06:50,  1.53it/s]
1|347|Loss: 5.9095964431762695:   5%|▌         | 346/6470 [07:11<1:06:50,  1.53it/s]
1|347|Loss: 5.9095964431762695:   5%|▌         | 347/6470 [07:11<1:14:35,  1.37it/s]
1|348|Loss: 5.600264072418213:   5%|▌         | 347/6470 [07:12<1:14:35,  1.37it/s] 
1|348|Loss: 5.600264072418213:   5%|▌         | 348/6470 [07:12<1:21:18,  1.25it/s]
1|349|Loss: 5.612281799316406:   5%|▌         | 348/6470 [07:13<1:21:18,  1.25it/s]
1|349|Loss: 5.612281799316406:   5%|▌         | 349/6470 [07:13<1:20:03,  1.27it/s]
1|350|Loss: 5.822343826293945:   5%|▌         | 349/6470 [07:13<1:20:03,  1.27it/s]
1|350|Loss: 5.822343826293945:   5%|▌         | 350/6470 [07:14<1:17:33,  1.32it/s]
1|351|Loss: 5.555905342102051:   5%|▌         | 350/6470 [07:14<1:17:33,  1.32it/s]
1|351|Loss: 5.555905342102051:   5%|▌         | 351/6470 [07:14<1:17:11,  1.32it/s]
1|352|Loss: 5.873291492462158:   5%|▌         | 351/6470 [07:15<1:17:11,  1.32it/s]
1|352|Loss: 5.873291492462158:   5%|▌         | 352/6470 [07:15<1:21:42,  1.25it/s]
1|353|Loss: 5.640453338623047:   5%|▌         | 352/6470 [07:16<1:21:42,  1.25it/s]
1|353|Loss: 5.640453338623047:   5%|▌         | 353/6470 [07:16<1:22:04,  1.24it/s]
1|354|Loss: 5.222876071929932:   5%|▌         | 353/6470 [07:16<1:22:04,  1.24it/s]
1|354|Loss: 5.222876071929932:   5%|▌         | 354/6470 [07:17<1:20:37,  1.26it/s]
1|355|Loss: 5.289863586425781:   5%|▌         | 354/6470 [07:17<1:20:37,  1.26it/s]
1|355|Loss: 5.289863586425781:   5%|▌         | 355/6470 [07:17<1:17:38,  1.31it/s]
1|356|Loss: 5.381932735443115:   5%|▌         | 355/6470 [07:18<1:17:38,  1.31it/s]
1|356|Loss: 5.381932735443115:   6%|▌         | 356/6470 [07:18<1:14:12,  1.37it/s]
1|357|Loss: 5.462170124053955:   6%|▌         | 356/6470 [07:18<1:14:12,  1.37it/s]
1|357|Loss: 5.462170124053955:   6%|▌         | 357/6470 [07:19<1:09:39,  1.46it/s]
1|358|Loss: 5.72653341293335:   6%|▌         | 357/6470 [07:19<1:09:39,  1.46it/s] 
1|358|Loss: 5.72653341293335:   6%|▌         | 358/6470 [07:19<1:12:26,  1.41it/s]
1|359|Loss: 5.531261920928955:   6%|▌         | 358/6470 [07:20<1:12:26,  1.41it/s]
1|359|Loss: 5.531261920928955:   6%|▌         | 359/6470 [07:20<1:13:08,  1.39it/s]
1|360|Loss: 5.47037410736084:   6%|▌         | 359/6470 [07:21<1:13:08,  1.39it/s] 
1|360|Loss: 5.47037410736084:   6%|▌         | 360/6470 [07:21<1:10:06,  1.45it/s]
1|361|Loss: 5.399463653564453:   6%|▌         | 360/6470 [07:21<1:10:06,  1.45it/s]
1|361|Loss: 5.399463653564453:   6%|▌         | 361/6470 [07:22<1:09:54,  1.46it/s]
1|362|Loss: 4.985212802886963:   6%|▌         | 361/6470 [07:22<1:09:54,  1.46it/s]
1|362|Loss: 4.985212802886963:   6%|▌         | 362/6470 [07:22<1:07:03,  1.52it/s]
1|363|Loss: 5.221169471740723:   6%|▌         | 362/6470 [07:22<1:07:03,  1.52it/s]
1|363|Loss: 5.221169471740723:   6%|▌         | 363/6470 [07:23<1:02:57,  1.62it/s]
1|364|Loss: 5.395838737487793:   6%|▌         | 363/6470 [07:23<1:02:57,  1.62it/s]
1|364|Loss: 5.395838737487793:   6%|▌         | 364/6470 [07:23<1:08:27,  1.49it/s]
1|365|Loss: 5.820916175842285:   6%|▌         | 364/6470 [07:24<1:08:27,  1.49it/s]
1|365|Loss: 5.820916175842285:   6%|▌         | 365/6470 [07:24<1:14:07,  1.37it/s]
1|366|Loss: 5.480217456817627:   6%|▌         | 365/6470 [07:25<1:14:07,  1.37it/s]
1|366|Loss: 5.480217456817627:   6%|▌         | 366/6470 [07:25<1:16:52,  1.32it/s]
1|367|Loss: 5.269069194793701:   6%|▌         | 366/6470 [07:26<1:16:52,  1.32it/s]
1|367|Loss: 5.269069194793701:   6%|▌         | 367/6470 [07:26<1:15:15,  1.35it/s]
1|368|Loss: 5.50141716003418:   6%|▌         | 367/6470 [07:26<1:15:15,  1.35it/s] 
1|368|Loss: 5.50141716003418:   6%|▌         | 368/6470 [07:27<1:19:29,  1.28it/s]
1|369|Loss: 5.565812587738037:   6%|▌         | 368/6470 [07:27<1:19:29,  1.28it/s]
1|369|Loss: 5.565812587738037:   6%|▌         | 369/6470 [07:28<1:20:47,  1.26it/s]
1|370|Loss: 5.509766101837158:   6%|▌         | 369/6470 [07:28<1:20:47,  1.26it/s]
1|370|Loss: 5.509766101837158:   6%|▌         | 370/6470 [07:28<1:24:57,  1.20it/s]
1|371|Loss: 5.1227521896362305:   6%|▌         | 370/6470 [07:29<1:24:57,  1.20it/s]
1|371|Loss: 5.1227521896362305:   6%|▌         | 371/6470 [07:29<1:21:16,  1.25it/s]
1|372|Loss: 5.299793720245361:   6%|▌         | 371/6470 [07:30<1:21:16,  1.25it/s] 
1|372|Loss: 5.299793720245361:   6%|▌         | 372/6470 [07:30<1:19:54,  1.27it/s]
1|373|Loss: 5.496092319488525:   6%|▌         | 372/6470 [07:30<1:19:54,  1.27it/s]
1|373|Loss: 5.496092319488525:   6%|▌         | 373/6470 [07:31<1:19:09,  1.28it/s]
1|374|Loss: 5.428674697875977:   6%|▌         | 373/6470 [07:31<1:19:09,  1.28it/s]
1|374|Loss: 5.428674697875977:   6%|▌         | 374/6470 [07:31<1:17:26,  1.31it/s]
1|375|Loss: 5.5427141189575195:   6%|▌         | 374/6470 [07:32<1:17:26,  1.31it/s]
1|375|Loss: 5.5427141189575195:   6%|▌         | 375/6470 [07:32<1:18:15,  1.30it/s]
1|376|Loss: 5.696809768676758:   6%|▌         | 375/6470 [07:33<1:18:15,  1.30it/s] 
1|376|Loss: 5.696809768676758:   6%|▌         | 376/6470 [07:33<1:16:16,  1.33it/s]
1|377|Loss: 5.368154525756836:   6%|▌         | 376/6470 [07:33<1:16:16,  1.33it/s]
1|377|Loss: 5.368154525756836:   6%|▌         | 377/6470 [07:34<1:14:00,  1.37it/s]
1|378|Loss: 5.587746620178223:   6%|▌         | 377/6470 [07:34<1:14:00,  1.37it/s]
1|378|Loss: 5.587746620178223:   6%|▌         | 378/6470 [07:34<1:15:25,  1.35it/s]
1|379|Loss: 5.6809773445129395:   6%|▌         | 378/6470 [07:35<1:15:25,  1.35it/s]
1|379|Loss: 5.6809773445129395:   6%|▌         | 379/6470 [07:35<1:14:08,  1.37it/s]
1|380|Loss: 5.305743217468262:   6%|▌         | 379/6470 [07:35<1:14:08,  1.37it/s] 
1|380|Loss: 5.305743217468262:   6%|▌         | 380/6470 [07:36<1:14:39,  1.36it/s]
1|381|Loss: 5.4774885177612305:   6%|▌         | 380/6470 [07:36<1:14:39,  1.36it/s]
1|381|Loss: 5.4774885177612305:   6%|▌         | 381/6470 [07:37<1:16:34,  1.33it/s]
1|382|Loss: 5.063885688781738:   6%|▌         | 381/6470 [07:37<1:16:34,  1.33it/s] 
1|382|Loss: 5.063885688781738:   6%|▌         | 382/6470 [07:37<1:15:35,  1.34it/s]
1|383|Loss: 4.2354607582092285:   6%|▌         | 382/6470 [07:38<1:15:35,  1.34it/s]
1|383|Loss: 4.2354607582092285:   6%|▌         | 383/6470 [07:38<1:06:53,  1.52it/s]
1|384|Loss: 5.554487228393555:   6%|▌         | 383/6470 [07:38<1:06:53,  1.52it/s] 
1|384|Loss: 5.554487228393555:   6%|▌         | 384/6470 [07:38<1:06:24,  1.53it/s]
1|385|Loss: 5.110655307769775:   6%|▌         | 384/6470 [07:39<1:06:24,  1.53it/s]
1|385|Loss: 5.110655307769775:   6%|▌         | 385/6470 [07:39<1:05:32,  1.55it/s]
1|386|Loss: 5.296600341796875:   6%|▌         | 385/6470 [07:39<1:05:32,  1.55it/s]
1|386|Loss: 5.296600341796875:   6%|▌         | 386/6470 [07:40<1:05:49,  1.54it/s]
1|387|Loss: 5.41945743560791:   6%|▌         | 386/6470 [07:40<1:05:49,  1.54it/s] 
1|387|Loss: 5.41945743560791:   6%|▌         | 387/6470 [07:40<1:08:49,  1.47it/s]
1|388|Loss: 5.5021209716796875:   6%|▌         | 387/6470 [07:41<1:08:49,  1.47it/s]
1|388|Loss: 5.5021209716796875:   6%|▌         | 388/6470 [07:41<1:15:23,  1.34it/s]
1|389|Loss: 5.683637619018555:   6%|▌         | 388/6470 [07:42<1:15:23,  1.34it/s] 
1|389|Loss: 5.683637619018555:   6%|▌         | 389/6470 [07:42<1:19:49,  1.27it/s]
1|390|Loss: 5.265662670135498:   6%|▌         | 389/6470 [07:43<1:19:49,  1.27it/s]
1|390|Loss: 5.265662670135498:   6%|▌         | 390/6470 [07:43<1:16:59,  1.32it/s]
1|391|Loss: 5.907064914703369:   6%|▌         | 390/6470 [07:43<1:16:59,  1.32it/s]
1|391|Loss: 5.907064914703369:   6%|▌         | 391/6470 [07:44<1:17:44,  1.30it/s]
1|392|Loss: 5.087512969970703:   6%|▌         | 391/6470 [07:44<1:17:44,  1.30it/s]
1|392|Loss: 5.087512969970703:   6%|▌         | 392/6470 [07:44<1:12:18,  1.40it/s]
1|393|Loss: 5.929271221160889:   6%|▌         | 392/6470 [07:45<1:12:18,  1.40it/s]
1|393|Loss: 5.929271221160889:   6%|▌         | 393/6470 [07:45<1:12:14,  1.40it/s]
1|394|Loss: 5.791534423828125:   6%|▌         | 393/6470 [07:45<1:12:14,  1.40it/s]
1|394|Loss: 5.791534423828125:   6%|▌         | 394/6470 [07:46<1:16:27,  1.32it/s]
1|395|Loss: 5.227432727813721:   6%|▌         | 394/6470 [07:46<1:16:27,  1.32it/s]
1|395|Loss: 5.227432727813721:   6%|▌         | 395/6470 [07:47<1:12:32,  1.40it/s]
1|396|Loss: 5.4707465171813965:   6%|▌         | 395/6470 [07:47<1:12:32,  1.40it/s]
1|396|Loss: 5.4707465171813965:   6%|▌         | 396/6470 [07:47<1:17:37,  1.30it/s]
1|397|Loss: 4.470016002655029:   6%|▌         | 396/6470 [07:48<1:17:37,  1.30it/s] 
1|397|Loss: 4.470016002655029:   6%|▌         | 397/6470 [07:48<1:12:53,  1.39it/s]
1|398|Loss: 5.43466854095459:   6%|▌         | 397/6470 [07:48<1:12:53,  1.39it/s] 
1|398|Loss: 5.43466854095459:   6%|▌         | 398/6470 [07:49<1:13:15,  1.38it/s]
1|399|Loss: 5.4577226638793945:   6%|▌         | 398/6470 [07:49<1:13:15,  1.38it/s]
1|399|Loss: 5.4577226638793945:   6%|▌         | 399/6470 [07:50<1:17:21,  1.31it/s]
1|400|Loss: 5.497666835784912:   6%|▌         | 399/6470 [07:50<1:17:21,  1.31it/s] 
1|400|Loss: 5.497666835784912:   6%|▌         | 400/6470 [07:50<1:18:15,  1.29it/s]
1|401|Loss: 5.928009986877441:   6%|▌         | 400/6470 [07:51<1:18:15,  1.29it/s]
1|401|Loss: 5.928009986877441:   6%|▌         | 401/6470 [07:51<1:22:26,  1.23it/s]
1|402|Loss: 5.056797504425049:   6%|▌         | 401/6470 [07:52<1:22:26,  1.23it/s]
1|402|Loss: 5.056797504425049:   6%|▌         | 402/6470 [07:52<1:18:30,  1.29it/s]
1|403|Loss: 5.334778785705566:   6%|▌         | 402/6470 [07:52<1:18:30,  1.29it/s]
1|403|Loss: 5.334778785705566:   6%|▌         | 403/6470 [07:53<1:11:42,  1.41it/s]
1|404|Loss: 5.748877048492432:   6%|▌         | 403/6470 [07:53<1:11:42,  1.41it/s]
1|404|Loss: 5.748877048492432:   6%|▌         | 404/6470 [07:53<1:13:24,  1.38it/s]
1|405|Loss: 5.272777080535889:   6%|▌         | 404/6470 [07:54<1:13:24,  1.38it/s]
1|405|Loss: 5.272777080535889:   6%|▋         | 405/6470 [07:54<1:11:05,  1.42it/s]
1|406|Loss: 5.757727146148682:   6%|▋         | 405/6470 [07:54<1:11:05,  1.42it/s]
1|406|Loss: 5.757727146148682:   6%|▋         | 406/6470 [07:55<1:15:21,  1.34it/s]
1|407|Loss: 5.845317363739014:   6%|▋         | 406/6470 [07:55<1:15:21,  1.34it/s]
1|407|Loss: 5.845317363739014:   6%|▋         | 407/6470 [07:56<1:19:28,  1.27it/s]
1|408|Loss: 5.425384521484375:   6%|▋         | 407/6470 [07:56<1:19:28,  1.27it/s]
1|408|Loss: 5.425384521484375:   6%|▋         | 408/6470 [07:56<1:19:16,  1.27it/s]
1|409|Loss: 5.579627990722656:   6%|▋         | 408/6470 [07:57<1:19:16,  1.27it/s]
1|409|Loss: 5.579627990722656:   6%|▋         | 409/6470 [07:57<1:18:44,  1.28it/s]
1|410|Loss: 5.3371124267578125:   6%|▋         | 409/6470 [07:58<1:18:44,  1.28it/s]
1|410|Loss: 5.3371124267578125:   6%|▋         | 410/6470 [07:58<1:16:08,  1.33it/s]
1|411|Loss: 4.852956771850586:   6%|▋         | 410/6470 [07:58<1:16:08,  1.33it/s] 
1|411|Loss: 4.852956771850586:   6%|▋         | 411/6470 [07:59<1:10:33,  1.43it/s]
1|412|Loss: 5.476583480834961:   6%|▋         | 411/6470 [07:59<1:10:33,  1.43it/s]
1|412|Loss: 5.476583480834961:   6%|▋         | 412/6470 [07:59<1:09:02,  1.46it/s]
1|413|Loss: 5.304172515869141:   6%|▋         | 412/6470 [08:00<1:09:02,  1.46it/s]
1|413|Loss: 5.304172515869141:   6%|▋         | 413/6470 [08:00<1:08:59,  1.46it/s]
1|414|Loss: 5.158614635467529:   6%|▋         | 413/6470 [08:00<1:08:59,  1.46it/s]
1|414|Loss: 5.158614635467529:   6%|▋         | 414/6470 [08:01<1:08:42,  1.47it/s]
1|415|Loss: 5.701577186584473:   6%|▋         | 414/6470 [08:01<1:08:42,  1.47it/s]
1|415|Loss: 5.701577186584473:   6%|▋         | 415/6470 [08:01<1:10:54,  1.42it/s]
1|416|Loss: 5.609804153442383:   6%|▋         | 415/6470 [08:02<1:10:54,  1.42it/s]
1|416|Loss: 5.609804153442383:   6%|▋         | 416/6470 [08:02<1:13:52,  1.37it/s]
1|417|Loss: 4.780465126037598:   6%|▋         | 416/6470 [08:02<1:13:52,  1.37it/s]
1|417|Loss: 4.780465126037598:   6%|▋         | 417/6470 [08:03<1:11:50,  1.40it/s]
1|418|Loss: 5.271884918212891:   6%|▋         | 417/6470 [08:03<1:11:50,  1.40it/s]
1|418|Loss: 5.271884918212891:   6%|▋         | 418/6470 [08:04<1:16:45,  1.31it/s]
1|419|Loss: 5.225771427154541:   6%|▋         | 418/6470 [08:04<1:16:45,  1.31it/s]
1|419|Loss: 5.225771427154541:   6%|▋         | 419/6470 [08:04<1:17:59,  1.29it/s]
1|420|Loss: 5.824793815612793:   6%|▋         | 419/6470 [08:05<1:17:59,  1.29it/s]
1|420|Loss: 5.824793815612793:   6%|▋         | 420/6470 [08:05<1:21:18,  1.24it/s]
1|421|Loss: 5.808602809906006:   6%|▋         | 420/6470 [08:06<1:21:18,  1.24it/s]
1|421|Loss: 5.808602809906006:   7%|▋         | 421/6470 [08:06<1:24:43,  1.19it/s]
1|422|Loss: 4.309171676635742:   7%|▋         | 421/6470 [08:07<1:24:43,  1.19it/s]
1|422|Loss: 4.309171676635742:   7%|▋         | 422/6470 [08:07<1:14:48,  1.35it/s]
1|423|Loss: 4.864187240600586:   7%|▋         | 422/6470 [08:07<1:14:48,  1.35it/s]
1|423|Loss: 4.864187240600586:   7%|▋         | 423/6470 [08:07<1:09:17,  1.45it/s]
1|424|Loss: 5.528370380401611:   7%|▋         | 423/6470 [08:08<1:09:17,  1.45it/s]
1|424|Loss: 5.528370380401611:   7%|▋         | 424/6470 [08:08<1:11:33,  1.41it/s]
1|425|Loss: 4.99350118637085:   7%|▋         | 424/6470 [08:08<1:11:33,  1.41it/s] 
1|425|Loss: 4.99350118637085:   7%|▋         | 425/6470 [08:09<1:11:43,  1.40it/s]
1|426|Loss: 5.53336238861084:   7%|▋         | 425/6470 [08:09<1:11:43,  1.40it/s]
1|426|Loss: 5.53336238861084:   7%|▋         | 426/6470 [08:09<1:11:32,  1.41it/s]
1|427|Loss: 5.428008079528809:   7%|▋         | 426/6470 [08:10<1:11:32,  1.41it/s]
1|427|Loss: 5.428008079528809:   7%|▋         | 427/6470 [08:10<1:12:48,  1.38it/s]
1|428|Loss: 5.444633960723877:   7%|▋         | 427/6470 [08:11<1:12:48,  1.38it/s]
1|428|Loss: 5.444633960723877:   7%|▋         | 428/6470 [08:11<1:14:41,  1.35it/s]
1|429|Loss: 4.815274715423584:   7%|▋         | 428/6470 [08:11<1:14:41,  1.35it/s]
1|429|Loss: 4.815274715423584:   7%|▋         | 429/6470 [08:12<1:11:20,  1.41it/s]
1|430|Loss: 5.28726053237915:   7%|▋         | 429/6470 [08:12<1:11:20,  1.41it/s] 
1|430|Loss: 5.28726053237915:   7%|▋         | 430/6470 [08:12<1:08:12,  1.48it/s]
1|431|Loss: 5.666355609893799:   7%|▋         | 430/6470 [08:13<1:08:12,  1.48it/s]
1|431|Loss: 5.666355609893799:   7%|▋         | 431/6470 [08:13<1:08:47,  1.46it/s]
1|432|Loss: 5.75726842880249:   7%|▋         | 431/6470 [08:13<1:08:47,  1.46it/s] 
1|432|Loss: 5.75726842880249:   7%|▋         | 432/6470 [08:14<1:15:56,  1.33it/s]
1|433|Loss: 4.9781389236450195:   7%|▋         | 432/6470 [08:14<1:15:56,  1.33it/s]
1|433|Loss: 4.9781389236450195:   7%|▋         | 433/6470 [08:15<1:18:50,  1.28it/s]
1|434|Loss: 4.914636611938477:   7%|▋         | 433/6470 [08:15<1:18:50,  1.28it/s] 
1|434|Loss: 4.914636611938477:   7%|▋         | 434/6470 [08:15<1:15:22,  1.33it/s]
1|435|Loss: 5.765984058380127:   7%|▋         | 434/6470 [08:16<1:15:22,  1.33it/s]
1|435|Loss: 5.765984058380127:   7%|▋         | 435/6470 [08:16<1:17:59,  1.29it/s]
1|436|Loss: 4.977652549743652:   7%|▋         | 435/6470 [08:17<1:17:59,  1.29it/s]
1|436|Loss: 4.977652549743652:   7%|▋         | 436/6470 [08:17<1:16:51,  1.31it/s]
1|437|Loss: 5.387964725494385:   7%|▋         | 436/6470 [08:17<1:16:51,  1.31it/s]
1|437|Loss: 5.387964725494385:   7%|▋         | 437/6470 [08:18<1:13:42,  1.36it/s]
1|438|Loss: 5.433880805969238:   7%|▋         | 437/6470 [08:18<1:13:42,  1.36it/s]
1|438|Loss: 5.433880805969238:   7%|▋         | 438/6470 [08:18<1:12:27,  1.39it/s]
1|439|Loss: 5.612411022186279:   7%|▋         | 438/6470 [08:19<1:12:27,  1.39it/s]
1|439|Loss: 5.612411022186279:   7%|▋         | 439/6470 [08:19<1:16:52,  1.31it/s]
1|440|Loss: 5.931272983551025:   7%|▋         | 439/6470 [08:20<1:16:52,  1.31it/s]
1|440|Loss: 5.931272983551025:   7%|▋         | 440/6470 [08:20<1:21:41,  1.23it/s]
1|441|Loss: 5.400498867034912:   7%|▋         | 440/6470 [08:21<1:21:41,  1.23it/s]
1|441|Loss: 5.400498867034912:   7%|▋         | 441/6470 [08:21<1:22:09,  1.22it/s]
1|442|Loss: 5.344727516174316:   7%|▋         | 441/6470 [08:21<1:22:09,  1.22it/s]
1|442|Loss: 5.344727516174316:   7%|▋         | 442/6470 [08:22<1:23:17,  1.21it/s]
1|443|Loss: 5.866415023803711:   7%|▋         | 442/6470 [08:22<1:23:17,  1.21it/s]
1|443|Loss: 5.866415023803711:   7%|▋         | 443/6470 [08:23<1:25:45,  1.17it/s]
1|444|Loss: 5.482367515563965:   7%|▋         | 443/6470 [08:23<1:25:45,  1.17it/s]
1|444|Loss: 5.482367515563965:   7%|▋         | 444/6470 [08:24<1:29:29,  1.12it/s]
1|445|Loss: 5.779191970825195:   7%|▋         | 444/6470 [08:24<1:29:29,  1.12it/s]
1|445|Loss: 5.779191970825195:   7%|▋         | 445/6470 [08:25<1:28:35,  1.13it/s]
1|446|Loss: 5.933222770690918:   7%|▋         | 445/6470 [08:25<1:28:35,  1.13it/s]
1|446|Loss: 5.933222770690918:   7%|▋         | 446/6470 [08:25<1:24:24,  1.19it/s]
1|447|Loss: 5.423274517059326:   7%|▋         | 446/6470 [08:26<1:24:24,  1.19it/s]
1|447|Loss: 5.423274517059326:   7%|▋         | 447/6470 [08:26<1:17:46,  1.29it/s]
1|448|Loss: 5.592013359069824:   7%|▋         | 447/6470 [08:26<1:17:46,  1.29it/s]
1|448|Loss: 5.592013359069824:   7%|▋         | 448/6470 [08:27<1:16:29,  1.31it/s]
1|449|Loss: 5.171943187713623:   7%|▋         | 448/6470 [08:27<1:16:29,  1.31it/s]
1|449|Loss: 5.171943187713623:   7%|▋         | 449/6470 [08:27<1:13:24,  1.37it/s]
1|450|Loss: 5.546588897705078:   7%|▋         | 449/6470 [08:28<1:13:24,  1.37it/s]
1|450|Loss: 5.546588897705078:   7%|▋         | 450/6470 [08:28<1:14:38,  1.34it/s]
1|451|Loss: 3.9185502529144287:   7%|▋         | 450/6470 [08:28<1:14:38,  1.34it/s]
1|451|Loss: 3.9185502529144287:   7%|▋         | 451/6470 [08:29<1:09:34,  1.44it/s]
1|452|Loss: 4.572867393493652:   7%|▋         | 451/6470 [08:29<1:09:34,  1.44it/s] 
1|452|Loss: 4.572867393493652:   7%|▋         | 452/6470 [08:29<1:01:07,  1.64it/s]
1|453|Loss: 5.719010829925537:   7%|▋         | 452/6470 [08:29<1:01:07,  1.64it/s]
1|453|Loss: 5.719010829925537:   7%|▋         | 453/6470 [08:30<1:00:16,  1.66it/s]
1|454|Loss: 5.607015609741211:   7%|▋         | 453/6470 [08:30<1:00:16,  1.66it/s]
1|454|Loss: 5.607015609741211:   7%|▋         | 454/6470 [08:30<1:03:19,  1.58it/s]
1|455|Loss: 5.860990524291992:   7%|▋         | 454/6470 [08:31<1:03:19,  1.58it/s]
1|455|Loss: 5.860990524291992:   7%|▋         | 455/6470 [08:31<1:12:12,  1.39it/s]
1|456|Loss: 5.220895767211914:   7%|▋         | 455/6470 [08:32<1:12:12,  1.39it/s]
1|456|Loss: 5.220895767211914:   7%|▋         | 456/6470 [08:32<1:13:45,  1.36it/s]
1|457|Loss: 5.666329860687256:   7%|▋         | 456/6470 [08:32<1:13:45,  1.36it/s]
1|457|Loss: 5.666329860687256:   7%|▋         | 457/6470 [08:33<1:13:04,  1.37it/s]
1|458|Loss: 5.157651901245117:   7%|▋         | 457/6470 [08:33<1:13:04,  1.37it/s]
1|458|Loss: 5.157651901245117:   7%|▋         | 458/6470 [08:34<1:14:46,  1.34it/s]
1|459|Loss: 5.341708660125732:   7%|▋         | 458/6470 [08:34<1:14:46,  1.34it/s]
1|459|Loss: 5.341708660125732:   7%|▋         | 459/6470 [08:34<1:13:43,  1.36it/s]
1|460|Loss: 5.954046726226807:   7%|▋         | 459/6470 [08:35<1:13:43,  1.36it/s]
1|460|Loss: 5.954046726226807:   7%|▋         | 460/6470 [08:35<1:18:11,  1.28it/s]
1|461|Loss: 5.426613807678223:   7%|▋         | 460/6470 [08:36<1:18:11,  1.28it/s]
1|461|Loss: 5.426613807678223:   7%|▋         | 461/6470 [08:36<1:18:01,  1.28it/s]
1|462|Loss: 5.463123321533203:   7%|▋         | 461/6470 [08:36<1:18:01,  1.28it/s]
1|462|Loss: 5.463123321533203:   7%|▋         | 462/6470 [08:37<1:18:01,  1.28it/s]
1|463|Loss: 5.599711894989014:   7%|▋         | 462/6470 [08:37<1:18:01,  1.28it/s]
1|463|Loss: 5.599711894989014:   7%|▋         | 463/6470 [08:37<1:15:44,  1.32it/s]
1|464|Loss: 5.725013256072998:   7%|▋         | 463/6470 [08:38<1:15:44,  1.32it/s]
1|464|Loss: 5.725013256072998:   7%|▋         | 464/6470 [08:38<1:13:11,  1.37it/s]
1|465|Loss: 5.901429653167725:   7%|▋         | 464/6470 [08:39<1:13:11,  1.37it/s]
1|465|Loss: 5.901429653167725:   7%|▋         | 465/6470 [08:39<1:16:22,  1.31it/s]
1|466|Loss: 5.459907054901123:   7%|▋         | 465/6470 [08:39<1:16:22,  1.31it/s]
1|466|Loss: 5.459907054901123:   7%|▋         | 466/6470 [08:40<1:18:37,  1.27it/s]
1|467|Loss: 5.415379524230957:   7%|▋         | 466/6470 [08:40<1:18:37,  1.27it/s]
1|467|Loss: 5.415379524230957:   7%|▋         | 467/6470 [08:41<1:19:48,  1.25it/s]
1|468|Loss: 5.220579147338867:   7%|▋         | 467/6470 [08:41<1:19:48,  1.25it/s]
1|468|Loss: 5.220579147338867:   7%|▋         | 468/6470 [08:41<1:14:06,  1.35it/s]
1|469|Loss: 5.376245021820068:   7%|▋         | 468/6470 [08:42<1:14:06,  1.35it/s]
1|469|Loss: 5.376245021820068:   7%|▋         | 469/6470 [08:42<1:09:46,  1.43it/s]
1|470|Loss: 5.045848369598389:   7%|▋         | 469/6470 [08:42<1:09:46,  1.43it/s]
1|470|Loss: 5.045848369598389:   7%|▋         | 470/6470 [08:42<1:07:28,  1.48it/s]
1|471|Loss: 5.297027587890625:   7%|▋         | 470/6470 [08:43<1:07:28,  1.48it/s]
1|471|Loss: 5.297027587890625:   7%|▋         | 471/6470 [08:43<1:09:18,  1.44it/s]
1|472|Loss: 5.695352077484131:   7%|▋         | 471/6470 [08:44<1:09:18,  1.44it/s]
1|472|Loss: 5.695352077484131:   7%|▋         | 472/6470 [08:44<1:14:24,  1.34it/s]
1|473|Loss: 5.412114143371582:   7%|▋         | 472/6470 [08:44<1:14:24,  1.34it/s]
1|473|Loss: 5.412114143371582:   7%|▋         | 473/6470 [08:45<1:12:23,  1.38it/s]
1|474|Loss: 5.847270965576172:   7%|▋         | 473/6470 [08:45<1:12:23,  1.38it/s]
1|474|Loss: 5.847270965576172:   7%|▋         | 474/6470 [08:45<1:12:44,  1.37it/s]
1|475|Loss: 5.1081223487854:   7%|▋         | 474/6470 [08:46<1:12:44,  1.37it/s]  
1|475|Loss: 5.1081223487854:   7%|▋         | 475/6470 [08:46<1:13:38,  1.36it/s]
1|476|Loss: 5.704952716827393:   7%|▋         | 475/6470 [08:47<1:13:38,  1.36it/s]
1|476|Loss: 5.704952716827393:   7%|▋         | 476/6470 [08:47<1:13:00,  1.37it/s]
1|477|Loss: 5.364786624908447:   7%|▋         | 476/6470 [08:47<1:13:00,  1.37it/s]
1|477|Loss: 5.364786624908447:   7%|▋         | 477/6470 [08:48<1:14:49,  1.33it/s]
1|478|Loss: 5.798461437225342:   7%|▋         | 477/6470 [08:48<1:14:49,  1.33it/s]
1|478|Loss: 5.798461437225342:   7%|▋         | 478/6470 [08:49<1:21:46,  1.22it/s]
1|479|Loss: 5.637391090393066:   7%|▋         | 478/6470 [08:49<1:21:46,  1.22it/s]
1|479|Loss: 5.637391090393066:   7%|▋         | 479/6470 [08:49<1:21:48,  1.22it/s]
1|480|Loss: 5.24155330657959:   7%|▋         | 479/6470 [08:50<1:21:48,  1.22it/s] 
1|480|Loss: 5.24155330657959:   7%|▋         | 480/6470 [08:50<1:16:45,  1.30it/s]
1|481|Loss: 5.380684852600098:   7%|▋         | 480/6470 [08:50<1:16:45,  1.30it/s]
1|481|Loss: 5.380684852600098:   7%|▋         | 481/6470 [08:51<1:13:16,  1.36it/s]
1|482|Loss: 5.461268901824951:   7%|▋         | 481/6470 [08:51<1:13:16,  1.36it/s]
1|482|Loss: 5.461268901824951:   7%|▋         | 482/6470 [08:52<1:13:47,  1.35it/s]
1|483|Loss: 5.2095232009887695:   7%|▋         | 482/6470 [08:52<1:13:47,  1.35it/s]
1|483|Loss: 5.2095232009887695:   7%|▋         | 483/6470 [08:52<1:13:02,  1.37it/s]
1|484|Loss: 5.644862651824951:   7%|▋         | 483/6470 [08:53<1:13:02,  1.37it/s] 
1|484|Loss: 5.644862651824951:   7%|▋         | 484/6470 [08:53<1:15:07,  1.33it/s]
1|485|Loss: 5.639408111572266:   7%|▋         | 484/6470 [08:53<1:15:07,  1.33it/s]
1|485|Loss: 5.639408111572266:   7%|▋         | 485/6470 [08:54<1:15:56,  1.31it/s]
1|486|Loss: 5.518553733825684:   7%|▋         | 485/6470 [08:54<1:15:56,  1.31it/s]
1|486|Loss: 5.518553733825684:   8%|▊         | 486/6470 [08:55<1:16:06,  1.31it/s]
1|487|Loss: 4.509305000305176:   8%|▊         | 486/6470 [08:55<1:16:06,  1.31it/s]
1|487|Loss: 4.509305000305176:   8%|▊         | 487/6470 [08:55<1:10:42,  1.41it/s]
1|488|Loss: 5.371571063995361:   8%|▊         | 487/6470 [08:56<1:10:42,  1.41it/s]
1|488|Loss: 5.371571063995361:   8%|▊         | 488/6470 [08:56<1:09:03,  1.44it/s]
1|489|Loss: 5.886227607727051:   8%|▊         | 488/6470 [08:56<1:09:03,  1.44it/s]
1|489|Loss: 5.886227607727051:   8%|▊         | 489/6470 [08:57<1:16:49,  1.30it/s]
1|490|Loss: 5.538754463195801:   8%|▊         | 489/6470 [08:57<1:16:49,  1.30it/s]
1|490|Loss: 5.538754463195801:   8%|▊         | 490/6470 [08:58<1:17:06,  1.29it/s]
1|491|Loss: 5.435714244842529:   8%|▊         | 490/6470 [08:58<1:17:06,  1.29it/s]
1|491|Loss: 5.435714244842529:   8%|▊         | 491/6470 [08:58<1:15:14,  1.32it/s]
1|492|Loss: 6.117063522338867:   8%|▊         | 491/6470 [08:59<1:15:14,  1.32it/s]
1|492|Loss: 6.117063522338867:   8%|▊         | 492/6470 [08:59<1:20:23,  1.24it/s]
1|493|Loss: 4.620728969573975:   8%|▊         | 492/6470 [09:00<1:20:23,  1.24it/s]
1|493|Loss: 4.620728969573975:   8%|▊         | 493/6470 [09:00<1:21:31,  1.22it/s]
1|494|Loss: 5.304385185241699:   8%|▊         | 493/6470 [09:00<1:21:31,  1.22it/s]
1|494|Loss: 5.304385185241699:   8%|▊         | 494/6470 [09:01<1:18:14,  1.27it/s]
1|495|Loss: 4.870848655700684:   8%|▊         | 494/6470 [09:01<1:18:14,  1.27it/s]
1|495|Loss: 4.870848655700684:   8%|▊         | 495/6470 [09:01<1:13:33,  1.35it/s]
1|496|Loss: 4.996723175048828:   8%|▊         | 495/6470 [09:02<1:13:33,  1.35it/s]
1|496|Loss: 4.996723175048828:   8%|▊         | 496/6470 [09:02<1:16:06,  1.31it/s]
1|497|Loss: 5.842445373535156:   8%|▊         | 496/6470 [09:03<1:16:06,  1.31it/s]
1|497|Loss: 5.842445373535156:   8%|▊         | 497/6470 [09:03<1:20:22,  1.24it/s]
1|498|Loss: 5.144033908843994:   8%|▊         | 497/6470 [09:04<1:20:22,  1.24it/s]
1|498|Loss: 5.144033908843994:   8%|▊         | 498/6470 [09:04<1:15:52,  1.31it/s]
1|499|Loss: 5.447689056396484:   8%|▊         | 498/6470 [09:04<1:15:52,  1.31it/s]
1|499|Loss: 5.447689056396484:   8%|▊         | 499/6470 [09:05<1:18:24,  1.27it/s]
1|500|Loss: 4.9210100173950195:   8%|▊         | 499/6470 [09:05<1:18:24,  1.27it/s]
1|500|Loss: 4.9210100173950195:   8%|▊         | 500/6470 [09:05<1:11:40,  1.39it/s]
1|501|Loss: 4.851637840270996:   8%|▊         | 500/6470 [09:06<1:11:40,  1.39it/s] 
1|501|Loss: 4.851637840270996:   8%|▊         | 501/6470 [09:06<1:09:16,  1.44it/s]
1|502|Loss: 5.618325710296631:   8%|▊         | 501/6470 [09:06<1:09:16,  1.44it/s]
1|502|Loss: 5.618325710296631:   8%|▊         | 502/6470 [09:07<1:09:55,  1.42it/s]
1|503|Loss: 4.833993911743164:   8%|▊         | 502/6470 [09:07<1:09:55,  1.42it/s]
1|503|Loss: 4.833993911743164:   8%|▊         | 503/6470 [09:07<1:06:11,  1.50it/s]
1|504|Loss: 5.613035202026367:   8%|▊         | 503/6470 [09:08<1:06:11,  1.50it/s]
1|504|Loss: 5.613035202026367:   8%|▊         | 504/6470 [09:08<1:11:23,  1.39it/s]
1|505|Loss: 5.364921569824219:   8%|▊         | 504/6470 [09:08<1:11:23,  1.39it/s]
1|505|Loss: 5.364921569824219:   8%|▊         | 505/6470 [09:09<1:13:20,  1.36it/s]
1|506|Loss: 5.2153825759887695:   8%|▊         | 505/6470 [09:09<1:13:20,  1.36it/s]
1|506|Loss: 5.2153825759887695:   8%|▊         | 506/6470 [09:09<1:11:00,  1.40it/s]
1|507|Loss: 5.705116271972656:   8%|▊         | 506/6470 [09:10<1:11:00,  1.40it/s] 
1|507|Loss: 5.705116271972656:   8%|▊         | 507/6470 [09:10<1:17:49,  1.28it/s]
1|508|Loss: 3.985438823699951:   8%|▊         | 507/6470 [09:11<1:17:49,  1.28it/s]
1|508|Loss: 3.985438823699951:   8%|▊         | 508/6470 [09:11<1:09:40,  1.43it/s]
1|509|Loss: 5.457404613494873:   8%|▊         | 508/6470 [09:11<1:09:40,  1.43it/s]
1|509|Loss: 5.457404613494873:   8%|▊         | 509/6470 [09:12<1:09:45,  1.42it/s]
1|510|Loss: 5.2147369384765625:   8%|▊         | 509/6470 [09:12<1:09:45,  1.42it/s]
1|510|Loss: 5.2147369384765625:   8%|▊         | 510/6470 [09:13<1:22:15,  1.21it/s]
1|511|Loss: 4.880239009857178:   8%|▊         | 510/6470 [09:13<1:22:15,  1.21it/s] 
1|511|Loss: 4.880239009857178:   8%|▊         | 511/6470 [09:13<1:21:20,  1.22it/s]
1|512|Loss: 5.564023494720459:   8%|▊         | 511/6470 [09:14<1:21:20,  1.22it/s]
1|512|Loss: 5.564023494720459:   8%|▊         | 512/6470 [09:14<1:26:01,  1.15it/s]
1|513|Loss: 5.417738914489746:   8%|▊         | 512/6470 [09:15<1:26:01,  1.15it/s]
1|513|Loss: 5.417738914489746:   8%|▊         | 513/6470 [09:15<1:26:50,  1.14it/s]
1|514|Loss: 5.083397388458252:   8%|▊         | 513/6470 [09:16<1:26:50,  1.14it/s]
1|514|Loss: 5.083397388458252:   8%|▊         | 514/6470 [09:16<1:22:00,  1.21it/s]
1|515|Loss: 5.173879623413086:   8%|▊         | 514/6470 [09:16<1:22:00,  1.21it/s]
1|515|Loss: 5.173879623413086:   8%|▊         | 515/6470 [09:17<1:17:34,  1.28it/s]
1|516|Loss: 5.852198600769043:   8%|▊         | 515/6470 [09:17<1:17:34,  1.28it/s]
1|516|Loss: 5.852198600769043:   8%|▊         | 516/6470 [09:18<1:17:46,  1.28it/s]
1|517|Loss: 5.677846431732178:   8%|▊         | 516/6470 [09:18<1:17:46,  1.28it/s]
1|517|Loss: 5.677846431732178:   8%|▊         | 517/6470 [09:18<1:21:59,  1.21it/s]
1|518|Loss: 5.525228500366211:   8%|▊         | 517/6470 [09:19<1:21:59,  1.21it/s]
1|518|Loss: 5.525228500366211:   8%|▊         | 518/6470 [09:19<1:24:51,  1.17it/s]
1|519|Loss: 5.187728404998779:   8%|▊         | 518/6470 [09:20<1:24:51,  1.17it/s]
1|519|Loss: 5.187728404998779:   8%|▊         | 519/6470 [09:20<1:22:53,  1.20it/s]
1|520|Loss: 5.542535305023193:   8%|▊         | 519/6470 [09:21<1:22:53,  1.20it/s]
1|520|Loss: 5.542535305023193:   8%|▊         | 520/6470 [09:21<1:23:16,  1.19it/s]
1|521|Loss: 5.050800323486328:   8%|▊         | 520/6470 [09:21<1:23:16,  1.19it/s]
1|521|Loss: 5.050800323486328:   8%|▊         | 521/6470 [09:22<1:18:54,  1.26it/s]
1|522|Loss: 5.740421295166016:   8%|▊         | 521/6470 [09:22<1:18:54,  1.26it/s]
1|522|Loss: 5.740421295166016:   8%|▊         | 522/6470 [09:23<1:20:55,  1.23it/s]
1|523|Loss: 5.5270843505859375:   8%|▊         | 522/6470 [09:23<1:20:55,  1.23it/s]
1|523|Loss: 5.5270843505859375:   8%|▊         | 523/6470 [09:23<1:18:52,  1.26it/s]
1|524|Loss: 5.1160993576049805:   8%|▊         | 523/6470 [09:24<1:18:52,  1.26it/s]
1|524|Loss: 5.1160993576049805:   8%|▊         | 524/6470 [09:24<1:15:25,  1.31it/s]
1|525|Loss: 5.539821147918701:   8%|▊         | 524/6470 [09:24<1:15:25,  1.31it/s] 
1|525|Loss: 5.539821147918701:   8%|▊         | 525/6470 [09:25<1:15:52,  1.31it/s]
1|526|Loss: 5.318568706512451:   8%|▊         | 525/6470 [09:25<1:15:52,  1.31it/s]
1|526|Loss: 5.318568706512451:   8%|▊         | 526/6470 [09:26<1:21:14,  1.22it/s]
1|527|Loss: 5.698144435882568:   8%|▊         | 526/6470 [09:26<1:21:14,  1.22it/s]
1|527|Loss: 5.698144435882568:   8%|▊         | 527/6470 [09:27<1:19:48,  1.24it/s]
1|528|Loss: 5.994936943054199:   8%|▊         | 527/6470 [09:27<1:19:48,  1.24it/s]
1|528|Loss: 5.994936943054199:   8%|▊         | 528/6470 [09:27<1:19:31,  1.25it/s]
1|529|Loss: 5.729415416717529:   8%|▊         | 528/6470 [09:28<1:19:31,  1.25it/s]
1|529|Loss: 5.729415416717529:   8%|▊         | 529/6470 [09:28<1:19:37,  1.24it/s]
1|530|Loss: 5.913659572601318:   8%|▊         | 529/6470 [09:29<1:19:37,  1.24it/s]
1|530|Loss: 5.913659572601318:   8%|▊         | 530/6470 [09:29<1:20:03,  1.24it/s]
1|531|Loss: 5.865533351898193:   8%|▊         | 530/6470 [09:29<1:20:03,  1.24it/s]
1|531|Loss: 5.865533351898193:   8%|▊         | 531/6470 [09:30<1:21:45,  1.21it/s]
1|532|Loss: 5.319387435913086:   8%|▊         | 531/6470 [09:30<1:21:45,  1.21it/s]
1|532|Loss: 5.319387435913086:   8%|▊         | 532/6470 [09:31<1:19:17,  1.25it/s]
1|533|Loss: 5.57859992980957:   8%|▊         | 532/6470 [09:31<1:19:17,  1.25it/s] 
1|533|Loss: 5.57859992980957:   8%|▊         | 533/6470 [09:31<1:19:10,  1.25it/s]
1|534|Loss: 5.7944560050964355:   8%|▊         | 533/6470 [09:32<1:19:10,  1.25it/s]
1|534|Loss: 5.7944560050964355:   8%|▊         | 534/6470 [09:32<1:25:43,  1.15it/s]
1|535|Loss: 5.180392265319824:   8%|▊         | 534/6470 [09:33<1:25:43,  1.15it/s] 
1|535|Loss: 5.180392265319824:   8%|▊         | 535/6470 [09:33<1:22:59,  1.19it/s]
1|536|Loss: 5.184136390686035:   8%|▊         | 535/6470 [09:34<1:22:59,  1.19it/s]
1|536|Loss: 5.184136390686035:   8%|▊         | 536/6470 [09:34<1:23:50,  1.18it/s]
1|537|Loss: 4.973743915557861:   8%|▊         | 536/6470 [09:34<1:23:50,  1.18it/s]
1|537|Loss: 4.973743915557861:   8%|▊         | 537/6470 [09:35<1:16:07,  1.30it/s]
1|538|Loss: 5.079006671905518:   8%|▊         | 537/6470 [09:35<1:16:07,  1.30it/s]
1|538|Loss: 5.079006671905518:   8%|▊         | 538/6470 [09:35<1:07:46,  1.46it/s]
1|539|Loss: 5.178671360015869:   8%|▊         | 538/6470 [09:35<1:07:46,  1.46it/s]
1|539|Loss: 5.178671360015869:   8%|▊         | 539/6470 [09:36<1:06:33,  1.49it/s]
1|540|Loss: 5.671057224273682:   8%|▊         | 539/6470 [09:36<1:06:33,  1.49it/s]
1|540|Loss: 5.671057224273682:   8%|▊         | 540/6470 [09:37<1:13:23,  1.35it/s]
1|541|Loss: 5.095906734466553:   8%|▊         | 540/6470 [09:37<1:13:23,  1.35it/s]
1|541|Loss: 5.095906734466553:   8%|▊         | 541/6470 [09:37<1:12:40,  1.36it/s]
1|542|Loss: 5.214864730834961:   8%|▊         | 541/6470 [09:38<1:12:40,  1.36it/s]
1|542|Loss: 5.214864730834961:   8%|▊         | 542/6470 [09:38<1:15:17,  1.31it/s]
1|543|Loss: 4.496190071105957:   8%|▊         | 542/6470 [09:39<1:15:17,  1.31it/s]
1|543|Loss: 4.496190071105957:   8%|▊         | 543/6470 [09:39<1:15:31,  1.31it/s]
1|544|Loss: 5.069795608520508:   8%|▊         | 543/6470 [09:39<1:15:31,  1.31it/s]
1|544|Loss: 5.069795608520508:   8%|▊         | 544/6470 [09:40<1:13:53,  1.34it/s]
1|545|Loss: 4.891145706176758:   8%|▊         | 544/6470 [09:40<1:13:53,  1.34it/s]
1|545|Loss: 4.891145706176758:   8%|▊         | 545/6470 [09:40<1:07:58,  1.45it/s]
1|546|Loss: 5.635434150695801:   8%|▊         | 545/6470 [09:41<1:07:58,  1.45it/s]
1|546|Loss: 5.635434150695801:   8%|▊         | 546/6470 [09:41<1:07:15,  1.47it/s]
1|547|Loss: 5.931945323944092:   8%|▊         | 546/6470 [09:41<1:07:15,  1.47it/s]
1|547|Loss: 5.931945323944092:   8%|▊         | 547/6470 [09:42<1:12:11,  1.37it/s]
1|548|Loss: 4.861852645874023:   8%|▊         | 547/6470 [09:42<1:12:11,  1.37it/s]
1|548|Loss: 4.861852645874023:   8%|▊         | 548/6470 [09:42<1:12:20,  1.36it/s]
1|549|Loss: 5.609127998352051:   8%|▊         | 548/6470 [09:43<1:12:20,  1.36it/s]
1|549|Loss: 5.609127998352051:   8%|▊         | 549/6470 [09:43<1:12:56,  1.35it/s]
1|550|Loss: 5.18697452545166:   8%|▊         | 549/6470 [09:44<1:12:56,  1.35it/s] 
1|550|Loss: 5.18697452545166:   9%|▊         | 550/6470 [09:44<1:11:30,  1.38it/s]
1|551|Loss: 5.822842597961426:   9%|▊         | 550/6470 [09:44<1:11:30,  1.38it/s]
1|551|Loss: 5.822842597961426:   9%|▊         | 551/6470 [09:45<1:16:21,  1.29it/s]
1|552|Loss: 5.193912506103516:   9%|▊         | 551/6470 [09:45<1:16:21,  1.29it/s]
1|552|Loss: 5.193912506103516:   9%|▊         | 552/6470 [09:46<1:18:00,  1.26it/s]
1|553|Loss: 5.297836780548096:   9%|▊         | 552/6470 [09:46<1:18:00,  1.26it/s]
1|553|Loss: 5.297836780548096:   9%|▊         | 553/6470 [09:46<1:12:15,  1.36it/s]
1|554|Loss: 5.366692066192627:   9%|▊         | 553/6470 [09:47<1:12:15,  1.36it/s]
1|554|Loss: 5.366692066192627:   9%|▊         | 554/6470 [09:47<1:10:07,  1.41it/s]
1|555|Loss: 5.52218770980835:   9%|▊         | 554/6470 [09:47<1:10:07,  1.41it/s] 
1|555|Loss: 5.52218770980835:   9%|▊         | 555/6470 [09:48<1:11:47,  1.37it/s]
1|556|Loss: 5.500850677490234:   9%|▊         | 555/6470 [09:48<1:11:47,  1.37it/s]
1|556|Loss: 5.500850677490234:   9%|▊         | 556/6470 [09:48<1:13:50,  1.33it/s]
1|557|Loss: 5.320433616638184:   9%|▊         | 556/6470 [09:49<1:13:50,  1.33it/s]
1|557|Loss: 5.320433616638184:   9%|▊         | 557/6470 [09:49<1:18:28,  1.26it/s]
1|558|Loss: 5.199936389923096:   9%|▊         | 557/6470 [09:50<1:18:28,  1.26it/s]
1|558|Loss: 5.199936389923096:   9%|▊         | 558/6470 [09:50<1:19:29,  1.24it/s]
1|559|Loss: 4.265129089355469:   9%|▊         | 558/6470 [09:51<1:19:29,  1.24it/s]
1|559|Loss: 4.265129089355469:   9%|▊         | 559/6470 [09:51<1:13:17,  1.34it/s]
1|560|Loss: 5.264007091522217:   9%|▊         | 559/6470 [09:51<1:13:17,  1.34it/s]
1|560|Loss: 5.264007091522217:   9%|▊         | 560/6470 [09:52<1:17:01,  1.28it/s]
1|561|Loss: 5.207897186279297:   9%|▊         | 560/6470 [09:52<1:17:01,  1.28it/s]
1|561|Loss: 5.207897186279297:   9%|▊         | 561/6470 [09:52<1:14:34,  1.32it/s]
1|562|Loss: 5.1512227058410645:   9%|▊         | 561/6470 [09:53<1:14:34,  1.32it/s]
1|562|Loss: 5.1512227058410645:   9%|▊         | 562/6470 [09:53<1:19:00,  1.25it/s]
1|563|Loss: 5.186619758605957:   9%|▊         | 562/6470 [09:54<1:19:00,  1.25it/s] 
1|563|Loss: 5.186619758605957:   9%|▊         | 563/6470 [09:54<1:19:31,  1.24it/s]
1|564|Loss: 5.375074863433838:   9%|▊         | 563/6470 [09:54<1:19:31,  1.24it/s]
1|564|Loss: 5.375074863433838:   9%|▊         | 564/6470 [09:55<1:15:04,  1.31it/s]
1|565|Loss: 4.907647609710693:   9%|▊         | 564/6470 [09:55<1:15:04,  1.31it/s]
1|565|Loss: 4.907647609710693:   9%|▊         | 565/6470 [09:55<1:10:40,  1.39it/s]
1|566|Loss: 5.0608391761779785:   9%|▊         | 565/6470 [09:56<1:10:40,  1.39it/s]
1|566|Loss: 5.0608391761779785:   9%|▊         | 566/6470 [09:56<1:09:43,  1.41it/s]
1|567|Loss: 5.476062774658203:   9%|▊         | 566/6470 [09:56<1:09:43,  1.41it/s] 
1|567|Loss: 5.476062774658203:   9%|▉         | 567/6470 [09:57<1:09:10,  1.42it/s]
1|568|Loss: 5.695864200592041:   9%|▉         | 567/6470 [09:57<1:09:10,  1.42it/s]
1|568|Loss: 5.695864200592041:   9%|▉         | 568/6470 [09:57<1:08:41,  1.43it/s]
1|569|Loss: 5.116230010986328:   9%|▉         | 568/6470 [09:58<1:08:41,  1.43it/s]
1|569|Loss: 5.116230010986328:   9%|▉         | 569/6470 [09:58<1:06:45,  1.47it/s]
1|570|Loss: 5.484655380249023:   9%|▉         | 569/6470 [09:58<1:06:45,  1.47it/s]
1|570|Loss: 5.484655380249023:   9%|▉         | 570/6470 [09:59<1:09:53,  1.41it/s]
1|571|Loss: 5.282200813293457:   9%|▉         | 570/6470 [09:59<1:09:53,  1.41it/s]
1|571|Loss: 5.282200813293457:   9%|▉         | 571/6470 [10:00<1:14:23,  1.32it/s]
1|572|Loss: 5.140493869781494:   9%|▉         | 571/6470 [10:00<1:14:23,  1.32it/s]
1|572|Loss: 5.140493869781494:   9%|▉         | 572/6470 [10:00<1:11:59,  1.37it/s]
1|573|Loss: 5.69390344619751:   9%|▉         | 572/6470 [10:01<1:11:59,  1.37it/s] 
1|573|Loss: 5.69390344619751:   9%|▉         | 573/6470 [10:01<1:15:46,  1.30it/s]
1|574|Loss: 4.858199119567871:   9%|▉         | 573/6470 [10:02<1:15:46,  1.30it/s]
1|574|Loss: 4.858199119567871:   9%|▉         | 574/6470 [10:02<1:13:58,  1.33it/s]
1|575|Loss: 5.4453535079956055:   9%|▉         | 574/6470 [10:02<1:13:58,  1.33it/s]
1|575|Loss: 5.4453535079956055:   9%|▉         | 575/6470 [10:03<1:13:29,  1.34it/s]
1|576|Loss: 4.666060447692871:   9%|▉         | 575/6470 [10:03<1:13:29,  1.34it/s] 
1|576|Loss: 4.666060447692871:   9%|▉         | 576/6470 [10:03<1:12:19,  1.36it/s]
1|577|Loss: 5.435704231262207:   9%|▉         | 576/6470 [10:04<1:12:19,  1.36it/s]
1|577|Loss: 5.435704231262207:   9%|▉         | 577/6470 [10:04<1:15:32,  1.30it/s]
1|578|Loss: 5.248461723327637:   9%|▉         | 577/6470 [10:04<1:15:32,  1.30it/s]
1|578|Loss: 5.248461723327637:   9%|▉         | 578/6470 [10:05<1:09:13,  1.42it/s]
1|579|Loss: 5.435723304748535:   9%|▉         | 578/6470 [10:05<1:09:13,  1.42it/s]
1|579|Loss: 5.435723304748535:   9%|▉         | 579/6470 [10:06<1:14:45,  1.31it/s]
1|580|Loss: 4.630306243896484:   9%|▉         | 579/6470 [10:06<1:14:45,  1.31it/s]
1|580|Loss: 4.630306243896484:   9%|▉         | 580/6470 [10:06<1:13:46,  1.33it/s]
1|581|Loss: 5.432604789733887:   9%|▉         | 580/6470 [10:07<1:13:46,  1.33it/s]
1|581|Loss: 5.432604789733887:   9%|▉         | 581/6470 [10:07<1:12:02,  1.36it/s]
1|582|Loss: 5.8164262771606445:   9%|▉         | 581/6470 [10:07<1:12:02,  1.36it/s]
1|582|Loss: 5.8164262771606445:   9%|▉         | 582/6470 [10:08<1:11:22,  1.37it/s]
1|583|Loss: 5.521543979644775:   9%|▉         | 582/6470 [10:08<1:11:22,  1.37it/s] 
1|583|Loss: 5.521543979644775:   9%|▉         | 583/6470 [10:09<1:12:00,  1.36it/s]
1|584|Loss: 5.368464469909668:   9%|▉         | 583/6470 [10:09<1:12:00,  1.36it/s]
1|584|Loss: 5.368464469909668:   9%|▉         | 584/6470 [10:09<1:13:53,  1.33it/s]
1|585|Loss: 5.251236438751221:   9%|▉         | 584/6470 [10:10<1:13:53,  1.33it/s]
1|585|Loss: 5.251236438751221:   9%|▉         | 585/6470 [10:10<1:16:02,  1.29it/s]
1|586|Loss: 5.36292839050293:   9%|▉         | 585/6470 [10:11<1:16:02,  1.29it/s] 
1|586|Loss: 5.36292839050293:   9%|▉         | 586/6470 [10:11<1:18:39,  1.25it/s]
1|587|Loss: 5.609087944030762:   9%|▉         | 586/6470 [10:11<1:18:39,  1.25it/s]
1|587|Loss: 5.609087944030762:   9%|▉         | 587/6470 [10:12<1:19:04,  1.24it/s]
1|588|Loss: 5.641037464141846:   9%|▉         | 587/6470 [10:12<1:19:04,  1.24it/s]
1|588|Loss: 5.641037464141846:   9%|▉         | 588/6470 [10:13<1:23:00,  1.18it/s]
1|589|Loss: 5.595703125:   9%|▉         | 588/6470 [10:13<1:23:00,  1.18it/s]      
1|589|Loss: 5.595703125:   9%|▉         | 589/6470 [10:14<1:23:17,  1.18it/s]
1|590|Loss: 5.482021331787109:   9%|▉         | 589/6470 [10:14<1:23:17,  1.18it/s]
1|590|Loss: 5.482021331787109:   9%|▉         | 590/6470 [10:14<1:22:01,  1.19it/s]
1|591|Loss: 5.49686861038208:   9%|▉         | 590/6470 [10:15<1:22:01,  1.19it/s] 
1|591|Loss: 5.49686861038208:   9%|▉         | 591/6470 [10:15<1:21:10,  1.21it/s]
1|592|Loss: 5.412060737609863:   9%|▉         | 591/6470 [10:16<1:21:10,  1.21it/s]
1|592|Loss: 5.412060737609863:   9%|▉         | 592/6470 [10:16<1:22:08,  1.19it/s]
1|593|Loss: 5.3764753341674805:   9%|▉         | 592/6470 [10:17<1:22:08,  1.19it/s]
1|593|Loss: 5.3764753341674805:   9%|▉         | 593/6470 [10:17<1:21:46,  1.20it/s]
1|594|Loss: 5.048027515411377:   9%|▉         | 593/6470 [10:17<1:21:46,  1.20it/s] 
1|594|Loss: 5.048027515411377:   9%|▉         | 594/6470 [10:18<1:17:47,  1.26it/s]
1|595|Loss: 5.399104595184326:   9%|▉         | 594/6470 [10:18<1:17:47,  1.26it/s]
1|595|Loss: 5.399104595184326:   9%|▉         | 595/6470 [10:18<1:15:06,  1.30it/s]
1|596|Loss: 5.228004455566406:   9%|▉         | 595/6470 [10:19<1:15:06,  1.30it/s]
1|596|Loss: 5.228004455566406:   9%|▉         | 596/6470 [10:19<1:15:21,  1.30it/s]
1|597|Loss: 4.968626022338867:   9%|▉         | 596/6470 [10:20<1:15:21,  1.30it/s]
1|597|Loss: 4.968626022338867:   9%|▉         | 597/6470 [10:20<1:12:05,  1.36it/s]
1|598|Loss: 5.284080505371094:   9%|▉         | 597/6470 [10:20<1:12:05,  1.36it/s]
1|598|Loss: 5.284080505371094:   9%|▉         | 598/6470 [10:20<1:10:48,  1.38it/s]
1|599|Loss: 5.573750972747803:   9%|▉         | 598/6470 [10:21<1:10:48,  1.38it/s]
1|599|Loss: 5.573750972747803:   9%|▉         | 599/6470 [10:21<1:10:40,  1.38it/s]
1|600|Loss: 5.011361598968506:   9%|▉         | 599/6470 [10:22<1:10:40,  1.38it/s]
1|600|Loss: 5.011361598968506:   9%|▉         | 600/6470 [10:22<1:07:56,  1.44it/s]
1|601|Loss: 5.479267120361328:   9%|▉         | 600/6470 [10:22<1:07:56,  1.44it/s]
1|601|Loss: 5.479267120361328:   9%|▉         | 601/6470 [10:23<1:09:46,  1.40it/s]
1|602|Loss: 5.682581424713135:   9%|▉         | 601/6470 [10:23<1:09:46,  1.40it/s]
1|602|Loss: 5.682581424713135:   9%|▉         | 602/6470 [10:24<1:15:22,  1.30it/s]
1|603|Loss: 5.1261796951293945:   9%|▉         | 602/6470 [10:24<1:15:22,  1.30it/s]
1|603|Loss: 5.1261796951293945:   9%|▉         | 603/6470 [10:24<1:20:27,  1.22it/s]
1|604|Loss: 5.166451454162598:   9%|▉         | 603/6470 [10:25<1:20:27,  1.22it/s] 
1|604|Loss: 5.166451454162598:   9%|▉         | 604/6470 [10:25<1:18:40,  1.24it/s]
1|605|Loss: 5.452523708343506:   9%|▉         | 604/6470 [10:26<1:18:40,  1.24it/s]
1|605|Loss: 5.452523708343506:   9%|▉         | 605/6470 [10:26<1:18:25,  1.25it/s]
1|606|Loss: 5.100858211517334:   9%|▉         | 605/6470 [10:26<1:18:25,  1.25it/s]
1|606|Loss: 5.100858211517334:   9%|▉         | 606/6470 [10:27<1:18:40,  1.24it/s]
1|607|Loss: 5.485628604888916:   9%|▉         | 606/6470 [10:27<1:18:40,  1.24it/s]
1|607|Loss: 5.485628604888916:   9%|▉         | 607/6470 [10:28<1:16:23,  1.28it/s]
1|608|Loss: 5.572479248046875:   9%|▉         | 607/6470 [10:28<1:16:23,  1.28it/s]
1|608|Loss: 5.572479248046875:   9%|▉         | 608/6470 [10:28<1:15:47,  1.29it/s]
1|609|Loss: 4.696810245513916:   9%|▉         | 608/6470 [10:29<1:15:47,  1.29it/s]
1|609|Loss: 4.696810245513916:   9%|▉         | 609/6470 [10:29<1:12:44,  1.34it/s]
1|610|Loss: 5.612606048583984:   9%|▉         | 609/6470 [10:29<1:12:44,  1.34it/s]
1|610|Loss: 5.612606048583984:   9%|▉         | 610/6470 [10:30<1:16:39,  1.27it/s]
1|611|Loss: 5.549259185791016:   9%|▉         | 610/6470 [10:30<1:16:39,  1.27it/s]
1|611|Loss: 5.549259185791016:   9%|▉         | 611/6470 [10:31<1:17:08,  1.27it/s]
1|612|Loss: 5.21829891204834:   9%|▉         | 611/6470 [10:31<1:17:08,  1.27it/s] 
1|612|Loss: 5.21829891204834:   9%|▉         | 612/6470 [10:31<1:17:26,  1.26it/s]
1|613|Loss: 5.563183307647705:   9%|▉         | 612/6470 [10:32<1:17:26,  1.26it/s]
1|613|Loss: 5.563183307647705:   9%|▉         | 613/6470 [10:32<1:20:27,  1.21it/s]
1|614|Loss: 5.285129070281982:   9%|▉         | 613/6470 [10:33<1:20:27,  1.21it/s]
1|614|Loss: 5.285129070281982:   9%|▉         | 614/6470 [10:33<1:20:12,  1.22it/s]
1|615|Loss: 4.854424953460693:   9%|▉         | 614/6470 [10:34<1:20:12,  1.22it/s]
1|615|Loss: 4.854424953460693:  10%|▉         | 615/6470 [10:34<1:17:21,  1.26it/s]
1|616|Loss: 5.635197639465332:  10%|▉         | 615/6470 [10:34<1:17:21,  1.26it/s]
1|616|Loss: 5.635197639465332:  10%|▉         | 616/6470 [10:35<1:17:38,  1.26it/s]
1|617|Loss: 4.384069919586182:  10%|▉         | 616/6470 [10:35<1:17:38,  1.26it/s]
1|617|Loss: 4.384069919586182:  10%|▉         | 617/6470 [10:35<1:13:22,  1.33it/s]
1|618|Loss: 5.214437484741211:  10%|▉         | 617/6470 [10:36<1:13:22,  1.33it/s]
1|618|Loss: 5.214437484741211:  10%|▉         | 618/6470 [10:36<1:10:12,  1.39it/s]
1|619|Loss: 5.125107288360596:  10%|▉         | 618/6470 [10:36<1:10:12,  1.39it/s]
1|619|Loss: 5.125107288360596:  10%|▉         | 619/6470 [10:37<1:08:57,  1.41it/s]
1|620|Loss: 5.550949573516846:  10%|▉         | 619/6470 [10:37<1:08:57,  1.41it/s]
1|620|Loss: 5.550949573516846:  10%|▉         | 620/6470 [10:38<1:18:29,  1.24it/s]
1|621|Loss: 5.484567165374756:  10%|▉         | 620/6470 [10:38<1:18:29,  1.24it/s]
1|621|Loss: 5.484567165374756:  10%|▉         | 621/6470 [10:39<1:21:36,  1.19it/s]
1|622|Loss: 5.369297981262207:  10%|▉         | 621/6470 [10:39<1:21:36,  1.19it/s]
1|622|Loss: 5.369297981262207:  10%|▉         | 622/6470 [10:39<1:18:49,  1.24it/s]
1|623|Loss: 4.548687934875488:  10%|▉         | 622/6470 [10:40<1:18:49,  1.24it/s]
1|623|Loss: 4.548687934875488:  10%|▉         | 623/6470 [10:40<1:11:19,  1.37it/s]
1|624|Loss: 5.250535488128662:  10%|▉         | 623/6470 [10:40<1:11:19,  1.37it/s]
1|624|Loss: 5.250535488128662:  10%|▉         | 624/6470 [10:41<1:08:17,  1.43it/s]
1|625|Loss: 5.491259574890137:  10%|▉         | 624/6470 [10:41<1:08:17,  1.43it/s]
1|625|Loss: 5.491259574890137:  10%|▉         | 625/6470 [10:41<1:11:08,  1.37it/s]
1|626|Loss: 4.6342668533325195:  10%|▉         | 625/6470 [10:42<1:11:08,  1.37it/s]
1|626|Loss: 4.6342668533325195:  10%|▉         | 626/6470 [10:42<1:07:43,  1.44it/s]
1|627|Loss: 5.068360328674316:  10%|▉         | 626/6470 [10:42<1:07:43,  1.44it/s] 
1|627|Loss: 5.068360328674316:  10%|▉         | 627/6470 [10:43<1:10:09,  1.39it/s]
1|628|Loss: 4.945695877075195:  10%|▉         | 627/6470 [10:43<1:10:09,  1.39it/s]
1|628|Loss: 4.945695877075195:  10%|▉         | 628/6470 [10:44<1:13:02,  1.33it/s]
1|629|Loss: 4.832034587860107:  10%|▉         | 628/6470 [10:44<1:13:02,  1.33it/s]
1|629|Loss: 4.832034587860107:  10%|▉         | 629/6470 [10:44<1:11:23,  1.36it/s]
1|630|Loss: 5.098811149597168:  10%|▉         | 629/6470 [10:45<1:11:23,  1.36it/s]
1|630|Loss: 5.098811149597168:  10%|▉         | 630/6470 [10:45<1:11:25,  1.36it/s]
1|631|Loss: 5.498658657073975:  10%|▉         | 630/6470 [10:45<1:11:25,  1.36it/s]
1|631|Loss: 5.498658657073975:  10%|▉         | 631/6470 [10:46<1:16:11,  1.28it/s]
1|632|Loss: 5.127059459686279:  10%|▉         | 631/6470 [10:46<1:16:11,  1.28it/s]
1|632|Loss: 5.127059459686279:  10%|▉         | 632/6470 [10:47<1:14:03,  1.31it/s]
1|633|Loss: 5.325395584106445:  10%|▉         | 632/6470 [10:47<1:14:03,  1.31it/s]
1|633|Loss: 5.325395584106445:  10%|▉         | 633/6470 [10:47<1:13:18,  1.33it/s]
1|634|Loss: 5.235452175140381:  10%|▉         | 633/6470 [10:48<1:13:18,  1.33it/s]
1|634|Loss: 5.235452175140381:  10%|▉         | 634/6470 [10:48<1:12:09,  1.35it/s]
1|635|Loss: 5.465634822845459:  10%|▉         | 634/6470 [10:49<1:12:09,  1.35it/s]
1|635|Loss: 5.465634822845459:  10%|▉         | 635/6470 [10:49<1:16:28,  1.27it/s]
1|636|Loss: 4.6173415184021:  10%|▉         | 635/6470 [10:49<1:16:28,  1.27it/s]  
1|636|Loss: 4.6173415184021:  10%|▉         | 636/6470 [10:50<1:11:40,  1.36it/s]
1|637|Loss: 4.839545726776123:  10%|▉         | 636/6470 [10:50<1:11:40,  1.36it/s]
1|637|Loss: 4.839545726776123:  10%|▉         | 637/6470 [10:50<1:12:18,  1.34it/s]
1|638|Loss: 5.134248733520508:  10%|▉         | 637/6470 [10:51<1:12:18,  1.34it/s]
1|638|Loss: 5.134248733520508:  10%|▉         | 638/6470 [10:51<1:09:58,  1.39it/s]
1|639|Loss: 5.525539875030518:  10%|▉         | 638/6470 [10:51<1:09:58,  1.39it/s]
1|639|Loss: 5.525539875030518:  10%|▉         | 639/6470 [10:52<1:14:47,  1.30it/s]
1|640|Loss: 5.32479190826416:  10%|▉         | 639/6470 [10:52<1:14:47,  1.30it/s] 
1|640|Loss: 5.32479190826416:  10%|▉         | 640/6470 [10:53<1:14:06,  1.31it/s]
1|641|Loss: 5.498141765594482:  10%|▉         | 640/6470 [10:53<1:14:06,  1.31it/s]
1|641|Loss: 5.498141765594482:  10%|▉         | 641/6470 [10:53<1:16:41,  1.27it/s]
1|642|Loss: 5.126455307006836:  10%|▉         | 641/6470 [10:54<1:16:41,  1.27it/s]
1|642|Loss: 5.126455307006836:  10%|▉         | 642/6470 [10:54<1:14:21,  1.31it/s]
1|643|Loss: 4.938323974609375:  10%|▉         | 642/6470 [10:55<1:14:21,  1.31it/s]
1|643|Loss: 4.938323974609375:  10%|▉         | 643/6470 [10:55<1:11:25,  1.36it/s]
1|644|Loss: 4.846402168273926:  10%|▉         | 643/6470 [10:55<1:11:25,  1.36it/s]
1|644|Loss: 4.846402168273926:  10%|▉         | 644/6470 [10:55<1:08:59,  1.41it/s]
1|645|Loss: 5.682754039764404:  10%|▉         | 644/6470 [10:56<1:08:59,  1.41it/s]
1|645|Loss: 5.682754039764404:  10%|▉         | 645/6470 [10:56<1:13:23,  1.32it/s]
1|646|Loss: 5.391602039337158:  10%|▉         | 645/6470 [10:57<1:13:23,  1.32it/s]
1|646|Loss: 5.391602039337158:  10%|▉         | 646/6470 [10:57<1:11:33,  1.36it/s]
1|647|Loss: 5.0603928565979:  10%|▉         | 646/6470 [10:57<1:11:33,  1.36it/s]  
1|647|Loss: 5.0603928565979:  10%|█         | 647/6470 [10:58<1:09:24,  1.40it/s]
1|648|Loss: 4.945554256439209:  10%|█         | 647/6470 [10:58<1:09:24,  1.40it/s]
1|648|Loss: 4.945554256439209:  10%|█         | 648/6470 [10:58<1:10:10,  1.38it/s]
1|649|Loss: 5.016765117645264:  10%|█         | 648/6470 [10:59<1:10:10,  1.38it/s]
1|649|Loss: 5.016765117645264:  10%|█         | 649/6470 [10:59<1:12:00,  1.35it/s]
1|650|Loss: 4.761125564575195:  10%|█         | 649/6470 [11:00<1:12:00,  1.35it/s]
1|650|Loss: 4.761125564575195:  10%|█         | 650/6470 [11:00<1:12:33,  1.34it/s]
1|651|Loss: 5.1278204917907715:  10%|█         | 650/6470 [11:00<1:12:33,  1.34it/s]
1|651|Loss: 5.1278204917907715:  10%|█         | 651/6470 [11:01<1:12:54,  1.33it/s]
1|652|Loss: 5.569110870361328:  10%|█         | 651/6470 [11:01<1:12:54,  1.33it/s] 
1|652|Loss: 5.569110870361328:  10%|█         | 652/6470 [11:02<1:16:09,  1.27it/s]
1|653|Loss: 5.0838751792907715:  10%|█         | 652/6470 [11:02<1:16:09,  1.27it/s]
1|653|Loss: 5.0838751792907715:  10%|█         | 653/6470 [11:03<1:18:51,  1.23it/s]
1|654|Loss: 5.264511585235596:  10%|█         | 653/6470 [11:03<1:18:51,  1.23it/s] 
1|654|Loss: 5.264511585235596:  10%|█         | 654/6470 [11:03<1:14:46,  1.30it/s]
1|655|Loss: 5.888613700866699:  10%|█         | 654/6470 [11:04<1:14:46,  1.30it/s]
1|655|Loss: 5.888613700866699:  10%|█         | 655/6470 [11:04<1:19:19,  1.22it/s]
1|656|Loss: 4.943739414215088:  10%|█         | 655/6470 [11:05<1:19:19,  1.22it/s]
1|656|Loss: 4.943739414215088:  10%|█         | 656/6470 [11:05<1:17:25,  1.25it/s]
1|657|Loss: 5.77833366394043:  10%|█         | 656/6470 [11:05<1:17:25,  1.25it/s] 
1|657|Loss: 5.77833366394043:  10%|█         | 657/6470 [11:06<1:20:48,  1.20it/s]
1|658|Loss: 5.431869029998779:  10%|█         | 657/6470 [11:06<1:20:48,  1.20it/s]
1|658|Loss: 5.431869029998779:  10%|█         | 658/6470 [11:07<1:18:08,  1.24it/s]
1|659|Loss: 4.835259914398193:  10%|█         | 658/6470 [11:07<1:18:08,  1.24it/s]
1|659|Loss: 4.835259914398193:  10%|█         | 659/6470 [11:07<1:13:00,  1.33it/s]
1|660|Loss: 5.664063453674316:  10%|█         | 659/6470 [11:08<1:13:00,  1.33it/s]
1|660|Loss: 5.664063453674316:  10%|█         | 660/6470 [11:08<1:13:33,  1.32it/s]
1|661|Loss: 5.317604064941406:  10%|█         | 660/6470 [11:08<1:13:33,  1.32it/s]
1|661|Loss: 5.317604064941406:  10%|█         | 661/6470 [11:09<1:14:00,  1.31it/s]
1|662|Loss: 5.256401062011719:  10%|█         | 661/6470 [11:09<1:14:00,  1.31it/s]
1|662|Loss: 5.256401062011719:  10%|█         | 662/6470 [11:09<1:12:02,  1.34it/s]
1|663|Loss: 5.202108383178711:  10%|█         | 662/6470 [11:10<1:12:02,  1.34it/s]
1|663|Loss: 5.202108383178711:  10%|█         | 663/6470 [11:10<1:08:59,  1.40it/s]
1|664|Loss: 5.414477825164795:  10%|█         | 663/6470 [11:10<1:08:59,  1.40it/s]
1|664|Loss: 5.414477825164795:  10%|█         | 664/6470 [11:11<1:09:31,  1.39it/s]
1|665|Loss: 4.943731307983398:  10%|█         | 664/6470 [11:11<1:09:31,  1.39it/s]
1|665|Loss: 4.943731307983398:  10%|█         | 665/6470 [11:12<1:10:33,  1.37it/s]
1|666|Loss: 4.839900493621826:  10%|█         | 665/6470 [11:12<1:10:33,  1.37it/s]
1|666|Loss: 4.839900493621826:  10%|█         | 666/6470 [11:12<1:06:21,  1.46it/s]
1|667|Loss: 5.810207366943359:  10%|█         | 666/6470 [11:13<1:06:21,  1.46it/s]
1|667|Loss: 5.810207366943359:  10%|█         | 667/6470 [11:13<1:10:04,  1.38it/s]
1|668|Loss: 5.5079264640808105:  10%|█         | 667/6470 [11:13<1:10:04,  1.38it/s]
1|668|Loss: 5.5079264640808105:  10%|█         | 668/6470 [11:14<1:14:37,  1.30it/s]
1|669|Loss: 5.642292022705078:  10%|█         | 668/6470 [11:14<1:14:37,  1.30it/s] 
1|669|Loss: 5.642292022705078:  10%|█         | 669/6470 [11:15<1:15:56,  1.27it/s]
1|670|Loss: 5.385451316833496:  10%|█         | 669/6470 [11:15<1:15:56,  1.27it/s]
1|670|Loss: 5.385451316833496:  10%|█         | 670/6470 [11:15<1:13:47,  1.31it/s]
1|671|Loss: 5.572310924530029:  10%|█         | 670/6470 [11:16<1:13:47,  1.31it/s]
1|671|Loss: 5.572310924530029:  10%|█         | 671/6470 [11:16<1:13:39,  1.31it/s]
1|672|Loss: 4.813906192779541:  10%|█         | 671/6470 [11:16<1:13:39,  1.31it/s]
1|672|Loss: 4.813906192779541:  10%|█         | 672/6470 [11:17<1:08:46,  1.41it/s]
1|673|Loss: 5.599424839019775:  10%|█         | 672/6470 [11:17<1:08:46,  1.41it/s]
1|673|Loss: 5.599424839019775:  10%|█         | 673/6470 [11:17<1:07:43,  1.43it/s]
1|674|Loss: 5.180171012878418:  10%|█         | 673/6470 [11:18<1:07:43,  1.43it/s]
1|674|Loss: 5.180171012878418:  10%|█         | 674/6470 [11:18<1:09:35,  1.39it/s]
1|675|Loss: 5.634079933166504:  10%|█         | 674/6470 [11:19<1:09:35,  1.39it/s]
1|675|Loss: 5.634079933166504:  10%|█         | 675/6470 [11:19<1:10:56,  1.36it/s]
1|676|Loss: 5.108024597167969:  10%|█         | 675/6470 [11:19<1:10:56,  1.36it/s]
1|676|Loss: 5.108024597167969:  10%|█         | 676/6470 [11:20<1:09:50,  1.38it/s]
1|677|Loss: 5.4727373123168945:  10%|█         | 676/6470 [11:20<1:09:50,  1.38it/s]
1|677|Loss: 5.4727373123168945:  10%|█         | 677/6470 [11:20<1:08:57,  1.40it/s]
1|678|Loss: 5.146080493927002:  10%|█         | 677/6470 [11:21<1:08:57,  1.40it/s] 
1|678|Loss: 5.146080493927002:  10%|█         | 678/6470 [11:21<1:10:46,  1.36it/s]
1|679|Loss: 5.5800886154174805:  10%|█         | 678/6470 [11:22<1:10:46,  1.36it/s]
1|679|Loss: 5.5800886154174805:  10%|█         | 679/6470 [11:22<1:17:36,  1.24it/s]
1|680|Loss: 4.844352722167969:  10%|█         | 679/6470 [11:22<1:17:36,  1.24it/s] 
1|680|Loss: 4.844352722167969:  11%|█         | 680/6470 [11:23<1:13:30,  1.31it/s]
1|681|Loss: 5.322681903839111:  11%|█         | 680/6470 [11:23<1:13:30,  1.31it/s]
1|681|Loss: 5.322681903839111:  11%|█         | 681/6470 [11:23<1:12:27,  1.33it/s]
1|682|Loss: 5.318181991577148:  11%|█         | 681/6470 [11:24<1:12:27,  1.33it/s]
1|682|Loss: 5.318181991577148:  11%|█         | 682/6470 [11:24<1:16:29,  1.26it/s]
1|683|Loss: 4.998460292816162:  11%|█         | 682/6470 [11:25<1:16:29,  1.26it/s]
1|683|Loss: 4.998460292816162:  11%|█         | 683/6470 [11:25<1:17:30,  1.24it/s]
1|684|Loss: 4.775598526000977:  11%|█         | 683/6470 [11:26<1:17:30,  1.24it/s]
1|684|Loss: 4.775598526000977:  11%|█         | 684/6470 [11:26<1:14:26,  1.30it/s]
1|685|Loss: 5.202014446258545:  11%|█         | 684/6470 [11:26<1:14:26,  1.30it/s]
1|685|Loss: 5.202014446258545:  11%|█         | 685/6470 [11:27<1:18:47,  1.22it/s]
1|686|Loss: 5.407200336456299:  11%|█         | 685/6470 [11:27<1:18:47,  1.22it/s]
1|686|Loss: 5.407200336456299:  11%|█         | 686/6470 [11:28<1:18:35,  1.23it/s]
1|687|Loss: 5.327414512634277:  11%|█         | 686/6470 [11:28<1:18:35,  1.23it/s]
1|687|Loss: 5.327414512634277:  11%|█         | 687/6470 [11:28<1:15:48,  1.27it/s]
1|688|Loss: 5.219651222229004:  11%|█         | 687/6470 [11:29<1:15:48,  1.27it/s]
1|688|Loss: 5.219651222229004:  11%|█         | 688/6470 [11:29<1:16:05,  1.27it/s]
1|689|Loss: 4.530085563659668:  11%|█         | 688/6470 [11:29<1:16:05,  1.27it/s]
1|689|Loss: 4.530085563659668:  11%|█         | 689/6470 [11:30<1:12:54,  1.32it/s]
1|690|Loss: 5.789369106292725:  11%|█         | 689/6470 [11:30<1:12:54,  1.32it/s]
1|690|Loss: 5.789369106292725:  11%|█         | 690/6470 [11:31<1:17:00,  1.25it/s]
1|691|Loss: 5.487215995788574:  11%|█         | 690/6470 [11:31<1:17:00,  1.25it/s]
1|691|Loss: 5.487215995788574:  11%|█         | 691/6470 [11:31<1:17:28,  1.24it/s]
1|692|Loss: 5.424281120300293:  11%|█         | 691/6470 [11:32<1:17:28,  1.24it/s]
1|692|Loss: 5.424281120300293:  11%|█         | 692/6470 [11:32<1:17:11,  1.25it/s]
1|693|Loss: 5.046437740325928:  11%|█         | 692/6470 [11:33<1:17:11,  1.25it/s]
1|693|Loss: 5.046437740325928:  11%|█         | 693/6470 [11:33<1:13:32,  1.31it/s]
1|694|Loss: 5.390150547027588:  11%|█         | 693/6470 [11:33<1:13:32,  1.31it/s]
1|694|Loss: 5.390150547027588:  11%|█         | 694/6470 [11:34<1:11:14,  1.35it/s]
1|695|Loss: 4.575216293334961:  11%|█         | 694/6470 [11:34<1:11:14,  1.35it/s]
1|695|Loss: 4.575216293334961:  11%|█         | 695/6470 [11:34<1:06:47,  1.44it/s]
1|696|Loss: 5.1749587059021:  11%|█         | 695/6470 [11:35<1:06:47,  1.44it/s]  
1|696|Loss: 5.1749587059021:  11%|█         | 696/6470 [11:35<1:11:42,  1.34it/s]
1|697|Loss: 5.085750102996826:  11%|█         | 696/6470 [11:36<1:11:42,  1.34it/s]
1|697|Loss: 5.085750102996826:  11%|█         | 697/6470 [11:36<1:10:44,  1.36it/s]
1|698|Loss: 4.78342866897583:  11%|█         | 697/6470 [11:36<1:10:44,  1.36it/s] 
1|698|Loss: 4.78342866897583:  11%|█         | 698/6470 [11:36<1:08:41,  1.40it/s]
1|699|Loss: 5.354487419128418:  11%|█         | 698/6470 [11:37<1:08:41,  1.40it/s]
1|699|Loss: 5.354487419128418:  11%|█         | 699/6470 [11:37<1:14:15,  1.30it/s]
1|700|Loss: 4.701541423797607:  11%|█         | 699/6470 [11:38<1:14:15,  1.30it/s]
1|700|Loss: 4.701541423797607:  11%|█         | 700/6470 [11:38<1:09:15,  1.39it/s]
1|701|Loss: 5.4669365882873535:  11%|█         | 700/6470 [11:38<1:09:15,  1.39it/s]
1|701|Loss: 5.4669365882873535:  11%|█         | 701/6470 [11:39<1:12:07,  1.33it/s]
1|702|Loss: 5.331884384155273:  11%|█         | 701/6470 [11:39<1:12:07,  1.33it/s] 
1|702|Loss: 5.331884384155273:  11%|█         | 702/6470 [11:40<1:18:55,  1.22it/s]
1|703|Loss: 4.81342077255249:  11%|█         | 702/6470 [11:40<1:18:55,  1.22it/s] 
1|703|Loss: 4.81342077255249:  11%|█         | 703/6470 [11:41<1:16:41,  1.25it/s]
1|704|Loss: 5.300023555755615:  11%|█         | 703/6470 [11:41<1:16:41,  1.25it/s]
1|704|Loss: 5.300023555755615:  11%|█         | 704/6470 [11:41<1:15:53,  1.27it/s]
1|705|Loss: 5.495643615722656:  11%|█         | 704/6470 [11:42<1:15:53,  1.27it/s]
1|705|Loss: 5.495643615722656:  11%|█         | 705/6470 [11:42<1:16:01,  1.26it/s]
1|706|Loss: 5.351819038391113:  11%|█         | 705/6470 [11:42<1:16:01,  1.26it/s]
1|706|Loss: 5.351819038391113:  11%|█         | 706/6470 [11:43<1:12:56,  1.32it/s]
1|707|Loss: 4.815001964569092:  11%|█         | 706/6470 [11:43<1:12:56,  1.32it/s]
1|707|Loss: 4.815001964569092:  11%|█         | 707/6470 [11:43<1:10:44,  1.36it/s]
1|708|Loss: 4.862123012542725:  11%|█         | 707/6470 [11:44<1:10:44,  1.36it/s]
1|708|Loss: 4.862123012542725:  11%|█         | 708/6470 [11:44<1:10:47,  1.36it/s]
1|709|Loss: 5.2757720947265625:  11%|█         | 708/6470 [11:45<1:10:47,  1.36it/s]
1|709|Loss: 5.2757720947265625:  11%|█         | 709/6470 [11:45<1:12:44,  1.32it/s]
1|710|Loss: 4.9104461669921875:  11%|█         | 709/6470 [11:45<1:12:44,  1.32it/s]
1|710|Loss: 4.9104461669921875:  11%|█         | 710/6470 [11:46<1:14:09,  1.29it/s]
1|711|Loss: 5.557596206665039:  11%|█         | 710/6470 [11:46<1:14:09,  1.29it/s] 
1|711|Loss: 5.557596206665039:  11%|█         | 711/6470 [11:47<1:17:47,  1.23it/s]
1|712|Loss: 5.550692081451416:  11%|█         | 711/6470 [11:47<1:17:47,  1.23it/s]
1|712|Loss: 5.550692081451416:  11%|█         | 712/6470 [11:48<1:21:38,  1.18it/s]
1|713|Loss: 4.873449325561523:  11%|█         | 712/6470 [11:48<1:21:38,  1.18it/s]
1|713|Loss: 4.873449325561523:  11%|█         | 713/6470 [11:48<1:19:33,  1.21it/s]
1|714|Loss: 4.8793840408325195:  11%|█         | 713/6470 [11:49<1:19:33,  1.21it/s]
1|714|Loss: 4.8793840408325195:  11%|█         | 714/6470 [11:49<1:12:54,  1.32it/s]
1|715|Loss: 5.742410659790039:  11%|█         | 714/6470 [11:49<1:12:54,  1.32it/s] 
1|715|Loss: 5.742410659790039:  11%|█         | 715/6470 [11:50<1:13:35,  1.30it/s]
1|716|Loss: 5.404948711395264:  11%|█         | 715/6470 [11:50<1:13:35,  1.30it/s]
1|716|Loss: 5.404948711395264:  11%|█         | 716/6470 [11:51<1:13:40,  1.30it/s]
1|717|Loss: 5.288538455963135:  11%|█         | 716/6470 [11:51<1:13:40,  1.30it/s]
1|717|Loss: 5.288538455963135:  11%|█         | 717/6470 [11:51<1:14:23,  1.29it/s]
1|718|Loss: 5.141791820526123:  11%|█         | 717/6470 [11:52<1:14:23,  1.29it/s]
1|718|Loss: 5.141791820526123:  11%|█         | 718/6470 [11:52<1:09:15,  1.38it/s]
1|719|Loss: 5.37257719039917:  11%|█         | 718/6470 [11:52<1:09:15,  1.38it/s] 
1|719|Loss: 5.37257719039917:  11%|█         | 719/6470 [11:53<1:10:41,  1.36it/s]
1|720|Loss: 5.059878349304199:  11%|█         | 719/6470 [11:53<1:10:41,  1.36it/s]
1|720|Loss: 5.059878349304199:  11%|█         | 720/6470 [11:54<1:12:51,  1.32it/s]
1|721|Loss: 4.9504804611206055:  11%|█         | 720/6470 [11:54<1:12:51,  1.32it/s]
1|721|Loss: 4.9504804611206055:  11%|█         | 721/6470 [11:54<1:11:35,  1.34it/s]
1|722|Loss: 4.716638565063477:  11%|█         | 721/6470 [11:55<1:11:35,  1.34it/s] 
1|722|Loss: 4.716638565063477:  11%|█         | 722/6470 [11:55<1:07:45,  1.41it/s]
1|723|Loss: 5.3116068840026855:  11%|█         | 722/6470 [11:55<1:07:45,  1.41it/s]
1|723|Loss: 5.3116068840026855:  11%|█         | 723/6470 [11:56<1:08:33,  1.40it/s]
1|724|Loss: 4.726253509521484:  11%|█         | 723/6470 [11:56<1:08:33,  1.40it/s] 
1|724|Loss: 4.726253509521484:  11%|█         | 724/6470 [11:56<1:11:59,  1.33it/s]
1|725|Loss: 5.517918109893799:  11%|█         | 724/6470 [11:57<1:11:59,  1.33it/s]
1|725|Loss: 5.517918109893799:  11%|█         | 725/6470 [11:57<1:14:18,  1.29it/s]
1|726|Loss: 4.803953647613525:  11%|█         | 725/6470 [11:58<1:14:18,  1.29it/s]
1|726|Loss: 4.803953647613525:  11%|█         | 726/6470 [11:58<1:09:34,  1.38it/s]
1|727|Loss: 4.692143440246582:  11%|█         | 726/6470 [11:58<1:09:34,  1.38it/s]
1|727|Loss: 4.692143440246582:  11%|█         | 727/6470 [11:59<1:06:15,  1.44it/s]
1|728|Loss: 5.203237056732178:  11%|█         | 727/6470 [11:59<1:06:15,  1.44it/s]
1|728|Loss: 5.203237056732178:  11%|█▏        | 728/6470 [11:59<1:04:02,  1.49it/s]
1|729|Loss: 5.096537113189697:  11%|█▏        | 728/6470 [11:59<1:04:02,  1.49it/s]
1|729|Loss: 5.096537113189697:  11%|█▏        | 729/6470 [12:00<1:02:01,  1.54it/s]
1|730|Loss: 5.167716026306152:  11%|█▏        | 729/6470 [12:00<1:02:01,  1.54it/s]
1|730|Loss: 5.167716026306152:  11%|█▏        | 730/6470 [12:01<1:16:39,  1.25it/s]
1|731|Loss: 4.793138027191162:  11%|█▏        | 730/6470 [12:01<1:16:39,  1.25it/s]
1|731|Loss: 4.793138027191162:  11%|█▏        | 731/6470 [12:02<1:18:50,  1.21it/s]
1|732|Loss: 5.03745698928833:  11%|█▏        | 731/6470 [12:02<1:18:50,  1.21it/s] 
1|732|Loss: 5.03745698928833:  11%|█▏        | 732/6470 [12:02<1:16:06,  1.26it/s]
1|733|Loss: 5.536983489990234:  11%|█▏        | 732/6470 [12:03<1:16:06,  1.26it/s]
1|733|Loss: 5.536983489990234:  11%|█▏        | 733/6470 [12:03<1:15:44,  1.26it/s]
1|734|Loss: 5.633242607116699:  11%|█▏        | 733/6470 [12:04<1:15:44,  1.26it/s]
1|734|Loss: 5.633242607116699:  11%|█▏        | 734/6470 [12:04<1:17:54,  1.23it/s]
1|735|Loss: 5.485710620880127:  11%|█▏        | 734/6470 [12:05<1:17:54,  1.23it/s]
1|735|Loss: 5.485710620880127:  11%|█▏        | 735/6470 [12:05<1:18:29,  1.22it/s]
1|736|Loss: 4.6153082847595215:  11%|█▏        | 735/6470 [12:05<1:18:29,  1.22it/s]
1|736|Loss: 4.6153082847595215:  11%|█▏        | 736/6470 [12:06<1:13:35,  1.30it/s]
1|737|Loss: 4.846163749694824:  11%|█▏        | 736/6470 [12:06<1:13:35,  1.30it/s] 
1|737|Loss: 4.846163749694824:  11%|█▏        | 737/6470 [12:06<1:09:27,  1.38it/s]
1|738|Loss: 5.407784461975098:  11%|█▏        | 737/6470 [12:07<1:09:27,  1.38it/s]
1|738|Loss: 5.407784461975098:  11%|█▏        | 738/6470 [12:07<1:12:12,  1.32it/s]
1|739|Loss: 5.367859840393066:  11%|█▏        | 738/6470 [12:08<1:12:12,  1.32it/s]
1|739|Loss: 5.367859840393066:  11%|█▏        | 739/6470 [12:08<1:13:50,  1.29it/s]
1|740|Loss: 5.1717448234558105:  11%|█▏        | 739/6470 [12:08<1:13:50,  1.29it/s]
1|740|Loss: 5.1717448234558105:  11%|█▏        | 740/6470 [12:09<1:12:34,  1.32it/s]
1|741|Loss: 5.778268814086914:  11%|█▏        | 740/6470 [12:09<1:12:34,  1.32it/s] 
1|741|Loss: 5.778268814086914:  11%|█▏        | 741/6470 [12:10<1:16:35,  1.25it/s]
1|742|Loss: 5.564521789550781:  11%|█▏        | 741/6470 [12:10<1:16:35,  1.25it/s]
1|742|Loss: 5.564521789550781:  11%|█▏        | 742/6470 [12:10<1:17:46,  1.23it/s]
1|743|Loss: 5.028493404388428:  11%|█▏        | 742/6470 [12:11<1:17:46,  1.23it/s]
1|743|Loss: 5.028493404388428:  11%|█▏        | 743/6470 [12:11<1:10:19,  1.36it/s]
1|744|Loss: 5.474637031555176:  11%|█▏        | 743/6470 [12:11<1:10:19,  1.36it/s]
1|744|Loss: 5.474637031555176:  11%|█▏        | 744/6470 [12:12<1:14:16,  1.28it/s]
1|745|Loss: 5.007040977478027:  11%|█▏        | 744/6470 [12:12<1:14:16,  1.28it/s]
1|745|Loss: 5.007040977478027:  12%|█▏        | 745/6470 [12:13<1:13:59,  1.29it/s]
1|746|Loss: 5.269833087921143:  12%|█▏        | 745/6470 [12:13<1:13:59,  1.29it/s]
1|746|Loss: 5.269833087921143:  12%|█▏        | 746/6470 [12:13<1:12:18,  1.32it/s]
1|747|Loss: 5.262751579284668:  12%|█▏        | 746/6470 [12:14<1:12:18,  1.32it/s]
1|747|Loss: 5.262751579284668:  12%|█▏        | 747/6470 [12:14<1:08:54,  1.38it/s]
1|748|Loss: 5.311534404754639:  12%|█▏        | 747/6470 [12:14<1:08:54,  1.38it/s]
1|748|Loss: 5.311534404754639:  12%|█▏        | 748/6470 [12:15<1:09:11,  1.38it/s]
1|749|Loss: 4.689733982086182:  12%|█▏        | 748/6470 [12:15<1:09:11,  1.38it/s]
1|749|Loss: 4.689733982086182:  12%|█▏        | 749/6470 [12:15<1:10:06,  1.36it/s]
1|750|Loss: 5.2148661613464355:  12%|█▏        | 749/6470 [12:16<1:10:06,  1.36it/s]
1|750|Loss: 5.2148661613464355:  12%|█▏        | 750/6470 [12:16<1:12:04,  1.32it/s]
1|751|Loss: 5.226097106933594:  12%|█▏        | 750/6470 [12:17<1:12:04,  1.32it/s] 
1|751|Loss: 5.226097106933594:  12%|█▏        | 751/6470 [12:17<1:10:43,  1.35it/s]
1|752|Loss: 5.148197174072266:  12%|█▏        | 751/6470 [12:17<1:10:43,  1.35it/s]
1|752|Loss: 5.148197174072266:  12%|█▏        | 752/6470 [12:18<1:14:15,  1.28it/s]
1|753|Loss: 5.658849239349365:  12%|█▏        | 752/6470 [12:18<1:14:15,  1.28it/s]
1|753|Loss: 5.658849239349365:  12%|█▏        | 753/6470 [12:19<1:17:14,  1.23it/s]
1|754|Loss: 5.333835601806641:  12%|█▏        | 753/6470 [12:19<1:17:14,  1.23it/s]
1|754|Loss: 5.333835601806641:  12%|█▏        | 754/6470 [12:20<1:20:20,  1.19it/s]
1|755|Loss: 5.0650739669799805:  12%|█▏        | 754/6470 [12:20<1:20:20,  1.19it/s]
1|755|Loss: 5.0650739669799805:  12%|█▏        | 755/6470 [12:20<1:17:51,  1.22it/s]
1|756|Loss: 4.865320682525635:  12%|█▏        | 755/6470 [12:21<1:17:51,  1.22it/s] 
1|756|Loss: 4.865320682525635:  12%|█▏        | 756/6470 [12:21<1:17:25,  1.23it/s]
1|757|Loss: 4.92718505859375:  12%|█▏        | 756/6470 [12:22<1:17:25,  1.23it/s] 
1|757|Loss: 4.92718505859375:  12%|█▏        | 757/6470 [12:22<1:15:00,  1.27it/s]
1|758|Loss: 5.017451286315918:  12%|█▏        | 757/6470 [12:22<1:15:00,  1.27it/s]
1|758|Loss: 5.017451286315918:  12%|█▏        | 758/6470 [12:23<1:13:48,  1.29it/s]
1|759|Loss: 4.786334037780762:  12%|█▏        | 758/6470 [12:23<1:13:48,  1.29it/s]
1|759|Loss: 4.786334037780762:  12%|█▏        | 759/6470 [12:23<1:10:30,  1.35it/s]
1|760|Loss: 5.02427339553833:  12%|█▏        | 759/6470 [12:24<1:10:30,  1.35it/s] 
1|760|Loss: 5.02427339553833:  12%|█▏        | 760/6470 [12:24<1:09:00,  1.38it/s]
1|761|Loss: 4.947105884552002:  12%|█▏        | 760/6470 [12:24<1:09:00,  1.38it/s]
1|761|Loss: 4.947105884552002:  12%|█▏        | 761/6470 [12:25<1:09:39,  1.37it/s]
1|762|Loss: 4.727417945861816:  12%|█▏        | 761/6470 [12:25<1:09:39,  1.37it/s]
1|762|Loss: 4.727417945861816:  12%|█▏        | 762/6470 [12:25<1:08:58,  1.38it/s]
1|763|Loss: 5.323975563049316:  12%|█▏        | 762/6470 [12:26<1:08:58,  1.38it/s]
1|763|Loss: 5.323975563049316:  12%|█▏        | 763/6470 [12:26<1:11:08,  1.34it/s]
1|764|Loss: 5.316952705383301:  12%|█▏        | 763/6470 [12:27<1:11:08,  1.34it/s]
1|764|Loss: 5.316952705383301:  12%|█▏        | 764/6470 [12:27<1:16:41,  1.24it/s]
1|765|Loss: 5.337224006652832:  12%|█▏        | 764/6470 [12:28<1:16:41,  1.24it/s]
1|765|Loss: 5.337224006652832:  12%|█▏        | 765/6470 [12:28<1:17:54,  1.22it/s]
1|766|Loss: 4.908186912536621:  12%|█▏        | 765/6470 [12:29<1:17:54,  1.22it/s]
1|766|Loss: 4.908186912536621:  12%|█▏        | 766/6470 [12:29<1:21:32,  1.17it/s]
1|767|Loss: 5.470173358917236:  12%|█▏        | 766/6470 [12:29<1:21:32,  1.17it/s]
1|767|Loss: 5.470173358917236:  12%|█▏        | 767/6470 [12:30<1:21:50,  1.16it/s]
1|768|Loss: 5.473245620727539:  12%|█▏        | 767/6470 [12:30<1:21:50,  1.16it/s]
1|768|Loss: 5.473245620727539:  12%|█▏        | 768/6470 [12:31<1:23:20,  1.14it/s]
1|769|Loss: 4.995161056518555:  12%|█▏        | 768/6470 [12:31<1:23:20,  1.14it/s]
1|769|Loss: 4.995161056518555:  12%|█▏        | 769/6470 [12:32<1:19:34,  1.19it/s]
1|770|Loss: 5.247954845428467:  12%|█▏        | 769/6470 [12:32<1:19:34,  1.19it/s]
1|770|Loss: 5.247954845428467:  12%|█▏        | 770/6470 [12:32<1:15:41,  1.25it/s]
1|771|Loss: 5.503551483154297:  12%|█▏        | 770/6470 [12:33<1:15:41,  1.25it/s]
1|771|Loss: 5.503551483154297:  12%|█▏        | 771/6470 [12:33<1:17:24,  1.23it/s]
1|772|Loss: 5.488414287567139:  12%|█▏        | 771/6470 [12:34<1:17:24,  1.23it/s]
1|772|Loss: 5.488414287567139:  12%|█▏        | 772/6470 [12:34<1:17:18,  1.23it/s]
1|773|Loss: 4.801214694976807:  12%|█▏        | 772/6470 [12:34<1:17:18,  1.23it/s]
1|773|Loss: 4.801214694976807:  12%|█▏        | 773/6470 [12:35<1:14:14,  1.28it/s]
1|774|Loss: 5.3645453453063965:  12%|█▏        | 773/6470 [12:35<1:14:14,  1.28it/s]
1|774|Loss: 5.3645453453063965:  12%|█▏        | 774/6470 [12:35<1:13:21,  1.29it/s]
1|775|Loss: 5.545258522033691:  12%|█▏        | 774/6470 [12:36<1:13:21,  1.29it/s] 
1|775|Loss: 5.545258522033691:  12%|█▏        | 775/6470 [12:36<1:14:39,  1.27it/s]
1|776|Loss: 5.156302452087402:  12%|█▏        | 775/6470 [12:37<1:14:39,  1.27it/s]
1|776|Loss: 5.156302452087402:  12%|█▏        | 776/6470 [12:37<1:15:29,  1.26it/s]
1|777|Loss: 5.519876003265381:  12%|█▏        | 776/6470 [12:37<1:15:29,  1.26it/s]
1|777|Loss: 5.519876003265381:  12%|█▏        | 777/6470 [12:38<1:15:52,  1.25it/s]
1|778|Loss: 5.248691082000732:  12%|█▏        | 777/6470 [12:38<1:15:52,  1.25it/s]
1|778|Loss: 5.248691082000732:  12%|█▏        | 778/6470 [12:39<1:16:21,  1.24it/s]
1|779|Loss: 5.66795539855957:  12%|█▏        | 778/6470 [12:39<1:16:21,  1.24it/s] 
1|779|Loss: 5.66795539855957:  12%|█▏        | 779/6470 [12:40<1:19:01,  1.20it/s]
1|780|Loss: 3.444636106491089:  12%|█▏        | 779/6470 [12:40<1:19:01,  1.20it/s]
1|780|Loss: 3.444636106491089:  12%|█▏        | 780/6470 [12:40<1:08:48,  1.38it/s]
1|781|Loss: 5.552541732788086:  12%|█▏        | 780/6470 [12:40<1:08:48,  1.38it/s]
1|781|Loss: 5.552541732788086:  12%|█▏        | 781/6470 [12:41<1:03:47,  1.49it/s]
1|782|Loss: 5.004039764404297:  12%|█▏        | 781/6470 [12:41<1:03:47,  1.49it/s]
1|782|Loss: 5.004039764404297:  12%|█▏        | 782/6470 [12:41<1:05:29,  1.45it/s]
1|783|Loss: 4.778889179229736:  12%|█▏        | 782/6470 [12:42<1:05:29,  1.45it/s]
1|783|Loss: 4.778889179229736:  12%|█▏        | 783/6470 [12:42<1:04:55,  1.46it/s]
1|784|Loss: 4.072962284088135:  12%|█▏        | 783/6470 [12:42<1:04:55,  1.46it/s]
1|784|Loss: 4.072962284088135:  12%|█▏        | 784/6470 [12:42<59:52,  1.58it/s]  
1|785|Loss: 5.574709415435791:  12%|█▏        | 784/6470 [12:43<59:52,  1.58it/s]
1|785|Loss: 5.574709415435791:  12%|█▏        | 785/6470 [12:43<1:04:07,  1.48it/s]
1|786|Loss: 5.362441062927246:  12%|█▏        | 785/6470 [12:44<1:04:07,  1.48it/s]
1|786|Loss: 5.362441062927246:  12%|█▏        | 786/6470 [12:44<1:10:19,  1.35it/s]
1|787|Loss: 5.027970790863037:  12%|█▏        | 786/6470 [12:45<1:10:19,  1.35it/s]
1|787|Loss: 5.027970790863037:  12%|█▏        | 787/6470 [12:45<1:08:00,  1.39it/s]
1|788|Loss: 5.220293998718262:  12%|█▏        | 787/6470 [12:45<1:08:00,  1.39it/s]
1|788|Loss: 5.220293998718262:  12%|█▏        | 788/6470 [12:46<1:11:44,  1.32it/s]
1|789|Loss: 5.538769245147705:  12%|█▏        | 788/6470 [12:46<1:11:44,  1.32it/s]
1|789|Loss: 5.538769245147705:  12%|█▏        | 789/6470 [12:46<1:13:52,  1.28it/s]
1|790|Loss: 6.038628101348877:  12%|█▏        | 789/6470 [12:47<1:13:52,  1.28it/s]
1|790|Loss: 6.038628101348877:  12%|█▏        | 790/6470 [12:47<1:19:23,  1.19it/s]
1|791|Loss: 5.440508842468262:  12%|█▏        | 790/6470 [12:48<1:19:23,  1.19it/s]
1|791|Loss: 5.440508842468262:  12%|█▏        | 791/6470 [12:48<1:20:05,  1.18it/s]
1|792|Loss: 4.894155025482178:  12%|█▏        | 791/6470 [12:49<1:20:05,  1.18it/s]
1|792|Loss: 4.894155025482178:  12%|█▏        | 792/6470 [12:49<1:12:02,  1.31it/s]
1|793|Loss: 5.709843635559082:  12%|█▏        | 792/6470 [12:49<1:12:02,  1.31it/s]
1|793|Loss: 5.709843635559082:  12%|█▏        | 793/6470 [12:50<1:11:39,  1.32it/s]
1|794|Loss: 5.156124114990234:  12%|█▏        | 793/6470 [12:50<1:11:39,  1.32it/s]
1|794|Loss: 5.156124114990234:  12%|█▏        | 794/6470 [12:50<1:12:02,  1.31it/s]
1|795|Loss: 5.056173801422119:  12%|█▏        | 794/6470 [12:51<1:12:02,  1.31it/s]
1|795|Loss: 5.056173801422119:  12%|█▏        | 795/6470 [12:51<1:07:38,  1.40it/s]
1|796|Loss: 5.101436614990234:  12%|█▏        | 795/6470 [12:51<1:07:38,  1.40it/s]
1|796|Loss: 5.101436614990234:  12%|█▏        | 796/6470 [12:52<1:04:12,  1.47it/s]
1|797|Loss: 5.35170316696167:  12%|█▏        | 796/6470 [12:52<1:04:12,  1.47it/s] 
1|797|Loss: 5.35170316696167:  12%|█▏        | 797/6470 [12:52<1:06:48,  1.42it/s]
1|798|Loss: 4.423119068145752:  12%|█▏        | 797/6470 [12:53<1:06:48,  1.42it/s]
1|798|Loss: 4.423119068145752:  12%|█▏        | 798/6470 [12:53<1:02:29,  1.51it/s]
1|799|Loss: 5.275019645690918:  12%|█▏        | 798/6470 [12:53<1:02:29,  1.51it/s]
1|799|Loss: 5.275019645690918:  12%|█▏        | 799/6470 [12:54<1:08:03,  1.39it/s]
1|800|Loss: 5.252682685852051:  12%|█▏        | 799/6470 [12:54<1:08:03,  1.39it/s]
1|800|Loss: 5.252682685852051:  12%|█▏        | 800/6470 [12:55<1:12:02,  1.31it/s]
1|801|Loss: 4.932631969451904:  12%|█▏        | 800/6470 [12:55<1:12:02,  1.31it/s]
1|801|Loss: 4.932631969451904:  12%|█▏        | 801/6470 [12:55<1:12:18,  1.31it/s]
1|802|Loss: 4.9904608726501465:  12%|█▏        | 801/6470 [12:56<1:12:18,  1.31it/s]
1|802|Loss: 4.9904608726501465:  12%|█▏        | 802/6470 [12:56<1:11:58,  1.31it/s]
1|803|Loss: 5.047650337219238:  12%|█▏        | 802/6470 [12:57<1:11:58,  1.31it/s] 
1|803|Loss: 5.047650337219238:  12%|█▏        | 803/6470 [12:57<1:12:47,  1.30it/s]
1|804|Loss: 5.458977222442627:  12%|█▏        | 803/6470 [12:57<1:12:47,  1.30it/s]
1|804|Loss: 5.458977222442627:  12%|█▏        | 804/6470 [12:58<1:13:29,  1.28it/s]
1|805|Loss: 5.2036285400390625:  12%|█▏        | 804/6470 [12:58<1:13:29,  1.28it/s]
1|805|Loss: 5.2036285400390625:  12%|█▏        | 805/6470 [12:59<1:13:23,  1.29it/s]
1|806|Loss: 5.553316116333008:  12%|█▏        | 805/6470 [12:59<1:13:23,  1.29it/s] 
1|806|Loss: 5.553316116333008:  12%|█▏        | 806/6470 [12:59<1:18:32,  1.20it/s]
1|807|Loss: 4.291637420654297:  12%|█▏        | 806/6470 [13:00<1:18:32,  1.20it/s]
1|807|Loss: 4.291637420654297:  12%|█▏        | 807/6470 [13:00<1:11:36,  1.32it/s]
1|808|Loss: 5.780128479003906:  12%|█▏        | 807/6470 [13:01<1:11:36,  1.32it/s]
1|808|Loss: 5.780128479003906:  12%|█▏        | 808/6470 [13:01<1:22:52,  1.14it/s]
1|809|Loss: 4.698262691497803:  12%|█▏        | 808/6470 [13:02<1:22:52,  1.14it/s]
1|809|Loss: 4.698262691497803:  13%|█▎        | 809/6470 [13:02<1:22:07,  1.15it/s]
1|810|Loss: 4.973056316375732:  13%|█▎        | 809/6470 [13:02<1:22:07,  1.15it/s]
1|810|Loss: 4.973056316375732:  13%|█▎        | 810/6470 [13:03<1:13:59,  1.27it/s]
1|811|Loss: 5.278341770172119:  13%|█▎        | 810/6470 [13:03<1:13:59,  1.27it/s]
1|811|Loss: 5.278341770172119:  13%|█▎        | 811/6470 [13:03<1:13:05,  1.29it/s]
1|812|Loss: 5.354401111602783:  13%|█▎        | 811/6470 [13:04<1:13:05,  1.29it/s]
1|812|Loss: 5.354401111602783:  13%|█▎        | 812/6470 [13:04<1:10:16,  1.34it/s]
1|813|Loss: 5.060739517211914:  13%|█▎        | 812/6470 [13:04<1:10:16,  1.34it/s]
1|813|Loss: 5.060739517211914:  13%|█▎        | 813/6470 [13:05<1:04:20,  1.47it/s]
1|814|Loss: 5.254265308380127:  13%|█▎        | 813/6470 [13:05<1:04:20,  1.47it/s]
1|814|Loss: 5.254265308380127:  13%|█▎        | 814/6470 [13:05<1:00:16,  1.56it/s]
1|815|Loss: 5.231261253356934:  13%|█▎        | 814/6470 [13:06<1:00:16,  1.56it/s]
1|815|Loss: 5.231261253356934:  13%|█▎        | 815/6470 [13:06<1:00:47,  1.55it/s]
1|816|Loss: 4.804144382476807:  13%|█▎        | 815/6470 [13:06<1:00:47,  1.55it/s]
1|816|Loss: 4.804144382476807:  13%|█▎        | 816/6470 [13:07<1:02:36,  1.51it/s]
1|817|Loss: 4.814953804016113:  13%|█▎        | 816/6470 [13:07<1:02:36,  1.51it/s]
1|817|Loss: 4.814953804016113:  13%|█▎        | 817/6470 [13:07<1:06:02,  1.43it/s]
1|818|Loss: 5.28950834274292:  13%|█▎        | 817/6470 [13:08<1:06:02,  1.43it/s] 
1|818|Loss: 5.28950834274292:  13%|█▎        | 818/6470 [13:08<1:08:48,  1.37it/s]
1|819|Loss: 4.446791172027588:  13%|█▎        | 818/6470 [13:08<1:08:48,  1.37it/s]
1|819|Loss: 4.446791172027588:  13%|█▎        | 819/6470 [13:09<1:04:11,  1.47it/s]
1|820|Loss: 5.389650821685791:  13%|█▎        | 819/6470 [13:09<1:04:11,  1.47it/s]
1|820|Loss: 5.389650821685791:  13%|█▎        | 820/6470 [13:09<1:05:40,  1.43it/s]
1|821|Loss: 5.1608710289001465:  13%|█▎        | 820/6470 [13:10<1:05:40,  1.43it/s]
1|821|Loss: 5.1608710289001465:  13%|█▎        | 821/6470 [13:10<1:09:47,  1.35it/s]
1|822|Loss: 4.946416854858398:  13%|█▎        | 821/6470 [13:11<1:09:47,  1.35it/s] 
1|822|Loss: 4.946416854858398:  13%|█▎        | 822/6470 [13:11<1:09:17,  1.36it/s]
1|823|Loss: 4.933332920074463:  13%|█▎        | 822/6470 [13:11<1:09:17,  1.36it/s]
1|823|Loss: 4.933332920074463:  13%|█▎        | 823/6470 [13:12<1:06:48,  1.41it/s]
1|824|Loss: 5.429958343505859:  13%|█▎        | 823/6470 [13:12<1:06:48,  1.41it/s]
1|824|Loss: 5.429958343505859:  13%|█▎        | 824/6470 [13:12<1:10:25,  1.34it/s]
1|825|Loss: 5.519584655761719:  13%|█▎        | 824/6470 [13:13<1:10:25,  1.34it/s]
1|825|Loss: 5.519584655761719:  13%|█▎        | 825/6470 [13:13<1:10:17,  1.34it/s]
1|826|Loss: 5.679227352142334:  13%|█▎        | 825/6470 [13:14<1:10:17,  1.34it/s]
1|826|Loss: 5.679227352142334:  13%|█▎        | 826/6470 [13:14<1:15:20,  1.25it/s]
1|827|Loss: 5.428067684173584:  13%|█▎        | 826/6470 [13:15<1:15:20,  1.25it/s]
1|827|Loss: 5.428067684173584:  13%|█▎        | 827/6470 [13:15<1:15:44,  1.24it/s]
1|828|Loss: 4.356750965118408:  13%|█▎        | 827/6470 [13:15<1:15:44,  1.24it/s]
1|828|Loss: 4.356750965118408:  13%|█▎        | 828/6470 [13:15<1:08:10,  1.38it/s]
1|829|Loss: 5.119475841522217:  13%|█▎        | 828/6470 [13:16<1:08:10,  1.38it/s]
1|829|Loss: 5.119475841522217:  13%|█▎        | 829/6470 [13:16<1:08:00,  1.38it/s]
1|830|Loss: 5.334329605102539:  13%|█▎        | 829/6470 [13:17<1:08:00,  1.38it/s]
1|830|Loss: 5.334329605102539:  13%|█▎        | 830/6470 [13:17<1:09:48,  1.35it/s]
1|831|Loss: 5.221779823303223:  13%|█▎        | 830/6470 [13:17<1:09:48,  1.35it/s]
1|831|Loss: 5.221779823303223:  13%|█▎        | 831/6470 [13:18<1:11:21,  1.32it/s]
1|832|Loss: 5.50247859954834:  13%|█▎        | 831/6470 [13:18<1:11:21,  1.32it/s] 
1|832|Loss: 5.50247859954834:  13%|█▎        | 832/6470 [13:19<1:11:35,  1.31it/s]
1|833|Loss: 5.661656379699707:  13%|█▎        | 832/6470 [13:19<1:11:35,  1.31it/s]
1|833|Loss: 5.661656379699707:  13%|█▎        | 833/6470 [13:19<1:16:10,  1.23it/s]
1|834|Loss: 4.718037128448486:  13%|█▎        | 833/6470 [13:20<1:16:10,  1.23it/s]
1|834|Loss: 4.718037128448486:  13%|█▎        | 834/6470 [13:20<1:13:07,  1.28it/s]
1|835|Loss: 5.2058186531066895:  13%|█▎        | 834/6470 [13:21<1:13:07,  1.28it/s]
1|835|Loss: 5.2058186531066895:  13%|█▎        | 835/6470 [13:21<1:12:57,  1.29it/s]
1|836|Loss: 5.194107532501221:  13%|█▎        | 835/6470 [13:21<1:12:57,  1.29it/s] 
1|836|Loss: 5.194107532501221:  13%|█▎        | 836/6470 [13:22<1:13:45,  1.27it/s]
1|837|Loss: 5.374629974365234:  13%|█▎        | 836/6470 [13:22<1:13:45,  1.27it/s]
1|837|Loss: 5.374629974365234:  13%|█▎        | 837/6470 [13:23<1:13:49,  1.27it/s]
1|838|Loss: 5.196591854095459:  13%|█▎        | 837/6470 [13:23<1:13:49,  1.27it/s]
1|838|Loss: 5.196591854095459:  13%|█▎        | 838/6470 [13:23<1:13:53,  1.27it/s]
1|839|Loss: 4.976978778839111:  13%|█▎        | 838/6470 [13:24<1:13:53,  1.27it/s]
1|839|Loss: 4.976978778839111:  13%|█▎        | 839/6470 [13:24<1:10:38,  1.33it/s]
1|840|Loss: 4.76797342300415:  13%|█▎        | 839/6470 [13:24<1:10:38,  1.33it/s] 
1|840|Loss: 4.76797342300415:  13%|█▎        | 840/6470 [13:25<1:10:44,  1.33it/s]
1|841|Loss: 5.55979061126709:  13%|█▎        | 840/6470 [13:25<1:10:44,  1.33it/s]
1|841|Loss: 5.55979061126709:  13%|█▎        | 841/6470 [13:26<1:13:54,  1.27it/s]
1|842|Loss: 5.172552108764648:  13%|█▎        | 841/6470 [13:26<1:13:54,  1.27it/s]
1|842|Loss: 5.172552108764648:  13%|█▎        | 842/6470 [13:27<1:17:39,  1.21it/s]
1|843|Loss: 5.5600128173828125:  13%|█▎        | 842/6470 [13:27<1:17:39,  1.21it/s]
1|843|Loss: 5.5600128173828125:  13%|█▎        | 843/6470 [13:28<1:23:14,  1.13it/s]
1|844|Loss: 5.487880706787109:  13%|█▎        | 843/6470 [13:28<1:23:14,  1.13it/s] 
1|844|Loss: 5.487880706787109:  13%|█▎        | 844/6470 [13:28<1:22:12,  1.14it/s]
1|845|Loss: 5.498204708099365:  13%|█▎        | 844/6470 [13:29<1:22:12,  1.14it/s]
1|845|Loss: 5.498204708099365:  13%|█▎        | 845/6470 [13:29<1:17:56,  1.20it/s]
1|846|Loss: 5.373988628387451:  13%|█▎        | 845/6470 [13:30<1:17:56,  1.20it/s]
1|846|Loss: 5.373988628387451:  13%|█▎        | 846/6470 [13:30<1:16:13,  1.23it/s]
1|847|Loss: 5.444551944732666:  13%|█▎        | 846/6470 [13:30<1:16:13,  1.23it/s]
1|847|Loss: 5.444551944732666:  13%|█▎        | 847/6470 [13:31<1:15:39,  1.24it/s]
1|848|Loss: 5.258175849914551:  13%|█▎        | 847/6470 [13:31<1:15:39,  1.24it/s]
1|848|Loss: 5.258175849914551:  13%|█▎        | 848/6470 [13:32<1:15:15,  1.24it/s]
1|849|Loss: 4.771291732788086:  13%|█▎        | 848/6470 [13:32<1:15:15,  1.24it/s]
1|849|Loss: 4.771291732788086:  13%|█▎        | 849/6470 [13:32<1:16:24,  1.23it/s]
1|850|Loss: 4.981473922729492:  13%|█▎        | 849/6470 [13:33<1:16:24,  1.23it/s]
1|850|Loss: 4.981473922729492:  13%|█▎        | 850/6470 [13:33<1:18:17,  1.20it/s]
1|851|Loss: 4.854066371917725:  13%|█▎        | 850/6470 [13:34<1:18:17,  1.20it/s]
1|851|Loss: 4.854066371917725:  13%|█▎        | 851/6470 [13:34<1:15:35,  1.24it/s]
1|852|Loss: 4.225384712219238:  13%|█▎        | 851/6470 [13:34<1:15:35,  1.24it/s]
1|852|Loss: 4.225384712219238:  13%|█▎        | 852/6470 [13:35<1:17:33,  1.21it/s]
1|853|Loss: 5.446647644042969:  13%|█▎        | 852/6470 [13:35<1:17:33,  1.21it/s]
1|853|Loss: 5.446647644042969:  13%|█▎        | 853/6470 [13:36<1:17:55,  1.20it/s]
1|854|Loss: 5.4610276222229:  13%|█▎        | 853/6470 [13:36<1:17:55,  1.20it/s]  
1|854|Loss: 5.4610276222229:  13%|█▎        | 854/6470 [13:37<1:17:23,  1.21it/s]
1|855|Loss: 5.436777591705322:  13%|█▎        | 854/6470 [13:37<1:17:23,  1.21it/s]
1|855|Loss: 5.436777591705322:  13%|█▎        | 855/6470 [13:37<1:14:47,  1.25it/s]
1|856|Loss: 4.930727005004883:  13%|█▎        | 855/6470 [13:38<1:14:47,  1.25it/s]
1|856|Loss: 4.930727005004883:  13%|█▎        | 856/6470 [13:38<1:08:38,  1.36it/s]
1|857|Loss: 4.669408321380615:  13%|█▎        | 856/6470 [13:38<1:08:38,  1.36it/s]
1|857|Loss: 4.669408321380615:  13%|█▎        | 857/6470 [13:38<1:01:59,  1.51it/s]
1|858|Loss: 4.544286251068115:  13%|█▎        | 857/6470 [13:39<1:01:59,  1.51it/s]
1|858|Loss: 4.544286251068115:  13%|█▎        | 858/6470 [13:39<1:04:02,  1.46it/s]
1|859|Loss: 4.860614776611328:  13%|█▎        | 858/6470 [13:39<1:04:02,  1.46it/s]
1|859|Loss: 4.860614776611328:  13%|█▎        | 859/6470 [13:40<1:04:36,  1.45it/s]
1|860|Loss: 4.852054119110107:  13%|█▎        | 859/6470 [13:40<1:04:36,  1.45it/s]
1|860|Loss: 4.852054119110107:  13%|█▎        | 860/6470 [13:40<1:01:30,  1.52it/s]
1|861|Loss: 5.1403961181640625:  13%|█▎        | 860/6470 [13:41<1:01:30,  1.52it/s]
1|861|Loss: 5.1403961181640625:  13%|█▎        | 861/6470 [13:41<1:04:26,  1.45it/s]
1|862|Loss: 4.955959320068359:  13%|█▎        | 861/6470 [13:42<1:04:26,  1.45it/s] 
1|862|Loss: 4.955959320068359:  13%|█▎        | 862/6470 [13:42<1:06:44,  1.40it/s]
1|863|Loss: 5.3583173751831055:  13%|█▎        | 862/6470 [13:42<1:06:44,  1.40it/s]
1|863|Loss: 5.3583173751831055:  13%|█▎        | 863/6470 [13:43<1:05:29,  1.43it/s]
1|864|Loss: 4.957246780395508:  13%|█▎        | 863/6470 [13:43<1:05:29,  1.43it/s] 
1|864|Loss: 4.957246780395508:  13%|█▎        | 864/6470 [13:43<1:05:12,  1.43it/s]
1|865|Loss: 5.382297039031982:  13%|█▎        | 864/6470 [13:44<1:05:12,  1.43it/s]
1|865|Loss: 5.382297039031982:  13%|█▎        | 865/6470 [13:44<1:09:12,  1.35it/s]
1|866|Loss: 5.305502414703369:  13%|█▎        | 865/6470 [13:45<1:09:12,  1.35it/s]
1|866|Loss: 5.305502414703369:  13%|█▎        | 866/6470 [13:45<1:12:48,  1.28it/s]
1|867|Loss: 5.221924304962158:  13%|█▎        | 866/6470 [13:45<1:12:48,  1.28it/s]
1|867|Loss: 5.221924304962158:  13%|█▎        | 867/6470 [13:46<1:16:11,  1.23it/s]
1|868|Loss: 5.27407693862915:  13%|█▎        | 867/6470 [13:46<1:16:11,  1.23it/s] 
1|868|Loss: 5.27407693862915:  13%|█▎        | 868/6470 [13:47<1:13:35,  1.27it/s]
1|869|Loss: 5.144892692565918:  13%|█▎        | 868/6470 [13:47<1:13:35,  1.27it/s]
1|869|Loss: 5.144892692565918:  13%|█▎        | 869/6470 [13:47<1:12:59,  1.28it/s]
1|870|Loss: 5.813358783721924:  13%|█▎        | 869/6470 [13:48<1:12:59,  1.28it/s]
1|870|Loss: 5.813358783721924:  13%|█▎        | 870/6470 [13:48<1:18:18,  1.19it/s]
1|871|Loss: 5.029597759246826:  13%|█▎        | 870/6470 [13:49<1:18:18,  1.19it/s]
1|871|Loss: 5.029597759246826:  13%|█▎        | 871/6470 [13:49<1:17:06,  1.21it/s]
1|872|Loss: 4.880209922790527:  13%|█▎        | 871/6470 [13:50<1:17:06,  1.21it/s]
1|872|Loss: 4.880209922790527:  13%|█▎        | 872/6470 [13:50<1:15:45,  1.23it/s]
1|873|Loss: 5.382526874542236:  13%|█▎        | 872/6470 [13:50<1:15:45,  1.23it/s]
1|873|Loss: 5.382526874542236:  13%|█▎        | 873/6470 [13:51<1:17:10,  1.21it/s]
1|874|Loss: 5.017911434173584:  13%|█▎        | 873/6470 [13:52<1:17:10,  1.21it/s]
1|874|Loss: 5.017911434173584:  14%|█▎        | 874/6470 [13:52<1:35:14,  1.02s/it]
1|875|Loss: 5.954817771911621:  14%|█▎        | 874/6470 [13:53<1:35:14,  1.02s/it]
1|875|Loss: 5.954817771911621:  14%|█▎        | 875/6470 [13:53<1:40:21,  1.08s/it]
1|876|Loss: 5.342129230499268:  14%|█▎        | 875/6470 [13:54<1:40:21,  1.08s/it]
1|876|Loss: 5.342129230499268:  14%|█▎        | 876/6470 [13:54<1:31:36,  1.02it/s]
1|877|Loss: 5.222705364227295:  14%|█▎        | 876/6470 [13:55<1:31:36,  1.02it/s]
1|877|Loss: 5.222705364227295:  14%|█▎        | 877/6470 [13:55<1:24:41,  1.10it/s]
1|878|Loss: 5.068163871765137:  14%|█▎        | 877/6470 [13:55<1:24:41,  1.10it/s]
1|878|Loss: 5.068163871765137:  14%|█▎        | 878/6470 [13:56<1:24:26,  1.10it/s]
1|879|Loss: 5.400509834289551:  14%|█▎        | 878/6470 [13:56<1:24:26,  1.10it/s]
1|879|Loss: 5.400509834289551:  14%|█▎        | 879/6470 [13:57<1:19:00,  1.18it/s]
1|880|Loss: 4.4576921463012695:  14%|█▎        | 879/6470 [13:57<1:19:00,  1.18it/s]
1|880|Loss: 4.4576921463012695:  14%|█▎        | 880/6470 [13:57<1:10:45,  1.32it/s]
1|881|Loss: 4.970792293548584:  14%|█▎        | 880/6470 [13:57<1:10:45,  1.32it/s] 
1|881|Loss: 4.970792293548584:  14%|█▎        | 881/6470 [13:58<1:06:51,  1.39it/s]
1|882|Loss: 4.956966876983643:  14%|█▎        | 881/6470 [13:58<1:06:51,  1.39it/s]
1|882|Loss: 4.956966876983643:  14%|█▎        | 882/6470 [13:59<1:11:02,  1.31it/s]
1|883|Loss: 5.09443473815918:  14%|█▎        | 882/6470 [13:59<1:11:02,  1.31it/s] 
1|883|Loss: 5.09443473815918:  14%|█▎        | 883/6470 [13:59<1:11:55,  1.29it/s]
1|884|Loss: 4.948795318603516:  14%|█▎        | 883/6470 [14:00<1:11:55,  1.29it/s]
1|884|Loss: 4.948795318603516:  14%|█▎        | 884/6470 [14:00<1:15:22,  1.24it/s]
1|885|Loss: 5.2570414543151855:  14%|█▎        | 884/6470 [14:01<1:15:22,  1.24it/s]
1|885|Loss: 5.2570414543151855:  14%|█▎        | 885/6470 [14:01<1:10:38,  1.32it/s]
1|886|Loss: 5.391348361968994:  14%|█▎        | 885/6470 [14:01<1:10:38,  1.32it/s] 
1|886|Loss: 5.391348361968994:  14%|█▎        | 886/6470 [14:02<1:13:03,  1.27it/s]
1|887|Loss: 5.001545429229736:  14%|█▎        | 886/6470 [14:02<1:13:03,  1.27it/s]
1|887|Loss: 5.001545429229736:  14%|█▎        | 887/6470 [14:02<1:09:58,  1.33it/s]
1|888|Loss: 5.327794551849365:  14%|█▎        | 887/6470 [14:03<1:09:58,  1.33it/s]
1|888|Loss: 5.327794551849365:  14%|█▎        | 888/6470 [14:03<1:10:42,  1.32it/s]
1|889|Loss: 5.267209529876709:  14%|█▎        | 888/6470 [14:04<1:10:42,  1.32it/s]
1|889|Loss: 5.267209529876709:  14%|█▎        | 889/6470 [14:04<1:14:01,  1.26it/s]
1|890|Loss: 4.95040225982666:  14%|█▎        | 889/6470 [14:05<1:14:01,  1.26it/s] 
1|890|Loss: 4.95040225982666:  14%|█▍        | 890/6470 [14:05<1:11:42,  1.30it/s]
1|891|Loss: 4.863961219787598:  14%|█▍        | 890/6470 [14:05<1:11:42,  1.30it/s]
1|891|Loss: 4.863961219787598:  14%|█▍        | 891/6470 [14:06<1:09:03,  1.35it/s]
1|892|Loss: 5.310206413269043:  14%|█▍        | 891/6470 [14:06<1:09:03,  1.35it/s]
1|892|Loss: 5.310206413269043:  14%|█▍        | 892/6470 [14:06<1:07:49,  1.37it/s]
1|893|Loss: 5.668123245239258:  14%|█▍        | 892/6470 [14:07<1:07:49,  1.37it/s]
1|893|Loss: 5.668123245239258:  14%|█▍        | 893/6470 [14:07<1:13:45,  1.26it/s]
1|894|Loss: 4.894466400146484:  14%|█▍        | 893/6470 [14:08<1:13:45,  1.26it/s]
1|894|Loss: 4.894466400146484:  14%|█▍        | 894/6470 [14:08<1:11:05,  1.31it/s]
1|895|Loss: 5.188134670257568:  14%|█▍        | 894/6470 [14:08<1:11:05,  1.31it/s]
1|895|Loss: 5.188134670257568:  14%|█▍        | 895/6470 [14:09<1:09:16,  1.34it/s]
1|896|Loss: 4.3440632820129395:  14%|█▍        | 895/6470 [14:09<1:09:16,  1.34it/s]
1|896|Loss: 4.3440632820129395:  14%|█▍        | 896/6470 [14:09<1:10:42,  1.31it/s]
1|897|Loss: 5.009603977203369:  14%|█▍        | 896/6470 [14:10<1:10:42,  1.31it/s] 
1|897|Loss: 5.009603977203369:  14%|█▍        | 897/6470 [14:10<1:11:59,  1.29it/s]
1|898|Loss: 5.408834934234619:  14%|█▍        | 897/6470 [14:11<1:11:59,  1.29it/s]
1|898|Loss: 5.408834934234619:  14%|█▍        | 898/6470 [14:11<1:12:58,  1.27it/s]
1|899|Loss: 5.495087623596191:  14%|█▍        | 898/6470 [14:11<1:12:58,  1.27it/s]
1|899|Loss: 5.495087623596191:  14%|█▍        | 899/6470 [14:12<1:15:29,  1.23it/s]
1|900|Loss: 5.4435648918151855:  14%|█▍        | 899/6470 [14:12<1:15:29,  1.23it/s]
1|900|Loss: 5.4435648918151855:  14%|█▍        | 900/6470 [14:13<1:13:07,  1.27it/s]
1|901|Loss: 5.050698757171631:  14%|█▍        | 900/6470 [14:13<1:13:07,  1.27it/s] 
1|901|Loss: 5.050698757171631:  14%|█▍        | 901/6470 [14:13<1:10:27,  1.32it/s]
1|902|Loss: 5.1605224609375:  14%|█▍        | 901/6470 [14:14<1:10:27,  1.32it/s]  
1|902|Loss: 5.1605224609375:  14%|█▍        | 902/6470 [14:14<1:11:53,  1.29it/s]
1|903|Loss: 5.137661457061768:  14%|█▍        | 902/6470 [14:14<1:11:53,  1.29it/s]
1|903|Loss: 5.137661457061768:  14%|█▍        | 903/6470 [14:15<1:10:14,  1.32it/s]
1|904|Loss: 5.645735263824463:  14%|█▍        | 903/6470 [14:15<1:10:14,  1.32it/s]
1|904|Loss: 5.645735263824463:  14%|█▍        | 904/6470 [14:16<1:13:21,  1.26it/s]
1|905|Loss: 5.085939407348633:  14%|█▍        | 904/6470 [14:16<1:13:21,  1.26it/s]
1|905|Loss: 5.085939407348633:  14%|█▍        | 905/6470 [14:16<1:13:23,  1.26it/s]
1|906|Loss: 4.246461868286133:  14%|█▍        | 905/6470 [14:17<1:13:23,  1.26it/s]
1|906|Loss: 4.246461868286133:  14%|█▍        | 906/6470 [14:17<1:06:16,  1.40it/s]
1|907|Loss: 4.96620512008667:  14%|█▍        | 906/6470 [14:17<1:06:16,  1.40it/s] 
1|907|Loss: 4.96620512008667:  14%|█▍        | 907/6470 [14:18<1:08:05,  1.36it/s]
1|908|Loss: 4.8864545822143555:  14%|█▍        | 907/6470 [14:18<1:08:05,  1.36it/s]
1|908|Loss: 4.8864545822143555:  14%|█▍        | 908/6470 [14:19<1:08:35,  1.35it/s]
1|909|Loss: 5.162713050842285:  14%|█▍        | 908/6470 [14:19<1:08:35,  1.35it/s] 
1|909|Loss: 5.162713050842285:  14%|█▍        | 909/6470 [14:19<1:07:05,  1.38it/s]
1|910|Loss: 4.928399562835693:  14%|█▍        | 909/6470 [14:20<1:07:05,  1.38it/s]
1|910|Loss: 4.928399562835693:  14%|█▍        | 910/6470 [14:20<1:07:44,  1.37it/s]
1|911|Loss: 5.389886856079102:  14%|█▍        | 910/6470 [14:20<1:07:44,  1.37it/s]
1|911|Loss: 5.389886856079102:  14%|█▍        | 911/6470 [14:21<1:11:05,  1.30it/s]
1|912|Loss: 5.197937965393066:  14%|█▍        | 911/6470 [14:21<1:11:05,  1.30it/s]
1|912|Loss: 5.197937965393066:  14%|█▍        | 912/6470 [14:22<1:12:27,  1.28it/s]
1|913|Loss: 5.433815002441406:  14%|█▍        | 912/6470 [14:22<1:12:27,  1.28it/s]
1|913|Loss: 5.433815002441406:  14%|█▍        | 913/6470 [14:22<1:12:06,  1.28it/s]
1|914|Loss: 5.053987503051758:  14%|█▍        | 913/6470 [14:23<1:12:06,  1.28it/s]
1|914|Loss: 5.053987503051758:  14%|█▍        | 914/6470 [14:23<1:10:10,  1.32it/s]
1|915|Loss: 5.095607280731201:  14%|█▍        | 914/6470 [14:24<1:10:10,  1.32it/s]
1|915|Loss: 5.095607280731201:  14%|█▍        | 915/6470 [14:24<1:11:09,  1.30it/s]
1|916|Loss: 4.980458736419678:  14%|█▍        | 915/6470 [14:24<1:11:09,  1.30it/s]
1|916|Loss: 4.980458736419678:  14%|█▍        | 916/6470 [14:25<1:10:10,  1.32it/s]
1|917|Loss: 4.858737945556641:  14%|█▍        | 916/6470 [14:25<1:10:10,  1.32it/s]
1|917|Loss: 4.858737945556641:  14%|█▍        | 917/6470 [14:25<1:08:05,  1.36it/s]
1|918|Loss: 4.939901351928711:  14%|█▍        | 917/6470 [14:26<1:08:05,  1.36it/s]
1|918|Loss: 4.939901351928711:  14%|█▍        | 918/6470 [14:26<1:06:05,  1.40it/s]
1|919|Loss: 4.877410411834717:  14%|█▍        | 918/6470 [14:26<1:06:05,  1.40it/s]
1|919|Loss: 4.877410411834717:  14%|█▍        | 919/6470 [14:27<1:02:04,  1.49it/s]
1|920|Loss: 5.4822916984558105:  14%|█▍        | 919/6470 [14:27<1:02:04,  1.49it/s]
1|920|Loss: 5.4822916984558105:  14%|█▍        | 920/6470 [14:27<1:03:41,  1.45it/s]
1|921|Loss: 5.509177207946777:  14%|█▍        | 920/6470 [14:28<1:03:41,  1.45it/s] 
1|921|Loss: 5.509177207946777:  14%|█▍        | 921/6470 [14:28<1:08:13,  1.36it/s]
1|922|Loss: 4.895814895629883:  14%|█▍        | 921/6470 [14:29<1:08:13,  1.36it/s]
1|922|Loss: 4.895814895629883:  14%|█▍        | 922/6470 [14:29<1:10:11,  1.32it/s]
1|923|Loss: 4.969324111938477:  14%|█▍        | 922/6470 [14:29<1:10:11,  1.32it/s]
1|923|Loss: 4.969324111938477:  14%|█▍        | 923/6470 [14:30<1:08:35,  1.35it/s]
1|924|Loss: 5.298583507537842:  14%|█▍        | 923/6470 [14:30<1:08:35,  1.35it/s]
1|924|Loss: 5.298583507537842:  14%|█▍        | 924/6470 [14:30<1:07:26,  1.37it/s]
1|925|Loss: 5.339652061462402:  14%|█▍        | 924/6470 [14:31<1:07:26,  1.37it/s]
1|925|Loss: 5.339652061462402:  14%|█▍        | 925/6470 [14:32<1:23:13,  1.11it/s]
1|926|Loss: 5.377033233642578:  14%|█▍        | 925/6470 [14:32<1:23:13,  1.11it/s]
1|926|Loss: 5.377033233642578:  14%|█▍        | 926/6470 [14:33<1:29:12,  1.04it/s]
1|927|Loss: 4.970113277435303:  14%|█▍        | 926/6470 [14:33<1:29:12,  1.04it/s]
1|927|Loss: 4.970113277435303:  14%|█▍        | 927/6470 [14:34<1:25:44,  1.08it/s]
1|928|Loss: 5.679239749908447:  14%|█▍        | 927/6470 [14:34<1:25:44,  1.08it/s]
1|928|Loss: 5.679239749908447:  14%|█▍        | 928/6470 [14:35<1:25:07,  1.09it/s]
1|929|Loss: 5.472445011138916:  14%|█▍        | 928/6470 [14:35<1:25:07,  1.09it/s]
1|929|Loss: 5.472445011138916:  14%|█▍        | 929/6470 [14:35<1:23:12,  1.11it/s]
1|930|Loss: 4.7386579513549805:  14%|█▍        | 929/6470 [14:36<1:23:12,  1.11it/s]
1|930|Loss: 4.7386579513549805:  14%|█▍        | 930/6470 [14:36<1:20:25,  1.15it/s]
1|931|Loss: 5.587419509887695:  14%|█▍        | 930/6470 [14:37<1:20:25,  1.15it/s] 
1|931|Loss: 5.587419509887695:  14%|█▍        | 931/6470 [14:37<1:15:55,  1.22it/s]
1|932|Loss: 5.563488483428955:  14%|█▍        | 931/6470 [14:37<1:15:55,  1.22it/s]
1|932|Loss: 5.563488483428955:  14%|█▍        | 932/6470 [14:38<1:12:39,  1.27it/s]
1|933|Loss: 5.217347145080566:  14%|█▍        | 932/6470 [14:38<1:12:39,  1.27it/s]
1|933|Loss: 5.217347145080566:  14%|█▍        | 933/6470 [14:38<1:15:04,  1.23it/s]
1|934|Loss: 4.964418411254883:  14%|█▍        | 933/6470 [14:39<1:15:04,  1.23it/s]
1|934|Loss: 4.964418411254883:  14%|█▍        | 934/6470 [14:39<1:18:30,  1.18it/s]
1|935|Loss: 5.284188747406006:  14%|█▍        | 934/6470 [14:40<1:18:30,  1.18it/s]
1|935|Loss: 5.284188747406006:  14%|█▍        | 935/6470 [14:40<1:14:57,  1.23it/s]
1|936|Loss: 5.280138969421387:  14%|█▍        | 935/6470 [14:41<1:14:57,  1.23it/s]
1|936|Loss: 5.280138969421387:  14%|█▍        | 936/6470 [14:41<1:13:56,  1.25it/s]
1|937|Loss: 4.868910789489746:  14%|█▍        | 936/6470 [14:41<1:13:56,  1.25it/s]
1|937|Loss: 4.868910789489746:  14%|█▍        | 937/6470 [14:42<1:11:09,  1.30it/s]
1|938|Loss: 5.361227035522461:  14%|█▍        | 937/6470 [14:42<1:11:09,  1.30it/s]
1|938|Loss: 5.361227035522461:  14%|█▍        | 938/6470 [14:42<1:11:02,  1.30it/s]
1|939|Loss: 4.668183326721191:  14%|█▍        | 938/6470 [14:43<1:11:02,  1.30it/s]
1|939|Loss: 4.668183326721191:  15%|█▍        | 939/6470 [14:43<1:11:55,  1.28it/s]
1|940|Loss: 5.393444061279297:  15%|█▍        | 939/6470 [14:44<1:11:55,  1.28it/s]
1|940|Loss: 5.393444061279297:  15%|█▍        | 940/6470 [14:44<1:11:41,  1.29it/s]
1|941|Loss: 4.961722373962402:  15%|█▍        | 940/6470 [14:44<1:11:41,  1.29it/s]
1|941|Loss: 4.961722373962402:  15%|█▍        | 941/6470 [14:45<1:11:34,  1.29it/s]
1|942|Loss: 5.197460651397705:  15%|█▍        | 941/6470 [14:45<1:11:34,  1.29it/s]
1|942|Loss: 5.197460651397705:  15%|█▍        | 942/6470 [14:45<1:12:04,  1.28it/s]
1|943|Loss: 5.005344390869141:  15%|█▍        | 942/6470 [14:46<1:12:04,  1.28it/s]
1|943|Loss: 5.005344390869141:  15%|█▍        | 943/6470 [14:46<1:10:24,  1.31it/s]
1|944|Loss: 4.936647891998291:  15%|█▍        | 943/6470 [14:47<1:10:24,  1.31it/s]
1|944|Loss: 4.936647891998291:  15%|█▍        | 944/6470 [14:47<1:10:41,  1.30it/s]
1|945|Loss: 5.022308349609375:  15%|█▍        | 944/6470 [14:47<1:10:41,  1.30it/s]
1|945|Loss: 5.022308349609375:  15%|█▍        | 945/6470 [14:48<1:11:11,  1.29it/s]
1|946|Loss: 5.203577041625977:  15%|█▍        | 945/6470 [14:48<1:11:11,  1.29it/s]
1|946|Loss: 5.203577041625977:  15%|█▍        | 946/6470 [14:49<1:13:38,  1.25it/s]
1|947|Loss: 5.10752010345459:  15%|█▍        | 946/6470 [14:49<1:13:38,  1.25it/s] 
1|947|Loss: 5.10752010345459:  15%|█▍        | 947/6470 [14:49<1:13:16,  1.26it/s]
1|948|Loss: 5.0132927894592285:  15%|█▍        | 947/6470 [14:50<1:13:16,  1.26it/s]
1|948|Loss: 5.0132927894592285:  15%|█▍        | 948/6470 [14:50<1:08:44,  1.34it/s]
1|949|Loss: 3.9568538665771484:  15%|█▍        | 948/6470 [14:50<1:08:44,  1.34it/s]
1|949|Loss: 3.9568538665771484:  15%|█▍        | 949/6470 [14:51<1:01:49,  1.49it/s]
1|950|Loss: 5.183533191680908:  15%|█▍        | 949/6470 [14:51<1:01:49,  1.49it/s] 
1|950|Loss: 5.183533191680908:  15%|█▍        | 950/6470 [14:51<1:04:00,  1.44it/s]
1|951|Loss: 5.530172348022461:  15%|█▍        | 950/6470 [14:52<1:04:00,  1.44it/s]
1|951|Loss: 5.530172348022461:  15%|█▍        | 951/6470 [14:52<1:11:35,  1.28it/s]
1|952|Loss: 5.284095764160156:  15%|█▍        | 951/6470 [14:53<1:11:35,  1.28it/s]
1|952|Loss: 5.284095764160156:  15%|█▍        | 952/6470 [14:53<1:16:25,  1.20it/s]
1|953|Loss: 4.948884963989258:  15%|█▍        | 952/6470 [14:54<1:16:25,  1.20it/s]
1|953|Loss: 4.948884963989258:  15%|█▍        | 953/6470 [14:54<1:13:26,  1.25it/s]
1|954|Loss: 5.064473628997803:  15%|█▍        | 953/6470 [14:54<1:13:26,  1.25it/s]
1|954|Loss: 5.064473628997803:  15%|█▍        | 954/6470 [14:55<1:13:03,  1.26it/s]
1|955|Loss: 5.08737325668335:  15%|█▍        | 954/6470 [14:55<1:13:03,  1.26it/s] 
1|955|Loss: 5.08737325668335:  15%|█▍        | 955/6470 [14:55<1:08:23,  1.34it/s]
1|956|Loss: 5.238171577453613:  15%|█▍        | 955/6470 [14:56<1:08:23,  1.34it/s]
1|956|Loss: 5.238171577453613:  15%|█▍        | 956/6470 [14:56<1:15:03,  1.22it/s]
1|957|Loss: 5.682981967926025:  15%|█▍        | 956/6470 [14:57<1:15:03,  1.22it/s]
1|957|Loss: 5.682981967926025:  15%|█▍        | 957/6470 [14:57<1:15:59,  1.21it/s]
1|958|Loss: 5.057596206665039:  15%|█▍        | 957/6470 [14:58<1:15:59,  1.21it/s]
1|958|Loss: 5.057596206665039:  15%|█▍        | 958/6470 [14:58<1:09:40,  1.32it/s]
1|959|Loss: 5.137867450714111:  15%|█▍        | 958/6470 [14:58<1:09:40,  1.32it/s]
1|959|Loss: 5.137867450714111:  15%|█▍        | 959/6470 [14:59<1:10:58,  1.29it/s]
1|960|Loss: 5.3332438468933105:  15%|█▍        | 959/6470 [14:59<1:10:58,  1.29it/s]
1|960|Loss: 5.3332438468933105:  15%|█▍        | 960/6470 [14:59<1:08:15,  1.35it/s]
1|961|Loss: 4.973443508148193:  15%|█▍        | 960/6470 [15:00<1:08:15,  1.35it/s] 
1|961|Loss: 4.973443508148193:  15%|█▍        | 961/6470 [15:00<1:05:36,  1.40it/s]
1|962|Loss: 4.445003509521484:  15%|█▍        | 961/6470 [15:00<1:05:36,  1.40it/s]
1|962|Loss: 4.445003509521484:  15%|█▍        | 962/6470 [15:01<1:03:33,  1.44it/s]
1|963|Loss: 4.657383918762207:  15%|█▍        | 962/6470 [15:01<1:03:33,  1.44it/s]
1|963|Loss: 4.657383918762207:  15%|█▍        | 963/6470 [15:01<1:04:37,  1.42it/s]
1|964|Loss: 5.10520076751709:  15%|█▍        | 963/6470 [15:02<1:04:37,  1.42it/s] 
1|964|Loss: 5.10520076751709:  15%|█▍        | 964/6470 [15:02<1:06:02,  1.39it/s]
1|965|Loss: 5.571455478668213:  15%|█▍        | 964/6470 [15:03<1:06:02,  1.39it/s]
1|965|Loss: 5.571455478668213:  15%|█▍        | 965/6470 [15:03<1:09:34,  1.32it/s]
1|966|Loss: 5.542193412780762:  15%|█▍        | 965/6470 [15:03<1:09:34,  1.32it/s]
1|966|Loss: 5.542193412780762:  15%|█▍        | 966/6470 [15:04<1:11:18,  1.29it/s]
1|967|Loss: 5.24943208694458:  15%|█▍        | 966/6470 [15:04<1:11:18,  1.29it/s] 
1|967|Loss: 5.24943208694458:  15%|█▍        | 967/6470 [15:05<1:13:29,  1.25it/s]
1|968|Loss: 4.904507160186768:  15%|█▍        | 967/6470 [15:05<1:13:29,  1.25it/s]
1|968|Loss: 4.904507160186768:  15%|█▍        | 968/6470 [15:05<1:07:53,  1.35it/s]
1|969|Loss: 4.797266483306885:  15%|█▍        | 968/6470 [15:06<1:07:53,  1.35it/s]
1|969|Loss: 4.797266483306885:  15%|█▍        | 969/6470 [15:06<1:04:51,  1.41it/s]
1|970|Loss: 5.296091556549072:  15%|█▍        | 969/6470 [15:06<1:04:51,  1.41it/s]
1|970|Loss: 5.296091556549072:  15%|█▍        | 970/6470 [15:07<1:05:52,  1.39it/s]
1|971|Loss: 4.7599778175354:  15%|█▍        | 970/6470 [15:07<1:05:52,  1.39it/s]  
1|971|Loss: 4.7599778175354:  15%|█▌        | 971/6470 [15:07<1:07:42,  1.35it/s]
1|972|Loss: 5.120316982269287:  15%|█▌        | 971/6470 [15:08<1:07:42,  1.35it/s]
1|972|Loss: 5.120316982269287:  15%|█▌        | 972/6470 [15:08<1:10:40,  1.30it/s]
1|973|Loss: 5.117179870605469:  15%|█▌        | 972/6470 [15:09<1:10:40,  1.30it/s]
1|973|Loss: 5.117179870605469:  15%|█▌        | 973/6470 [15:09<1:09:00,  1.33it/s]
1|974|Loss: 4.948509216308594:  15%|█▌        | 973/6470 [15:09<1:09:00,  1.33it/s]
1|974|Loss: 4.948509216308594:  15%|█▌        | 974/6470 [15:10<1:10:48,  1.29it/s]
1|975|Loss: 4.709219455718994:  15%|█▌        | 974/6470 [15:10<1:10:48,  1.29it/s]
1|975|Loss: 4.709219455718994:  15%|█▌        | 975/6470 [15:10<1:09:56,  1.31it/s]
1|976|Loss: 3.6869707107543945:  15%|█▌        | 975/6470 [15:11<1:09:56,  1.31it/s]
1|976|Loss: 3.6869707107543945:  15%|█▌        | 976/6470 [15:11<1:06:01,  1.39it/s]
1|977|Loss: 5.202908039093018:  15%|█▌        | 976/6470 [15:11<1:06:01,  1.39it/s] 
1|977|Loss: 5.202908039093018:  15%|█▌        | 977/6470 [15:12<1:04:15,  1.42it/s]
1|978|Loss: 5.087667942047119:  15%|█▌        | 977/6470 [15:12<1:04:15,  1.42it/s]
1|978|Loss: 5.087667942047119:  15%|█▌        | 978/6470 [15:12<1:04:58,  1.41it/s]
1|979|Loss: 5.212259769439697:  15%|█▌        | 978/6470 [15:13<1:04:58,  1.41it/s]
1|979|Loss: 5.212259769439697:  15%|█▌        | 979/6470 [15:13<1:07:00,  1.37it/s]
1|980|Loss: 5.166768550872803:  15%|█▌        | 979/6470 [15:14<1:07:00,  1.37it/s]
1|980|Loss: 5.166768550872803:  15%|█▌        | 980/6470 [15:14<1:07:41,  1.35it/s]
1|981|Loss: 5.420701026916504:  15%|█▌        | 980/6470 [15:14<1:07:41,  1.35it/s]
1|981|Loss: 5.420701026916504:  15%|█▌        | 981/6470 [15:15<1:06:21,  1.38it/s]
1|982|Loss: 4.725162506103516:  15%|█▌        | 981/6470 [15:15<1:06:21,  1.38it/s]
1|982|Loss: 4.725162506103516:  15%|█▌        | 982/6470 [15:15<1:01:12,  1.49it/s]
1|983|Loss: 5.164677143096924:  15%|█▌        | 982/6470 [15:16<1:01:12,  1.49it/s]
1|983|Loss: 5.164677143096924:  15%|█▌        | 983/6470 [15:16<1:05:47,  1.39it/s]
1|984|Loss: 5.52095365524292:  15%|█▌        | 983/6470 [15:17<1:05:47,  1.39it/s] 
1|984|Loss: 5.52095365524292:  15%|█▌        | 984/6470 [15:17<1:13:37,  1.24it/s]
1|985|Loss: 4.87300968170166:  15%|█▌        | 984/6470 [15:18<1:13:37,  1.24it/s]
1|985|Loss: 4.87300968170166:  15%|█▌        | 985/6470 [15:18<1:11:14,  1.28it/s]
1|986|Loss: 4.9674553871154785:  15%|█▌        | 985/6470 [15:18<1:11:14,  1.28it/s]
1|986|Loss: 4.9674553871154785:  15%|█▌        | 986/6470 [15:19<1:09:40,  1.31it/s]
1|987|Loss: 5.588778495788574:  15%|█▌        | 986/6470 [15:19<1:09:40,  1.31it/s] 
1|987|Loss: 5.588778495788574:  15%|█▌        | 987/6470 [15:19<1:10:24,  1.30it/s]
1|988|Loss: 5.25922155380249:  15%|█▌        | 987/6470 [15:20<1:10:24,  1.30it/s] 
1|988|Loss: 5.25922155380249:  15%|█▌        | 988/6470 [15:20<1:11:49,  1.27it/s]
1|989|Loss: 4.701908111572266:  15%|█▌        | 988/6470 [15:21<1:11:49,  1.27it/s]
1|989|Loss: 4.701908111572266:  15%|█▌        | 989/6470 [15:21<1:10:07,  1.30it/s]
1|990|Loss: 5.528533935546875:  15%|█▌        | 989/6470 [15:21<1:10:07,  1.30it/s]
1|990|Loss: 5.528533935546875:  15%|█▌        | 990/6470 [15:22<1:10:23,  1.30it/s]
1|991|Loss: 4.303916931152344:  15%|█▌        | 990/6470 [15:22<1:10:23,  1.30it/s]
1|991|Loss: 4.303916931152344:  15%|█▌        | 991/6470 [15:22<1:06:48,  1.37it/s]
1|992|Loss: 4.827807426452637:  15%|█▌        | 991/6470 [15:23<1:06:48,  1.37it/s]
1|992|Loss: 4.827807426452637:  15%|█▌        | 992/6470 [15:23<1:05:16,  1.40it/s]
1|993|Loss: 5.349340915679932:  15%|█▌        | 992/6470 [15:23<1:05:16,  1.40it/s]
1|993|Loss: 5.349340915679932:  15%|█▌        | 993/6470 [15:24<1:06:27,  1.37it/s]
1|994|Loss: 4.929642200469971:  15%|█▌        | 993/6470 [15:24<1:06:27,  1.37it/s]
1|994|Loss: 4.929642200469971:  15%|█▌        | 994/6470 [15:24<1:05:54,  1.38it/s]
1|995|Loss: 4.88464879989624:  15%|█▌        | 994/6470 [15:25<1:05:54,  1.38it/s] 
1|995|Loss: 4.88464879989624:  15%|█▌        | 995/6470 [15:25<1:02:53,  1.45it/s]
1|996|Loss: 5.445326805114746:  15%|█▌        | 995/6470 [15:25<1:02:53,  1.45it/s]
1|996|Loss: 5.445326805114746:  15%|█▌        | 996/6470 [15:26<1:03:46,  1.43it/s]
1|997|Loss: 5.305709362030029:  15%|█▌        | 996/6470 [15:26<1:03:46,  1.43it/s]
1|997|Loss: 5.305709362030029:  15%|█▌        | 997/6470 [15:27<1:09:07,  1.32it/s]
1|998|Loss: 4.44908332824707:  15%|█▌        | 997/6470 [15:27<1:09:07,  1.32it/s] 
1|998|Loss: 4.44908332824707:  15%|█▌        | 998/6470 [15:27<1:05:06,  1.40it/s]
1|999|Loss: 5.241370677947998:  15%|█▌        | 998/6470 [15:28<1:05:06,  1.40it/s]
1|999|Loss: 5.241370677947998:  15%|█▌        | 999/6470 [15:28<1:02:53,  1.45it/s]
1|1000|Loss: 4.620686054229736:  15%|█▌        | 999/6470 [15:28<1:02:53,  1.45it/s]
1|1000|Loss: 4.620686054229736:  15%|█▌        | 1000/6470 [15:29<1:03:08,  1.44it/s]
1|1001|Loss: 5.278007507324219:  15%|█▌        | 1000/6470 [15:29<1:03:08,  1.44it/s]
1|1001|Loss: 5.278007507324219:  15%|█▌        | 1001/6470 [15:29<1:07:46,  1.34it/s]
1|1002|Loss: 4.990810394287109:  15%|█▌        | 1001/6470 [15:30<1:07:46,  1.34it/s]
1|1002|Loss: 4.990810394287109:  15%|█▌        | 1002/6470 [15:30<1:07:26,  1.35it/s]
1|1003|Loss: 4.644108772277832:  15%|█▌        | 1002/6470 [15:31<1:07:26,  1.35it/s]
1|1003|Loss: 4.644108772277832:  16%|█▌        | 1003/6470 [15:31<1:03:16,  1.44it/s]
1|1004|Loss: 4.60257625579834:  16%|█▌        | 1003/6470 [15:31<1:03:16,  1.44it/s] 
1|1004|Loss: 4.60257625579834:  16%|█▌        | 1004/6470 [15:31<1:01:48,  1.47it/s]
1|1005|Loss: 3.8800978660583496:  16%|█▌        | 1004/6470 [15:32<1:01:48,  1.47it/s]
1|1005|Loss: 3.8800978660583496:  16%|█▌        | 1005/6470 [15:32<57:19,  1.59it/s]  
1|1006|Loss: 5.332036018371582:  16%|█▌        | 1005/6470 [15:32<57:19,  1.59it/s] 
1|1006|Loss: 5.332036018371582:  16%|█▌        | 1006/6470 [15:33<58:53,  1.55it/s]
1|1007|Loss: 4.660665035247803:  16%|█▌        | 1006/6470 [15:33<58:53,  1.55it/s]
1|1007|Loss: 4.660665035247803:  16%|█▌        | 1007/6470 [15:33<1:00:07,  1.51it/s]
1|1008|Loss: 4.893197536468506:  16%|█▌        | 1007/6470 [15:34<1:00:07,  1.51it/s]
1|1008|Loss: 4.893197536468506:  16%|█▌        | 1008/6470 [15:34<58:59,  1.54it/s]  
1|1009|Loss: 5.363673210144043:  16%|█▌        | 1008/6470 [15:34<58:59,  1.54it/s]
1|1009|Loss: 5.363673210144043:  16%|█▌        | 1009/6470 [15:35<1:04:28,  1.41it/s]
1|1010|Loss: 4.375876426696777:  16%|█▌        | 1009/6470 [15:35<1:04:28,  1.41it/s]
1|1010|Loss: 4.375876426696777:  16%|█▌        | 1010/6470 [15:35<1:03:23,  1.44it/s]
1|1011|Loss: 5.623371124267578:  16%|█▌        | 1010/6470 [15:36<1:03:23,  1.44it/s]
1|1011|Loss: 5.623371124267578:  16%|█▌        | 1011/6470 [15:36<1:09:07,  1.32it/s]
1|1012|Loss: 5.070591449737549:  16%|█▌        | 1011/6470 [15:37<1:09:07,  1.32it/s]
1|1012|Loss: 5.070591449737549:  16%|█▌        | 1012/6470 [15:37<1:08:10,  1.33it/s]
1|1013|Loss: 5.139857769012451:  16%|█▌        | 1012/6470 [15:37<1:08:10,  1.33it/s]
1|1013|Loss: 5.139857769012451:  16%|█▌        | 1013/6470 [15:38<1:08:57,  1.32it/s]
1|1014|Loss: 5.0182719230651855:  16%|█▌        | 1013/6470 [15:38<1:08:57,  1.32it/s]
1|1014|Loss: 5.0182719230651855:  16%|█▌        | 1014/6470 [15:39<1:07:15,  1.35it/s]
1|1015|Loss: 5.476492881774902:  16%|█▌        | 1014/6470 [15:39<1:07:15,  1.35it/s] 
1|1015|Loss: 5.476492881774902:  16%|█▌        | 1015/6470 [15:39<1:07:13,  1.35it/s]
1|1016|Loss: 6.160192489624023:  16%|█▌        | 1015/6470 [15:40<1:07:13,  1.35it/s]
1|1016|Loss: 6.160192489624023:  16%|█▌        | 1016/6470 [15:40<1:10:34,  1.29it/s]
1|1017|Loss: 4.847967147827148:  16%|█▌        | 1016/6470 [15:41<1:10:34,  1.29it/s]
1|1017|Loss: 4.847967147827148:  16%|█▌        | 1017/6470 [15:41<1:15:44,  1.20it/s]
1|1018|Loss: 5.5506205558776855:  16%|█▌        | 1017/6470 [15:42<1:15:44,  1.20it/s]
1|1018|Loss: 5.5506205558776855:  16%|█▌        | 1018/6470 [15:42<1:17:02,  1.18it/s]
1|1019|Loss: 5.016202926635742:  16%|█▌        | 1018/6470 [15:42<1:17:02,  1.18it/s] 
1|1019|Loss: 5.016202926635742:  16%|█▌        | 1019/6470 [15:43<1:14:23,  1.22it/s]
1|1020|Loss: 5.062036991119385:  16%|█▌        | 1019/6470 [15:43<1:14:23,  1.22it/s]
1|1020|Loss: 5.062036991119385:  16%|█▌        | 1020/6470 [15:43<1:09:29,  1.31it/s]
1|1021|Loss: 5.136591911315918:  16%|█▌        | 1020/6470 [15:44<1:09:29,  1.31it/s]
1|1021|Loss: 5.136591911315918:  16%|█▌        | 1021/6470 [15:44<1:06:52,  1.36it/s]
1|1022|Loss: 5.325132846832275:  16%|█▌        | 1021/6470 [15:45<1:06:52,  1.36it/s]
1|1022|Loss: 5.325132846832275:  16%|█▌        | 1022/6470 [15:45<1:09:22,  1.31it/s]
1|1023|Loss: 5.267472267150879:  16%|█▌        | 1022/6470 [15:45<1:09:22,  1.31it/s]
1|1023|Loss: 5.267472267150879:  16%|█▌        | 1023/6470 [15:46<1:08:30,  1.32it/s]
1|1024|Loss: 4.851212024688721:  16%|█▌        | 1023/6470 [15:46<1:08:30,  1.32it/s]
1|1024|Loss: 4.851212024688721:  16%|█▌        | 1024/6470 [15:46<1:08:07,  1.33it/s]
1|1025|Loss: 4.970585346221924:  16%|█▌        | 1024/6470 [15:47<1:08:07,  1.33it/s]
1|1025|Loss: 4.970585346221924:  16%|█▌        | 1025/6470 [15:47<1:08:02,  1.33it/s]
1|1026|Loss: 4.994704723358154:  16%|█▌        | 1025/6470 [15:48<1:08:02,  1.33it/s]
1|1026|Loss: 4.994704723358154:  16%|█▌        | 1026/6470 [15:48<1:07:14,  1.35it/s]
1|1027|Loss: 4.860224723815918:  16%|█▌        | 1026/6470 [15:48<1:07:14,  1.35it/s]
1|1027|Loss: 4.860224723815918:  16%|█▌        | 1027/6470 [15:49<1:06:03,  1.37it/s]
1|1028|Loss: 5.15086555480957:  16%|█▌        | 1027/6470 [15:49<1:06:03,  1.37it/s] 
1|1028|Loss: 5.15086555480957:  16%|█▌        | 1028/6470 [15:49<1:10:42,  1.28it/s]
1|1029|Loss: 5.094442367553711:  16%|█▌        | 1028/6470 [15:50<1:10:42,  1.28it/s]
1|1029|Loss: 5.094442367553711:  16%|█▌        | 1029/6470 [15:50<1:08:27,  1.32it/s]
1|1030|Loss: 4.735212326049805:  16%|█▌        | 1029/6470 [15:50<1:08:27,  1.32it/s]
1|1030|Loss: 4.735212326049805:  16%|█▌        | 1030/6470 [15:51<1:03:55,  1.42it/s]
1|1031|Loss: 4.767551422119141:  16%|█▌        | 1030/6470 [15:51<1:03:55,  1.42it/s]
1|1031|Loss: 4.767551422119141:  16%|█▌        | 1031/6470 [15:51<1:01:45,  1.47it/s]
1|1032|Loss: 5.346795082092285:  16%|█▌        | 1031/6470 [15:52<1:01:45,  1.47it/s]
1|1032|Loss: 5.346795082092285:  16%|█▌        | 1032/6470 [15:52<1:07:43,  1.34it/s]
1|1033|Loss: 5.202760219573975:  16%|█▌        | 1032/6470 [15:53<1:07:43,  1.34it/s]
1|1033|Loss: 5.202760219573975:  16%|█▌        | 1033/6470 [15:53<1:10:18,  1.29it/s]
1|1034|Loss: 5.058894157409668:  16%|█▌        | 1033/6470 [15:53<1:10:18,  1.29it/s]
1|1034|Loss: 5.058894157409668:  16%|█▌        | 1034/6470 [15:54<1:07:29,  1.34it/s]
1|1035|Loss: 5.189888000488281:  16%|█▌        | 1034/6470 [15:54<1:07:29,  1.34it/s]
1|1035|Loss: 5.189888000488281:  16%|█▌        | 1035/6470 [15:55<1:09:21,  1.31it/s]
1|1036|Loss: 4.960878372192383:  16%|█▌        | 1035/6470 [15:55<1:09:21,  1.31it/s]
1|1036|Loss: 4.960878372192383:  16%|█▌        | 1036/6470 [15:55<1:10:24,  1.29it/s]
1|1037|Loss: 5.09475040435791:  16%|█▌        | 1036/6470 [15:56<1:10:24,  1.29it/s] 
1|1037|Loss: 5.09475040435791:  16%|█▌        | 1037/6470 [15:56<1:08:17,  1.33it/s]
1|1038|Loss: 5.040708065032959:  16%|█▌        | 1037/6470 [15:57<1:08:17,  1.33it/s]
1|1038|Loss: 5.040708065032959:  16%|█▌        | 1038/6470 [15:57<1:10:08,  1.29it/s]
1|1039|Loss: 4.731176853179932:  16%|█▌        | 1038/6470 [15:57<1:10:08,  1.29it/s]
1|1039|Loss: 4.731176853179932:  16%|█▌        | 1039/6470 [15:58<1:07:30,  1.34it/s]
1|1040|Loss: 5.446556091308594:  16%|█▌        | 1039/6470 [15:58<1:07:30,  1.34it/s]
1|1040|Loss: 5.446556091308594:  16%|█▌        | 1040/6470 [15:58<1:10:22,  1.29it/s]
1|1041|Loss: 5.554371356964111:  16%|█▌        | 1040/6470 [15:59<1:10:22,  1.29it/s]
1|1041|Loss: 5.554371356964111:  16%|█▌        | 1041/6470 [15:59<1:11:25,  1.27it/s]
1|1042|Loss: 5.4407243728637695:  16%|█▌        | 1041/6470 [16:00<1:11:25,  1.27it/s]
1|1042|Loss: 5.4407243728637695:  16%|█▌        | 1042/6470 [16:00<1:11:06,  1.27it/s]
1|1043|Loss: 5.520183563232422:  16%|█▌        | 1042/6470 [16:01<1:11:06,  1.27it/s] 
1|1043|Loss: 5.520183563232422:  16%|█▌        | 1043/6470 [16:01<1:14:33,  1.21it/s]
1|1044|Loss: 4.723823547363281:  16%|█▌        | 1043/6470 [16:01<1:14:33,  1.21it/s]
1|1044|Loss: 4.723823547363281:  16%|█▌        | 1044/6470 [16:02<1:11:52,  1.26it/s]
1|1045|Loss: 4.9307780265808105:  16%|█▌        | 1044/6470 [16:02<1:11:52,  1.26it/s]
1|1045|Loss: 4.9307780265808105:  16%|█▌        | 1045/6470 [16:03<1:16:31,  1.18it/s]
1|1046|Loss: 5.07102632522583:  16%|█▌        | 1045/6470 [16:03<1:16:31,  1.18it/s]  
1|1046|Loss: 5.07102632522583:  16%|█▌        | 1046/6470 [16:03<1:15:25,  1.20it/s]
1|1047|Loss: 5.02051305770874:  16%|█▌        | 1046/6470 [16:04<1:15:25,  1.20it/s]
1|1047|Loss: 5.02051305770874:  16%|█▌        | 1047/6470 [16:04<1:15:45,  1.19it/s]
1|1048|Loss: 5.407169342041016:  16%|█▌        | 1047/6470 [16:05<1:15:45,  1.19it/s]
1|1048|Loss: 5.407169342041016:  16%|█▌        | 1048/6470 [16:05<1:14:16,  1.22it/s]
1|1049|Loss: 5.215499401092529:  16%|█▌        | 1048/6470 [16:05<1:14:16,  1.22it/s]
1|1049|Loss: 5.215499401092529:  16%|█▌        | 1049/6470 [16:06<1:10:42,  1.28it/s]
1|1050|Loss: 4.862400531768799:  16%|█▌        | 1049/6470 [16:06<1:10:42,  1.28it/s]
1|1050|Loss: 4.862400531768799:  16%|█▌        | 1050/6470 [16:07<1:09:34,  1.30it/s]
1|1051|Loss: 5.466357707977295:  16%|█▌        | 1050/6470 [16:07<1:09:34,  1.30it/s]
1|1051|Loss: 5.466357707977295:  16%|█▌        | 1051/6470 [16:08<1:15:25,  1.20it/s]
1|1052|Loss: 5.10164737701416:  16%|█▌        | 1051/6470 [16:08<1:15:25,  1.20it/s] 
1|1052|Loss: 5.10164737701416:  16%|█▋        | 1052/6470 [16:08<1:14:51,  1.21it/s]
1|1053|Loss: 4.448996543884277:  16%|█▋        | 1052/6470 [16:09<1:14:51,  1.21it/s]
1|1053|Loss: 4.448996543884277:  16%|█▋        | 1053/6470 [16:09<1:06:49,  1.35it/s]
1|1054|Loss: 5.332292556762695:  16%|█▋        | 1053/6470 [16:09<1:06:49,  1.35it/s]
1|1054|Loss: 5.332292556762695:  16%|█▋        | 1054/6470 [16:10<1:06:18,  1.36it/s]
1|1055|Loss: 5.060762405395508:  16%|█▋        | 1054/6470 [16:10<1:06:18,  1.36it/s]
1|1055|Loss: 5.060762405395508:  16%|█▋        | 1055/6470 [16:10<1:05:22,  1.38it/s]
1|1056|Loss: 5.000399112701416:  16%|█▋        | 1055/6470 [16:11<1:05:22,  1.38it/s]
1|1056|Loss: 5.000399112701416:  16%|█▋        | 1056/6470 [16:11<1:04:39,  1.40it/s]
1|1057|Loss: 3.8955349922180176:  16%|█▋        | 1056/6470 [16:11<1:04:39,  1.40it/s]
1|1057|Loss: 3.8955349922180176:  16%|█▋        | 1057/6470 [16:12<1:00:07,  1.50it/s]
1|1058|Loss: 4.852310657501221:  16%|█▋        | 1057/6470 [16:12<1:00:07,  1.50it/s] 
1|1058|Loss: 4.852310657501221:  16%|█▋        | 1058/6470 [16:12<58:49,  1.53it/s]  
1|1059|Loss: 5.0359392166137695:  16%|█▋        | 1058/6470 [16:13<58:49,  1.53it/s]
1|1059|Loss: 5.0359392166137695:  16%|█▋        | 1059/6470 [16:13<59:18,  1.52it/s]
1|1060|Loss: 5.067965984344482:  16%|█▋        | 1059/6470 [16:13<59:18,  1.52it/s] 
1|1060|Loss: 5.067965984344482:  16%|█▋        | 1060/6470 [16:14<1:00:14,  1.50it/s]
1|1061|Loss: 5.373074054718018:  16%|█▋        | 1060/6470 [16:14<1:00:14,  1.50it/s]
1|1061|Loss: 5.373074054718018:  16%|█▋        | 1061/6470 [16:14<1:05:39,  1.37it/s]
1|1062|Loss: 5.221950531005859:  16%|█▋        | 1061/6470 [16:15<1:05:39,  1.37it/s]
1|1062|Loss: 5.221950531005859:  16%|█▋        | 1062/6470 [16:15<1:11:49,  1.26it/s]
1|1063|Loss: 5.624630451202393:  16%|█▋        | 1062/6470 [16:16<1:11:49,  1.26it/s]
1|1063|Loss: 5.624630451202393:  16%|█▋        | 1063/6470 [16:16<1:16:36,  1.18it/s]
1|1064|Loss: 4.797082901000977:  16%|█▋        | 1063/6470 [16:17<1:16:36,  1.18it/s]
1|1064|Loss: 4.797082901000977:  16%|█▋        | 1064/6470 [16:17<1:11:20,  1.26it/s]
1|1065|Loss: 4.937548637390137:  16%|█▋        | 1064/6470 [16:17<1:11:20,  1.26it/s]
1|1065|Loss: 4.937548637390137:  16%|█▋        | 1065/6470 [16:18<1:07:25,  1.34it/s]
1|1066|Loss: 5.172311782836914:  16%|█▋        | 1065/6470 [16:18<1:07:25,  1.34it/s]
1|1066|Loss: 5.172311782836914:  16%|█▋        | 1066/6470 [16:18<1:06:05,  1.36it/s]
1|1067|Loss: 5.300012111663818:  16%|█▋        | 1066/6470 [16:19<1:06:05,  1.36it/s]
1|1067|Loss: 5.300012111663818:  16%|█▋        | 1067/6470 [16:19<1:11:27,  1.26it/s]
1|1068|Loss: 5.264061450958252:  16%|█▋        | 1067/6470 [16:20<1:11:27,  1.26it/s]
1|1068|Loss: 5.264061450958252:  17%|█▋        | 1068/6470 [16:20<1:07:37,  1.33it/s]
1|1069|Loss: 4.647559642791748:  17%|█▋        | 1068/6470 [16:20<1:07:37,  1.33it/s]
1|1069|Loss: 4.647559642791748:  17%|█▋        | 1069/6470 [16:20<1:03:47,  1.41it/s]
1|1070|Loss: 5.023626327514648:  17%|█▋        | 1069/6470 [16:21<1:03:47,  1.41it/s]
1|1070|Loss: 5.023626327514648:  17%|█▋        | 1070/6470 [16:21<1:03:16,  1.42it/s]
1|1071|Loss: 5.173550128936768:  17%|█▋        | 1070/6470 [16:22<1:03:16,  1.42it/s]
1|1071|Loss: 5.173550128936768:  17%|█▋        | 1071/6470 [16:22<1:00:58,  1.48it/s]
1|1072|Loss: 5.094869136810303:  17%|█▋        | 1071/6470 [16:22<1:00:58,  1.48it/s]
1|1072|Loss: 5.094869136810303:  17%|█▋        | 1072/6470 [16:22<1:00:56,  1.48it/s]
1|1073|Loss: 5.441095352172852:  17%|█▋        | 1072/6470 [16:23<1:00:56,  1.48it/s]
1|1073|Loss: 5.441095352172852:  17%|█▋        | 1073/6470 [16:23<1:02:59,  1.43it/s]
1|1074|Loss: 4.921294689178467:  17%|█▋        | 1073/6470 [16:24<1:02:59,  1.43it/s]
1|1074|Loss: 4.921294689178467:  17%|█▋        | 1074/6470 [16:24<1:01:49,  1.45it/s]
1|1075|Loss: 5.283716678619385:  17%|█▋        | 1074/6470 [16:24<1:01:49,  1.45it/s]
1|1075|Loss: 5.283716678619385:  17%|█▋        | 1075/6470 [16:25<1:02:09,  1.45it/s]
1|1076|Loss: 5.616933822631836:  17%|█▋        | 1075/6470 [16:25<1:02:09,  1.45it/s]
1|1076|Loss: 5.616933822631836:  17%|█▋        | 1076/6470 [16:25<1:04:54,  1.38it/s]
1|1077|Loss: 5.556154251098633:  17%|█▋        | 1076/6470 [16:26<1:04:54,  1.38it/s]
1|1077|Loss: 5.556154251098633:  17%|█▋        | 1077/6470 [16:26<1:05:17,  1.38it/s]
1|1078|Loss: 5.920745372772217:  17%|█▋        | 1077/6470 [16:27<1:05:17,  1.38it/s]
1|1078|Loss: 5.920745372772217:  17%|█▋        | 1078/6470 [16:27<1:19:13,  1.13it/s]
1|1079|Loss: 3.981187582015991:  17%|█▋        | 1078/6470 [16:28<1:19:13,  1.13it/s]
1|1079|Loss: 3.981187582015991:  17%|█▋        | 1079/6470 [16:28<1:15:16,  1.19it/s]
1|1080|Loss: 5.218023300170898:  17%|█▋        | 1079/6470 [16:28<1:15:16,  1.19it/s]
1|1080|Loss: 5.218023300170898:  17%|█▋        | 1080/6470 [16:29<1:13:19,  1.23it/s]
1|1081|Loss: 5.051525592803955:  17%|█▋        | 1080/6470 [16:29<1:13:19,  1.23it/s]
1|1081|Loss: 5.051525592803955:  17%|█▋        | 1081/6470 [16:30<1:10:29,  1.27it/s]
1|1082|Loss: 5.035562992095947:  17%|█▋        | 1081/6470 [16:30<1:10:29,  1.27it/s]
1|1082|Loss: 5.035562992095947:  17%|█▋        | 1082/6470 [16:30<1:13:08,  1.23it/s]
1|1083|Loss: 5.223138809204102:  17%|█▋        | 1082/6470 [16:31<1:13:08,  1.23it/s]
1|1083|Loss: 5.223138809204102:  17%|█▋        | 1083/6470 [16:31<1:13:43,  1.22it/s]
1|1084|Loss: 5.304084777832031:  17%|█▋        | 1083/6470 [16:32<1:13:43,  1.22it/s]
1|1084|Loss: 5.304084777832031:  17%|█▋        | 1084/6470 [16:32<1:12:20,  1.24it/s]
1|1085|Loss: 5.135820388793945:  17%|█▋        | 1084/6470 [16:32<1:12:20,  1.24it/s]
1|1085|Loss: 5.135820388793945:  17%|█▋        | 1085/6470 [16:33<1:09:50,  1.29it/s]
1|1086|Loss: 4.613682746887207:  17%|█▋        | 1085/6470 [16:33<1:09:50,  1.29it/s]
1|1086|Loss: 4.613682746887207:  17%|█▋        | 1086/6470 [16:33<1:05:26,  1.37it/s]
1|1087|Loss: 5.659032821655273:  17%|█▋        | 1086/6470 [16:34<1:05:26,  1.37it/s]
1|1087|Loss: 5.659032821655273:  17%|█▋        | 1087/6470 [16:34<1:06:10,  1.36it/s]
1|1088|Loss: 5.173786640167236:  17%|█▋        | 1087/6470 [16:35<1:06:10,  1.36it/s]
1|1088|Loss: 5.173786640167236:  17%|█▋        | 1088/6470 [16:35<1:07:34,  1.33it/s]
1|1089|Loss: 4.994651794433594:  17%|█▋        | 1088/6470 [16:35<1:07:34,  1.33it/s]
1|1089|Loss: 4.994651794433594:  17%|█▋        | 1089/6470 [16:36<1:02:54,  1.43it/s]
1|1090|Loss: 4.847129821777344:  17%|█▋        | 1089/6470 [16:36<1:02:54,  1.43it/s]
1|1090|Loss: 4.847129821777344:  17%|█▋        | 1090/6470 [16:36<1:02:46,  1.43it/s]
1|1091|Loss: 5.679340362548828:  17%|█▋        | 1090/6470 [16:37<1:02:46,  1.43it/s]
1|1091|Loss: 5.679340362548828:  17%|█▋        | 1091/6470 [16:37<1:08:02,  1.32it/s]
1|1092|Loss: 4.916650772094727:  17%|█▋        | 1091/6470 [16:38<1:08:02,  1.32it/s]
1|1092|Loss: 4.916650772094727:  17%|█▋        | 1092/6470 [16:38<1:09:40,  1.29it/s]
1|1093|Loss: 4.246484756469727:  17%|█▋        | 1092/6470 [16:38<1:09:40,  1.29it/s]
1|1093|Loss: 4.246484756469727:  17%|█▋        | 1093/6470 [16:39<1:04:47,  1.38it/s]
1|1094|Loss: 4.903238296508789:  17%|█▋        | 1093/6470 [16:39<1:04:47,  1.38it/s]
1|1094|Loss: 4.903238296508789:  17%|█▋        | 1094/6470 [16:39<1:00:50,  1.47it/s]
1|1095|Loss: 3.8768210411071777:  17%|█▋        | 1094/6470 [16:39<1:00:50,  1.47it/s]
1|1095|Loss: 3.8768210411071777:  17%|█▋        | 1095/6470 [16:40<56:02,  1.60it/s]  
1|1096|Loss: 5.3872809410095215:  17%|█▋        | 1095/6470 [16:40<56:02,  1.60it/s]
1|1096|Loss: 5.3872809410095215:  17%|█▋        | 1096/6470 [16:40<1:02:03,  1.44it/s]
1|1097|Loss: 5.52497673034668:  17%|█▋        | 1096/6470 [16:41<1:02:03,  1.44it/s]  
1|1097|Loss: 5.52497673034668:  17%|█▋        | 1097/6470 [16:41<1:07:41,  1.32it/s]
1|1098|Loss: 4.507845878601074:  17%|█▋        | 1097/6470 [16:42<1:07:41,  1.32it/s]
1|1098|Loss: 4.507845878601074:  17%|█▋        | 1098/6470 [16:42<1:04:03,  1.40it/s]
1|1099|Loss: 5.415037631988525:  17%|█▋        | 1098/6470 [16:42<1:04:03,  1.40it/s]
1|1099|Loss: 5.415037631988525:  17%|█▋        | 1099/6470 [16:43<1:02:56,  1.42it/s]
1|1100|Loss: 4.322084426879883:  17%|█▋        | 1099/6470 [16:43<1:02:56,  1.42it/s]
1|1100|Loss: 4.322084426879883:  17%|█▋        | 1100/6470 [16:43<1:03:22,  1.41it/s]
1|1101|Loss: 4.946712970733643:  17%|█▋        | 1100/6470 [16:44<1:03:22,  1.41it/s]
1|1101|Loss: 4.946712970733643:  17%|█▋        | 1101/6470 [16:44<1:04:27,  1.39it/s]
1|1102|Loss: 4.8217620849609375:  17%|█▋        | 1101/6470 [16:45<1:04:27,  1.39it/s]
1|1102|Loss: 4.8217620849609375:  17%|█▋        | 1102/6470 [16:45<1:04:11,  1.39it/s]
1|1103|Loss: 5.309300422668457:  17%|█▋        | 1102/6470 [16:45<1:04:11,  1.39it/s] 
1|1103|Loss: 5.309300422668457:  17%|█▋        | 1103/6470 [16:46<1:03:07,  1.42it/s]
1|1104|Loss: 5.533336639404297:  17%|█▋        | 1103/6470 [16:46<1:03:07,  1.42it/s]
1|1104|Loss: 5.533336639404297:  17%|█▋        | 1104/6470 [16:47<1:14:49,  1.20it/s]
1|1105|Loss: 4.99615478515625:  17%|█▋        | 1104/6470 [16:47<1:14:49,  1.20it/s] 
1|1105|Loss: 4.99615478515625:  17%|█▋        | 1105/6470 [16:47<1:13:14,  1.22it/s]
1|1106|Loss: 4.773468971252441:  17%|█▋        | 1105/6470 [16:48<1:13:14,  1.22it/s]
1|1106|Loss: 4.773468971252441:  17%|█▋        | 1106/6470 [16:48<1:14:45,  1.20it/s]
1|1107|Loss: 5.201586723327637:  17%|█▋        | 1106/6470 [16:49<1:14:45,  1.20it/s]
1|1107|Loss: 5.201586723327637:  17%|█▋        | 1107/6470 [16:49<1:12:44,  1.23it/s]
1|1108|Loss: 4.7036848068237305:  17%|█▋        | 1107/6470 [16:49<1:12:44,  1.23it/s]
1|1108|Loss: 4.7036848068237305:  17%|█▋        | 1108/6470 [16:50<1:08:12,  1.31it/s]
1|1109|Loss: 4.813915729522705:  17%|█▋        | 1108/6470 [16:50<1:08:12,  1.31it/s] 
1|1109|Loss: 4.813915729522705:  17%|█▋        | 1109/6470 [16:50<1:06:07,  1.35it/s]
1|1110|Loss: 5.051959991455078:  17%|█▋        | 1109/6470 [16:51<1:06:07,  1.35it/s]
1|1110|Loss: 5.051959991455078:  17%|█▋        | 1110/6470 [16:51<1:06:10,  1.35it/s]
1|1111|Loss: 5.336190223693848:  17%|█▋        | 1110/6470 [16:52<1:06:10,  1.35it/s]
1|1111|Loss: 5.336190223693848:  17%|█▋        | 1111/6470 [16:52<1:07:34,  1.32it/s]
1|1112|Loss: 5.273159980773926:  17%|█▋        | 1111/6470 [16:52<1:07:34,  1.32it/s]
1|1112|Loss: 5.273159980773926:  17%|█▋        | 1112/6470 [16:53<1:08:30,  1.30it/s]
1|1113|Loss: 5.308159351348877:  17%|█▋        | 1112/6470 [16:53<1:08:30,  1.30it/s]
1|1113|Loss: 5.308159351348877:  17%|█▋        | 1113/6470 [16:54<1:12:38,  1.23it/s]
1|1114|Loss: 5.486394882202148:  17%|█▋        | 1113/6470 [16:54<1:12:38,  1.23it/s]
1|1114|Loss: 5.486394882202148:  17%|█▋        | 1114/6470 [16:55<1:15:40,  1.18it/s]
1|1115|Loss: 4.608723163604736:  17%|█▋        | 1114/6470 [16:55<1:15:40,  1.18it/s]
1|1115|Loss: 4.608723163604736:  17%|█▋        | 1115/6470 [16:55<1:14:00,  1.21it/s]
1|1116|Loss: 4.850037097930908:  17%|█▋        | 1115/6470 [16:56<1:14:00,  1.21it/s]
1|1116|Loss: 4.850037097930908:  17%|█▋        | 1116/6470 [16:56<1:09:03,  1.29it/s]
1|1117|Loss: 5.555197238922119:  17%|█▋        | 1116/6470 [16:56<1:09:03,  1.29it/s]
1|1117|Loss: 5.555197238922119:  17%|█▋        | 1117/6470 [16:57<1:06:33,  1.34it/s]
1|1118|Loss: 4.441101551055908:  17%|█▋        | 1117/6470 [16:57<1:06:33,  1.34it/s]
1|1118|Loss: 4.441101551055908:  17%|█▋        | 1118/6470 [16:57<1:03:13,  1.41it/s]
1|1119|Loss: 4.446625232696533:  17%|█▋        | 1118/6470 [16:58<1:03:13,  1.41it/s]
1|1119|Loss: 4.446625232696533:  17%|█▋        | 1119/6470 [16:58<58:42,  1.52it/s]  
1|1120|Loss: 5.233987808227539:  17%|█▋        | 1119/6470 [16:58<58:42,  1.52it/s]
1|1120|Loss: 5.233987808227539:  17%|█▋        | 1120/6470 [16:59<1:01:16,  1.46it/s]
1|1121|Loss: 4.943981647491455:  17%|█▋        | 1120/6470 [16:59<1:01:16,  1.46it/s]
1|1121|Loss: 4.943981647491455:  17%|█▋        | 1121/6470 [16:59<1:01:41,  1.45it/s]
1|1122|Loss: 4.508963584899902:  17%|█▋        | 1121/6470 [17:00<1:01:41,  1.45it/s]
1|1122|Loss: 4.508963584899902:  17%|█▋        | 1122/6470 [17:00<59:08,  1.51it/s]  
1|1123|Loss: 4.963556289672852:  17%|█▋        | 1122/6470 [17:00<59:08,  1.51it/s]
1|1123|Loss: 4.963556289672852:  17%|█▋        | 1123/6470 [17:01<1:00:43,  1.47it/s]
1|1124|Loss: 4.789652347564697:  17%|█▋        | 1123/6470 [17:01<1:00:43,  1.47it/s]
1|1124|Loss: 4.789652347564697:  17%|█▋        | 1124/6470 [17:01<59:22,  1.50it/s]  
1|1125|Loss: 5.1682868003845215:  17%|█▋        | 1124/6470 [17:02<59:22,  1.50it/s]
1|1125|Loss: 5.1682868003845215:  17%|█▋        | 1125/6470 [17:02<1:00:51,  1.46it/s]
1|1126|Loss: 5.212246417999268:  17%|█▋        | 1125/6470 [17:03<1:00:51,  1.46it/s] 
1|1126|Loss: 5.212246417999268:  17%|█▋        | 1126/6470 [17:03<1:15:02,  1.19it/s]
1|1127|Loss: 5.28847599029541:  17%|█▋        | 1126/6470 [17:04<1:15:02,  1.19it/s] 
1|1127|Loss: 5.28847599029541:  17%|█▋        | 1127/6470 [17:04<1:16:46,  1.16it/s]
1|1128|Loss: 4.871777534484863:  17%|█▋        | 1127/6470 [17:05<1:16:46,  1.16it/s]
1|1128|Loss: 4.871777534484863:  17%|█▋        | 1128/6470 [17:05<1:13:44,  1.21it/s]
1|1129|Loss: 4.458423137664795:  17%|█▋        | 1128/6470 [17:05<1:13:44,  1.21it/s]
1|1129|Loss: 4.458423137664795:  17%|█▋        | 1129/6470 [17:06<1:10:46,  1.26it/s]
1|1130|Loss: 4.724917888641357:  17%|█▋        | 1129/6470 [17:06<1:10:46,  1.26it/s]
1|1130|Loss: 4.724917888641357:  17%|█▋        | 1130/6470 [17:06<1:09:17,  1.28it/s]
1|1131|Loss: 4.701244831085205:  17%|█▋        | 1130/6470 [17:07<1:09:17,  1.28it/s]
1|1131|Loss: 4.701244831085205:  17%|█▋        | 1131/6470 [17:07<1:07:36,  1.32it/s]
1|1132|Loss: 5.525669574737549:  17%|█▋        | 1131/6470 [17:07<1:07:36,  1.32it/s]
1|1132|Loss: 5.525669574737549:  17%|█▋        | 1132/6470 [17:08<1:09:01,  1.29it/s]
1|1133|Loss: 5.457925796508789:  17%|█▋        | 1132/6470 [17:08<1:09:01,  1.29it/s]
1|1133|Loss: 5.457925796508789:  18%|█▊        | 1133/6470 [17:09<1:10:17,  1.27it/s]
1|1134|Loss: 4.846338748931885:  18%|█▊        | 1133/6470 [17:09<1:10:17,  1.27it/s]
1|1134|Loss: 4.846338748931885:  18%|█▊        | 1134/6470 [17:10<1:15:59,  1.17it/s]
1|1135|Loss: 4.914979457855225:  18%|█▊        | 1134/6470 [17:10<1:15:59,  1.17it/s]
1|1135|Loss: 4.914979457855225:  18%|█▊        | 1135/6470 [17:10<1:13:51,  1.20it/s]
1|1136|Loss: 4.698550701141357:  18%|█▊        | 1135/6470 [17:11<1:13:51,  1.20it/s]
1|1136|Loss: 4.698550701141357:  18%|█▊        | 1136/6470 [17:11<1:10:41,  1.26it/s]
1|1137|Loss: 5.659379005432129:  18%|█▊        | 1136/6470 [17:12<1:10:41,  1.26it/s]
1|1137|Loss: 5.659379005432129:  18%|█▊        | 1137/6470 [17:12<1:09:48,  1.27it/s]
1|1138|Loss: 4.9927754402160645:  18%|█▊        | 1137/6470 [17:12<1:09:48,  1.27it/s]
1|1138|Loss: 4.9927754402160645:  18%|█▊        | 1138/6470 [17:13<1:07:41,  1.31it/s]
1|1139|Loss: 5.048996925354004:  18%|█▊        | 1138/6470 [17:14<1:07:41,  1.31it/s] 
1|1139|Loss: 5.048996925354004:  18%|█▊        | 1139/6470 [17:14<1:31:34,  1.03s/it]
1|1140|Loss: 4.632837772369385:  18%|█▊        | 1139/6470 [17:15<1:31:34,  1.03s/it]
1|1140|Loss: 4.632837772369385:  18%|█▊        | 1140/6470 [17:15<1:20:10,  1.11it/s]
1|1141|Loss: 5.01173734664917:  18%|█▊        | 1140/6470 [17:15<1:20:10,  1.11it/s] 
1|1141|Loss: 5.01173734664917:  18%|█▊        | 1141/6470 [17:16<1:13:16,  1.21it/s]
1|1142|Loss: 5.188048362731934:  18%|█▊        | 1141/6470 [17:16<1:13:16,  1.21it/s]
1|1142|Loss: 5.188048362731934:  18%|█▊        | 1142/6470 [17:16<1:08:12,  1.30it/s]
1|1143|Loss: 5.27966833114624:  18%|█▊        | 1142/6470 [17:16<1:08:12,  1.30it/s] 
1|1143|Loss: 5.27966833114624:  18%|█▊        | 1143/6470 [17:17<59:44,  1.49it/s]  
1|1144|Loss: 5.671422004699707:  18%|█▊        | 1143/6470 [17:17<59:44,  1.49it/s]
1|1144|Loss: 5.671422004699707:  18%|█▊        | 1144/6470 [17:17<1:00:10,  1.48it/s]
1|1145|Loss: 5.0997490882873535:  18%|█▊        | 1144/6470 [17:18<1:00:10,  1.48it/s]
1|1145|Loss: 5.0997490882873535:  18%|█▊        | 1145/6470 [17:18<1:00:45,  1.46it/s]
1|1146|Loss: 5.177976131439209:  18%|█▊        | 1145/6470 [17:18<1:00:45,  1.46it/s] 
1|1146|Loss: 5.177976131439209:  18%|█▊        | 1146/6470 [17:19<1:02:52,  1.41it/s]
1|1147|Loss: 5.58982515335083:  18%|█▊        | 1146/6470 [17:19<1:02:52,  1.41it/s] 
1|1147|Loss: 5.58982515335083:  18%|█▊        | 1147/6470 [17:20<1:06:31,  1.33it/s]
1|1148|Loss: 4.949589252471924:  18%|█▊        | 1147/6470 [17:20<1:06:31,  1.33it/s]
1|1148|Loss: 4.949589252471924:  18%|█▊        | 1148/6470 [17:20<1:07:27,  1.31it/s]
1|1149|Loss: 5.1838765144348145:  18%|█▊        | 1148/6470 [17:21<1:07:27,  1.31it/s]
1|1149|Loss: 5.1838765144348145:  18%|█▊        | 1149/6470 [17:21<1:07:26,  1.32it/s]
1|1150|Loss: 5.762479782104492:  18%|█▊        | 1149/6470 [17:22<1:07:26,  1.32it/s] 
1|1150|Loss: 5.762479782104492:  18%|█▊        | 1150/6470 [17:22<1:21:49,  1.08it/s]
1|1151|Loss: 5.041145324707031:  18%|█▊        | 1150/6470 [17:23<1:21:49,  1.08it/s]
1|1151|Loss: 5.041145324707031:  18%|█▊        | 1151/6470 [17:23<1:21:15,  1.09it/s]
1|1152|Loss: 5.362344264984131:  18%|█▊        | 1151/6470 [17:24<1:21:15,  1.09it/s]
1|1152|Loss: 5.362344264984131:  18%|█▊        | 1152/6470 [17:24<1:20:40,  1.10it/s]
1|1153|Loss: 4.857501983642578:  18%|█▊        | 1152/6470 [17:25<1:20:40,  1.10it/s]
1|1153|Loss: 4.857501983642578:  18%|█▊        | 1153/6470 [17:25<1:14:45,  1.19it/s]
1|1154|Loss: 4.383788108825684:  18%|█▊        | 1153/6470 [17:25<1:14:45,  1.19it/s]
1|1154|Loss: 4.383788108825684:  18%|█▊        | 1154/6470 [17:26<1:10:15,  1.26it/s]
1|1155|Loss: 5.238054275512695:  18%|█▊        | 1154/6470 [17:26<1:10:15,  1.26it/s]
1|1155|Loss: 5.238054275512695:  18%|█▊        | 1155/6470 [17:26<1:06:42,  1.33it/s]
1|1156|Loss: 4.962399482727051:  18%|█▊        | 1155/6470 [17:27<1:06:42,  1.33it/s]
1|1156|Loss: 4.962399482727051:  18%|█▊        | 1156/6470 [17:27<1:02:07,  1.43it/s]
1|1157|Loss: 4.812907695770264:  18%|█▊        | 1156/6470 [17:27<1:02:07,  1.43it/s]
1|1157|Loss: 4.812907695770264:  18%|█▊        | 1157/6470 [17:28<1:00:41,  1.46it/s]
1|1158|Loss: 5.370114803314209:  18%|█▊        | 1157/6470 [17:28<1:00:41,  1.46it/s]
1|1158|Loss: 5.370114803314209:  18%|█▊        | 1158/6470 [17:28<1:01:17,  1.44it/s]
1|1159|Loss: 5.240358829498291:  18%|█▊        | 1158/6470 [17:29<1:01:17,  1.44it/s]
1|1159|Loss: 5.240358829498291:  18%|█▊        | 1159/6470 [17:29<1:01:47,  1.43it/s]
1|1160|Loss: 4.231897830963135:  18%|█▊        | 1159/6470 [17:29<1:01:47,  1.43it/s]
1|1160|Loss: 4.231897830963135:  18%|█▊        | 1160/6470 [17:29<54:34,  1.62it/s]  
1|1161|Loss: 5.4666595458984375:  18%|█▊        | 1160/6470 [17:30<54:34,  1.62it/s]
1|1161|Loss: 5.4666595458984375:  18%|█▊        | 1161/6470 [17:30<56:58,  1.55it/s]
1|1162|Loss: 5.128293037414551:  18%|█▊        | 1161/6470 [17:31<56:58,  1.55it/s] 
1|1162|Loss: 5.128293037414551:  18%|█▊        | 1162/6470 [17:31<1:04:30,  1.37it/s]
1|1163|Loss: 4.9333271980285645:  18%|█▊        | 1162/6470 [17:31<1:04:30,  1.37it/s]
1|1163|Loss: 4.9333271980285645:  18%|█▊        | 1163/6470 [17:32<1:05:39,  1.35it/s]
1|1164|Loss: 4.341230392456055:  18%|█▊        | 1163/6470 [17:32<1:05:39,  1.35it/s] 
1|1164|Loss: 4.341230392456055:  18%|█▊        | 1164/6470 [17:33<1:05:49,  1.34it/s]
1|1165|Loss: 5.224513053894043:  18%|█▊        | 1164/6470 [17:33<1:05:49,  1.34it/s]
1|1165|Loss: 5.224513053894043:  18%|█▊        | 1165/6470 [17:33<1:04:25,  1.37it/s]
1|1166|Loss: 4.667504787445068:  18%|█▊        | 1165/6470 [17:34<1:04:25,  1.37it/s]
1|1166|Loss: 4.667504787445068:  18%|█▊        | 1166/6470 [17:34<1:01:11,  1.44it/s]
1|1167|Loss: 5.223702430725098:  18%|█▊        | 1166/6470 [17:34<1:01:11,  1.44it/s]
1|1167|Loss: 5.223702430725098:  18%|█▊        | 1167/6470 [17:35<1:04:16,  1.38it/s]
1|1168|Loss: 5.190558433532715:  18%|█▊        | 1167/6470 [17:35<1:04:16,  1.38it/s]
1|1168|Loss: 5.190558433532715:  18%|█▊        | 1168/6470 [17:35<1:07:43,  1.30it/s]
1|1169|Loss: 5.488102436065674:  18%|█▊        | 1168/6470 [17:36<1:07:43,  1.30it/s]
1|1169|Loss: 5.488102436065674:  18%|█▊        | 1169/6470 [17:36<1:11:24,  1.24it/s]
1|1170|Loss: 6.039536952972412:  18%|█▊        | 1169/6470 [17:37<1:11:24,  1.24it/s]
1|1170|Loss: 6.039536952972412:  18%|█▊        | 1170/6470 [17:38<1:32:15,  1.04s/it]
1|1171|Loss: 5.274531841278076:  18%|█▊        | 1170/6470 [17:39<1:32:15,  1.04s/it]
1|1171|Loss: 5.274531841278076:  18%|█▊        | 1171/6470 [17:39<1:31:16,  1.03s/it]
1|1172|Loss: 4.683337211608887:  18%|█▊        | 1171/6470 [17:39<1:31:16,  1.03s/it]
1|1172|Loss: 4.683337211608887:  18%|█▊        | 1172/6470 [17:40<1:22:26,  1.07it/s]
1|1173|Loss: 4.891923904418945:  18%|█▊        | 1172/6470 [17:40<1:22:26,  1.07it/s]
1|1173|Loss: 4.891923904418945:  18%|█▊        | 1173/6470 [17:40<1:16:15,  1.16it/s]
1|1174|Loss: 5.181328296661377:  18%|█▊        | 1173/6470 [17:41<1:16:15,  1.16it/s]
1|1174|Loss: 5.181328296661377:  18%|█▊        | 1174/6470 [17:41<1:13:40,  1.20it/s]
1|1175|Loss: 5.121070384979248:  18%|█▊        | 1174/6470 [17:42<1:13:40,  1.20it/s]
1|1175|Loss: 5.121070384979248:  18%|█▊        | 1175/6470 [17:42<1:12:47,  1.21it/s]
1|1176|Loss: 5.754456520080566:  18%|█▊        | 1175/6470 [17:42<1:12:47,  1.21it/s]
1|1176|Loss: 5.754456520080566:  18%|█▊        | 1176/6470 [17:43<1:16:52,  1.15it/s]
1|1177|Loss: 4.952175617218018:  18%|█▊        | 1176/6470 [17:43<1:16:52,  1.15it/s]
1|1177|Loss: 4.952175617218018:  18%|█▊        | 1177/6470 [17:44<1:13:50,  1.19it/s]
1|1178|Loss: 5.006951808929443:  18%|█▊        | 1177/6470 [17:44<1:13:50,  1.19it/s]
1|1178|Loss: 5.006951808929443:  18%|█▊        | 1178/6470 [17:44<1:11:30,  1.23it/s]
1|1179|Loss: 5.1419291496276855:  18%|█▊        | 1178/6470 [17:45<1:11:30,  1.23it/s]
1|1179|Loss: 5.1419291496276855:  18%|█▊        | 1179/6470 [17:45<1:10:09,  1.26it/s]
1|1180|Loss: 5.157105445861816:  18%|█▊        | 1179/6470 [17:46<1:10:09,  1.26it/s] 
1|1180|Loss: 5.157105445861816:  18%|█▊        | 1180/6470 [17:46<1:07:27,  1.31it/s]
1|1181|Loss: 4.214907169342041:  18%|█▊        | 1180/6470 [17:46<1:07:27,  1.31it/s]
1|1181|Loss: 4.214907169342041:  18%|█▊        | 1181/6470 [17:47<1:04:13,  1.37it/s]
1|1182|Loss: 5.113836288452148:  18%|█▊        | 1181/6470 [17:47<1:04:13,  1.37it/s]
1|1182|Loss: 5.113836288452148:  18%|█▊        | 1182/6470 [17:47<1:03:01,  1.40it/s]
1|1183|Loss: 5.2114996910095215:  18%|█▊        | 1182/6470 [17:48<1:03:01,  1.40it/s]
1|1183|Loss: 5.2114996910095215:  18%|█▊        | 1183/6470 [17:48<1:07:44,  1.30it/s]
1|1184|Loss: 4.258905410766602:  18%|█▊        | 1183/6470 [17:49<1:07:44,  1.30it/s] 
1|1184|Loss: 4.258905410766602:  18%|█▊        | 1184/6470 [17:49<1:04:01,  1.38it/s]
1|1185|Loss: 4.989101409912109:  18%|█▊        | 1184/6470 [17:49<1:04:01,  1.38it/s]
1|1185|Loss: 4.989101409912109:  18%|█▊        | 1185/6470 [17:49<59:31,  1.48it/s]  
1|1186|Loss: 5.301366806030273:  18%|█▊        | 1185/6470 [17:50<59:31,  1.48it/s]
1|1186|Loss: 5.301366806030273:  18%|█▊        | 1186/6470 [17:50<1:04:26,  1.37it/s]
1|1187|Loss: 5.430302619934082:  18%|█▊        | 1186/6470 [17:51<1:04:26,  1.37it/s]
1|1187|Loss: 5.430302619934082:  18%|█▊        | 1187/6470 [17:51<1:09:54,  1.26it/s]
1|1188|Loss: 5.14141321182251:  18%|█▊        | 1187/6470 [17:52<1:09:54,  1.26it/s] 
1|1188|Loss: 5.14141321182251:  18%|█▊        | 1188/6470 [17:52<1:08:55,  1.28it/s]
1|1189|Loss: 4.826003074645996:  18%|█▊        | 1188/6470 [17:52<1:08:55,  1.28it/s]
1|1189|Loss: 4.826003074645996:  18%|█▊        | 1189/6470 [17:53<1:06:23,  1.33it/s]
1|1190|Loss: 4.8402814865112305:  18%|█▊        | 1189/6470 [17:53<1:06:23,  1.33it/s]
1|1190|Loss: 4.8402814865112305:  18%|█▊        | 1190/6470 [17:53<1:03:32,  1.38it/s]
1|1191|Loss: 4.482444763183594:  18%|█▊        | 1190/6470 [17:54<1:03:32,  1.38it/s] 
1|1191|Loss: 4.482444763183594:  18%|█▊        | 1191/6470 [17:54<58:02,  1.52it/s]  
1|1192|Loss: 5.18630313873291:  18%|█▊        | 1191/6470 [17:54<58:02,  1.52it/s] 
1|1192|Loss: 5.18630313873291:  18%|█▊        | 1192/6470 [17:55<1:01:29,  1.43it/s]
1|1193|Loss: 4.608379364013672:  18%|█▊        | 1192/6470 [17:55<1:01:29,  1.43it/s]
1|1193|Loss: 4.608379364013672:  18%|█▊        | 1193/6470 [17:55<1:04:07,  1.37it/s]
1|1194|Loss: 5.166022300720215:  18%|█▊        | 1193/6470 [17:56<1:04:07,  1.37it/s]
1|1194|Loss: 5.166022300720215:  18%|█▊        | 1194/6470 [17:56<1:01:51,  1.42it/s]
1|1195|Loss: 4.861145973205566:  18%|█▊        | 1194/6470 [17:56<1:01:51,  1.42it/s]
1|1195|Loss: 4.861145973205566:  18%|█▊        | 1195/6470 [17:57<1:00:28,  1.45it/s]
1|1196|Loss: 4.880294322967529:  18%|█▊        | 1195/6470 [17:57<1:00:28,  1.45it/s]
1|1196|Loss: 4.880294322967529:  18%|█▊        | 1196/6470 [17:57<1:01:52,  1.42it/s]
1|1197|Loss: 4.880544662475586:  18%|█▊        | 1196/6470 [17:58<1:01:52,  1.42it/s]
1|1197|Loss: 4.880544662475586:  19%|█▊        | 1197/6470 [17:58<1:06:51,  1.31it/s]
1|1198|Loss: 4.56682014465332:  19%|█▊        | 1197/6470 [17:59<1:06:51,  1.31it/s] 
1|1198|Loss: 4.56682014465332:  19%|█▊        | 1198/6470 [17:59<1:06:10,  1.33it/s]
1|1199|Loss: 5.237998008728027:  19%|█▊        | 1198/6470 [17:59<1:06:10,  1.33it/s]
1|1199|Loss: 5.237998008728027:  19%|█▊        | 1199/6470 [18:00<1:04:00,  1.37it/s]
1|1200|Loss: 5.442763328552246:  19%|█▊        | 1199/6470 [18:00<1:04:00,  1.37it/s]
1|1200|Loss: 5.442763328552246:  19%|█▊        | 1200/6470 [18:01<1:09:33,  1.26it/s]
1|1201|Loss: 5.243850231170654:  19%|█▊        | 1200/6470 [18:01<1:09:33,  1.26it/s]
1|1201|Loss: 5.243850231170654:  19%|█▊        | 1201/6470 [18:02<1:12:30,  1.21it/s]
1|1202|Loss: 5.0861124992370605:  19%|█▊        | 1201/6470 [18:02<1:12:30,  1.21it/s]
1|1202|Loss: 5.0861124992370605:  19%|█▊        | 1202/6470 [18:02<1:15:14,  1.17it/s]
1|1203|Loss: 4.881364822387695:  19%|█▊        | 1202/6470 [18:03<1:15:14,  1.17it/s] 
1|1203|Loss: 4.881364822387695:  19%|█▊        | 1203/6470 [18:03<1:11:18,  1.23it/s]
1|1204|Loss: 4.9756927490234375:  19%|█▊        | 1203/6470 [18:04<1:11:18,  1.23it/s]
1|1204|Loss: 4.9756927490234375:  19%|█▊        | 1204/6470 [18:04<1:12:28,  1.21it/s]
1|1205|Loss: 5.333990573883057:  19%|█▊        | 1204/6470 [18:04<1:12:28,  1.21it/s] 
1|1205|Loss: 5.333990573883057:  19%|█▊        | 1205/6470 [18:05<1:14:15,  1.18it/s]
1|1206|Loss: 4.847487926483154:  19%|█▊        | 1205/6470 [18:05<1:14:15,  1.18it/s]
1|1206|Loss: 4.847487926483154:  19%|█▊        | 1206/6470 [18:06<1:13:20,  1.20it/s]
1|1207|Loss: 4.89023494720459:  19%|█▊        | 1206/6470 [18:06<1:13:20,  1.20it/s] 
1|1207|Loss: 4.89023494720459:  19%|█▊        | 1207/6470 [18:07<1:13:35,  1.19it/s]
1|1208|Loss: 4.934288024902344:  19%|█▊        | 1207/6470 [18:07<1:13:35,  1.19it/s]
1|1208|Loss: 4.934288024902344:  19%|█▊        | 1208/6470 [18:07<1:12:45,  1.21it/s]
1|1209|Loss: 4.785810470581055:  19%|█▊        | 1208/6470 [18:08<1:12:45,  1.21it/s]
1|1209|Loss: 4.785810470581055:  19%|█▊        | 1209/6470 [18:08<1:10:55,  1.24it/s]
1|1210|Loss: 4.705643653869629:  19%|█▊        | 1209/6470 [18:08<1:10:55,  1.24it/s]
1|1210|Loss: 4.705643653869629:  19%|█▊        | 1210/6470 [18:09<1:05:20,  1.34it/s]
1|1211|Loss: 4.999284267425537:  19%|█▊        | 1210/6470 [18:09<1:05:20,  1.34it/s]
1|1211|Loss: 4.999284267425537:  19%|█▊        | 1211/6470 [18:09<1:03:30,  1.38it/s]
1|1212|Loss: 4.970063209533691:  19%|█▊        | 1211/6470 [18:10<1:03:30,  1.38it/s]
1|1212|Loss: 4.970063209533691:  19%|█▊        | 1212/6470 [18:10<1:04:54,  1.35it/s]
1|1213|Loss: 5.022704601287842:  19%|█▊        | 1212/6470 [18:11<1:04:54,  1.35it/s]
1|1213|Loss: 5.022704601287842:  19%|█▊        | 1213/6470 [18:11<1:06:15,  1.32it/s]
1|1214|Loss: 5.225892066955566:  19%|█▊        | 1213/6470 [18:11<1:06:15,  1.32it/s]
1|1214|Loss: 5.225892066955566:  19%|█▉        | 1214/6470 [18:12<1:10:06,  1.25it/s]
1|1215|Loss: 5.137231349945068:  19%|█▉        | 1214/6470 [18:12<1:10:06,  1.25it/s]
1|1215|Loss: 5.137231349945068:  19%|█▉        | 1215/6470 [18:13<1:12:16,  1.21it/s]
1|1216|Loss: 5.04833459854126:  19%|█▉        | 1215/6470 [18:13<1:12:16,  1.21it/s] 
1|1216|Loss: 5.04833459854126:  19%|█▉        | 1216/6470 [18:14<1:12:13,  1.21it/s]
1|1217|Loss: 4.989017009735107:  19%|█▉        | 1216/6470 [18:14<1:12:13,  1.21it/s]
1|1217|Loss: 4.989017009735107:  19%|█▉        | 1217/6470 [18:14<1:10:53,  1.24it/s]
1|1218|Loss: 4.968618869781494:  19%|█▉        | 1217/6470 [18:15<1:10:53,  1.24it/s]
1|1218|Loss: 4.968618869781494:  19%|█▉        | 1218/6470 [18:15<1:05:46,  1.33it/s]
1|1219|Loss: 5.389319896697998:  19%|█▉        | 1218/6470 [18:15<1:05:46,  1.33it/s]
1|1219|Loss: 5.389319896697998:  19%|█▉        | 1219/6470 [18:16<1:02:52,  1.39it/s]
1|1220|Loss: 5.259135723114014:  19%|█▉        | 1219/6470 [18:16<1:02:52,  1.39it/s]
1|1220|Loss: 5.259135723114014:  19%|█▉        | 1220/6470 [18:17<1:08:50,  1.27it/s]
1|1221|Loss: 5.348784923553467:  19%|█▉        | 1220/6470 [18:17<1:08:50,  1.27it/s]
1|1221|Loss: 5.348784923553467:  19%|█▉        | 1221/6470 [18:17<1:10:01,  1.25it/s]
1|1222|Loss: 5.370367050170898:  19%|█▉        | 1221/6470 [18:18<1:10:01,  1.25it/s]
1|1222|Loss: 5.370367050170898:  19%|█▉        | 1222/6470 [18:18<1:09:48,  1.25it/s]
1|1223|Loss: 5.82347297668457:  19%|█▉        | 1222/6470 [18:19<1:09:48,  1.25it/s] 
1|1223|Loss: 5.82347297668457:  19%|█▉        | 1223/6470 [18:19<1:14:03,  1.18it/s]
1|1224|Loss: 5.435611724853516:  19%|█▉        | 1223/6470 [18:20<1:14:03,  1.18it/s]
1|1224|Loss: 5.435611724853516:  19%|█▉        | 1224/6470 [18:20<1:15:47,  1.15it/s]
1|1225|Loss: 4.882147789001465:  19%|█▉        | 1224/6470 [18:20<1:15:47,  1.15it/s]
1|1225|Loss: 4.882147789001465:  19%|█▉        | 1225/6470 [18:21<1:10:51,  1.23it/s]
1|1226|Loss: 5.476885795593262:  19%|█▉        | 1225/6470 [18:21<1:10:51,  1.23it/s]
1|1226|Loss: 5.476885795593262:  19%|█▉        | 1226/6470 [18:22<1:12:13,  1.21it/s]
1|1227|Loss: 5.99390172958374:  19%|█▉        | 1226/6470 [18:22<1:12:13,  1.21it/s] 
1|1227|Loss: 5.99390172958374:  19%|█▉        | 1227/6470 [18:23<1:18:39,  1.11it/s]
1|1228|Loss: 5.337514877319336:  19%|█▉        | 1227/6470 [18:23<1:18:39,  1.11it/s]
1|1228|Loss: 5.337514877319336:  19%|█▉        | 1228/6470 [18:24<1:21:51,  1.07it/s]
1|1229|Loss: 5.536855697631836:  19%|█▉        | 1228/6470 [18:24<1:21:51,  1.07it/s]
1|1229|Loss: 5.536855697631836:  19%|█▉        | 1229/6470 [18:25<1:19:41,  1.10it/s]
1|1230|Loss: 5.573211193084717:  19%|█▉        | 1229/6470 [18:25<1:19:41,  1.10it/s]
1|1230|Loss: 5.573211193084717:  19%|█▉        | 1230/6470 [18:25<1:17:07,  1.13it/s]
1|1231|Loss: 5.2457170486450195:  19%|█▉        | 1230/6470 [18:26<1:17:07,  1.13it/s]
1|1231|Loss: 5.2457170486450195:  19%|█▉        | 1231/6470 [18:26<1:11:30,  1.22it/s]
1|1232|Loss: 3.550488233566284:  19%|█▉        | 1231/6470 [18:26<1:11:30,  1.22it/s] 
1|1232|Loss: 3.550488233566284:  19%|█▉        | 1232/6470 [18:26<1:01:16,  1.42it/s]
1|1233|Loss: 4.969585418701172:  19%|█▉        | 1232/6470 [18:27<1:01:16,  1.42it/s]
1|1233|Loss: 4.969585418701172:  19%|█▉        | 1233/6470 [18:27<59:41,  1.46it/s]  
1|1234|Loss: 4.88390588760376:  19%|█▉        | 1233/6470 [18:28<59:41,  1.46it/s] 
1|1234|Loss: 4.88390588760376:  19%|█▉        | 1234/6470 [18:28<1:09:49,  1.25it/s]
1|1235|Loss: 5.463140964508057:  19%|█▉        | 1234/6470 [18:29<1:09:49,  1.25it/s]
1|1235|Loss: 5.463140964508057:  19%|█▉        | 1235/6470 [18:29<1:14:54,  1.16it/s]
1|1236|Loss: 4.100780010223389:  19%|█▉        | 1235/6470 [18:29<1:14:54,  1.16it/s]
1|1236|Loss: 4.100780010223389:  19%|█▉        | 1236/6470 [18:30<1:04:57,  1.34it/s]
1|1237|Loss: 4.155149459838867:  19%|█▉        | 1236/6470 [18:30<1:04:57,  1.34it/s]
1|1237|Loss: 4.155149459838867:  19%|█▉        | 1237/6470 [18:30<59:04,  1.48it/s]  
1|1238|Loss: 5.256834983825684:  19%|█▉        | 1237/6470 [18:31<59:04,  1.48it/s]
1|1238|Loss: 5.256834983825684:  19%|█▉        | 1238/6470 [18:31<1:00:07,  1.45it/s]
1|1239|Loss: 5.36829948425293:  19%|█▉        | 1238/6470 [18:31<1:00:07,  1.45it/s] 
1|1239|Loss: 5.36829948425293:  19%|█▉        | 1239/6470 [18:31<58:11,  1.50it/s]  
1|1240|Loss: 5.474311828613281:  19%|█▉        | 1239/6470 [18:32<58:11,  1.50it/s]
1|1240|Loss: 5.474311828613281:  19%|█▉        | 1240/6470 [18:32<1:00:30,  1.44it/s]
1|1241|Loss: 5.82414436340332:  19%|█▉        | 1240/6470 [18:33<1:00:30,  1.44it/s] 
1|1241|Loss: 5.82414436340332:  19%|█▉        | 1241/6470 [18:33<1:03:09,  1.38it/s]
1|1242|Loss: 5.048603534698486:  19%|█▉        | 1241/6470 [18:33<1:03:09,  1.38it/s]
1|1242|Loss: 5.048603534698486:  19%|█▉        | 1242/6470 [18:34<59:09,  1.47it/s]  
1|1243|Loss: 5.178728103637695:  19%|█▉        | 1242/6470 [18:34<59:09,  1.47it/s]
1|1243|Loss: 5.178728103637695:  19%|█▉        | 1243/6470 [18:34<58:29,  1.49it/s]
1|1244|Loss: 5.223867893218994:  19%|█▉        | 1243/6470 [18:35<58:29,  1.49it/s]
1|1244|Loss: 5.223867893218994:  19%|█▉        | 1244/6470 [18:35<1:00:48,  1.43it/s]
1|1245|Loss: 5.408742427825928:  19%|█▉        | 1244/6470 [18:35<1:00:48,  1.43it/s]
1|1245|Loss: 5.408742427825928:  19%|█▉        | 1245/6470 [18:36<1:00:46,  1.43it/s]
1|1246|Loss: 4.640544414520264:  19%|█▉        | 1245/6470 [18:36<1:00:46,  1.43it/s]
1|1246|Loss: 4.640544414520264:  19%|█▉        | 1246/6470 [18:36<59:58,  1.45it/s]  
1|1247|Loss: 5.256528854370117:  19%|█▉        | 1246/6470 [18:37<59:58,  1.45it/s]
1|1247|Loss: 5.256528854370117:  19%|█▉        | 1247/6470 [18:37<1:01:14,  1.42it/s]
1|1248|Loss: 5.529438018798828:  19%|█▉        | 1247/6470 [18:38<1:01:14,  1.42it/s]
1|1248|Loss: 5.529438018798828:  19%|█▉        | 1248/6470 [18:38<1:05:57,  1.32it/s]
1|1249|Loss: 5.219566822052002:  19%|█▉        | 1248/6470 [18:39<1:05:57,  1.32it/s]
1|1249|Loss: 5.219566822052002:  19%|█▉        | 1249/6470 [18:39<1:09:17,  1.26it/s]
1|1250|Loss: 5.3478593826293945:  19%|█▉        | 1249/6470 [18:39<1:09:17,  1.26it/s]
1|1250|Loss: 5.3478593826293945:  19%|█▉        | 1250/6470 [18:40<1:09:34,  1.25it/s]
1|1251|Loss: 5.241257667541504:  19%|█▉        | 1250/6470 [18:40<1:09:34,  1.25it/s] 
1|1251|Loss: 5.241257667541504:  19%|█▉        | 1251/6470 [18:40<1:07:23,  1.29it/s]
1|1252|Loss: 5.423116207122803:  19%|█▉        | 1251/6470 [18:41<1:07:23,  1.29it/s]
1|1252|Loss: 5.423116207122803:  19%|█▉        | 1252/6470 [18:41<1:07:12,  1.29it/s]
1|1253|Loss: 5.259155750274658:  19%|█▉        | 1252/6470 [18:42<1:07:12,  1.29it/s]
1|1253|Loss: 5.259155750274658:  19%|█▉        | 1253/6470 [18:42<1:05:16,  1.33it/s]
1|1254|Loss: 5.468133449554443:  19%|█▉        | 1253/6470 [18:42<1:05:16,  1.33it/s]
1|1254|Loss: 5.468133449554443:  19%|█▉        | 1254/6470 [18:43<1:07:19,  1.29it/s]
1|1255|Loss: 5.316504955291748:  19%|█▉        | 1254/6470 [18:43<1:07:19,  1.29it/s]
1|1255|Loss: 5.316504955291748:  19%|█▉        | 1255/6470 [18:43<1:05:48,  1.32it/s]
1|1256|Loss: 4.934748649597168:  19%|█▉        | 1255/6470 [18:44<1:05:48,  1.32it/s]
1|1256|Loss: 4.934748649597168:  19%|█▉        | 1256/6470 [18:44<1:08:44,  1.26it/s]
1|1257|Loss: 5.359617710113525:  19%|█▉        | 1256/6470 [18:45<1:08:44,  1.26it/s]
1|1257|Loss: 5.359617710113525:  19%|█▉        | 1257/6470 [18:45<1:06:57,  1.30it/s]
1|1258|Loss: 5.467870235443115:  19%|█▉        | 1257/6470 [18:46<1:06:57,  1.30it/s]
1|1258|Loss: 5.467870235443115:  19%|█▉        | 1258/6470 [18:46<1:10:00,  1.24it/s]
1|1259|Loss: 5.155881404876709:  19%|█▉        | 1258/6470 [18:46<1:10:00,  1.24it/s]
1|1259|Loss: 5.155881404876709:  19%|█▉        | 1259/6470 [18:47<1:09:10,  1.26it/s]
1|1260|Loss: 6.241896152496338:  19%|█▉        | 1259/6470 [18:47<1:09:10,  1.26it/s]
1|1260|Loss: 6.241896152496338:  19%|█▉        | 1260/6470 [18:48<1:19:41,  1.09it/s]
1|1261|Loss: 5.468957424163818:  19%|█▉        | 1260/6470 [18:48<1:19:41,  1.09it/s]
1|1261|Loss: 5.468957424163818:  19%|█▉        | 1261/6470 [18:49<1:17:32,  1.12it/s]
1|1262|Loss: 5.356865406036377:  19%|█▉        | 1261/6470 [18:49<1:17:32,  1.12it/s]
1|1262|Loss: 5.356865406036377:  20%|█▉        | 1262/6470 [18:50<1:14:11,  1.17it/s]
1|1263|Loss: 5.284598350524902:  20%|█▉        | 1262/6470 [18:50<1:14:11,  1.17it/s]
1|1263|Loss: 5.284598350524902:  20%|█▉        | 1263/6470 [18:50<1:09:45,  1.24it/s]
1|1264|Loss: 5.397931098937988:  20%|█▉        | 1263/6470 [18:51<1:09:45,  1.24it/s]
1|1264|Loss: 5.397931098937988:  20%|█▉        | 1264/6470 [18:51<1:09:10,  1.25it/s]
1|1265|Loss: 4.434213638305664:  20%|█▉        | 1264/6470 [18:51<1:09:10,  1.25it/s]
1|1265|Loss: 4.434213638305664:  20%|█▉        | 1265/6470 [18:52<1:04:09,  1.35it/s]
1|1266|Loss: 4.831677436828613:  20%|█▉        | 1265/6470 [18:52<1:04:09,  1.35it/s]
1|1266|Loss: 4.831677436828613:  20%|█▉        | 1266/6470 [18:52<1:06:07,  1.31it/s]
1|1267|Loss: 5.028735637664795:  20%|█▉        | 1266/6470 [18:53<1:06:07,  1.31it/s]
1|1267|Loss: 5.028735637664795:  20%|█▉        | 1267/6470 [18:53<1:06:18,  1.31it/s]
1|1268|Loss: 5.210615158081055:  20%|█▉        | 1267/6470 [18:54<1:06:18,  1.31it/s]
1|1268|Loss: 5.210615158081055:  20%|█▉        | 1268/6470 [18:54<1:06:58,  1.29it/s]
1|1269|Loss: 5.724403381347656:  20%|█▉        | 1268/6470 [18:54<1:06:58,  1.29it/s]
1|1269|Loss: 5.724403381347656:  20%|█▉        | 1269/6470 [18:55<1:05:51,  1.32it/s]
1|1270|Loss: 4.974681377410889:  20%|█▉        | 1269/6470 [18:55<1:05:51,  1.32it/s]
1|1270|Loss: 4.974681377410889:  20%|█▉        | 1270/6470 [18:55<1:01:00,  1.42it/s]
1|1271|Loss: 5.748997211456299:  20%|█▉        | 1270/6470 [18:56<1:01:00,  1.42it/s]
1|1271|Loss: 5.748997211456299:  20%|█▉        | 1271/6470 [18:56<1:07:20,  1.29it/s]
1|1272|Loss: 5.8909010887146:  20%|█▉        | 1271/6470 [18:57<1:07:20,  1.29it/s]  
1|1272|Loss: 5.8909010887146:  20%|█▉        | 1272/6470 [18:57<1:11:41,  1.21it/s]
1|1273|Loss: 5.2849884033203125:  20%|█▉        | 1272/6470 [18:58<1:11:41,  1.21it/s]
1|1273|Loss: 5.2849884033203125:  20%|█▉        | 1273/6470 [18:58<1:12:54,  1.19it/s]
1|1274|Loss: 5.428703784942627:  20%|█▉        | 1273/6470 [18:58<1:12:54,  1.19it/s] 
1|1274|Loss: 5.428703784942627:  20%|█▉        | 1274/6470 [18:59<1:11:29,  1.21it/s]
1|1275|Loss: 5.0309858322143555:  20%|█▉        | 1274/6470 [18:59<1:11:29,  1.21it/s]
1|1275|Loss: 5.0309858322143555:  20%|█▉        | 1275/6470 [19:00<1:10:45,  1.22it/s]
1|1276|Loss: 5.356889247894287:  20%|█▉        | 1275/6470 [19:00<1:10:45,  1.22it/s] 
1|1276|Loss: 5.356889247894287:  20%|█▉        | 1276/6470 [19:00<1:10:30,  1.23it/s]
1|1277|Loss: 5.41193962097168:  20%|█▉        | 1276/6470 [19:01<1:10:30,  1.23it/s] 
1|1277|Loss: 5.41193962097168:  20%|█▉        | 1277/6470 [19:01<1:08:11,  1.27it/s]
1|1278|Loss: 5.52785062789917:  20%|█▉        | 1277/6470 [19:02<1:08:11,  1.27it/s]
1|1278|Loss: 5.52785062789917:  20%|█▉        | 1278/6470 [19:02<1:11:36,  1.21it/s]
1|1279|Loss: 5.466170787811279:  20%|█▉        | 1278/6470 [19:03<1:11:36,  1.21it/s]
1|1279|Loss: 5.466170787811279:  20%|█▉        | 1279/6470 [19:03<1:14:31,  1.16it/s]
1|1280|Loss: 4.838337421417236:  20%|█▉        | 1279/6470 [19:03<1:14:31,  1.16it/s]
1|1280|Loss: 4.838337421417236:  20%|█▉        | 1280/6470 [19:04<1:08:38,  1.26it/s]
1|1281|Loss: 5.177889347076416:  20%|█▉        | 1280/6470 [19:04<1:08:38,  1.26it/s]
1|1281|Loss: 5.177889347076416:  20%|█▉        | 1281/6470 [19:04<1:04:48,  1.33it/s]
1|1282|Loss: 5.006211757659912:  20%|█▉        | 1281/6470 [19:05<1:04:48,  1.33it/s]
1|1282|Loss: 5.006211757659912:  20%|█▉        | 1282/6470 [19:05<1:03:02,  1.37it/s]
1|1283|Loss: 5.584716796875:  20%|█▉        | 1282/6470 [19:05<1:03:02,  1.37it/s]   
1|1283|Loss: 5.584716796875:  20%|█▉        | 1283/6470 [19:06<1:04:30,  1.34it/s]
1|1284|Loss: 5.046987533569336:  20%|█▉        | 1283/6470 [19:06<1:04:30,  1.34it/s]
1|1284|Loss: 5.046987533569336:  20%|█▉        | 1284/6470 [19:06<1:02:25,  1.38it/s]
1|1285|Loss: 5.874090671539307:  20%|█▉        | 1284/6470 [19:07<1:02:25,  1.38it/s]
1|1285|Loss: 5.874090671539307:  20%|█▉        | 1285/6470 [19:07<1:06:47,  1.29it/s]
1|1286|Loss: 4.76835823059082:  20%|█▉        | 1285/6470 [19:08<1:06:47,  1.29it/s] 
1|1286|Loss: 4.76835823059082:  20%|█▉        | 1286/6470 [19:08<1:03:44,  1.36it/s]
1|1287|Loss: 4.850039005279541:  20%|█▉        | 1286/6470 [19:08<1:03:44,  1.36it/s]
1|1287|Loss: 4.850039005279541:  20%|█▉        | 1287/6470 [19:09<1:03:06,  1.37it/s]
1|1288|Loss: 4.488128185272217:  20%|█▉        | 1287/6470 [19:09<1:03:06,  1.37it/s]
1|1288|Loss: 4.488128185272217:  20%|█▉        | 1288/6470 [19:09<57:59,  1.49it/s]  
1|1289|Loss: 4.707962512969971:  20%|█▉        | 1288/6470 [19:10<57:59,  1.49it/s]
1|1289|Loss: 4.707962512969971:  20%|█▉        | 1289/6470 [19:10<55:11,  1.56it/s]
1|1290|Loss: 5.486250400543213:  20%|█▉        | 1289/6470 [19:10<55:11,  1.56it/s]
1|1290|Loss: 5.486250400543213:  20%|█▉        | 1290/6470 [19:11<57:57,  1.49it/s]
1|1291|Loss: 5.14794397354126:  20%|█▉        | 1290/6470 [19:11<57:57,  1.49it/s] 
1|1291|Loss: 5.14794397354126:  20%|█▉        | 1291/6470 [19:11<1:00:18,  1.43it/s]
1|1292|Loss: 5.81742000579834:  20%|█▉        | 1291/6470 [19:12<1:00:18,  1.43it/s]
1|1292|Loss: 5.81742000579834:  20%|█▉        | 1292/6470 [19:12<1:08:31,  1.26it/s]
1|1293|Loss: 5.413368225097656:  20%|█▉        | 1292/6470 [19:13<1:08:31,  1.26it/s]
1|1293|Loss: 5.413368225097656:  20%|█▉        | 1293/6470 [19:13<1:10:36,  1.22it/s]
1|1294|Loss: 5.710625171661377:  20%|█▉        | 1293/6470 [19:14<1:10:36,  1.22it/s]
1|1294|Loss: 5.710625171661377:  20%|██        | 1294/6470 [19:15<1:26:38,  1.00s/it]
1|1295|Loss: 5.26390266418457:  20%|██        | 1294/6470 [19:15<1:26:38,  1.00s/it] 
1|1295|Loss: 5.26390266418457:  20%|██        | 1295/6470 [19:16<1:26:28,  1.00s/it]
1|1296|Loss: 4.8438029289245605:  20%|██        | 1295/6470 [19:16<1:26:28,  1.00s/it]
1|1296|Loss: 4.8438029289245605:  20%|██        | 1296/6470 [19:16<1:20:59,  1.06it/s]
1|1297|Loss: 4.899810314178467:  20%|██        | 1296/6470 [19:17<1:20:59,  1.06it/s] 
1|1297|Loss: 4.899810314178467:  20%|██        | 1297/6470 [19:17<1:17:11,  1.12it/s]
1|1298|Loss: 5.2310404777526855:  20%|██        | 1297/6470 [19:18<1:17:11,  1.12it/s]
1|1298|Loss: 5.2310404777526855:  20%|██        | 1298/6470 [19:18<1:11:32,  1.20it/s]
1|1299|Loss: 4.7224273681640625:  20%|██        | 1298/6470 [19:18<1:11:32,  1.20it/s]
1|1299|Loss: 4.7224273681640625:  20%|██        | 1299/6470 [19:18<1:06:07,  1.30it/s]
1|1300|Loss: 5.329744815826416:  20%|██        | 1299/6470 [19:19<1:06:07,  1.30it/s] 
1|1300|Loss: 5.329744815826416:  20%|██        | 1300/6470 [19:19<1:04:25,  1.34it/s]
1|1301|Loss: 5.487841606140137:  20%|██        | 1300/6470 [19:20<1:04:25,  1.34it/s]
1|1301|Loss: 5.487841606140137:  20%|██        | 1301/6470 [19:20<1:07:08,  1.28it/s]
1|1302|Loss: 5.3061370849609375:  20%|██        | 1301/6470 [19:21<1:07:08,  1.28it/s]
1|1302|Loss: 5.3061370849609375:  20%|██        | 1302/6470 [19:21<1:09:54,  1.23it/s]
1|1303|Loss: 5.342339992523193:  20%|██        | 1302/6470 [19:21<1:09:54,  1.23it/s] 
1|1303|Loss: 5.342339992523193:  20%|██        | 1303/6470 [19:22<1:12:44,  1.18it/s]
1|1304|Loss: 4.65507173538208:  20%|██        | 1303/6470 [19:22<1:12:44,  1.18it/s] 
1|1304|Loss: 4.65507173538208:  20%|██        | 1304/6470 [19:23<1:07:39,  1.27it/s]
1|1305|Loss: 5.47456169128418:  20%|██        | 1304/6470 [19:23<1:07:39,  1.27it/s]
1|1305|Loss: 5.47456169128418:  20%|██        | 1305/6470 [19:23<1:06:22,  1.30it/s]
1|1306|Loss: 5.459225177764893:  20%|██        | 1305/6470 [19:24<1:06:22,  1.30it/s]
1|1306|Loss: 5.459225177764893:  20%|██        | 1306/6470 [19:24<1:07:30,  1.27it/s]
1|1307|Loss: 5.650866985321045:  20%|██        | 1306/6470 [19:25<1:07:30,  1.27it/s]
1|1307|Loss: 5.650866985321045:  20%|██        | 1307/6470 [19:25<1:10:55,  1.21it/s]
1|1308|Loss: 5.582210540771484:  20%|██        | 1307/6470 [19:25<1:10:55,  1.21it/s]
1|1308|Loss: 5.582210540771484:  20%|██        | 1308/6470 [19:26<1:09:08,  1.24it/s]
1|1309|Loss: 5.731757640838623:  20%|██        | 1308/6470 [19:26<1:09:08,  1.24it/s]
1|1309|Loss: 5.731757640838623:  20%|██        | 1309/6470 [19:27<1:09:04,  1.25it/s]
1|1310|Loss: 5.172431468963623:  20%|██        | 1309/6470 [19:27<1:09:04,  1.25it/s]
1|1310|Loss: 5.172431468963623:  20%|██        | 1310/6470 [19:27<1:09:14,  1.24it/s]
1|1311|Loss: 4.518978595733643:  20%|██        | 1310/6470 [19:28<1:09:14,  1.24it/s]
1|1311|Loss: 4.518978595733643:  20%|██        | 1311/6470 [19:28<1:04:35,  1.33it/s]
1|1312|Loss: 5.532994270324707:  20%|██        | 1311/6470 [19:28<1:04:35,  1.33it/s]
1|1312|Loss: 5.532994270324707:  20%|██        | 1312/6470 [19:29<1:10:17,  1.22it/s]
1|1313|Loss: 5.726708889007568:  20%|██        | 1312/6470 [19:29<1:10:17,  1.22it/s]
1|1313|Loss: 5.726708889007568:  20%|██        | 1313/6470 [19:30<1:14:31,  1.15it/s]
1|1314|Loss: 4.7498931884765625:  20%|██        | 1313/6470 [19:30<1:14:31,  1.15it/s]
1|1314|Loss: 4.7498931884765625:  20%|██        | 1314/6470 [19:31<1:13:03,  1.18it/s]
1|1315|Loss: 4.7837371826171875:  20%|██        | 1314/6470 [19:31<1:13:03,  1.18it/s]
1|1315|Loss: 4.7837371826171875:  20%|██        | 1315/6470 [19:31<1:08:16,  1.26it/s]
1|1316|Loss: 5.44094181060791:  20%|██        | 1315/6470 [19:32<1:08:16,  1.26it/s]  
1|1316|Loss: 5.44094181060791:  20%|██        | 1316/6470 [19:32<1:11:41,  1.20it/s]
1|1317|Loss: 5.0380449295043945:  20%|██        | 1316/6470 [19:33<1:11:41,  1.20it/s]
1|1317|Loss: 5.0380449295043945:  20%|██        | 1317/6470 [19:33<1:07:01,  1.28it/s]
1|1318|Loss: 4.429440498352051:  20%|██        | 1317/6470 [19:33<1:07:01,  1.28it/s] 
1|1318|Loss: 4.429440498352051:  20%|██        | 1318/6470 [19:34<1:06:00,  1.30it/s]
1|1319|Loss: 4.754300117492676:  20%|██        | 1318/6470 [19:34<1:06:00,  1.30it/s]
1|1319|Loss: 4.754300117492676:  20%|██        | 1319/6470 [19:34<1:03:58,  1.34it/s]
1|1320|Loss: 5.406382083892822:  20%|██        | 1319/6470 [19:35<1:03:58,  1.34it/s]
1|1320|Loss: 5.406382083892822:  20%|██        | 1320/6470 [19:35<1:04:50,  1.32it/s]
1|1321|Loss: 4.840616226196289:  20%|██        | 1320/6470 [19:36<1:04:50,  1.32it/s]
1|1321|Loss: 4.840616226196289:  20%|██        | 1321/6470 [19:36<1:07:25,  1.27it/s]
1|1322|Loss: 5.2118096351623535:  20%|██        | 1321/6470 [19:37<1:07:25,  1.27it/s]
1|1322|Loss: 5.2118096351623535:  20%|██        | 1322/6470 [19:37<1:10:07,  1.22it/s]
1|1323|Loss: 5.171910285949707:  20%|██        | 1322/6470 [19:37<1:10:07,  1.22it/s] 
1|1323|Loss: 5.171910285949707:  20%|██        | 1323/6470 [19:38<1:12:26,  1.18it/s]
1|1324|Loss: 4.801393985748291:  20%|██        | 1323/6470 [19:38<1:12:26,  1.18it/s]
1|1324|Loss: 4.801393985748291:  20%|██        | 1324/6470 [19:39<1:13:53,  1.16it/s]
1|1325|Loss: 5.283027648925781:  20%|██        | 1324/6470 [19:39<1:13:53,  1.16it/s]
1|1325|Loss: 5.283027648925781:  20%|██        | 1325/6470 [19:40<1:14:56,  1.14it/s]
1|1326|Loss: 5.057816505432129:  20%|██        | 1325/6470 [19:40<1:14:56,  1.14it/s]
1|1326|Loss: 5.057816505432129:  20%|██        | 1326/6470 [19:40<1:11:33,  1.20it/s]
1|1327|Loss: 4.990675449371338:  20%|██        | 1326/6470 [19:41<1:11:33,  1.20it/s]
1|1327|Loss: 4.990675449371338:  21%|██        | 1327/6470 [19:41<1:11:54,  1.19it/s]
1|1328|Loss: 4.952035903930664:  21%|██        | 1327/6470 [19:42<1:11:54,  1.19it/s]
1|1328|Loss: 4.952035903930664:  21%|██        | 1328/6470 [19:42<1:11:28,  1.20it/s]
1|1329|Loss: 5.153169631958008:  21%|██        | 1328/6470 [19:43<1:11:28,  1.20it/s]
1|1329|Loss: 5.153169631958008:  21%|██        | 1329/6470 [19:43<1:12:03,  1.19it/s]
1|1330|Loss: 4.810207843780518:  21%|██        | 1329/6470 [19:43<1:12:03,  1.19it/s]
1|1330|Loss: 4.810207843780518:  21%|██        | 1330/6470 [19:44<1:09:30,  1.23it/s]
1|1331|Loss: 4.829325199127197:  21%|██        | 1330/6470 [19:44<1:09:30,  1.23it/s]
1|1331|Loss: 4.829325199127197:  21%|██        | 1331/6470 [19:44<1:08:32,  1.25it/s]
1|1332|Loss: 5.349762439727783:  21%|██        | 1331/6470 [19:45<1:08:32,  1.25it/s]
1|1332|Loss: 5.349762439727783:  21%|██        | 1332/6470 [19:45<1:09:50,  1.23it/s]
1|1333|Loss: 5.080216884613037:  21%|██        | 1332/6470 [19:46<1:09:50,  1.23it/s]
1|1333|Loss: 5.080216884613037:  21%|██        | 1333/6470 [19:46<1:10:13,  1.22it/s]
1|1334|Loss: 4.691391468048096:  21%|██        | 1333/6470 [19:46<1:10:13,  1.22it/s]
1|1334|Loss: 4.691391468048096:  21%|██        | 1334/6470 [19:47<1:04:06,  1.34it/s]
1|1335|Loss: 5.164997577667236:  21%|██        | 1334/6470 [19:47<1:04:06,  1.34it/s]
1|1335|Loss: 5.164997577667236:  21%|██        | 1335/6470 [19:47<59:36,  1.44it/s]  
1|1336|Loss: 5.254321575164795:  21%|██        | 1335/6470 [19:48<59:36,  1.44it/s]
1|1336|Loss: 5.254321575164795:  21%|██        | 1336/6470 [19:48<1:03:30,  1.35it/s]
1|1337|Loss: 4.99591588973999:  21%|██        | 1336/6470 [19:49<1:03:30,  1.35it/s] 
1|1337|Loss: 4.99591588973999:  21%|██        | 1337/6470 [19:49<1:03:10,  1.35it/s]
1|1338|Loss: 5.16185188293457:  21%|██        | 1337/6470 [19:49<1:03:10,  1.35it/s]
1|1338|Loss: 5.16185188293457:  21%|██        | 1338/6470 [19:50<1:04:27,  1.33it/s]
1|1339|Loss: 4.995237350463867:  21%|██        | 1338/6470 [19:50<1:04:27,  1.33it/s]
1|1339|Loss: 4.995237350463867:  21%|██        | 1339/6470 [19:50<1:02:05,  1.38it/s]
1|1340|Loss: 5.23856782913208:  21%|██        | 1339/6470 [19:51<1:02:05,  1.38it/s] 
1|1340|Loss: 5.23856782913208:  21%|██        | 1340/6470 [19:51<1:03:35,  1.34it/s]
1|1341|Loss: 5.516235828399658:  21%|██        | 1340/6470 [19:52<1:03:35,  1.34it/s]
1|1341|Loss: 5.516235828399658:  21%|██        | 1341/6470 [19:52<1:07:48,  1.26it/s]
1|1342|Loss: 5.243888854980469:  21%|██        | 1341/6470 [19:52<1:07:48,  1.26it/s]
1|1342|Loss: 5.243888854980469:  21%|██        | 1342/6470 [19:53<1:08:58,  1.24it/s]
1|1343|Loss: 4.212539196014404:  21%|██        | 1342/6470 [19:53<1:08:58,  1.24it/s]
1|1343|Loss: 4.212539196014404:  21%|██        | 1343/6470 [19:54<1:06:04,  1.29it/s]
1|1344|Loss: 5.098612308502197:  21%|██        | 1343/6470 [19:54<1:06:04,  1.29it/s]
1|1344|Loss: 5.098612308502197:  21%|██        | 1344/6470 [19:54<1:07:46,  1.26it/s]
1|1345|Loss: 4.839677810668945:  21%|██        | 1344/6470 [19:55<1:07:46,  1.26it/s]
1|1345|Loss: 4.839677810668945:  21%|██        | 1345/6470 [19:55<1:03:36,  1.34it/s]
1|1346|Loss: 5.321206092834473:  21%|██        | 1345/6470 [19:55<1:03:36,  1.34it/s]
1|1346|Loss: 5.321206092834473:  21%|██        | 1346/6470 [19:56<1:03:56,  1.34it/s]
1|1347|Loss: 5.087517261505127:  21%|██        | 1346/6470 [19:56<1:03:56,  1.34it/s]
1|1347|Loss: 5.087517261505127:  21%|██        | 1347/6470 [19:56<1:02:52,  1.36it/s]
1|1348|Loss: 5.4318623542785645:  21%|██        | 1347/6470 [19:57<1:02:52,  1.36it/s]
1|1348|Loss: 5.4318623542785645:  21%|██        | 1348/6470 [19:57<1:04:15,  1.33it/s]
1|1349|Loss: 5.01657772064209:  21%|██        | 1348/6470 [19:58<1:04:15,  1.33it/s]  
1|1349|Loss: 5.01657772064209:  21%|██        | 1349/6470 [19:58<1:00:14,  1.42it/s]
1|1350|Loss: 5.62166690826416:  21%|██        | 1349/6470 [19:58<1:00:14,  1.42it/s]
1|1350|Loss: 5.62166690826416:  21%|██        | 1350/6470 [19:59<1:04:14,  1.33it/s]
1|1351|Loss: 5.177751064300537:  21%|██        | 1350/6470 [19:59<1:04:14,  1.33it/s]
1|1351|Loss: 5.177751064300537:  21%|██        | 1351/6470 [20:00<1:09:33,  1.23it/s]
1|1352|Loss: 5.1133551597595215:  21%|██        | 1351/6470 [20:00<1:09:33,  1.23it/s]
1|1352|Loss: 5.1133551597595215:  21%|██        | 1352/6470 [20:01<1:09:39,  1.22it/s]
1|1353|Loss: 4.685092926025391:  21%|██        | 1352/6470 [20:01<1:09:39,  1.22it/s] 
1|1353|Loss: 4.685092926025391:  21%|██        | 1353/6470 [20:01<1:04:53,  1.31it/s]
1|1354|Loss: 5.145389080047607:  21%|██        | 1353/6470 [20:02<1:04:53,  1.31it/s]
1|1354|Loss: 5.145389080047607:  21%|██        | 1354/6470 [20:02<1:06:58,  1.27it/s]
1|1355|Loss: 4.922844886779785:  21%|██        | 1354/6470 [20:03<1:06:58,  1.27it/s]
1|1355|Loss: 4.922844886779785:  21%|██        | 1355/6470 [20:03<1:14:27,  1.15it/s]
1|1356|Loss: 5.007555961608887:  21%|██        | 1355/6470 [20:03<1:14:27,  1.15it/s]
1|1356|Loss: 5.007555961608887:  21%|██        | 1356/6470 [20:04<1:09:18,  1.23it/s]
1|1357|Loss: 5.4655256271362305:  21%|██        | 1356/6470 [20:04<1:09:18,  1.23it/s]
1|1357|Loss: 5.4655256271362305:  21%|██        | 1357/6470 [20:04<1:04:51,  1.31it/s]
1|1358|Loss: 5.593006134033203:  21%|██        | 1357/6470 [20:05<1:04:51,  1.31it/s] 
1|1358|Loss: 5.593006134033203:  21%|██        | 1358/6470 [20:05<1:07:59,  1.25it/s]
1|1359|Loss: 5.112639427185059:  21%|██        | 1358/6470 [20:06<1:07:59,  1.25it/s]
1|1359|Loss: 5.112639427185059:  21%|██        | 1359/6470 [20:06<1:08:03,  1.25it/s]
1|1360|Loss: 5.3678789138793945:  21%|██        | 1359/6470 [20:07<1:08:03,  1.25it/s]
1|1360|Loss: 5.3678789138793945:  21%|██        | 1360/6470 [20:07<1:10:15,  1.21it/s]
1|1361|Loss: 4.221972942352295:  21%|██        | 1360/6470 [20:07<1:10:15,  1.21it/s] 
1|1361|Loss: 4.221972942352295:  21%|██        | 1361/6470 [20:07<1:02:00,  1.37it/s]
1|1362|Loss: 5.15015172958374:  21%|██        | 1361/6470 [20:08<1:02:00,  1.37it/s] 
1|1362|Loss: 5.15015172958374:  21%|██        | 1362/6470 [20:08<59:09,  1.44it/s]  
1|1363|Loss: 5.027256011962891:  21%|██        | 1362/6470 [20:08<59:09,  1.44it/s]
1|1363|Loss: 5.027256011962891:  21%|██        | 1363/6470 [20:09<1:01:17,  1.39it/s]
1|1364|Loss: 5.114017009735107:  21%|██        | 1363/6470 [20:09<1:01:17,  1.39it/s]
1|1364|Loss: 5.114017009735107:  21%|██        | 1364/6470 [20:10<1:08:26,  1.24it/s]
1|1365|Loss: 5.68148946762085:  21%|██        | 1364/6470 [20:11<1:08:26,  1.24it/s] 
1|1365|Loss: 5.68148946762085:  21%|██        | 1365/6470 [20:11<1:20:03,  1.06it/s]
1|1366|Loss: 5.438793182373047:  21%|██        | 1365/6470 [20:12<1:20:03,  1.06it/s]
1|1366|Loss: 5.438793182373047:  21%|██        | 1366/6470 [20:12<1:19:37,  1.07it/s]
1|1367|Loss: 5.176249980926514:  21%|██        | 1366/6470 [20:12<1:19:37,  1.07it/s]
1|1367|Loss: 5.176249980926514:  21%|██        | 1367/6470 [20:13<1:15:24,  1.13it/s]
1|1368|Loss: 5.091771125793457:  21%|██        | 1367/6470 [20:13<1:15:24,  1.13it/s]
1|1368|Loss: 5.091771125793457:  21%|██        | 1368/6470 [20:13<1:10:04,  1.21it/s]
1|1369|Loss: 4.818010330200195:  21%|██        | 1368/6470 [20:14<1:10:04,  1.21it/s]
1|1369|Loss: 4.818010330200195:  21%|██        | 1369/6470 [20:14<1:03:34,  1.34it/s]
1|1370|Loss: 5.1215314865112305:  21%|██        | 1369/6470 [20:14<1:03:34,  1.34it/s]
1|1370|Loss: 5.1215314865112305:  21%|██        | 1370/6470 [20:15<1:03:12,  1.34it/s]
1|1371|Loss: 5.251302242279053:  21%|██        | 1370/6470 [20:15<1:03:12,  1.34it/s] 
1|1371|Loss: 5.251302242279053:  21%|██        | 1371/6470 [20:16<1:04:16,  1.32it/s]
1|1372|Loss: 5.018741607666016:  21%|██        | 1371/6470 [20:16<1:04:16,  1.32it/s]
1|1372|Loss: 5.018741607666016:  21%|██        | 1372/6470 [20:16<1:05:29,  1.30it/s]
1|1373|Loss: 5.4813737869262695:  21%|██        | 1372/6470 [20:17<1:05:29,  1.30it/s]
1|1373|Loss: 5.4813737869262695:  21%|██        | 1373/6470 [20:17<1:10:58,  1.20it/s]
1|1374|Loss: 5.342748165130615:  21%|██        | 1373/6470 [20:18<1:10:58,  1.20it/s] 
1|1374|Loss: 5.342748165130615:  21%|██        | 1374/6470 [20:18<1:14:11,  1.14it/s]
1|1375|Loss: 4.5705790519714355:  21%|██        | 1374/6470 [20:19<1:14:11,  1.14it/s]
1|1375|Loss: 4.5705790519714355:  21%|██▏       | 1375/6470 [20:19<1:09:42,  1.22it/s]
1|1376|Loss: 4.96467399597168:  21%|██▏       | 1375/6470 [20:19<1:09:42,  1.22it/s]  
1|1376|Loss: 4.96467399597168:  21%|██▏       | 1376/6470 [20:20<1:05:58,  1.29it/s]
1|1377|Loss: 5.0122857093811035:  21%|██▏       | 1376/6470 [20:20<1:05:58,  1.29it/s]
1|1377|Loss: 5.0122857093811035:  21%|██▏       | 1377/6470 [20:21<1:08:30,  1.24it/s]
1|1378|Loss: 5.47745418548584:  21%|██▏       | 1377/6470 [20:21<1:08:30,  1.24it/s]  
1|1378|Loss: 5.47745418548584:  21%|██▏       | 1378/6470 [20:21<1:06:51,  1.27it/s]
1|1379|Loss: 5.5439252853393555:  21%|██▏       | 1378/6470 [20:22<1:06:51,  1.27it/s]
1|1379|Loss: 5.5439252853393555:  21%|██▏       | 1379/6470 [20:22<1:06:46,  1.27it/s]
1|1380|Loss: 4.841860294342041:  21%|██▏       | 1379/6470 [20:23<1:06:46,  1.27it/s] 
1|1380|Loss: 4.841860294342041:  21%|██▏       | 1380/6470 [20:23<1:08:17,  1.24it/s]
1|1381|Loss: 5.386633396148682:  21%|██▏       | 1380/6470 [20:23<1:08:17,  1.24it/s]
1|1381|Loss: 5.386633396148682:  21%|██▏       | 1381/6470 [20:24<1:12:49,  1.16it/s]
1|1382|Loss: 4.590892314910889:  21%|██▏       | 1381/6470 [20:24<1:12:49,  1.16it/s]
1|1382|Loss: 4.590892314910889:  21%|██▏       | 1382/6470 [20:25<1:11:29,  1.19it/s]
1|1383|Loss: 4.819864749908447:  21%|██▏       | 1382/6470 [20:25<1:11:29,  1.19it/s]
1|1383|Loss: 4.819864749908447:  21%|██▏       | 1383/6470 [20:26<1:10:02,  1.21it/s]
1|1384|Loss: 4.940854072570801:  21%|██▏       | 1383/6470 [20:26<1:10:02,  1.21it/s]
1|1384|Loss: 4.940854072570801:  21%|██▏       | 1384/6470 [20:26<1:07:08,  1.26it/s]
1|1385|Loss: 4.7357258796691895:  21%|██▏       | 1384/6470 [20:27<1:07:08,  1.26it/s]
1|1385|Loss: 4.7357258796691895:  21%|██▏       | 1385/6470 [20:27<1:03:17,  1.34it/s]
1|1386|Loss: 4.5713348388671875:  21%|██▏       | 1385/6470 [20:27<1:03:17,  1.34it/s]
1|1386|Loss: 4.5713348388671875:  21%|██▏       | 1386/6470 [20:28<1:01:39,  1.37it/s]
1|1387|Loss: 4.906405925750732:  21%|██▏       | 1386/6470 [20:28<1:01:39,  1.37it/s] 
1|1387|Loss: 4.906405925750732:  21%|██▏       | 1387/6470 [20:28<1:00:29,  1.40it/s]
1|1388|Loss: 4.404752254486084:  21%|██▏       | 1387/6470 [20:29<1:00:29,  1.40it/s]
1|1388|Loss: 4.404752254486084:  21%|██▏       | 1388/6470 [20:29<55:38,  1.52it/s]  
1|1389|Loss: 4.604936599731445:  21%|██▏       | 1388/6470 [20:29<55:38,  1.52it/s]
1|1389|Loss: 4.604936599731445:  21%|██▏       | 1389/6470 [20:29<55:26,  1.53it/s]
1|1390|Loss: 5.190474510192871:  21%|██▏       | 1389/6470 [20:30<55:26,  1.53it/s]
1|1390|Loss: 5.190474510192871:  21%|██▏       | 1390/6470 [20:30<1:01:08,  1.38it/s]
1|1391|Loss: 5.3734517097473145:  21%|██▏       | 1390/6470 [20:31<1:01:08,  1.38it/s]
1|1391|Loss: 5.3734517097473145:  21%|██▏       | 1391/6470 [20:31<1:05:13,  1.30it/s]
1|1392|Loss: 5.423923015594482:  21%|██▏       | 1391/6470 [20:32<1:05:13,  1.30it/s] 
1|1392|Loss: 5.423923015594482:  22%|██▏       | 1392/6470 [20:32<1:04:34,  1.31it/s]
1|1393|Loss: 4.956614971160889:  22%|██▏       | 1392/6470 [20:32<1:04:34,  1.31it/s]
1|1393|Loss: 4.956614971160889:  22%|██▏       | 1393/6470 [20:33<1:04:56,  1.30it/s]
1|1394|Loss: 4.998533725738525:  22%|██▏       | 1393/6470 [20:33<1:04:56,  1.30it/s]
1|1394|Loss: 4.998533725738525:  22%|██▏       | 1394/6470 [20:33<1:05:50,  1.28it/s]
1|1395|Loss: 4.3815460205078125:  22%|██▏       | 1394/6470 [20:34<1:05:50,  1.28it/s]
1|1395|Loss: 4.3815460205078125:  22%|██▏       | 1395/6470 [20:34<1:04:00,  1.32it/s]
1|1396|Loss: 4.561927318572998:  22%|██▏       | 1395/6470 [20:35<1:04:00,  1.32it/s] 
1|1396|Loss: 4.561927318572998:  22%|██▏       | 1396/6470 [20:35<1:04:48,  1.30it/s]
1|1397|Loss: 4.870641708374023:  22%|██▏       | 1396/6470 [20:35<1:04:48,  1.30it/s]
1|1397|Loss: 4.870641708374023:  22%|██▏       | 1397/6470 [20:36<1:05:23,  1.29it/s]
1|1398|Loss: 4.43265962600708:  22%|██▏       | 1397/6470 [20:36<1:05:23,  1.29it/s] 
1|1398|Loss: 4.43265962600708:  22%|██▏       | 1398/6470 [20:36<1:01:22,  1.38it/s]
1|1399|Loss: 5.2141008377075195:  22%|██▏       | 1398/6470 [20:37<1:01:22,  1.38it/s]
1|1399|Loss: 5.2141008377075195:  22%|██▏       | 1399/6470 [20:37<59:11,  1.43it/s]  
1|1400|Loss: 5.290701866149902:  22%|██▏       | 1399/6470 [20:37<59:11,  1.43it/s] 
1|1400|Loss: 5.290701866149902:  22%|██▏       | 1400/6470 [20:38<1:01:46,  1.37it/s]
1|1401|Loss: 3.934368848800659:  22%|██▏       | 1400/6470 [20:38<1:01:46,  1.37it/s]
1|1401|Loss: 3.934368848800659:  22%|██▏       | 1401/6470 [20:38<55:34,  1.52it/s]  
1|1402|Loss: 4.663698196411133:  22%|██▏       | 1401/6470 [20:39<55:34,  1.52it/s]
1|1402|Loss: 4.663698196411133:  22%|██▏       | 1402/6470 [20:39<57:35,  1.47it/s]
1|1403|Loss: 4.904525279998779:  22%|██▏       | 1402/6470 [20:40<57:35,  1.47it/s]
1|1403|Loss: 4.904525279998779:  22%|██▏       | 1403/6470 [20:40<1:01:00,  1.38it/s]
1|1404|Loss: 4.709682464599609:  22%|██▏       | 1403/6470 [20:40<1:01:00,  1.38it/s]
1|1404|Loss: 4.709682464599609:  22%|██▏       | 1404/6470 [20:40<58:01,  1.45it/s]  
1|1405|Loss: 5.434725761413574:  22%|██▏       | 1404/6470 [20:41<58:01,  1.45it/s]
1|1405|Loss: 5.434725761413574:  22%|██▏       | 1405/6470 [20:41<56:26,  1.50it/s]
1|1406|Loss: 5.615944862365723:  22%|██▏       | 1405/6470 [20:42<56:26,  1.50it/s]
1|1406|Loss: 5.615944862365723:  22%|██▏       | 1406/6470 [20:42<58:44,  1.44it/s]
1|1407|Loss: 5.120071887969971:  22%|██▏       | 1406/6470 [20:42<58:44,  1.44it/s]
1|1407|Loss: 5.120071887969971:  22%|██▏       | 1407/6470 [20:43<1:01:42,  1.37it/s]
1|1408|Loss: 5.0778374671936035:  22%|██▏       | 1407/6470 [20:43<1:01:42,  1.37it/s]
1|1408|Loss: 5.0778374671936035:  22%|██▏       | 1408/6470 [20:43<1:01:16,  1.38it/s]
1|1409|Loss: 5.807650089263916:  22%|██▏       | 1408/6470 [20:44<1:01:16,  1.38it/s] 
1|1409|Loss: 5.807650089263916:  22%|██▏       | 1409/6470 [20:44<1:04:58,  1.30it/s]
1|1410|Loss: 4.7739949226379395:  22%|██▏       | 1409/6470 [20:45<1:04:58,  1.30it/s]
1|1410|Loss: 4.7739949226379395:  22%|██▏       | 1410/6470 [20:45<1:01:10,  1.38it/s]
1|1411|Loss: 5.610060214996338:  22%|██▏       | 1410/6470 [20:45<1:01:10,  1.38it/s] 
1|1411|Loss: 5.610060214996338:  22%|██▏       | 1411/6470 [20:46<59:04,  1.43it/s]  
1|1412|Loss: 4.921205043792725:  22%|██▏       | 1411/6470 [20:46<59:04,  1.43it/s]
1|1412|Loss: 4.921205043792725:  22%|██▏       | 1412/6470 [20:46<1:00:58,  1.38it/s]
1|1413|Loss: 4.738728046417236:  22%|██▏       | 1412/6470 [20:47<1:00:58,  1.38it/s]
1|1413|Loss: 4.738728046417236:  22%|██▏       | 1413/6470 [20:47<58:48,  1.43it/s]  
1|1414|Loss: 5.49049711227417:  22%|██▏       | 1413/6470 [20:47<58:48,  1.43it/s] 
1|1414|Loss: 5.49049711227417:  22%|██▏       | 1414/6470 [20:48<1:04:59,  1.30it/s]
1|1415|Loss: 4.616142749786377:  22%|██▏       | 1414/6470 [20:48<1:04:59,  1.30it/s]
1|1415|Loss: 4.616142749786377:  22%|██▏       | 1415/6470 [20:49<1:04:38,  1.30it/s]
1|1416|Loss: 5.174586772918701:  22%|██▏       | 1415/6470 [20:49<1:04:38,  1.30it/s]
1|1416|Loss: 5.174586772918701:  22%|██▏       | 1416/6470 [20:49<1:04:52,  1.30it/s]
1|1417|Loss: 5.20295524597168:  22%|██▏       | 1416/6470 [20:50<1:04:52,  1.30it/s] 
1|1417|Loss: 5.20295524597168:  22%|██▏       | 1417/6470 [20:50<1:07:26,  1.25it/s]
1|1418|Loss: 4.9502739906311035:  22%|██▏       | 1417/6470 [20:51<1:07:26,  1.25it/s]
1|1418|Loss: 4.9502739906311035:  22%|██▏       | 1418/6470 [20:51<1:06:51,  1.26it/s]
1|1419|Loss: 5.437039375305176:  22%|██▏       | 1418/6470 [20:52<1:06:51,  1.26it/s] 
1|1419|Loss: 5.437039375305176:  22%|██▏       | 1419/6470 [20:52<1:09:10,  1.22it/s]
1|1420|Loss: 4.82609224319458:  22%|██▏       | 1419/6470 [20:52<1:09:10,  1.22it/s] 
1|1420|Loss: 4.82609224319458:  22%|██▏       | 1420/6470 [20:53<1:07:35,  1.25it/s]
1|1421|Loss: 5.348557472229004:  22%|██▏       | 1420/6470 [20:53<1:07:35,  1.25it/s]
1|1421|Loss: 5.348557472229004:  22%|██▏       | 1421/6470 [20:54<1:07:35,  1.25it/s]
1|1422|Loss: 5.0741472244262695:  22%|██▏       | 1421/6470 [20:54<1:07:35,  1.25it/s]
1|1422|Loss: 5.0741472244262695:  22%|██▏       | 1422/6470 [20:54<1:09:52,  1.20it/s]
1|1423|Loss: 5.482860088348389:  22%|██▏       | 1422/6470 [20:55<1:09:52,  1.20it/s] 
1|1423|Loss: 5.482860088348389:  22%|██▏       | 1423/6470 [20:55<1:09:31,  1.21it/s]
1|1424|Loss: 4.893527507781982:  22%|██▏       | 1423/6470 [20:56<1:09:31,  1.21it/s]
1|1424|Loss: 4.893527507781982:  22%|██▏       | 1424/6470 [20:56<1:05:27,  1.28it/s]
1|1425|Loss: 5.04876184463501:  22%|██▏       | 1424/6470 [20:56<1:05:27,  1.28it/s] 
1|1425|Loss: 5.04876184463501:  22%|██▏       | 1425/6470 [20:57<1:07:21,  1.25it/s]
1|1426|Loss: 5.519512176513672:  22%|██▏       | 1425/6470 [20:57<1:07:21,  1.25it/s]
1|1426|Loss: 5.519512176513672:  22%|██▏       | 1426/6470 [20:58<1:09:33,  1.21it/s]
1|1427|Loss: 5.1933770179748535:  22%|██▏       | 1426/6470 [20:58<1:09:33,  1.21it/s]
1|1427|Loss: 5.1933770179748535:  22%|██▏       | 1427/6470 [20:58<1:07:14,  1.25it/s]
1|1428|Loss: 4.88418436050415:  22%|██▏       | 1427/6470 [20:59<1:07:14,  1.25it/s]  
1|1428|Loss: 4.88418436050415:  22%|██▏       | 1428/6470 [20:59<1:05:01,  1.29it/s]
1|1429|Loss: 5.461588382720947:  22%|██▏       | 1428/6470 [21:00<1:05:01,  1.29it/s]
1|1429|Loss: 5.461588382720947:  22%|██▏       | 1429/6470 [21:00<1:08:39,  1.22it/s]
1|1430|Loss: 4.7003173828125:  22%|██▏       | 1429/6470 [21:00<1:08:39,  1.22it/s]  
1|1430|Loss: 4.7003173828125:  22%|██▏       | 1430/6470 [21:01<1:08:28,  1.23it/s]
1|1431|Loss: 5.220778465270996:  22%|██▏       | 1430/6470 [21:01<1:08:28,  1.23it/s]
1|1431|Loss: 5.220778465270996:  22%|██▏       | 1431/6470 [21:01<1:04:43,  1.30it/s]
1|1432|Loss: 5.593029022216797:  22%|██▏       | 1431/6470 [21:02<1:04:43,  1.30it/s]
1|1432|Loss: 5.593029022216797:  22%|██▏       | 1432/6470 [21:02<1:04:49,  1.30it/s]
1|1433|Loss: 5.056102275848389:  22%|██▏       | 1432/6470 [21:03<1:04:49,  1.30it/s]
1|1433|Loss: 5.056102275848389:  22%|██▏       | 1433/6470 [21:03<1:03:06,  1.33it/s]
1|1434|Loss: 5.089758396148682:  22%|██▏       | 1433/6470 [21:03<1:03:06,  1.33it/s]
1|1434|Loss: 5.089758396148682:  22%|██▏       | 1434/6470 [21:04<1:01:15,  1.37it/s]
1|1435|Loss: 4.772385120391846:  22%|██▏       | 1434/6470 [21:04<1:01:15,  1.37it/s]
1|1435|Loss: 4.772385120391846:  22%|██▏       | 1435/6470 [21:04<59:46,  1.40it/s]  
1|1436|Loss: 5.080549716949463:  22%|██▏       | 1435/6470 [21:05<59:46,  1.40it/s]
1|1436|Loss: 5.080549716949463:  22%|██▏       | 1436/6470 [21:05<56:45,  1.48it/s]
1|1437|Loss: 3.993051290512085:  22%|██▏       | 1436/6470 [21:05<56:45,  1.48it/s]
1|1437|Loss: 3.993051290512085:  22%|██▏       | 1437/6470 [21:05<52:18,  1.60it/s]
1|1438|Loss: 4.987251281738281:  22%|██▏       | 1437/6470 [21:06<52:18,  1.60it/s]
1|1438|Loss: 4.987251281738281:  22%|██▏       | 1438/6470 [21:06<53:07,  1.58it/s]
1|1439|Loss: 5.3376641273498535:  22%|██▏       | 1438/6470 [21:07<53:07,  1.58it/s]
1|1439|Loss: 5.3376641273498535:  22%|██▏       | 1439/6470 [21:07<1:01:15,  1.37it/s]
1|1440|Loss: 4.91042947769165:  22%|██▏       | 1439/6470 [21:08<1:01:15,  1.37it/s]  
1|1440|Loss: 4.91042947769165:  22%|██▏       | 1440/6470 [21:08<1:02:15,  1.35it/s]
1|1441|Loss: 4.9231109619140625:  22%|██▏       | 1440/6470 [21:08<1:02:15,  1.35it/s]
1|1441|Loss: 4.9231109619140625:  22%|██▏       | 1441/6470 [21:08<59:47,  1.40it/s]  
1|1442|Loss: 4.603603839874268:  22%|██▏       | 1441/6470 [21:09<59:47,  1.40it/s] 
1|1442|Loss: 4.603603839874268:  22%|██▏       | 1442/6470 [21:09<55:52,  1.50it/s]
1|1443|Loss: 4.969595432281494:  22%|██▏       | 1442/6470 [21:09<55:52,  1.50it/s]
1|1443|Loss: 4.969595432281494:  22%|██▏       | 1443/6470 [21:10<56:06,  1.49it/s]
1|1444|Loss: 4.693012714385986:  22%|██▏       | 1443/6470 [21:10<56:06,  1.49it/s]
1|1444|Loss: 4.693012714385986:  22%|██▏       | 1444/6470 [21:10<58:53,  1.42it/s]
1|1445|Loss: 4.621679306030273:  22%|██▏       | 1444/6470 [21:11<58:53,  1.42it/s]
1|1445|Loss: 4.621679306030273:  22%|██▏       | 1445/6470 [21:11<56:15,  1.49it/s]
1|1446|Loss: 4.718147277832031:  22%|██▏       | 1445/6470 [21:11<56:15,  1.49it/s]
1|1446|Loss: 4.718147277832031:  22%|██▏       | 1446/6470 [21:12<53:50,  1.56it/s]
1|1447|Loss: 5.408814430236816:  22%|██▏       | 1446/6470 [21:12<53:50,  1.56it/s]
1|1447|Loss: 5.408814430236816:  22%|██▏       | 1447/6470 [21:12<58:23,  1.43it/s]
1|1448|Loss: 4.437899589538574:  22%|██▏       | 1447/6470 [21:13<58:23,  1.43it/s]
1|1448|Loss: 4.437899589538574:  22%|██▏       | 1448/6470 [21:13<55:10,  1.52it/s]
1|1449|Loss: 4.860703468322754:  22%|██▏       | 1448/6470 [21:13<55:10,  1.52it/s]
1|1449|Loss: 4.860703468322754:  22%|██▏       | 1449/6470 [21:14<55:57,  1.50it/s]
1|1450|Loss: 4.308241367340088:  22%|██▏       | 1449/6470 [21:14<55:57,  1.50it/s]
1|1450|Loss: 4.308241367340088:  22%|██▏       | 1450/6470 [21:14<56:02,  1.49it/s]
1|1451|Loss: 5.190607070922852:  22%|██▏       | 1450/6470 [21:15<56:02,  1.49it/s]
1|1451|Loss: 5.190607070922852:  22%|██▏       | 1451/6470 [21:15<1:01:38,  1.36it/s]
1|1452|Loss: 5.541688442230225:  22%|██▏       | 1451/6470 [21:16<1:01:38,  1.36it/s]
1|1452|Loss: 5.541688442230225:  22%|██▏       | 1452/6470 [21:16<1:07:13,  1.24it/s]
1|1453|Loss: 4.797906875610352:  22%|██▏       | 1452/6470 [21:17<1:07:13,  1.24it/s]
1|1453|Loss: 4.797906875610352:  22%|██▏       | 1453/6470 [21:17<1:04:41,  1.29it/s]
1|1454|Loss: 5.437178134918213:  22%|██▏       | 1453/6470 [21:17<1:04:41,  1.29it/s]
1|1454|Loss: 5.437178134918213:  22%|██▏       | 1454/6470 [21:18<1:02:04,  1.35it/s]
1|1455|Loss: 4.281017303466797:  22%|██▏       | 1454/6470 [21:18<1:02:04,  1.35it/s]
1|1455|Loss: 4.281017303466797:  22%|██▏       | 1455/6470 [21:18<1:00:27,  1.38it/s]
1|1456|Loss: 4.579724311828613:  22%|██▏       | 1455/6470 [21:19<1:00:27,  1.38it/s]
1|1456|Loss: 4.579724311828613:  23%|██▎       | 1456/6470 [21:19<57:21,  1.46it/s]  
1|1457|Loss: 4.777480125427246:  23%|██▎       | 1456/6470 [21:19<57:21,  1.46it/s]
1|1457|Loss: 4.777480125427246:  23%|██▎       | 1457/6470 [21:20<59:03,  1.41it/s]
1|1458|Loss: 4.789614200592041:  23%|██▎       | 1457/6470 [21:20<59:03,  1.41it/s]
1|1458|Loss: 4.789614200592041:  23%|██▎       | 1458/6470 [21:21<1:02:54,  1.33it/s]
1|1459|Loss: 4.593945503234863:  23%|██▎       | 1458/6470 [21:21<1:02:54,  1.33it/s]
1|1459|Loss: 4.593945503234863:  23%|██▎       | 1459/6470 [21:21<59:27,  1.40it/s]  
1|1460|Loss: 4.827998638153076:  23%|██▎       | 1459/6470 [21:22<59:27,  1.40it/s]
1|1460|Loss: 4.827998638153076:  23%|██▎       | 1460/6470 [21:22<59:20,  1.41it/s]
1|1461|Loss: 4.969372272491455:  23%|██▎       | 1460/6470 [21:22<59:20,  1.41it/s]
1|1461|Loss: 4.969372272491455:  23%|██▎       | 1461/6470 [21:23<1:03:22,  1.32it/s]
1|1462|Loss: 5.378926753997803:  23%|██▎       | 1461/6470 [21:23<1:03:22,  1.32it/s]
1|1462|Loss: 5.378926753997803:  23%|██▎       | 1462/6470 [21:24<1:05:10,  1.28it/s]
1|1463|Loss: 4.854235649108887:  23%|██▎       | 1462/6470 [21:24<1:05:10,  1.28it/s]
1|1463|Loss: 4.854235649108887:  23%|██▎       | 1463/6470 [21:24<1:03:44,  1.31it/s]
1|1464|Loss: 4.946134090423584:  23%|██▎       | 1463/6470 [21:25<1:03:44,  1.31it/s]
1|1464|Loss: 4.946134090423584:  23%|██▎       | 1464/6470 [21:25<1:03:23,  1.32it/s]
1|1465|Loss: 5.337272644042969:  23%|██▎       | 1464/6470 [21:25<1:03:23,  1.32it/s]
1|1465|Loss: 5.337272644042969:  23%|██▎       | 1465/6470 [21:26<1:03:43,  1.31it/s]
1|1466|Loss: 5.379128456115723:  23%|██▎       | 1465/6470 [21:26<1:03:43,  1.31it/s]
1|1466|Loss: 5.379128456115723:  23%|██▎       | 1466/6470 [21:27<1:07:11,  1.24it/s]
1|1467|Loss: 5.285252571105957:  23%|██▎       | 1466/6470 [21:27<1:07:11,  1.24it/s]
1|1467|Loss: 5.285252571105957:  23%|██▎       | 1467/6470 [21:28<1:07:23,  1.24it/s]
1|1468|Loss: 4.3971757888793945:  23%|██▎       | 1467/6470 [21:28<1:07:23,  1.24it/s]
1|1468|Loss: 4.3971757888793945:  23%|██▎       | 1468/6470 [21:28<1:03:52,  1.31it/s]
1|1469|Loss: 5.075925827026367:  23%|██▎       | 1468/6470 [21:29<1:03:52,  1.31it/s] 
1|1469|Loss: 5.075925827026367:  23%|██▎       | 1469/6470 [21:29<1:01:34,  1.35it/s]
1|1470|Loss: 4.861908435821533:  23%|██▎       | 1469/6470 [21:29<1:01:34,  1.35it/s]
1|1470|Loss: 4.861908435821533:  23%|██▎       | 1470/6470 [21:30<1:05:45,  1.27it/s]
1|1471|Loss: 4.471346855163574:  23%|██▎       | 1470/6470 [21:30<1:05:45,  1.27it/s]
1|1471|Loss: 4.471346855163574:  23%|██▎       | 1471/6470 [21:31<1:06:26,  1.25it/s]
1|1472|Loss: 4.90428352355957:  23%|██▎       | 1471/6470 [21:31<1:06:26,  1.25it/s] 
1|1472|Loss: 4.90428352355957:  23%|██▎       | 1472/6470 [21:31<1:08:10,  1.22it/s]
1|1473|Loss: 5.237273216247559:  23%|██▎       | 1472/6470 [21:32<1:08:10,  1.22it/s]
1|1473|Loss: 5.237273216247559:  23%|██▎       | 1473/6470 [21:32<1:08:48,  1.21it/s]
1|1474|Loss: 4.9630608558654785:  23%|██▎       | 1473/6470 [21:33<1:08:48,  1.21it/s]
1|1474|Loss: 4.9630608558654785:  23%|██▎       | 1474/6470 [21:33<1:03:13,  1.32it/s]
1|1475|Loss: 5.017241954803467:  23%|██▎       | 1474/6470 [21:33<1:03:13,  1.32it/s] 
1|1475|Loss: 5.017241954803467:  23%|██▎       | 1475/6470 [21:34<1:02:15,  1.34it/s]
1|1476|Loss: 4.481811046600342:  23%|██▎       | 1475/6470 [21:34<1:02:15,  1.34it/s]
1|1476|Loss: 4.481811046600342:  23%|██▎       | 1476/6470 [21:34<1:01:21,  1.36it/s]
1|1477|Loss: 5.299872875213623:  23%|██▎       | 1476/6470 [21:35<1:01:21,  1.36it/s]
1|1477|Loss: 5.299872875213623:  23%|██▎       | 1477/6470 [21:35<1:02:26,  1.33it/s]
1|1478|Loss: 4.729557991027832:  23%|██▎       | 1477/6470 [21:36<1:02:26,  1.33it/s]
1|1478|Loss: 4.729557991027832:  23%|██▎       | 1478/6470 [21:36<1:02:52,  1.32it/s]
1|1479|Loss: 4.993478775024414:  23%|██▎       | 1478/6470 [21:36<1:02:52,  1.32it/s]
1|1479|Loss: 4.993478775024414:  23%|██▎       | 1479/6470 [21:37<1:03:04,  1.32it/s]
1|1480|Loss: 4.63534688949585:  23%|██▎       | 1479/6470 [21:37<1:03:04,  1.32it/s] 
1|1480|Loss: 4.63534688949585:  23%|██▎       | 1480/6470 [21:37<1:03:10,  1.32it/s]
1|1481|Loss: 4.518458366394043:  23%|██▎       | 1480/6470 [21:38<1:03:10,  1.32it/s]
1|1481|Loss: 4.518458366394043:  23%|██▎       | 1481/6470 [21:38<1:00:23,  1.38it/s]
1|1482|Loss: 5.311977386474609:  23%|██▎       | 1481/6470 [21:38<1:00:23,  1.38it/s]
1|1482|Loss: 5.311977386474609:  23%|██▎       | 1482/6470 [21:39<1:01:25,  1.35it/s]
1|1483|Loss: 5.043342113494873:  23%|██▎       | 1482/6470 [21:39<1:01:25,  1.35it/s]
1|1483|Loss: 5.043342113494873:  23%|██▎       | 1483/6470 [21:40<1:02:07,  1.34it/s]
1|1484|Loss: 4.918396472930908:  23%|██▎       | 1483/6470 [21:40<1:02:07,  1.34it/s]
1|1484|Loss: 4.918396472930908:  23%|██▎       | 1484/6470 [21:40<59:19,  1.40it/s]  
1|1485|Loss: 4.816133975982666:  23%|██▎       | 1484/6470 [21:41<59:19,  1.40it/s]
1|1485|Loss: 4.816133975982666:  23%|██▎       | 1485/6470 [21:41<56:55,  1.46it/s]
1|1486|Loss: 5.0453314781188965:  23%|██▎       | 1485/6470 [21:41<56:55,  1.46it/s]
1|1486|Loss: 5.0453314781188965:  23%|██▎       | 1486/6470 [21:41<55:01,  1.51it/s]
1|1487|Loss: 5.2889604568481445:  23%|██▎       | 1486/6470 [21:42<55:01,  1.51it/s]
1|1487|Loss: 5.2889604568481445:  23%|██▎       | 1487/6470 [21:42<55:49,  1.49it/s]
1|1488|Loss: 4.369231224060059:  23%|██▎       | 1487/6470 [21:43<55:49,  1.49it/s] 
1|1488|Loss: 4.369231224060059:  23%|██▎       | 1488/6470 [21:43<53:39,  1.55it/s]
1|1489|Loss: 5.065928936004639:  23%|██▎       | 1488/6470 [21:43<53:39,  1.55it/s]
1|1489|Loss: 5.065928936004639:  23%|██▎       | 1489/6470 [21:43<53:41,  1.55it/s]
1|1490|Loss: 5.366151332855225:  23%|██▎       | 1489/6470 [21:44<53:41,  1.55it/s]
1|1490|Loss: 5.366151332855225:  23%|██▎       | 1490/6470 [21:44<56:17,  1.47it/s]
1|1491|Loss: 5.121876239776611:  23%|██▎       | 1490/6470 [21:45<56:17,  1.47it/s]
1|1491|Loss: 5.121876239776611:  23%|██▎       | 1491/6470 [21:45<57:15,  1.45it/s]
1|1492|Loss: 4.861780643463135:  23%|██▎       | 1491/6470 [21:45<57:15,  1.45it/s]
1|1492|Loss: 4.861780643463135:  23%|██▎       | 1492/6470 [21:46<57:54,  1.43it/s]
1|1493|Loss: 4.759934425354004:  23%|██▎       | 1492/6470 [21:46<57:54,  1.43it/s]
1|1493|Loss: 4.759934425354004:  23%|██▎       | 1493/6470 [21:46<59:06,  1.40it/s]
1|1494|Loss: 4.903611660003662:  23%|██▎       | 1493/6470 [21:47<59:06,  1.40it/s]
1|1494|Loss: 4.903611660003662:  23%|██▎       | 1494/6470 [21:47<59:05,  1.40it/s]
1|1495|Loss: 4.636834621429443:  23%|██▎       | 1494/6470 [21:47<59:05,  1.40it/s]
1|1495|Loss: 4.636834621429443:  23%|██▎       | 1495/6470 [21:48<57:21,  1.45it/s]
1|1496|Loss: 5.082964897155762:  23%|██▎       | 1495/6470 [21:48<57:21,  1.45it/s]
1|1496|Loss: 5.082964897155762:  23%|██▎       | 1496/6470 [21:48<59:18,  1.40it/s]
1|1497|Loss: 4.902812957763672:  23%|██▎       | 1496/6470 [21:49<59:18,  1.40it/s]
1|1497|Loss: 4.902812957763672:  23%|██▎       | 1497/6470 [21:49<56:58,  1.45it/s]
1|1498|Loss: 5.082154750823975:  23%|██▎       | 1497/6470 [21:49<56:58,  1.45it/s]
1|1498|Loss: 5.082154750823975:  23%|██▎       | 1498/6470 [21:50<58:30,  1.42it/s]
1|1499|Loss: 5.317807674407959:  23%|██▎       | 1498/6470 [21:50<58:30,  1.42it/s]
1|1499|Loss: 5.317807674407959:  23%|██▎       | 1499/6470 [21:51<1:02:51,  1.32it/s]
1|1500|Loss: 4.704019546508789:  23%|██▎       | 1499/6470 [21:51<1:02:51,  1.32it/s]
1|1500|Loss: 4.704019546508789:  23%|██▎       | 1500/6470 [21:51<1:03:36,  1.30it/s]
1|1501|Loss: 5.102283954620361:  23%|██▎       | 1500/6470 [21:52<1:03:36,  1.30it/s]
1|1501|Loss: 5.102283954620361:  23%|██▎       | 1501/6470 [21:52<1:08:32,  1.21it/s]
1|1502|Loss: 5.3177266120910645:  23%|██▎       | 1501/6470 [21:53<1:08:32,  1.21it/s]
1|1502|Loss: 5.3177266120910645:  23%|██▎       | 1502/6470 [21:54<1:18:07,  1.06it/s]
1|1503|Loss: 4.8812150955200195:  23%|██▎       | 1502/6470 [21:54<1:18:07,  1.06it/s]
1|1503|Loss: 4.8812150955200195:  23%|██▎       | 1503/6470 [21:54<1:12:02,  1.15it/s]
1|1504|Loss: 5.219593524932861:  23%|██▎       | 1503/6470 [21:55<1:12:02,  1.15it/s] 
1|1504|Loss: 5.219593524932861:  23%|██▎       | 1504/6470 [21:55<1:09:19,  1.19it/s]
1|1505|Loss: 4.703258514404297:  23%|██▎       | 1504/6470 [21:56<1:09:19,  1.19it/s]
1|1505|Loss: 4.703258514404297:  23%|██▎       | 1505/6470 [21:56<1:04:14,  1.29it/s]
1|1506|Loss: 5.117099761962891:  23%|██▎       | 1505/6470 [21:56<1:04:14,  1.29it/s]
1|1506|Loss: 5.117099761962891:  23%|██▎       | 1506/6470 [21:56<1:00:22,  1.37it/s]
1|1507|Loss: 5.048677444458008:  23%|██▎       | 1506/6470 [21:57<1:00:22,  1.37it/s]
1|1507|Loss: 5.048677444458008:  23%|██▎       | 1507/6470 [21:57<1:01:37,  1.34it/s]
1|1508|Loss: 4.799874782562256:  23%|██▎       | 1507/6470 [21:58<1:01:37,  1.34it/s]
1|1508|Loss: 4.799874782562256:  23%|██▎       | 1508/6470 [21:58<59:45,  1.38it/s]  
1|1509|Loss: 5.043576717376709:  23%|██▎       | 1508/6470 [21:58<59:45,  1.38it/s]
1|1509|Loss: 5.043576717376709:  23%|██▎       | 1509/6470 [21:58<57:20,  1.44it/s]
1|1510|Loss: 5.343689441680908:  23%|██▎       | 1509/6470 [21:59<57:20,  1.44it/s]
1|1510|Loss: 5.343689441680908:  23%|██▎       | 1510/6470 [21:59<58:44,  1.41it/s]
1|1511|Loss: 5.110406398773193:  23%|██▎       | 1510/6470 [22:00<58:44,  1.41it/s]
1|1511|Loss: 5.110406398773193:  23%|██▎       | 1511/6470 [22:00<59:48,  1.38it/s]
1|1512|Loss: 5.418425559997559:  23%|██▎       | 1511/6470 [22:00<59:48,  1.38it/s]
1|1512|Loss: 5.418425559997559:  23%|██▎       | 1512/6470 [22:01<1:01:46,  1.34it/s]
1|1513|Loss: 5.557167053222656:  23%|██▎       | 1512/6470 [22:01<1:01:46,  1.34it/s]
1|1513|Loss: 5.557167053222656:  23%|██▎       | 1513/6470 [22:02<1:06:38,  1.24it/s]
1|1514|Loss: 5.192817687988281:  23%|██▎       | 1513/6470 [22:02<1:06:38,  1.24it/s]
1|1514|Loss: 5.192817687988281:  23%|██▎       | 1514/6470 [22:03<1:09:46,  1.18it/s]
1|1515|Loss: 4.989720821380615:  23%|██▎       | 1514/6470 [22:03<1:09:46,  1.18it/s]
1|1515|Loss: 4.989720821380615:  23%|██▎       | 1515/6470 [22:03<1:04:24,  1.28it/s]
1|1516|Loss: 4.593389511108398:  23%|██▎       | 1515/6470 [22:04<1:04:24,  1.28it/s]
1|1516|Loss: 4.593389511108398:  23%|██▎       | 1516/6470 [22:04<1:02:40,  1.32it/s]
1|1517|Loss: 5.078245162963867:  23%|██▎       | 1516/6470 [22:04<1:02:40,  1.32it/s]
1|1517|Loss: 5.078245162963867:  23%|██▎       | 1517/6470 [22:05<58:55,  1.40it/s]  
1|1518|Loss: 4.977163314819336:  23%|██▎       | 1517/6470 [22:05<58:55,  1.40it/s]
1|1518|Loss: 4.977163314819336:  23%|██▎       | 1518/6470 [22:05<59:59,  1.38it/s]
1|1519|Loss: 4.862263202667236:  23%|██▎       | 1518/6470 [22:06<59:59,  1.38it/s]
1|1519|Loss: 4.862263202667236:  23%|██▎       | 1519/6470 [22:06<59:52,  1.38it/s]
1|1520|Loss: 4.312783718109131:  23%|██▎       | 1519/6470 [22:06<59:52,  1.38it/s]
1|1520|Loss: 4.312783718109131:  23%|██▎       | 1520/6470 [22:07<58:00,  1.42it/s]
1|1521|Loss: 4.823154926300049:  23%|██▎       | 1520/6470 [22:07<58:00,  1.42it/s]
1|1521|Loss: 4.823154926300049:  24%|██▎       | 1521/6470 [22:07<56:45,  1.45it/s]
1|1522|Loss: 4.957106590270996:  24%|██▎       | 1521/6470 [22:08<56:45,  1.45it/s]
1|1522|Loss: 4.957106590270996:  24%|██▎       | 1522/6470 [22:08<56:58,  1.45it/s]
1|1523|Loss: 5.086127281188965:  24%|██▎       | 1522/6470 [22:08<56:58,  1.45it/s]
1|1523|Loss: 5.086127281188965:  24%|██▎       | 1523/6470 [22:09<58:30,  1.41it/s]
1|1524|Loss: 5.610914707183838:  24%|██▎       | 1523/6470 [22:09<58:30,  1.41it/s]
1|1524|Loss: 5.610914707183838:  24%|██▎       | 1524/6470 [22:10<1:01:09,  1.35it/s]
1|1525|Loss: 4.9500651359558105:  24%|██▎       | 1524/6470 [22:10<1:01:09,  1.35it/s]
1|1525|Loss: 4.9500651359558105:  24%|██▎       | 1525/6470 [22:10<1:01:02,  1.35it/s]
1|1526|Loss: 5.26507043838501:  24%|██▎       | 1525/6470 [22:11<1:01:02,  1.35it/s]  
1|1526|Loss: 5.26507043838501:  24%|██▎       | 1526/6470 [22:11<1:04:07,  1.29it/s]
1|1527|Loss: 5.433947563171387:  24%|██▎       | 1526/6470 [22:12<1:04:07,  1.29it/s]
1|1527|Loss: 5.433947563171387:  24%|██▎       | 1527/6470 [22:12<1:06:35,  1.24it/s]
1|1528|Loss: 4.836202144622803:  24%|██▎       | 1527/6470 [22:13<1:06:35,  1.24it/s]
1|1528|Loss: 4.836202144622803:  24%|██▎       | 1528/6470 [22:13<1:04:25,  1.28it/s]
1|1529|Loss: 4.887305736541748:  24%|██▎       | 1528/6470 [22:13<1:04:25,  1.28it/s]
1|1529|Loss: 4.887305736541748:  24%|██▎       | 1529/6470 [22:14<1:04:13,  1.28it/s]
1|1530|Loss: 4.310898303985596:  24%|██▎       | 1529/6470 [22:14<1:04:13,  1.28it/s]
1|1530|Loss: 4.310898303985596:  24%|██▎       | 1530/6470 [22:14<1:01:12,  1.35it/s]
1|1531|Loss: 4.926581859588623:  24%|██▎       | 1530/6470 [22:15<1:01:12,  1.35it/s]
1|1531|Loss: 4.926581859588623:  24%|██▎       | 1531/6470 [22:15<1:00:46,  1.35it/s]
1|1532|Loss: 5.17212438583374:  24%|██▎       | 1531/6470 [22:15<1:00:46,  1.35it/s] 
1|1532|Loss: 5.17212438583374:  24%|██▎       | 1532/6470 [22:16<59:51,  1.37it/s]  
1|1533|Loss: 4.816970348358154:  24%|██▎       | 1532/6470 [22:16<59:51,  1.37it/s]
1|1533|Loss: 4.816970348358154:  24%|██▎       | 1533/6470 [22:16<57:44,  1.42it/s]
1|1534|Loss: 5.078226089477539:  24%|██▎       | 1533/6470 [22:17<57:44,  1.42it/s]
1|1534|Loss: 5.078226089477539:  24%|██▎       | 1534/6470 [22:17<56:38,  1.45it/s]
1|1535|Loss: 5.4100871086120605:  24%|██▎       | 1534/6470 [22:17<56:38,  1.45it/s]
1|1535|Loss: 5.4100871086120605:  24%|██▎       | 1535/6470 [22:18<58:52,  1.40it/s]
1|1536|Loss: 5.045543193817139:  24%|██▎       | 1535/6470 [22:18<58:52,  1.40it/s] 
1|1536|Loss: 5.045543193817139:  24%|██▎       | 1536/6470 [22:19<1:00:56,  1.35it/s]
1|1537|Loss: 5.1016435623168945:  24%|██▎       | 1536/6470 [22:19<1:00:56,  1.35it/s]
1|1537|Loss: 5.1016435623168945:  24%|██▍       | 1537/6470 [22:19<59:31,  1.38it/s]  
1|1538|Loss: 4.652373313903809:  24%|██▍       | 1537/6470 [22:20<59:31,  1.38it/s] 
1|1538|Loss: 4.652373313903809:  24%|██▍       | 1538/6470 [22:20<58:03,  1.42it/s]
1|1539|Loss: 5.086601257324219:  24%|██▍       | 1538/6470 [22:20<58:03,  1.42it/s]
1|1539|Loss: 5.086601257324219:  24%|██▍       | 1539/6470 [22:21<1:00:05,  1.37it/s]
1|1540|Loss: 4.657416343688965:  24%|██▍       | 1539/6470 [22:21<1:00:05,  1.37it/s]
1|1540|Loss: 4.657416343688965:  24%|██▍       | 1540/6470 [22:22<1:01:18,  1.34it/s]
1|1541|Loss: 4.830776214599609:  24%|██▍       | 1540/6470 [22:22<1:01:18,  1.34it/s]
1|1541|Loss: 4.830776214599609:  24%|██▍       | 1541/6470 [22:22<1:01:26,  1.34it/s]
1|1542|Loss: 3.925400733947754:  24%|██▍       | 1541/6470 [22:23<1:01:26,  1.34it/s]
1|1542|Loss: 3.925400733947754:  24%|██▍       | 1542/6470 [22:23<58:11,  1.41it/s]  
1|1543|Loss: 4.7595415115356445:  24%|██▍       | 1542/6470 [22:23<58:11,  1.41it/s]
1|1543|Loss: 4.7595415115356445:  24%|██▍       | 1543/6470 [22:24<57:22,  1.43it/s]
1|1544|Loss: 5.233638763427734:  24%|██▍       | 1543/6470 [22:24<57:22,  1.43it/s] 
1|1544|Loss: 5.233638763427734:  24%|██▍       | 1544/6470 [22:24<58:57,  1.39it/s]
1|1545|Loss: 5.087804317474365:  24%|██▍       | 1544/6470 [22:25<58:57,  1.39it/s]
1|1545|Loss: 5.087804317474365:  24%|██▍       | 1545/6470 [22:25<1:00:59,  1.35it/s]
1|1546|Loss: 4.979197025299072:  24%|██▍       | 1545/6470 [22:26<1:00:59,  1.35it/s]
1|1546|Loss: 4.979197025299072:  24%|██▍       | 1546/6470 [22:26<1:00:40,  1.35it/s]
1|1547|Loss: 4.815568447113037:  24%|██▍       | 1546/6470 [22:26<1:00:40,  1.35it/s]
1|1547|Loss: 4.815568447113037:  24%|██▍       | 1547/6470 [22:27<1:00:48,  1.35it/s]
1|1548|Loss: 5.3171772956848145:  24%|██▍       | 1547/6470 [22:27<1:00:48,  1.35it/s]
1|1548|Loss: 5.3171772956848145:  24%|██▍       | 1548/6470 [22:27<1:01:11,  1.34it/s]
1|1549|Loss: 4.8636579513549805:  24%|██▍       | 1548/6470 [22:28<1:01:11,  1.34it/s]
1|1549|Loss: 4.8636579513549805:  24%|██▍       | 1549/6470 [22:28<1:01:50,  1.33it/s]
1|1550|Loss: 4.986620903015137:  24%|██▍       | 1549/6470 [22:29<1:01:50,  1.33it/s] 
1|1550|Loss: 4.986620903015137:  24%|██▍       | 1550/6470 [22:29<1:00:54,  1.35it/s]
1|1551|Loss: 4.909533977508545:  24%|██▍       | 1550/6470 [22:29<1:00:54,  1.35it/s]
1|1551|Loss: 4.909533977508545:  24%|██▍       | 1551/6470 [22:29<57:57,  1.41it/s]  
1|1552|Loss: 5.378458023071289:  24%|██▍       | 1551/6470 [22:30<57:57,  1.41it/s]
1|1552|Loss: 5.378458023071289:  24%|██▍       | 1552/6470 [22:30<1:01:37,  1.33it/s]
1|1553|Loss: 5.237524509429932:  24%|██▍       | 1552/6470 [22:31<1:01:37,  1.33it/s]
1|1553|Loss: 5.237524509429932:  24%|██▍       | 1553/6470 [22:31<1:05:55,  1.24it/s]
1|1554|Loss: 5.4420552253723145:  24%|██▍       | 1553/6470 [22:32<1:05:55,  1.24it/s]
1|1554|Loss: 5.4420552253723145:  24%|██▍       | 1554/6470 [22:32<1:08:53,  1.19it/s]
1|1555|Loss: 5.406954288482666:  24%|██▍       | 1554/6470 [22:33<1:08:53,  1.19it/s] 
1|1555|Loss: 5.406954288482666:  24%|██▍       | 1555/6470 [22:33<1:09:44,  1.17it/s]
1|1556|Loss: 4.401863098144531:  24%|██▍       | 1555/6470 [22:33<1:09:44,  1.17it/s]
1|1556|Loss: 4.401863098144531:  24%|██▍       | 1556/6470 [22:34<1:04:52,  1.26it/s]
1|1557|Loss: 4.4955878257751465:  24%|██▍       | 1556/6470 [22:34<1:04:52,  1.26it/s]
1|1557|Loss: 4.4955878257751465:  24%|██▍       | 1557/6470 [22:34<1:01:46,  1.33it/s]
1|1558|Loss: 5.093407154083252:  24%|██▍       | 1557/6470 [22:35<1:01:46,  1.33it/s] 
1|1558|Loss: 5.093407154083252:  24%|██▍       | 1558/6470 [22:35<1:02:43,  1.31it/s]
1|1559|Loss: 5.154027938842773:  24%|██▍       | 1558/6470 [22:36<1:02:43,  1.31it/s]
1|1559|Loss: 5.154027938842773:  24%|██▍       | 1559/6470 [22:36<1:03:49,  1.28it/s]
1|1560|Loss: 4.892890930175781:  24%|██▍       | 1559/6470 [22:36<1:03:49,  1.28it/s]
1|1560|Loss: 4.892890930175781:  24%|██▍       | 1560/6470 [22:37<1:06:41,  1.23it/s]
1|1561|Loss: 4.976705074310303:  24%|██▍       | 1560/6470 [22:37<1:06:41,  1.23it/s]
1|1561|Loss: 4.976705074310303:  24%|██▍       | 1561/6470 [22:38<1:05:00,  1.26it/s]
1|1562|Loss: 5.406408309936523:  24%|██▍       | 1561/6470 [22:38<1:05:00,  1.26it/s]
1|1562|Loss: 5.406408309936523:  24%|██▍       | 1562/6470 [22:38<1:05:48,  1.24it/s]
1|1563|Loss: 4.622292995452881:  24%|██▍       | 1562/6470 [22:39<1:05:48,  1.24it/s]
1|1563|Loss: 4.622292995452881:  24%|██▍       | 1563/6470 [22:39<1:02:05,  1.32it/s]
1|1564|Loss: 5.22581148147583:  24%|██▍       | 1563/6470 [22:39<1:02:05,  1.32it/s] 
1|1564|Loss: 5.22581148147583:  24%|██▍       | 1564/6470 [22:40<1:01:19,  1.33it/s]
1|1565|Loss: 5.062490940093994:  24%|██▍       | 1564/6470 [22:40<1:01:19,  1.33it/s]
1|1565|Loss: 5.062490940093994:  24%|██▍       | 1565/6470 [22:41<1:01:34,  1.33it/s]
1|1566|Loss: 5.037630081176758:  24%|██▍       | 1565/6470 [22:41<1:01:34,  1.33it/s]
1|1566|Loss: 5.037630081176758:  24%|██▍       | 1566/6470 [22:41<1:02:33,  1.31it/s]
1|1567|Loss: 5.425173759460449:  24%|██▍       | 1566/6470 [22:42<1:02:33,  1.31it/s]
1|1567|Loss: 5.425173759460449:  24%|██▍       | 1567/6470 [22:42<1:04:50,  1.26it/s]
1|1568|Loss: 4.72787618637085:  24%|██▍       | 1567/6470 [22:43<1:04:50,  1.26it/s] 
1|1568|Loss: 4.72787618637085:  24%|██▍       | 1568/6470 [22:43<1:05:40,  1.24it/s]
1|1569|Loss: 4.296340465545654:  24%|██▍       | 1568/6470 [22:43<1:05:40,  1.24it/s]
1|1569|Loss: 4.296340465545654:  24%|██▍       | 1569/6470 [22:44<1:02:17,  1.31it/s]
1|1570|Loss: 4.621351718902588:  24%|██▍       | 1569/6470 [22:44<1:02:17,  1.31it/s]
1|1570|Loss: 4.621351718902588:  24%|██▍       | 1570/6470 [22:45<1:03:42,  1.28it/s]
1|1571|Loss: 3.7394063472747803:  24%|██▍       | 1570/6470 [22:45<1:03:42,  1.28it/s]
1|1571|Loss: 3.7394063472747803:  24%|██▍       | 1571/6470 [22:45<59:20,  1.38it/s]  
1|1572|Loss: 4.885801315307617:  24%|██▍       | 1571/6470 [22:46<59:20,  1.38it/s] 
1|1572|Loss: 4.885801315307617:  24%|██▍       | 1572/6470 [22:46<58:49,  1.39it/s]
1|1573|Loss: 5.33651876449585:  24%|██▍       | 1572/6470 [22:46<58:49,  1.39it/s] 
1|1573|Loss: 5.33651876449585:  24%|██▍       | 1573/6470 [22:47<1:02:10,  1.31it/s]
1|1574|Loss: 4.863540172576904:  24%|██▍       | 1573/6470 [22:47<1:02:10,  1.31it/s]
1|1574|Loss: 4.863540172576904:  24%|██▍       | 1574/6470 [22:48<1:03:36,  1.28it/s]
1|1575|Loss: 5.026980400085449:  24%|██▍       | 1574/6470 [22:48<1:03:36,  1.28it/s]
1|1575|Loss: 5.026980400085449:  24%|██▍       | 1575/6470 [22:48<1:02:28,  1.31it/s]
1|1576|Loss: 4.714412689208984:  24%|██▍       | 1575/6470 [22:49<1:02:28,  1.31it/s]
1|1576|Loss: 4.714412689208984:  24%|██▍       | 1576/6470 [22:49<1:03:01,  1.29it/s]
1|1577|Loss: 5.126096725463867:  24%|██▍       | 1576/6470 [22:49<1:03:01,  1.29it/s]
1|1577|Loss: 5.126096725463867:  24%|██▍       | 1577/6470 [22:50<1:00:25,  1.35it/s]
1|1578|Loss: 5.181324481964111:  24%|██▍       | 1577/6470 [22:50<1:00:25,  1.35it/s]
1|1578|Loss: 5.181324481964111:  24%|██▍       | 1578/6470 [22:51<1:01:17,  1.33it/s]
1|1579|Loss: 5.259474754333496:  24%|██▍       | 1578/6470 [22:51<1:01:17,  1.33it/s]
1|1579|Loss: 5.259474754333496:  24%|██▍       | 1579/6470 [22:51<1:02:31,  1.30it/s]
1|1580|Loss: 4.529947757720947:  24%|██▍       | 1579/6470 [22:52<1:02:31,  1.30it/s]
1|1580|Loss: 4.529947757720947:  24%|██▍       | 1580/6470 [22:52<1:00:02,  1.36it/s]
1|1581|Loss: 5.292140007019043:  24%|██▍       | 1580/6470 [22:52<1:00:02,  1.36it/s]
1|1581|Loss: 5.292140007019043:  24%|██▍       | 1581/6470 [22:53<1:00:10,  1.35it/s]
1|1582|Loss: 4.875998497009277:  24%|██▍       | 1581/6470 [22:53<1:00:10,  1.35it/s]
1|1582|Loss: 4.875998497009277:  24%|██▍       | 1582/6470 [22:53<58:05,  1.40it/s]  
1|1583|Loss: 5.039625644683838:  24%|██▍       | 1582/6470 [22:54<58:05,  1.40it/s]
1|1583|Loss: 5.039625644683838:  24%|██▍       | 1583/6470 [22:54<58:29,  1.39it/s]
1|1584|Loss: 3.8715927600860596:  24%|██▍       | 1583/6470 [22:54<58:29,  1.39it/s]
1|1584|Loss: 3.8715927600860596:  24%|██▍       | 1584/6470 [22:55<55:11,  1.48it/s]
1|1585|Loss: 5.4345831871032715:  24%|██▍       | 1584/6470 [22:55<55:11,  1.48it/s]
1|1585|Loss: 5.4345831871032715:  24%|██▍       | 1585/6470 [22:56<59:20,  1.37it/s]
1|1586|Loss: 4.7606120109558105:  24%|██▍       | 1585/6470 [22:56<59:20,  1.37it/s]
1|1586|Loss: 4.7606120109558105:  25%|██▍       | 1586/6470 [22:56<1:00:39,  1.34it/s]
1|1587|Loss: 5.351412296295166:  25%|██▍       | 1586/6470 [22:57<1:00:39,  1.34it/s] 
1|1587|Loss: 5.351412296295166:  25%|██▍       | 1587/6470 [22:57<1:02:16,  1.31it/s]
1|1588|Loss: 4.7515082359313965:  25%|██▍       | 1587/6470 [22:57<1:02:16,  1.31it/s]
1|1588|Loss: 4.7515082359313965:  25%|██▍       | 1588/6470 [22:58<55:44,  1.46it/s]  
1|1589|Loss: 5.442857265472412:  25%|██▍       | 1588/6470 [22:58<55:44,  1.46it/s] 
1|1589|Loss: 5.442857265472412:  25%|██▍       | 1589/6470 [22:59<1:01:14,  1.33it/s]
1|1590|Loss: 4.6667914390563965:  25%|██▍       | 1589/6470 [22:59<1:01:14,  1.33it/s]
1|1590|Loss: 4.6667914390563965:  25%|██▍       | 1590/6470 [22:59<1:03:31,  1.28it/s]
1|1591|Loss: 5.337241172790527:  25%|██▍       | 1590/6470 [23:00<1:03:31,  1.28it/s] 
1|1591|Loss: 5.337241172790527:  25%|██▍       | 1591/6470 [23:00<1:03:58,  1.27it/s]
1|1592|Loss: 4.529200553894043:  25%|██▍       | 1591/6470 [23:01<1:03:58,  1.27it/s]
1|1592|Loss: 4.529200553894043:  25%|██▍       | 1592/6470 [23:01<1:07:00,  1.21it/s]
1|1593|Loss: 4.582864761352539:  25%|██▍       | 1592/6470 [23:02<1:07:00,  1.21it/s]
1|1593|Loss: 4.582864761352539:  25%|██▍       | 1593/6470 [23:02<1:05:55,  1.23it/s]
1|1594|Loss: 5.26484489440918:  25%|██▍       | 1593/6470 [23:02<1:05:55,  1.23it/s] 
1|1594|Loss: 5.26484489440918:  25%|██▍       | 1594/6470 [23:03<1:04:50,  1.25it/s]
1|1595|Loss: 5.308984279632568:  25%|██▍       | 1594/6470 [23:03<1:04:50,  1.25it/s]
1|1595|Loss: 5.308984279632568:  25%|██▍       | 1595/6470 [23:03<1:04:52,  1.25it/s]
1|1596|Loss: 4.531942367553711:  25%|██▍       | 1595/6470 [23:04<1:04:52,  1.25it/s]
1|1596|Loss: 4.531942367553711:  25%|██▍       | 1596/6470 [23:04<1:04:59,  1.25it/s]
1|1597|Loss: 5.2249755859375:  25%|██▍       | 1596/6470 [23:05<1:04:59,  1.25it/s]  
1|1597|Loss: 5.2249755859375:  25%|██▍       | 1597/6470 [23:05<1:04:16,  1.26it/s]
1|1598|Loss: 4.960761070251465:  25%|██▍       | 1597/6470 [23:05<1:04:16,  1.26it/s]
1|1598|Loss: 4.960761070251465:  25%|██▍       | 1598/6470 [23:06<1:04:06,  1.27it/s]
1|1599|Loss: 5.411720275878906:  25%|██▍       | 1598/6470 [23:06<1:04:06,  1.27it/s]
1|1599|Loss: 5.411720275878906:  25%|██▍       | 1599/6470 [23:07<1:04:28,  1.26it/s]
1|1600|Loss: 5.146651744842529:  25%|██▍       | 1599/6470 [23:07<1:04:28,  1.26it/s]
1|1600|Loss: 5.146651744842529:  25%|██▍       | 1600/6470 [23:07<1:05:45,  1.23it/s]
1|1601|Loss: 4.126594543457031:  25%|██▍       | 1600/6470 [23:08<1:05:45,  1.23it/s]
1|1601|Loss: 4.126594543457031:  25%|██▍       | 1601/6470 [23:08<1:00:52,  1.33it/s]
1|1602|Loss: 4.9345703125:  25%|██▍       | 1601/6470 [23:08<1:00:52,  1.33it/s]     
1|1602|Loss: 4.9345703125:  25%|██▍       | 1602/6470 [23:09<1:00:02,  1.35it/s]
1|1603|Loss: 4.2879533767700195:  25%|██▍       | 1602/6470 [23:09<1:00:02,  1.35it/s]
1|1603|Loss: 4.2879533767700195:  25%|██▍       | 1603/6470 [23:09<59:08,  1.37it/s]  
1|1604|Loss: 4.96567964553833:  25%|██▍       | 1603/6470 [23:10<59:08,  1.37it/s]  
1|1604|Loss: 4.96567964553833:  25%|██▍       | 1604/6470 [23:10<1:00:27,  1.34it/s]
1|1605|Loss: 4.8447442054748535:  25%|██▍       | 1604/6470 [23:11<1:00:27,  1.34it/s]
1|1605|Loss: 4.8447442054748535:  25%|██▍       | 1605/6470 [23:11<1:01:54,  1.31it/s]
1|1606|Loss: 4.636611461639404:  25%|██▍       | 1605/6470 [23:11<1:01:54,  1.31it/s] 
1|1606|Loss: 4.636611461639404:  25%|██▍       | 1606/6470 [23:12<56:55,  1.42it/s]  
1|1607|Loss: 5.103053569793701:  25%|██▍       | 1606/6470 [23:12<56:55,  1.42it/s]
1|1607|Loss: 5.103053569793701:  25%|██▍       | 1607/6470 [23:12<54:25,  1.49it/s]
1|1608|Loss: 5.36824893951416:  25%|██▍       | 1607/6470 [23:13<54:25,  1.49it/s] 
1|1608|Loss: 5.36824893951416:  25%|██▍       | 1608/6470 [23:13<59:24,  1.36it/s]
1|1609|Loss: 4.913968563079834:  25%|██▍       | 1608/6470 [23:14<59:24,  1.36it/s]
1|1609|Loss: 4.913968563079834:  25%|██▍       | 1609/6470 [23:14<1:03:47,  1.27it/s]
1|1610|Loss: 5.361828327178955:  25%|██▍       | 1609/6470 [23:15<1:03:47,  1.27it/s]
1|1610|Loss: 5.361828327178955:  25%|██▍       | 1610/6470 [23:15<1:07:20,  1.20it/s]
1|1611|Loss: 4.65300989151001:  25%|██▍       | 1610/6470 [23:15<1:07:20,  1.20it/s] 
1|1611|Loss: 4.65300989151001:  25%|██▍       | 1611/6470 [23:16<1:02:30,  1.30it/s]
1|1612|Loss: 5.120466709136963:  25%|██▍       | 1611/6470 [23:16<1:02:30,  1.30it/s]
1|1612|Loss: 5.120466709136963:  25%|██▍       | 1612/6470 [23:16<59:59,  1.35it/s]  
1|1613|Loss: 4.56010627746582:  25%|██▍       | 1612/6470 [23:17<59:59,  1.35it/s] 
1|1613|Loss: 4.56010627746582:  25%|██▍       | 1613/6470 [23:17<57:02,  1.42it/s]
1|1614|Loss: 4.9669084548950195:  25%|██▍       | 1613/6470 [23:17<57:02,  1.42it/s]
1|1614|Loss: 4.9669084548950195:  25%|██▍       | 1614/6470 [23:18<56:31,  1.43it/s]
1|1615|Loss: 5.121506690979004:  25%|██▍       | 1614/6470 [23:18<56:31,  1.43it/s] 
1|1615|Loss: 5.121506690979004:  25%|██▍       | 1615/6470 [23:18<55:20,  1.46it/s]
1|1616|Loss: 4.978365898132324:  25%|██▍       | 1615/6470 [23:19<55:20,  1.46it/s]
1|1616|Loss: 4.978365898132324:  25%|██▍       | 1616/6470 [23:19<52:46,  1.53it/s]
1|1617|Loss: 4.968018054962158:  25%|██▍       | 1616/6470 [23:19<52:46,  1.53it/s]
1|1617|Loss: 4.968018054962158:  25%|██▍       | 1617/6470 [23:20<55:13,  1.46it/s]
1|1618|Loss: 5.227860927581787:  25%|██▍       | 1617/6470 [23:20<55:13,  1.46it/s]
1|1618|Loss: 5.227860927581787:  25%|██▌       | 1618/6470 [23:20<1:00:32,  1.34it/s]
1|1619|Loss: 5.037115097045898:  25%|██▌       | 1618/6470 [23:21<1:00:32,  1.34it/s]
1|1619|Loss: 5.037115097045898:  25%|██▌       | 1619/6470 [23:21<1:00:13,  1.34it/s]
1|1620|Loss: 5.281020164489746:  25%|██▌       | 1619/6470 [23:22<1:00:13,  1.34it/s]
1|1620|Loss: 5.281020164489746:  25%|██▌       | 1620/6470 [23:22<1:01:22,  1.32it/s]
1|1621|Loss: 5.030029773712158:  25%|██▌       | 1620/6470 [23:22<1:01:22,  1.32it/s]
1|1621|Loss: 5.030029773712158:  25%|██▌       | 1621/6470 [23:23<1:02:26,  1.29it/s]
1|1622|Loss: 5.201103210449219:  25%|██▌       | 1621/6470 [23:23<1:02:26,  1.29it/s]
1|1622|Loss: 5.201103210449219:  25%|██▌       | 1622/6470 [23:23<1:00:32,  1.33it/s]
1|1623|Loss: 4.9550862312316895:  25%|██▌       | 1622/6470 [23:24<1:00:32,  1.33it/s]
1|1623|Loss: 4.9550862312316895:  25%|██▌       | 1623/6470 [23:24<1:01:22,  1.32it/s]
1|1624|Loss: 5.154857635498047:  25%|██▌       | 1623/6470 [23:25<1:01:22,  1.32it/s] 
1|1624|Loss: 5.154857635498047:  25%|██▌       | 1624/6470 [23:25<1:04:45,  1.25it/s]
1|1625|Loss: 6.3975138664245605:  25%|██▌       | 1624/6470 [23:26<1:04:45,  1.25it/s]
1|1625|Loss: 6.3975138664245605:  25%|██▌       | 1625/6470 [23:26<1:05:27,  1.23it/s]
1|1626|Loss: 5.940799236297607:  25%|██▌       | 1625/6470 [23:26<1:05:27,  1.23it/s] 
1|1626|Loss: 5.940799236297607:  25%|██▌       | 1626/6470 [23:27<1:07:44,  1.19it/s]
1|1627|Loss: 5.346569061279297:  25%|██▌       | 1626/6470 [23:27<1:07:44,  1.19it/s]
1|1627|Loss: 5.346569061279297:  25%|██▌       | 1627/6470 [23:28<1:04:24,  1.25it/s]
1|1628|Loss: 5.39933443069458:  25%|██▌       | 1627/6470 [23:28<1:04:24,  1.25it/s] 
1|1628|Loss: 5.39933443069458:  25%|██▌       | 1628/6470 [23:28<1:06:08,  1.22it/s]
1|1629|Loss: 4.596079349517822:  25%|██▌       | 1628/6470 [23:29<1:06:08,  1.22it/s]
1|1629|Loss: 4.596079349517822:  25%|██▌       | 1629/6470 [23:29<1:01:15,  1.32it/s]
1|1630|Loss: 4.847562313079834:  25%|██▌       | 1629/6470 [23:29<1:01:15,  1.32it/s]
1|1630|Loss: 4.847562313079834:  25%|██▌       | 1630/6470 [23:30<1:01:06,  1.32it/s]
1|1631|Loss: 4.729304790496826:  25%|██▌       | 1630/6470 [23:30<1:01:06,  1.32it/s]
1|1631|Loss: 4.729304790496826:  25%|██▌       | 1631/6470 [23:31<1:03:23,  1.27it/s]
1|1632|Loss: 5.42565393447876:  25%|██▌       | 1631/6470 [23:31<1:03:23,  1.27it/s] 
1|1632|Loss: 5.42565393447876:  25%|██▌       | 1632/6470 [23:32<1:08:12,  1.18it/s]
1|1633|Loss: 4.762849807739258:  25%|██▌       | 1632/6470 [23:32<1:08:12,  1.18it/s]
1|1633|Loss: 4.762849807739258:  25%|██▌       | 1633/6470 [23:33<1:10:47,  1.14it/s]
1|1634|Loss: 4.828693866729736:  25%|██▌       | 1633/6470 [23:33<1:10:47,  1.14it/s]
1|1634|Loss: 4.828693866729736:  25%|██▌       | 1634/6470 [23:33<1:07:15,  1.20it/s]
1|1635|Loss: 4.2877583503723145:  25%|██▌       | 1634/6470 [23:34<1:07:15,  1.20it/s]
1|1635|Loss: 4.2877583503723145:  25%|██▌       | 1635/6470 [23:34<1:00:36,  1.33it/s]
1|1636|Loss: 4.5444159507751465:  25%|██▌       | 1635/6470 [23:34<1:00:36,  1.33it/s]
1|1636|Loss: 4.5444159507751465:  25%|██▌       | 1636/6470 [23:34<55:36,  1.45it/s]  
1|1637|Loss: 4.774986267089844:  25%|██▌       | 1636/6470 [23:35<55:36,  1.45it/s] 
1|1637|Loss: 4.774986267089844:  25%|██▌       | 1637/6470 [23:35<54:39,  1.47it/s]
1|1638|Loss: 5.300380706787109:  25%|██▌       | 1637/6470 [23:36<54:39,  1.47it/s]
1|1638|Loss: 5.300380706787109:  25%|██▌       | 1638/6470 [23:36<56:38,  1.42it/s]
1|1639|Loss: 4.985599994659424:  25%|██▌       | 1638/6470 [23:36<56:38,  1.42it/s]
1|1639|Loss: 4.985599994659424:  25%|██▌       | 1639/6470 [23:37<1:01:40,  1.31it/s]
1|1640|Loss: 4.181095600128174:  25%|██▌       | 1639/6470 [23:37<1:01:40,  1.31it/s]
1|1640|Loss: 4.181095600128174:  25%|██▌       | 1640/6470 [23:37<59:56,  1.34it/s]  
1|1641|Loss: 4.821814060211182:  25%|██▌       | 1640/6470 [23:38<59:56,  1.34it/s]
1|1641|Loss: 4.821814060211182:  25%|██▌       | 1641/6470 [23:38<58:27,  1.38it/s]
1|1642|Loss: 4.817452430725098:  25%|██▌       | 1641/6470 [23:39<58:27,  1.38it/s]
1|1642|Loss: 4.817452430725098:  25%|██▌       | 1642/6470 [23:39<58:58,  1.36it/s]
1|1643|Loss: 4.695510387420654:  25%|██▌       | 1642/6470 [23:39<58:58,  1.36it/s]
1|1643|Loss: 4.695510387420654:  25%|██▌       | 1643/6470 [23:40<58:28,  1.38it/s]
1|1644|Loss: 4.532629013061523:  25%|██▌       | 1643/6470 [23:40<58:28,  1.38it/s]
1|1644|Loss: 4.532629013061523:  25%|██▌       | 1644/6470 [23:40<56:21,  1.43it/s]
1|1645|Loss: 5.226356029510498:  25%|██▌       | 1644/6470 [23:41<56:21,  1.43it/s]
1|1645|Loss: 5.226356029510498:  25%|██▌       | 1645/6470 [23:41<57:48,  1.39it/s]
1|1646|Loss: 4.617908954620361:  25%|██▌       | 1645/6470 [23:41<57:48,  1.39it/s]
1|1646|Loss: 4.617908954620361:  25%|██▌       | 1646/6470 [23:42<58:44,  1.37it/s]
1|1647|Loss: 4.923166751861572:  25%|██▌       | 1646/6470 [23:42<58:44,  1.37it/s]
1|1647|Loss: 4.923166751861572:  25%|██▌       | 1647/6470 [23:43<58:18,  1.38it/s]
1|1648|Loss: 4.856832981109619:  25%|██▌       | 1647/6470 [23:43<58:18,  1.38it/s]
1|1648|Loss: 4.856832981109619:  25%|██▌       | 1648/6470 [23:43<1:00:57,  1.32it/s]
1|1649|Loss: 5.127013206481934:  25%|██▌       | 1648/6470 [23:44<1:00:57,  1.32it/s]
1|1649|Loss: 5.127013206481934:  25%|██▌       | 1649/6470 [23:44<1:01:32,  1.31it/s]
1|1650|Loss: 5.281871318817139:  25%|██▌       | 1649/6470 [23:45<1:01:32,  1.31it/s]
1|1650|Loss: 5.281871318817139:  26%|██▌       | 1650/6470 [23:45<1:02:32,  1.28it/s]
1|1651|Loss: 4.601633548736572:  26%|██▌       | 1650/6470 [23:45<1:02:32,  1.28it/s]
1|1651|Loss: 4.601633548736572:  26%|██▌       | 1651/6470 [23:46<1:01:25,  1.31it/s]
1|1652|Loss: 5.186320781707764:  26%|██▌       | 1651/6470 [23:46<1:01:25,  1.31it/s]
1|1652|Loss: 5.186320781707764:  26%|██▌       | 1652/6470 [23:46<1:01:45,  1.30it/s]
1|1653|Loss: 4.252360820770264:  26%|██▌       | 1652/6470 [23:47<1:01:45,  1.30it/s]
1|1653|Loss: 4.252360820770264:  26%|██▌       | 1653/6470 [23:47<57:01,  1.41it/s]  
1|1654|Loss: 4.914821624755859:  26%|██▌       | 1653/6470 [23:47<57:01,  1.41it/s]
1|1654|Loss: 4.914821624755859:  26%|██▌       | 1654/6470 [23:48<59:42,  1.34it/s]
1|1655|Loss: 4.913224697113037:  26%|██▌       | 1654/6470 [23:48<59:42,  1.34it/s]
1|1655|Loss: 4.913224697113037:  26%|██▌       | 1655/6470 [23:49<1:04:22,  1.25it/s]
1|1656|Loss: 5.402722358703613:  26%|██▌       | 1655/6470 [23:49<1:04:22,  1.25it/s]
1|1656|Loss: 5.402722358703613:  26%|██▌       | 1656/6470 [23:50<1:07:19,  1.19it/s]
1|1657|Loss: 4.566112041473389:  26%|██▌       | 1656/6470 [23:50<1:07:19,  1.19it/s]
1|1657|Loss: 4.566112041473389:  26%|██▌       | 1657/6470 [23:50<1:05:11,  1.23it/s]
1|1658|Loss: 4.917697429656982:  26%|██▌       | 1657/6470 [23:51<1:05:11,  1.23it/s]
1|1658|Loss: 4.917697429656982:  26%|██▌       | 1658/6470 [23:51<1:04:38,  1.24it/s]
1|1659|Loss: 4.61029577255249:  26%|██▌       | 1658/6470 [23:52<1:04:38,  1.24it/s] 
1|1659|Loss: 4.61029577255249:  26%|██▌       | 1659/6470 [23:52<1:05:34,  1.22it/s]
1|1660|Loss: 5.0311279296875:  26%|██▌       | 1659/6470 [23:53<1:05:34,  1.22it/s] 
1|1660|Loss: 5.0311279296875:  26%|██▌       | 1660/6470 [23:53<1:05:34,  1.22it/s]
1|1661|Loss: 4.983843803405762:  26%|██▌       | 1660/6470 [23:53<1:05:34,  1.22it/s]
1|1661|Loss: 4.983843803405762:  26%|██▌       | 1661/6470 [23:54<1:02:54,  1.27it/s]
1|1662|Loss: 5.092051982879639:  26%|██▌       | 1661/6470 [23:54<1:02:54,  1.27it/s]
1|1662|Loss: 5.092051982879639:  26%|██▌       | 1662/6470 [23:54<1:00:29,  1.32it/s]
1|1663|Loss: 4.755511283874512:  26%|██▌       | 1662/6470 [23:55<1:00:29,  1.32it/s]
1|1663|Loss: 4.755511283874512:  26%|██▌       | 1663/6470 [23:55<1:01:10,  1.31it/s]
1|1664|Loss: 5.547111988067627:  26%|██▌       | 1663/6470 [23:56<1:01:10,  1.31it/s]
1|1664|Loss: 5.547111988067627:  26%|██▌       | 1664/6470 [23:56<1:07:09,  1.19it/s]
1|1665|Loss: 5.597397804260254:  26%|██▌       | 1664/6470 [23:57<1:07:09,  1.19it/s]
1|1665|Loss: 5.597397804260254:  26%|██▌       | 1665/6470 [23:57<1:10:40,  1.13it/s]
1|1666|Loss: 4.860858917236328:  26%|██▌       | 1665/6470 [23:58<1:10:40,  1.13it/s]
1|1666|Loss: 4.860858917236328:  26%|██▌       | 1666/6470 [23:58<1:07:34,  1.18it/s]
1|1667|Loss: 5.022996425628662:  26%|██▌       | 1666/6470 [23:58<1:07:34,  1.18it/s]
1|1667|Loss: 5.022996425628662:  26%|██▌       | 1667/6470 [23:59<1:05:01,  1.23it/s]
1|1668|Loss: 4.616703033447266:  26%|██▌       | 1667/6470 [23:59<1:05:01,  1.23it/s]
1|1668|Loss: 4.616703033447266:  26%|██▌       | 1668/6470 [23:59<58:12,  1.37it/s]  
1|1669|Loss: 4.995319366455078:  26%|██▌       | 1668/6470 [24:00<58:12,  1.37it/s]
1|1669|Loss: 4.995319366455078:  26%|██▌       | 1669/6470 [24:00<1:00:19,  1.33it/s]
1|1670|Loss: 5.449856758117676:  26%|██▌       | 1669/6470 [24:00<1:00:19,  1.33it/s]
1|1670|Loss: 5.449856758117676:  26%|██▌       | 1670/6470 [24:01<1:03:51,  1.25it/s]
1|1671|Loss: 5.4263691902160645:  26%|██▌       | 1670/6470 [24:01<1:03:51,  1.25it/s]
1|1671|Loss: 5.4263691902160645:  26%|██▌       | 1671/6470 [24:02<1:06:46,  1.20it/s]
1|1672|Loss: 4.62283182144165:  26%|██▌       | 1671/6470 [24:02<1:06:46,  1.20it/s]  
1|1672|Loss: 4.62283182144165:  26%|██▌       | 1672/6470 [24:02<1:01:47,  1.29it/s]
1|1673|Loss: 4.727748394012451:  26%|██▌       | 1672/6470 [24:03<1:01:47,  1.29it/s]
1|1673|Loss: 4.727748394012451:  26%|██▌       | 1673/6470 [24:03<1:01:16,  1.30it/s]
1|1674|Loss: 5.200191497802734:  26%|██▌       | 1673/6470 [24:04<1:01:16,  1.30it/s]
1|1674|Loss: 5.200191497802734:  26%|██▌       | 1674/6470 [24:04<1:01:32,  1.30it/s]
1|1675|Loss: 4.548878192901611:  26%|██▌       | 1674/6470 [24:04<1:01:32,  1.30it/s]
1|1675|Loss: 4.548878192901611:  26%|██▌       | 1675/6470 [24:05<57:38,  1.39it/s]  
1|1676|Loss: 4.782780647277832:  26%|██▌       | 1675/6470 [24:05<57:38,  1.39it/s]
1|1676|Loss: 4.782780647277832:  26%|██▌       | 1676/6470 [24:05<55:46,  1.43it/s]
1|1677|Loss: 5.077944755554199:  26%|██▌       | 1676/6470 [24:06<55:46,  1.43it/s]
1|1677|Loss: 5.077944755554199:  26%|██▌       | 1677/6470 [24:06<59:54,  1.33it/s]
1|1678|Loss: 5.3650617599487305:  26%|██▌       | 1677/6470 [24:07<59:54,  1.33it/s]
1|1678|Loss: 5.3650617599487305:  26%|██▌       | 1678/6470 [24:07<1:03:56,  1.25it/s]
1|1679|Loss: 5.315333366394043:  26%|██▌       | 1678/6470 [24:07<1:03:56,  1.25it/s] 
1|1679|Loss: 5.315333366394043:  26%|██▌       | 1679/6470 [24:08<1:07:11,  1.19it/s]
1|1680|Loss: 5.335399150848389:  26%|██▌       | 1679/6470 [24:08<1:07:11,  1.19it/s]
1|1680|Loss: 5.335399150848389:  26%|██▌       | 1680/6470 [24:09<1:10:17,  1.14it/s]
1|1681|Loss: 4.992084503173828:  26%|██▌       | 1680/6470 [24:09<1:10:17,  1.14it/s]
1|1681|Loss: 4.992084503173828:  26%|██▌       | 1681/6470 [24:10<1:09:56,  1.14it/s]
1|1682|Loss: 5.113529205322266:  26%|██▌       | 1681/6470 [24:10<1:09:56,  1.14it/s]
1|1682|Loss: 5.113529205322266:  26%|██▌       | 1682/6470 [24:11<1:09:52,  1.14it/s]
1|1683|Loss: 4.0473527908325195:  26%|██▌       | 1682/6470 [24:11<1:09:52,  1.14it/s]
1|1683|Loss: 4.0473527908325195:  26%|██▌       | 1683/6470 [24:11<1:03:30,  1.26it/s]
1|1684|Loss: 5.216270923614502:  26%|██▌       | 1683/6470 [24:12<1:03:30,  1.26it/s] 
1|1684|Loss: 5.216270923614502:  26%|██▌       | 1684/6470 [24:12<1:01:46,  1.29it/s]
1|1685|Loss: 4.896644115447998:  26%|██▌       | 1684/6470 [24:12<1:01:46,  1.29it/s]
1|1685|Loss: 4.896644115447998:  26%|██▌       | 1685/6470 [24:13<1:01:36,  1.29it/s]
1|1686|Loss: 5.0127129554748535:  26%|██▌       | 1685/6470 [24:13<1:01:36,  1.29it/s]
1|1686|Loss: 5.0127129554748535:  26%|██▌       | 1686/6470 [24:13<1:00:31,  1.32it/s]
1|1687|Loss: 4.260533332824707:  26%|██▌       | 1686/6470 [24:14<1:00:31,  1.32it/s] 
1|1687|Loss: 4.260533332824707:  26%|██▌       | 1687/6470 [24:14<1:00:27,  1.32it/s]
1|1688|Loss: 5.167516231536865:  26%|██▌       | 1687/6470 [24:15<1:00:27,  1.32it/s]
1|1688|Loss: 5.167516231536865:  26%|██▌       | 1688/6470 [24:15<1:03:49,  1.25it/s]
1|1689|Loss: 5.151127815246582:  26%|██▌       | 1688/6470 [24:16<1:03:49,  1.25it/s]
1|1689|Loss: 5.151127815246582:  26%|██▌       | 1689/6470 [24:16<1:05:08,  1.22it/s]
1|1690|Loss: 3.57346773147583:  26%|██▌       | 1689/6470 [24:16<1:05:08,  1.22it/s] 
1|1690|Loss: 3.57346773147583:  26%|██▌       | 1690/6470 [24:16<56:41,  1.41it/s]  
1|1691|Loss: 5.228078842163086:  26%|██▌       | 1690/6470 [24:17<56:41,  1.41it/s]
1|1691|Loss: 5.228078842163086:  26%|██▌       | 1691/6470 [24:17<57:47,  1.38it/s]
1|1692|Loss: 5.352578639984131:  26%|██▌       | 1691/6470 [24:18<57:47,  1.38it/s]
1|1692|Loss: 5.352578639984131:  26%|██▌       | 1692/6470 [24:18<1:03:09,  1.26it/s]
1|1693|Loss: 5.142791748046875:  26%|██▌       | 1692/6470 [24:19<1:03:09,  1.26it/s]
1|1693|Loss: 5.142791748046875:  26%|██▌       | 1693/6470 [24:19<1:06:23,  1.20it/s]
1|1694|Loss: 4.77101993560791:  26%|██▌       | 1693/6470 [24:19<1:06:23,  1.20it/s] 
1|1694|Loss: 4.77101993560791:  26%|██▌       | 1694/6470 [24:20<1:05:49,  1.21it/s]
1|1695|Loss: 4.566845893859863:  26%|██▌       | 1694/6470 [24:20<1:05:49,  1.21it/s]
1|1695|Loss: 4.566845893859863:  26%|██▌       | 1695/6470 [24:21<1:04:09,  1.24it/s]
1|1696|Loss: 4.970268726348877:  26%|██▌       | 1695/6470 [24:21<1:04:09,  1.24it/s]
1|1696|Loss: 4.970268726348877:  26%|██▌       | 1696/6470 [24:21<1:04:53,  1.23it/s]
1|1697|Loss: 5.120678424835205:  26%|██▌       | 1696/6470 [24:22<1:04:53,  1.23it/s]
1|1697|Loss: 5.120678424835205:  26%|██▌       | 1697/6470 [24:22<1:04:08,  1.24it/s]
1|1698|Loss: 4.809289455413818:  26%|██▌       | 1697/6470 [24:23<1:04:08,  1.24it/s]
1|1698|Loss: 4.809289455413818:  26%|██▌       | 1698/6470 [24:23<1:05:47,  1.21it/s]
1|1699|Loss: 4.575539588928223:  26%|██▌       | 1698/6470 [24:24<1:05:47,  1.21it/s]
1|1699|Loss: 4.575539588928223:  26%|██▋       | 1699/6470 [24:24<1:05:15,  1.22it/s]
1|1700|Loss: 4.996840000152588:  26%|██▋       | 1699/6470 [24:24<1:05:15,  1.22it/s]
1|1700|Loss: 4.996840000152588:  26%|██▋       | 1700/6470 [24:25<1:06:47,  1.19it/s]
1|1701|Loss: 5.4006218910217285:  26%|██▋       | 1700/6470 [24:25<1:06:47,  1.19it/s]
1|1701|Loss: 5.4006218910217285:  26%|██▋       | 1701/6470 [24:26<1:06:48,  1.19it/s]
1|1702|Loss: 4.382570743560791:  26%|██▋       | 1701/6470 [24:26<1:06:48,  1.19it/s] 
1|1702|Loss: 4.382570743560791:  26%|██▋       | 1702/6470 [24:26<1:04:17,  1.24it/s]
1|1703|Loss: 4.761341571807861:  26%|██▋       | 1702/6470 [24:27<1:04:17,  1.24it/s]
1|1703|Loss: 4.761341571807861:  26%|██▋       | 1703/6470 [24:27<1:01:27,  1.29it/s]
1|1704|Loss: 5.536947727203369:  26%|██▋       | 1703/6470 [24:27<1:01:27,  1.29it/s]
1|1704|Loss: 5.536947727203369:  26%|██▋       | 1704/6470 [24:28<1:01:15,  1.30it/s]
1|1705|Loss: 5.401932716369629:  26%|██▋       | 1704/6470 [24:28<1:01:15,  1.30it/s]
1|1705|Loss: 5.401932716369629:  26%|██▋       | 1705/6470 [24:29<1:02:25,  1.27it/s]
1|1706|Loss: 4.879430294036865:  26%|██▋       | 1705/6470 [24:29<1:02:25,  1.27it/s]
1|1706|Loss: 4.879430294036865:  26%|██▋       | 1706/6470 [24:29<1:01:23,  1.29it/s]
1|1707|Loss: 4.500516414642334:  26%|██▋       | 1706/6470 [24:30<1:01:23,  1.29it/s]
1|1707|Loss: 4.500516414642334:  26%|██▋       | 1707/6470 [24:30<58:02,  1.37it/s]  
1|1708|Loss: 4.987974166870117:  26%|██▋       | 1707/6470 [24:30<58:02,  1.37it/s]
1|1708|Loss: 4.987974166870117:  26%|██▋       | 1708/6470 [24:31<58:35,  1.35it/s]
1|1709|Loss: 4.490630626678467:  26%|██▋       | 1708/6470 [24:31<58:35,  1.35it/s]
1|1709|Loss: 4.490630626678467:  26%|██▋       | 1709/6470 [24:32<59:41,  1.33it/s]
1|1710|Loss: 4.68341064453125:  26%|██▋       | 1709/6470 [24:32<59:41,  1.33it/s] 
1|1710|Loss: 4.68341064453125:  26%|██▋       | 1710/6470 [24:32<1:00:58,  1.30it/s]
1|1711|Loss: 4.403144359588623:  26%|██▋       | 1710/6470 [24:33<1:00:58,  1.30it/s]
1|1711|Loss: 4.403144359588623:  26%|██▋       | 1711/6470 [24:33<55:48,  1.42it/s]  
1|1712|Loss: 4.71170711517334:  26%|██▋       | 1711/6470 [24:33<55:48,  1.42it/s] 
1|1712|Loss: 4.71170711517334:  26%|██▋       | 1712/6470 [24:34<53:59,  1.47it/s]
1|1713|Loss: 5.13882303237915:  26%|██▋       | 1712/6470 [24:34<53:59,  1.47it/s]
1|1713|Loss: 5.13882303237915:  26%|██▋       | 1713/6470 [24:34<58:19,  1.36it/s]
1|1714|Loss: 5.75327205657959:  26%|██▋       | 1713/6470 [24:35<58:19,  1.36it/s]
1|1714|Loss: 5.75327205657959:  26%|██▋       | 1714/6470 [24:35<1:06:01,  1.20it/s]
1|1715|Loss: 5.009687900543213:  26%|██▋       | 1714/6470 [24:36<1:06:01,  1.20it/s]
1|1715|Loss: 5.009687900543213:  27%|██▋       | 1715/6470 [24:36<1:09:42,  1.14it/s]
1|1716|Loss: 4.975920677185059:  27%|██▋       | 1715/6470 [24:37<1:09:42,  1.14it/s]
1|1716|Loss: 4.975920677185059:  27%|██▋       | 1716/6470 [24:37<1:09:00,  1.15it/s]
1|1717|Loss: 5.3358917236328125:  27%|██▋       | 1716/6470 [24:38<1:09:00,  1.15it/s]
1|1717|Loss: 5.3358917236328125:  27%|██▋       | 1717/6470 [24:38<1:10:56,  1.12it/s]
1|1718|Loss: 4.2091498374938965:  27%|██▋       | 1717/6470 [24:39<1:10:56,  1.12it/s]
1|1718|Loss: 4.2091498374938965:  27%|██▋       | 1718/6470 [24:39<1:03:35,  1.25it/s]
1|1719|Loss: 5.308660984039307:  27%|██▋       | 1718/6470 [24:39<1:03:35,  1.25it/s] 
1|1719|Loss: 5.308660984039307:  27%|██▋       | 1719/6470 [24:40<1:02:01,  1.28it/s]
1|1720|Loss: 4.581623554229736:  27%|██▋       | 1719/6470 [24:40<1:02:01,  1.28it/s]
1|1720|Loss: 4.581623554229736:  27%|██▋       | 1720/6470 [24:40<1:00:34,  1.31it/s]
1|1721|Loss: 4.50102424621582:  27%|██▋       | 1720/6470 [24:41<1:00:34,  1.31it/s] 
1|1721|Loss: 4.50102424621582:  27%|██▋       | 1721/6470 [24:41<56:22,  1.40it/s]  
1|1722|Loss: 4.760962009429932:  27%|██▋       | 1721/6470 [24:41<56:22,  1.40it/s]
1|1722|Loss: 4.760962009429932:  27%|██▋       | 1722/6470 [24:42<54:54,  1.44it/s]
1|1723|Loss: 5.027639389038086:  27%|██▋       | 1722/6470 [24:42<54:54,  1.44it/s]
1|1723|Loss: 5.027639389038086:  27%|██▋       | 1723/6470 [24:42<55:15,  1.43it/s]
1|1724|Loss: 4.762590408325195:  27%|██▋       | 1723/6470 [24:43<55:15,  1.43it/s]
1|1724|Loss: 4.762590408325195:  27%|██▋       | 1724/6470 [24:43<1:00:05,  1.32it/s]
1|1725|Loss: 5.092555046081543:  27%|██▋       | 1724/6470 [24:44<1:00:05,  1.32it/s]
1|1725|Loss: 5.092555046081543:  27%|██▋       | 1725/6470 [24:44<1:01:53,  1.28it/s]
1|1726|Loss: 4.6616597175598145:  27%|██▋       | 1725/6470 [24:44<1:01:53,  1.28it/s]
1|1726|Loss: 4.6616597175598145:  27%|██▋       | 1726/6470 [24:45<1:00:37,  1.30it/s]
1|1727|Loss: 4.7212233543396:  27%|██▋       | 1726/6470 [24:45<1:00:37,  1.30it/s]   
1|1727|Loss: 4.7212233543396:  27%|██▋       | 1727/6470 [24:45<59:02,  1.34it/s]  
1|1728|Loss: 4.709179878234863:  27%|██▋       | 1727/6470 [24:46<59:02,  1.34it/s]
1|1728|Loss: 4.709179878234863:  27%|██▋       | 1728/6470 [24:46<57:24,  1.38it/s]
1|1729|Loss: 5.459099769592285:  27%|██▋       | 1728/6470 [24:47<57:24,  1.38it/s]
1|1729|Loss: 5.459099769592285:  27%|██▋       | 1729/6470 [24:47<1:01:28,  1.29it/s]
1|1730|Loss: 4.022057056427002:  27%|██▋       | 1729/6470 [24:47<1:01:28,  1.29it/s]
1|1730|Loss: 4.022057056427002:  27%|██▋       | 1730/6470 [24:48<56:59,  1.39it/s]  
1|1731|Loss: 4.639261245727539:  27%|██▋       | 1730/6470 [24:48<56:59,  1.39it/s]
1|1731|Loss: 4.639261245727539:  27%|██▋       | 1731/6470 [24:48<54:59,  1.44it/s]
1|1732|Loss: 4.889997959136963:  27%|██▋       | 1731/6470 [24:49<54:59,  1.44it/s]
1|1732|Loss: 4.889997959136963:  27%|██▋       | 1732/6470 [24:49<56:59,  1.39it/s]
1|1733|Loss: 4.76204776763916:  27%|██▋       | 1732/6470 [24:49<56:59,  1.39it/s] 
1|1733|Loss: 4.76204776763916:  27%|██▋       | 1733/6470 [24:50<57:59,  1.36it/s]
1|1734|Loss: 4.408921241760254:  27%|██▋       | 1733/6470 [24:50<57:59,  1.36it/s]
1|1734|Loss: 4.408921241760254:  27%|██▋       | 1734/6470 [24:50<55:13,  1.43it/s]
1|1735|Loss: 4.85889196395874:  27%|██▋       | 1734/6470 [24:51<55:13,  1.43it/s] 
1|1735|Loss: 4.85889196395874:  27%|██▋       | 1735/6470 [24:51<55:39,  1.42it/s]
1|1736|Loss: 5.271814823150635:  27%|██▋       | 1735/6470 [24:52<55:39,  1.42it/s]
1|1736|Loss: 5.271814823150635:  27%|██▋       | 1736/6470 [24:52<57:50,  1.36it/s]
1|1737|Loss: 5.255559921264648:  27%|██▋       | 1736/6470 [24:52<57:50,  1.36it/s]
1|1737|Loss: 5.255559921264648:  27%|██▋       | 1737/6470 [24:53<57:41,  1.37it/s]
1|1738|Loss: 5.08394718170166:  27%|██▋       | 1737/6470 [24:53<57:41,  1.37it/s] 
1|1738|Loss: 5.08394718170166:  27%|██▋       | 1738/6470 [24:53<58:20,  1.35it/s]
1|1739|Loss: 5.220420837402344:  27%|██▋       | 1738/6470 [24:54<58:20,  1.35it/s]
1|1739|Loss: 5.220420837402344:  27%|██▋       | 1739/6470 [24:54<1:00:45,  1.30it/s]
1|1740|Loss: 4.91049861907959:  27%|██▋       | 1739/6470 [24:55<1:00:45,  1.30it/s] 
1|1740|Loss: 4.91049861907959:  27%|██▋       | 1740/6470 [24:55<59:58,  1.31it/s]  
1|1741|Loss: 4.939528942108154:  27%|██▋       | 1740/6470 [24:55<59:58,  1.31it/s]
1|1741|Loss: 4.939528942108154:  27%|██▋       | 1741/6470 [24:56<58:45,  1.34it/s]
1|1742|Loss: 4.487040996551514:  27%|██▋       | 1741/6470 [24:56<58:45,  1.34it/s]
1|1742|Loss: 4.487040996551514:  27%|██▋       | 1742/6470 [24:56<55:08,  1.43it/s]
1|1743|Loss: 4.951235294342041:  27%|██▋       | 1742/6470 [24:57<55:08,  1.43it/s]
1|1743|Loss: 4.951235294342041:  27%|██▋       | 1743/6470 [24:57<56:44,  1.39it/s]
1|1744|Loss: 4.920300006866455:  27%|██▋       | 1743/6470 [24:57<56:44,  1.39it/s]
1|1744|Loss: 4.920300006866455:  27%|██▋       | 1744/6470 [24:58<56:55,  1.38it/s]
1|1745|Loss: 4.5548834800720215:  27%|██▋       | 1744/6470 [24:58<56:55,  1.38it/s]
1|1745|Loss: 4.5548834800720215:  27%|██▋       | 1745/6470 [24:58<54:19,  1.45it/s]
1|1746|Loss: 5.20632791519165:  27%|██▋       | 1745/6470 [24:59<54:19,  1.45it/s]  
1|1746|Loss: 5.20632791519165:  27%|██▋       | 1746/6470 [24:59<55:15,  1.42it/s]
1|1747|Loss: 5.326307773590088:  27%|██▋       | 1746/6470 [25:00<55:15,  1.42it/s]
1|1747|Loss: 5.326307773590088:  27%|██▋       | 1747/6470 [25:00<1:00:02,  1.31it/s]
1|1748|Loss: 4.934054374694824:  27%|██▋       | 1747/6470 [25:00<1:00:02,  1.31it/s]
1|1748|Loss: 4.934054374694824:  27%|██▋       | 1748/6470 [25:01<1:01:05,  1.29it/s]
1|1749|Loss: 4.546851634979248:  27%|██▋       | 1748/6470 [25:01<1:01:05,  1.29it/s]
1|1749|Loss: 4.546851634979248:  27%|██▋       | 1749/6470 [25:02<59:11,  1.33it/s]  
1|1750|Loss: 4.905669212341309:  27%|██▋       | 1749/6470 [25:02<59:11,  1.33it/s]
1|1750|Loss: 4.905669212341309:  27%|██▋       | 1750/6470 [25:02<1:01:40,  1.28it/s]
1|1751|Loss: 4.891421318054199:  27%|██▋       | 1750/6470 [25:03<1:01:40,  1.28it/s]
1|1751|Loss: 4.891421318054199:  27%|██▋       | 1751/6470 [25:03<1:00:13,  1.31it/s]
1|1752|Loss: 4.685229778289795:  27%|██▋       | 1751/6470 [25:03<1:00:13,  1.31it/s]
1|1752|Loss: 4.685229778289795:  27%|██▋       | 1752/6470 [25:04<58:20,  1.35it/s]  
1|1753|Loss: 4.382863521575928:  27%|██▋       | 1752/6470 [25:04<58:20,  1.35it/s]
1|1753|Loss: 4.382863521575928:  27%|██▋       | 1753/6470 [25:04<54:39,  1.44it/s]
1|1754|Loss: 4.268833160400391:  27%|██▋       | 1753/6470 [25:05<54:39,  1.44it/s]
1|1754|Loss: 4.268833160400391:  27%|██▋       | 1754/6470 [25:05<51:29,  1.53it/s]
1|1755|Loss: 4.362635612487793:  27%|██▋       | 1754/6470 [25:05<51:29,  1.53it/s]
1|1755|Loss: 4.362635612487793:  27%|██▋       | 1755/6470 [25:05<47:40,  1.65it/s]
1|1756|Loss: 4.835260391235352:  27%|██▋       | 1755/6470 [25:06<47:40,  1.65it/s]
1|1756|Loss: 4.835260391235352:  27%|██▋       | 1756/6470 [25:06<50:13,  1.56it/s]
1|1757|Loss: 5.193789958953857:  27%|██▋       | 1756/6470 [25:07<50:13,  1.56it/s]
1|1757|Loss: 5.193789958953857:  27%|██▋       | 1757/6470 [25:07<56:17,  1.40it/s]
1|1758|Loss: 4.892556667327881:  27%|██▋       | 1757/6470 [25:08<56:17,  1.40it/s]
1|1758|Loss: 4.892556667327881:  27%|██▋       | 1758/6470 [25:08<57:15,  1.37it/s]
1|1759|Loss: 4.951153755187988:  27%|██▋       | 1758/6470 [25:08<57:15,  1.37it/s]
1|1759|Loss: 4.951153755187988:  27%|██▋       | 1759/6470 [25:09<57:51,  1.36it/s]
1|1760|Loss: 4.796426296234131:  27%|██▋       | 1759/6470 [25:09<57:51,  1.36it/s]
1|1760|Loss: 4.796426296234131:  27%|██▋       | 1760/6470 [25:09<1:01:31,  1.28it/s]
1|1761|Loss: 5.2405242919921875:  27%|██▋       | 1760/6470 [25:10<1:01:31,  1.28it/s]
1|1761|Loss: 5.2405242919921875:  27%|██▋       | 1761/6470 [25:10<1:04:49,  1.21it/s]
1|1762|Loss: 4.567843914031982:  27%|██▋       | 1761/6470 [25:11<1:04:49,  1.21it/s] 
1|1762|Loss: 4.567843914031982:  27%|██▋       | 1762/6470 [25:11<1:02:49,  1.25it/s]
1|1763|Loss: 4.632692337036133:  27%|██▋       | 1762/6470 [25:12<1:02:49,  1.25it/s]
1|1763|Loss: 4.632692337036133:  27%|██▋       | 1763/6470 [25:12<1:02:17,  1.26it/s]
1|1764|Loss: 5.274379730224609:  27%|██▋       | 1763/6470 [25:12<1:02:17,  1.26it/s]
1|1764|Loss: 5.274379730224609:  27%|██▋       | 1764/6470 [25:13<1:03:44,  1.23it/s]
1|1765|Loss: 5.0223069190979:  27%|██▋       | 1764/6470 [25:13<1:03:44,  1.23it/s]  
1|1765|Loss: 5.0223069190979:  27%|██▋       | 1765/6470 [25:14<1:03:55,  1.23it/s]
1|1766|Loss: 4.99437141418457:  27%|██▋       | 1765/6470 [25:14<1:03:55,  1.23it/s]
1|1766|Loss: 4.99437141418457:  27%|██▋       | 1766/6470 [25:14<1:01:23,  1.28it/s]
1|1767|Loss: 4.961633682250977:  27%|██▋       | 1766/6470 [25:15<1:01:23,  1.28it/s]
1|1767|Loss: 4.961633682250977:  27%|██▋       | 1767/6470 [25:15<1:03:48,  1.23it/s]
1|1768|Loss: 5.0078630447387695:  27%|██▋       | 1767/6470 [25:16<1:03:48,  1.23it/s]
1|1768|Loss: 5.0078630447387695:  27%|██▋       | 1768/6470 [25:16<1:06:08,  1.18it/s]
1|1769|Loss: 4.695223808288574:  27%|██▋       | 1768/6470 [25:17<1:06:08,  1.18it/s] 
1|1769|Loss: 4.695223808288574:  27%|██▋       | 1769/6470 [25:17<1:05:04,  1.20it/s]
1|1770|Loss: 4.272402286529541:  27%|██▋       | 1769/6470 [25:17<1:05:04,  1.20it/s]
1|1770|Loss: 4.272402286529541:  27%|██▋       | 1770/6470 [25:17<59:27,  1.32it/s]  
1|1771|Loss: 4.867706775665283:  27%|██▋       | 1770/6470 [25:18<59:27,  1.32it/s]
1|1771|Loss: 4.867706775665283:  27%|██▋       | 1771/6470 [25:18<58:46,  1.33it/s]
1|1772|Loss: 4.954766750335693:  27%|██▋       | 1771/6470 [25:19<58:46,  1.33it/s]
1|1772|Loss: 4.954766750335693:  27%|██▋       | 1772/6470 [25:19<59:10,  1.32it/s]
1|1773|Loss: 4.482272148132324:  27%|██▋       | 1772/6470 [25:19<59:10,  1.32it/s]
1|1773|Loss: 4.482272148132324:  27%|██▋       | 1773/6470 [25:20<55:19,  1.41it/s]
1|1774|Loss: 3.9607529640197754:  27%|██▋       | 1773/6470 [25:20<55:19,  1.41it/s]
1|1774|Loss: 3.9607529640197754:  27%|██▋       | 1774/6470 [25:20<50:20,  1.55it/s]
1|1775|Loss: 5.04662561416626:  27%|██▋       | 1774/6470 [25:20<50:20,  1.55it/s]  
1|1775|Loss: 5.04662561416626:  27%|██▋       | 1775/6470 [25:21<52:16,  1.50it/s]
1|1776|Loss: 4.571303367614746:  27%|██▋       | 1775/6470 [25:21<52:16,  1.50it/s]
1|1776|Loss: 4.571303367614746:  27%|██▋       | 1776/6470 [25:22<53:34,  1.46it/s]
1|1777|Loss: 3.989222526550293:  27%|██▋       | 1776/6470 [25:22<53:34,  1.46it/s]
1|1777|Loss: 3.989222526550293:  27%|██▋       | 1777/6470 [25:22<49:44,  1.57it/s]
1|1778|Loss: 4.7809343338012695:  27%|██▋       | 1777/6470 [25:22<49:44,  1.57it/s]
1|1778|Loss: 4.7809343338012695:  27%|██▋       | 1778/6470 [25:23<50:07,  1.56it/s]
1|1779|Loss: 2.9533169269561768:  27%|██▋       | 1778/6470 [25:23<50:07,  1.56it/s]
1|1779|Loss: 2.9533169269561768:  27%|██▋       | 1779/6470 [25:23<47:57,  1.63it/s]
1|1780|Loss: 5.388723850250244:  27%|██▋       | 1779/6470 [25:24<47:57,  1.63it/s] 
1|1780|Loss: 5.388723850250244:  28%|██▊       | 1780/6470 [25:24<53:51,  1.45it/s]
1|1781|Loss: 5.50615119934082:  28%|██▊       | 1780/6470 [25:25<53:51,  1.45it/s] 
1|1781|Loss: 5.50615119934082:  28%|██▊       | 1781/6470 [25:25<59:50,  1.31it/s]
1|1782|Loss: 5.1048903465271:  28%|██▊       | 1781/6470 [25:26<59:50,  1.31it/s] 
1|1782|Loss: 5.1048903465271:  28%|██▊       | 1782/6470 [25:26<1:00:40,  1.29it/s]
1|1783|Loss: 4.552786350250244:  28%|██▊       | 1782/6470 [25:26<1:00:40,  1.29it/s]
1|1783|Loss: 4.552786350250244:  28%|██▊       | 1783/6470 [25:27<57:41,  1.35it/s]  
1|1784|Loss: 4.9592719078063965:  28%|██▊       | 1783/6470 [25:27<57:41,  1.35it/s]
1|1784|Loss: 4.9592719078063965:  28%|██▊       | 1784/6470 [25:27<59:51,  1.30it/s]
1|1785|Loss: 5.333460807800293:  28%|██▊       | 1784/6470 [25:28<59:51,  1.30it/s] 
1|1785|Loss: 5.333460807800293:  28%|██▊       | 1785/6470 [25:28<1:00:49,  1.28it/s]
1|1786|Loss: 4.47512149810791:  28%|██▊       | 1785/6470 [25:29<1:00:49,  1.28it/s] 
1|1786|Loss: 4.47512149810791:  28%|██▊       | 1786/6470 [25:29<59:08,  1.32it/s]  
1|1787|Loss: 4.777003765106201:  28%|██▊       | 1786/6470 [25:29<59:08,  1.32it/s]
1|1787|Loss: 4.777003765106201:  28%|██▊       | 1787/6470 [25:30<59:18,  1.32it/s]
1|1788|Loss: 4.706060886383057:  28%|██▊       | 1787/6470 [25:30<59:18,  1.32it/s]
1|1788|Loss: 4.706060886383057:  28%|██▊       | 1788/6470 [25:30<1:01:27,  1.27it/s]
1|1789|Loss: 5.281774997711182:  28%|██▊       | 1788/6470 [25:31<1:01:27,  1.27it/s]
1|1789|Loss: 5.281774997711182:  28%|██▊       | 1789/6470 [25:32<1:07:31,  1.16it/s]
1|1790|Loss: 3.841752290725708:  28%|██▊       | 1789/6470 [25:32<1:07:31,  1.16it/s]
1|1790|Loss: 3.841752290725708:  28%|██▊       | 1790/6470 [25:32<59:55,  1.30it/s]  
1|1791|Loss: 4.810812950134277:  28%|██▊       | 1790/6470 [25:32<59:55,  1.30it/s]
1|1791|Loss: 4.810812950134277:  28%|██▊       | 1791/6470 [25:33<56:45,  1.37it/s]
1|1792|Loss: 4.432427406311035:  28%|██▊       | 1791/6470 [25:33<56:45,  1.37it/s]
1|1792|Loss: 4.432427406311035:  28%|██▊       | 1792/6470 [25:33<54:11,  1.44it/s]
1|1793|Loss: 4.7678303718566895:  28%|██▊       | 1792/6470 [25:34<54:11,  1.44it/s]
1|1793|Loss: 4.7678303718566895:  28%|██▊       | 1793/6470 [25:34<53:17,  1.46it/s]
1|1794|Loss: 5.33889627456665:  28%|██▊       | 1793/6470 [25:34<53:17,  1.46it/s]  
1|1794|Loss: 5.33889627456665:  28%|██▊       | 1794/6470 [25:35<57:15,  1.36it/s]
1|1795|Loss: 5.044769763946533:  28%|██▊       | 1794/6470 [25:35<57:15,  1.36it/s]
1|1795|Loss: 5.044769763946533:  28%|██▊       | 1795/6470 [25:36<1:02:28,  1.25it/s]
1|1796|Loss: 5.30889368057251:  28%|██▊       | 1795/6470 [25:36<1:02:28,  1.25it/s] 
1|1796|Loss: 5.30889368057251:  28%|██▊       | 1796/6470 [25:37<1:03:50,  1.22it/s]
1|1797|Loss: 5.226224422454834:  28%|██▊       | 1796/6470 [25:37<1:03:50,  1.22it/s]
1|1797|Loss: 5.226224422454834:  28%|██▊       | 1797/6470 [25:37<1:00:10,  1.29it/s]
1|1798|Loss: 4.76179838180542:  28%|██▊       | 1797/6470 [25:38<1:00:10,  1.29it/s] 
1|1798|Loss: 4.76179838180542:  28%|██▊       | 1798/6470 [25:38<58:03,  1.34it/s]  
1|1799|Loss: 5.268750190734863:  28%|██▊       | 1798/6470 [25:38<58:03,  1.34it/s]
1|1799|Loss: 5.268750190734863:  28%|██▊       | 1799/6470 [25:39<1:01:39,  1.26it/s]
1|1800|Loss: 4.681768894195557:  28%|██▊       | 1799/6470 [25:39<1:01:39,  1.26it/s]
1|1800|Loss: 4.681768894195557:  28%|██▊       | 1800/6470 [25:40<1:00:19,  1.29it/s]
1|1801|Loss: 4.02803373336792:  28%|██▊       | 1800/6470 [25:40<1:00:19,  1.29it/s] 
1|1801|Loss: 4.02803373336792:  28%|██▊       | 1801/6470 [25:40<55:20,  1.41it/s]  
1|1802|Loss: 5.162520885467529:  28%|██▊       | 1801/6470 [25:41<55:20,  1.41it/s]
1|1802|Loss: 5.162520885467529:  28%|██▊       | 1802/6470 [25:41<53:46,  1.45it/s]
1|1803|Loss: 5.147670745849609:  28%|██▊       | 1802/6470 [25:41<53:46,  1.45it/s]
1|1803|Loss: 5.147670745849609:  28%|██▊       | 1803/6470 [25:42<58:50,  1.32it/s]
1|1804|Loss: 4.788311958312988:  28%|██▊       | 1803/6470 [25:42<58:50,  1.32it/s]
1|1804|Loss: 4.788311958312988:  28%|██▊       | 1804/6470 [25:42<58:18,  1.33it/s]
1|1805|Loss: 4.962204933166504:  28%|██▊       | 1804/6470 [25:43<58:18,  1.33it/s]
1|1805|Loss: 4.962204933166504:  28%|██▊       | 1805/6470 [25:43<58:20,  1.33it/s]
1|1806|Loss: 4.715915203094482:  28%|██▊       | 1805/6470 [25:44<58:20,  1.33it/s]
1|1806|Loss: 4.715915203094482:  28%|██▊       | 1806/6470 [25:44<59:10,  1.31it/s]
1|1807|Loss: 3.963188409805298:  28%|██▊       | 1806/6470 [25:44<59:10,  1.31it/s]
1|1807|Loss: 3.963188409805298:  28%|██▊       | 1807/6470 [25:45<55:45,  1.39it/s]
1|1808|Loss: 4.471463680267334:  28%|██▊       | 1807/6470 [25:45<55:45,  1.39it/s]
1|1808|Loss: 4.471463680267334:  28%|██▊       | 1808/6470 [25:45<53:41,  1.45it/s]
1|1809|Loss: 5.098952293395996:  28%|██▊       | 1808/6470 [25:46<53:41,  1.45it/s]
1|1809|Loss: 5.098952293395996:  28%|██▊       | 1809/6470 [25:46<57:52,  1.34it/s]
1|1810|Loss: 4.9535322189331055:  28%|██▊       | 1809/6470 [25:47<57:52,  1.34it/s]
1|1810|Loss: 4.9535322189331055:  28%|██▊       | 1810/6470 [25:47<56:29,  1.37it/s]
1|1811|Loss: 4.239253997802734:  28%|██▊       | 1810/6470 [25:47<56:29,  1.37it/s] 
1|1811|Loss: 4.239253997802734:  28%|██▊       | 1811/6470 [25:47<52:57,  1.47it/s]
1|1812|Loss: 5.012612342834473:  28%|██▊       | 1811/6470 [25:48<52:57,  1.47it/s]
1|1812|Loss: 5.012612342834473:  28%|██▊       | 1812/6470 [25:48<52:57,  1.47it/s]
1|1813|Loss: 4.56345796585083:  28%|██▊       | 1812/6470 [25:48<52:57,  1.47it/s] 
1|1813|Loss: 4.56345796585083:  28%|██▊       | 1813/6470 [25:49<53:26,  1.45it/s]
1|1814|Loss: 4.572864055633545:  28%|██▊       | 1813/6470 [25:49<53:26,  1.45it/s]
1|1814|Loss: 4.572864055633545:  28%|██▊       | 1814/6470 [25:49<52:16,  1.48it/s]
1|1815|Loss: 4.742664813995361:  28%|██▊       | 1814/6470 [25:50<52:16,  1.48it/s]
1|1815|Loss: 4.742664813995361:  28%|██▊       | 1815/6470 [25:50<51:49,  1.50it/s]
1|1816|Loss: 5.117136478424072:  28%|██▊       | 1815/6470 [25:50<51:49,  1.50it/s]
1|1816|Loss: 5.117136478424072:  28%|██▊       | 1816/6470 [25:51<53:43,  1.44it/s]
1|1817|Loss: 3.420591354370117:  28%|██▊       | 1816/6470 [25:51<53:43,  1.44it/s]
1|1817|Loss: 3.420591354370117:  28%|██▊       | 1817/6470 [25:51<51:41,  1.50it/s]
1|1818|Loss: 4.703716278076172:  28%|██▊       | 1817/6470 [25:52<51:41,  1.50it/s]
1|1818|Loss: 4.703716278076172:  28%|██▊       | 1818/6470 [25:52<49:12,  1.58it/s]
1|1819|Loss: 4.6584343910217285:  28%|██▊       | 1818/6470 [25:52<49:12,  1.58it/s]
1|1819|Loss: 4.6584343910217285:  28%|██▊       | 1819/6470 [25:53<46:39,  1.66it/s]
1|1820|Loss: 4.93129301071167:  28%|██▊       | 1819/6470 [25:53<46:39,  1.66it/s]  
1|1820|Loss: 4.93129301071167:  28%|██▊       | 1820/6470 [25:53<49:51,  1.55it/s]
1|1821|Loss: 5.249684810638428:  28%|██▊       | 1820/6470 [25:54<49:51,  1.55it/s]
1|1821|Loss: 5.249684810638428:  28%|██▊       | 1821/6470 [25:54<1:03:06,  1.23it/s]
1|1822|Loss: 4.691854476928711:  28%|██▊       | 1821/6470 [25:55<1:03:06,  1.23it/s]
1|1822|Loss: 4.691854476928711:  28%|██▊       | 1822/6470 [25:55<1:01:30,  1.26it/s]
1|1823|Loss: 4.676339626312256:  28%|██▊       | 1822/6470 [25:56<1:01:30,  1.26it/s]
1|1823|Loss: 4.676339626312256:  28%|██▊       | 1823/6470 [25:56<59:52,  1.29it/s]  
1|1824|Loss: 4.373099327087402:  28%|██▊       | 1823/6470 [25:56<59:52,  1.29it/s]
1|1824|Loss: 4.373099327087402:  28%|██▊       | 1824/6470 [25:57<57:49,  1.34it/s]
1|1825|Loss: 4.904026985168457:  28%|██▊       | 1824/6470 [25:57<57:49,  1.34it/s]
1|1825|Loss: 4.904026985168457:  28%|██▊       | 1825/6470 [25:57<56:40,  1.37it/s]
1|1826|Loss: 4.516726016998291:  28%|██▊       | 1825/6470 [25:58<56:40,  1.37it/s]
1|1826|Loss: 4.516726016998291:  28%|██▊       | 1826/6470 [25:58<53:31,  1.45it/s]
1|1827|Loss: 5.082115650177002:  28%|██▊       | 1826/6470 [25:58<53:31,  1.45it/s]
1|1827|Loss: 5.082115650177002:  28%|██▊       | 1827/6470 [25:59<51:07,  1.51it/s]
1|1828|Loss: 4.866512298583984:  28%|██▊       | 1827/6470 [25:59<51:07,  1.51it/s]
1|1828|Loss: 4.866512298583984:  28%|██▊       | 1828/6470 [25:59<52:52,  1.46it/s]
1|1829|Loss: 5.262931823730469:  28%|██▊       | 1828/6470 [26:00<52:52,  1.46it/s]
1|1829|Loss: 5.262931823730469:  28%|██▊       | 1829/6470 [26:00<54:49,  1.41it/s]
1|1830|Loss: 5.110801696777344:  28%|██▊       | 1829/6470 [26:00<54:49,  1.41it/s]
1|1830|Loss: 5.110801696777344:  28%|██▊       | 1830/6470 [26:01<57:03,  1.36it/s]
1|1831|Loss: 5.008248329162598:  28%|██▊       | 1830/6470 [26:01<57:03,  1.36it/s]
1|1831|Loss: 5.008248329162598:  28%|██▊       | 1831/6470 [26:02<57:57,  1.33it/s]
1|1832|Loss: 5.030551910400391:  28%|██▊       | 1831/6470 [26:02<57:57,  1.33it/s]
1|1832|Loss: 5.030551910400391:  28%|██▊       | 1832/6470 [26:02<58:39,  1.32it/s]
1|1833|Loss: 5.137796878814697:  28%|██▊       | 1832/6470 [26:03<58:39,  1.32it/s]
1|1833|Loss: 5.137796878814697:  28%|██▊       | 1833/6470 [26:03<57:12,  1.35it/s]
1|1834|Loss: 4.897740840911865:  28%|██▊       | 1833/6470 [26:04<57:12,  1.35it/s]
1|1834|Loss: 4.897740840911865:  28%|██▊       | 1834/6470 [26:04<1:00:02,  1.29it/s]
1|1835|Loss: 4.994190216064453:  28%|██▊       | 1834/6470 [26:04<1:00:02,  1.29it/s]
1|1835|Loss: 4.994190216064453:  28%|██▊       | 1835/6470 [26:05<57:09,  1.35it/s]  
1|1836|Loss: 4.834111213684082:  28%|██▊       | 1835/6470 [26:05<57:09,  1.35it/s]
1|1836|Loss: 4.834111213684082:  28%|██▊       | 1836/6470 [26:05<57:28,  1.34it/s]
1|1837|Loss: 4.455090522766113:  28%|██▊       | 1836/6470 [26:06<57:28,  1.34it/s]
1|1837|Loss: 4.455090522766113:  28%|██▊       | 1837/6470 [26:06<1:00:20,  1.28it/s]
1|1838|Loss: 4.975029945373535:  28%|██▊       | 1837/6470 [26:07<1:00:20,  1.28it/s]
1|1838|Loss: 4.975029945373535:  28%|██▊       | 1838/6470 [26:07<58:52,  1.31it/s]  
1|1839|Loss: 5.005648612976074:  28%|██▊       | 1838/6470 [26:07<58:52,  1.31it/s]
1|1839|Loss: 5.005648612976074:  28%|██▊       | 1839/6470 [26:08<1:00:49,  1.27it/s]
1|1840|Loss: 4.693655490875244:  28%|██▊       | 1839/6470 [26:08<1:00:49,  1.27it/s]
1|1840|Loss: 4.693655490875244:  28%|██▊       | 1840/6470 [26:09<1:02:47,  1.23it/s]
1|1841|Loss: 4.7805585861206055:  28%|██▊       | 1840/6470 [26:09<1:02:47,  1.23it/s]
1|1841|Loss: 4.7805585861206055:  28%|██▊       | 1841/6470 [26:10<1:06:18,  1.16it/s]
1|1842|Loss: 5.1333746910095215:  28%|██▊       | 1841/6470 [26:10<1:06:18,  1.16it/s]
1|1842|Loss: 5.1333746910095215:  28%|██▊       | 1842/6470 [26:11<1:08:18,  1.13it/s]
1|1843|Loss: 5.1209611892700195:  28%|██▊       | 1842/6470 [26:11<1:08:18,  1.13it/s]
1|1843|Loss: 5.1209611892700195:  28%|██▊       | 1843/6470 [26:11<1:07:40,  1.14it/s]
1|1844|Loss: 4.751475811004639:  28%|██▊       | 1843/6470 [26:12<1:07:40,  1.14it/s] 
1|1844|Loss: 4.751475811004639:  29%|██▊       | 1844/6470 [26:12<1:05:27,  1.18it/s]
1|1845|Loss: 4.866180896759033:  29%|██▊       | 1844/6470 [26:13<1:05:27,  1.18it/s]
1|1845|Loss: 4.866180896759033:  29%|██▊       | 1845/6470 [26:13<1:05:59,  1.17it/s]
1|1846|Loss: 4.730040073394775:  29%|██▊       | 1845/6470 [26:13<1:05:59,  1.17it/s]
1|1846|Loss: 4.730040073394775:  29%|██▊       | 1846/6470 [26:14<1:02:29,  1.23it/s]
1|1847|Loss: 4.746648788452148:  29%|██▊       | 1846/6470 [26:14<1:02:29,  1.23it/s]
1|1847|Loss: 4.746648788452148:  29%|██▊       | 1847/6470 [26:14<59:48,  1.29it/s]  
1|1848|Loss: 5.067188739776611:  29%|██▊       | 1847/6470 [26:15<59:48,  1.29it/s]
1|1848|Loss: 5.067188739776611:  29%|██▊       | 1848/6470 [26:15<1:00:01,  1.28it/s]
1|1849|Loss: 4.750893592834473:  29%|██▊       | 1848/6470 [26:16<1:00:01,  1.28it/s]
1|1849|Loss: 4.750893592834473:  29%|██▊       | 1849/6470 [26:16<1:00:47,  1.27it/s]
1|1850|Loss: 4.960609436035156:  29%|██▊       | 1849/6470 [26:16<1:00:47,  1.27it/s]
1|1850|Loss: 4.960609436035156:  29%|██▊       | 1850/6470 [26:17<58:17,  1.32it/s]  
1|1851|Loss: 3.916001796722412:  29%|██▊       | 1850/6470 [26:17<58:17,  1.32it/s]
1|1851|Loss: 3.916001796722412:  29%|██▊       | 1851/6470 [26:17<52:31,  1.47it/s]
1|1852|Loss: 4.886096477508545:  29%|██▊       | 1851/6470 [26:18<52:31,  1.47it/s]
1|1852|Loss: 4.886096477508545:  29%|██▊       | 1852/6470 [26:18<53:58,  1.43it/s]
1|1853|Loss: 4.9017462730407715:  29%|██▊       | 1852/6470 [26:18<53:58,  1.43it/s]
1|1853|Loss: 4.9017462730407715:  29%|██▊       | 1853/6470 [26:19<53:18,  1.44it/s]
1|1854|Loss: 4.772246360778809:  29%|██▊       | 1853/6470 [26:19<53:18,  1.44it/s] 
1|1854|Loss: 4.772246360778809:  29%|██▊       | 1854/6470 [26:19<53:15,  1.44it/s]
1|1855|Loss: 4.750368118286133:  29%|██▊       | 1854/6470 [26:20<53:15,  1.44it/s]
1|1855|Loss: 4.750368118286133:  29%|██▊       | 1855/6470 [26:20<52:05,  1.48it/s]
1|1856|Loss: 4.633184909820557:  29%|██▊       | 1855/6470 [26:20<52:05,  1.48it/s]
1|1856|Loss: 4.633184909820557:  29%|██▊       | 1856/6470 [26:21<53:25,  1.44it/s]
1|1857|Loss: 4.762345314025879:  29%|██▊       | 1856/6470 [26:21<53:25,  1.44it/s]
1|1857|Loss: 4.762345314025879:  29%|██▊       | 1857/6470 [26:21<51:24,  1.50it/s]
1|1858|Loss: 4.720088958740234:  29%|██▊       | 1857/6470 [26:22<51:24,  1.50it/s]
1|1858|Loss: 4.720088958740234:  29%|██▊       | 1858/6470 [26:22<53:40,  1.43it/s]
1|1859|Loss: 4.958878517150879:  29%|██▊       | 1858/6470 [26:23<53:40,  1.43it/s]
1|1859|Loss: 4.958878517150879:  29%|██▊       | 1859/6470 [26:23<54:22,  1.41it/s]
1|1860|Loss: 4.070259094238281:  29%|██▊       | 1859/6470 [26:23<54:22,  1.41it/s]
1|1860|Loss: 4.070259094238281:  29%|██▊       | 1860/6470 [26:23<50:47,  1.51it/s]
1|1861|Loss: 4.948444843292236:  29%|██▊       | 1860/6470 [26:24<50:47,  1.51it/s]
1|1861|Loss: 4.948444843292236:  29%|██▉       | 1861/6470 [26:24<53:11,  1.44it/s]
1|1862|Loss: 5.062022686004639:  29%|██▉       | 1861/6470 [26:25<53:11,  1.44it/s]
1|1862|Loss: 5.062022686004639:  29%|██▉       | 1862/6470 [26:25<57:52,  1.33it/s]
1|1863|Loss: 5.175358295440674:  29%|██▉       | 1862/6470 [26:26<57:52,  1.33it/s]
1|1863|Loss: 5.175358295440674:  29%|██▉       | 1863/6470 [26:26<58:41,  1.31it/s]
1|1864|Loss: 4.734996795654297:  29%|██▉       | 1863/6470 [26:26<58:41,  1.31it/s]
1|1864|Loss: 4.734996795654297:  29%|██▉       | 1864/6470 [26:27<57:29,  1.34it/s]
1|1865|Loss: 4.874465465545654:  29%|██▉       | 1864/6470 [26:27<57:29,  1.34it/s]
1|1865|Loss: 4.874465465545654:  29%|██▉       | 1865/6470 [26:27<57:34,  1.33it/s]
1|1866|Loss: 4.734516620635986:  29%|██▉       | 1865/6470 [26:28<57:34,  1.33it/s]
1|1866|Loss: 4.734516620635986:  29%|██▉       | 1866/6470 [26:28<55:12,  1.39it/s]
1|1867|Loss: 4.640482425689697:  29%|██▉       | 1866/6470 [26:28<55:12,  1.39it/s]
1|1867|Loss: 4.640482425689697:  29%|██▉       | 1867/6470 [26:29<55:34,  1.38it/s]
1|1868|Loss: 5.224170207977295:  29%|██▉       | 1867/6470 [26:29<55:34,  1.38it/s]
1|1868|Loss: 5.224170207977295:  29%|██▉       | 1868/6470 [26:30<59:39,  1.29it/s]
1|1869|Loss: 5.117776870727539:  29%|██▉       | 1868/6470 [26:30<59:39,  1.29it/s]
1|1869|Loss: 5.117776870727539:  29%|██▉       | 1869/6470 [26:31<1:02:30,  1.23it/s]
1|1870|Loss: 4.91146183013916:  29%|██▉       | 1869/6470 [26:31<1:02:30,  1.23it/s] 
1|1870|Loss: 4.91146183013916:  29%|██▉       | 1870/6470 [26:31<1:02:54,  1.22it/s]
1|1871|Loss: 4.887763500213623:  29%|██▉       | 1870/6470 [26:32<1:02:54,  1.22it/s]
1|1871|Loss: 4.887763500213623:  29%|██▉       | 1871/6470 [26:32<1:02:00,  1.24it/s]
1|1872|Loss: 5.177425861358643:  29%|██▉       | 1871/6470 [26:33<1:02:00,  1.24it/s]
1|1872|Loss: 5.177425861358643:  29%|██▉       | 1872/6470 [26:33<1:00:55,  1.26it/s]
1|1873|Loss: 4.744492053985596:  29%|██▉       | 1872/6470 [26:33<1:00:55,  1.26it/s]
1|1873|Loss: 4.744492053985596:  29%|██▉       | 1873/6470 [26:34<1:00:27,  1.27it/s]
1|1874|Loss: 4.402763843536377:  29%|██▉       | 1873/6470 [26:34<1:00:27,  1.27it/s]
1|1874|Loss: 4.402763843536377:  29%|██▉       | 1874/6470 [26:35<1:02:12,  1.23it/s]
1|1875|Loss: 4.877931594848633:  29%|██▉       | 1874/6470 [26:35<1:02:12,  1.23it/s]
1|1875|Loss: 4.877931594848633:  29%|██▉       | 1875/6470 [26:35<59:32,  1.29it/s]  
1|1876|Loss: 4.979436874389648:  29%|██▉       | 1875/6470 [26:36<59:32,  1.29it/s]
1|1876|Loss: 4.979436874389648:  29%|██▉       | 1876/6470 [26:36<57:34,  1.33it/s]
1|1877|Loss: 4.858129978179932:  29%|██▉       | 1876/6470 [26:36<57:34,  1.33it/s]
1|1877|Loss: 4.858129978179932:  29%|██▉       | 1877/6470 [26:37<1:00:14,  1.27it/s]
1|1878|Loss: 5.126821994781494:  29%|██▉       | 1877/6470 [26:37<1:00:14,  1.27it/s]
1|1878|Loss: 5.126821994781494:  29%|██▉       | 1878/6470 [26:38<1:03:35,  1.20it/s]
1|1879|Loss: 5.230401992797852:  29%|██▉       | 1878/6470 [26:38<1:03:35,  1.20it/s]
1|1879|Loss: 5.230401992797852:  29%|██▉       | 1879/6470 [26:39<1:03:46,  1.20it/s]
1|1880|Loss: 5.1308183670043945:  29%|██▉       | 1879/6470 [26:39<1:03:46,  1.20it/s]
1|1880|Loss: 5.1308183670043945:  29%|██▉       | 1880/6470 [26:39<1:00:19,  1.27it/s]
1|1881|Loss: 4.680689334869385:  29%|██▉       | 1880/6470 [26:40<1:00:19,  1.27it/s] 
1|1881|Loss: 4.680689334869385:  29%|██▉       | 1881/6470 [26:40<59:33,  1.28it/s]  
1|1882|Loss: 5.129324436187744:  29%|██▉       | 1881/6470 [26:40<59:33,  1.28it/s]
1|1882|Loss: 5.129324436187744:  29%|██▉       | 1882/6470 [26:41<59:57,  1.28it/s]
1|1883|Loss: 4.794610977172852:  29%|██▉       | 1882/6470 [26:41<59:57,  1.28it/s]
1|1883|Loss: 4.794610977172852:  29%|██▉       | 1883/6470 [26:42<59:38,  1.28it/s]
1|1884|Loss: 4.900151252746582:  29%|██▉       | 1883/6470 [26:42<59:38,  1.28it/s]
1|1884|Loss: 4.900151252746582:  29%|██▉       | 1884/6470 [26:42<1:02:19,  1.23it/s]
1|1885|Loss: 4.663589000701904:  29%|██▉       | 1884/6470 [26:43<1:02:19,  1.23it/s]
1|1885|Loss: 4.663589000701904:  29%|██▉       | 1885/6470 [26:43<1:01:12,  1.25it/s]
1|1886|Loss: 4.966624736785889:  29%|██▉       | 1885/6470 [26:44<1:01:12,  1.25it/s]
1|1886|Loss: 4.966624736785889:  29%|██▉       | 1886/6470 [26:44<1:01:01,  1.25it/s]
1|1887|Loss: 4.936028480529785:  29%|██▉       | 1886/6470 [26:44<1:01:01,  1.25it/s]
1|1887|Loss: 4.936028480529785:  29%|██▉       | 1887/6470 [26:45<1:01:32,  1.24it/s]
1|1888|Loss: 4.5253682136535645:  29%|██▉       | 1887/6470 [26:45<1:01:32,  1.24it/s]
1|1888|Loss: 4.5253682136535645:  29%|██▉       | 1888/6470 [26:46<1:03:53,  1.20it/s]
1|1889|Loss: 4.7892889976501465:  29%|██▉       | 1888/6470 [26:46<1:03:53,  1.20it/s]
1|1889|Loss: 4.7892889976501465:  29%|██▉       | 1889/6470 [26:46<1:01:34,  1.24it/s]
1|1890|Loss: 5.02838659286499:  29%|██▉       | 1889/6470 [26:47<1:01:34,  1.24it/s]  
1|1890|Loss: 5.02838659286499:  29%|██▉       | 1890/6470 [26:47<1:03:27,  1.20it/s]
1|1891|Loss: 5.114438056945801:  29%|██▉       | 1890/6470 [26:48<1:03:27,  1.20it/s]
1|1891|Loss: 5.114438056945801:  29%|██▉       | 1891/6470 [26:48<1:03:46,  1.20it/s]
1|1892|Loss: 4.917705535888672:  29%|██▉       | 1891/6470 [26:49<1:03:46,  1.20it/s]
1|1892|Loss: 4.917705535888672:  29%|██▉       | 1892/6470 [26:49<1:03:42,  1.20it/s]
1|1893|Loss: 4.5463643074035645:  29%|██▉       | 1892/6470 [26:50<1:03:42,  1.20it/s]
1|1893|Loss: 4.5463643074035645:  29%|██▉       | 1893/6470 [26:50<1:02:58,  1.21it/s]
1|1894|Loss: 4.690234661102295:  29%|██▉       | 1893/6470 [26:50<1:02:58,  1.21it/s] 
1|1894|Loss: 4.690234661102295:  29%|██▉       | 1894/6470 [26:51<1:00:51,  1.25it/s]
1|1895|Loss: 5.2356085777282715:  29%|██▉       | 1894/6470 [26:51<1:00:51,  1.25it/s]
1|1895|Loss: 5.2356085777282715:  29%|██▉       | 1895/6470 [26:51<1:00:26,  1.26it/s]
1|1896|Loss: 5.005707263946533:  29%|██▉       | 1895/6470 [26:52<1:00:26,  1.26it/s] 
1|1896|Loss: 5.005707263946533:  29%|██▉       | 1896/6470 [26:52<59:48,  1.27it/s]  
1|1897|Loss: 4.9636006355285645:  29%|██▉       | 1896/6470 [26:53<59:48,  1.27it/s]
1|1897|Loss: 4.9636006355285645:  29%|██▉       | 1897/6470 [26:53<59:17,  1.29it/s]
1|1898|Loss: 4.777237415313721:  29%|██▉       | 1897/6470 [26:53<59:17,  1.29it/s] 
1|1898|Loss: 4.777237415313721:  29%|██▉       | 1898/6470 [26:54<1:02:08,  1.23it/s]
1|1899|Loss: 4.534135818481445:  29%|██▉       | 1898/6470 [26:54<1:02:08,  1.23it/s]
1|1899|Loss: 4.534135818481445:  29%|██▉       | 1899/6470 [26:55<1:01:41,  1.23it/s]
1|1900|Loss: 4.75107479095459:  29%|██▉       | 1899/6470 [26:55<1:01:41,  1.23it/s] 
1|1900|Loss: 4.75107479095459:  29%|██▉       | 1900/6470 [26:56<1:05:51,  1.16it/s]
1|1901|Loss: 4.753730773925781:  29%|██▉       | 1900/6470 [26:56<1:05:51,  1.16it/s]
1|1901|Loss: 4.753730773925781:  29%|██▉       | 1901/6470 [26:56<1:03:59,  1.19it/s]
1|1902|Loss: 5.022869110107422:  29%|██▉       | 1901/6470 [26:57<1:03:59,  1.19it/s]
1|1902|Loss: 5.022869110107422:  29%|██▉       | 1902/6470 [26:57<1:02:05,  1.23it/s]
1|1903|Loss: 5.063500881195068:  29%|██▉       | 1902/6470 [26:58<1:02:05,  1.23it/s]
1|1903|Loss: 5.063500881195068:  29%|██▉       | 1903/6470 [26:58<1:00:45,  1.25it/s]
1|1904|Loss: 4.9882330894470215:  29%|██▉       | 1903/6470 [26:58<1:00:45,  1.25it/s]
1|1904|Loss: 4.9882330894470215:  29%|██▉       | 1904/6470 [26:59<1:00:55,  1.25it/s]
1|1905|Loss: 4.518192291259766:  29%|██▉       | 1904/6470 [26:59<1:00:55,  1.25it/s] 
1|1905|Loss: 4.518192291259766:  29%|██▉       | 1905/6470 [27:00<1:02:20,  1.22it/s]
1|1906|Loss: 5.049877643585205:  29%|██▉       | 1905/6470 [27:00<1:02:20,  1.22it/s]
1|1906|Loss: 5.049877643585205:  29%|██▉       | 1906/6470 [27:00<1:02:15,  1.22it/s]
1|1907|Loss: 4.947010517120361:  29%|██▉       | 1906/6470 [27:01<1:02:15,  1.22it/s]
1|1907|Loss: 4.947010517120361:  29%|██▉       | 1907/6470 [27:01<1:03:16,  1.20it/s]
1|1908|Loss: 5.010522842407227:  29%|██▉       | 1907/6470 [27:02<1:03:16,  1.20it/s]
1|1908|Loss: 5.010522842407227:  29%|██▉       | 1908/6470 [27:02<1:03:59,  1.19it/s]
1|1909|Loss: 4.358541965484619:  29%|██▉       | 1908/6470 [27:03<1:03:59,  1.19it/s]
1|1909|Loss: 4.358541965484619:  30%|██▉       | 1909/6470 [27:03<59:40,  1.27it/s]  
1|1910|Loss: 4.531242370605469:  30%|██▉       | 1909/6470 [27:03<59:40,  1.27it/s]
1|1910|Loss: 4.531242370605469:  30%|██▉       | 1910/6470 [27:04<58:46,  1.29it/s]
1|1911|Loss: 4.513711929321289:  30%|██▉       | 1910/6470 [27:04<58:46,  1.29it/s]
1|1911|Loss: 4.513711929321289:  30%|██▉       | 1911/6470 [27:04<54:06,  1.40it/s]
1|1912|Loss: 5.011270046234131:  30%|██▉       | 1911/6470 [27:04<54:06,  1.40it/s]
1|1912|Loss: 5.011270046234131:  30%|██▉       | 1912/6470 [27:05<53:49,  1.41it/s]
1|1913|Loss: 4.569122791290283:  30%|██▉       | 1912/6470 [27:05<53:49,  1.41it/s]
1|1913|Loss: 4.569122791290283:  30%|██▉       | 1913/6470 [27:05<52:44,  1.44it/s]
1|1914|Loss: 5.181248664855957:  30%|██▉       | 1913/6470 [27:06<52:44,  1.44it/s]
1|1914|Loss: 5.181248664855957:  30%|██▉       | 1914/6470 [27:06<54:25,  1.39it/s]
1|1915|Loss: 4.743128299713135:  30%|██▉       | 1914/6470 [27:07<54:25,  1.39it/s]
1|1915|Loss: 4.743128299713135:  30%|██▉       | 1915/6470 [27:07<54:20,  1.40it/s]
1|1916|Loss: 4.843085289001465:  30%|██▉       | 1915/6470 [27:07<54:20,  1.40it/s]
1|1916|Loss: 4.843085289001465:  30%|██▉       | 1916/6470 [27:08<58:26,  1.30it/s]
1|1917|Loss: 5.0537919998168945:  30%|██▉       | 1916/6470 [27:08<58:26,  1.30it/s]
1|1917|Loss: 5.0537919998168945:  30%|██▉       | 1917/6470 [27:09<1:01:36,  1.23it/s]
1|1918|Loss: 4.849306106567383:  30%|██▉       | 1917/6470 [27:09<1:01:36,  1.23it/s] 
1|1918|Loss: 4.849306106567383:  30%|██▉       | 1918/6470 [27:09<59:37,  1.27it/s]  
1|1919|Loss: 4.780431270599365:  30%|██▉       | 1918/6470 [27:10<59:37,  1.27it/s]
1|1919|Loss: 4.780431270599365:  30%|██▉       | 1919/6470 [27:10<58:54,  1.29it/s]
1|1920|Loss: 4.162721633911133:  30%|██▉       | 1919/6470 [27:11<58:54,  1.29it/s]
1|1920|Loss: 4.162721633911133:  30%|██▉       | 1920/6470 [27:11<53:30,  1.42it/s]
1|1921|Loss: 4.701174736022949:  30%|██▉       | 1920/6470 [27:11<53:30,  1.42it/s]
1|1921|Loss: 4.701174736022949:  30%|██▉       | 1921/6470 [27:11<54:14,  1.40it/s]
1|1922|Loss: 4.886640548706055:  30%|██▉       | 1921/6470 [27:12<54:14,  1.40it/s]
1|1922|Loss: 4.886640548706055:  30%|██▉       | 1922/6470 [27:12<58:14,  1.30it/s]
1|1923|Loss: 5.008878231048584:  30%|██▉       | 1922/6470 [27:13<58:14,  1.30it/s]
1|1923|Loss: 5.008878231048584:  30%|██▉       | 1923/6470 [27:13<1:01:31,  1.23it/s]
1|1924|Loss: 4.920267105102539:  30%|██▉       | 1923/6470 [27:14<1:01:31,  1.23it/s]
1|1924|Loss: 4.920267105102539:  30%|██▉       | 1924/6470 [27:14<59:42,  1.27it/s]  
1|1925|Loss: 4.106424808502197:  30%|██▉       | 1924/6470 [27:14<59:42,  1.27it/s]
1|1925|Loss: 4.106424808502197:  30%|██▉       | 1925/6470 [27:15<54:59,  1.38it/s]
1|1926|Loss: 4.807155609130859:  30%|██▉       | 1925/6470 [27:15<54:59,  1.38it/s]
1|1926|Loss: 4.807155609130859:  30%|██▉       | 1926/6470 [27:15<55:46,  1.36it/s]
1|1927|Loss: 4.4244465827941895:  30%|██▉       | 1926/6470 [27:16<55:46,  1.36it/s]
1|1927|Loss: 4.4244465827941895:  30%|██▉       | 1927/6470 [27:16<55:05,  1.37it/s]
1|1928|Loss: 4.715002536773682:  30%|██▉       | 1927/6470 [27:17<55:05,  1.37it/s] 
1|1928|Loss: 4.715002536773682:  30%|██▉       | 1928/6470 [27:17<58:11,  1.30it/s]
1|1929|Loss: 4.159977436065674:  30%|██▉       | 1928/6470 [27:17<58:11,  1.30it/s]
1|1929|Loss: 4.159977436065674:  30%|██▉       | 1929/6470 [27:18<58:29,  1.29it/s]
1|1930|Loss: 4.340827941894531:  30%|██▉       | 1929/6470 [27:18<58:29,  1.29it/s]
1|1930|Loss: 4.340827941894531:  30%|██▉       | 1930/6470 [27:18<56:34,  1.34it/s]
1|1931|Loss: 4.740940570831299:  30%|██▉       | 1930/6470 [27:19<56:34,  1.34it/s]
1|1931|Loss: 4.740940570831299:  30%|██▉       | 1931/6470 [27:19<53:53,  1.40it/s]
1|1932|Loss: 4.198634624481201:  30%|██▉       | 1931/6470 [27:19<53:53,  1.40it/s]
1|1932|Loss: 4.198634624481201:  30%|██▉       | 1932/6470 [27:20<53:21,  1.42it/s]
1|1933|Loss: 4.700112819671631:  30%|██▉       | 1932/6470 [27:20<53:21,  1.42it/s]
1|1933|Loss: 4.700112819671631:  30%|██▉       | 1933/6470 [27:20<52:55,  1.43it/s]
1|1934|Loss: 4.0282063484191895:  30%|██▉       | 1933/6470 [27:21<52:55,  1.43it/s]
1|1934|Loss: 4.0282063484191895:  30%|██▉       | 1934/6470 [27:21<50:51,  1.49it/s]
1|1935|Loss: 5.1252875328063965:  30%|██▉       | 1934/6470 [27:21<50:51,  1.49it/s]
1|1935|Loss: 5.1252875328063965:  30%|██▉       | 1935/6470 [27:22<54:22,  1.39it/s]
1|1936|Loss: 4.925983905792236:  30%|██▉       | 1935/6470 [27:22<54:22,  1.39it/s] 
1|1936|Loss: 4.925983905792236:  30%|██▉       | 1936/6470 [27:23<56:55,  1.33it/s]
1|1937|Loss: 5.029117107391357:  30%|██▉       | 1936/6470 [27:23<56:55,  1.33it/s]
1|1937|Loss: 5.029117107391357:  30%|██▉       | 1937/6470 [27:23<56:11,  1.34it/s]
1|1938|Loss: 4.9177751541137695:  30%|██▉       | 1937/6470 [27:24<56:11,  1.34it/s]
1|1938|Loss: 4.9177751541137695:  30%|██▉       | 1938/6470 [27:24<56:18,  1.34it/s]
1|1939|Loss: 4.945585250854492:  30%|██▉       | 1938/6470 [27:25<56:18,  1.34it/s] 
1|1939|Loss: 4.945585250854492:  30%|██▉       | 1939/6470 [27:25<58:41,  1.29it/s]
1|1940|Loss: 4.841538906097412:  30%|██▉       | 1939/6470 [27:25<58:41,  1.29it/s]
1|1940|Loss: 4.841538906097412:  30%|██▉       | 1940/6470 [27:26<57:15,  1.32it/s]
1|1941|Loss: 4.959519863128662:  30%|██▉       | 1940/6470 [27:26<57:15,  1.32it/s]
1|1941|Loss: 4.959519863128662:  30%|███       | 1941/6470 [27:27<1:00:33,  1.25it/s]
1|1942|Loss: 4.930429935455322:  30%|███       | 1941/6470 [27:27<1:00:33,  1.25it/s]
1|1942|Loss: 4.930429935455322:  30%|███       | 1942/6470 [27:28<1:02:49,  1.20it/s]
1|1943|Loss: 4.7208404541015625:  30%|███       | 1942/6470 [27:28<1:02:49,  1.20it/s]
1|1943|Loss: 4.7208404541015625:  30%|███       | 1943/6470 [27:28<58:08,  1.30it/s]  
1|1944|Loss: 5.1584906578063965:  30%|███       | 1943/6470 [27:29<58:08,  1.30it/s]
1|1944|Loss: 5.1584906578063965:  30%|███       | 1944/6470 [27:29<59:33,  1.27it/s]
1|1945|Loss: 5.138605117797852:  30%|███       | 1944/6470 [27:30<59:33,  1.27it/s] 
1|1945|Loss: 5.138605117797852:  30%|███       | 1945/6470 [27:30<1:02:41,  1.20it/s]
1|1946|Loss: 5.111215591430664:  30%|███       | 1945/6470 [27:30<1:02:41,  1.20it/s]
1|1946|Loss: 5.111215591430664:  30%|███       | 1946/6470 [27:31<1:02:08,  1.21it/s]
1|1947|Loss: 4.7980170249938965:  30%|███       | 1946/6470 [27:31<1:02:08,  1.21it/s]
1|1947|Loss: 4.7980170249938965:  30%|███       | 1947/6470 [27:31<1:00:45,  1.24it/s]
1|1948|Loss: 4.881308555603027:  30%|███       | 1947/6470 [27:32<1:00:45,  1.24it/s] 
1|1948|Loss: 4.881308555603027:  30%|███       | 1948/6470 [27:32<59:41,  1.26it/s]  
1|1949|Loss: 4.343343257904053:  30%|███       | 1948/6470 [27:33<59:41,  1.26it/s]
1|1949|Loss: 4.343343257904053:  30%|███       | 1949/6470 [27:33<59:18,  1.27it/s]
1|1950|Loss: 4.511829853057861:  30%|███       | 1949/6470 [27:33<59:18,  1.27it/s]
1|1950|Loss: 4.511829853057861:  30%|███       | 1950/6470 [27:34<1:00:36,  1.24it/s]
1|1951|Loss: 4.521218776702881:  30%|███       | 1950/6470 [27:34<1:00:36,  1.24it/s]
1|1951|Loss: 4.521218776702881:  30%|███       | 1951/6470 [27:34<56:29,  1.33it/s]  
1|1952|Loss: 4.624436855316162:  30%|███       | 1951/6470 [27:35<56:29,  1.33it/s]
1|1952|Loss: 4.624436855316162:  30%|███       | 1952/6470 [27:35<58:54,  1.28it/s]
1|1953|Loss: 4.244802951812744:  30%|███       | 1952/6470 [27:36<58:54,  1.28it/s]
1|1953|Loss: 4.244802951812744:  30%|███       | 1953/6470 [27:36<54:59,  1.37it/s]
1|1954|Loss: 4.959560394287109:  30%|███       | 1953/6470 [27:36<54:59,  1.37it/s]
1|1954|Loss: 4.959560394287109:  30%|███       | 1954/6470 [27:37<52:06,  1.44it/s]
1|1955|Loss: 4.43599271774292:  30%|███       | 1954/6470 [27:37<52:06,  1.44it/s] 
1|1955|Loss: 4.43599271774292:  30%|███       | 1955/6470 [27:37<50:08,  1.50it/s]
1|1956|Loss: 4.3551740646362305:  30%|███       | 1955/6470 [27:37<50:08,  1.50it/s]
1|1956|Loss: 4.3551740646362305:  30%|███       | 1956/6470 [27:38<47:29,  1.58it/s]
1|1957|Loss: 4.88593864440918:  30%|███       | 1956/6470 [27:38<47:29,  1.58it/s]  
1|1957|Loss: 4.88593864440918:  30%|███       | 1957/6470 [27:38<46:19,  1.62it/s]
1|1958|Loss: 4.509881973266602:  30%|███       | 1957/6470 [27:39<46:19,  1.62it/s]
1|1958|Loss: 4.509881973266602:  30%|███       | 1958/6470 [27:39<49:18,  1.53it/s]
1|1959|Loss: 3.863927125930786:  30%|███       | 1958/6470 [27:39<49:18,  1.53it/s]
1|1959|Loss: 3.863927125930786:  30%|███       | 1959/6470 [27:40<50:28,  1.49it/s]
1|1960|Loss: 4.5915961265563965:  30%|███       | 1959/6470 [27:40<50:28,  1.49it/s]
1|1960|Loss: 4.5915961265563965:  30%|███       | 1960/6470 [27:41<54:55,  1.37it/s]
1|1961|Loss: 5.161881923675537:  30%|███       | 1960/6470 [27:41<54:55,  1.37it/s] 
1|1961|Loss: 5.161881923675537:  30%|███       | 1961/6470 [27:41<57:10,  1.31it/s]
1|1962|Loss: 3.6963658332824707:  30%|███       | 1961/6470 [27:42<57:10,  1.31it/s]
1|1962|Loss: 3.6963658332824707:  30%|███       | 1962/6470 [27:42<55:17,  1.36it/s]
1|1963|Loss: 4.813211441040039:  30%|███       | 1962/6470 [27:42<55:17,  1.36it/s] 
1|1963|Loss: 4.813211441040039:  30%|███       | 1963/6470 [27:43<52:31,  1.43it/s]
1|1964|Loss: 4.382133960723877:  30%|███       | 1963/6470 [27:43<52:31,  1.43it/s]
1|1964|Loss: 4.382133960723877:  30%|███       | 1964/6470 [27:43<52:29,  1.43it/s]
1|1965|Loss: 4.836086273193359:  30%|███       | 1964/6470 [27:44<52:29,  1.43it/s]
1|1965|Loss: 4.836086273193359:  30%|███       | 1965/6470 [27:44<54:30,  1.38it/s]
1|1966|Loss: 4.791568279266357:  30%|███       | 1965/6470 [27:45<54:30,  1.38it/s]
1|1966|Loss: 4.791568279266357:  30%|███       | 1966/6470 [27:45<56:16,  1.33it/s]
1|1967|Loss: 4.595444679260254:  30%|███       | 1966/6470 [27:45<56:16,  1.33it/s]
1|1967|Loss: 4.595444679260254:  30%|███       | 1967/6470 [27:46<55:52,  1.34it/s]
1|1968|Loss: 4.714697360992432:  30%|███       | 1967/6470 [27:46<55:52,  1.34it/s]
1|1968|Loss: 4.714697360992432:  30%|███       | 1968/6470 [27:47<56:04,  1.34it/s]
1|1969|Loss: 4.605732440948486:  30%|███       | 1968/6470 [27:47<56:04,  1.34it/s]
1|1969|Loss: 4.605732440948486:  30%|███       | 1969/6470 [27:47<1:00:39,  1.24it/s]
1|1970|Loss: 4.92574405670166:  30%|███       | 1969/6470 [27:48<1:00:39,  1.24it/s] 
1|1970|Loss: 4.92574405670166:  30%|███       | 1970/6470 [27:48<1:01:53,  1.21it/s]
1|1971|Loss: 4.497678279876709:  30%|███       | 1970/6470 [27:49<1:01:53,  1.21it/s]
1|1971|Loss: 4.497678279876709:  30%|███       | 1971/6470 [27:49<1:01:28,  1.22it/s]
1|1972|Loss: 5.160934925079346:  30%|███       | 1971/6470 [27:50<1:01:28,  1.22it/s]
1|1972|Loss: 5.160934925079346:  30%|███       | 1972/6470 [27:50<1:00:53,  1.23it/s]
1|1973|Loss: 4.231438636779785:  30%|███       | 1972/6470 [27:50<1:00:53,  1.23it/s]
1|1973|Loss: 4.231438636779785:  30%|███       | 1973/6470 [27:51<58:13,  1.29it/s]  
1|1974|Loss: 4.735618591308594:  30%|███       | 1973/6470 [27:51<58:13,  1.29it/s]
1|1974|Loss: 4.735618591308594:  31%|███       | 1974/6470 [27:51<58:32,  1.28it/s]
1|1975|Loss: 4.758961200714111:  31%|███       | 1974/6470 [27:52<58:32,  1.28it/s]
1|1975|Loss: 4.758961200714111:  31%|███       | 1975/6470 [27:52<1:00:59,  1.23it/s]
1|1976|Loss: 4.83824348449707:  31%|███       | 1975/6470 [27:53<1:00:59,  1.23it/s] 
1|1976|Loss: 4.83824348449707:  31%|███       | 1976/6470 [27:53<1:00:28,  1.24it/s]
1|1977|Loss: 4.359246730804443:  31%|███       | 1976/6470 [27:53<1:00:28,  1.24it/s]
1|1977|Loss: 4.359246730804443:  31%|███       | 1977/6470 [27:54<55:36,  1.35it/s]  
1|1978|Loss: 4.819316387176514:  31%|███       | 1977/6470 [27:54<55:36,  1.35it/s]
1|1978|Loss: 4.819316387176514:  31%|███       | 1978/6470 [27:54<55:52,  1.34it/s]
1|1979|Loss: 4.359622478485107:  31%|███       | 1978/6470 [27:55<55:52,  1.34it/s]
1|1979|Loss: 4.359622478485107:  31%|███       | 1979/6470 [27:55<55:28,  1.35it/s]
1|1980|Loss: 4.576103210449219:  31%|███       | 1979/6470 [27:56<55:28,  1.35it/s]
1|1980|Loss: 4.576103210449219:  31%|███       | 1980/6470 [27:56<53:31,  1.40it/s]
1|1981|Loss: 4.918634414672852:  31%|███       | 1980/6470 [27:56<53:31,  1.40it/s]
1|1981|Loss: 4.918634414672852:  31%|███       | 1981/6470 [27:57<57:35,  1.30it/s]
1|1982|Loss: 4.382486343383789:  31%|███       | 1981/6470 [27:57<57:35,  1.30it/s]
1|1982|Loss: 4.382486343383789:  31%|███       | 1982/6470 [27:58<1:02:51,  1.19it/s]
1|1983|Loss: 4.829205513000488:  31%|███       | 1982/6470 [27:58<1:02:51,  1.19it/s]
1|1983|Loss: 4.829205513000488:  31%|███       | 1983/6470 [27:59<1:02:53,  1.19it/s]
1|1984|Loss: 4.752636909484863:  31%|███       | 1983/6470 [27:59<1:02:53,  1.19it/s]
1|1984|Loss: 4.752636909484863:  31%|███       | 1984/6470 [27:59<1:03:44,  1.17it/s]
1|1985|Loss: 4.664055824279785:  31%|███       | 1984/6470 [28:00<1:03:44,  1.17it/s]
1|1985|Loss: 4.664055824279785:  31%|███       | 1985/6470 [28:00<1:04:19,  1.16it/s]
1|1986|Loss: 4.7918524742126465:  31%|███       | 1985/6470 [28:01<1:04:19,  1.16it/s]
1|1986|Loss: 4.7918524742126465:  31%|███       | 1986/6470 [28:01<58:40,  1.27it/s]  
1|1987|Loss: 4.603951454162598:  31%|███       | 1986/6470 [28:01<58:40,  1.27it/s] 
1|1987|Loss: 4.603951454162598:  31%|███       | 1987/6470 [28:02<56:04,  1.33it/s]
1|1988|Loss: 4.994110107421875:  31%|███       | 1987/6470 [28:02<56:04,  1.33it/s]
1|1988|Loss: 4.994110107421875:  31%|███       | 1988/6470 [28:02<56:48,  1.32it/s]
1|1989|Loss: 4.594790935516357:  31%|███       | 1988/6470 [28:03<56:48,  1.32it/s]
1|1989|Loss: 4.594790935516357:  31%|███       | 1989/6470 [28:03<57:25,  1.30it/s]
1|1990|Loss: 4.800445556640625:  31%|███       | 1989/6470 [28:04<57:25,  1.30it/s]
1|1990|Loss: 4.800445556640625:  31%|███       | 1990/6470 [28:04<59:15,  1.26it/s]
1|1991|Loss: 4.903623580932617:  31%|███       | 1990/6470 [28:04<59:15,  1.26it/s]
1|1991|Loss: 4.903623580932617:  31%|███       | 1991/6470 [28:05<58:49,  1.27it/s]
1|1992|Loss: 4.607656002044678:  31%|███       | 1991/6470 [28:05<58:49,  1.27it/s]
1|1992|Loss: 4.607656002044678:  31%|███       | 1992/6470 [28:05<54:21,  1.37it/s]
1|1993|Loss: 4.025506019592285:  31%|███       | 1992/6470 [28:06<54:21,  1.37it/s]
1|1993|Loss: 4.025506019592285:  31%|███       | 1993/6470 [28:06<49:10,  1.52it/s]
1|1994|Loss: 4.996114253997803:  31%|███       | 1993/6470 [28:06<49:10,  1.52it/s]
1|1994|Loss: 4.996114253997803:  31%|███       | 1994/6470 [28:07<50:26,  1.48it/s]
1|1995|Loss: 4.944242000579834:  31%|███       | 1994/6470 [28:07<50:26,  1.48it/s]
1|1995|Loss: 4.944242000579834:  31%|███       | 1995/6470 [28:07<51:30,  1.45it/s]
1|1996|Loss: 4.750770092010498:  31%|███       | 1995/6470 [28:08<51:30,  1.45it/s]
1|1996|Loss: 4.750770092010498:  31%|███       | 1996/6470 [28:08<51:20,  1.45it/s]
1|1997|Loss: 5.012516975402832:  31%|███       | 1996/6470 [28:08<51:20,  1.45it/s]
1|1997|Loss: 5.012516975402832:  31%|███       | 1997/6470 [28:09<53:11,  1.40it/s]
1|1998|Loss: 4.3854498863220215:  31%|███       | 1997/6470 [28:09<53:11,  1.40it/s]
1|1998|Loss: 4.3854498863220215:  31%|███       | 1998/6470 [28:10<56:17,  1.32it/s]
1|1999|Loss: 4.430286884307861:  31%|███       | 1998/6470 [28:10<56:17,  1.32it/s] 
1|1999|Loss: 4.430286884307861:  31%|███       | 1999/6470 [28:10<53:28,  1.39it/s]
1|2000|Loss: 4.603545665740967:  31%|███       | 1999/6470 [28:11<53:28,  1.39it/s]
1|2000|Loss: 4.603545665740967:  31%|███       | 2000/6470 [28:11<51:03,  1.46it/s]
1|2001|Loss: 5.070216178894043:  31%|███       | 2000/6470 [28:11<51:03,  1.46it/s]
1|2001|Loss: 5.070216178894043:  31%|███       | 2001/6470 [28:12<52:57,  1.41it/s]
1|2002|Loss: 3.8941972255706787:  31%|███       | 2001/6470 [28:12<52:57,  1.41it/s]
1|2002|Loss: 3.8941972255706787:  31%|███       | 2002/6470 [28:12<47:33,  1.57it/s]
1|2003|Loss: 4.499204635620117:  31%|███       | 2002/6470 [28:12<47:33,  1.57it/s] 
1|2003|Loss: 4.499204635620117:  31%|███       | 2003/6470 [28:13<49:05,  1.52it/s]
1|2004|Loss: 4.982021808624268:  31%|███       | 2003/6470 [28:13<49:05,  1.52it/s]
1|2004|Loss: 4.982021808624268:  31%|███       | 2004/6470 [28:14<54:13,  1.37it/s]
1|2005|Loss: 4.996500492095947:  31%|███       | 2004/6470 [28:14<54:13,  1.37it/s]
1|2005|Loss: 4.996500492095947:  31%|███       | 2005/6470 [28:15<59:04,  1.26it/s]
1|2006|Loss: 5.099322319030762:  31%|███       | 2005/6470 [28:15<59:04,  1.26it/s]
1|2006|Loss: 5.099322319030762:  31%|███       | 2006/6470 [28:16<1:01:44,  1.20it/s]
1|2007|Loss: 4.2250800132751465:  31%|███       | 2006/6470 [28:16<1:01:44,  1.20it/s]
1|2007|Loss: 4.2250800132751465:  31%|███       | 2007/6470 [28:16<59:04,  1.26it/s]  
1|2008|Loss: 4.262435436248779:  31%|███       | 2007/6470 [28:17<59:04,  1.26it/s] 
1|2008|Loss: 4.262435436248779:  31%|███       | 2008/6470 [28:17<58:27,  1.27it/s]
1|2009|Loss: 4.634154796600342:  31%|███       | 2008/6470 [28:18<58:27,  1.27it/s]
1|2009|Loss: 4.634154796600342:  31%|███       | 2009/6470 [28:18<57:56,  1.28it/s]
1|2010|Loss: 4.45595121383667:  31%|███       | 2009/6470 [28:18<57:56,  1.28it/s] 
1|2010|Loss: 4.45595121383667:  31%|███       | 2010/6470 [28:19<57:25,  1.29it/s]
1|2011|Loss: 4.3288893699646:  31%|███       | 2010/6470 [28:19<57:25,  1.29it/s] 
1|2011|Loss: 4.3288893699646:  31%|███       | 2011/6470 [28:19<55:59,  1.33it/s]
1|2012|Loss: 4.440206527709961:  31%|███       | 2011/6470 [28:20<55:59,  1.33it/s]
1|2012|Loss: 4.440206527709961:  31%|███       | 2012/6470 [28:20<53:45,  1.38it/s]
1|2013|Loss: 4.932809829711914:  31%|███       | 2012/6470 [28:20<53:45,  1.38it/s]
1|2013|Loss: 4.932809829711914:  31%|███       | 2013/6470 [28:21<55:47,  1.33it/s]
1|2014|Loss: 4.94246244430542:  31%|███       | 2013/6470 [28:21<55:47,  1.33it/s] 
1|2014|Loss: 4.94246244430542:  31%|███       | 2014/6470 [28:22<56:37,  1.31it/s]
1|2015|Loss: 4.159670352935791:  31%|███       | 2014/6470 [28:22<56:37,  1.31it/s]
1|2015|Loss: 4.159670352935791:  31%|███       | 2015/6470 [28:22<57:04,  1.30it/s]
1|2016|Loss: 4.611878395080566:  31%|███       | 2015/6470 [28:23<57:04,  1.30it/s]
1|2016|Loss: 4.611878395080566:  31%|███       | 2016/6470 [28:23<53:34,  1.39it/s]
1|2017|Loss: 4.354329586029053:  31%|███       | 2016/6470 [28:23<53:34,  1.39it/s]
1|2017|Loss: 4.354329586029053:  31%|███       | 2017/6470 [28:24<52:15,  1.42it/s]
1|2018|Loss: 4.745022773742676:  31%|███       | 2017/6470 [28:24<52:15,  1.42it/s]
1|2018|Loss: 4.745022773742676:  31%|███       | 2018/6470 [28:24<54:00,  1.37it/s]
1|2019|Loss: 4.720463275909424:  31%|███       | 2018/6470 [28:25<54:00,  1.37it/s]
1|2019|Loss: 4.720463275909424:  31%|███       | 2019/6470 [28:25<52:28,  1.41it/s]
1|2020|Loss: 5.0668206214904785:  31%|███       | 2019/6470 [28:25<52:28,  1.41it/s]
1|2020|Loss: 5.0668206214904785:  31%|███       | 2020/6470 [28:26<54:53,  1.35it/s]
1|2021|Loss: 4.553141117095947:  31%|███       | 2020/6470 [28:26<54:53,  1.35it/s] 
1|2021|Loss: 4.553141117095947:  31%|███       | 2021/6470 [28:27<57:34,  1.29it/s]
1|2022|Loss: 3.7123310565948486:  31%|███       | 2021/6470 [28:27<57:34,  1.29it/s]
1|2022|Loss: 3.7123310565948486:  31%|███▏      | 2022/6470 [28:27<51:24,  1.44it/s]
1|2023|Loss: 4.768063545227051:  31%|███▏      | 2022/6470 [28:28<51:24,  1.44it/s] 
1|2023|Loss: 4.768063545227051:  31%|███▏      | 2023/6470 [28:28<50:25,  1.47it/s]
1|2024|Loss: 5.198605060577393:  31%|███▏      | 2023/6470 [28:28<50:25,  1.47it/s]
1|2024|Loss: 5.198605060577393:  31%|███▏      | 2024/6470 [28:29<54:47,  1.35it/s]
1|2025|Loss: 4.506339073181152:  31%|███▏      | 2024/6470 [28:29<54:47,  1.35it/s]
1|2025|Loss: 4.506339073181152:  31%|███▏      | 2025/6470 [28:30<56:22,  1.31it/s]
1|2026|Loss: 5.140873908996582:  31%|███▏      | 2025/6470 [28:30<56:22,  1.31it/s]
1|2026|Loss: 5.140873908996582:  31%|███▏      | 2026/6470 [28:30<56:34,  1.31it/s]
1|2027|Loss: 4.940193176269531:  31%|███▏      | 2026/6470 [28:31<56:34,  1.31it/s]
1|2027|Loss: 4.940193176269531:  31%|███▏      | 2027/6470 [28:31<58:26,  1.27it/s]
1|2028|Loss: 4.515279293060303:  31%|███▏      | 2027/6470 [28:32<58:26,  1.27it/s]
1|2028|Loss: 4.515279293060303:  31%|███▏      | 2028/6470 [28:32<57:12,  1.29it/s]
1|2029|Loss: 4.154665470123291:  31%|███▏      | 2028/6470 [28:32<57:12,  1.29it/s]
1|2029|Loss: 4.154665470123291:  31%|███▏      | 2029/6470 [28:33<53:35,  1.38it/s]
1|2030|Loss: 4.726309299468994:  31%|███▏      | 2029/6470 [28:33<53:35,  1.38it/s]
1|2030|Loss: 4.726309299468994:  31%|███▏      | 2030/6470 [28:33<54:31,  1.36it/s]
1|2031|Loss: 4.544002056121826:  31%|███▏      | 2030/6470 [28:34<54:31,  1.36it/s]
1|2031|Loss: 4.544002056121826:  31%|███▏      | 2031/6470 [28:34<53:43,  1.38it/s]
1|2032|Loss: 5.014767169952393:  31%|███▏      | 2031/6470 [28:34<53:43,  1.38it/s]
1|2032|Loss: 5.014767169952393:  31%|███▏      | 2032/6470 [28:35<58:21,  1.27it/s]
1|2033|Loss: 5.066415786743164:  31%|███▏      | 2032/6470 [28:35<58:21,  1.27it/s]
1|2033|Loss: 5.066415786743164:  31%|███▏      | 2033/6470 [28:36<1:01:54,  1.19it/s]
1|2034|Loss: 4.828640937805176:  31%|███▏      | 2033/6470 [28:36<1:01:54,  1.19it/s]
1|2034|Loss: 4.828640937805176:  31%|███▏      | 2034/6470 [28:37<59:22,  1.25it/s]  
1|2035|Loss: 4.795395851135254:  31%|███▏      | 2034/6470 [28:37<59:22,  1.25it/s]
1|2035|Loss: 4.795395851135254:  31%|███▏      | 2035/6470 [28:37<58:45,  1.26it/s]
1|2036|Loss: 4.74796724319458:  31%|███▏      | 2035/6470 [28:38<58:45,  1.26it/s] 
1|2036|Loss: 4.74796724319458:  31%|███▏      | 2036/6470 [28:38<1:00:04,  1.23it/s]
1|2037|Loss: 4.331359386444092:  31%|███▏      | 2036/6470 [28:39<1:00:04,  1.23it/s]
1|2037|Loss: 4.331359386444092:  31%|███▏      | 2037/6470 [28:39<57:55,  1.28it/s]  
1|2038|Loss: 4.995499134063721:  31%|███▏      | 2037/6470 [28:39<57:55,  1.28it/s]
1|2038|Loss: 4.995499134063721:  31%|███▏      | 2038/6470 [28:40<58:03,  1.27it/s]
1|2039|Loss: 4.961095333099365:  31%|███▏      | 2038/6470 [28:40<58:03,  1.27it/s]
1|2039|Loss: 4.961095333099365:  32%|███▏      | 2039/6470 [28:41<58:40,  1.26it/s]
1|2040|Loss: 4.455604076385498:  32%|███▏      | 2039/6470 [28:41<58:40,  1.26it/s]
1|2040|Loss: 4.455604076385498:  32%|███▏      | 2040/6470 [28:41<56:45,  1.30it/s]
1|2041|Loss: 4.008322715759277:  32%|███▏      | 2040/6470 [28:42<56:45,  1.30it/s]
1|2041|Loss: 4.008322715759277:  32%|███▏      | 2041/6470 [28:42<52:08,  1.42it/s]
1|2042|Loss: 4.630858898162842:  32%|███▏      | 2041/6470 [28:42<52:08,  1.42it/s]
1|2042|Loss: 4.630858898162842:  32%|███▏      | 2042/6470 [28:43<51:15,  1.44it/s]
1|2043|Loss: 4.476683139801025:  32%|███▏      | 2042/6470 [28:43<51:15,  1.44it/s]
1|2043|Loss: 4.476683139801025:  32%|███▏      | 2043/6470 [28:43<52:42,  1.40it/s]
1|2044|Loss: 4.6865553855896:  32%|███▏      | 2043/6470 [28:44<52:42,  1.40it/s]  
1|2044|Loss: 4.6865553855896:  32%|███▏      | 2044/6470 [28:44<57:09,  1.29it/s]
1|2045|Loss: 4.626902103424072:  32%|███▏      | 2044/6470 [28:45<57:09,  1.29it/s]
1|2045|Loss: 4.626902103424072:  32%|███▏      | 2045/6470 [28:45<58:47,  1.25it/s]
1|2046|Loss: 4.82757043838501:  32%|███▏      | 2045/6470 [28:45<58:47,  1.25it/s] 
1|2046|Loss: 4.82757043838501:  32%|███▏      | 2046/6470 [28:46<56:53,  1.30it/s]
1|2047|Loss: 4.664708614349365:  32%|███▏      | 2046/6470 [28:46<56:53,  1.30it/s]
1|2047|Loss: 4.664708614349365:  32%|███▏      | 2047/6470 [28:47<58:28,  1.26it/s]
1|2048|Loss: 5.127233982086182:  32%|███▏      | 2047/6470 [28:47<58:28,  1.26it/s]
1|2048|Loss: 5.127233982086182:  32%|███▏      | 2048/6470 [28:47<1:00:51,  1.21it/s]
1|2049|Loss: 4.641616344451904:  32%|███▏      | 2048/6470 [28:48<1:00:51,  1.21it/s]
1|2049|Loss: 4.641616344451904:  32%|███▏      | 2049/6470 [28:48<58:29,  1.26it/s]  
1|2050|Loss: 5.294253826141357:  32%|███▏      | 2049/6470 [28:49<58:29,  1.26it/s]
1|2050|Loss: 5.294253826141357:  32%|███▏      | 2050/6470 [28:49<1:00:14,  1.22it/s]
1|2051|Loss: 4.976721286773682:  32%|███▏      | 2050/6470 [28:50<1:00:14,  1.22it/s]
1|2051|Loss: 4.976721286773682:  32%|███▏      | 2051/6470 [28:50<1:02:25,  1.18it/s]
1|2052|Loss: 5.082939624786377:  32%|███▏      | 2051/6470 [28:50<1:02:25,  1.18it/s]
1|2052|Loss: 5.082939624786377:  32%|███▏      | 2052/6470 [28:51<1:02:22,  1.18it/s]
1|2053|Loss: 4.997297763824463:  32%|███▏      | 2052/6470 [28:51<1:02:22,  1.18it/s]
1|2053|Loss: 4.997297763824463:  32%|███▏      | 2053/6470 [28:52<1:02:52,  1.17it/s]
1|2054|Loss: 4.322334289550781:  32%|███▏      | 2053/6470 [28:52<1:02:52,  1.17it/s]
1|2054|Loss: 4.322334289550781:  32%|███▏      | 2054/6470 [28:52<59:46,  1.23it/s]  
1|2055|Loss: 5.057184219360352:  32%|███▏      | 2054/6470 [28:53<59:46,  1.23it/s]
1|2055|Loss: 5.057184219360352:  32%|███▏      | 2055/6470 [28:53<1:02:22,  1.18it/s]
1|2056|Loss: 4.888563632965088:  32%|███▏      | 2055/6470 [28:54<1:02:22,  1.18it/s]
1|2056|Loss: 4.888563632965088:  32%|███▏      | 2056/6470 [28:54<1:04:08,  1.15it/s]
1|2057|Loss: 4.942014694213867:  32%|███▏      | 2056/6470 [28:55<1:04:08,  1.15it/s]
1|2057|Loss: 4.942014694213867:  32%|███▏      | 2057/6470 [28:55<1:01:24,  1.20it/s]
1|2058|Loss: 5.006298542022705:  32%|███▏      | 2057/6470 [28:56<1:01:24,  1.20it/s]
1|2058|Loss: 5.006298542022705:  32%|███▏      | 2058/6470 [28:56<1:03:59,  1.15it/s]
1|2059|Loss: 4.257785320281982:  32%|███▏      | 2058/6470 [28:56<1:03:59,  1.15it/s]
1|2059|Loss: 4.257785320281982:  32%|███▏      | 2059/6470 [28:57<1:01:35,  1.19it/s]
1|2060|Loss: 4.405270099639893:  32%|███▏      | 2059/6470 [28:57<1:01:35,  1.19it/s]
1|2060|Loss: 4.405270099639893:  32%|███▏      | 2060/6470 [28:57<56:42,  1.30it/s]  
1|2061|Loss: 4.450846195220947:  32%|███▏      | 2060/6470 [28:58<56:42,  1.30it/s]
1|2061|Loss: 4.450846195220947:  32%|███▏      | 2061/6470 [28:58<54:44,  1.34it/s]
1|2062|Loss: 4.7693586349487305:  32%|███▏      | 2061/6470 [28:59<54:44,  1.34it/s]
1|2062|Loss: 4.7693586349487305:  32%|███▏      | 2062/6470 [28:59<57:29,  1.28it/s]
1|2063|Loss: 4.8660125732421875:  32%|███▏      | 2062/6470 [28:59<57:29,  1.28it/s]
1|2063|Loss: 4.8660125732421875:  32%|███▏      | 2063/6470 [29:00<56:03,  1.31it/s]
1|2064|Loss: 4.430721759796143:  32%|███▏      | 2063/6470 [29:00<56:03,  1.31it/s] 
1|2064|Loss: 4.430721759796143:  32%|███▏      | 2064/6470 [29:00<54:36,  1.34it/s]
1|2065|Loss: 4.570536136627197:  32%|███▏      | 2064/6470 [29:01<54:36,  1.34it/s]
1|2065|Loss: 4.570536136627197:  32%|███▏      | 2065/6470 [29:01<53:45,  1.37it/s]
1|2066|Loss: 4.418663024902344:  32%|███▏      | 2065/6470 [29:01<53:45,  1.37it/s]
1|2066|Loss: 4.418663024902344:  32%|███▏      | 2066/6470 [29:02<51:50,  1.42it/s]
1|2067|Loss: 4.701064109802246:  32%|███▏      | 2066/6470 [29:02<51:50,  1.42it/s]
1|2067|Loss: 4.701064109802246:  32%|███▏      | 2067/6470 [29:02<52:48,  1.39it/s]
1|2068|Loss: 4.512167453765869:  32%|███▏      | 2067/6470 [29:03<52:48,  1.39it/s]
1|2068|Loss: 4.512167453765869:  32%|███▏      | 2068/6470 [29:03<55:54,  1.31it/s]
1|2069|Loss: 5.2137451171875:  32%|███▏      | 2068/6470 [29:04<55:54,  1.31it/s]  
1|2069|Loss: 5.2137451171875:  32%|███▏      | 2069/6470 [29:04<55:19,  1.33it/s]
1|2070|Loss: 5.298881530761719:  32%|███▏      | 2069/6470 [29:04<55:19,  1.33it/s]
1|2070|Loss: 5.298881530761719:  32%|███▏      | 2070/6470 [29:05<58:13,  1.26it/s]
1|2071|Loss: 4.811887264251709:  32%|███▏      | 2070/6470 [29:05<58:13,  1.26it/s]
1|2071|Loss: 4.811887264251709:  32%|███▏      | 2071/6470 [29:06<57:09,  1.28it/s]
1|2072|Loss: 5.869930267333984:  32%|███▏      | 2071/6470 [29:06<57:09,  1.28it/s]
1|2072|Loss: 5.869930267333984:  32%|███▏      | 2072/6470 [29:07<58:43,  1.25it/s]
1|2073|Loss: 5.710726261138916:  32%|███▏      | 2072/6470 [29:07<58:43,  1.25it/s]
1|2073|Loss: 5.710726261138916:  32%|███▏      | 2073/6470 [29:07<59:09,  1.24it/s]
1|2074|Loss: 4.844525337219238:  32%|███▏      | 2073/6470 [29:08<59:09,  1.24it/s]
1|2074|Loss: 4.844525337219238:  32%|███▏      | 2074/6470 [29:08<58:21,  1.26it/s]
1|2075|Loss: 4.670273780822754:  32%|███▏      | 2074/6470 [29:09<58:21,  1.26it/s]
1|2075|Loss: 4.670273780822754:  32%|███▏      | 2075/6470 [29:09<58:22,  1.25it/s]
1|2076|Loss: 4.71882963180542:  32%|███▏      | 2075/6470 [29:09<58:22,  1.25it/s] 
1|2076|Loss: 4.71882963180542:  32%|███▏      | 2076/6470 [29:10<58:51,  1.24it/s]
1|2077|Loss: 4.363955497741699:  32%|███▏      | 2076/6470 [29:10<58:51,  1.24it/s]
1|2077|Loss: 4.363955497741699:  32%|███▏      | 2077/6470 [29:11<1:01:08,  1.20it/s]
1|2078|Loss: 4.828411102294922:  32%|███▏      | 2077/6470 [29:11<1:01:08,  1.20it/s]
1|2078|Loss: 4.828411102294922:  32%|███▏      | 2078/6470 [29:11<1:00:21,  1.21it/s]
1|2079|Loss: 5.064642906188965:  32%|███▏      | 2078/6470 [29:12<1:00:21,  1.21it/s]
1|2079|Loss: 5.064642906188965:  32%|███▏      | 2079/6470 [29:12<59:04,  1.24it/s]  
1|2080|Loss: 5.059035778045654:  32%|███▏      | 2079/6470 [29:13<59:04,  1.24it/s]
1|2080|Loss: 5.059035778045654:  32%|███▏      | 2080/6470 [29:13<58:41,  1.25it/s]
1|2081|Loss: 4.739441394805908:  32%|███▏      | 2080/6470 [29:13<58:41,  1.25it/s]
1|2081|Loss: 4.739441394805908:  32%|███▏      | 2081/6470 [29:14<56:21,  1.30it/s]
1|2082|Loss: 4.436370372772217:  32%|███▏      | 2081/6470 [29:14<56:21,  1.30it/s]
1|2082|Loss: 4.436370372772217:  32%|███▏      | 2082/6470 [29:14<55:48,  1.31it/s]
1|2083|Loss: 4.826611518859863:  32%|███▏      | 2082/6470 [29:15<55:48,  1.31it/s]
1|2083|Loss: 4.826611518859863:  32%|███▏      | 2083/6470 [29:15<58:31,  1.25it/s]
1|2084|Loss: 4.544292449951172:  32%|███▏      | 2083/6470 [29:16<58:31,  1.25it/s]
1|2084|Loss: 4.544292449951172:  32%|███▏      | 2084/6470 [29:16<58:13,  1.26it/s]
1|2085|Loss: 4.596356391906738:  32%|███▏      | 2084/6470 [29:17<58:13,  1.26it/s]
1|2085|Loss: 4.596356391906738:  32%|███▏      | 2085/6470 [29:17<1:00:07,  1.22it/s]
1|2086|Loss: 4.934671401977539:  32%|███▏      | 2085/6470 [29:17<1:00:07,  1.22it/s]
1|2086|Loss: 4.934671401977539:  32%|███▏      | 2086/6470 [29:18<1:01:34,  1.19it/s]
1|2087|Loss: 3.1321616172790527:  32%|███▏      | 2086/6470 [29:18<1:01:34,  1.19it/s]
1|2087|Loss: 3.1321616172790527:  32%|███▏      | 2087/6470 [29:18<53:25,  1.37it/s]  
1|2088|Loss: 3.9860899448394775:  32%|███▏      | 2087/6470 [29:19<53:25,  1.37it/s]
1|2088|Loss: 3.9860899448394775:  32%|███▏      | 2088/6470 [29:19<50:43,  1.44it/s]
1|2089|Loss: 5.073370456695557:  32%|███▏      | 2088/6470 [29:19<50:43,  1.44it/s] 
1|2089|Loss: 5.073370456695557:  32%|███▏      | 2089/6470 [29:20<55:41,  1.31it/s]
1|2090|Loss: 4.904321193695068:  32%|███▏      | 2089/6470 [29:20<55:41,  1.31it/s]
1|2090|Loss: 4.904321193695068:  32%|███▏      | 2090/6470 [29:21<59:02,  1.24it/s]
1|2091|Loss: 4.694737911224365:  32%|███▏      | 2090/6470 [29:21<59:02,  1.24it/s]
1|2091|Loss: 4.694737911224365:  32%|███▏      | 2091/6470 [29:22<59:22,  1.23it/s]
1|2092|Loss: 4.743601322174072:  32%|███▏      | 2091/6470 [29:22<59:22,  1.23it/s]
1|2092|Loss: 4.743601322174072:  32%|███▏      | 2092/6470 [29:22<56:55,  1.28it/s]
1|2093|Loss: 4.743026256561279:  32%|███▏      | 2092/6470 [29:23<56:55,  1.28it/s]
1|2093|Loss: 4.743026256561279:  32%|███▏      | 2093/6470 [29:23<56:18,  1.30it/s]
1|2094|Loss: 4.49983549118042:  32%|███▏      | 2093/6470 [29:24<56:18,  1.30it/s] 
1|2094|Loss: 4.49983549118042:  32%|███▏      | 2094/6470 [29:24<56:26,  1.29it/s]
1|2095|Loss: 4.377139091491699:  32%|███▏      | 2094/6470 [29:24<56:26,  1.29it/s]
1|2095|Loss: 4.377139091491699:  32%|███▏      | 2095/6470 [29:25<56:21,  1.29it/s]
1|2096|Loss: 3.505425453186035:  32%|███▏      | 2095/6470 [29:25<56:21,  1.29it/s]
1|2096|Loss: 3.505425453186035:  32%|███▏      | 2096/6470 [29:25<51:02,  1.43it/s]
1|2097|Loss: 4.75175666809082:  32%|███▏      | 2096/6470 [29:26<51:02,  1.43it/s] 
1|2097|Loss: 4.75175666809082:  32%|███▏      | 2097/6470 [29:26<49:51,  1.46it/s]
1|2098|Loss: 3.887338161468506:  32%|███▏      | 2097/6470 [29:26<49:51,  1.46it/s]
1|2098|Loss: 3.887338161468506:  32%|███▏      | 2098/6470 [29:26<49:47,  1.46it/s]
1|2099|Loss: 4.552371501922607:  32%|███▏      | 2098/6470 [29:27<49:47,  1.46it/s]
1|2099|Loss: 4.552371501922607:  32%|███▏      | 2099/6470 [29:27<48:03,  1.52it/s]
1|2100|Loss: 4.57327938079834:  32%|███▏      | 2099/6470 [29:27<48:03,  1.52it/s] 
1|2100|Loss: 4.57327938079834:  32%|███▏      | 2100/6470 [29:28<47:59,  1.52it/s]
1|2101|Loss: 4.761475563049316:  32%|███▏      | 2100/6470 [29:28<47:59,  1.52it/s]
1|2101|Loss: 4.761475563049316:  32%|███▏      | 2101/6470 [29:29<49:54,  1.46it/s]
1|2102|Loss: 4.916047096252441:  32%|███▏      | 2101/6470 [29:29<49:54,  1.46it/s]
1|2102|Loss: 4.916047096252441:  32%|███▏      | 2102/6470 [29:29<52:07,  1.40it/s]
1|2103|Loss: 4.326394081115723:  32%|███▏      | 2102/6470 [29:30<52:07,  1.40it/s]
1|2103|Loss: 4.326394081115723:  33%|███▎      | 2103/6470 [29:30<54:10,  1.34it/s]
1|2104|Loss: 4.902428150177002:  33%|███▎      | 2103/6470 [29:31<54:10,  1.34it/s]
1|2104|Loss: 4.902428150177002:  33%|███▎      | 2104/6470 [29:31<57:26,  1.27it/s]
1|2105|Loss: 4.560391426086426:  33%|███▎      | 2104/6470 [29:31<57:26,  1.27it/s]
1|2105|Loss: 4.560391426086426:  33%|███▎      | 2105/6470 [29:32<59:54,  1.21it/s]
1|2106|Loss: 5.027873516082764:  33%|███▎      | 2105/6470 [29:32<59:54,  1.21it/s]
1|2106|Loss: 5.027873516082764:  33%|███▎      | 2106/6470 [29:33<59:44,  1.22it/s]
1|2107|Loss: 4.255978584289551:  33%|███▎      | 2106/6470 [29:33<59:44,  1.22it/s]
1|2107|Loss: 4.255978584289551:  33%|███▎      | 2107/6470 [29:34<59:03,  1.23it/s]
1|2108|Loss: 4.913733959197998:  33%|███▎      | 2107/6470 [29:34<59:03,  1.23it/s]
1|2108|Loss: 4.913733959197998:  33%|███▎      | 2108/6470 [29:34<57:10,  1.27it/s]
1|2109|Loss: 4.78033971786499:  33%|███▎      | 2108/6470 [29:35<57:10,  1.27it/s] 
1|2109|Loss: 4.78033971786499:  33%|███▎      | 2109/6470 [29:35<58:23,  1.24it/s]
1|2110|Loss: 4.526660919189453:  33%|███▎      | 2109/6470 [29:36<58:23,  1.24it/s]
1|2110|Loss: 4.526660919189453:  33%|███▎      | 2110/6470 [29:36<58:40,  1.24it/s]
1|2111|Loss: 4.831823825836182:  33%|███▎      | 2110/6470 [29:36<58:40,  1.24it/s]
1|2111|Loss: 4.831823825836182:  33%|███▎      | 2111/6470 [29:37<1:00:18,  1.20it/s]
1|2112|Loss: 4.9782328605651855:  33%|███▎      | 2111/6470 [29:37<1:00:18,  1.20it/s]
1|2112|Loss: 4.9782328605651855:  33%|███▎      | 2112/6470 [29:38<1:03:38,  1.14it/s]
1|2113|Loss: 4.781454563140869:  33%|███▎      | 2112/6470 [29:38<1:03:38,  1.14it/s] 
1|2113|Loss: 4.781454563140869:  33%|███▎      | 2113/6470 [29:39<1:02:44,  1.16it/s]
1|2114|Loss: 4.171412944793701:  33%|███▎      | 2113/6470 [29:39<1:02:44,  1.16it/s]
1|2114|Loss: 4.171412944793701:  33%|███▎      | 2114/6470 [29:39<1:02:28,  1.16it/s]
1|2115|Loss: 4.2126874923706055:  33%|███▎      | 2114/6470 [29:40<1:02:28,  1.16it/s]
1|2115|Loss: 4.2126874923706055:  33%|███▎      | 2115/6470 [29:40<57:48,  1.26it/s]  
1|2116|Loss: 4.922954559326172:  33%|███▎      | 2115/6470 [29:40<57:48,  1.26it/s] 
1|2116|Loss: 4.922954559326172:  33%|███▎      | 2116/6470 [29:41<56:55,  1.27it/s]
1|2117|Loss: 4.646845817565918:  33%|███▎      | 2116/6470 [29:41<56:55,  1.27it/s]
1|2117|Loss: 4.646845817565918:  33%|███▎      | 2117/6470 [29:42<58:39,  1.24it/s]
1|2118|Loss: 4.516308784484863:  33%|███▎      | 2117/6470 [29:42<58:39,  1.24it/s]
1|2118|Loss: 4.516308784484863:  33%|███▎      | 2118/6470 [29:43<1:01:04,  1.19it/s]
1|2119|Loss: 4.614883899688721:  33%|███▎      | 2118/6470 [29:43<1:01:04,  1.19it/s]
1|2119|Loss: 4.614883899688721:  33%|███▎      | 2119/6470 [29:43<1:00:15,  1.20it/s]
1|2120|Loss: 4.597227096557617:  33%|███▎      | 2119/6470 [29:44<1:00:15,  1.20it/s]
1|2120|Loss: 4.597227096557617:  33%|███▎      | 2120/6470 [29:44<57:02,  1.27it/s]  
1|2121|Loss: 4.241832733154297:  33%|███▎      | 2120/6470 [29:45<57:02,  1.27it/s]
1|2121|Loss: 4.241832733154297:  33%|███▎      | 2121/6470 [29:45<54:59,  1.32it/s]
1|2122|Loss: 5.00538969039917:  33%|███▎      | 2121/6470 [29:45<54:59,  1.32it/s] 
1|2122|Loss: 5.00538969039917:  33%|███▎      | 2122/6470 [29:46<55:45,  1.30it/s]
1|2123|Loss: 3.8084239959716797:  33%|███▎      | 2122/6470 [29:46<55:45,  1.30it/s]
1|2123|Loss: 3.8084239959716797:  33%|███▎      | 2123/6470 [29:46<49:21,  1.47it/s]
1|2124|Loss: 4.278555870056152:  33%|███▎      | 2123/6470 [29:46<49:21,  1.47it/s] 
1|2124|Loss: 4.278555870056152:  33%|███▎      | 2124/6470 [29:47<47:09,  1.54it/s]
1|2125|Loss: 4.237179279327393:  33%|███▎      | 2124/6470 [29:47<47:09,  1.54it/s]
1|2125|Loss: 4.237179279327393:  33%|███▎      | 2125/6470 [29:47<46:31,  1.56it/s]
1|2126|Loss: 4.704911708831787:  33%|███▎      | 2125/6470 [29:48<46:31,  1.56it/s]
1|2126|Loss: 4.704911708831787:  33%|███▎      | 2126/6470 [29:48<50:37,  1.43it/s]
1|2127|Loss: 4.535848617553711:  33%|███▎      | 2126/6470 [29:49<50:37,  1.43it/s]
1|2127|Loss: 4.535848617553711:  33%|███▎      | 2127/6470 [29:49<49:30,  1.46it/s]
1|2128|Loss: 4.165378093719482:  33%|███▎      | 2127/6470 [29:49<49:30,  1.46it/s]
1|2128|Loss: 4.165378093719482:  33%|███▎      | 2128/6470 [29:49<47:51,  1.51it/s]
1|2129|Loss: 4.359060287475586:  33%|███▎      | 2128/6470 [29:50<47:51,  1.51it/s]
1|2129|Loss: 4.359060287475586:  33%|███▎      | 2129/6470 [29:50<48:26,  1.49it/s]
1|2130|Loss: 4.950026512145996:  33%|███▎      | 2129/6470 [29:51<48:26,  1.49it/s]
1|2130|Loss: 4.950026512145996:  33%|███▎      | 2130/6470 [29:51<53:47,  1.34it/s]
1|2131|Loss: 4.7708635330200195:  33%|███▎      | 2130/6470 [29:52<53:47,  1.34it/s]
1|2131|Loss: 4.7708635330200195:  33%|███▎      | 2131/6470 [29:52<57:19,  1.26it/s]
1|2132|Loss: 4.662044048309326:  33%|███▎      | 2131/6470 [29:52<57:19,  1.26it/s] 
1|2132|Loss: 4.662044048309326:  33%|███▎      | 2132/6470 [29:53<55:44,  1.30it/s]
1|2133|Loss: 4.263569355010986:  33%|███▎      | 2132/6470 [29:53<55:44,  1.30it/s]
1|2133|Loss: 4.263569355010986:  33%|███▎      | 2133/6470 [29:53<52:56,  1.37it/s]
1|2134|Loss: 4.667465686798096:  33%|███▎      | 2133/6470 [29:54<52:56,  1.37it/s]
1|2134|Loss: 4.667465686798096:  33%|███▎      | 2134/6470 [29:54<52:04,  1.39it/s]
1|2135|Loss: 4.839487075805664:  33%|███▎      | 2134/6470 [29:54<52:04,  1.39it/s]
1|2135|Loss: 4.839487075805664:  33%|███▎      | 2135/6470 [29:55<55:53,  1.29it/s]
1|2136|Loss: 4.640219688415527:  33%|███▎      | 2135/6470 [29:55<55:53,  1.29it/s]
1|2136|Loss: 4.640219688415527:  33%|███▎      | 2136/6470 [29:56<57:06,  1.26it/s]
1|2137|Loss: 4.564962387084961:  33%|███▎      | 2136/6470 [29:56<57:06,  1.26it/s]
1|2137|Loss: 4.564962387084961:  33%|███▎      | 2137/6470 [29:56<57:33,  1.25it/s]
1|2138|Loss: 4.9159626960754395:  33%|███▎      | 2137/6470 [29:57<57:33,  1.25it/s]
1|2138|Loss: 4.9159626960754395:  33%|███▎      | 2138/6470 [29:57<1:00:36,  1.19it/s]
1|2139|Loss: 4.2512712478637695:  33%|███▎      | 2138/6470 [29:58<1:00:36,  1.19it/s]
1|2139|Loss: 4.2512712478637695:  33%|███▎      | 2139/6470 [29:58<57:41,  1.25it/s]  
1|2140|Loss: 5.023204803466797:  33%|███▎      | 2139/6470 [29:59<57:41,  1.25it/s] 
1|2140|Loss: 5.023204803466797:  33%|███▎      | 2140/6470 [29:59<59:52,  1.21it/s]
1|2141|Loss: 5.125802516937256:  33%|███▎      | 2140/6470 [30:00<59:52,  1.21it/s]
1|2141|Loss: 5.125802516937256:  33%|███▎      | 2141/6470 [30:00<1:01:52,  1.17it/s]
1|2142|Loss: 4.269064426422119:  33%|███▎      | 2141/6470 [30:00<1:01:52,  1.17it/s]
1|2142|Loss: 4.269064426422119:  33%|███▎      | 2142/6470 [30:01<58:44,  1.23it/s]  
1|2143|Loss: 4.779566287994385:  33%|███▎      | 2142/6470 [30:01<58:44,  1.23it/s]
1|2143|Loss: 4.779566287994385:  33%|███▎      | 2143/6470 [30:01<56:23,  1.28it/s]
1|2144|Loss: 4.236965656280518:  33%|███▎      | 2143/6470 [30:02<56:23,  1.28it/s]
1|2144|Loss: 4.236965656280518:  33%|███▎      | 2144/6470 [30:02<56:17,  1.28it/s]
1|2145|Loss: 4.369197368621826:  33%|███▎      | 2144/6470 [30:03<56:17,  1.28it/s]
1|2145|Loss: 4.369197368621826:  33%|███▎      | 2145/6470 [30:03<55:10,  1.31it/s]
1|2146|Loss: 4.394051551818848:  33%|███▎      | 2145/6470 [30:03<55:10,  1.31it/s]
1|2146|Loss: 4.394051551818848:  33%|███▎      | 2146/6470 [30:04<54:58,  1.31it/s]
1|2147|Loss: 4.290215969085693:  33%|███▎      | 2146/6470 [30:04<54:58,  1.31it/s]
1|2147|Loss: 4.290215969085693:  33%|███▎      | 2147/6470 [30:04<53:49,  1.34it/s]
1|2148|Loss: 4.776644229888916:  33%|███▎      | 2147/6470 [30:05<53:49,  1.34it/s]
1|2148|Loss: 4.776644229888916:  33%|███▎      | 2148/6470 [30:05<57:16,  1.26it/s]
1|2149|Loss: 4.3853583335876465:  33%|███▎      | 2148/6470 [30:06<57:16,  1.26it/s]
1|2149|Loss: 4.3853583335876465:  33%|███▎      | 2149/6470 [30:06<58:19,  1.23it/s]
1|2150|Loss: 4.25027322769165:  33%|███▎      | 2149/6470 [30:07<58:19,  1.23it/s]  
1|2150|Loss: 4.25027322769165:  33%|███▎      | 2150/6470 [30:07<56:11,  1.28it/s]
1|2151|Loss: 3.312127113342285:  33%|███▎      | 2150/6470 [30:07<56:11,  1.28it/s]
1|2151|Loss: 3.312127113342285:  33%|███▎      | 2151/6470 [30:07<48:49,  1.47it/s]
1|2152|Loss: 4.051576614379883:  33%|███▎      | 2151/6470 [30:08<48:49,  1.47it/s]
1|2152|Loss: 4.051576614379883:  33%|███▎      | 2152/6470 [30:08<47:28,  1.52it/s]
1|2153|Loss: 4.030419826507568:  33%|███▎      | 2152/6470 [30:08<47:28,  1.52it/s]
1|2153|Loss: 4.030419826507568:  33%|███▎      | 2153/6470 [30:08<44:53,  1.60it/s]
1|2154|Loss: 4.606896877288818:  33%|███▎      | 2153/6470 [30:09<44:53,  1.60it/s]
1|2154|Loss: 4.606896877288818:  33%|███▎      | 2154/6470 [30:09<46:29,  1.55it/s]
1|2155|Loss: 4.268981456756592:  33%|███▎      | 2154/6470 [30:09<46:29,  1.55it/s]
1|2155|Loss: 4.268981456756592:  33%|███▎      | 2155/6470 [30:10<44:06,  1.63it/s]
1|2156|Loss: 4.218482494354248:  33%|███▎      | 2155/6470 [30:10<44:06,  1.63it/s]
1|2156|Loss: 4.218482494354248:  33%|███▎      | 2156/6470 [30:10<48:35,  1.48it/s]
1|2157|Loss: 4.734086990356445:  33%|███▎      | 2156/6470 [30:11<48:35,  1.48it/s]
1|2157|Loss: 4.734086990356445:  33%|███▎      | 2157/6470 [30:11<50:43,  1.42it/s]
1|2158|Loss: 5.0577826499938965:  33%|███▎      | 2157/6470 [30:12<50:43,  1.42it/s]
1|2158|Loss: 5.0577826499938965:  33%|███▎      | 2158/6470 [30:12<53:14,  1.35it/s]
1|2159|Loss: 4.919459819793701:  33%|███▎      | 2158/6470 [30:13<53:14,  1.35it/s] 
1|2159|Loss: 4.919459819793701:  33%|███▎      | 2159/6470 [30:13<55:07,  1.30it/s]
1|2160|Loss: 4.6845221519470215:  33%|███▎      | 2159/6470 [30:13<55:07,  1.30it/s]
1|2160|Loss: 4.6845221519470215:  33%|███▎      | 2160/6470 [30:14<52:43,  1.36it/s]
1|2161|Loss: 3.9094650745391846:  33%|███▎      | 2160/6470 [30:14<52:43,  1.36it/s]
1|2161|Loss: 3.9094650745391846:  33%|███▎      | 2161/6470 [30:14<53:00,  1.35it/s]
1|2162|Loss: 4.86754035949707:  33%|███▎      | 2161/6470 [30:15<53:00,  1.35it/s]  
1|2162|Loss: 4.86754035949707:  33%|███▎      | 2162/6470 [30:15<54:26,  1.32it/s]
1|2163|Loss: 4.998618125915527:  33%|███▎      | 2162/6470 [30:16<54:26,  1.32it/s]
1|2163|Loss: 4.998618125915527:  33%|███▎      | 2163/6470 [30:16<56:50,  1.26it/s]
1|2164|Loss: 4.604111194610596:  33%|███▎      | 2163/6470 [30:16<56:50,  1.26it/s]
1|2164|Loss: 4.604111194610596:  33%|███▎      | 2164/6470 [30:17<55:18,  1.30it/s]
1|2165|Loss: 4.196420192718506:  33%|███▎      | 2164/6470 [30:17<55:18,  1.30it/s]
1|2165|Loss: 4.196420192718506:  33%|███▎      | 2165/6470 [30:18<1:01:21,  1.17it/s]
1|2166|Loss: 6.131997585296631:  33%|███▎      | 2165/6470 [30:18<1:01:21,  1.17it/s]
1|2166|Loss: 6.131997585296631:  33%|███▎      | 2166/6470 [30:19<1:01:52,  1.16it/s]
1|2167|Loss: 5.873985290527344:  33%|███▎      | 2166/6470 [30:19<1:01:52,  1.16it/s]
1|2167|Loss: 5.873985290527344:  33%|███▎      | 2167/6470 [30:19<55:03,  1.30it/s]  
1|2168|Loss: 5.331331253051758:  33%|███▎      | 2167/6470 [30:20<55:03,  1.30it/s]
1|2168|Loss: 5.331331253051758:  34%|███▎      | 2168/6470 [30:20<51:54,  1.38it/s]
1|2169|Loss: 5.162450313568115:  34%|███▎      | 2168/6470 [30:20<51:54,  1.38it/s]
1|2169|Loss: 5.162450313568115:  34%|███▎      | 2169/6470 [30:20<49:31,  1.45it/s]
1|2170|Loss: 5.112900257110596:  34%|███▎      | 2169/6470 [30:21<49:31,  1.45it/s]
1|2170|Loss: 5.112900257110596:  34%|███▎      | 2170/6470 [30:21<49:25,  1.45it/s]
1|2171|Loss: 4.382884502410889:  34%|███▎      | 2170/6470 [30:21<49:25,  1.45it/s]
1|2171|Loss: 4.382884502410889:  34%|███▎      | 2171/6470 [30:22<48:19,  1.48it/s]
1|2172|Loss: 4.865356922149658:  34%|███▎      | 2171/6470 [30:22<48:19,  1.48it/s]
1|2172|Loss: 4.865356922149658:  34%|███▎      | 2172/6470 [30:22<47:09,  1.52it/s]
1|2173|Loss: 5.262212753295898:  34%|███▎      | 2172/6470 [30:23<47:09,  1.52it/s]
1|2173|Loss: 5.262212753295898:  34%|███▎      | 2173/6470 [30:23<51:40,  1.39it/s]
1|2174|Loss: 4.847404479980469:  34%|███▎      | 2173/6470 [30:24<51:40,  1.39it/s]
1|2174|Loss: 4.847404479980469:  34%|███▎      | 2174/6470 [30:24<52:09,  1.37it/s]
1|2175|Loss: 4.569468021392822:  34%|███▎      | 2174/6470 [30:24<52:09,  1.37it/s]
1|2175|Loss: 4.569468021392822:  34%|███▎      | 2175/6470 [30:25<53:24,  1.34it/s]
1|2176|Loss: 5.016727924346924:  34%|███▎      | 2175/6470 [30:25<53:24,  1.34it/s]
1|2176|Loss: 5.016727924346924:  34%|███▎      | 2176/6470 [30:26<57:03,  1.25it/s]
1|2177|Loss: 4.087893962860107:  34%|███▎      | 2176/6470 [30:26<57:03,  1.25it/s]
1|2177|Loss: 4.087893962860107:  34%|███▎      | 2177/6470 [30:26<56:06,  1.28it/s]
1|2178|Loss: 4.044397354125977:  34%|███▎      | 2177/6470 [30:27<56:06,  1.28it/s]
1|2178|Loss: 4.044397354125977:  34%|███▎      | 2178/6470 [30:27<50:25,  1.42it/s]
1|2179|Loss: 4.454919815063477:  34%|███▎      | 2178/6470 [30:27<50:25,  1.42it/s]
1|2179|Loss: 4.454919815063477:  34%|███▎      | 2179/6470 [30:28<50:06,  1.43it/s]
1|2180|Loss: 4.768804550170898:  34%|███▎      | 2179/6470 [30:28<50:06,  1.43it/s]
1|2180|Loss: 4.768804550170898:  34%|███▎      | 2180/6470 [30:28<50:45,  1.41it/s]
1|2181|Loss: 4.449602127075195:  34%|███▎      | 2180/6470 [30:29<50:45,  1.41it/s]
1|2181|Loss: 4.449602127075195:  34%|███▎      | 2181/6470 [30:29<52:29,  1.36it/s]
1|2182|Loss: 4.650279521942139:  34%|███▎      | 2181/6470 [30:30<52:29,  1.36it/s]
1|2182|Loss: 4.650279521942139:  34%|███▎      | 2182/6470 [30:30<54:03,  1.32it/s]
1|2183|Loss: 4.448374271392822:  34%|███▎      | 2182/6470 [30:30<54:03,  1.32it/s]
1|2183|Loss: 4.448374271392822:  34%|███▎      | 2183/6470 [30:31<52:31,  1.36it/s]
1|2184|Loss: 4.840047836303711:  34%|███▎      | 2183/6470 [30:31<52:31,  1.36it/s]
1|2184|Loss: 4.840047836303711:  34%|███▍      | 2184/6470 [30:31<51:13,  1.39it/s]
1|2185|Loss: 4.580573081970215:  34%|███▍      | 2184/6470 [30:32<51:13,  1.39it/s]
1|2185|Loss: 4.580573081970215:  34%|███▍      | 2185/6470 [30:32<52:33,  1.36it/s]
1|2186|Loss: 4.539567947387695:  34%|███▍      | 2185/6470 [30:33<52:33,  1.36it/s]
1|2186|Loss: 4.539567947387695:  34%|███▍      | 2186/6470 [30:33<54:24,  1.31it/s]
1|2187|Loss: 4.6630473136901855:  34%|███▍      | 2186/6470 [30:33<54:24,  1.31it/s]
1|2187|Loss: 4.6630473136901855:  34%|███▍      | 2187/6470 [30:34<55:14,  1.29it/s]
1|2188|Loss: 4.714292526245117:  34%|███▍      | 2187/6470 [30:34<55:14,  1.29it/s] 
1|2188|Loss: 4.714292526245117:  34%|███▍      | 2188/6470 [30:34<55:03,  1.30it/s]
1|2189|Loss: 4.372790813446045:  34%|███▍      | 2188/6470 [30:35<55:03,  1.30it/s]
1|2189|Loss: 4.372790813446045:  34%|███▍      | 2189/6470 [30:35<53:22,  1.34it/s]
1|2190|Loss: 3.6167328357696533:  34%|███▍      | 2189/6470 [30:36<53:22,  1.34it/s]
1|2190|Loss: 3.6167328357696533:  34%|███▍      | 2190/6470 [30:36<48:24,  1.47it/s]
1|2191|Loss: 4.683857440948486:  34%|███▍      | 2190/6470 [30:36<48:24,  1.47it/s] 
1|2191|Loss: 4.683857440948486:  34%|███▍      | 2191/6470 [30:36<49:40,  1.44it/s]
1|2192|Loss: 4.685129165649414:  34%|███▍      | 2191/6470 [30:37<49:40,  1.44it/s]
1|2192|Loss: 4.685129165649414:  34%|███▍      | 2192/6470 [30:37<49:51,  1.43it/s]
1|2193|Loss: 5.257781028747559:  34%|███▍      | 2192/6470 [30:38<49:51,  1.43it/s]
1|2193|Loss: 5.257781028747559:  34%|███▍      | 2193/6470 [30:38<54:41,  1.30it/s]
1|2194|Loss: 4.741687774658203:  34%|███▍      | 2193/6470 [30:39<54:41,  1.30it/s]
1|2194|Loss: 4.741687774658203:  34%|███▍      | 2194/6470 [30:39<56:17,  1.27it/s]
1|2195|Loss: 4.250307559967041:  34%|███▍      | 2194/6470 [30:39<56:17,  1.27it/s]
1|2195|Loss: 4.250307559967041:  34%|███▍      | 2195/6470 [30:40<54:58,  1.30it/s]
1|2196|Loss: 4.51971960067749:  34%|███▍      | 2195/6470 [30:40<54:58,  1.30it/s] 
1|2196|Loss: 4.51971960067749:  34%|███▍      | 2196/6470 [30:40<52:09,  1.37it/s]
1|2197|Loss: 4.690065383911133:  34%|███▍      | 2196/6470 [30:41<52:09,  1.37it/s]
1|2197|Loss: 4.690065383911133:  34%|███▍      | 2197/6470 [30:41<54:45,  1.30it/s]
1|2198|Loss: 4.161244869232178:  34%|███▍      | 2197/6470 [30:42<54:45,  1.30it/s]
1|2198|Loss: 4.161244869232178:  34%|███▍      | 2198/6470 [30:42<55:30,  1.28it/s]
1|2199|Loss: 4.401716232299805:  34%|███▍      | 2198/6470 [30:42<55:30,  1.28it/s]
1|2199|Loss: 4.401716232299805:  34%|███▍      | 2199/6470 [30:43<53:56,  1.32it/s]
1|2200|Loss: 4.527199745178223:  34%|███▍      | 2199/6470 [30:43<53:56,  1.32it/s]
1|2200|Loss: 4.527199745178223:  34%|███▍      | 2200/6470 [30:43<53:51,  1.32it/s]
1|2201|Loss: 4.137553691864014:  34%|███▍      | 2200/6470 [30:44<53:51,  1.32it/s]
1|2201|Loss: 4.137553691864014:  34%|███▍      | 2201/6470 [30:44<49:38,  1.43it/s]
1|2202|Loss: 4.61073112487793:  34%|███▍      | 2201/6470 [30:44<49:38,  1.43it/s] 
1|2202|Loss: 4.61073112487793:  34%|███▍      | 2202/6470 [30:45<48:26,  1.47it/s]
1|2203|Loss: 4.6022443771362305:  34%|███▍      | 2202/6470 [30:45<48:26,  1.47it/s]
1|2203|Loss: 4.6022443771362305:  34%|███▍      | 2203/6470 [30:45<50:01,  1.42it/s]
1|2204|Loss: 4.730962753295898:  34%|███▍      | 2203/6470 [30:46<50:01,  1.42it/s] 
1|2204|Loss: 4.730962753295898:  34%|███▍      | 2204/6470 [30:46<51:59,  1.37it/s]
1|2205|Loss: 4.625973224639893:  34%|███▍      | 2204/6470 [30:47<51:59,  1.37it/s]
1|2205|Loss: 4.625973224639893:  34%|███▍      | 2205/6470 [30:47<51:22,  1.38it/s]
1|2206|Loss: 5.096763610839844:  34%|███▍      | 2205/6470 [30:47<51:22,  1.38it/s]
1|2206|Loss: 5.096763610839844:  34%|███▍      | 2206/6470 [30:48<53:05,  1.34it/s]
1|2207|Loss: 4.1931939125061035:  34%|███▍      | 2206/6470 [30:48<53:05,  1.34it/s]
1|2207|Loss: 4.1931939125061035:  34%|███▍      | 2207/6470 [30:48<52:46,  1.35it/s]
1|2208|Loss: 4.5015997886657715:  34%|███▍      | 2207/6470 [30:49<52:46,  1.35it/s]
1|2208|Loss: 4.5015997886657715:  34%|███▍      | 2208/6470 [30:49<51:36,  1.38it/s]
1|2209|Loss: 4.334253787994385:  34%|███▍      | 2208/6470 [30:49<51:36,  1.38it/s] 
1|2209|Loss: 4.334253787994385:  34%|███▍      | 2209/6470 [30:50<48:11,  1.47it/s]
1|2210|Loss: 4.4592695236206055:  34%|███▍      | 2209/6470 [30:50<48:11,  1.47it/s]
1|2210|Loss: 4.4592695236206055:  34%|███▍      | 2210/6470 [30:50<51:01,  1.39it/s]
1|2211|Loss: 4.186504364013672:  34%|███▍      | 2210/6470 [30:51<51:01,  1.39it/s] 
1|2211|Loss: 4.186504364013672:  34%|███▍      | 2211/6470 [30:51<49:02,  1.45it/s]
1|2212|Loss: 4.37858772277832:  34%|███▍      | 2211/6470 [30:51<49:02,  1.45it/s] 
1|2212|Loss: 4.37858772277832:  34%|███▍      | 2212/6470 [30:52<50:34,  1.40it/s]
1|2213|Loss: 5.1433868408203125:  34%|███▍      | 2212/6470 [30:52<50:34,  1.40it/s]
1|2213|Loss: 5.1433868408203125:  34%|███▍      | 2213/6470 [30:53<54:27,  1.30it/s]
1|2214|Loss: 4.446285724639893:  34%|███▍      | 2213/6470 [30:53<54:27,  1.30it/s] 
1|2214|Loss: 4.446285724639893:  34%|███▍      | 2214/6470 [30:54<54:51,  1.29it/s]
1|2215|Loss: 4.653257369995117:  34%|███▍      | 2214/6470 [30:54<54:51,  1.29it/s]
1|2215|Loss: 4.653257369995117:  34%|███▍      | 2215/6470 [30:54<57:06,  1.24it/s]
1|2216|Loss: 4.475744247436523:  34%|███▍      | 2215/6470 [30:55<57:06,  1.24it/s]
1|2216|Loss: 4.475744247436523:  34%|███▍      | 2216/6470 [30:55<57:22,  1.24it/s]
1|2217|Loss: 4.792026996612549:  34%|███▍      | 2216/6470 [30:56<57:22,  1.24it/s]
1|2217|Loss: 4.792026996612549:  34%|███▍      | 2217/6470 [30:56<56:35,  1.25it/s]
1|2218|Loss: 4.907054424285889:  34%|███▍      | 2217/6470 [30:56<56:35,  1.25it/s]
1|2218|Loss: 4.907054424285889:  34%|███▍      | 2218/6470 [30:57<56:03,  1.26it/s]
1|2219|Loss: 4.743969440460205:  34%|███▍      | 2218/6470 [30:57<56:03,  1.26it/s]
1|2219|Loss: 4.743969440460205:  34%|███▍      | 2219/6470 [30:58<56:28,  1.25it/s]
1|2220|Loss: 4.766871452331543:  34%|███▍      | 2219/6470 [30:58<56:28,  1.25it/s]
1|2220|Loss: 4.766871452331543:  34%|███▍      | 2220/6470 [30:58<56:37,  1.25it/s]
1|2221|Loss: 4.272496223449707:  34%|███▍      | 2220/6470 [30:59<56:37,  1.25it/s]
1|2221|Loss: 4.272496223449707:  34%|███▍      | 2221/6470 [30:59<53:31,  1.32it/s]
1|2222|Loss: 4.322777271270752:  34%|███▍      | 2221/6470 [30:59<53:31,  1.32it/s]
1|2222|Loss: 4.322777271270752:  34%|███▍      | 2222/6470 [31:00<51:48,  1.37it/s]
1|2223|Loss: 4.954014778137207:  34%|███▍      | 2222/6470 [31:00<51:48,  1.37it/s]
1|2223|Loss: 4.954014778137207:  34%|███▍      | 2223/6470 [31:01<54:50,  1.29it/s]
1|2224|Loss: 5.018333911895752:  34%|███▍      | 2223/6470 [31:01<54:50,  1.29it/s]
1|2224|Loss: 5.018333911895752:  34%|███▍      | 2224/6470 [31:01<57:11,  1.24it/s]
1|2225|Loss: 4.50993013381958:  34%|███▍      | 2224/6470 [31:02<57:11,  1.24it/s] 
1|2225|Loss: 4.50993013381958:  34%|███▍      | 2225/6470 [31:02<54:38,  1.29it/s]
1|2226|Loss: 4.124431133270264:  34%|███▍      | 2225/6470 [31:03<54:38,  1.29it/s]
1|2226|Loss: 4.124431133270264:  34%|███▍      | 2226/6470 [31:03<51:21,  1.38it/s]
1|2227|Loss: 4.033305644989014:  34%|███▍      | 2226/6470 [31:03<51:21,  1.38it/s]
1|2227|Loss: 4.033305644989014:  34%|███▍      | 2227/6470 [31:03<48:40,  1.45it/s]
1|2228|Loss: 4.912182331085205:  34%|███▍      | 2227/6470 [31:04<48:40,  1.45it/s]
1|2228|Loss: 4.912182331085205:  34%|███▍      | 2228/6470 [31:04<55:02,  1.28it/s]
1|2229|Loss: 4.704214096069336:  34%|███▍      | 2228/6470 [31:05<55:02,  1.28it/s]
1|2229|Loss: 4.704214096069336:  34%|███▍      | 2229/6470 [31:05<56:43,  1.25it/s]
1|2230|Loss: 4.722804546356201:  34%|███▍      | 2229/6470 [31:06<56:43,  1.25it/s]
1|2230|Loss: 4.722804546356201:  34%|███▍      | 2230/6470 [31:06<52:47,  1.34it/s]
1|2231|Loss: 5.073528289794922:  34%|███▍      | 2230/6470 [31:06<52:47,  1.34it/s]
1|2231|Loss: 5.073528289794922:  34%|███▍      | 2231/6470 [31:07<52:59,  1.33it/s]
1|2232|Loss: 4.915069103240967:  34%|███▍      | 2231/6470 [31:07<52:59,  1.33it/s]
1|2232|Loss: 4.915069103240967:  34%|███▍      | 2232/6470 [31:07<55:12,  1.28it/s]
1|2233|Loss: 4.61982536315918:  34%|███▍      | 2232/6470 [31:08<55:12,  1.28it/s] 
1|2233|Loss: 4.61982536315918:  35%|███▍      | 2233/6470 [31:08<53:53,  1.31it/s]
1|2234|Loss: 5.087625026702881:  35%|███▍      | 2233/6470 [31:09<53:53,  1.31it/s]
1|2234|Loss: 5.087625026702881:  35%|███▍      | 2234/6470 [31:09<56:51,  1.24it/s]
1|2235|Loss: 4.930630207061768:  35%|███▍      | 2234/6470 [31:10<56:51,  1.24it/s]
1|2235|Loss: 4.930630207061768:  35%|███▍      | 2235/6470 [31:10<59:08,  1.19it/s]
1|2236|Loss: 4.479995250701904:  35%|███▍      | 2235/6470 [31:10<59:08,  1.19it/s]
1|2236|Loss: 4.479995250701904:  35%|███▍      | 2236/6470 [31:11<56:52,  1.24it/s]
1|2237|Loss: 3.931947708129883:  35%|███▍      | 2236/6470 [31:11<56:52,  1.24it/s]
1|2237|Loss: 3.931947708129883:  35%|███▍      | 2237/6470 [31:11<50:50,  1.39it/s]
1|2238|Loss: 4.238712310791016:  35%|███▍      | 2237/6470 [31:12<50:50,  1.39it/s]
1|2238|Loss: 4.238712310791016:  35%|███▍      | 2238/6470 [31:12<47:01,  1.50it/s]
1|2239|Loss: 4.539337158203125:  35%|███▍      | 2238/6470 [31:12<47:01,  1.50it/s]
1|2239|Loss: 4.539337158203125:  35%|███▍      | 2239/6470 [31:13<48:25,  1.46it/s]
1|2240|Loss: 4.199169635772705:  35%|███▍      | 2239/6470 [31:13<48:25,  1.46it/s]
1|2240|Loss: 4.199169635772705:  35%|███▍      | 2240/6470 [31:13<49:05,  1.44it/s]
1|2241|Loss: 4.570164680480957:  35%|███▍      | 2240/6470 [31:14<49:05,  1.44it/s]
1|2241|Loss: 4.570164680480957:  35%|███▍      | 2241/6470 [31:14<52:07,  1.35it/s]
1|2242|Loss: 4.2430291175842285:  35%|███▍      | 2241/6470 [31:15<52:07,  1.35it/s]
1|2242|Loss: 4.2430291175842285:  35%|███▍      | 2242/6470 [31:15<50:49,  1.39it/s]
1|2243|Loss: 4.059286117553711:  35%|███▍      | 2242/6470 [31:15<50:49,  1.39it/s] 
1|2243|Loss: 4.059286117553711:  35%|███▍      | 2243/6470 [31:15<49:22,  1.43it/s]
1|2244|Loss: 4.404403209686279:  35%|███▍      | 2243/6470 [31:16<49:22,  1.43it/s]
1|2244|Loss: 4.404403209686279:  35%|███▍      | 2244/6470 [31:16<54:22,  1.30it/s]
1|2245|Loss: 4.02859354019165:  35%|███▍      | 2244/6470 [31:17<54:22,  1.30it/s] 
1|2245|Loss: 4.02859354019165:  35%|███▍      | 2245/6470 [31:17<54:24,  1.29it/s]
1|2246|Loss: 4.753051280975342:  35%|███▍      | 2245/6470 [31:18<54:24,  1.29it/s]
1|2246|Loss: 4.753051280975342:  35%|███▍      | 2246/6470 [31:18<57:33,  1.22it/s]
1|2247|Loss: 4.225448131561279:  35%|███▍      | 2246/6470 [31:19<57:33,  1.22it/s]
1|2247|Loss: 4.225448131561279:  35%|███▍      | 2247/6470 [31:19<55:33,  1.27it/s]
1|2248|Loss: 4.619615077972412:  35%|███▍      | 2247/6470 [31:19<55:33,  1.27it/s]
1|2248|Loss: 4.619615077972412:  35%|███▍      | 2248/6470 [31:20<54:48,  1.28it/s]
1|2249|Loss: 4.929762840270996:  35%|███▍      | 2248/6470 [31:20<54:48,  1.28it/s]
1|2249|Loss: 4.929762840270996:  35%|███▍      | 2249/6470 [31:20<55:09,  1.28it/s]
1|2250|Loss: 4.970005035400391:  35%|███▍      | 2249/6470 [31:21<55:09,  1.28it/s]
1|2250|Loss: 4.970005035400391:  35%|███▍      | 2250/6470 [31:21<51:39,  1.36it/s]
1|2251|Loss: 4.161904335021973:  35%|███▍      | 2250/6470 [31:21<51:39,  1.36it/s]
1|2251|Loss: 4.161904335021973:  35%|███▍      | 2251/6470 [31:22<49:34,  1.42it/s]
1|2252|Loss: 4.516464710235596:  35%|███▍      | 2251/6470 [31:22<49:34,  1.42it/s]
1|2252|Loss: 4.516464710235596:  35%|███▍      | 2252/6470 [31:22<52:55,  1.33it/s]
1|2253|Loss: 4.7731218338012695:  35%|███▍      | 2252/6470 [31:23<52:55,  1.33it/s]
1|2253|Loss: 4.7731218338012695:  35%|███▍      | 2253/6470 [31:23<54:23,  1.29it/s]
1|2254|Loss: 4.889891624450684:  35%|███▍      | 2253/6470 [31:24<54:23,  1.29it/s] 
1|2254|Loss: 4.889891624450684:  35%|███▍      | 2254/6470 [31:24<54:37,  1.29it/s]
1|2255|Loss: 4.449739933013916:  35%|███▍      | 2254/6470 [31:24<54:37,  1.29it/s]
1|2255|Loss: 4.449739933013916:  35%|███▍      | 2255/6470 [31:25<54:54,  1.28it/s]
1|2256|Loss: 4.005634784698486:  35%|███▍      | 2255/6470 [31:25<54:54,  1.28it/s]
1|2256|Loss: 4.005634784698486:  35%|███▍      | 2256/6470 [31:25<51:00,  1.38it/s]
1|2257|Loss: 4.151175498962402:  35%|███▍      | 2256/6470 [31:26<51:00,  1.38it/s]
1|2257|Loss: 4.151175498962402:  35%|███▍      | 2257/6470 [31:26<48:23,  1.45it/s]
1|2258|Loss: 3.2579097747802734:  35%|███▍      | 2257/6470 [31:26<48:23,  1.45it/s]
1|2258|Loss: 3.2579097747802734:  35%|███▍      | 2258/6470 [31:27<44:36,  1.57it/s]
1|2259|Loss: 4.43716287612915:  35%|███▍      | 2258/6470 [31:27<44:36,  1.57it/s]  
1|2259|Loss: 4.43716287612915:  35%|███▍      | 2259/6470 [31:27<44:34,  1.57it/s]
1|2260|Loss: 4.734658718109131:  35%|███▍      | 2259/6470 [31:28<44:34,  1.57it/s]
1|2260|Loss: 4.734658718109131:  35%|███▍      | 2260/6470 [31:28<47:01,  1.49it/s]
1|2261|Loss: 4.802090167999268:  35%|███▍      | 2260/6470 [31:28<47:01,  1.49it/s]
1|2261|Loss: 4.802090167999268:  35%|███▍      | 2261/6470 [31:29<49:12,  1.43it/s]
1|2262|Loss: 4.377850532531738:  35%|███▍      | 2261/6470 [31:29<49:12,  1.43it/s]
1|2262|Loss: 4.377850532531738:  35%|███▍      | 2262/6470 [31:30<52:20,  1.34it/s]
1|2263|Loss: 4.291651725769043:  35%|███▍      | 2262/6470 [31:30<52:20,  1.34it/s]
1|2263|Loss: 4.291651725769043:  35%|███▍      | 2263/6470 [31:30<49:31,  1.42it/s]
1|2264|Loss: 4.783527851104736:  35%|███▍      | 2263/6470 [31:31<49:31,  1.42it/s]
1|2264|Loss: 4.783527851104736:  35%|███▍      | 2264/6470 [31:31<52:21,  1.34it/s]
1|2265|Loss: 4.83928108215332:  35%|███▍      | 2264/6470 [31:32<52:21,  1.34it/s] 
1|2265|Loss: 4.83928108215332:  35%|███▌      | 2265/6470 [31:32<56:47,  1.23it/s]
1|2266|Loss: 4.29826021194458:  35%|███▌      | 2265/6470 [31:32<56:47,  1.23it/s]
1|2266|Loss: 4.29826021194458:  35%|███▌      | 2266/6470 [31:33<55:16,  1.27it/s]
1|2267|Loss: 4.418262958526611:  35%|███▌      | 2266/6470 [31:33<55:16,  1.27it/s]
1|2267|Loss: 4.418262958526611:  35%|███▌      | 2267/6470 [31:33<50:40,  1.38it/s]
1|2268|Loss: 3.6808762550354004:  35%|███▌      | 2267/6470 [31:34<50:40,  1.38it/s]
1|2268|Loss: 3.6808762550354004:  35%|███▌      | 2268/6470 [31:34<49:31,  1.41it/s]
1|2269|Loss: 4.842172145843506:  35%|███▌      | 2268/6470 [31:34<49:31,  1.41it/s] 
1|2269|Loss: 4.842172145843506:  35%|███▌      | 2269/6470 [31:35<52:33,  1.33it/s]
1|2270|Loss: 4.2928361892700195:  35%|███▌      | 2269/6470 [31:35<52:33,  1.33it/s]
1|2270|Loss: 4.2928361892700195:  35%|███▌      | 2270/6470 [31:36<50:57,  1.37it/s]
1|2271|Loss: 4.769676208496094:  35%|███▌      | 2270/6470 [31:36<50:57,  1.37it/s] 
1|2271|Loss: 4.769676208496094:  35%|███▌      | 2271/6470 [31:36<50:04,  1.40it/s]
1|2272|Loss: 4.171824932098389:  35%|███▌      | 2271/6470 [31:37<50:04,  1.40it/s]
1|2272|Loss: 4.171824932098389:  35%|███▌      | 2272/6470 [31:37<50:06,  1.40it/s]
1|2273|Loss: 4.6791276931762695:  35%|███▌      | 2272/6470 [31:37<50:06,  1.40it/s]
1|2273|Loss: 4.6791276931762695:  35%|███▌      | 2273/6470 [31:38<50:52,  1.38it/s]
1|2274|Loss: 4.248213768005371:  35%|███▌      | 2273/6470 [31:38<50:52,  1.38it/s] 
1|2274|Loss: 4.248213768005371:  35%|███▌      | 2274/6470 [31:38<51:36,  1.36it/s]
1|2275|Loss: 4.901913166046143:  35%|███▌      | 2274/6470 [31:39<51:36,  1.36it/s]
1|2275|Loss: 4.901913166046143:  35%|███▌      | 2275/6470 [31:39<54:43,  1.28it/s]
1|2276|Loss: 4.611388683319092:  35%|███▌      | 2275/6470 [31:40<54:43,  1.28it/s]
1|2276|Loss: 4.611388683319092:  35%|███▌      | 2276/6470 [31:40<57:22,  1.22it/s]
1|2277|Loss: 4.687268257141113:  35%|███▌      | 2276/6470 [31:41<57:22,  1.22it/s]
1|2277|Loss: 4.687268257141113:  35%|███▌      | 2277/6470 [31:41<59:32,  1.17it/s]
1|2278|Loss: 4.779814720153809:  35%|███▌      | 2277/6470 [31:42<59:32,  1.17it/s]
1|2278|Loss: 4.779814720153809:  35%|███▌      | 2278/6470 [31:42<57:38,  1.21it/s]
1|2279|Loss: 4.0663533210754395:  35%|███▌      | 2278/6470 [31:42<57:38,  1.21it/s]
1|2279|Loss: 4.0663533210754395:  35%|███▌      | 2279/6470 [31:43<56:58,  1.23it/s]
1|2280|Loss: 4.479558944702148:  35%|███▌      | 2279/6470 [31:43<56:58,  1.23it/s] 
1|2280|Loss: 4.479558944702148:  35%|███▌      | 2280/6470 [31:44<57:10,  1.22it/s]
1|2281|Loss: 4.38119649887085:  35%|███▌      | 2280/6470 [31:44<57:10,  1.22it/s] 
1|2281|Loss: 4.38119649887085:  35%|███▌      | 2281/6470 [31:44<54:53,  1.27it/s]
1|2282|Loss: 4.397964000701904:  35%|███▌      | 2281/6470 [31:45<54:53,  1.27it/s]
1|2282|Loss: 4.397964000701904:  35%|███▌      | 2282/6470 [31:45<54:02,  1.29it/s]
1|2283|Loss: 4.381885051727295:  35%|███▌      | 2282/6470 [31:45<54:02,  1.29it/s]
1|2283|Loss: 4.381885051727295:  35%|███▌      | 2283/6470 [31:46<55:57,  1.25it/s]
1|2284|Loss: 4.955790996551514:  35%|███▌      | 2283/6470 [31:46<55:57,  1.25it/s]
1|2284|Loss: 4.955790996551514:  35%|███▌      | 2284/6470 [31:47<54:39,  1.28it/s]
1|2285|Loss: 4.4754228591918945:  35%|███▌      | 2284/6470 [31:47<54:39,  1.28it/s]
1|2285|Loss: 4.4754228591918945:  35%|███▌      | 2285/6470 [31:47<53:59,  1.29it/s]
1|2286|Loss: 4.223106861114502:  35%|███▌      | 2285/6470 [31:48<53:59,  1.29it/s] 
1|2286|Loss: 4.223106861114502:  35%|███▌      | 2286/6470 [31:48<51:34,  1.35it/s]
1|2287|Loss: 4.381472110748291:  35%|███▌      | 2286/6470 [31:48<51:34,  1.35it/s]
1|2287|Loss: 4.381472110748291:  35%|███▌      | 2287/6470 [31:49<48:31,  1.44it/s]
1|2288|Loss: 4.407419681549072:  35%|███▌      | 2287/6470 [31:49<48:31,  1.44it/s]
1|2288|Loss: 4.407419681549072:  35%|███▌      | 2288/6470 [31:49<50:08,  1.39it/s]
1|2289|Loss: 4.1650543212890625:  35%|███▌      | 2288/6470 [31:50<50:08,  1.39it/s]
1|2289|Loss: 4.1650543212890625:  35%|███▌      | 2289/6470 [31:50<53:46,  1.30it/s]
1|2290|Loss: 3.949906587600708:  35%|███▌      | 2289/6470 [31:51<53:46,  1.30it/s] 
1|2290|Loss: 3.949906587600708:  35%|███▌      | 2290/6470 [31:51<52:30,  1.33it/s]
1|2291|Loss: 4.299445152282715:  35%|███▌      | 2290/6470 [31:51<52:30,  1.33it/s]
1|2291|Loss: 4.299445152282715:  35%|███▌      | 2291/6470 [31:52<54:06,  1.29it/s]
1|2292|Loss: 4.1019368171691895:  35%|███▌      | 2291/6470 [31:52<54:06,  1.29it/s]
1|2292|Loss: 4.1019368171691895:  35%|███▌      | 2292/6470 [31:52<52:07,  1.34it/s]
1|2293|Loss: 4.506968975067139:  35%|███▌      | 2292/6470 [31:53<52:07,  1.34it/s] 
1|2293|Loss: 4.506968975067139:  35%|███▌      | 2293/6470 [31:53<52:34,  1.32it/s]
1|2294|Loss: 4.193725109100342:  35%|███▌      | 2293/6470 [31:54<52:34,  1.32it/s]
1|2294|Loss: 4.193725109100342:  35%|███▌      | 2294/6470 [31:54<51:52,  1.34it/s]
1|2295|Loss: 4.176212787628174:  35%|███▌      | 2294/6470 [31:54<51:52,  1.34it/s]
1|2295|Loss: 4.176212787628174:  35%|███▌      | 2295/6470 [31:55<50:42,  1.37it/s]
1|2296|Loss: 4.107534885406494:  35%|███▌      | 2295/6470 [31:55<50:42,  1.37it/s]
1|2296|Loss: 4.107534885406494:  35%|███▌      | 2296/6470 [31:55<48:56,  1.42it/s]
1|2297|Loss: 4.662285327911377:  35%|███▌      | 2296/6470 [31:56<48:56,  1.42it/s]
1|2297|Loss: 4.662285327911377:  36%|███▌      | 2297/6470 [31:56<50:11,  1.39it/s]
1|2298|Loss: 4.707197666168213:  36%|███▌      | 2297/6470 [31:57<50:11,  1.39it/s]
1|2298|Loss: 4.707197666168213:  36%|███▌      | 2298/6470 [31:57<54:46,  1.27it/s]
1|2299|Loss: 4.550867080688477:  36%|███▌      | 2298/6470 [31:58<54:46,  1.27it/s]
1|2299|Loss: 4.550867080688477:  36%|███▌      | 2299/6470 [31:58<58:49,  1.18it/s]
1|2300|Loss: 4.817346572875977:  36%|███▌      | 2299/6470 [31:59<58:49,  1.18it/s]
1|2300|Loss: 4.817346572875977:  36%|███▌      | 2300/6470 [31:59<1:00:05,  1.16it/s]
1|2301|Loss: 4.689504146575928:  36%|███▌      | 2300/6470 [31:59<1:00:05,  1.16it/s]
1|2301|Loss: 4.689504146575928:  36%|███▌      | 2301/6470 [32:00<59:19,  1.17it/s]  
1|2302|Loss: 4.090219974517822:  36%|███▌      | 2301/6470 [32:00<59:19,  1.17it/s]
1|2302|Loss: 4.090219974517822:  36%|███▌      | 2302/6470 [32:00<54:49,  1.27it/s]
1|2303|Loss: 4.078366279602051:  36%|███▌      | 2302/6470 [32:01<54:49,  1.27it/s]
1|2303|Loss: 4.078366279602051:  36%|███▌      | 2303/6470 [32:01<50:32,  1.37it/s]
1|2304|Loss: 4.634000778198242:  36%|███▌      | 2303/6470 [32:01<50:32,  1.37it/s]
1|2304|Loss: 4.634000778198242:  36%|███▌      | 2304/6470 [32:02<51:12,  1.36it/s]
1|2305|Loss: 4.262722015380859:  36%|███▌      | 2304/6470 [32:02<51:12,  1.36it/s]
1|2305|Loss: 4.262722015380859:  36%|███▌      | 2305/6470 [32:02<49:38,  1.40it/s]
1|2306|Loss: 4.495357990264893:  36%|███▌      | 2305/6470 [32:03<49:38,  1.40it/s]
1|2306|Loss: 4.495357990264893:  36%|███▌      | 2306/6470 [32:03<48:52,  1.42it/s]
1|2307|Loss: 5.037970066070557:  36%|███▌      | 2306/6470 [32:04<48:52,  1.42it/s]
1|2307|Loss: 5.037970066070557:  36%|███▌      | 2307/6470 [32:04<52:26,  1.32it/s]
1|2308|Loss: 4.335791110992432:  36%|███▌      | 2307/6470 [32:04<52:26,  1.32it/s]
1|2308|Loss: 4.335791110992432:  36%|███▌      | 2308/6470 [32:05<51:49,  1.34it/s]
1|2309|Loss: 3.9647181034088135:  36%|███▌      | 2308/6470 [32:05<51:49,  1.34it/s]
1|2309|Loss: 3.9647181034088135:  36%|███▌      | 2309/6470 [32:05<50:19,  1.38it/s]
1|2310|Loss: 4.266229152679443:  36%|███▌      | 2309/6470 [32:06<50:19,  1.38it/s] 
1|2310|Loss: 4.266229152679443:  36%|███▌      | 2310/6470 [32:06<48:25,  1.43it/s]
1|2311|Loss: 4.395913124084473:  36%|███▌      | 2310/6470 [32:06<48:25,  1.43it/s]
1|2311|Loss: 4.395913124084473:  36%|███▌      | 2311/6470 [32:07<49:44,  1.39it/s]
1|2312|Loss: 4.6907148361206055:  36%|███▌      | 2311/6470 [32:07<49:44,  1.39it/s]
1|2312|Loss: 4.6907148361206055:  36%|███▌      | 2312/6470 [32:08<53:54,  1.29it/s]
1|2313|Loss: 4.0678324699401855:  36%|███▌      | 2312/6470 [32:08<53:54,  1.29it/s]
1|2313|Loss: 4.0678324699401855:  36%|███▌      | 2313/6470 [32:08<51:46,  1.34it/s]
1|2314|Loss: 3.4922449588775635:  36%|███▌      | 2313/6470 [32:09<51:46,  1.34it/s]
1|2314|Loss: 3.4922449588775635:  36%|███▌      | 2314/6470 [32:09<49:01,  1.41it/s]
1|2315|Loss: 4.376204013824463:  36%|███▌      | 2314/6470 [32:09<49:01,  1.41it/s] 
1|2315|Loss: 4.376204013824463:  36%|███▌      | 2315/6470 [32:10<49:49,  1.39it/s]
1|2316|Loss: 3.7443881034851074:  36%|███▌      | 2315/6470 [32:10<49:49,  1.39it/s]
1|2316|Loss: 3.7443881034851074:  36%|███▌      | 2316/6470 [32:10<47:08,  1.47it/s]
1|2317|Loss: 4.043197154998779:  36%|███▌      | 2316/6470 [32:11<47:08,  1.47it/s] 
1|2317|Loss: 4.043197154998779:  36%|███▌      | 2317/6470 [32:11<46:43,  1.48it/s]
1|2318|Loss: 4.385286808013916:  36%|███▌      | 2317/6470 [32:11<46:43,  1.48it/s]
1|2318|Loss: 4.385286808013916:  36%|███▌      | 2318/6470 [32:12<47:32,  1.46it/s]
1|2319|Loss: 4.4537153244018555:  36%|███▌      | 2318/6470 [32:12<47:32,  1.46it/s]
1|2319|Loss: 4.4537153244018555:  36%|███▌      | 2319/6470 [32:12<49:42,  1.39it/s]
1|2320|Loss: 4.419322490692139:  36%|███▌      | 2319/6470 [32:13<49:42,  1.39it/s] 
1|2320|Loss: 4.419322490692139:  36%|███▌      | 2320/6470 [32:13<50:58,  1.36it/s]
1|2321|Loss: 4.566569805145264:  36%|███▌      | 2320/6470 [32:14<50:58,  1.36it/s]
1|2321|Loss: 4.566569805145264:  36%|███▌      | 2321/6470 [32:14<51:25,  1.34it/s]
1|2322|Loss: 4.268582344055176:  36%|███▌      | 2321/6470 [32:14<51:25,  1.34it/s]
1|2322|Loss: 4.268582344055176:  36%|███▌      | 2322/6470 [32:15<52:20,  1.32it/s]
1|2323|Loss: 4.145210266113281:  36%|███▌      | 2322/6470 [32:15<52:20,  1.32it/s]
1|2323|Loss: 4.145210266113281:  36%|███▌      | 2323/6470 [32:16<53:01,  1.30it/s]
1|2324|Loss: 4.577236175537109:  36%|███▌      | 2323/6470 [32:16<53:01,  1.30it/s]
1|2324|Loss: 4.577236175537109:  36%|███▌      | 2324/6470 [32:16<53:37,  1.29it/s]
1|2325|Loss: 4.738955020904541:  36%|███▌      | 2324/6470 [32:17<53:37,  1.29it/s]
1|2325|Loss: 4.738955020904541:  36%|███▌      | 2325/6470 [32:17<53:27,  1.29it/s]
1|2326|Loss: 5.051583766937256:  36%|███▌      | 2325/6470 [32:18<53:27,  1.29it/s]
1|2326|Loss: 5.051583766937256:  36%|███▌      | 2326/6470 [32:18<56:31,  1.22it/s]
1|2327|Loss: 4.533127784729004:  36%|███▌      | 2326/6470 [32:19<56:31,  1.22it/s]
1|2327|Loss: 4.533127784729004:  36%|███▌      | 2327/6470 [32:19<59:24,  1.16it/s]
1|2328|Loss: 4.04428243637085:  36%|███▌      | 2327/6470 [32:20<59:24,  1.16it/s] 
1|2328|Loss: 4.04428243637085:  36%|███▌      | 2328/6470 [32:20<58:44,  1.18it/s]
1|2329|Loss: 5.075565814971924:  36%|███▌      | 2328/6470 [32:20<58:44,  1.18it/s]
1|2329|Loss: 5.075565814971924:  36%|███▌      | 2329/6470 [32:21<1:01:51,  1.12it/s]
1|2330|Loss: 4.474106311798096:  36%|███▌      | 2329/6470 [32:21<1:01:51,  1.12it/s]
1|2330|Loss: 4.474106311798096:  36%|███▌      | 2330/6470 [32:22<1:03:14,  1.09it/s]
1|2331|Loss: 4.434894561767578:  36%|███▌      | 2330/6470 [32:22<1:03:14,  1.09it/s]
1|2331|Loss: 4.434894561767578:  36%|███▌      | 2331/6470 [32:23<58:42,  1.18it/s]  
1|2332|Loss: 4.2303056716918945:  36%|███▌      | 2331/6470 [32:23<58:42,  1.18it/s]
1|2332|Loss: 4.2303056716918945:  36%|███▌      | 2332/6470 [32:23<56:02,  1.23it/s]
1|2333|Loss: 4.507936000823975:  36%|███▌      | 2332/6470 [32:24<56:02,  1.23it/s] 
1|2333|Loss: 4.507936000823975:  36%|███▌      | 2333/6470 [32:24<55:26,  1.24it/s]
1|2334|Loss: 4.559267997741699:  36%|███▌      | 2333/6470 [32:24<55:26,  1.24it/s]
1|2334|Loss: 4.559267997741699:  36%|███▌      | 2334/6470 [32:25<54:40,  1.26it/s]
1|2335|Loss: 4.126304626464844:  36%|███▌      | 2334/6470 [32:25<54:40,  1.26it/s]
1|2335|Loss: 4.126304626464844:  36%|███▌      | 2335/6470 [32:25<51:31,  1.34it/s]
1|2336|Loss: 4.707137107849121:  36%|███▌      | 2335/6470 [32:26<51:31,  1.34it/s]
1|2336|Loss: 4.707137107849121:  36%|███▌      | 2336/6470 [32:26<54:53,  1.26it/s]
1|2337|Loss: 4.40123176574707:  36%|███▌      | 2336/6470 [32:27<54:53,  1.26it/s] 
1|2337|Loss: 4.40123176574707:  36%|███▌      | 2337/6470 [32:27<54:39,  1.26it/s]
1|2338|Loss: 4.500685691833496:  36%|███▌      | 2337/6470 [32:28<54:39,  1.26it/s]
1|2338|Loss: 4.500685691833496:  36%|███▌      | 2338/6470 [32:28<56:27,  1.22it/s]
1|2339|Loss: 4.570767402648926:  36%|███▌      | 2338/6470 [32:28<56:27,  1.22it/s]
1|2339|Loss: 4.570767402648926:  36%|███▌      | 2339/6470 [32:29<57:46,  1.19it/s]
1|2340|Loss: 3.698014497756958:  36%|███▌      | 2339/6470 [32:29<57:46,  1.19it/s]
1|2340|Loss: 3.698014497756958:  36%|███▌      | 2340/6470 [32:30<54:34,  1.26it/s]
1|2341|Loss: 4.093430995941162:  36%|███▌      | 2340/6470 [32:30<54:34,  1.26it/s]
1|2341|Loss: 4.093430995941162:  36%|███▌      | 2341/6470 [32:30<53:14,  1.29it/s]
1|2342|Loss: 4.058367729187012:  36%|███▌      | 2341/6470 [32:31<53:14,  1.29it/s]
1|2342|Loss: 4.058367729187012:  36%|███▌      | 2342/6470 [32:31<59:18,  1.16it/s]
1|2343|Loss: 4.347651958465576:  36%|███▌      | 2342/6470 [32:32<59:18,  1.16it/s]
1|2343|Loss: 4.347651958465576:  36%|███▌      | 2343/6470 [32:32<55:37,  1.24it/s]
1|2344|Loss: 4.026272773742676:  36%|███▌      | 2343/6470 [32:32<55:37,  1.24it/s]
1|2344|Loss: 4.026272773742676:  36%|███▌      | 2344/6470 [32:33<52:45,  1.30it/s]
1|2345|Loss: 4.397124767303467:  36%|███▌      | 2344/6470 [32:33<52:45,  1.30it/s]
1|2345|Loss: 4.397124767303467:  36%|███▌      | 2345/6470 [32:33<51:06,  1.35it/s]
1|2346|Loss: 4.437066555023193:  36%|███▌      | 2345/6470 [32:34<51:06,  1.35it/s]
1|2346|Loss: 4.437066555023193:  36%|███▋      | 2346/6470 [32:34<53:53,  1.28it/s]
1|2347|Loss: 4.913909435272217:  36%|███▋      | 2346/6470 [32:35<53:53,  1.28it/s]
1|2347|Loss: 4.913909435272217:  36%|███▋      | 2347/6470 [32:35<54:50,  1.25it/s]
1|2348|Loss: 4.522950172424316:  36%|███▋      | 2347/6470 [32:36<54:50,  1.25it/s]
1|2348|Loss: 4.522950172424316:  36%|███▋      | 2348/6470 [32:36<57:02,  1.20it/s]
1|2349|Loss: 4.443497657775879:  36%|███▋      | 2348/6470 [32:37<57:02,  1.20it/s]
1|2349|Loss: 4.443497657775879:  36%|███▋      | 2349/6470 [32:37<56:58,  1.21it/s]
1|2350|Loss: 4.240196704864502:  36%|███▋      | 2349/6470 [32:37<56:58,  1.21it/s]
1|2350|Loss: 4.240196704864502:  36%|███▋      | 2350/6470 [32:38<53:04,  1.29it/s]
1|2351|Loss: 4.637772560119629:  36%|███▋      | 2350/6470 [32:38<53:04,  1.29it/s]
1|2351|Loss: 4.637772560119629:  36%|███▋      | 2351/6470 [32:38<51:18,  1.34it/s]
1|2352|Loss: 4.0616230964660645:  36%|███▋      | 2351/6470 [32:39<51:18,  1.34it/s]
1|2352|Loss: 4.0616230964660645:  36%|███▋      | 2352/6470 [32:39<48:08,  1.43it/s]
1|2353|Loss: 3.496950387954712:  36%|███▋      | 2352/6470 [32:39<48:08,  1.43it/s] 
1|2353|Loss: 3.496950387954712:  36%|███▋      | 2353/6470 [32:39<46:13,  1.48it/s]
1|2354|Loss: 4.4092936515808105:  36%|███▋      | 2353/6470 [32:40<46:13,  1.48it/s]
1|2354|Loss: 4.4092936515808105:  36%|███▋      | 2354/6470 [32:40<44:18,  1.55it/s]
1|2355|Loss: 3.342844247817993:  36%|███▋      | 2354/6470 [32:40<44:18,  1.55it/s] 
1|2355|Loss: 3.342844247817993:  36%|███▋      | 2355/6470 [32:40<41:12,  1.66it/s]
1|2356|Loss: 4.038485527038574:  36%|███▋      | 2355/6470 [32:41<41:12,  1.66it/s]
1|2356|Loss: 4.038485527038574:  36%|███▋      | 2356/6470 [32:41<40:57,  1.67it/s]
1|2357|Loss: 4.28773307800293:  36%|███▋      | 2356/6470 [32:41<40:57,  1.67it/s] 
1|2357|Loss: 4.28773307800293:  36%|███▋      | 2357/6470 [32:42<45:46,  1.50it/s]
1|2358|Loss: 4.506740570068359:  36%|███▋      | 2357/6470 [32:42<45:46,  1.50it/s]
1|2358|Loss: 4.506740570068359:  36%|███▋      | 2358/6470 [32:43<48:26,  1.41it/s]
1|2359|Loss: 4.758767604827881:  36%|███▋      | 2358/6470 [32:43<48:26,  1.41it/s]
1|2359|Loss: 4.758767604827881:  36%|███▋      | 2359/6470 [32:43<48:44,  1.41it/s]
1|2360|Loss: 4.443474292755127:  36%|███▋      | 2359/6470 [32:44<48:44,  1.41it/s]
1|2360|Loss: 4.443474292755127:  36%|███▋      | 2360/6470 [32:44<52:08,  1.31it/s]
1|2361|Loss: 4.210275173187256:  36%|███▋      | 2360/6470 [32:45<52:08,  1.31it/s]
1|2361|Loss: 4.210275173187256:  36%|███▋      | 2361/6470 [32:45<53:05,  1.29it/s]
1|2362|Loss: 4.199355602264404:  36%|███▋      | 2361/6470 [32:46<53:05,  1.29it/s]
1|2362|Loss: 4.199355602264404:  37%|███▋      | 2362/6470 [32:46<53:40,  1.28it/s]
1|2363|Loss: 4.722002029418945:  37%|███▋      | 2362/6470 [32:46<53:40,  1.28it/s]
1|2363|Loss: 4.722002029418945:  37%|███▋      | 2363/6470 [32:47<55:08,  1.24it/s]
1|2364|Loss: 4.141756057739258:  37%|███▋      | 2363/6470 [32:47<55:08,  1.24it/s]
1|2364|Loss: 4.141756057739258:  37%|███▋      | 2364/6470 [32:48<54:35,  1.25it/s]
1|2365|Loss: 4.676270484924316:  37%|███▋      | 2364/6470 [32:48<54:35,  1.25it/s]
1|2365|Loss: 4.676270484924316:  37%|███▋      | 2365/6470 [32:48<54:15,  1.26it/s]
1|2366|Loss: 4.458611965179443:  37%|███▋      | 2365/6470 [32:49<54:15,  1.26it/s]
1|2366|Loss: 4.458611965179443:  37%|███▋      | 2366/6470 [32:49<56:47,  1.20it/s]
1|2367|Loss: 4.211419105529785:  37%|███▋      | 2366/6470 [32:50<56:47,  1.20it/s]
1|2367|Loss: 4.211419105529785:  37%|███▋      | 2367/6470 [32:50<55:10,  1.24it/s]
1|2368|Loss: 4.256415367126465:  37%|███▋      | 2367/6470 [32:50<55:10,  1.24it/s]
1|2368|Loss: 4.256415367126465:  37%|███▋      | 2368/6470 [32:51<56:53,  1.20it/s]
1|2369|Loss: 4.044301986694336:  37%|███▋      | 2368/6470 [32:51<56:53,  1.20it/s]
1|2369|Loss: 4.044301986694336:  37%|███▋      | 2369/6470 [32:52<57:29,  1.19it/s]
1|2370|Loss: 4.366623401641846:  37%|███▋      | 2369/6470 [32:52<57:29,  1.19it/s]
1|2370|Loss: 4.366623401641846:  37%|███▋      | 2370/6470 [32:52<51:34,  1.32it/s]
1|2371|Loss: 4.458895206451416:  37%|███▋      | 2370/6470 [32:53<51:34,  1.32it/s]
1|2371|Loss: 4.458895206451416:  37%|███▋      | 2371/6470 [32:53<49:20,  1.38it/s]
1|2372|Loss: 4.2641282081604:  37%|███▋      | 2371/6470 [32:53<49:20,  1.38it/s]  
1|2372|Loss: 4.2641282081604:  37%|███▋      | 2372/6470 [32:54<47:27,  1.44it/s]
1|2373|Loss: 4.287390232086182:  37%|███▋      | 2372/6470 [32:54<47:27,  1.44it/s]
1|2373|Loss: 4.287390232086182:  37%|███▋      | 2373/6470 [32:54<46:46,  1.46it/s]
1|2374|Loss: 3.8604283332824707:  37%|███▋      | 2373/6470 [32:55<46:46,  1.46it/s]
1|2374|Loss: 3.8604283332824707:  37%|███▋      | 2374/6470 [32:55<46:21,  1.47it/s]
1|2375|Loss: 4.176723003387451:  37%|███▋      | 2374/6470 [32:55<46:21,  1.47it/s] 
1|2375|Loss: 4.176723003387451:  37%|███▋      | 2375/6470 [32:56<45:20,  1.51it/s]
1|2376|Loss: 3.943289041519165:  37%|███▋      | 2375/6470 [32:56<45:20,  1.51it/s]
1|2376|Loss: 3.943289041519165:  37%|███▋      | 2376/6470 [32:56<45:23,  1.50it/s]
1|2377|Loss: 4.820227146148682:  37%|███▋      | 2376/6470 [32:57<45:23,  1.50it/s]
1|2377|Loss: 4.820227146148682:  37%|███▋      | 2377/6470 [32:57<45:00,  1.52it/s]
1|2378|Loss: 4.010146141052246:  37%|███▋      | 2377/6470 [32:57<45:00,  1.52it/s]
1|2378|Loss: 4.010146141052246:  37%|███▋      | 2378/6470 [32:58<44:57,  1.52it/s]
1|2379|Loss: 4.679326057434082:  37%|███▋      | 2378/6470 [32:58<44:57,  1.52it/s]
1|2379|Loss: 4.679326057434082:  37%|███▋      | 2379/6470 [32:58<43:56,  1.55it/s]
1|2380|Loss: 4.310914516448975:  37%|███▋      | 2379/6470 [32:58<43:56,  1.55it/s]
1|2380|Loss: 4.310914516448975:  37%|███▋      | 2380/6470 [32:59<44:54,  1.52it/s]
1|2381|Loss: 4.348454475402832:  37%|███▋      | 2380/6470 [32:59<44:54,  1.52it/s]
1|2381|Loss: 4.348454475402832:  37%|███▋      | 2381/6470 [32:59<45:32,  1.50it/s]
1|2382|Loss: 4.981403350830078:  37%|███▋      | 2381/6470 [33:00<45:32,  1.50it/s]
1|2382|Loss: 4.981403350830078:  37%|███▋      | 2382/6470 [33:00<48:16,  1.41it/s]
1|2383|Loss: 4.459673881530762:  37%|███▋      | 2382/6470 [33:01<48:16,  1.41it/s]
1|2383|Loss: 4.459673881530762:  37%|███▋      | 2383/6470 [33:01<50:30,  1.35it/s]
1|2384|Loss: 3.995026111602783:  37%|███▋      | 2383/6470 [33:02<50:30,  1.35it/s]
1|2384|Loss: 3.995026111602783:  37%|███▋      | 2384/6470 [33:02<49:44,  1.37it/s]
1|2385|Loss: 4.494912147521973:  37%|███▋      | 2384/6470 [33:02<49:44,  1.37it/s]
1|2385|Loss: 4.494912147521973:  37%|███▋      | 2385/6470 [33:03<50:24,  1.35it/s]
1|2386|Loss: 4.266088485717773:  37%|███▋      | 2385/6470 [33:03<50:24,  1.35it/s]
1|2386|Loss: 4.266088485717773:  37%|███▋      | 2386/6470 [33:03<48:41,  1.40it/s]
1|2387|Loss: 4.421868801116943:  37%|███▋      | 2386/6470 [33:04<48:41,  1.40it/s]
1|2387|Loss: 4.421868801116943:  37%|███▋      | 2387/6470 [33:04<49:12,  1.38it/s]
1|2388|Loss: 4.421457290649414:  37%|███▋      | 2387/6470 [33:04<49:12,  1.38it/s]
1|2388|Loss: 4.421457290649414:  37%|███▋      | 2388/6470 [33:05<47:46,  1.42it/s]
1|2389|Loss: 3.446035861968994:  37%|███▋      | 2388/6470 [33:05<47:46,  1.42it/s]
1|2389|Loss: 3.446035861968994:  37%|███▋      | 2389/6470 [33:05<42:08,  1.61it/s]
1|2390|Loss: 4.85651159286499:  37%|███▋      | 2389/6470 [33:05<42:08,  1.61it/s] 
1|2390|Loss: 4.85651159286499:  37%|███▋      | 2390/6470 [33:06<46:45,  1.45it/s]
1|2391|Loss: 3.4886553287506104:  37%|███▋      | 2390/6470 [33:06<46:45,  1.45it/s]
1|2391|Loss: 3.4886553287506104:  37%|███▋      | 2391/6470 [33:07<44:46,  1.52it/s]
1|2392|Loss: 4.190117835998535:  37%|███▋      | 2391/6470 [33:07<44:46,  1.52it/s] 
1|2392|Loss: 4.190117835998535:  37%|███▋      | 2392/6470 [33:07<46:00,  1.48it/s]
1|2393|Loss: 4.188443183898926:  37%|███▋      | 2392/6470 [33:08<46:00,  1.48it/s]
1|2393|Loss: 4.188443183898926:  37%|███▋      | 2393/6470 [33:08<44:37,  1.52it/s]
1|2394|Loss: 4.4075775146484375:  37%|███▋      | 2393/6470 [33:08<44:37,  1.52it/s]
1|2394|Loss: 4.4075775146484375:  37%|███▋      | 2394/6470 [33:09<48:44,  1.39it/s]
1|2395|Loss: 4.418619632720947:  37%|███▋      | 2394/6470 [33:09<48:44,  1.39it/s] 
1|2395|Loss: 4.418619632720947:  37%|███▋      | 2395/6470 [33:09<49:32,  1.37it/s]
1|2396|Loss: 4.388126373291016:  37%|███▋      | 2395/6470 [33:10<49:32,  1.37it/s]
1|2396|Loss: 4.388126373291016:  37%|███▋      | 2396/6470 [33:10<50:12,  1.35it/s]
1|2397|Loss: 3.853344202041626:  37%|███▋      | 2396/6470 [33:11<50:12,  1.35it/s]
1|2397|Loss: 3.853344202041626:  37%|███▋      | 2397/6470 [33:11<50:33,  1.34it/s]
1|2398|Loss: 3.9592268466949463:  37%|███▋      | 2397/6470 [33:11<50:33,  1.34it/s]
1|2398|Loss: 3.9592268466949463:  37%|███▋      | 2398/6470 [33:12<49:16,  1.38it/s]
1|2399|Loss: 4.612338066101074:  37%|███▋      | 2398/6470 [33:12<49:16,  1.38it/s] 
1|2399|Loss: 4.612338066101074:  37%|███▋      | 2399/6470 [33:12<49:44,  1.36it/s]
1|2400|Loss: 4.077324867248535:  37%|███▋      | 2399/6470 [33:13<49:44,  1.36it/s]
1|2400|Loss: 4.077324867248535:  37%|███▋      | 2400/6470 [33:13<48:52,  1.39it/s]
1|2401|Loss: 4.49127721786499:  37%|███▋      | 2400/6470 [33:13<48:52,  1.39it/s] 
1|2401|Loss: 4.49127721786499:  37%|███▋      | 2401/6470 [33:14<48:29,  1.40it/s]
1|2402|Loss: 4.2238640785217285:  37%|███▋      | 2401/6470 [33:14<48:29,  1.40it/s]
1|2402|Loss: 4.2238640785217285:  37%|███▋      | 2402/6470 [33:14<46:56,  1.44it/s]
1|2403|Loss: 4.359887599945068:  37%|███▋      | 2402/6470 [33:15<46:56,  1.44it/s] 
1|2403|Loss: 4.359887599945068:  37%|███▋      | 2403/6470 [33:15<49:34,  1.37it/s]
1|2404|Loss: 4.552347660064697:  37%|███▋      | 2403/6470 [33:16<49:34,  1.37it/s]
1|2404|Loss: 4.552347660064697:  37%|███▋      | 2404/6470 [33:16<50:24,  1.34it/s]
1|2405|Loss: 4.651130676269531:  37%|███▋      | 2404/6470 [33:17<50:24,  1.34it/s]
1|2405|Loss: 4.651130676269531:  37%|███▋      | 2405/6470 [33:17<55:23,  1.22it/s]
1|2406|Loss: 4.415604591369629:  37%|███▋      | 2405/6470 [33:17<55:23,  1.22it/s]
1|2406|Loss: 4.415604591369629:  37%|███▋      | 2406/6470 [33:18<54:34,  1.24it/s]
1|2407|Loss: 4.6410722732543945:  37%|███▋      | 2406/6470 [33:18<54:34,  1.24it/s]
1|2407|Loss: 4.6410722732543945:  37%|███▋      | 2407/6470 [33:19<53:28,  1.27it/s]
1|2408|Loss: 4.631606101989746:  37%|███▋      | 2407/6470 [33:19<53:28,  1.27it/s] 
1|2408|Loss: 4.631606101989746:  37%|███▋      | 2408/6470 [33:19<51:22,  1.32it/s]
1|2409|Loss: 4.624361038208008:  37%|███▋      | 2408/6470 [33:20<51:22,  1.32it/s]
1|2409|Loss: 4.624361038208008:  37%|███▋      | 2409/6470 [33:20<53:08,  1.27it/s]
1|2410|Loss: 4.611135482788086:  37%|███▋      | 2409/6470 [33:21<53:08,  1.27it/s]
1|2410|Loss: 4.611135482788086:  37%|███▋      | 2410/6470 [33:21<53:51,  1.26it/s]
1|2411|Loss: 3.63718843460083:  37%|███▋      | 2410/6470 [33:21<53:51,  1.26it/s] 
1|2411|Loss: 3.63718843460083:  37%|███▋      | 2411/6470 [33:22<51:02,  1.33it/s]
1|2412|Loss: 4.950681209564209:  37%|███▋      | 2411/6470 [33:22<51:02,  1.33it/s]
1|2412|Loss: 4.950681209564209:  37%|███▋      | 2412/6470 [33:22<54:38,  1.24it/s]
1|2413|Loss: 4.50325870513916:  37%|███▋      | 2412/6470 [33:23<54:38,  1.24it/s] 
1|2413|Loss: 4.50325870513916:  37%|███▋      | 2413/6470 [33:24<59:44,  1.13it/s]
1|2414|Loss: 4.678188800811768:  37%|███▋      | 2413/6470 [33:24<59:44,  1.13it/s]
1|2414|Loss: 4.678188800811768:  37%|███▋      | 2414/6470 [33:25<1:01:43,  1.10it/s]
1|2415|Loss: 4.226397514343262:  37%|███▋      | 2414/6470 [33:25<1:01:43,  1.10it/s]
1|2415|Loss: 4.226397514343262:  37%|███▋      | 2415/6470 [33:25<58:12,  1.16it/s]  
1|2416|Loss: 4.765214920043945:  37%|███▋      | 2415/6470 [33:26<58:12,  1.16it/s]
1|2416|Loss: 4.765214920043945:  37%|███▋      | 2416/6470 [33:26<56:35,  1.19it/s]
1|2417|Loss: 4.764944553375244:  37%|███▋      | 2416/6470 [33:26<56:35,  1.19it/s]
1|2417|Loss: 4.764944553375244:  37%|███▋      | 2417/6470 [33:27<56:04,  1.20it/s]
1|2418|Loss: 3.810948133468628:  37%|███▋      | 2417/6470 [33:27<56:04,  1.20it/s]
1|2418|Loss: 3.810948133468628:  37%|███▋      | 2418/6470 [33:28<54:07,  1.25it/s]
1|2419|Loss: 3.573410749435425:  37%|███▋      | 2418/6470 [33:28<54:07,  1.25it/s]
1|2419|Loss: 3.573410749435425:  37%|███▋      | 2419/6470 [33:28<47:12,  1.43it/s]
1|2420|Loss: 4.2888593673706055:  37%|███▋      | 2419/6470 [33:28<47:12,  1.43it/s]
1|2420|Loss: 4.2888593673706055:  37%|███▋      | 2420/6470 [33:29<45:41,  1.48it/s]
1|2421|Loss: 4.325859546661377:  37%|███▋      | 2420/6470 [33:29<45:41,  1.48it/s] 
1|2421|Loss: 4.325859546661377:  37%|███▋      | 2421/6470 [33:29<47:31,  1.42it/s]
1|2422|Loss: 4.761496543884277:  37%|███▋      | 2421/6470 [33:30<47:31,  1.42it/s]
1|2422|Loss: 4.761496543884277:  37%|███▋      | 2422/6470 [33:30<48:40,  1.39it/s]
1|2423|Loss: 4.60936975479126:  37%|███▋      | 2422/6470 [33:31<48:40,  1.39it/s] 
1|2423|Loss: 4.60936975479126:  37%|███▋      | 2423/6470 [33:31<51:36,  1.31it/s]
1|2424|Loss: 4.715193748474121:  37%|███▋      | 2423/6470 [33:32<51:36,  1.31it/s]
1|2424|Loss: 4.715193748474121:  37%|███▋      | 2424/6470 [33:32<54:30,  1.24it/s]
1|2425|Loss: 4.137026786804199:  37%|███▋      | 2424/6470 [33:32<54:30,  1.24it/s]
1|2425|Loss: 4.137026786804199:  37%|███▋      | 2425/6470 [33:33<56:33,  1.19it/s]
1|2426|Loss: 4.456124782562256:  37%|███▋      | 2425/6470 [33:33<56:33,  1.19it/s]
1|2426|Loss: 4.456124782562256:  37%|███▋      | 2426/6470 [33:34<54:08,  1.24it/s]
1|2427|Loss: 3.9363815784454346:  37%|███▋      | 2426/6470 [33:34<54:08,  1.24it/s]
1|2427|Loss: 3.9363815784454346:  38%|███▊      | 2427/6470 [33:34<49:50,  1.35it/s]
1|2428|Loss: 3.9650461673736572:  38%|███▊      | 2427/6470 [33:35<49:50,  1.35it/s]
1|2428|Loss: 3.9650461673736572:  38%|███▊      | 2428/6470 [33:35<50:02,  1.35it/s]
1|2429|Loss: 4.597536563873291:  38%|███▊      | 2428/6470 [33:35<50:02,  1.35it/s] 
1|2429|Loss: 4.597536563873291:  38%|███▊      | 2429/6470 [33:36<50:55,  1.32it/s]
1|2430|Loss: 4.721949577331543:  38%|███▊      | 2429/6470 [33:36<50:55,  1.32it/s]
1|2430|Loss: 4.721949577331543:  38%|███▊      | 2430/6470 [33:37<52:20,  1.29it/s]
1|2431|Loss: 4.663471221923828:  38%|███▊      | 2430/6470 [33:37<52:20,  1.29it/s]
1|2431|Loss: 4.663471221923828:  38%|███▊      | 2431/6470 [33:37<52:59,  1.27it/s]
1|2432|Loss: 4.399860858917236:  38%|███▊      | 2431/6470 [33:38<52:59,  1.27it/s]
1|2432|Loss: 4.399860858917236:  38%|███▊      | 2432/6470 [33:38<53:14,  1.26it/s]
1|2433|Loss: 4.216787338256836:  38%|███▊      | 2432/6470 [33:39<53:14,  1.26it/s]
1|2433|Loss: 4.216787338256836:  38%|███▊      | 2433/6470 [33:39<57:03,  1.18it/s]
1|2434|Loss: 4.177308559417725:  38%|███▊      | 2433/6470 [33:40<57:03,  1.18it/s]
1|2434|Loss: 4.177308559417725:  38%|███▊      | 2434/6470 [33:40<57:21,  1.17it/s]
1|2435|Loss: 4.867818832397461:  38%|███▊      | 2434/6470 [33:41<57:21,  1.17it/s]
1|2435|Loss: 4.867818832397461:  38%|███▊      | 2435/6470 [33:41<1:02:14,  1.08it/s]
1|2436|Loss: 4.823277950286865:  38%|███▊      | 2435/6470 [33:42<1:02:14,  1.08it/s]
1|2436|Loss: 4.823277950286865:  38%|███▊      | 2436/6470 [33:42<1:02:24,  1.08it/s]
1|2437|Loss: 4.82684850692749:  38%|███▊      | 2436/6470 [33:42<1:02:24,  1.08it/s] 
1|2437|Loss: 4.82684850692749:  38%|███▊      | 2437/6470 [33:43<59:00,  1.14it/s]  
1|2438|Loss: 4.163213729858398:  38%|███▊      | 2437/6470 [33:43<59:00,  1.14it/s]
1|2438|Loss: 4.163213729858398:  38%|███▊      | 2438/6470 [33:44<56:22,  1.19it/s]
1|2439|Loss: 4.8682403564453125:  38%|███▊      | 2438/6470 [33:44<56:22,  1.19it/s]
1|2439|Loss: 4.8682403564453125:  38%|███▊      | 2439/6470 [33:45<58:48,  1.14it/s]
1|2440|Loss: 4.12214469909668:  38%|███▊      | 2439/6470 [33:45<58:48,  1.14it/s]  
1|2440|Loss: 4.12214469909668:  38%|███▊      | 2440/6470 [33:45<58:00,  1.16it/s]
1|2441|Loss: 4.373893737792969:  38%|███▊      | 2440/6470 [33:46<58:00,  1.16it/s]
1|2441|Loss: 4.373893737792969:  38%|███▊      | 2441/6470 [33:46<56:41,  1.18it/s]
1|2442|Loss: 4.178497791290283:  38%|███▊      | 2441/6470 [33:47<56:41,  1.18it/s]
1|2442|Loss: 4.178497791290283:  38%|███▊      | 2442/6470 [33:47<52:51,  1.27it/s]
1|2443|Loss: 4.551174640655518:  38%|███▊      | 2442/6470 [33:47<52:51,  1.27it/s]
1|2443|Loss: 4.551174640655518:  38%|███▊      | 2443/6470 [33:48<54:37,  1.23it/s]
1|2444|Loss: 3.717794895172119:  38%|███▊      | 2443/6470 [33:48<54:37,  1.23it/s]
1|2444|Loss: 3.717794895172119:  38%|███▊      | 2444/6470 [33:49<54:49,  1.22it/s]
1|2445|Loss: 3.8137574195861816:  38%|███▊      | 2444/6470 [33:49<54:49,  1.22it/s]
1|2445|Loss: 3.8137574195861816:  38%|███▊      | 2445/6470 [33:49<49:25,  1.36it/s]
1|2446|Loss: 4.069401264190674:  38%|███▊      | 2445/6470 [33:49<49:25,  1.36it/s] 
1|2446|Loss: 4.069401264190674:  38%|███▊      | 2446/6470 [33:50<46:34,  1.44it/s]
1|2447|Loss: 4.346704483032227:  38%|███▊      | 2446/6470 [33:50<46:34,  1.44it/s]
1|2447|Loss: 4.346704483032227:  38%|███▊      | 2447/6470 [33:50<47:48,  1.40it/s]
1|2448|Loss: 4.627779960632324:  38%|███▊      | 2447/6470 [33:51<47:48,  1.40it/s]
1|2448|Loss: 4.627779960632324:  38%|███▊      | 2448/6470 [33:51<49:21,  1.36it/s]
1|2449|Loss: 3.83870267868042:  38%|███▊      | 2448/6470 [33:52<49:21,  1.36it/s] 
1|2449|Loss: 3.83870267868042:  38%|███▊      | 2449/6470 [33:52<47:00,  1.43it/s]
1|2450|Loss: 4.444507598876953:  38%|███▊      | 2449/6470 [33:52<47:00,  1.43it/s]
1|2450|Loss: 4.444507598876953:  38%|███▊      | 2450/6470 [33:53<49:00,  1.37it/s]
1|2451|Loss: 4.053186416625977:  38%|███▊      | 2450/6470 [33:53<49:00,  1.37it/s]
1|2451|Loss: 4.053186416625977:  38%|███▊      | 2451/6470 [33:53<50:45,  1.32it/s]
1|2452|Loss: 4.3091511726379395:  38%|███▊      | 2451/6470 [33:54<50:45,  1.32it/s]
1|2452|Loss: 4.3091511726379395:  38%|███▊      | 2452/6470 [33:54<51:38,  1.30it/s]
1|2453|Loss: 4.494777202606201:  38%|███▊      | 2452/6470 [33:55<51:38,  1.30it/s] 
1|2453|Loss: 4.494777202606201:  38%|███▊      | 2453/6470 [33:55<53:38,  1.25it/s]
1|2454|Loss: 4.0104217529296875:  38%|███▊      | 2453/6470 [33:56<53:38,  1.25it/s]
1|2454|Loss: 4.0104217529296875:  38%|███▊      | 2454/6470 [33:56<53:12,  1.26it/s]
1|2455|Loss: 4.40674352645874:  38%|███▊      | 2454/6470 [33:56<53:12,  1.26it/s]  
1|2455|Loss: 4.40674352645874:  38%|███▊      | 2455/6470 [33:57<52:30,  1.27it/s]
1|2456|Loss: 5.068017482757568:  38%|███▊      | 2455/6470 [33:57<52:30,  1.27it/s]
1|2456|Loss: 5.068017482757568:  38%|███▊      | 2456/6470 [33:58<56:03,  1.19it/s]
1|2457|Loss: 4.348203659057617:  38%|███▊      | 2456/6470 [33:58<56:03,  1.19it/s]
1|2457|Loss: 4.348203659057617:  38%|███▊      | 2457/6470 [33:58<54:42,  1.22it/s]
1|2458|Loss: 4.159536838531494:  38%|███▊      | 2457/6470 [33:59<54:42,  1.22it/s]
1|2458|Loss: 4.159536838531494:  38%|███▊      | 2458/6470 [33:59<53:31,  1.25it/s]
1|2459|Loss: 4.599704742431641:  38%|███▊      | 2458/6470 [34:00<53:31,  1.25it/s]
1|2459|Loss: 4.599704742431641:  38%|███▊      | 2459/6470 [34:00<51:34,  1.30it/s]
1|2460|Loss: 4.9330549240112305:  38%|███▊      | 2459/6470 [34:00<51:34,  1.30it/s]
1|2460|Loss: 4.9330549240112305:  38%|███▊      | 2460/6470 [34:01<54:36,  1.22it/s]
1|2461|Loss: 4.282344341278076:  38%|███▊      | 2460/6470 [34:01<54:36,  1.22it/s] 
1|2461|Loss: 4.282344341278076:  38%|███▊      | 2461/6470 [34:02<53:04,  1.26it/s]
1|2462|Loss: 4.0369873046875:  38%|███▊      | 2461/6470 [34:02<53:04,  1.26it/s]  
1|2462|Loss: 4.0369873046875:  38%|███▊      | 2462/6470 [34:02<52:03,  1.28it/s]
1|2463|Loss: 5.398966312408447:  38%|███▊      | 2462/6470 [34:03<52:03,  1.28it/s]
1|2463|Loss: 5.398966312408447:  38%|███▊      | 2463/6470 [34:03<55:36,  1.20it/s]
1|2464|Loss: 4.655509948730469:  38%|███▊      | 2463/6470 [34:04<55:36,  1.20it/s]
1|2464|Loss: 4.655509948730469:  38%|███▊      | 2464/6470 [34:04<54:27,  1.23it/s]
1|2465|Loss: 4.610982894897461:  38%|███▊      | 2464/6470 [34:04<54:27,  1.23it/s]
1|2465|Loss: 4.610982894897461:  38%|███▊      | 2465/6470 [34:05<53:21,  1.25it/s]
1|2466|Loss: 4.0486040115356445:  38%|███▊      | 2465/6470 [34:05<53:21,  1.25it/s]
1|2466|Loss: 4.0486040115356445:  38%|███▊      | 2466/6470 [34:06<52:26,  1.27it/s]
1|2467|Loss: 4.893353462219238:  38%|███▊      | 2466/6470 [34:06<52:26,  1.27it/s] 
1|2467|Loss: 4.893353462219238:  38%|███▊      | 2467/6470 [34:06<54:34,  1.22it/s]
1|2468|Loss: 4.623818874359131:  38%|███▊      | 2467/6470 [34:07<54:34,  1.22it/s]
1|2468|Loss: 4.623818874359131:  38%|███▊      | 2468/6470 [34:07<54:57,  1.21it/s]
1|2469|Loss: 4.769859313964844:  38%|███▊      | 2468/6470 [34:08<54:57,  1.21it/s]
1|2469|Loss: 4.769859313964844:  38%|███▊      | 2469/6470 [34:08<54:48,  1.22it/s]
1|2470|Loss: 4.054125785827637:  38%|███▊      | 2469/6470 [34:08<54:48,  1.22it/s]
1|2470|Loss: 4.054125785827637:  38%|███▊      | 2470/6470 [34:09<53:01,  1.26it/s]
1|2471|Loss: 4.264698028564453:  38%|███▊      | 2470/6470 [34:09<53:01,  1.26it/s]
1|2471|Loss: 4.264698028564453:  38%|███▊      | 2471/6470 [34:09<50:51,  1.31it/s]
1|2472|Loss: 4.03123140335083:  38%|███▊      | 2471/6470 [34:10<50:51,  1.31it/s] 
1|2472|Loss: 4.03123140335083:  38%|███▊      | 2472/6470 [34:10<50:20,  1.32it/s]
1|2473|Loss: 4.047743320465088:  38%|███▊      | 2472/6470 [34:11<50:20,  1.32it/s]
1|2473|Loss: 4.047743320465088:  38%|███▊      | 2473/6470 [34:11<50:52,  1.31it/s]
1|2474|Loss: 4.898281574249268:  38%|███▊      | 2473/6470 [34:12<50:52,  1.31it/s]
1|2474|Loss: 4.898281574249268:  38%|███▊      | 2474/6470 [34:12<53:56,  1.23it/s]
1|2475|Loss: 4.079779624938965:  38%|███▊      | 2474/6470 [34:12<53:56,  1.23it/s]
1|2475|Loss: 4.079779624938965:  38%|███▊      | 2475/6470 [34:13<52:26,  1.27it/s]
1|2476|Loss: 4.615967750549316:  38%|███▊      | 2475/6470 [34:13<52:26,  1.27it/s]
1|2476|Loss: 4.615967750549316:  38%|███▊      | 2476/6470 [34:13<51:42,  1.29it/s]
1|2477|Loss: 4.261115074157715:  38%|███▊      | 2476/6470 [34:14<51:42,  1.29it/s]
1|2477|Loss: 4.261115074157715:  38%|███▊      | 2477/6470 [34:14<50:28,  1.32it/s]
1|2478|Loss: 4.3178582191467285:  38%|███▊      | 2477/6470 [34:15<50:28,  1.32it/s]
1|2478|Loss: 4.3178582191467285:  38%|███▊      | 2478/6470 [34:15<49:24,  1.35it/s]
1|2479|Loss: 3.7848403453826904:  38%|███▊      | 2478/6470 [34:15<49:24,  1.35it/s]
1|2479|Loss: 3.7848403453826904:  38%|███▊      | 2479/6470 [34:15<45:08,  1.47it/s]
1|2480|Loss: 4.284090518951416:  38%|███▊      | 2479/6470 [34:16<45:08,  1.47it/s] 
1|2480|Loss: 4.284090518951416:  38%|███▊      | 2480/6470 [34:16<45:26,  1.46it/s]
1|2481|Loss: 4.664743423461914:  38%|███▊      | 2480/6470 [34:17<45:26,  1.46it/s]
1|2481|Loss: 4.664743423461914:  38%|███▊      | 2481/6470 [34:17<48:00,  1.38it/s]
1|2482|Loss: 4.255521297454834:  38%|███▊      | 2481/6470 [34:17<48:00,  1.38it/s]
1|2482|Loss: 4.255521297454834:  38%|███▊      | 2482/6470 [34:18<48:30,  1.37it/s]
1|2483|Loss: 4.2100090980529785:  38%|███▊      | 2482/6470 [34:18<48:30,  1.37it/s]
1|2483|Loss: 4.2100090980529785:  38%|███▊      | 2483/6470 [34:19<51:42,  1.28it/s]
1|2484|Loss: 4.6663947105407715:  38%|███▊      | 2483/6470 [34:19<51:42,  1.28it/s]
1|2484|Loss: 4.6663947105407715:  38%|███▊      | 2484/6470 [34:19<52:59,  1.25it/s]
1|2485|Loss: 3.4877827167510986:  38%|███▊      | 2484/6470 [34:20<52:59,  1.25it/s]
1|2485|Loss: 3.4877827167510986:  38%|███▊      | 2485/6470 [34:20<51:08,  1.30it/s]
1|2486|Loss: 4.5390119552612305:  38%|███▊      | 2485/6470 [34:21<51:08,  1.30it/s]
1|2486|Loss: 4.5390119552612305:  38%|███▊      | 2486/6470 [34:21<52:15,  1.27it/s]
1|2487|Loss: 4.226600646972656:  38%|███▊      | 2486/6470 [34:21<52:15,  1.27it/s] 
1|2487|Loss: 4.226600646972656:  38%|███▊      | 2487/6470 [34:22<52:39,  1.26it/s]
1|2488|Loss: 4.269367218017578:  38%|███▊      | 2487/6470 [34:22<52:39,  1.26it/s]
1|2488|Loss: 4.269367218017578:  38%|███▊      | 2488/6470 [34:22<51:23,  1.29it/s]
1|2489|Loss: 4.533302307128906:  38%|███▊      | 2488/6470 [34:23<51:23,  1.29it/s]
1|2489|Loss: 4.533302307128906:  38%|███▊      | 2489/6470 [34:23<51:45,  1.28it/s]
1|2490|Loss: 4.132348537445068:  38%|███▊      | 2489/6470 [34:24<51:45,  1.28it/s]
1|2490|Loss: 4.132348537445068:  38%|███▊      | 2490/6470 [34:24<50:47,  1.31it/s]
1|2491|Loss: 4.170551300048828:  38%|███▊      | 2490/6470 [34:24<50:47,  1.31it/s]
1|2491|Loss: 4.170551300048828:  39%|███▊      | 2491/6470 [34:25<49:36,  1.34it/s]
1|2492|Loss: 3.728576421737671:  39%|███▊      | 2491/6470 [34:25<49:36,  1.34it/s]
1|2492|Loss: 3.728576421737671:  39%|███▊      | 2492/6470 [34:25<48:09,  1.38it/s]
1|2493|Loss: 4.192000389099121:  39%|███▊      | 2492/6470 [34:26<48:09,  1.38it/s]
1|2493|Loss: 4.192000389099121:  39%|███▊      | 2493/6470 [34:26<45:26,  1.46it/s]
1|2494|Loss: 4.40061092376709:  39%|███▊      | 2493/6470 [34:26<45:26,  1.46it/s] 
1|2494|Loss: 4.40061092376709:  39%|███▊      | 2494/6470 [34:27<46:17,  1.43it/s]
1|2495|Loss: 4.431842803955078:  39%|███▊      | 2494/6470 [34:27<46:17,  1.43it/s]
1|2495|Loss: 4.431842803955078:  39%|███▊      | 2495/6470 [34:27<48:15,  1.37it/s]
1|2496|Loss: 4.012765407562256:  39%|███▊      | 2495/6470 [34:28<48:15,  1.37it/s]
1|2496|Loss: 4.012765407562256:  39%|███▊      | 2496/6470 [34:28<47:14,  1.40it/s]
1|2497|Loss: 4.403838634490967:  39%|███▊      | 2496/6470 [34:29<47:14,  1.40it/s]
1|2497|Loss: 4.403838634490967:  39%|███▊      | 2497/6470 [34:29<48:18,  1.37it/s]
1|2498|Loss: 4.561600685119629:  39%|███▊      | 2497/6470 [34:29<48:18,  1.37it/s]
1|2498|Loss: 4.561600685119629:  39%|███▊      | 2498/6470 [34:30<51:25,  1.29it/s]
1|2499|Loss: 3.969094753265381:  39%|███▊      | 2498/6470 [34:30<51:25,  1.29it/s]
1|2499|Loss: 3.969094753265381:  39%|███▊      | 2499/6470 [34:30<49:36,  1.33it/s]
1|2500|Loss: 4.437841892242432:  39%|███▊      | 2499/6470 [34:31<49:36,  1.33it/s]
1|2500|Loss: 4.437841892242432:  39%|███▊      | 2500/6470 [34:31<49:21,  1.34it/s]
1|2501|Loss: 4.092336654663086:  39%|███▊      | 2500/6470 [34:32<49:21,  1.34it/s]
1|2501|Loss: 4.092336654663086:  39%|███▊      | 2501/6470 [34:32<48:36,  1.36it/s]
1|2502|Loss: 4.444873332977295:  39%|███▊      | 2501/6470 [34:32<48:36,  1.36it/s]
1|2502|Loss: 4.444873332977295:  39%|███▊      | 2502/6470 [34:33<53:05,  1.25it/s]
1|2503|Loss: 4.4503254890441895:  39%|███▊      | 2502/6470 [34:33<53:05,  1.25it/s]
1|2503|Loss: 4.4503254890441895:  39%|███▊      | 2503/6470 [34:34<55:47,  1.19it/s]
1|2504|Loss: 3.8294589519500732:  39%|███▊      | 2503/6470 [34:34<55:47,  1.19it/s]
1|2504|Loss: 3.8294589519500732:  39%|███▊      | 2504/6470 [34:35<52:28,  1.26it/s]
1|2505|Loss: 3.9533865451812744:  39%|███▊      | 2504/6470 [34:35<52:28,  1.26it/s]
1|2505|Loss: 3.9533865451812744:  39%|███▊      | 2505/6470 [34:35<48:52,  1.35it/s]
1|2506|Loss: 3.698178768157959:  39%|███▊      | 2505/6470 [34:35<48:52,  1.35it/s] 
1|2506|Loss: 3.698178768157959:  39%|███▊      | 2506/6470 [34:36<44:24,  1.49it/s]
1|2507|Loss: 4.1715779304504395:  39%|███▊      | 2506/6470 [34:36<44:24,  1.49it/s]
1|2507|Loss: 4.1715779304504395:  39%|███▊      | 2507/6470 [34:36<43:23,  1.52it/s]
1|2508|Loss: 4.038144111633301:  39%|███▊      | 2507/6470 [34:37<43:23,  1.52it/s] 
1|2508|Loss: 4.038144111633301:  39%|███▉      | 2508/6470 [34:37<44:16,  1.49it/s]
1|2509|Loss: 3.270885944366455:  39%|███▉      | 2508/6470 [34:37<44:16,  1.49it/s]
1|2509|Loss: 3.270885944366455:  39%|███▉      | 2509/6470 [34:38<43:41,  1.51it/s]
1|2510|Loss: 4.2388153076171875:  39%|███▉      | 2509/6470 [34:38<43:41,  1.51it/s]
1|2510|Loss: 4.2388153076171875:  39%|███▉      | 2510/6470 [34:38<45:29,  1.45it/s]
1|2511|Loss: 4.56335973739624:  39%|███▉      | 2510/6470 [34:39<45:29,  1.45it/s]  
1|2511|Loss: 4.56335973739624:  39%|███▉      | 2511/6470 [34:39<44:50,  1.47it/s]
1|2512|Loss: 4.264776229858398:  39%|███▉      | 2511/6470 [34:39<44:50,  1.47it/s]
1|2512|Loss: 4.264776229858398:  39%|███▉      | 2512/6470 [34:40<46:40,  1.41it/s]
1|2513|Loss: 4.6956892013549805:  39%|███▉      | 2512/6470 [34:40<46:40,  1.41it/s]
1|2513|Loss: 4.6956892013549805:  39%|███▉      | 2513/6470 [34:41<51:18,  1.29it/s]
1|2514|Loss: 4.420858860015869:  39%|███▉      | 2513/6470 [34:41<51:18,  1.29it/s] 
1|2514|Loss: 4.420858860015869:  39%|███▉      | 2514/6470 [34:42<54:45,  1.20it/s]
1|2515|Loss: 4.177216529846191:  39%|███▉      | 2514/6470 [34:42<54:45,  1.20it/s]
1|2515|Loss: 4.177216529846191:  39%|███▉      | 2515/6470 [34:42<51:27,  1.28it/s]
1|2516|Loss: 4.008087158203125:  39%|███▉      | 2515/6470 [34:43<51:27,  1.28it/s]
1|2516|Loss: 4.008087158203125:  39%|███▉      | 2516/6470 [34:43<48:37,  1.36it/s]
1|2517|Loss: 4.45444393157959:  39%|███▉      | 2516/6470 [34:43<48:37,  1.36it/s] 
1|2517|Loss: 4.45444393157959:  39%|███▉      | 2517/6470 [34:44<49:16,  1.34it/s]
1|2518|Loss: 4.1230950355529785:  39%|███▉      | 2517/6470 [34:44<49:16,  1.34it/s]
1|2518|Loss: 4.1230950355529785:  39%|███▉      | 2518/6470 [34:45<50:13,  1.31it/s]
1|2519|Loss: 4.0185089111328125:  39%|███▉      | 2518/6470 [34:45<50:13,  1.31it/s]
1|2519|Loss: 4.0185089111328125:  39%|███▉      | 2519/6470 [34:45<50:59,  1.29it/s]
1|2520|Loss: 4.237961292266846:  39%|███▉      | 2519/6470 [34:46<50:59,  1.29it/s] 
1|2520|Loss: 4.237961292266846:  39%|███▉      | 2520/6470 [34:46<51:35,  1.28it/s]
1|2521|Loss: 4.523571968078613:  39%|███▉      | 2520/6470 [34:47<51:35,  1.28it/s]
1|2521|Loss: 4.523571968078613:  39%|███▉      | 2521/6470 [34:47<55:57,  1.18it/s]
1|2522|Loss: 4.945390701293945:  39%|███▉      | 2521/6470 [34:48<55:57,  1.18it/s]
1|2522|Loss: 4.945390701293945:  39%|███▉      | 2522/6470 [34:48<59:30,  1.11it/s]
1|2523|Loss: 4.202311038970947:  39%|███▉      | 2522/6470 [34:49<59:30,  1.11it/s]
1|2523|Loss: 4.202311038970947:  39%|███▉      | 2523/6470 [34:49<58:24,  1.13it/s]
1|2524|Loss: 4.5729451179504395:  39%|███▉      | 2523/6470 [34:49<58:24,  1.13it/s]
1|2524|Loss: 4.5729451179504395:  39%|███▉      | 2524/6470 [34:50<53:53,  1.22it/s]
1|2525|Loss: 4.5401153564453125:  39%|███▉      | 2524/6470 [34:50<53:53,  1.22it/s]
1|2525|Loss: 4.5401153564453125:  39%|███▉      | 2525/6470 [34:51<54:26,  1.21it/s]
1|2526|Loss: 4.2290215492248535:  39%|███▉      | 2525/6470 [34:51<54:26,  1.21it/s]
1|2526|Loss: 4.2290215492248535:  39%|███▉      | 2526/6470 [34:51<50:58,  1.29it/s]
1|2527|Loss: 4.686309814453125:  39%|███▉      | 2526/6470 [34:52<50:58,  1.29it/s] 
1|2527|Loss: 4.686309814453125:  39%|███▉      | 2527/6470 [34:52<48:38,  1.35it/s]
1|2528|Loss: 4.877195835113525:  39%|███▉      | 2527/6470 [34:52<48:38,  1.35it/s]
1|2528|Loss: 4.877195835113525:  39%|███▉      | 2528/6470 [34:53<50:00,  1.31it/s]
1|2529|Loss: 4.5555644035339355:  39%|███▉      | 2528/6470 [34:53<50:00,  1.31it/s]
1|2529|Loss: 4.5555644035339355:  39%|███▉      | 2529/6470 [34:53<51:03,  1.29it/s]
1|2530|Loss: 4.160679817199707:  39%|███▉      | 2529/6470 [34:54<51:03,  1.29it/s] 
1|2530|Loss: 4.160679817199707:  39%|███▉      | 2530/6470 [34:54<47:42,  1.38it/s]
1|2531|Loss: 4.40584659576416:  39%|███▉      | 2530/6470 [34:54<47:42,  1.38it/s] 
1|2531|Loss: 4.40584659576416:  39%|███▉      | 2531/6470 [34:55<46:34,  1.41it/s]
1|2532|Loss: 4.003934383392334:  39%|███▉      | 2531/6470 [34:55<46:34,  1.41it/s]
1|2532|Loss: 4.003934383392334:  39%|███▉      | 2532/6470 [34:56<48:04,  1.37it/s]
1|2533|Loss: 4.384208679199219:  39%|███▉      | 2532/6470 [34:56<48:04,  1.37it/s]
1|2533|Loss: 4.384208679199219:  39%|███▉      | 2533/6470 [34:56<50:22,  1.30it/s]
1|2534|Loss: 4.231726169586182:  39%|███▉      | 2533/6470 [34:57<50:22,  1.30it/s]
1|2534|Loss: 4.231726169586182:  39%|███▉      | 2534/6470 [34:57<51:08,  1.28it/s]
1|2535|Loss: 3.6652626991271973:  39%|███▉      | 2534/6470 [34:58<51:08,  1.28it/s]
1|2535|Loss: 3.6652626991271973:  39%|███▉      | 2535/6470 [34:58<47:25,  1.38it/s]
1|2536|Loss: 4.14353084564209:  39%|███▉      | 2535/6470 [34:58<47:25,  1.38it/s]  
1|2536|Loss: 4.14353084564209:  39%|███▉      | 2536/6470 [34:59<47:14,  1.39it/s]
1|2537|Loss: 4.072688579559326:  39%|███▉      | 2536/6470 [34:59<47:14,  1.39it/s]
1|2537|Loss: 4.072688579559326:  39%|███▉      | 2537/6470 [34:59<50:32,  1.30it/s]
1|2538|Loss: 3.49346923828125:  39%|███▉      | 2537/6470 [35:00<50:32,  1.30it/s] 
1|2538|Loss: 3.49346923828125:  39%|███▉      | 2538/6470 [35:00<45:39,  1.44it/s]
1|2539|Loss: 4.869886875152588:  39%|███▉      | 2538/6470 [35:00<45:39,  1.44it/s]
1|2539|Loss: 4.869886875152588:  39%|███▉      | 2539/6470 [35:01<46:20,  1.41it/s]
1|2540|Loss: 4.66549015045166:  39%|███▉      | 2539/6470 [35:01<46:20,  1.41it/s] 
1|2540|Loss: 4.66549015045166:  39%|███▉      | 2540/6470 [35:02<49:57,  1.31it/s]
1|2541|Loss: 4.5778703689575195:  39%|███▉      | 2540/6470 [35:02<49:57,  1.31it/s]
1|2541|Loss: 4.5778703689575195:  39%|███▉      | 2541/6470 [35:02<53:40,  1.22it/s]
1|2542|Loss: 4.0841498374938965:  39%|███▉      | 2541/6470 [35:03<53:40,  1.22it/s]
1|2542|Loss: 4.0841498374938965:  39%|███▉      | 2542/6470 [35:03<52:37,  1.24it/s]
1|2543|Loss: 4.570873737335205:  39%|███▉      | 2542/6470 [35:04<52:37,  1.24it/s] 
1|2543|Loss: 4.570873737335205:  39%|███▉      | 2543/6470 [35:04<51:41,  1.27it/s]
1|2544|Loss: 4.949811935424805:  39%|███▉      | 2543/6470 [35:05<51:41,  1.27it/s]
1|2544|Loss: 4.949811935424805:  39%|███▉      | 2544/6470 [35:05<58:16,  1.12it/s]
1|2545|Loss: 3.999884843826294:  39%|███▉      | 2544/6470 [35:06<58:16,  1.12it/s]
1|2545|Loss: 3.999884843826294:  39%|███▉      | 2545/6470 [35:06<54:15,  1.21it/s]
1|2546|Loss: 4.075326919555664:  39%|███▉      | 2545/6470 [35:06<54:15,  1.21it/s]
1|2546|Loss: 4.075326919555664:  39%|███▉      | 2546/6470 [35:07<52:35,  1.24it/s]
1|2547|Loss: 3.795673370361328:  39%|███▉      | 2546/6470 [35:07<52:35,  1.24it/s]
1|2547|Loss: 3.795673370361328:  39%|███▉      | 2547/6470 [35:07<51:33,  1.27it/s]
1|2548|Loss: 4.046022415161133:  39%|███▉      | 2547/6470 [35:08<51:33,  1.27it/s]
1|2548|Loss: 4.046022415161133:  39%|███▉      | 2548/6470 [35:08<50:53,  1.28it/s]
1|2549|Loss: 3.649876594543457:  39%|███▉      | 2548/6470 [35:08<50:53,  1.28it/s]
1|2549|Loss: 3.649876594543457:  39%|███▉      | 2549/6470 [35:09<48:58,  1.33it/s]
1|2550|Loss: 4.19044828414917:  39%|███▉      | 2549/6470 [35:09<48:58,  1.33it/s] 
1|2550|Loss: 4.19044828414917:  39%|███▉      | 2550/6470 [35:09<46:53,  1.39it/s]
1|2551|Loss: 4.309271812438965:  39%|███▉      | 2550/6470 [35:10<46:53,  1.39it/s]
1|2551|Loss: 4.309271812438965:  39%|███▉      | 2551/6470 [35:10<45:19,  1.44it/s]
1|2552|Loss: 3.873896360397339:  39%|███▉      | 2551/6470 [35:10<45:19,  1.44it/s]
1|2552|Loss: 3.873896360397339:  39%|███▉      | 2552/6470 [35:11<44:47,  1.46it/s]
1|2553|Loss: 4.191488265991211:  39%|███▉      | 2552/6470 [35:11<44:47,  1.46it/s]
1|2553|Loss: 4.191488265991211:  39%|███▉      | 2553/6470 [35:11<42:32,  1.53it/s]
1|2554|Loss: 4.013682842254639:  39%|███▉      | 2553/6470 [35:12<42:32,  1.53it/s]
1|2554|Loss: 4.013682842254639:  39%|███▉      | 2554/6470 [35:12<42:57,  1.52it/s]
1|2555|Loss: 3.2202558517456055:  39%|███▉      | 2554/6470 [35:12<42:57,  1.52it/s]
1|2555|Loss: 3.2202558517456055:  39%|███▉      | 2555/6470 [35:13<40:53,  1.60it/s]
1|2556|Loss: 3.954421281814575:  39%|███▉      | 2555/6470 [35:13<40:53,  1.60it/s] 
1|2556|Loss: 3.954421281814575:  40%|███▉      | 2556/6470 [35:13<41:37,  1.57it/s]
1|2557|Loss: 4.239681243896484:  40%|███▉      | 2556/6470 [35:14<41:37,  1.57it/s]
1|2557|Loss: 4.239681243896484:  40%|███▉      | 2557/6470 [35:14<41:52,  1.56it/s]
1|2558|Loss: 3.8905179500579834:  40%|███▉      | 2557/6470 [35:14<41:52,  1.56it/s]
1|2558|Loss: 3.8905179500579834:  40%|███▉      | 2558/6470 [35:14<42:16,  1.54it/s]
1|2559|Loss: 3.8641104698181152:  40%|███▉      | 2558/6470 [35:15<42:16,  1.54it/s]
1|2559|Loss: 3.8641104698181152:  40%|███▉      | 2559/6470 [35:15<42:48,  1.52it/s]
1|2560|Loss: 4.157017230987549:  40%|███▉      | 2559/6470 [35:16<42:48,  1.52it/s] 
1|2560|Loss: 4.157017230987549:  40%|███▉      | 2560/6470 [35:16<46:35,  1.40it/s]
1|2561|Loss: 4.428857803344727:  40%|███▉      | 2560/6470 [35:16<46:35,  1.40it/s]
1|2561|Loss: 4.428857803344727:  40%|███▉      | 2561/6470 [35:17<47:03,  1.38it/s]
1|2562|Loss: 4.3304877281188965:  40%|███▉      | 2561/6470 [35:17<47:03,  1.38it/s]
1|2562|Loss: 4.3304877281188965:  40%|███▉      | 2562/6470 [35:17<45:08,  1.44it/s]
1|2563|Loss: 4.615057468414307:  40%|███▉      | 2562/6470 [35:18<45:08,  1.44it/s] 
1|2563|Loss: 4.615057468414307:  40%|███▉      | 2563/6470 [35:18<47:31,  1.37it/s]
1|2564|Loss: 3.118778944015503:  40%|███▉      | 2563/6470 [35:19<47:31,  1.37it/s]
1|2564|Loss: 3.118778944015503:  40%|███▉      | 2564/6470 [35:19<44:16,  1.47it/s]
1|2565|Loss: 4.240452766418457:  40%|███▉      | 2564/6470 [35:19<44:16,  1.47it/s]
1|2565|Loss: 4.240452766418457:  40%|███▉      | 2565/6470 [35:19<43:01,  1.51it/s]
1|2566|Loss: 4.298559665679932:  40%|███▉      | 2565/6470 [35:20<43:01,  1.51it/s]
1|2566|Loss: 4.298559665679932:  40%|███▉      | 2566/6470 [35:20<45:20,  1.44it/s]
1|2567|Loss: 4.350934028625488:  40%|███▉      | 2566/6470 [35:21<45:20,  1.44it/s]
1|2567|Loss: 4.350934028625488:  40%|███▉      | 2567/6470 [35:21<46:45,  1.39it/s]
1|2568|Loss: 4.3757100105285645:  40%|███▉      | 2567/6470 [35:21<46:45,  1.39it/s]
1|2568|Loss: 4.3757100105285645:  40%|███▉      | 2568/6470 [35:22<47:43,  1.36it/s]
1|2569|Loss: 4.1327223777771:  40%|███▉      | 2568/6470 [35:22<47:43,  1.36it/s]   
1|2569|Loss: 4.1327223777771:  40%|███▉      | 2569/6470 [35:22<46:48,  1.39it/s]
1|2570|Loss: 4.474556922912598:  40%|███▉      | 2569/6470 [35:23<46:48,  1.39it/s]
1|2570|Loss: 4.474556922912598:  40%|███▉      | 2570/6470 [35:23<46:23,  1.40it/s]
1|2571|Loss: 4.218978404998779:  40%|███▉      | 2570/6470 [35:23<46:23,  1.40it/s]
1|2571|Loss: 4.218978404998779:  40%|███▉      | 2571/6470 [35:24<46:17,  1.40it/s]
1|2572|Loss: 4.471031188964844:  40%|███▉      | 2571/6470 [35:24<46:17,  1.40it/s]
1|2572|Loss: 4.471031188964844:  40%|███▉      | 2572/6470 [35:25<47:28,  1.37it/s]
1|2573|Loss: 4.369685173034668:  40%|███▉      | 2572/6470 [35:25<47:28,  1.37it/s]
1|2573|Loss: 4.369685173034668:  40%|███▉      | 2573/6470 [35:25<48:15,  1.35it/s]
1|2574|Loss: 3.8918917179107666:  40%|███▉      | 2573/6470 [35:26<48:15,  1.35it/s]
1|2574|Loss: 3.8918917179107666:  40%|███▉      | 2574/6470 [35:26<49:05,  1.32it/s]
1|2575|Loss: 4.536164283752441:  40%|███▉      | 2574/6470 [35:27<49:05,  1.32it/s] 
1|2575|Loss: 4.536164283752441:  40%|███▉      | 2575/6470 [35:27<49:24,  1.31it/s]
1|2576|Loss: 4.184412479400635:  40%|███▉      | 2575/6470 [35:27<49:24,  1.31it/s]
1|2576|Loss: 4.184412479400635:  40%|███▉      | 2576/6470 [35:28<49:51,  1.30it/s]
1|2577|Loss: 4.091911315917969:  40%|███▉      | 2576/6470 [35:28<49:51,  1.30it/s]
1|2577|Loss: 4.091911315917969:  40%|███▉      | 2577/6470 [35:28<48:53,  1.33it/s]
1|2578|Loss: 3.967383623123169:  40%|███▉      | 2577/6470 [35:29<48:53,  1.33it/s]
1|2578|Loss: 3.967383623123169:  40%|███▉      | 2578/6470 [35:29<47:57,  1.35it/s]
1|2579|Loss: 3.7787225246429443:  40%|███▉      | 2578/6470 [35:29<47:57,  1.35it/s]
1|2579|Loss: 3.7787225246429443:  40%|███▉      | 2579/6470 [35:30<46:55,  1.38it/s]
1|2580|Loss: 4.076198577880859:  40%|███▉      | 2579/6470 [35:30<46:55,  1.38it/s] 
1|2580|Loss: 4.076198577880859:  40%|███▉      | 2580/6470 [35:30<46:23,  1.40it/s]
1|2581|Loss: 4.553375244140625:  40%|███▉      | 2580/6470 [35:31<46:23,  1.40it/s]
1|2581|Loss: 4.553375244140625:  40%|███▉      | 2581/6470 [35:31<47:38,  1.36it/s]
1|2582|Loss: 4.615078449249268:  40%|███▉      | 2581/6470 [35:32<47:38,  1.36it/s]
1|2582|Loss: 4.615078449249268:  40%|███▉      | 2582/6470 [35:32<49:12,  1.32it/s]
1|2583|Loss: 4.881384372711182:  40%|███▉      | 2582/6470 [35:33<49:12,  1.32it/s]
1|2583|Loss: 4.881384372711182:  40%|███▉      | 2583/6470 [35:33<52:46,  1.23it/s]
1|2584|Loss: 4.276851654052734:  40%|███▉      | 2583/6470 [35:33<52:46,  1.23it/s]
1|2584|Loss: 4.276851654052734:  40%|███▉      | 2584/6470 [35:34<51:55,  1.25it/s]
1|2585|Loss: 4.23475980758667:  40%|███▉      | 2584/6470 [35:34<51:55,  1.25it/s] 
1|2585|Loss: 4.23475980758667:  40%|███▉      | 2585/6470 [35:35<51:26,  1.26it/s]
1|2586|Loss: 4.733723163604736:  40%|███▉      | 2585/6470 [35:35<51:26,  1.26it/s]
1|2586|Loss: 4.733723163604736:  40%|███▉      | 2586/6470 [35:35<50:51,  1.27it/s]
1|2587|Loss: 4.1357598304748535:  40%|███▉      | 2586/6470 [35:36<50:51,  1.27it/s]
1|2587|Loss: 4.1357598304748535:  40%|███▉      | 2587/6470 [35:36<50:28,  1.28it/s]
1|2588|Loss: 2.7276089191436768:  40%|███▉      | 2587/6470 [35:36<50:28,  1.28it/s]
1|2588|Loss: 2.7276089191436768:  40%|████      | 2588/6470 [35:37<44:01,  1.47it/s]
1|2589|Loss: 4.468102931976318:  40%|████      | 2588/6470 [35:37<44:01,  1.47it/s] 
1|2589|Loss: 4.468102931976318:  40%|████      | 2589/6470 [35:37<45:52,  1.41it/s]
1|2590|Loss: 4.36815881729126:  40%|████      | 2589/6470 [35:38<45:52,  1.41it/s] 
1|2590|Loss: 4.36815881729126:  40%|████      | 2590/6470 [35:38<46:35,  1.39it/s]
1|2591|Loss: 3.8399593830108643:  40%|████      | 2590/6470 [35:38<46:35,  1.39it/s]
1|2591|Loss: 3.8399593830108643:  40%|████      | 2591/6470 [35:39<46:21,  1.39it/s]
1|2592|Loss: 4.559841156005859:  40%|████      | 2591/6470 [35:39<46:21,  1.39it/s] 
1|2592|Loss: 4.559841156005859:  40%|████      | 2592/6470 [35:40<46:50,  1.38it/s]
1|2593|Loss: 4.359592914581299:  40%|████      | 2592/6470 [35:40<46:50,  1.38it/s]
1|2593|Loss: 4.359592914581299:  40%|████      | 2593/6470 [35:40<47:35,  1.36it/s]
1|2594|Loss: 4.180266380310059:  40%|████      | 2593/6470 [35:41<47:35,  1.36it/s]
1|2594|Loss: 4.180266380310059:  40%|████      | 2594/6470 [35:41<46:00,  1.40it/s]
1|2595|Loss: 4.597426891326904:  40%|████      | 2594/6470 [35:41<46:00,  1.40it/s]
1|2595|Loss: 4.597426891326904:  40%|████      | 2595/6470 [35:42<49:40,  1.30it/s]
1|2596|Loss: 4.454766273498535:  40%|████      | 2595/6470 [35:42<49:40,  1.30it/s]
1|2596|Loss: 4.454766273498535:  40%|████      | 2596/6470 [35:43<55:30,  1.16it/s]
1|2597|Loss: 4.475356101989746:  40%|████      | 2596/6470 [35:43<55:30,  1.16it/s]
1|2597|Loss: 4.475356101989746:  40%|████      | 2597/6470 [35:44<53:39,  1.20it/s]
1|2598|Loss: 3.9334352016448975:  40%|████      | 2597/6470 [35:44<53:39,  1.20it/s]
1|2598|Loss: 3.9334352016448975:  40%|████      | 2598/6470 [35:44<48:47,  1.32it/s]
1|2599|Loss: 4.073073387145996:  40%|████      | 2598/6470 [35:45<48:47,  1.32it/s] 
1|2599|Loss: 4.073073387145996:  40%|████      | 2599/6470 [35:45<45:52,  1.41it/s]
1|2600|Loss: 4.5068206787109375:  40%|████      | 2599/6470 [35:45<45:52,  1.41it/s]
1|2600|Loss: 4.5068206787109375:  40%|████      | 2600/6470 [35:46<47:21,  1.36it/s]
1|2601|Loss: 4.035952091217041:  40%|████      | 2600/6470 [35:46<47:21,  1.36it/s] 
1|2601|Loss: 4.035952091217041:  40%|████      | 2601/6470 [35:47<50:55,  1.27it/s]
1|2602|Loss: 3.6944878101348877:  40%|████      | 2601/6470 [35:47<50:55,  1.27it/s]
1|2602|Loss: 3.6944878101348877:  40%|████      | 2602/6470 [35:47<49:43,  1.30it/s]
1|2603|Loss: 4.114649772644043:  40%|████      | 2602/6470 [35:48<49:43,  1.30it/s] 
1|2603|Loss: 4.114649772644043:  40%|████      | 2603/6470 [35:48<49:01,  1.31it/s]
1|2604|Loss: 4.629021167755127:  40%|████      | 2603/6470 [35:49<49:01,  1.31it/s]
1|2604|Loss: 4.629021167755127:  40%|████      | 2604/6470 [35:49<51:22,  1.25it/s]
1|2605|Loss: 4.191944599151611:  40%|████      | 2604/6470 [35:49<51:22,  1.25it/s]
1|2605|Loss: 4.191944599151611:  40%|████      | 2605/6470 [35:50<50:28,  1.28it/s]
1|2606|Loss: 4.232478141784668:  40%|████      | 2605/6470 [35:50<50:28,  1.28it/s]
1|2606|Loss: 4.232478141784668:  40%|████      | 2606/6470 [35:51<52:05,  1.24it/s]
1|2607|Loss: 3.642930746078491:  40%|████      | 2606/6470 [35:51<52:05,  1.24it/s]
1|2607|Loss: 3.642930746078491:  40%|████      | 2607/6470 [35:51<49:00,  1.31it/s]
1|2608|Loss: 4.035150051116943:  40%|████      | 2607/6470 [35:52<49:00,  1.31it/s]
1|2608|Loss: 4.035150051116943:  40%|████      | 2608/6470 [35:52<48:53,  1.32it/s]
1|2609|Loss: 4.2052154541015625:  40%|████      | 2608/6470 [35:52<48:53,  1.32it/s]
1|2609|Loss: 4.2052154541015625:  40%|████      | 2609/6470 [35:53<48:19,  1.33it/s]
1|2610|Loss: 4.455789089202881:  40%|████      | 2609/6470 [35:53<48:19,  1.33it/s] 
1|2610|Loss: 4.455789089202881:  40%|████      | 2610/6470 [35:53<47:01,  1.37it/s]
1|2611|Loss: 3.943911552429199:  40%|████      | 2610/6470 [35:54<47:01,  1.37it/s]
1|2611|Loss: 3.943911552429199:  40%|████      | 2611/6470 [35:54<45:57,  1.40it/s]
1|2612|Loss: 4.164718151092529:  40%|████      | 2611/6470 [35:54<45:57,  1.40it/s]
1|2612|Loss: 4.164718151092529:  40%|████      | 2612/6470 [35:55<45:17,  1.42it/s]
1|2613|Loss: 4.091245174407959:  40%|████      | 2612/6470 [35:55<45:17,  1.42it/s]
1|2613|Loss: 4.091245174407959:  40%|████      | 2613/6470 [35:55<44:51,  1.43it/s]
1|2614|Loss: 4.119307994842529:  40%|████      | 2613/6470 [35:56<44:51,  1.43it/s]
1|2614|Loss: 4.119307994842529:  40%|████      | 2614/6470 [35:56<46:19,  1.39it/s]
1|2615|Loss: 4.763193130493164:  40%|████      | 2614/6470 [35:57<46:19,  1.39it/s]
1|2615|Loss: 4.763193130493164:  40%|████      | 2615/6470 [35:57<50:13,  1.28it/s]
1|2616|Loss: 4.759117126464844:  40%|████      | 2615/6470 [35:58<50:13,  1.28it/s]
1|2616|Loss: 4.759117126464844:  40%|████      | 2616/6470 [35:58<51:23,  1.25it/s]
1|2617|Loss: 4.557048797607422:  40%|████      | 2616/6470 [35:58<51:23,  1.25it/s]
1|2617|Loss: 4.557048797607422:  40%|████      | 2617/6470 [35:59<50:50,  1.26it/s]
1|2618|Loss: 3.068500280380249:  40%|████      | 2617/6470 [35:59<50:50,  1.26it/s]
1|2618|Loss: 3.068500280380249:  40%|████      | 2618/6470 [35:59<44:36,  1.44it/s]
1|2619|Loss: 3.982180595397949:  40%|████      | 2618/6470 [36:00<44:36,  1.44it/s]
1|2619|Loss: 3.982180595397949:  40%|████      | 2619/6470 [36:00<44:35,  1.44it/s]
1|2620|Loss: 3.863079786300659:  40%|████      | 2619/6470 [36:00<44:35,  1.44it/s]
1|2620|Loss: 3.863079786300659:  40%|████      | 2620/6470 [36:01<45:50,  1.40it/s]
1|2621|Loss: 4.065112590789795:  40%|████      | 2620/6470 [36:01<45:50,  1.40it/s]
1|2621|Loss: 4.065112590789795:  41%|████      | 2621/6470 [36:01<47:55,  1.34it/s]
1|2622|Loss: 4.472235679626465:  41%|████      | 2621/6470 [36:02<47:55,  1.34it/s]
1|2622|Loss: 4.472235679626465:  41%|████      | 2622/6470 [36:02<51:02,  1.26it/s]
1|2623|Loss: 4.43416166305542:  41%|████      | 2622/6470 [36:03<51:02,  1.26it/s] 
1|2623|Loss: 4.43416166305542:  41%|████      | 2623/6470 [36:03<52:04,  1.23it/s]
1|2624|Loss: 4.287778854370117:  41%|████      | 2623/6470 [36:04<52:04,  1.23it/s]
1|2624|Loss: 4.287778854370117:  41%|████      | 2624/6470 [36:04<51:50,  1.24it/s]
1|2625|Loss: 3.6447017192840576:  41%|████      | 2624/6470 [36:04<51:50,  1.24it/s]
1|2625|Loss: 3.6447017192840576:  41%|████      | 2625/6470 [36:05<49:30,  1.29it/s]
1|2626|Loss: 4.234892845153809:  41%|████      | 2625/6470 [36:05<49:30,  1.29it/s] 
1|2626|Loss: 4.234892845153809:  41%|████      | 2626/6470 [36:06<51:09,  1.25it/s]
1|2627|Loss: 3.201000213623047:  41%|████      | 2626/6470 [36:06<51:09,  1.25it/s]
1|2627|Loss: 3.201000213623047:  41%|████      | 2627/6470 [36:06<45:23,  1.41it/s]
1|2628|Loss: 4.357348918914795:  41%|████      | 2627/6470 [36:06<45:23,  1.41it/s]
1|2628|Loss: 4.357348918914795:  41%|████      | 2628/6470 [36:07<45:12,  1.42it/s]
1|2629|Loss: 4.192975997924805:  41%|████      | 2628/6470 [36:07<45:12,  1.42it/s]
1|2629|Loss: 4.192975997924805:  41%|████      | 2629/6470 [36:08<49:18,  1.30it/s]
1|2630|Loss: 4.1804327964782715:  41%|████      | 2629/6470 [36:08<49:18,  1.30it/s]
1|2630|Loss: 4.1804327964782715:  41%|████      | 2630/6470 [36:09<52:56,  1.21it/s]
1|2631|Loss: 4.174871921539307:  41%|████      | 2630/6470 [36:09<52:56,  1.21it/s] 
1|2631|Loss: 4.174871921539307:  41%|████      | 2631/6470 [36:09<51:40,  1.24it/s]
1|2632|Loss: 3.873100996017456:  41%|████      | 2631/6470 [36:10<51:40,  1.24it/s]
1|2632|Loss: 3.873100996017456:  41%|████      | 2632/6470 [36:10<47:06,  1.36it/s]
1|2633|Loss: 3.992372989654541:  41%|████      | 2632/6470 [36:10<47:06,  1.36it/s]
1|2633|Loss: 3.992372989654541:  41%|████      | 2633/6470 [36:11<48:22,  1.32it/s]
1|2634|Loss: 4.1164774894714355:  41%|████      | 2633/6470 [36:11<48:22,  1.32it/s]
1|2634|Loss: 4.1164774894714355:  41%|████      | 2634/6470 [36:12<50:57,  1.25it/s]
1|2635|Loss: 4.0218915939331055:  41%|████      | 2634/6470 [36:12<50:57,  1.25it/s]
1|2635|Loss: 4.0218915939331055:  41%|████      | 2635/6470 [36:12<48:56,  1.31it/s]
1|2636|Loss: 3.689598321914673:  41%|████      | 2635/6470 [36:13<48:56,  1.31it/s] 
1|2636|Loss: 3.689598321914673:  41%|████      | 2636/6470 [36:13<48:05,  1.33it/s]
1|2637|Loss: 4.015457630157471:  41%|████      | 2636/6470 [36:14<48:05,  1.33it/s]
1|2637|Loss: 4.015457630157471:  41%|████      | 2637/6470 [36:14<48:49,  1.31it/s]
1|2638|Loss: 4.850780963897705:  41%|████      | 2637/6470 [36:14<48:49,  1.31it/s]
1|2638|Loss: 4.850780963897705:  41%|████      | 2638/6470 [36:15<49:59,  1.28it/s]
1|2639|Loss: 4.203866004943848:  41%|████      | 2638/6470 [36:15<49:59,  1.28it/s]
1|2639|Loss: 4.203866004943848:  41%|████      | 2639/6470 [36:16<49:52,  1.28it/s]
1|2640|Loss: 3.2015011310577393:  41%|████      | 2639/6470 [36:16<49:52,  1.28it/s]
1|2640|Loss: 3.2015011310577393:  41%|████      | 2640/6470 [36:16<43:25,  1.47it/s]
1|2641|Loss: 3.9000942707061768:  41%|████      | 2640/6470 [36:16<43:25,  1.47it/s]
1|2641|Loss: 3.9000942707061768:  41%|████      | 2641/6470 [36:17<44:00,  1.45it/s]
1|2642|Loss: 4.075713157653809:  41%|████      | 2641/6470 [36:17<44:00,  1.45it/s] 
1|2642|Loss: 4.075713157653809:  41%|████      | 2642/6470 [36:17<44:42,  1.43it/s]
1|2643|Loss: 3.940366506576538:  41%|████      | 2642/6470 [36:18<44:42,  1.43it/s]
1|2643|Loss: 3.940366506576538:  41%|████      | 2643/6470 [36:18<44:16,  1.44it/s]
1|2644|Loss: 4.155117034912109:  41%|████      | 2643/6470 [36:18<44:16,  1.44it/s]
1|2644|Loss: 4.155117034912109:  41%|████      | 2644/6470 [36:19<44:30,  1.43it/s]
1|2645|Loss: 4.57941198348999:  41%|████      | 2644/6470 [36:19<44:30,  1.43it/s] 
1|2645|Loss: 4.57941198348999:  41%|████      | 2645/6470 [36:20<45:45,  1.39it/s]
1|2646|Loss: 4.359495162963867:  41%|████      | 2645/6470 [36:20<45:45,  1.39it/s]
1|2646|Loss: 4.359495162963867:  41%|████      | 2646/6470 [36:20<46:45,  1.36it/s]
1|2647|Loss: 4.1744914054870605:  41%|████      | 2646/6470 [36:21<46:45,  1.36it/s]
1|2647|Loss: 4.1744914054870605:  41%|████      | 2647/6470 [36:21<45:49,  1.39it/s]
1|2648|Loss: 4.206617832183838:  41%|████      | 2647/6470 [36:21<45:49,  1.39it/s] 
1|2648|Loss: 4.206617832183838:  41%|████      | 2648/6470 [36:22<46:46,  1.36it/s]
1|2649|Loss: 3.8671555519104004:  41%|████      | 2648/6470 [36:22<46:46,  1.36it/s]
1|2649|Loss: 3.8671555519104004:  41%|████      | 2649/6470 [36:22<46:00,  1.38it/s]
1|2650|Loss: 3.840653657913208:  41%|████      | 2649/6470 [36:23<46:00,  1.38it/s] 
1|2650|Loss: 3.840653657913208:  41%|████      | 2650/6470 [36:23<46:20,  1.37it/s]
1|2651|Loss: 4.357181072235107:  41%|████      | 2650/6470 [36:24<46:20,  1.37it/s]
1|2651|Loss: 4.357181072235107:  41%|████      | 2651/6470 [36:24<47:30,  1.34it/s]
1|2652|Loss: 4.303518772125244:  41%|████      | 2651/6470 [36:24<47:30,  1.34it/s]
1|2652|Loss: 4.303518772125244:  41%|████      | 2652/6470 [36:25<50:41,  1.26it/s]
1|2653|Loss: 4.404846668243408:  41%|████      | 2652/6470 [36:25<50:41,  1.26it/s]
1|2653|Loss: 4.404846668243408:  41%|████      | 2653/6470 [36:26<49:57,  1.27it/s]
1|2654|Loss: 4.443698406219482:  41%|████      | 2653/6470 [36:26<49:57,  1.27it/s]
1|2654|Loss: 4.443698406219482:  41%|████      | 2654/6470 [36:27<51:03,  1.25it/s]
1|2655|Loss: 4.021615505218506:  41%|████      | 2654/6470 [36:27<51:03,  1.25it/s]
1|2655|Loss: 4.021615505218506:  41%|████      | 2655/6470 [36:27<52:17,  1.22it/s]
1|2656|Loss: 4.104684829711914:  41%|████      | 2655/6470 [36:28<52:17,  1.22it/s]
1|2656|Loss: 4.104684829711914:  41%|████      | 2656/6470 [36:28<50:13,  1.27it/s]
1|2657|Loss: 3.8680317401885986:  41%|████      | 2656/6470 [36:28<50:13,  1.27it/s]
1|2657|Loss: 3.8680317401885986:  41%|████      | 2657/6470 [36:29<47:50,  1.33it/s]
1|2658|Loss: 4.184014320373535:  41%|████      | 2657/6470 [36:29<47:50,  1.33it/s] 
1|2658|Loss: 4.184014320373535:  41%|████      | 2658/6470 [36:29<46:13,  1.37it/s]
1|2659|Loss: 4.425799369812012:  41%|████      | 2658/6470 [36:30<46:13,  1.37it/s]
1|2659|Loss: 4.425799369812012:  41%|████      | 2659/6470 [36:30<48:56,  1.30it/s]
1|2660|Loss: 4.402050495147705:  41%|████      | 2659/6470 [36:31<48:56,  1.30it/s]
1|2660|Loss: 4.402050495147705:  41%|████      | 2660/6470 [36:31<51:48,  1.23it/s]
1|2661|Loss: 3.4508626461029053:  41%|████      | 2660/6470 [36:32<51:48,  1.23it/s]
1|2661|Loss: 3.4508626461029053:  41%|████      | 2661/6470 [36:32<51:41,  1.23it/s]
1|2662|Loss: 4.557891368865967:  41%|████      | 2661/6470 [36:32<51:41,  1.23it/s] 
1|2662|Loss: 4.557891368865967:  41%|████      | 2662/6470 [36:33<51:26,  1.23it/s]
1|2663|Loss: 4.544424533843994:  41%|████      | 2662/6470 [36:33<51:26,  1.23it/s]
1|2663|Loss: 4.544424533843994:  41%|████      | 2663/6470 [36:34<50:42,  1.25it/s]
1|2664|Loss: 4.382371425628662:  41%|████      | 2663/6470 [36:34<50:42,  1.25it/s]
1|2664|Loss: 4.382371425628662:  41%|████      | 2664/6470 [36:34<49:04,  1.29it/s]
1|2665|Loss: 4.1798996925354:  41%|████      | 2664/6470 [36:35<49:04,  1.29it/s]  
1|2665|Loss: 4.1798996925354:  41%|████      | 2665/6470 [36:35<50:10,  1.26it/s]
1|2666|Loss: 3.933147668838501:  41%|████      | 2665/6470 [36:36<50:10,  1.26it/s]
1|2666|Loss: 3.933147668838501:  41%|████      | 2666/6470 [36:36<47:29,  1.34it/s]
1|2667|Loss: 4.317831516265869:  41%|████      | 2666/6470 [36:36<47:29,  1.34it/s]
1|2667|Loss: 4.317831516265869:  41%|████      | 2667/6470 [36:37<47:46,  1.33it/s]
1|2668|Loss: 4.136623382568359:  41%|████      | 2667/6470 [36:37<47:46,  1.33it/s]
1|2668|Loss: 4.136623382568359:  41%|████      | 2668/6470 [36:37<47:19,  1.34it/s]
1|2669|Loss: 3.4917104244232178:  41%|████      | 2668/6470 [36:38<47:19,  1.34it/s]
1|2669|Loss: 3.4917104244232178:  41%|████▏     | 2669/6470 [36:38<45:13,  1.40it/s]
1|2670|Loss: 3.8401072025299072:  41%|████▏     | 2669/6470 [36:38<45:13,  1.40it/s]
1|2670|Loss: 3.8401072025299072:  41%|████▏     | 2670/6470 [36:39<45:22,  1.40it/s]
1|2671|Loss: 3.619314670562744:  41%|████▏     | 2670/6470 [36:39<45:22,  1.40it/s] 
1|2671|Loss: 3.619314670562744:  41%|████▏     | 2671/6470 [36:39<44:01,  1.44it/s]
1|2672|Loss: 4.572930812835693:  41%|████▏     | 2671/6470 [36:40<44:01,  1.44it/s]
1|2672|Loss: 4.572930812835693:  41%|████▏     | 2672/6470 [36:40<44:38,  1.42it/s]
1|2673|Loss: 5.011483192443848:  41%|████▏     | 2672/6470 [36:40<44:38,  1.42it/s]
1|2673|Loss: 5.011483192443848:  41%|████▏     | 2673/6470 [36:41<47:12,  1.34it/s]
1|2674|Loss: 4.578220367431641:  41%|████▏     | 2673/6470 [36:41<47:12,  1.34it/s]
1|2674|Loss: 4.578220367431641:  41%|████▏     | 2674/6470 [36:42<46:41,  1.36it/s]
1|2675|Loss: 3.8982467651367188:  41%|████▏     | 2674/6470 [36:42<46:41,  1.36it/s]
1|2675|Loss: 3.8982467651367188:  41%|████▏     | 2675/6470 [36:42<42:26,  1.49it/s]
1|2676|Loss: 4.526614665985107:  41%|████▏     | 2675/6470 [36:43<42:26,  1.49it/s] 
1|2676|Loss: 4.526614665985107:  41%|████▏     | 2676/6470 [36:43<46:43,  1.35it/s]
1|2677|Loss: 3.3668627738952637:  41%|████▏     | 2676/6470 [36:43<46:43,  1.35it/s]
1|2677|Loss: 3.3668627738952637:  41%|████▏     | 2677/6470 [36:44<43:07,  1.47it/s]
1|2678|Loss: 4.743989944458008:  41%|████▏     | 2677/6470 [36:44<43:07,  1.47it/s] 
1|2678|Loss: 4.743989944458008:  41%|████▏     | 2678/6470 [36:44<43:12,  1.46it/s]
1|2679|Loss: 3.7871780395507812:  41%|████▏     | 2678/6470 [36:45<43:12,  1.46it/s]
1|2679|Loss: 3.7871780395507812:  41%|████▏     | 2679/6470 [36:45<44:51,  1.41it/s]
1|2680|Loss: 4.739719867706299:  41%|████▏     | 2679/6470 [36:45<44:51,  1.41it/s] 
1|2680|Loss: 4.739719867706299:  41%|████▏     | 2680/6470 [36:46<45:58,  1.37it/s]
1|2681|Loss: 4.002449035644531:  41%|████▏     | 2680/6470 [36:46<45:58,  1.37it/s]
1|2681|Loss: 4.002449035644531:  41%|████▏     | 2681/6470 [36:47<47:57,  1.32it/s]
1|2682|Loss: 3.832805871963501:  41%|████▏     | 2681/6470 [36:47<47:57,  1.32it/s]
1|2682|Loss: 3.832805871963501:  41%|████▏     | 2682/6470 [36:47<45:28,  1.39it/s]
1|2683|Loss: 4.353486061096191:  41%|████▏     | 2682/6470 [36:48<45:28,  1.39it/s]
1|2683|Loss: 4.353486061096191:  41%|████▏     | 2683/6470 [36:48<46:12,  1.37it/s]
1|2684|Loss: 3.905257225036621:  41%|████▏     | 2683/6470 [36:48<46:12,  1.37it/s]
1|2684|Loss: 3.905257225036621:  41%|████▏     | 2684/6470 [36:49<48:40,  1.30it/s]
1|2685|Loss: 3.81564998626709:  41%|████▏     | 2684/6470 [36:49<48:40,  1.30it/s] 
1|2685|Loss: 3.81564998626709:  41%|████▏     | 2685/6470 [36:49<45:55,  1.37it/s]
1|2686|Loss: 4.493804454803467:  41%|████▏     | 2685/6470 [36:50<45:55,  1.37it/s]
1|2686|Loss: 4.493804454803467:  42%|████▏     | 2686/6470 [36:50<46:32,  1.36it/s]
1|2687|Loss: 4.035717487335205:  42%|████▏     | 2686/6470 [36:51<46:32,  1.36it/s]
1|2687|Loss: 4.035717487335205:  42%|████▏     | 2687/6470 [36:51<47:37,  1.32it/s]
1|2688|Loss: 4.082304000854492:  42%|████▏     | 2687/6470 [36:51<47:37,  1.32it/s]
1|2688|Loss: 4.082304000854492:  42%|████▏     | 2688/6470 [36:52<47:00,  1.34it/s]
1|2689|Loss: 4.475947380065918:  42%|████▏     | 2688/6470 [36:52<47:00,  1.34it/s]
1|2689|Loss: 4.475947380065918:  42%|████▏     | 2689/6470 [36:53<50:45,  1.24it/s]
1|2690|Loss: 4.557693004608154:  42%|████▏     | 2689/6470 [36:53<50:45,  1.24it/s]
1|2690|Loss: 4.557693004608154:  42%|████▏     | 2690/6470 [36:54<52:54,  1.19it/s]
1|2691|Loss: 4.782763957977295:  42%|████▏     | 2690/6470 [36:54<52:54,  1.19it/s]
1|2691|Loss: 4.782763957977295:  42%|████▏     | 2691/6470 [36:55<54:15,  1.16it/s]
1|2692|Loss: 4.726193904876709:  42%|████▏     | 2691/6470 [36:55<54:15,  1.16it/s]
1|2692|Loss: 4.726193904876709:  42%|████▏     | 2692/6470 [36:55<53:42,  1.17it/s]
1|2693|Loss: 3.966703414916992:  42%|████▏     | 2692/6470 [36:56<53:42,  1.17it/s]
1|2693|Loss: 3.966703414916992:  42%|████▏     | 2693/6470 [36:56<50:31,  1.25it/s]
1|2694|Loss: 4.225470066070557:  42%|████▏     | 2693/6470 [36:56<50:31,  1.25it/s]
1|2694|Loss: 4.225470066070557:  42%|████▏     | 2694/6470 [36:57<49:13,  1.28it/s]
1|2695|Loss: 4.608399868011475:  42%|████▏     | 2694/6470 [36:57<49:13,  1.28it/s]
1|2695|Loss: 4.608399868011475:  42%|████▏     | 2695/6470 [36:58<50:57,  1.23it/s]
1|2696|Loss: 4.340460777282715:  42%|████▏     | 2695/6470 [36:58<50:57,  1.23it/s]
1|2696|Loss: 4.340460777282715:  42%|████▏     | 2696/6470 [36:58<50:20,  1.25it/s]
1|2697|Loss: 4.33040714263916:  42%|████▏     | 2696/6470 [36:59<50:20,  1.25it/s] 
1|2697|Loss: 4.33040714263916:  42%|████▏     | 2697/6470 [36:59<51:28,  1.22it/s]
1|2698|Loss: 4.319905757904053:  42%|████▏     | 2697/6470 [37:00<51:28,  1.22it/s]
1|2698|Loss: 4.319905757904053:  42%|████▏     | 2698/6470 [37:00<52:41,  1.19it/s]
1|2699|Loss: 4.434638023376465:  42%|████▏     | 2698/6470 [37:01<52:41,  1.19it/s]
1|2699|Loss: 4.434638023376465:  42%|████▏     | 2699/6470 [37:01<54:21,  1.16it/s]
1|2700|Loss: 4.046444416046143:  42%|████▏     | 2699/6470 [37:02<54:21,  1.16it/s]
1|2700|Loss: 4.046444416046143:  42%|████▏     | 2700/6470 [37:02<52:04,  1.21it/s]
1|2701|Loss: 4.508556842803955:  42%|████▏     | 2700/6470 [37:02<52:04,  1.21it/s]
1|2701|Loss: 4.508556842803955:  42%|████▏     | 2701/6470 [37:03<52:07,  1.21it/s]
1|2702|Loss: 4.537334442138672:  42%|████▏     | 2701/6470 [37:03<52:07,  1.21it/s]
1|2702|Loss: 4.537334442138672:  42%|████▏     | 2702/6470 [37:04<51:39,  1.22it/s]
1|2703|Loss: 4.671072959899902:  42%|████▏     | 2702/6470 [37:04<51:39,  1.22it/s]
1|2703|Loss: 4.671072959899902:  42%|████▏     | 2703/6470 [37:04<51:34,  1.22it/s]
1|2704|Loss: 4.376542568206787:  42%|████▏     | 2703/6470 [37:05<51:34,  1.22it/s]
1|2704|Loss: 4.376542568206787:  42%|████▏     | 2704/6470 [37:05<50:51,  1.23it/s]
1|2705|Loss: 2.7826709747314453:  42%|████▏     | 2704/6470 [37:05<50:51,  1.23it/s]
1|2705|Loss: 2.7826709747314453:  42%|████▏     | 2705/6470 [37:06<44:02,  1.42it/s]
1|2706|Loss: 5.64291524887085:  42%|████▏     | 2705/6470 [37:06<44:02,  1.42it/s]  
1|2706|Loss: 5.64291524887085:  42%|████▏     | 2706/6470 [37:06<48:29,  1.29it/s]
1|2707|Loss: 3.7620930671691895:  42%|████▏     | 2706/6470 [37:07<48:29,  1.29it/s]
1|2707|Loss: 3.7620930671691895:  42%|████▏     | 2707/6470 [37:07<46:24,  1.35it/s]
1|2708|Loss: 5.766101360321045:  42%|████▏     | 2707/6470 [37:08<46:24,  1.35it/s] 
1|2708|Loss: 5.766101360321045:  42%|████▏     | 2708/6470 [37:08<49:10,  1.28it/s]
1|2709|Loss: 4.836706161499023:  42%|████▏     | 2708/6470 [37:08<49:10,  1.28it/s]
1|2709|Loss: 4.836706161499023:  42%|████▏     | 2709/6470 [37:09<48:16,  1.30it/s]
1|2710|Loss: 3.8731143474578857:  42%|████▏     | 2709/6470 [37:09<48:16,  1.30it/s]
1|2710|Loss: 3.8731143474578857:  42%|████▏     | 2710/6470 [37:09<45:35,  1.37it/s]
1|2711|Loss: 4.844658851623535:  42%|████▏     | 2710/6470 [37:10<45:35,  1.37it/s] 
1|2711|Loss: 4.844658851623535:  42%|████▏     | 2711/6470 [37:10<44:21,  1.41it/s]
1|2712|Loss: 4.711084365844727:  42%|████▏     | 2711/6470 [37:10<44:21,  1.41it/s]
1|2712|Loss: 4.711084365844727:  42%|████▏     | 2712/6470 [37:11<46:09,  1.36it/s]
1|2713|Loss: 4.083827018737793:  42%|████▏     | 2712/6470 [37:11<46:09,  1.36it/s]
1|2713|Loss: 4.083827018737793:  42%|████▏     | 2713/6470 [37:12<47:00,  1.33it/s]
1|2714|Loss: 4.443286418914795:  42%|████▏     | 2713/6470 [37:12<47:00,  1.33it/s]
1|2714|Loss: 4.443286418914795:  42%|████▏     | 2714/6470 [37:12<47:42,  1.31it/s]
1|2715|Loss: 4.152273654937744:  42%|████▏     | 2714/6470 [37:13<47:42,  1.31it/s]
1|2715|Loss: 4.152273654937744:  42%|████▏     | 2715/6470 [37:13<45:55,  1.36it/s]
1|2716|Loss: 5.209043979644775:  42%|████▏     | 2715/6470 [37:14<45:55,  1.36it/s]
1|2716|Loss: 5.209043979644775:  42%|████▏     | 2716/6470 [37:14<50:33,  1.24it/s]
1|2717|Loss: 4.211713790893555:  42%|████▏     | 2716/6470 [37:15<50:33,  1.24it/s]
1|2717|Loss: 4.211713790893555:  42%|████▏     | 2717/6470 [37:15<48:41,  1.28it/s]
1|2718|Loss: 3.8351032733917236:  42%|████▏     | 2717/6470 [37:15<48:41,  1.28it/s]
1|2718|Loss: 3.8351032733917236:  42%|████▏     | 2718/6470 [37:15<42:26,  1.47it/s]
1|2719|Loss: 4.254301071166992:  42%|████▏     | 2718/6470 [37:16<42:26,  1.47it/s] 
1|2719|Loss: 4.254301071166992:  42%|████▏     | 2719/6470 [37:16<43:29,  1.44it/s]
1|2720|Loss: 3.2623918056488037:  42%|████▏     | 2719/6470 [37:16<43:29,  1.44it/s]
1|2720|Loss: 3.2623918056488037:  42%|████▏     | 2720/6470 [37:17<42:27,  1.47it/s]
1|2721|Loss: 4.211732387542725:  42%|████▏     | 2720/6470 [37:17<42:27,  1.47it/s] 
1|2721|Loss: 4.211732387542725:  42%|████▏     | 2721/6470 [37:17<43:11,  1.45it/s]
1|2722|Loss: 4.0225605964660645:  42%|████▏     | 2721/6470 [37:18<43:11,  1.45it/s]
1|2722|Loss: 4.0225605964660645:  42%|████▏     | 2722/6470 [37:18<43:07,  1.45it/s]
1|2723|Loss: 4.184802532196045:  42%|████▏     | 2722/6470 [37:18<43:07,  1.45it/s] 
1|2723|Loss: 4.184802532196045:  42%|████▏     | 2723/6470 [37:19<41:35,  1.50it/s]
1|2724|Loss: 4.015427589416504:  42%|████▏     | 2723/6470 [37:19<41:35,  1.50it/s]
1|2724|Loss: 4.015427589416504:  42%|████▏     | 2724/6470 [37:19<43:16,  1.44it/s]
1|2725|Loss: 3.9932162761688232:  42%|████▏     | 2724/6470 [37:20<43:16,  1.44it/s]
1|2725|Loss: 3.9932162761688232:  42%|████▏     | 2725/6470 [37:20<43:39,  1.43it/s]
1|2726|Loss: 3.972177028656006:  42%|████▏     | 2725/6470 [37:20<43:39,  1.43it/s] 
1|2726|Loss: 3.972177028656006:  42%|████▏     | 2726/6470 [37:21<41:18,  1.51it/s]
1|2727|Loss: 4.521966457366943:  42%|████▏     | 2726/6470 [37:21<41:18,  1.51it/s]
1|2727|Loss: 4.521966457366943:  42%|████▏     | 2727/6470 [37:21<43:06,  1.45it/s]
1|2728|Loss: 3.8968348503112793:  42%|████▏     | 2727/6470 [37:22<43:06,  1.45it/s]
1|2728|Loss: 3.8968348503112793:  42%|████▏     | 2728/6470 [37:22<44:51,  1.39it/s]
1|2729|Loss: 3.59027361869812:  42%|████▏     | 2728/6470 [37:23<44:51,  1.39it/s]  
1|2729|Loss: 3.59027361869812:  42%|████▏     | 2729/6470 [37:23<43:34,  1.43it/s]
1|2730|Loss: 4.24292516708374:  42%|████▏     | 2729/6470 [37:23<43:34,  1.43it/s]
1|2730|Loss: 4.24292516708374:  42%|████▏     | 2730/6470 [37:24<44:39,  1.40it/s]
1|2731|Loss: 4.213404178619385:  42%|████▏     | 2730/6470 [37:24<44:39,  1.40it/s]
1|2731|Loss: 4.213404178619385:  42%|████▏     | 2731/6470 [37:24<45:39,  1.36it/s]
1|2732|Loss: 4.038444995880127:  42%|████▏     | 2731/6470 [37:25<45:39,  1.36it/s]
1|2732|Loss: 4.038444995880127:  42%|████▏     | 2732/6470 [37:25<45:22,  1.37it/s]
1|2733|Loss: 4.386899471282959:  42%|████▏     | 2732/6470 [37:26<45:22,  1.37it/s]
1|2733|Loss: 4.386899471282959:  42%|████▏     | 2733/6470 [37:26<44:42,  1.39it/s]
1|2734|Loss: 4.186074256896973:  42%|████▏     | 2733/6470 [37:26<44:42,  1.39it/s]
1|2734|Loss: 4.186074256896973:  42%|████▏     | 2734/6470 [37:27<44:27,  1.40it/s]
1|2735|Loss: 4.104050636291504:  42%|████▏     | 2734/6470 [37:27<44:27,  1.40it/s]
1|2735|Loss: 4.104050636291504:  42%|████▏     | 2735/6470 [37:27<42:42,  1.46it/s]
1|2736|Loss: 4.3284711837768555:  42%|████▏     | 2735/6470 [37:28<42:42,  1.46it/s]
1|2736|Loss: 4.3284711837768555:  42%|████▏     | 2736/6470 [37:28<44:19,  1.40it/s]
1|2737|Loss: 4.510136604309082:  42%|████▏     | 2736/6470 [37:28<44:19,  1.40it/s] 
1|2737|Loss: 4.510136604309082:  42%|████▏     | 2737/6470 [37:29<45:41,  1.36it/s]
1|2738|Loss: 4.002171516418457:  42%|████▏     | 2737/6470 [37:29<45:41,  1.36it/s]
1|2738|Loss: 4.002171516418457:  42%|████▏     | 2738/6470 [37:30<47:06,  1.32it/s]
1|2739|Loss: 4.0883588790893555:  42%|████▏     | 2738/6470 [37:30<47:06,  1.32it/s]
1|2739|Loss: 4.0883588790893555:  42%|████▏     | 2739/6470 [37:30<48:14,  1.29it/s]
1|2740|Loss: 4.118159770965576:  42%|████▏     | 2739/6470 [37:31<48:14,  1.29it/s] 
1|2740|Loss: 4.118159770965576:  42%|████▏     | 2740/6470 [37:31<48:48,  1.27it/s]
1|2741|Loss: 3.6752500534057617:  42%|████▏     | 2740/6470 [37:32<48:48,  1.27it/s]
1|2741|Loss: 3.6752500534057617:  42%|████▏     | 2741/6470 [37:32<45:22,  1.37it/s]
1|2742|Loss: 4.262792587280273:  42%|████▏     | 2741/6470 [37:32<45:22,  1.37it/s] 
1|2742|Loss: 4.262792587280273:  42%|████▏     | 2742/6470 [37:33<45:48,  1.36it/s]
1|2743|Loss: 4.38555383682251:  42%|████▏     | 2742/6470 [37:33<45:48,  1.36it/s] 
1|2743|Loss: 4.38555383682251:  42%|████▏     | 2743/6470 [37:33<47:02,  1.32it/s]
1|2744|Loss: 4.017820358276367:  42%|████▏     | 2743/6470 [37:34<47:02,  1.32it/s]
1|2744|Loss: 4.017820358276367:  42%|████▏     | 2744/6470 [37:34<46:35,  1.33it/s]
1|2745|Loss: 3.3201215267181396:  42%|████▏     | 2744/6470 [37:34<46:35,  1.33it/s]
1|2745|Loss: 3.3201215267181396:  42%|████▏     | 2745/6470 [37:35<42:25,  1.46it/s]
1|2746|Loss: 4.149592399597168:  42%|████▏     | 2745/6470 [37:35<42:25,  1.46it/s] 
1|2746|Loss: 4.149592399597168:  42%|████▏     | 2746/6470 [37:35<43:10,  1.44it/s]
1|2747|Loss: 4.1499104499816895:  42%|████▏     | 2746/6470 [37:36<43:10,  1.44it/s]
1|2747|Loss: 4.1499104499816895:  42%|████▏     | 2747/6470 [37:36<46:56,  1.32it/s]
1|2748|Loss: 3.8030598163604736:  42%|████▏     | 2747/6470 [37:37<46:56,  1.32it/s]
1|2748|Loss: 3.8030598163604736:  42%|████▏     | 2748/6470 [37:37<48:32,  1.28it/s]
1|2749|Loss: 4.455605506896973:  42%|████▏     | 2748/6470 [37:37<48:32,  1.28it/s] 
1|2749|Loss: 4.455605506896973:  42%|████▏     | 2749/6470 [37:38<47:10,  1.31it/s]
1|2750|Loss: 3.871311664581299:  42%|████▏     | 2749/6470 [37:38<47:10,  1.31it/s]
1|2750|Loss: 3.871311664581299:  43%|████▎     | 2750/6470 [37:38<45:26,  1.36it/s]
1|2751|Loss: 4.36777400970459:  43%|████▎     | 2750/6470 [37:39<45:26,  1.36it/s] 
1|2751|Loss: 4.36777400970459:  43%|████▎     | 2751/6470 [37:39<46:08,  1.34it/s]
1|2752|Loss: 3.9660027027130127:  43%|████▎     | 2751/6470 [37:40<46:08,  1.34it/s]
1|2752|Loss: 3.9660027027130127:  43%|████▎     | 2752/6470 [37:40<44:53,  1.38it/s]
1|2753|Loss: 4.127408504486084:  43%|████▎     | 2752/6470 [37:40<44:53,  1.38it/s] 
1|2753|Loss: 4.127408504486084:  43%|████▎     | 2753/6470 [37:41<45:40,  1.36it/s]
1|2754|Loss: 4.226918697357178:  43%|████▎     | 2753/6470 [37:41<45:40,  1.36it/s]
1|2754|Loss: 4.226918697357178:  43%|████▎     | 2754/6470 [37:41<46:52,  1.32it/s]
1|2755|Loss: 4.493648529052734:  43%|████▎     | 2754/6470 [37:42<46:52,  1.32it/s]
1|2755|Loss: 4.493648529052734:  43%|████▎     | 2755/6470 [37:42<48:56,  1.27it/s]
1|2756|Loss: 4.113360404968262:  43%|████▎     | 2755/6470 [37:43<48:56,  1.27it/s]
1|2756|Loss: 4.113360404968262:  43%|████▎     | 2756/6470 [37:43<49:25,  1.25it/s]
1|2757|Loss: 4.596752166748047:  43%|████▎     | 2756/6470 [37:44<49:25,  1.25it/s]
1|2757|Loss: 4.596752166748047:  43%|████▎     | 2757/6470 [37:44<48:10,  1.28it/s]
1|2758|Loss: 4.024542331695557:  43%|████▎     | 2757/6470 [37:44<48:10,  1.28it/s]
1|2758|Loss: 4.024542331695557:  43%|████▎     | 2758/6470 [37:45<50:32,  1.22it/s]
1|2759|Loss: 4.656187057495117:  43%|████▎     | 2758/6470 [37:45<50:32,  1.22it/s]
1|2759|Loss: 4.656187057495117:  43%|████▎     | 2759/6470 [37:46<53:14,  1.16it/s]
1|2760|Loss: 5.022215366363525:  43%|████▎     | 2759/6470 [37:46<53:14,  1.16it/s]
1|2760|Loss: 5.022215366363525:  43%|████▎     | 2760/6470 [37:47<55:33,  1.11it/s]
1|2761|Loss: 4.215420246124268:  43%|████▎     | 2760/6470 [37:47<55:33,  1.11it/s]
1|2761|Loss: 4.215420246124268:  43%|████▎     | 2761/6470 [37:48<56:47,  1.09it/s]
1|2762|Loss: 3.7680037021636963:  43%|████▎     | 2761/6470 [37:48<56:47,  1.09it/s]
1|2762|Loss: 3.7680037021636963:  43%|████▎     | 2762/6470 [37:48<53:18,  1.16it/s]
1|2763|Loss: 4.380611896514893:  43%|████▎     | 2762/6470 [37:49<53:18,  1.16it/s] 
1|2763|Loss: 4.380611896514893:  43%|████▎     | 2763/6470 [37:49<51:59,  1.19it/s]
1|2764|Loss: 3.4294588565826416:  43%|████▎     | 2763/6470 [37:50<51:59,  1.19it/s]
1|2764|Loss: 3.4294588565826416:  43%|████▎     | 2764/6470 [37:50<49:21,  1.25it/s]
1|2765|Loss: 4.470267295837402:  43%|████▎     | 2764/6470 [37:50<49:21,  1.25it/s] 
1|2765|Loss: 4.470267295837402:  43%|████▎     | 2765/6470 [37:51<47:21,  1.30it/s]
1|2766|Loss: 3.738582134246826:  43%|████▎     | 2765/6470 [37:51<47:21,  1.30it/s]
1|2766|Loss: 3.738582134246826:  43%|████▎     | 2766/6470 [37:51<43:50,  1.41it/s]
1|2767|Loss: 4.318953514099121:  43%|████▎     | 2766/6470 [37:52<43:50,  1.41it/s]
1|2767|Loss: 4.318953514099121:  43%|████▎     | 2767/6470 [37:52<44:34,  1.38it/s]
1|2768|Loss: 3.7378859519958496:  43%|████▎     | 2767/6470 [37:52<44:34,  1.38it/s]
1|2768|Loss: 3.7378859519958496:  43%|████▎     | 2768/6470 [37:52<41:19,  1.49it/s]
1|2769|Loss: 3.8693175315856934:  43%|████▎     | 2768/6470 [37:53<41:19,  1.49it/s]
1|2769|Loss: 3.8693175315856934:  43%|████▎     | 2769/6470 [37:53<39:51,  1.55it/s]
1|2770|Loss: 4.1132025718688965:  43%|████▎     | 2769/6470 [37:53<39:51,  1.55it/s]
1|2770|Loss: 4.1132025718688965:  43%|████▎     | 2770/6470 [37:54<40:45,  1.51it/s]
1|2771|Loss: 4.343291759490967:  43%|████▎     | 2770/6470 [37:54<40:45,  1.51it/s] 
1|2771|Loss: 4.343291759490967:  43%|████▎     | 2771/6470 [37:55<43:11,  1.43it/s]
1|2772|Loss: 4.2118706703186035:  43%|████▎     | 2771/6470 [37:55<43:11,  1.43it/s]
1|2772|Loss: 4.2118706703186035:  43%|████▎     | 2772/6470 [37:55<45:07,  1.37it/s]
1|2773|Loss: 4.132319927215576:  43%|████▎     | 2772/6470 [37:56<45:07,  1.37it/s] 
1|2773|Loss: 4.132319927215576:  43%|████▎     | 2773/6470 [37:56<46:18,  1.33it/s]
1|2774|Loss: 4.474695205688477:  43%|████▎     | 2773/6470 [37:57<46:18,  1.33it/s]
1|2774|Loss: 4.474695205688477:  43%|████▎     | 2774/6470 [37:57<45:50,  1.34it/s]
1|2775|Loss: 4.302671909332275:  43%|████▎     | 2774/6470 [37:57<45:50,  1.34it/s]
1|2775|Loss: 4.302671909332275:  43%|████▎     | 2775/6470 [37:58<47:54,  1.29it/s]
1|2776|Loss: 4.065959453582764:  43%|████▎     | 2775/6470 [37:58<47:54,  1.29it/s]
1|2776|Loss: 4.065959453582764:  43%|████▎     | 2776/6470 [37:58<47:07,  1.31it/s]
1|2777|Loss: 3.887432336807251:  43%|████▎     | 2776/6470 [37:59<47:07,  1.31it/s]
1|2777|Loss: 3.887432336807251:  43%|████▎     | 2777/6470 [37:59<44:43,  1.38it/s]
1|2778|Loss: 4.3519816398620605:  43%|████▎     | 2777/6470 [37:59<44:43,  1.38it/s]
1|2778|Loss: 4.3519816398620605:  43%|████▎     | 2778/6470 [38:00<44:06,  1.40it/s]
1|2779|Loss: 4.699892044067383:  43%|████▎     | 2778/6470 [38:00<44:06,  1.40it/s] 
1|2779|Loss: 4.699892044067383:  43%|████▎     | 2779/6470 [38:01<48:30,  1.27it/s]
1|2780|Loss: 3.972276449203491:  43%|████▎     | 2779/6470 [38:01<48:30,  1.27it/s]
1|2780|Loss: 3.972276449203491:  43%|████▎     | 2780/6470 [38:02<49:32,  1.24it/s]
1|2781|Loss: 4.380733966827393:  43%|████▎     | 2780/6470 [38:02<49:32,  1.24it/s]
1|2781|Loss: 4.380733966827393:  43%|████▎     | 2781/6470 [38:02<49:28,  1.24it/s]
1|2782|Loss: 3.304729461669922:  43%|████▎     | 2781/6470 [38:03<49:28,  1.24it/s]
1|2782|Loss: 3.304729461669922:  43%|████▎     | 2782/6470 [38:03<43:08,  1.42it/s]
1|2783|Loss: 4.204593658447266:  43%|████▎     | 2782/6470 [38:03<43:08,  1.42it/s]
1|2783|Loss: 4.204593658447266:  43%|████▎     | 2783/6470 [38:04<43:24,  1.42it/s]
1|2784|Loss: 3.43762469291687:  43%|████▎     | 2783/6470 [38:04<43:24,  1.42it/s] 
1|2784|Loss: 3.43762469291687:  43%|████▎     | 2784/6470 [38:04<43:24,  1.42it/s]
1|2785|Loss: 3.873296022415161:  43%|████▎     | 2784/6470 [38:05<43:24,  1.42it/s]
1|2785|Loss: 3.873296022415161:  43%|████▎     | 2785/6470 [38:05<44:36,  1.38it/s]
1|2786|Loss: 4.3907151222229:  43%|████▎     | 2785/6470 [38:05<44:36,  1.38it/s]  
1|2786|Loss: 4.3907151222229:  43%|████▎     | 2786/6470 [38:06<44:01,  1.39it/s]
1|2787|Loss: 4.173532962799072:  43%|████▎     | 2786/6470 [38:06<44:01,  1.39it/s]
1|2787|Loss: 4.173532962799072:  43%|████▎     | 2787/6470 [38:07<47:29,  1.29it/s]
1|2788|Loss: 4.365286827087402:  43%|████▎     | 2787/6470 [38:07<47:29,  1.29it/s]
1|2788|Loss: 4.365286827087402:  43%|████▎     | 2788/6470 [38:08<49:40,  1.24it/s]
1|2789|Loss: 4.059629440307617:  43%|████▎     | 2788/6470 [38:08<49:40,  1.24it/s]
1|2789|Loss: 4.059629440307617:  43%|████▎     | 2789/6470 [38:08<47:55,  1.28it/s]
1|2790|Loss: 4.24057149887085:  43%|████▎     | 2789/6470 [38:09<47:55,  1.28it/s] 
1|2790|Loss: 4.24057149887085:  43%|████▎     | 2790/6470 [38:09<48:44,  1.26it/s]
1|2791|Loss: 3.7791495323181152:  43%|████▎     | 2790/6470 [38:10<48:44,  1.26it/s]
1|2791|Loss: 3.7791495323181152:  43%|████▎     | 2791/6470 [38:10<51:12,  1.20it/s]
1|2792|Loss: 3.659491539001465:  43%|████▎     | 2791/6470 [38:10<51:12,  1.20it/s] 
1|2792|Loss: 3.659491539001465:  43%|████▎     | 2792/6470 [38:11<49:04,  1.25it/s]
1|2793|Loss: 4.082653522491455:  43%|████▎     | 2792/6470 [38:11<49:04,  1.25it/s]
1|2793|Loss: 4.082653522491455:  43%|████▎     | 2793/6470 [38:11<47:03,  1.30it/s]
1|2794|Loss: 4.043684482574463:  43%|████▎     | 2793/6470 [38:12<47:03,  1.30it/s]
1|2794|Loss: 4.043684482574463:  43%|████▎     | 2794/6470 [38:12<46:58,  1.30it/s]
1|2795|Loss: 4.061278820037842:  43%|████▎     | 2794/6470 [38:13<46:58,  1.30it/s]
1|2795|Loss: 4.061278820037842:  43%|████▎     | 2795/6470 [38:13<45:38,  1.34it/s]
1|2796|Loss: 4.110863208770752:  43%|████▎     | 2795/6470 [38:13<45:38,  1.34it/s]
1|2796|Loss: 4.110863208770752:  43%|████▎     | 2796/6470 [38:14<46:24,  1.32it/s]
1|2797|Loss: 4.192824363708496:  43%|████▎     | 2796/6470 [38:14<46:24,  1.32it/s]
1|2797|Loss: 4.192824363708496:  43%|████▎     | 2797/6470 [38:14<45:07,  1.36it/s]
1|2798|Loss: 4.267408847808838:  43%|████▎     | 2797/6470 [38:15<45:07,  1.36it/s]
1|2798|Loss: 4.267408847808838:  43%|████▎     | 2798/6470 [38:15<43:16,  1.41it/s]
1|2799|Loss: 4.3110785484313965:  43%|████▎     | 2798/6470 [38:15<43:16,  1.41it/s]
1|2799|Loss: 4.3110785484313965:  43%|████▎     | 2799/6470 [38:16<44:18,  1.38it/s]
1|2800|Loss: 3.8795206546783447:  43%|████▎     | 2799/6470 [38:16<44:18,  1.38it/s]
1|2800|Loss: 3.8795206546783447:  43%|████▎     | 2800/6470 [38:16<43:54,  1.39it/s]
1|2801|Loss: 3.7742981910705566:  43%|████▎     | 2800/6470 [38:17<43:54,  1.39it/s]
1|2801|Loss: 3.7742981910705566:  43%|████▎     | 2801/6470 [38:17<46:25,  1.32it/s]
1|2802|Loss: 4.781187057495117:  43%|████▎     | 2801/6470 [38:18<46:25,  1.32it/s] 
1|2802|Loss: 4.781187057495117:  43%|████▎     | 2802/6470 [38:18<49:43,  1.23it/s]
1|2803|Loss: 3.4284000396728516:  43%|████▎     | 2802/6470 [38:19<49:43,  1.23it/s]
1|2803|Loss: 3.4284000396728516:  43%|████▎     | 2803/6470 [38:19<48:52,  1.25it/s]
1|2804|Loss: 3.9567956924438477:  43%|████▎     | 2803/6470 [38:19<48:52,  1.25it/s]
1|2804|Loss: 3.9567956924438477:  43%|████▎     | 2804/6470 [38:20<45:31,  1.34it/s]
1|2805|Loss: 3.670539379119873:  43%|████▎     | 2804/6470 [38:20<45:31,  1.34it/s] 
1|2805|Loss: 3.670539379119873:  43%|████▎     | 2805/6470 [38:20<42:01,  1.45it/s]
1|2806|Loss: 4.337193965911865:  43%|████▎     | 2805/6470 [38:21<42:01,  1.45it/s]
1|2806|Loss: 4.337193965911865:  43%|████▎     | 2806/6470 [38:21<43:09,  1.42it/s]
1|2807|Loss: 4.420831203460693:  43%|████▎     | 2806/6470 [38:21<43:09,  1.42it/s]
1|2807|Loss: 4.420831203460693:  43%|████▎     | 2807/6470 [38:22<45:13,  1.35it/s]
1|2808|Loss: 4.762723922729492:  43%|████▎     | 2807/6470 [38:22<45:13,  1.35it/s]
1|2808|Loss: 4.762723922729492:  43%|████▎     | 2808/6470 [38:23<48:44,  1.25it/s]
1|2809|Loss: 3.5629396438598633:  43%|████▎     | 2808/6470 [38:23<48:44,  1.25it/s]
1|2809|Loss: 3.5629396438598633:  43%|████▎     | 2809/6470 [38:23<46:25,  1.31it/s]
1|2810|Loss: 3.826112747192383:  43%|████▎     | 2809/6470 [38:24<46:25,  1.31it/s] 
1|2810|Loss: 3.826112747192383:  43%|████▎     | 2810/6470 [38:24<43:47,  1.39it/s]
1|2811|Loss: 5.2357916831970215:  43%|████▎     | 2810/6470 [38:25<43:47,  1.39it/s]
1|2811|Loss: 5.2357916831970215:  43%|████▎     | 2811/6470 [38:25<56:55,  1.07it/s]
1|2812|Loss: 4.068277359008789:  43%|████▎     | 2811/6470 [38:26<56:55,  1.07it/s] 
1|2812|Loss: 4.068277359008789:  43%|████▎     | 2812/6470 [38:26<57:05,  1.07it/s]
1|2813|Loss: 4.335984230041504:  43%|████▎     | 2812/6470 [38:27<57:05,  1.07it/s]
1|2813|Loss: 4.335984230041504:  43%|████▎     | 2813/6470 [38:27<54:26,  1.12it/s]
1|2814|Loss: 3.941309690475464:  43%|████▎     | 2813/6470 [38:28<54:26,  1.12it/s]
1|2814|Loss: 3.941309690475464:  43%|████▎     | 2814/6470 [38:28<50:15,  1.21it/s]
1|2815|Loss: 4.373548984527588:  43%|████▎     | 2814/6470 [38:28<50:15,  1.21it/s]
1|2815|Loss: 4.373548984527588:  44%|████▎     | 2815/6470 [38:29<49:16,  1.24it/s]
1|2816|Loss: 4.017000675201416:  44%|████▎     | 2815/6470 [38:29<49:16,  1.24it/s]
1|2816|Loss: 4.017000675201416:  44%|████▎     | 2816/6470 [38:29<47:55,  1.27it/s]
1|2817|Loss: 3.947960615158081:  44%|████▎     | 2816/6470 [38:30<47:55,  1.27it/s]
1|2817|Loss: 3.947960615158081:  44%|████▎     | 2817/6470 [38:30<43:59,  1.38it/s]
1|2818|Loss: 4.344359874725342:  44%|████▎     | 2817/6470 [38:30<43:59,  1.38it/s]
1|2818|Loss: 4.344359874725342:  44%|████▎     | 2818/6470 [38:31<43:53,  1.39it/s]
1|2819|Loss: 3.72816801071167:  44%|████▎     | 2818/6470 [38:31<43:53,  1.39it/s] 
1|2819|Loss: 3.72816801071167:  44%|████▎     | 2819/6470 [38:31<40:33,  1.50it/s]
1|2820|Loss: 3.940702438354492:  44%|████▎     | 2819/6470 [38:32<40:33,  1.50it/s]
1|2820|Loss: 3.940702438354492:  44%|████▎     | 2820/6470 [38:32<39:50,  1.53it/s]
1|2821|Loss: 4.091364860534668:  44%|████▎     | 2820/6470 [38:32<39:50,  1.53it/s]
1|2821|Loss: 4.091364860534668:  44%|████▎     | 2821/6470 [38:33<42:21,  1.44it/s]
1|2822|Loss: 4.0626325607299805:  44%|████▎     | 2821/6470 [38:33<42:21,  1.44it/s]
1|2822|Loss: 4.0626325607299805:  44%|████▎     | 2822/6470 [38:33<42:37,  1.43it/s]
1|2823|Loss: 4.197271347045898:  44%|████▎     | 2822/6470 [38:34<42:37,  1.43it/s] 
1|2823|Loss: 4.197271347045898:  44%|████▎     | 2823/6470 [38:34<42:03,  1.45it/s]
1|2824|Loss: 4.366764545440674:  44%|████▎     | 2823/6470 [38:34<42:03,  1.45it/s]
1|2824|Loss: 4.366764545440674:  44%|████▎     | 2824/6470 [38:35<44:32,  1.36it/s]
1|2825|Loss: 3.996495246887207:  44%|████▎     | 2824/6470 [38:35<44:32,  1.36it/s]
1|2825|Loss: 3.996495246887207:  44%|████▎     | 2825/6470 [38:36<45:32,  1.33it/s]
1|2826|Loss: 4.046561241149902:  44%|████▎     | 2825/6470 [38:36<45:32,  1.33it/s]
1|2826|Loss: 4.046561241149902:  44%|████▎     | 2826/6470 [38:36<43:50,  1.39it/s]
1|2827|Loss: 4.404221534729004:  44%|████▎     | 2826/6470 [38:37<43:50,  1.39it/s]
1|2827|Loss: 4.404221534729004:  44%|████▎     | 2827/6470 [38:37<45:37,  1.33it/s]
1|2828|Loss: 3.7948169708251953:  44%|████▎     | 2827/6470 [38:37<45:37,  1.33it/s]
1|2828|Loss: 3.7948169708251953:  44%|████▎     | 2828/6470 [38:38<43:25,  1.40it/s]
1|2829|Loss: 3.433577537536621:  44%|████▎     | 2828/6470 [38:38<43:25,  1.40it/s] 
1|2829|Loss: 3.433577537536621:  44%|████▎     | 2829/6470 [38:38<41:33,  1.46it/s]
1|2830|Loss: 4.061591148376465:  44%|████▎     | 2829/6470 [38:39<41:33,  1.46it/s]
1|2830|Loss: 4.061591148376465:  44%|████▎     | 2830/6470 [38:39<41:28,  1.46it/s]
1|2831|Loss: 4.063116550445557:  44%|████▎     | 2830/6470 [38:39<41:28,  1.46it/s]
1|2831|Loss: 4.063116550445557:  44%|████▍     | 2831/6470 [38:40<42:46,  1.42it/s]
1|2832|Loss: 4.560510158538818:  44%|████▍     | 2831/6470 [38:40<42:46,  1.42it/s]
1|2832|Loss: 4.560510158538818:  44%|████▍     | 2832/6470 [38:41<44:52,  1.35it/s]
1|2833|Loss: 4.243647575378418:  44%|████▍     | 2832/6470 [38:41<44:52,  1.35it/s]
1|2833|Loss: 4.243647575378418:  44%|████▍     | 2833/6470 [38:41<44:21,  1.37it/s]
1|2834|Loss: 3.727471113204956:  44%|████▍     | 2833/6470 [38:42<44:21,  1.37it/s]
1|2834|Loss: 3.727471113204956:  44%|████▍     | 2834/6470 [38:42<44:28,  1.36it/s]
1|2835|Loss: 4.182504653930664:  44%|████▍     | 2834/6470 [38:42<44:28,  1.36it/s]
1|2835|Loss: 4.182504653930664:  44%|████▍     | 2835/6470 [38:43<45:38,  1.33it/s]
1|2836|Loss: 4.138566970825195:  44%|████▍     | 2835/6470 [38:43<45:38,  1.33it/s]
1|2836|Loss: 4.138566970825195:  44%|████▍     | 2836/6470 [38:44<48:31,  1.25it/s]
1|2837|Loss: 4.4756083488464355:  44%|████▍     | 2836/6470 [38:44<48:31,  1.25it/s]
1|2837|Loss: 4.4756083488464355:  44%|████▍     | 2837/6470 [38:45<49:26,  1.22it/s]
1|2838|Loss: 4.22342586517334:  44%|████▍     | 2837/6470 [38:45<49:26,  1.22it/s]  
1|2838|Loss: 4.22342586517334:  44%|████▍     | 2838/6470 [38:45<47:47,  1.27it/s]
1|2839|Loss: 4.047216892242432:  44%|████▍     | 2838/6470 [38:46<47:47,  1.27it/s]
1|2839|Loss: 4.047216892242432:  44%|████▍     | 2839/6470 [38:46<48:32,  1.25it/s]
1|2840|Loss: 4.235690116882324:  44%|████▍     | 2839/6470 [38:47<48:32,  1.25it/s]
1|2840|Loss: 4.235690116882324:  44%|████▍     | 2840/6470 [38:47<48:15,  1.25it/s]
1|2841|Loss: 4.123205184936523:  44%|████▍     | 2840/6470 [38:47<48:15,  1.25it/s]
1|2841|Loss: 4.123205184936523:  44%|████▍     | 2841/6470 [38:48<50:12,  1.20it/s]
1|2842|Loss: 4.425403118133545:  44%|████▍     | 2841/6470 [38:48<50:12,  1.20it/s]
1|2842|Loss: 4.425403118133545:  44%|████▍     | 2842/6470 [38:49<51:20,  1.18it/s]
1|2843|Loss: 4.03876256942749:  44%|████▍     | 2842/6470 [38:49<51:20,  1.18it/s] 
1|2843|Loss: 4.03876256942749:  44%|████▍     | 2843/6470 [38:49<47:14,  1.28it/s]
1|2844|Loss: 4.2544403076171875:  44%|████▍     | 2843/6470 [38:50<47:14,  1.28it/s]
1|2844|Loss: 4.2544403076171875:  44%|████▍     | 2844/6470 [38:50<46:27,  1.30it/s]
1|2845|Loss: 4.089507102966309:  44%|████▍     | 2844/6470 [38:51<46:27,  1.30it/s] 
1|2845|Loss: 4.089507102966309:  44%|████▍     | 2845/6470 [38:51<45:14,  1.34it/s]
1|2846|Loss: 3.90518856048584:  44%|████▍     | 2845/6470 [38:51<45:14,  1.34it/s] 
1|2846|Loss: 3.90518856048584:  44%|████▍     | 2846/6470 [38:51<43:59,  1.37it/s]
1|2847|Loss: 4.288803577423096:  44%|████▍     | 2846/6470 [38:52<43:59,  1.37it/s]
1|2847|Loss: 4.288803577423096:  44%|████▍     | 2847/6470 [38:52<44:49,  1.35it/s]
1|2848|Loss: 4.278128147125244:  44%|████▍     | 2847/6470 [38:53<44:49,  1.35it/s]
1|2848|Loss: 4.278128147125244:  44%|████▍     | 2848/6470 [38:53<45:19,  1.33it/s]
1|2849|Loss: 4.113713264465332:  44%|████▍     | 2848/6470 [38:53<45:19,  1.33it/s]
1|2849|Loss: 4.113713264465332:  44%|████▍     | 2849/6470 [38:54<44:13,  1.36it/s]
1|2850|Loss: 3.635068893432617:  44%|████▍     | 2849/6470 [38:54<44:13,  1.36it/s]
1|2850|Loss: 3.635068893432617:  44%|████▍     | 2850/6470 [38:54<43:33,  1.38it/s]
1|2851|Loss: 4.649890899658203:  44%|████▍     | 2850/6470 [38:55<43:33,  1.38it/s]
1|2851|Loss: 4.649890899658203:  44%|████▍     | 2851/6470 [38:55<47:16,  1.28it/s]
1|2852|Loss: 4.402257919311523:  44%|████▍     | 2851/6470 [38:56<47:16,  1.28it/s]
1|2852|Loss: 4.402257919311523:  44%|████▍     | 2852/6470 [38:56<48:04,  1.25it/s]
1|2853|Loss: 4.098967552185059:  44%|████▍     | 2852/6470 [38:57<48:04,  1.25it/s]
1|2853|Loss: 4.098967552185059:  44%|████▍     | 2853/6470 [38:57<46:39,  1.29it/s]
1|2854|Loss: 4.093681812286377:  44%|████▍     | 2853/6470 [38:57<46:39,  1.29it/s]
1|2854|Loss: 4.093681812286377:  44%|████▍     | 2854/6470 [38:58<46:17,  1.30it/s]
1|2855|Loss: 4.177534103393555:  44%|████▍     | 2854/6470 [38:58<46:17,  1.30it/s]
1|2855|Loss: 4.177534103393555:  44%|████▍     | 2855/6470 [38:59<48:18,  1.25it/s]
1|2856|Loss: 3.924391746520996:  44%|████▍     | 2855/6470 [38:59<48:18,  1.25it/s]
1|2856|Loss: 3.924391746520996:  44%|████▍     | 2856/6470 [38:59<48:18,  1.25it/s]
1|2857|Loss: 4.681149005889893:  44%|████▍     | 2856/6470 [39:00<48:18,  1.25it/s]
1|2857|Loss: 4.681149005889893:  44%|████▍     | 2857/6470 [39:00<50:25,  1.19it/s]
1|2858|Loss: 4.267444133758545:  44%|████▍     | 2857/6470 [39:01<50:25,  1.19it/s]
1|2858|Loss: 4.267444133758545:  44%|████▍     | 2858/6470 [39:01<50:22,  1.20it/s]
1|2859|Loss: 4.256056308746338:  44%|████▍     | 2858/6470 [39:02<50:22,  1.20it/s]
1|2859|Loss: 4.256056308746338:  44%|████▍     | 2859/6470 [39:02<49:14,  1.22it/s]
1|2860|Loss: 3.715116024017334:  44%|████▍     | 2859/6470 [39:02<49:14,  1.22it/s]
1|2860|Loss: 3.715116024017334:  44%|████▍     | 2860/6470 [39:03<46:24,  1.30it/s]
1|2861|Loss: 3.6500325202941895:  44%|████▍     | 2860/6470 [39:03<46:24,  1.30it/s]
1|2861|Loss: 3.6500325202941895:  44%|████▍     | 2861/6470 [39:03<43:56,  1.37it/s]
1|2862|Loss: 4.205315589904785:  44%|████▍     | 2861/6470 [39:04<43:56,  1.37it/s] 
1|2862|Loss: 4.205315589904785:  44%|████▍     | 2862/6470 [39:04<42:51,  1.40it/s]
1|2863|Loss: 4.421847820281982:  44%|████▍     | 2862/6470 [39:04<42:51,  1.40it/s]
1|2863|Loss: 4.421847820281982:  44%|████▍     | 2863/6470 [39:05<43:35,  1.38it/s]
1|2864|Loss: 4.291705131530762:  44%|████▍     | 2863/6470 [39:05<43:35,  1.38it/s]
1|2864|Loss: 4.291705131530762:  44%|████▍     | 2864/6470 [39:05<44:15,  1.36it/s]
1|2865|Loss: 3.5218758583068848:  44%|████▍     | 2864/6470 [39:06<44:15,  1.36it/s]
1|2865|Loss: 3.5218758583068848:  44%|████▍     | 2865/6470 [39:06<41:27,  1.45it/s]
1|2866|Loss: 4.269209384918213:  44%|████▍     | 2865/6470 [39:06<41:27,  1.45it/s] 
1|2866|Loss: 4.269209384918213:  44%|████▍     | 2866/6470 [39:07<44:27,  1.35it/s]
1|2867|Loss: 4.132718563079834:  44%|████▍     | 2866/6470 [39:07<44:27,  1.35it/s]
1|2867|Loss: 4.132718563079834:  44%|████▍     | 2867/6470 [39:08<44:28,  1.35it/s]
1|2868|Loss: 4.144627571105957:  44%|████▍     | 2867/6470 [39:08<44:28,  1.35it/s]
1|2868|Loss: 4.144627571105957:  44%|████▍     | 2868/6470 [39:08<44:39,  1.34it/s]
1|2869|Loss: 3.685497999191284:  44%|████▍     | 2868/6470 [39:09<44:39,  1.34it/s]
1|2869|Loss: 3.685497999191284:  44%|████▍     | 2869/6470 [39:09<41:15,  1.45it/s]
1|2870|Loss: 4.084831237792969:  44%|████▍     | 2869/6470 [39:09<41:15,  1.45it/s]
1|2870|Loss: 4.084831237792969:  44%|████▍     | 2870/6470 [39:09<40:39,  1.48it/s]
1|2871|Loss: 4.23062801361084:  44%|████▍     | 2870/6470 [39:10<40:39,  1.48it/s] 
1|2871|Loss: 4.23062801361084:  44%|████▍     | 2871/6470 [39:10<44:16,  1.35it/s]
1|2872|Loss: 3.898148775100708:  44%|████▍     | 2871/6470 [39:11<44:16,  1.35it/s]
1|2872|Loss: 3.898148775100708:  44%|████▍     | 2872/6470 [39:11<45:58,  1.30it/s]
1|2873|Loss: 4.379604339599609:  44%|████▍     | 2872/6470 [39:12<45:58,  1.30it/s]
1|2873|Loss: 4.379604339599609:  44%|████▍     | 2873/6470 [39:12<47:42,  1.26it/s]
1|2874|Loss: 4.235180854797363:  44%|████▍     | 2873/6470 [39:13<47:42,  1.26it/s]
1|2874|Loss: 4.235180854797363:  44%|████▍     | 2874/6470 [39:13<49:43,  1.21it/s]
1|2875|Loss: 3.7381794452667236:  44%|████▍     | 2874/6470 [39:13<49:43,  1.21it/s]
1|2875|Loss: 3.7381794452667236:  44%|████▍     | 2875/6470 [39:14<46:17,  1.29it/s]
1|2876|Loss: 4.184052467346191:  44%|████▍     | 2875/6470 [39:14<46:17,  1.29it/s] 
1|2876|Loss: 4.184052467346191:  44%|████▍     | 2876/6470 [39:14<45:15,  1.32it/s]
1|2877|Loss: 4.468729496002197:  44%|████▍     | 2876/6470 [39:15<45:15,  1.32it/s]
1|2877|Loss: 4.468729496002197:  44%|████▍     | 2877/6470 [39:15<48:21,  1.24it/s]
1|2878|Loss: 4.2083868980407715:  44%|████▍     | 2877/6470 [39:16<48:21,  1.24it/s]
1|2878|Loss: 4.2083868980407715:  44%|████▍     | 2878/6470 [39:16<50:35,  1.18it/s]
1|2879|Loss: 4.545684337615967:  44%|████▍     | 2878/6470 [39:17<50:35,  1.18it/s] 
1|2879|Loss: 4.545684337615967:  44%|████▍     | 2879/6470 [39:17<49:39,  1.21it/s]
1|2880|Loss: 4.131464958190918:  44%|████▍     | 2879/6470 [39:18<49:39,  1.21it/s]
1|2880|Loss: 4.131464958190918:  45%|████▍     | 2880/6470 [39:18<51:21,  1.17it/s]
1|2881|Loss: 3.546022653579712:  45%|████▍     | 2880/6470 [39:18<51:21,  1.17it/s]
1|2881|Loss: 3.546022653579712:  45%|████▍     | 2881/6470 [39:19<49:05,  1.22it/s]
1|2882|Loss: 3.475919485092163:  45%|████▍     | 2881/6470 [39:19<49:05,  1.22it/s]
1|2882|Loss: 3.475919485092163:  45%|████▍     | 2882/6470 [39:19<43:48,  1.36it/s]
1|2883|Loss: 3.525831699371338:  45%|████▍     | 2882/6470 [39:20<43:48,  1.36it/s]
1|2883|Loss: 3.525831699371338:  45%|████▍     | 2883/6470 [39:20<42:49,  1.40it/s]
1|2884|Loss: 3.4762237071990967:  45%|████▍     | 2883/6470 [39:20<42:49,  1.40it/s]
1|2884|Loss: 3.4762237071990967:  45%|████▍     | 2884/6470 [39:20<40:38,  1.47it/s]
1|2885|Loss: 3.949699640274048:  45%|████▍     | 2884/6470 [39:21<40:38,  1.47it/s] 
1|2885|Loss: 3.949699640274048:  45%|████▍     | 2885/6470 [39:21<40:02,  1.49it/s]
1|2886|Loss: 4.491135597229004:  45%|████▍     | 2885/6470 [39:22<40:02,  1.49it/s]
1|2886|Loss: 4.491135597229004:  45%|████▍     | 2886/6470 [39:22<41:40,  1.43it/s]
1|2887|Loss: 4.270879745483398:  45%|████▍     | 2886/6470 [39:22<41:40,  1.43it/s]
1|2887|Loss: 4.270879745483398:  45%|████▍     | 2887/6470 [39:23<45:02,  1.33it/s]
1|2888|Loss: 3.8540046215057373:  45%|████▍     | 2887/6470 [39:23<45:02,  1.33it/s]
1|2888|Loss: 3.8540046215057373:  45%|████▍     | 2888/6470 [39:24<46:18,  1.29it/s]
1|2889|Loss: 3.9943437576293945:  45%|████▍     | 2888/6470 [39:24<46:18,  1.29it/s]
1|2889|Loss: 3.9943437576293945:  45%|████▍     | 2889/6470 [39:24<48:52,  1.22it/s]
1|2890|Loss: 3.439250946044922:  45%|████▍     | 2889/6470 [39:25<48:52,  1.22it/s] 
1|2890|Loss: 3.439250946044922:  45%|████▍     | 2890/6470 [39:25<47:04,  1.27it/s]
1|2891|Loss: 4.039182186126709:  45%|████▍     | 2890/6470 [39:26<47:04,  1.27it/s]
1|2891|Loss: 4.039182186126709:  45%|████▍     | 2891/6470 [39:26<44:24,  1.34it/s]
1|2892|Loss: 3.9747025966644287:  45%|████▍     | 2891/6470 [39:26<44:24,  1.34it/s]
1|2892|Loss: 3.9747025966644287:  45%|████▍     | 2892/6470 [39:27<43:15,  1.38it/s]
1|2893|Loss: 4.090397357940674:  45%|████▍     | 2892/6470 [39:27<43:15,  1.38it/s] 
1|2893|Loss: 4.090397357940674:  45%|████▍     | 2893/6470 [39:27<41:26,  1.44it/s]
1|2894|Loss: 3.8323514461517334:  45%|████▍     | 2893/6470 [39:28<41:26,  1.44it/s]
1|2894|Loss: 3.8323514461517334:  45%|████▍     | 2894/6470 [39:28<40:39,  1.47it/s]
1|2895|Loss: 3.7093141078948975:  45%|████▍     | 2894/6470 [39:28<40:39,  1.47it/s]
1|2895|Loss: 3.7093141078948975:  45%|████▍     | 2895/6470 [39:28<40:31,  1.47it/s]
1|2896|Loss: 4.029814720153809:  45%|████▍     | 2895/6470 [39:29<40:31,  1.47it/s] 
1|2896|Loss: 4.029814720153809:  45%|████▍     | 2896/6470 [39:29<41:18,  1.44it/s]
1|2897|Loss: 4.060997009277344:  45%|████▍     | 2896/6470 [39:30<41:18,  1.44it/s]
1|2897|Loss: 4.060997009277344:  45%|████▍     | 2897/6470 [39:30<43:25,  1.37it/s]
1|2898|Loss: 4.641971588134766:  45%|████▍     | 2897/6470 [39:31<43:25,  1.37it/s]
1|2898|Loss: 4.641971588134766:  45%|████▍     | 2898/6470 [39:31<49:49,  1.20it/s]
1|2899|Loss: 4.3811845779418945:  45%|████▍     | 2898/6470 [39:32<49:49,  1.20it/s]
1|2899|Loss: 4.3811845779418945:  45%|████▍     | 2899/6470 [39:32<52:37,  1.13it/s]
1|2900|Loss: 4.363766670227051:  45%|████▍     | 2899/6470 [39:33<52:37,  1.13it/s] 
1|2900|Loss: 4.363766670227051:  45%|████▍     | 2900/6470 [39:33<51:22,  1.16it/s]
1|2901|Loss: 3.3210537433624268:  45%|████▍     | 2900/6470 [39:33<51:22,  1.16it/s]
1|2901|Loss: 3.3210537433624268:  45%|████▍     | 2901/6470 [39:34<48:18,  1.23it/s]
1|2902|Loss: 3.940545082092285:  45%|████▍     | 2901/6470 [39:34<48:18,  1.23it/s] 
1|2902|Loss: 3.940545082092285:  45%|████▍     | 2902/6470 [39:34<46:18,  1.28it/s]
1|2903|Loss: 3.7209062576293945:  45%|████▍     | 2902/6470 [39:35<46:18,  1.28it/s]
1|2903|Loss: 3.7209062576293945:  45%|████▍     | 2903/6470 [39:35<44:58,  1.32it/s]
1|2904|Loss: 3.8629510402679443:  45%|████▍     | 2903/6470 [39:35<44:58,  1.32it/s]
1|2904|Loss: 3.8629510402679443:  45%|████▍     | 2904/6470 [39:36<42:18,  1.40it/s]
1|2905|Loss: 4.0657501220703125:  45%|████▍     | 2904/6470 [39:36<42:18,  1.40it/s]
1|2905|Loss: 4.0657501220703125:  45%|████▍     | 2905/6470 [39:36<40:55,  1.45it/s]
1|2906|Loss: 4.385887145996094:  45%|████▍     | 2905/6470 [39:37<40:55,  1.45it/s] 
1|2906|Loss: 4.385887145996094:  45%|████▍     | 2906/6470 [39:37<40:02,  1.48it/s]
1|2907|Loss: 4.126544952392578:  45%|████▍     | 2906/6470 [39:37<40:02,  1.48it/s]
1|2907|Loss: 4.126544952392578:  45%|████▍     | 2907/6470 [39:38<40:14,  1.48it/s]
1|2908|Loss: 3.682852268218994:  45%|████▍     | 2907/6470 [39:38<40:14,  1.48it/s]
1|2908|Loss: 3.682852268218994:  45%|████▍     | 2908/6470 [39:38<41:28,  1.43it/s]
1|2909|Loss: 3.5890305042266846:  45%|████▍     | 2908/6470 [39:39<41:28,  1.43it/s]
1|2909|Loss: 3.5890305042266846:  45%|████▍     | 2909/6470 [39:39<42:24,  1.40it/s]
1|2910|Loss: 4.4314188957214355:  45%|████▍     | 2909/6470 [39:40<42:24,  1.40it/s]
1|2910|Loss: 4.4314188957214355:  45%|████▍     | 2910/6470 [39:40<44:52,  1.32it/s]
1|2911|Loss: 4.013341426849365:  45%|████▍     | 2910/6470 [39:40<44:52,  1.32it/s] 
1|2911|Loss: 4.013341426849365:  45%|████▍     | 2911/6470 [39:41<45:49,  1.29it/s]
1|2912|Loss: 4.50511360168457:  45%|████▍     | 2911/6470 [39:41<45:49,  1.29it/s] 
1|2912|Loss: 4.50511360168457:  45%|████▌     | 2912/6470 [39:42<48:17,  1.23it/s]
1|2913|Loss: 3.7819571495056152:  45%|████▌     | 2912/6470 [39:42<48:17,  1.23it/s]
1|2913|Loss: 3.7819571495056152:  45%|████▌     | 2913/6470 [39:42<48:30,  1.22it/s]
1|2914|Loss: 4.1656575202941895:  45%|████▌     | 2913/6470 [39:43<48:30,  1.22it/s]
1|2914|Loss: 4.1656575202941895:  45%|████▌     | 2914/6470 [39:43<48:15,  1.23it/s]
1|2915|Loss: 4.451877117156982:  45%|████▌     | 2914/6470 [39:44<48:15,  1.23it/s] 
1|2915|Loss: 4.451877117156982:  45%|████▌     | 2915/6470 [39:44<49:36,  1.19it/s]
1|2916|Loss: 4.075017929077148:  45%|████▌     | 2915/6470 [39:45<49:36,  1.19it/s]
1|2916|Loss: 4.075017929077148:  45%|████▌     | 2916/6470 [39:45<50:19,  1.18it/s]
1|2917|Loss: 3.8823165893554688:  45%|████▌     | 2916/6470 [39:45<50:19,  1.18it/s]
1|2917|Loss: 3.8823165893554688:  45%|████▌     | 2917/6470 [39:46<47:46,  1.24it/s]
1|2918|Loss: 4.064777374267578:  45%|████▌     | 2917/6470 [39:46<47:46,  1.24it/s] 
1|2918|Loss: 4.064777374267578:  45%|████▌     | 2918/6470 [39:46<45:44,  1.29it/s]
1|2919|Loss: 4.252045631408691:  45%|████▌     | 2918/6470 [39:47<45:44,  1.29it/s]
1|2919|Loss: 4.252045631408691:  45%|████▌     | 2919/6470 [39:47<47:27,  1.25it/s]
1|2920|Loss: 3.5349185466766357:  45%|████▌     | 2919/6470 [39:48<47:27,  1.25it/s]
1|2920|Loss: 3.5349185466766357:  45%|████▌     | 2920/6470 [39:48<45:09,  1.31it/s]
1|2921|Loss: 4.300512313842773:  45%|████▌     | 2920/6470 [39:48<45:09,  1.31it/s] 
1|2921|Loss: 4.300512313842773:  45%|████▌     | 2921/6470 [39:49<47:16,  1.25it/s]
1|2922|Loss: 4.054931640625:  45%|████▌     | 2921/6470 [39:49<47:16,  1.25it/s]   
1|2922|Loss: 4.054931640625:  45%|████▌     | 2922/6470 [39:50<47:50,  1.24it/s]
1|2923|Loss: 4.1068525314331055:  45%|████▌     | 2922/6470 [39:50<47:50,  1.24it/s]
1|2923|Loss: 4.1068525314331055:  45%|████▌     | 2923/6470 [39:51<47:49,  1.24it/s]
1|2924|Loss: 3.361772060394287:  45%|████▌     | 2923/6470 [39:51<47:49,  1.24it/s] 
1|2924|Loss: 3.361772060394287:  45%|████▌     | 2924/6470 [39:51<47:30,  1.24it/s]
1|2925|Loss: 4.408644199371338:  45%|████▌     | 2924/6470 [39:52<47:30,  1.24it/s]
1|2925|Loss: 4.408644199371338:  45%|████▌     | 2925/6470 [39:52<46:04,  1.28it/s]
1|2926|Loss: 3.5344443321228027:  45%|████▌     | 2925/6470 [39:52<46:04,  1.28it/s]
1|2926|Loss: 3.5344443321228027:  45%|████▌     | 2926/6470 [39:53<43:16,  1.37it/s]
1|2927|Loss: 3.9856553077697754:  45%|████▌     | 2926/6470 [39:53<43:16,  1.37it/s]
1|2927|Loss: 3.9856553077697754:  45%|████▌     | 2927/6470 [39:53<41:47,  1.41it/s]
1|2928|Loss: 4.13138484954834:  45%|████▌     | 2927/6470 [39:54<41:47,  1.41it/s]  
1|2928|Loss: 4.13138484954834:  45%|████▌     | 2928/6470 [39:54<46:46,  1.26it/s]
1|2929|Loss: 3.869619607925415:  45%|████▌     | 2928/6470 [39:55<46:46,  1.26it/s]
1|2929|Loss: 3.869619607925415:  45%|████▌     | 2929/6470 [39:55<49:40,  1.19it/s]
1|2930|Loss: 4.173179626464844:  45%|████▌     | 2929/6470 [39:56<49:40,  1.19it/s]
1|2930|Loss: 4.173179626464844:  45%|████▌     | 2930/6470 [39:56<47:52,  1.23it/s]
1|2931|Loss: 4.650327682495117:  45%|████▌     | 2930/6470 [39:56<47:52,  1.23it/s]
1|2931|Loss: 4.650327682495117:  45%|████▌     | 2931/6470 [39:57<46:29,  1.27it/s]
1|2932|Loss: 4.608364582061768:  45%|████▌     | 2931/6470 [39:57<46:29,  1.27it/s]
1|2932|Loss: 4.608364582061768:  45%|████▌     | 2932/6470 [39:57<45:07,  1.31it/s]
1|2933|Loss: 3.4594831466674805:  45%|████▌     | 2932/6470 [39:58<45:07,  1.31it/s]
1|2933|Loss: 3.4594831466674805:  45%|████▌     | 2933/6470 [39:58<46:31,  1.27it/s]
1|2934|Loss: 3.6238861083984375:  45%|████▌     | 2933/6470 [39:59<46:31,  1.27it/s]
1|2934|Loss: 3.6238861083984375:  45%|████▌     | 2934/6470 [39:59<44:34,  1.32it/s]
1|2935|Loss: 4.260143280029297:  45%|████▌     | 2934/6470 [39:59<44:34,  1.32it/s] 
1|2935|Loss: 4.260143280029297:  45%|████▌     | 2935/6470 [40:00<44:07,  1.34it/s]
1|2936|Loss: 3.759442090988159:  45%|████▌     | 2935/6470 [40:00<44:07,  1.34it/s]
1|2936|Loss: 3.759442090988159:  45%|████▌     | 2936/6470 [40:00<43:02,  1.37it/s]
1|2937|Loss: 3.9461469650268555:  45%|████▌     | 2936/6470 [40:01<43:02,  1.37it/s]
1|2937|Loss: 3.9461469650268555:  45%|████▌     | 2937/6470 [40:01<41:12,  1.43it/s]
1|2938|Loss: 3.267303228378296:  45%|████▌     | 2937/6470 [40:01<41:12,  1.43it/s] 
1|2938|Loss: 3.267303228378296:  45%|████▌     | 2938/6470 [40:02<38:39,  1.52it/s]
1|2939|Loss: 3.618356704711914:  45%|████▌     | 2938/6470 [40:02<38:39,  1.52it/s]
1|2939|Loss: 3.618356704711914:  45%|████▌     | 2939/6470 [40:02<37:12,  1.58it/s]
1|2940|Loss: 4.299381732940674:  45%|████▌     | 2939/6470 [40:03<37:12,  1.58it/s]
1|2940|Loss: 4.299381732940674:  45%|████▌     | 2940/6470 [40:03<40:40,  1.45it/s]
1|2941|Loss: 3.6016578674316406:  45%|████▌     | 2940/6470 [40:03<40:40,  1.45it/s]
1|2941|Loss: 3.6016578674316406:  45%|████▌     | 2941/6470 [40:04<42:07,  1.40it/s]
1|2942|Loss: 4.653281211853027:  45%|████▌     | 2941/6470 [40:04<42:07,  1.40it/s] 
1|2942|Loss: 4.653281211853027:  45%|████▌     | 2942/6470 [40:05<45:09,  1.30it/s]
1|2943|Loss: 4.480607986450195:  45%|████▌     | 2942/6470 [40:05<45:09,  1.30it/s]
1|2943|Loss: 4.480607986450195:  45%|████▌     | 2943/6470 [40:06<48:25,  1.21it/s]
1|2944|Loss: 4.37150239944458:  45%|████▌     | 2943/6470 [40:06<48:25,  1.21it/s] 
1|2944|Loss: 4.37150239944458:  46%|████▌     | 2944/6470 [40:06<49:16,  1.19it/s]
1|2945|Loss: 3.947133779525757:  46%|████▌     | 2944/6470 [40:07<49:16,  1.19it/s]
1|2945|Loss: 3.947133779525757:  46%|████▌     | 2945/6470 [40:07<45:38,  1.29it/s]
1|2946|Loss: 4.258351802825928:  46%|████▌     | 2945/6470 [40:07<45:38,  1.29it/s]
1|2946|Loss: 4.258351802825928:  46%|████▌     | 2946/6470 [40:08<42:39,  1.38it/s]
1|2947|Loss: 3.687375545501709:  46%|████▌     | 2946/6470 [40:08<42:39,  1.38it/s]
1|2947|Loss: 3.687375545501709:  46%|████▌     | 2947/6470 [40:08<40:46,  1.44it/s]
1|2948|Loss: 4.012365818023682:  46%|████▌     | 2947/6470 [40:09<40:46,  1.44it/s]
1|2948|Loss: 4.012365818023682:  46%|████▌     | 2948/6470 [40:09<41:30,  1.41it/s]
1|2949|Loss: 4.1874680519104:  46%|████▌     | 2948/6470 [40:09<41:30,  1.41it/s]  
1|2949|Loss: 4.1874680519104:  46%|████▌     | 2949/6470 [40:10<42:59,  1.37it/s]
1|2950|Loss: 3.7305471897125244:  46%|████▌     | 2949/6470 [40:10<42:59,  1.37it/s]
1|2950|Loss: 3.7305471897125244:  46%|████▌     | 2950/6470 [40:11<44:13,  1.33it/s]
1|2951|Loss: 3.9882750511169434:  46%|████▌     | 2950/6470 [40:11<44:13,  1.33it/s]
1|2951|Loss: 3.9882750511169434:  46%|████▌     | 2951/6470 [40:11<45:02,  1.30it/s]
1|2952|Loss: 3.8329365253448486:  46%|████▌     | 2951/6470 [40:12<45:02,  1.30it/s]
1|2952|Loss: 3.8329365253448486:  46%|████▌     | 2952/6470 [40:12<44:51,  1.31it/s]
1|2953|Loss: 3.978912591934204:  46%|████▌     | 2952/6470 [40:13<44:51,  1.31it/s] 
1|2953|Loss: 3.978912591934204:  46%|████▌     | 2953/6470 [40:13<44:44,  1.31it/s]
1|2954|Loss: 4.152771472930908:  46%|████▌     | 2953/6470 [40:13<44:44,  1.31it/s]
1|2954|Loss: 4.152771472930908:  46%|████▌     | 2954/6470 [40:14<46:42,  1.25it/s]
1|2955|Loss: 3.2599408626556396:  46%|████▌     | 2954/6470 [40:14<46:42,  1.25it/s]
1|2955|Loss: 3.2599408626556396:  46%|████▌     | 2955/6470 [40:14<41:12,  1.42it/s]
1|2956|Loss: 3.93571400642395:  46%|████▌     | 2955/6470 [40:15<41:12,  1.42it/s]  
1|2956|Loss: 3.93571400642395:  46%|████▌     | 2956/6470 [40:15<38:40,  1.51it/s]
1|2957|Loss: 4.138517379760742:  46%|████▌     | 2956/6470 [40:15<38:40,  1.51it/s]
1|2957|Loss: 4.138517379760742:  46%|████▌     | 2957/6470 [40:16<41:35,  1.41it/s]
1|2958|Loss: 4.412384033203125:  46%|████▌     | 2957/6470 [40:16<41:35,  1.41it/s]
1|2958|Loss: 4.412384033203125:  46%|████▌     | 2958/6470 [40:17<43:01,  1.36it/s]
1|2959|Loss: 3.421497106552124:  46%|████▌     | 2958/6470 [40:17<43:01,  1.36it/s]
1|2959|Loss: 3.421497106552124:  46%|████▌     | 2959/6470 [40:17<40:31,  1.44it/s]
1|2960|Loss: 4.133057594299316:  46%|████▌     | 2959/6470 [40:17<40:31,  1.44it/s]
1|2960|Loss: 4.133057594299316:  46%|████▌     | 2960/6470 [40:18<39:23,  1.48it/s]
1|2961|Loss: 4.061723232269287:  46%|████▌     | 2960/6470 [40:18<39:23,  1.48it/s]
1|2961|Loss: 4.061723232269287:  46%|████▌     | 2961/6470 [40:18<38:45,  1.51it/s]
1|2962|Loss: 4.135809421539307:  46%|████▌     | 2961/6470 [40:19<38:45,  1.51it/s]
1|2962|Loss: 4.135809421539307:  46%|████▌     | 2962/6470 [40:19<39:58,  1.46it/s]
1|2963|Loss: 3.900161027908325:  46%|████▌     | 2962/6470 [40:20<39:58,  1.46it/s]
1|2963|Loss: 3.900161027908325:  46%|████▌     | 2963/6470 [40:20<40:35,  1.44it/s]
1|2964|Loss: 4.315924644470215:  46%|████▌     | 2963/6470 [40:20<40:35,  1.44it/s]
1|2964|Loss: 4.315924644470215:  46%|████▌     | 2964/6470 [40:21<40:21,  1.45it/s]
1|2965|Loss: 4.045265197753906:  46%|████▌     | 2964/6470 [40:21<40:21,  1.45it/s]
1|2965|Loss: 4.045265197753906:  46%|████▌     | 2965/6470 [40:21<40:29,  1.44it/s]
1|2966|Loss: 4.32642936706543:  46%|████▌     | 2965/6470 [40:22<40:29,  1.44it/s] 
1|2966|Loss: 4.32642936706543:  46%|████▌     | 2966/6470 [40:22<41:57,  1.39it/s]
1|2967|Loss: 3.587770938873291:  46%|████▌     | 2966/6470 [40:22<41:57,  1.39it/s]
1|2967|Loss: 3.587770938873291:  46%|████▌     | 2967/6470 [40:22<37:47,  1.54it/s]
1|2968|Loss: 4.1513285636901855:  46%|████▌     | 2967/6470 [40:23<37:47,  1.54it/s]
1|2968|Loss: 4.1513285636901855:  46%|████▌     | 2968/6470 [40:23<41:27,  1.41it/s]
1|2969|Loss: 4.2218828201293945:  46%|████▌     | 2968/6470 [40:24<41:27,  1.41it/s]
1|2969|Loss: 4.2218828201293945:  46%|████▌     | 2969/6470 [40:24<43:12,  1.35it/s]
1|2970|Loss: 4.14213752746582:  46%|████▌     | 2969/6470 [40:25<43:12,  1.35it/s]  
1|2970|Loss: 4.14213752746582:  46%|████▌     | 2970/6470 [40:25<44:21,  1.31it/s]
1|2971|Loss: 4.258807182312012:  46%|████▌     | 2970/6470 [40:25<44:21,  1.31it/s]
1|2971|Loss: 4.258807182312012:  46%|████▌     | 2971/6470 [40:26<45:12,  1.29it/s]
1|2972|Loss: 3.355368137359619:  46%|████▌     | 2971/6470 [40:26<45:12,  1.29it/s]
1|2972|Loss: 3.355368137359619:  46%|████▌     | 2972/6470 [40:26<43:56,  1.33it/s]
1|2973|Loss: 3.877678155899048:  46%|████▌     | 2972/6470 [40:27<43:56,  1.33it/s]
1|2973|Loss: 3.877678155899048:  46%|████▌     | 2973/6470 [40:27<44:14,  1.32it/s]
1|2974|Loss: 3.806070327758789:  46%|████▌     | 2973/6470 [40:28<44:14,  1.32it/s]
1|2974|Loss: 3.806070327758789:  46%|████▌     | 2974/6470 [40:28<43:04,  1.35it/s]
1|2975|Loss: 3.7013449668884277:  46%|████▌     | 2974/6470 [40:28<43:04,  1.35it/s]
1|2975|Loss: 3.7013449668884277:  46%|████▌     | 2975/6470 [40:29<42:59,  1.35it/s]
1|2976|Loss: 4.162820339202881:  46%|████▌     | 2975/6470 [40:29<42:59,  1.35it/s] 
1|2976|Loss: 4.162820339202881:  46%|████▌     | 2976/6470 [40:29<43:49,  1.33it/s]
1|2977|Loss: 3.9557511806488037:  46%|████▌     | 2976/6470 [40:30<43:49,  1.33it/s]
1|2977|Loss: 3.9557511806488037:  46%|████▌     | 2977/6470 [40:30<43:15,  1.35it/s]
1|2978|Loss: 3.8783137798309326:  46%|████▌     | 2977/6470 [40:31<43:15,  1.35it/s]
1|2978|Loss: 3.8783137798309326:  46%|████▌     | 2978/6470 [40:31<43:42,  1.33it/s]
1|2979|Loss: 4.130062103271484:  46%|████▌     | 2978/6470 [40:31<43:42,  1.33it/s] 
1|2979|Loss: 4.130062103271484:  46%|████▌     | 2979/6470 [40:32<46:22,  1.25it/s]
1|2980|Loss: 4.119286060333252:  46%|████▌     | 2979/6470 [40:32<46:22,  1.25it/s]
1|2980|Loss: 4.119286060333252:  46%|████▌     | 2980/6470 [40:33<45:47,  1.27it/s]
1|2981|Loss: 3.1179301738739014:  46%|████▌     | 2980/6470 [40:33<45:47,  1.27it/s]
1|2981|Loss: 3.1179301738739014:  46%|████▌     | 2981/6470 [40:33<41:11,  1.41it/s]
1|2982|Loss: 3.4935238361358643:  46%|████▌     | 2981/6470 [40:33<41:11,  1.41it/s]
1|2982|Loss: 3.4935238361358643:  46%|████▌     | 2982/6470 [40:34<41:19,  1.41it/s]
1|2983|Loss: 4.228529453277588:  46%|████▌     | 2982/6470 [40:34<41:19,  1.41it/s] 
1|2983|Loss: 4.228529453277588:  46%|████▌     | 2983/6470 [40:35<42:47,  1.36it/s]
1|2984|Loss: 3.7091801166534424:  46%|████▌     | 2983/6470 [40:35<42:47,  1.36it/s]
1|2984|Loss: 3.7091801166534424:  46%|████▌     | 2984/6470 [40:35<43:52,  1.32it/s]
1|2985|Loss: 4.335305213928223:  46%|████▌     | 2984/6470 [40:36<43:52,  1.32it/s] 
1|2985|Loss: 4.335305213928223:  46%|████▌     | 2985/6470 [40:36<44:55,  1.29it/s]
1|2986|Loss: 4.125843048095703:  46%|████▌     | 2985/6470 [40:37<44:55,  1.29it/s]
1|2986|Loss: 4.125843048095703:  46%|████▌     | 2986/6470 [40:37<47:17,  1.23it/s]
1|2987|Loss: 4.203516483306885:  46%|████▌     | 2986/6470 [40:38<47:17,  1.23it/s]
1|2987|Loss: 4.203516483306885:  46%|████▌     | 2987/6470 [40:38<46:25,  1.25it/s]
1|2988|Loss: 3.140625:  46%|████▌     | 2987/6470 [40:38<46:25,  1.25it/s]         
1|2988|Loss: 3.140625:  46%|████▌     | 2988/6470 [40:39<42:34,  1.36it/s]
1|2989|Loss: 3.9845871925354004:  46%|████▌     | 2988/6470 [40:39<42:34,  1.36it/s]
1|2989|Loss: 3.9845871925354004:  46%|████▌     | 2989/6470 [40:39<41:22,  1.40it/s]
1|2990|Loss: 3.269221067428589:  46%|████▌     | 2989/6470 [40:40<41:22,  1.40it/s] 
1|2990|Loss: 3.269221067428589:  46%|████▌     | 2990/6470 [40:40<40:13,  1.44it/s]
1|2991|Loss: 3.645092248916626:  46%|████▌     | 2990/6470 [40:40<40:13,  1.44it/s]
1|2991|Loss: 3.645092248916626:  46%|████▌     | 2991/6470 [40:40<39:34,  1.47it/s]
1|2992|Loss: 4.294040679931641:  46%|████▌     | 2991/6470 [40:41<39:34,  1.47it/s]
1|2992|Loss: 4.294040679931641:  46%|████▌     | 2992/6470 [40:41<41:29,  1.40it/s]
1|2993|Loss: 4.115614891052246:  46%|████▌     | 2992/6470 [40:42<41:29,  1.40it/s]
1|2993|Loss: 4.115614891052246:  46%|████▋     | 2993/6470 [40:42<41:16,  1.40it/s]
1|2994|Loss: 4.571591377258301:  46%|████▋     | 2993/6470 [40:42<41:16,  1.40it/s]
1|2994|Loss: 4.571591377258301:  46%|████▋     | 2994/6470 [40:43<43:59,  1.32it/s]
1|2995|Loss: 3.6613235473632812:  46%|████▋     | 2994/6470 [40:43<43:59,  1.32it/s]
1|2995|Loss: 3.6613235473632812:  46%|████▋     | 2995/6470 [40:43<41:33,  1.39it/s]
1|2996|Loss: 4.309657096862793:  46%|████▋     | 2995/6470 [40:44<41:33,  1.39it/s] 
1|2996|Loss: 4.309657096862793:  46%|████▋     | 2996/6470 [40:44<42:05,  1.38it/s]
1|2997|Loss: 4.085971832275391:  46%|████▋     | 2996/6470 [40:45<42:05,  1.38it/s]
1|2997|Loss: 4.085971832275391:  46%|████▋     | 2997/6470 [40:45<43:20,  1.34it/s]
1|2998|Loss: 3.8447985649108887:  46%|████▋     | 2997/6470 [40:45<43:20,  1.34it/s]
1|2998|Loss: 3.8447985649108887:  46%|████▋     | 2998/6470 [40:46<43:00,  1.35it/s]
1|2999|Loss: 4.146988868713379:  46%|████▋     | 2998/6470 [40:46<43:00,  1.35it/s] 
1|2999|Loss: 4.146988868713379:  46%|████▋     | 2999/6470 [40:46<41:58,  1.38it/s]
1|3000|Loss: 4.394527912139893:  46%|████▋     | 2999/6470 [40:47<41:58,  1.38it/s]
1|3000|Loss: 4.394527912139893:  46%|████▋     | 3000/6470 [40:47<42:31,  1.36it/s]
1|3001|Loss: 5.05198335647583:  46%|████▋     | 3000/6470 [40:48<42:31,  1.36it/s] 
1|3001|Loss: 5.05198335647583:  46%|████▋     | 3001/6470 [40:48<45:16,  1.28it/s]
1|3002|Loss: 3.9356703758239746:  46%|████▋     | 3001/6470 [40:49<45:16,  1.28it/s]
1|3002|Loss: 3.9356703758239746:  46%|████▋     | 3002/6470 [40:49<44:44,  1.29it/s]
1|3003|Loss: 3.5232646465301514:  46%|████▋     | 3002/6470 [40:49<44:44,  1.29it/s]
1|3003|Loss: 3.5232646465301514:  46%|████▋     | 3003/6470 [40:50<43:06,  1.34it/s]
1|3004|Loss: 3.623760461807251:  46%|████▋     | 3003/6470 [40:50<43:06,  1.34it/s] 
1|3004|Loss: 3.623760461807251:  46%|████▋     | 3004/6470 [40:50<41:18,  1.40it/s]
1|3005|Loss: 3.1440279483795166:  46%|████▋     | 3004/6470 [40:50<41:18,  1.40it/s]
1|3005|Loss: 3.1440279483795166:  46%|████▋     | 3005/6470 [40:51<38:08,  1.51it/s]
1|3006|Loss: 4.063755512237549:  46%|████▋     | 3005/6470 [40:51<38:08,  1.51it/s] 
1|3006|Loss: 4.063755512237549:  46%|████▋     | 3006/6470 [40:51<38:48,  1.49it/s]
1|3007|Loss: 2.4578869342803955:  46%|████▋     | 3006/6470 [40:52<38:48,  1.49it/s]
1|3007|Loss: 2.4578869342803955:  46%|████▋     | 3007/6470 [40:52<35:06,  1.64it/s]
1|3008|Loss: 3.840250253677368:  46%|████▋     | 3007/6470 [40:52<35:06,  1.64it/s] 
1|3008|Loss: 3.840250253677368:  46%|████▋     | 3008/6470 [40:52<35:38,  1.62it/s]
1|3009|Loss: 4.047394275665283:  46%|████▋     | 3008/6470 [40:53<35:38,  1.62it/s]
1|3009|Loss: 4.047394275665283:  47%|████▋     | 3009/6470 [40:53<39:58,  1.44it/s]
1|3010|Loss: 4.221040725708008:  47%|████▋     | 3009/6470 [40:54<39:58,  1.44it/s]
1|3010|Loss: 4.221040725708008:  47%|████▋     | 3010/6470 [40:54<42:06,  1.37it/s]
1|3011|Loss: 3.897519111633301:  47%|████▋     | 3010/6470 [40:55<42:06,  1.37it/s]
1|3011|Loss: 3.897519111633301:  47%|████▋     | 3011/6470 [40:55<44:09,  1.31it/s]
1|3012|Loss: 3.1652886867523193:  47%|████▋     | 3011/6470 [40:55<44:09,  1.31it/s]
1|3012|Loss: 3.1652886867523193:  47%|████▋     | 3012/6470 [40:56<43:07,  1.34it/s]
1|3013|Loss: 3.965542793273926:  47%|████▋     | 3012/6470 [40:56<43:07,  1.34it/s] 
1|3013|Loss: 3.965542793273926:  47%|████▋     | 3013/6470 [40:57<46:00,  1.25it/s]
1|3014|Loss: 4.073278427124023:  47%|████▋     | 3013/6470 [40:57<46:00,  1.25it/s]
1|3014|Loss: 4.073278427124023:  47%|████▋     | 3014/6470 [40:57<46:43,  1.23it/s]
1|3015|Loss: 4.039538860321045:  47%|████▋     | 3014/6470 [40:58<46:43,  1.23it/s]
1|3015|Loss: 4.039538860321045:  47%|████▋     | 3015/6470 [40:58<44:18,  1.30it/s]
1|3016|Loss: 3.645073652267456:  47%|████▋     | 3015/6470 [40:59<44:18,  1.30it/s]
1|3016|Loss: 3.645073652267456:  47%|████▋     | 3016/6470 [40:59<43:38,  1.32it/s]
1|3017|Loss: 3.4187111854553223:  47%|████▋     | 3016/6470 [40:59<43:38,  1.32it/s]
1|3017|Loss: 3.4187111854553223:  47%|████▋     | 3017/6470 [41:00<42:26,  1.36it/s]
1|3018|Loss: 3.7657432556152344:  47%|████▋     | 3017/6470 [41:00<42:26,  1.36it/s]
1|3018|Loss: 3.7657432556152344:  47%|████▋     | 3018/6470 [41:00<41:12,  1.40it/s]
1|3019|Loss: 3.857002019882202:  47%|████▋     | 3018/6470 [41:01<41:12,  1.40it/s] 
1|3019|Loss: 3.857002019882202:  47%|████▋     | 3019/6470 [41:01<39:44,  1.45it/s]
1|3020|Loss: 3.7261550426483154:  47%|████▋     | 3019/6470 [41:01<39:44,  1.45it/s]
1|3020|Loss: 3.7261550426483154:  47%|████▋     | 3020/6470 [41:02<40:59,  1.40it/s]
1|3021|Loss: 4.248289108276367:  47%|████▋     | 3020/6470 [41:02<40:59,  1.40it/s] 
1|3021|Loss: 4.248289108276367:  47%|████▋     | 3021/6470 [41:02<42:48,  1.34it/s]
1|3022|Loss: 4.208076477050781:  47%|████▋     | 3021/6470 [41:03<42:48,  1.34it/s]
1|3022|Loss: 4.208076477050781:  47%|████▋     | 3022/6470 [41:03<44:01,  1.31it/s]
1|3023|Loss: 4.081170082092285:  47%|████▋     | 3022/6470 [41:04<44:01,  1.31it/s]
1|3023|Loss: 4.081170082092285:  47%|████▋     | 3023/6470 [41:04<44:15,  1.30it/s]
1|3024|Loss: 4.047793865203857:  47%|████▋     | 3023/6470 [41:04<44:15,  1.30it/s]
1|3024|Loss: 4.047793865203857:  47%|████▋     | 3024/6470 [41:05<44:29,  1.29it/s]
1|3025|Loss: 3.3160760402679443:  47%|████▋     | 3024/6470 [41:05<44:29,  1.29it/s]
1|3025|Loss: 3.3160760402679443:  47%|████▋     | 3025/6470 [41:06<42:42,  1.34it/s]
1|3026|Loss: 3.7205636501312256:  47%|████▋     | 3025/6470 [41:06<42:42,  1.34it/s]
1|3026|Loss: 3.7205636501312256:  47%|████▋     | 3026/6470 [41:06<39:26,  1.46it/s]
1|3027|Loss: 4.12637186050415:  47%|████▋     | 3026/6470 [41:06<39:26,  1.46it/s]  
1|3027|Loss: 4.12637186050415:  47%|████▋     | 3027/6470 [41:07<40:48,  1.41it/s]
1|3028|Loss: 3.7384748458862305:  47%|████▋     | 3027/6470 [41:07<40:48,  1.41it/s]
1|3028|Loss: 3.7384748458862305:  47%|████▋     | 3028/6470 [41:08<42:11,  1.36it/s]
1|3029|Loss: 3.737619638442993:  47%|████▋     | 3028/6470 [41:08<42:11,  1.36it/s] 
1|3029|Loss: 3.737619638442993:  47%|████▋     | 3029/6470 [41:08<41:32,  1.38it/s]
1|3030|Loss: 3.5418429374694824:  47%|████▋     | 3029/6470 [41:09<41:32,  1.38it/s]
1|3030|Loss: 3.5418429374694824:  47%|████▋     | 3030/6470 [41:09<39:00,  1.47it/s]
1|3031|Loss: 4.108321666717529:  47%|████▋     | 3030/6470 [41:09<39:00,  1.47it/s] 
1|3031|Loss: 4.108321666717529:  47%|████▋     | 3031/6470 [41:10<40:14,  1.42it/s]
1|3032|Loss: 4.082006931304932:  47%|████▋     | 3031/6470 [41:10<40:14,  1.42it/s]
1|3032|Loss: 4.082006931304932:  47%|████▋     | 3032/6470 [41:10<41:21,  1.39it/s]
1|3033|Loss: 4.271818161010742:  47%|████▋     | 3032/6470 [41:11<41:21,  1.39it/s]
1|3033|Loss: 4.271818161010742:  47%|████▋     | 3033/6470 [41:11<43:32,  1.32it/s]
1|3034|Loss: 4.4825758934021:  47%|████▋     | 3033/6470 [41:12<43:32,  1.32it/s]  
1|3034|Loss: 4.4825758934021:  47%|████▋     | 3034/6470 [41:12<44:03,  1.30it/s]
1|3035|Loss: 3.3975837230682373:  47%|████▋     | 3034/6470 [41:12<44:03,  1.30it/s]
1|3035|Loss: 3.3975837230682373:  47%|████▋     | 3035/6470 [41:13<42:58,  1.33it/s]
1|3036|Loss: 3.9593493938446045:  47%|████▋     | 3035/6470 [41:13<42:58,  1.33it/s]
1|3036|Loss: 3.9593493938446045:  47%|████▋     | 3036/6470 [41:14<42:53,  1.33it/s]
1|3037|Loss: 2.7600173950195312:  47%|████▋     | 3036/6470 [41:14<42:53,  1.33it/s]
1|3037|Loss: 2.7600173950195312:  47%|████▋     | 3037/6470 [41:14<39:06,  1.46it/s]
1|3038|Loss: 3.4812123775482178:  47%|████▋     | 3037/6470 [41:14<39:06,  1.46it/s]
1|3038|Loss: 3.4812123775482178:  47%|████▋     | 3038/6470 [41:15<36:01,  1.59it/s]
1|3039|Loss: 4.142998218536377:  47%|████▋     | 3038/6470 [41:15<36:01,  1.59it/s] 
1|3039|Loss: 4.142998218536377:  47%|████▋     | 3039/6470 [41:15<36:35,  1.56it/s]
1|3040|Loss: 3.616858959197998:  47%|████▋     | 3039/6470 [41:16<36:35,  1.56it/s]
1|3040|Loss: 3.616858959197998:  47%|████▋     | 3040/6470 [41:16<38:53,  1.47it/s]
1|3041|Loss: 3.728574275970459:  47%|████▋     | 3040/6470 [41:16<38:53,  1.47it/s]
1|3041|Loss: 3.728574275970459:  47%|████▋     | 3041/6470 [41:17<40:41,  1.40it/s]
1|3042|Loss: 3.8020012378692627:  47%|████▋     | 3041/6470 [41:17<40:41,  1.40it/s]
1|3042|Loss: 3.8020012378692627:  47%|████▋     | 3042/6470 [41:18<42:03,  1.36it/s]
1|3043|Loss: 3.7527496814727783:  47%|████▋     | 3042/6470 [41:18<42:03,  1.36it/s]
1|3043|Loss: 3.7527496814727783:  47%|████▋     | 3043/6470 [41:18<42:34,  1.34it/s]
1|3044|Loss: 4.700105667114258:  47%|████▋     | 3043/6470 [41:19<42:34,  1.34it/s] 
1|3044|Loss: 4.700105667114258:  47%|████▋     | 3044/6470 [41:19<45:25,  1.26it/s]
1|3045|Loss: 4.210269927978516:  47%|████▋     | 3044/6470 [41:20<45:25,  1.26it/s]
1|3045|Loss: 4.210269927978516:  47%|████▋     | 3045/6470 [41:20<47:50,  1.19it/s]
1|3046|Loss: 3.1325674057006836:  47%|████▋     | 3045/6470 [41:21<47:50,  1.19it/s]
1|3046|Loss: 3.1325674057006836:  47%|████▋     | 3046/6470 [41:21<43:27,  1.31it/s]
1|3047|Loss: 4.066211700439453:  47%|████▋     | 3046/6470 [41:21<43:27,  1.31it/s] 
1|3047|Loss: 4.066211700439453:  47%|████▋     | 3047/6470 [41:21<40:27,  1.41it/s]
1|3048|Loss: 3.7754275798797607:  47%|████▋     | 3047/6470 [41:22<40:27,  1.41it/s]
1|3048|Loss: 3.7754275798797607:  47%|████▋     | 3048/6470 [41:22<41:35,  1.37it/s]
1|3049|Loss: 4.193380832672119:  47%|████▋     | 3048/6470 [41:23<41:35,  1.37it/s] 
1|3049|Loss: 4.193380832672119:  47%|████▋     | 3049/6470 [41:23<44:48,  1.27it/s]
1|3050|Loss: 3.800774335861206:  47%|████▋     | 3049/6470 [41:24<44:48,  1.27it/s]
1|3050|Loss: 3.800774335861206:  47%|████▋     | 3050/6470 [41:24<46:31,  1.23it/s]
1|3051|Loss: 4.255690097808838:  47%|████▋     | 3050/6470 [41:24<46:31,  1.23it/s]
1|3051|Loss: 4.255690097808838:  47%|████▋     | 3051/6470 [41:25<47:42,  1.19it/s]
1|3052|Loss: 3.222426652908325:  47%|████▋     | 3051/6470 [41:25<47:42,  1.19it/s]
1|3052|Loss: 3.222426652908325:  47%|████▋     | 3052/6470 [41:26<44:55,  1.27it/s]
1|3053|Loss: 3.6047353744506836:  47%|████▋     | 3052/6470 [41:26<44:55,  1.27it/s]
1|3053|Loss: 3.6047353744506836:  47%|████▋     | 3053/6470 [41:26<43:56,  1.30it/s]
1|3054|Loss: 3.8287010192871094:  47%|████▋     | 3053/6470 [41:27<43:56,  1.30it/s]
1|3054|Loss: 3.8287010192871094:  47%|████▋     | 3054/6470 [41:27<42:33,  1.34it/s]
1|3055|Loss: 4.1494574546813965:  47%|████▋     | 3054/6470 [41:27<42:33,  1.34it/s]
1|3055|Loss: 4.1494574546813965:  47%|████▋     | 3055/6470 [41:28<44:26,  1.28it/s]
1|3056|Loss: 3.769821882247925:  47%|████▋     | 3055/6470 [41:28<44:26,  1.28it/s] 
1|3056|Loss: 3.769821882247925:  47%|████▋     | 3056/6470 [41:29<44:47,  1.27it/s]
1|3057|Loss: 3.5802505016326904:  47%|████▋     | 3056/6470 [41:29<44:47,  1.27it/s]
1|3057|Loss: 3.5802505016326904:  47%|████▋     | 3057/6470 [41:29<45:35,  1.25it/s]
1|3058|Loss: 4.37013053894043:  47%|████▋     | 3057/6470 [41:30<45:35,  1.25it/s]  
1|3058|Loss: 4.37013053894043:  47%|████▋     | 3058/6470 [41:30<47:46,  1.19it/s]
1|3059|Loss: 4.101025581359863:  47%|████▋     | 3058/6470 [41:31<47:46,  1.19it/s]
1|3059|Loss: 4.101025581359863:  47%|████▋     | 3059/6470 [41:31<49:24,  1.15it/s]
1|3060|Loss: 4.088815689086914:  47%|████▋     | 3059/6470 [41:32<49:24,  1.15it/s]
1|3060|Loss: 4.088815689086914:  47%|████▋     | 3060/6470 [41:32<48:46,  1.17it/s]
1|3061|Loss: 3.308730125427246:  47%|████▋     | 3060/6470 [41:33<48:46,  1.17it/s]
1|3061|Loss: 3.308730125427246:  47%|████▋     | 3061/6470 [41:33<45:39,  1.24it/s]
1|3062|Loss: 3.201869249343872:  47%|████▋     | 3061/6470 [41:33<45:39,  1.24it/s]
1|3062|Loss: 3.201869249343872:  47%|████▋     | 3062/6470 [41:33<43:39,  1.30it/s]
1|3063|Loss: 3.220839738845825:  47%|████▋     | 3062/6470 [41:34<43:39,  1.30it/s]
1|3063|Loss: 3.220839738845825:  47%|████▋     | 3063/6470 [41:34<42:40,  1.33it/s]
1|3064|Loss: 3.5573554039001465:  47%|████▋     | 3063/6470 [41:35<42:40,  1.33it/s]
1|3064|Loss: 3.5573554039001465:  47%|████▋     | 3064/6470 [41:35<44:07,  1.29it/s]
1|3065|Loss: 2.5771403312683105:  47%|████▋     | 3064/6470 [41:35<44:07,  1.29it/s]
1|3065|Loss: 2.5771403312683105:  47%|████▋     | 3065/6470 [41:36<40:16,  1.41it/s]
1|3066|Loss: 4.250051021575928:  47%|████▋     | 3065/6470 [41:36<40:16,  1.41it/s] 
1|3066|Loss: 4.250051021575928:  47%|████▋     | 3066/6470 [41:36<42:48,  1.33it/s]
1|3067|Loss: 3.6613845825195312:  47%|████▋     | 3066/6470 [41:37<42:48,  1.33it/s]
1|3067|Loss: 3.6613845825195312:  47%|████▋     | 3067/6470 [41:37<40:58,  1.38it/s]
1|3068|Loss: 3.8609297275543213:  47%|████▋     | 3067/6470 [41:37<40:58,  1.38it/s]
1|3068|Loss: 3.8609297275543213:  47%|████▋     | 3068/6470 [41:38<41:35,  1.36it/s]
1|3069|Loss: 3.3782896995544434:  47%|████▋     | 3068/6470 [41:38<41:35,  1.36it/s]
1|3069|Loss: 3.3782896995544434:  47%|████▋     | 3069/6470 [41:39<42:12,  1.34it/s]
1|3070|Loss: 3.230926513671875:  47%|████▋     | 3069/6470 [41:39<42:12,  1.34it/s] 
1|3070|Loss: 3.230926513671875:  47%|████▋     | 3070/6470 [41:39<39:36,  1.43it/s]
1|3071|Loss: 3.0833423137664795:  47%|████▋     | 3070/6470 [41:40<39:36,  1.43it/s]
1|3071|Loss: 3.0833423137664795:  47%|████▋     | 3071/6470 [41:40<37:36,  1.51it/s]
1|3072|Loss: 3.7027456760406494:  47%|████▋     | 3071/6470 [41:40<37:36,  1.51it/s]
1|3072|Loss: 3.7027456760406494:  47%|████▋     | 3072/6470 [41:41<38:41,  1.46it/s]
1|3073|Loss: 3.5827744007110596:  47%|████▋     | 3072/6470 [41:41<38:41,  1.46it/s]
1|3073|Loss: 3.5827744007110596:  47%|████▋     | 3073/6470 [41:41<38:52,  1.46it/s]
1|3074|Loss: 4.592604160308838:  47%|████▋     | 3073/6470 [41:42<38:52,  1.46it/s] 
1|3074|Loss: 4.592604160308838:  48%|████▊     | 3074/6470 [41:42<39:07,  1.45it/s]
1|3075|Loss: 4.971042156219482:  48%|████▊     | 3074/6470 [41:42<39:07,  1.45it/s]
1|3075|Loss: 4.971042156219482:  48%|████▊     | 3075/6470 [41:43<44:44,  1.26it/s]
1|3076|Loss: 4.219371318817139:  48%|████▊     | 3075/6470 [41:43<44:44,  1.26it/s]
1|3076|Loss: 4.219371318817139:  48%|████▊     | 3076/6470 [41:44<45:39,  1.24it/s]
1|3077|Loss: 3.819953680038452:  48%|████▊     | 3076/6470 [41:44<45:39,  1.24it/s]
1|3077|Loss: 3.819953680038452:  48%|████▊     | 3077/6470 [41:45<44:15,  1.28it/s]
1|3078|Loss: 3.940828561782837:  48%|████▊     | 3077/6470 [41:45<44:15,  1.28it/s]
1|3078|Loss: 3.940828561782837:  48%|████▊     | 3078/6470 [41:45<45:23,  1.25it/s]
1|3079|Loss: 3.2905731201171875:  48%|████▊     | 3078/6470 [41:46<45:23,  1.25it/s]
1|3079|Loss: 3.2905731201171875:  48%|████▊     | 3079/6470 [41:46<47:07,  1.20it/s]
1|3080|Loss: 3.220816135406494:  48%|████▊     | 3079/6470 [41:47<47:07,  1.20it/s] 
1|3080|Loss: 3.220816135406494:  48%|████▊     | 3080/6470 [41:47<45:48,  1.23it/s]
1|3081|Loss: 3.975142478942871:  48%|████▊     | 3080/6470 [41:47<45:48,  1.23it/s]
1|3081|Loss: 3.975142478942871:  48%|████▊     | 3081/6470 [41:48<45:30,  1.24it/s]
1|3082|Loss: 3.272881269454956:  48%|████▊     | 3081/6470 [41:48<45:30,  1.24it/s]
1|3082|Loss: 3.272881269454956:  48%|████▊     | 3082/6470 [41:49<46:32,  1.21it/s]
1|3083|Loss: 3.9188151359558105:  48%|████▊     | 3082/6470 [41:49<46:32,  1.21it/s]
1|3083|Loss: 3.9188151359558105:  48%|████▊     | 3083/6470 [41:50<47:16,  1.19it/s]
1|3084|Loss: 3.6779966354370117:  48%|████▊     | 3083/6470 [41:50<47:16,  1.19it/s]
1|3084|Loss: 3.6779966354370117:  48%|████▊     | 3084/6470 [41:50<48:26,  1.16it/s]
1|3085|Loss: 3.9859440326690674:  48%|████▊     | 3084/6470 [41:51<48:26,  1.16it/s]
1|3085|Loss: 3.9859440326690674:  48%|████▊     | 3085/6470 [41:51<48:11,  1.17it/s]
1|3086|Loss: 3.1240150928497314:  48%|████▊     | 3085/6470 [41:52<48:11,  1.17it/s]
1|3086|Loss: 3.1240150928497314:  48%|████▊     | 3086/6470 [41:52<49:08,  1.15it/s]
1|3087|Loss: 2.9983086585998535:  48%|████▊     | 3086/6470 [41:53<49:08,  1.15it/s]
1|3087|Loss: 2.9983086585998535:  48%|████▊     | 3087/6470 [41:53<48:35,  1.16it/s]
1|3088|Loss: 3.348698377609253:  48%|████▊     | 3087/6470 [41:53<48:35,  1.16it/s] 
1|3088|Loss: 3.348698377609253:  48%|████▊     | 3088/6470 [41:54<45:16,  1.24it/s]
1|3089|Loss: 3.6158628463745117:  48%|████▊     | 3088/6470 [41:54<45:16,  1.24it/s]
1|3089|Loss: 3.6158628463745117:  48%|████▊     | 3089/6470 [41:54<44:19,  1.27it/s]
1|3090|Loss: 3.0417473316192627:  48%|████▊     | 3089/6470 [41:55<44:19,  1.27it/s]
1|3090|Loss: 3.0417473316192627:  48%|████▊     | 3090/6470 [41:55<41:48,  1.35it/s]
1|3091|Loss: 3.1071085929870605:  48%|████▊     | 3090/6470 [41:56<41:48,  1.35it/s]
1|3091|Loss: 3.1071085929870605:  48%|████▊     | 3091/6470 [41:56<43:20,  1.30it/s]
1|3092|Loss: 2.5318291187286377:  48%|████▊     | 3091/6470 [41:56<43:20,  1.30it/s]
1|3092|Loss: 2.5318291187286377:  48%|████▊     | 3092/6470 [41:57<42:42,  1.32it/s]
1|3093|Loss: 3.4122493267059326:  48%|████▊     | 3092/6470 [41:57<42:42,  1.32it/s]
1|3093|Loss: 3.4122493267059326:  48%|████▊     | 3093/6470 [41:57<43:06,  1.31it/s]
1|3094|Loss: 2.434474229812622:  48%|████▊     | 3093/6470 [41:58<43:06,  1.31it/s] 
1|3094|Loss: 2.434474229812622:  48%|████▊     | 3094/6470 [41:58<40:30,  1.39it/s]
1|3095|Loss: 2.962949752807617:  48%|████▊     | 3094/6470 [41:58<40:30,  1.39it/s]
1|3095|Loss: 2.962949752807617:  48%|████▊     | 3095/6470 [41:59<39:42,  1.42it/s]
1|3096|Loss: 2.7661678791046143:  48%|████▊     | 3095/6470 [41:59<39:42,  1.42it/s]
1|3096|Loss: 2.7661678791046143:  48%|████▊     | 3096/6470 [42:00<41:06,  1.37it/s]
1|3097|Loss: 2.788365602493286:  48%|████▊     | 3096/6470 [42:00<41:06,  1.37it/s] 
1|3097|Loss: 2.788365602493286:  48%|████▊     | 3097/6470 [42:00<40:01,  1.40it/s]
1|3098|Loss: 3.620835542678833:  48%|████▊     | 3097/6470 [42:01<40:01,  1.40it/s]
1|3098|Loss: 3.620835542678833:  48%|████▊     | 3098/6470 [42:01<40:26,  1.39it/s]
1|3099|Loss: 3.0089876651763916:  48%|████▊     | 3098/6470 [42:01<40:26,  1.39it/s]
1|3099|Loss: 3.0089876651763916:  48%|████▊     | 3099/6470 [42:02<42:28,  1.32it/s]
1|3100|Loss: 1.7413698434829712:  48%|████▊     | 3099/6470 [42:02<42:28,  1.32it/s]
1|3100|Loss: 1.7413698434829712:  48%|████▊     | 3100/6470 [42:02<38:05,  1.47it/s]
1|3101|Loss: 1.9211281538009644:  48%|████▊     | 3100/6470 [42:03<38:05,  1.47it/s]
1|3101|Loss: 1.9211281538009644:  48%|████▊     | 3101/6470 [42:03<35:21,  1.59it/s]
1|3102|Loss: 3.1344246864318848:  48%|████▊     | 3101/6470 [42:03<35:21,  1.59it/s]
1|3102|Loss: 3.1344246864318848:  48%|████▊     | 3102/6470 [42:04<40:11,  1.40it/s]
1|3103|Loss: 3.253805160522461:  48%|████▊     | 3102/6470 [42:04<40:11,  1.40it/s] 
1|3103|Loss: 3.253805160522461:  48%|████▊     | 3103/6470 [42:05<45:15,  1.24it/s]
1|3104|Loss: 2.218660354614258:  48%|████▊     | 3103/6470 [42:05<45:15,  1.24it/s]
1|3104|Loss: 2.218660354614258:  48%|████▊     | 3104/6470 [42:06<46:02,  1.22it/s]
1|3105|Loss: 2.7053756713867188:  48%|████▊     | 3104/6470 [42:06<46:02,  1.22it/s]
1|3105|Loss: 2.7053756713867188:  48%|████▊     | 3105/6470 [42:06<45:55,  1.22it/s]
1|3106|Loss: 3.0652108192443848:  48%|████▊     | 3105/6470 [42:07<45:55,  1.22it/s]
1|3106|Loss: 3.0652108192443848:  48%|████▊     | 3106/6470 [42:07<47:00,  1.19it/s]
1|3107|Loss: 2.660538911819458:  48%|████▊     | 3106/6470 [42:08<47:00,  1.19it/s] 
1|3107|Loss: 2.660538911819458:  48%|████▊     | 3107/6470 [42:08<46:48,  1.20it/s]
1|3108|Loss: 2.6247918605804443:  48%|████▊     | 3107/6470 [42:09<46:48,  1.20it/s]
1|3108|Loss: 2.6247918605804443:  48%|████▊     | 3108/6470 [42:09<45:55,  1.22it/s]
1|3109|Loss: 2.693549633026123:  48%|████▊     | 3108/6470 [42:09<45:55,  1.22it/s] 
1|3109|Loss: 2.693549633026123:  48%|████▊     | 3109/6470 [42:10<45:39,  1.23it/s]
1|3110|Loss: 2.0365610122680664:  48%|████▊     | 3109/6470 [42:10<45:39,  1.23it/s]
1|3110|Loss: 2.0365610122680664:  48%|████▊     | 3110/6470 [42:10<42:49,  1.31it/s]
1|3111|Loss: 1.6275006532669067:  48%|████▊     | 3110/6470 [42:11<42:49,  1.31it/s]
1|3111|Loss: 1.6275006532669067:  48%|████▊     | 3111/6470 [42:11<37:24,  1.50it/s]
1|3112|Loss: 2.6191415786743164:  48%|████▊     | 3111/6470 [42:11<37:24,  1.50it/s]
1|3112|Loss: 2.6191415786743164:  48%|████▊     | 3112/6470 [42:11<35:48,  1.56it/s]
1|3113|Loss: 3.1866443157196045:  48%|████▊     | 3112/6470 [42:12<35:48,  1.56it/s]
1|3113|Loss: 3.1866443157196045:  48%|████▊     | 3113/6470 [42:12<39:35,  1.41it/s]
1|3114|Loss: 2.0265471935272217:  48%|████▊     | 3113/6470 [42:13<39:35,  1.41it/s]
1|3114|Loss: 2.0265471935272217:  48%|████▊     | 3114/6470 [42:13<41:34,  1.35it/s]
1|3115|Loss: 2.556004524230957:  48%|████▊     | 3114/6470 [42:13<41:34,  1.35it/s] 
1|3115|Loss: 2.556004524230957:  48%|████▊     | 3115/6470 [42:14<41:14,  1.36it/s]
1|3116|Loss: 1.6183929443359375:  48%|████▊     | 3115/6470 [42:14<41:14,  1.36it/s]
1|3116|Loss: 1.6183929443359375:  48%|████▊     | 3116/6470 [42:14<38:25,  1.45it/s]
1|3117|Loss: 2.487744092941284:  48%|████▊     | 3116/6470 [42:15<38:25,  1.45it/s] 
1|3117|Loss: 2.487744092941284:  48%|████▊     | 3117/6470 [42:15<39:05,  1.43it/s]
1|3118|Loss: 2.539273500442505:  48%|████▊     | 3117/6470 [42:16<39:05,  1.43it/s]
1|3118|Loss: 2.539273500442505:  48%|████▊     | 3118/6470 [42:16<40:39,  1.37it/s]
1|3119|Loss: 2.212038516998291:  48%|████▊     | 3118/6470 [42:16<40:39,  1.37it/s]
1|3119|Loss: 2.212038516998291:  48%|████▊     | 3119/6470 [42:17<42:10,  1.32it/s]
1|3120|Loss: 2.0340542793273926:  48%|████▊     | 3119/6470 [42:17<42:10,  1.32it/s]
1|3120|Loss: 2.0340542793273926:  48%|████▊     | 3120/6470 [42:17<41:20,  1.35it/s]
1|3121|Loss: 1.9897881746292114:  48%|████▊     | 3120/6470 [42:18<41:20,  1.35it/s]
1|3121|Loss: 1.9897881746292114:  48%|████▊     | 3121/6470 [42:18<39:29,  1.41it/s]
1|3122|Loss: 2.0522658824920654:  48%|████▊     | 3121/6470 [42:18<39:29,  1.41it/s]
1|3122|Loss: 2.0522658824920654:  48%|████▊     | 3122/6470 [42:19<38:59,  1.43it/s]
1|3123|Loss: 2.375898838043213:  48%|████▊     | 3122/6470 [42:19<38:59,  1.43it/s] 
1|3123|Loss: 2.375898838043213:  48%|████▊     | 3123/6470 [42:19<39:55,  1.40it/s]
1|3124|Loss: 2.0655276775360107:  48%|████▊     | 3123/6470 [42:20<39:55,  1.40it/s]
1|3124|Loss: 2.0655276775360107:  48%|████▊     | 3124/6470 [42:20<42:22,  1.32it/s]
1|3125|Loss: 2.3331832885742188:  48%|████▊     | 3124/6470 [42:21<42:22,  1.32it/s]
1|3125|Loss: 2.3331832885742188:  48%|████▊     | 3125/6470 [42:21<44:49,  1.24it/s]
1|3126|Loss: 1.8714029788970947:  48%|████▊     | 3125/6470 [42:22<44:49,  1.24it/s]
1|3126|Loss: 1.8714029788970947:  48%|████▊     | 3126/6470 [42:22<43:21,  1.29it/s]
1|3127|Loss: 2.055116891860962:  48%|████▊     | 3126/6470 [42:22<43:21,  1.29it/s] 
1|3127|Loss: 2.055116891860962:  48%|████▊     | 3127/6470 [42:23<44:16,  1.26it/s]
1|3128|Loss: 1.730453610420227:  48%|████▊     | 3127/6470 [42:23<44:16,  1.26it/s]
1|3128|Loss: 1.730453610420227:  48%|████▊     | 3128/6470 [42:24<45:45,  1.22it/s]
1|3129|Loss: 2.059220314025879:  48%|████▊     | 3128/6470 [42:24<45:45,  1.22it/s]
1|3129|Loss: 2.059220314025879:  48%|████▊     | 3129/6470 [42:24<42:47,  1.30it/s]
1|3130|Loss: 2.4592268466949463:  48%|████▊     | 3129/6470 [42:25<42:47,  1.30it/s]
1|3130|Loss: 2.4592268466949463:  48%|████▊     | 3130/6470 [42:25<42:01,  1.32it/s]
1|3131|Loss: 1.4009617567062378:  48%|████▊     | 3130/6470 [42:25<42:01,  1.32it/s]
1|3131|Loss: 1.4009617567062378:  48%|████▊     | 3131/6470 [42:26<41:01,  1.36it/s]
1|3132|Loss: 1.8932008743286133:  48%|████▊     | 3131/6470 [42:26<41:01,  1.36it/s]
1|3132|Loss: 1.8932008743286133:  48%|████▊     | 3132/6470 [42:26<41:19,  1.35it/s]
1|3133|Loss: 1.9901546239852905:  48%|████▊     | 3132/6470 [42:27<41:19,  1.35it/s]
1|3133|Loss: 1.9901546239852905:  48%|████▊     | 3133/6470 [42:27<38:41,  1.44it/s]
1|3134|Loss: 2.145620346069336:  48%|████▊     | 3133/6470 [42:27<38:41,  1.44it/s] 
1|3134|Loss: 2.145620346069336:  48%|████▊     | 3134/6470 [42:28<39:02,  1.42it/s]
1|3135|Loss: 1.6784197092056274:  48%|████▊     | 3134/6470 [42:28<39:02,  1.42it/s]
1|3135|Loss: 1.6784197092056274:  48%|████▊     | 3135/6470 [42:28<38:44,  1.43it/s]
1|3136|Loss: 1.9758447408676147:  48%|████▊     | 3135/6470 [42:29<38:44,  1.43it/s]
1|3136|Loss: 1.9758447408676147:  48%|████▊     | 3136/6470 [42:29<37:35,  1.48it/s]
1|3137|Loss: 1.7298345565795898:  48%|████▊     | 3136/6470 [42:30<37:35,  1.48it/s]
1|3137|Loss: 1.7298345565795898:  48%|████▊     | 3137/6470 [42:30<40:02,  1.39it/s]
1|3138|Loss: 2.1811933517456055:  48%|████▊     | 3137/6470 [42:30<40:02,  1.39it/s]
1|3138|Loss: 2.1811933517456055:  49%|████▊     | 3138/6470 [42:31<39:51,  1.39it/s]
1|3139|Loss: 1.852769136428833:  49%|████▊     | 3138/6470 [42:31<39:51,  1.39it/s] 
1|3139|Loss: 1.852769136428833:  49%|████▊     | 3139/6470 [42:32<42:40,  1.30it/s]
1|3140|Loss: 1.316152811050415:  49%|████▊     | 3139/6470 [42:32<42:40,  1.30it/s]
1|3140|Loss: 1.316152811050415:  49%|████▊     | 3140/6470 [42:32<40:55,  1.36it/s]
1|3141|Loss: 1.65479576587677:  49%|████▊     | 3140/6470 [42:33<40:55,  1.36it/s] 
1|3141|Loss: 1.65479576587677:  49%|████▊     | 3141/6470 [42:33<40:55,  1.36it/s]
1|3142|Loss: 1.8263522386550903:  49%|████▊     | 3141/6470 [42:33<40:55,  1.36it/s]
1|3142|Loss: 1.8263522386550903:  49%|████▊     | 3142/6470 [42:34<41:18,  1.34it/s]
1|3143|Loss: 1.6901612281799316:  49%|████▊     | 3142/6470 [42:34<41:18,  1.34it/s]
1|3143|Loss: 1.6901612281799316:  49%|████▊     | 3143/6470 [42:34<40:50,  1.36it/s]
1|3144|Loss: 1.7926312685012817:  49%|████▊     | 3143/6470 [42:35<40:50,  1.36it/s]
1|3144|Loss: 1.7926312685012817:  49%|████▊     | 3144/6470 [42:35<41:46,  1.33it/s]
1|3145|Loss: 1.7318352460861206:  49%|████▊     | 3144/6470 [42:36<41:46,  1.33it/s]
1|3145|Loss: 1.7318352460861206:  49%|████▊     | 3145/6470 [42:36<40:00,  1.39it/s]
1|3146|Loss: 1.6007771492004395:  49%|████▊     | 3145/6470 [42:36<40:00,  1.39it/s]
1|3146|Loss: 1.6007771492004395:  49%|████▊     | 3146/6470 [42:36<37:29,  1.48it/s]
1|3147|Loss: 1.9233994483947754:  49%|████▊     | 3146/6470 [42:37<37:29,  1.48it/s]
1|3147|Loss: 1.9233994483947754:  49%|████▊     | 3147/6470 [42:37<38:17,  1.45it/s]
1|3148|Loss: 1.8326985836029053:  49%|████▊     | 3147/6470 [42:38<38:17,  1.45it/s]
1|3148|Loss: 1.8326985836029053:  49%|████▊     | 3148/6470 [42:38<40:07,  1.38it/s]
1|3149|Loss: 1.8501845598220825:  49%|████▊     | 3148/6470 [42:38<40:07,  1.38it/s]
1|3149|Loss: 1.8501845598220825:  49%|████▊     | 3149/6470 [42:39<41:49,  1.32it/s]
1|3150|Loss: 1.5123798847198486:  49%|████▊     | 3149/6470 [42:39<41:49,  1.32it/s]
1|3150|Loss: 1.5123798847198486:  49%|████▊     | 3150/6470 [42:39<41:15,  1.34it/s]
1|3151|Loss: 1.3134418725967407:  49%|████▊     | 3150/6470 [42:40<41:15,  1.34it/s]
1|3151|Loss: 1.3134418725967407:  49%|████▊     | 3151/6470 [42:40<41:45,  1.32it/s]
1|3152|Loss: 1.884824275970459:  49%|████▊     | 3151/6470 [42:41<41:45,  1.32it/s] 
1|3152|Loss: 1.884824275970459:  49%|████▊     | 3152/6470 [42:41<43:57,  1.26it/s]
1|3153|Loss: 1.6108055114746094:  49%|████▊     | 3152/6470 [42:42<43:57,  1.26it/s]
1|3153|Loss: 1.6108055114746094:  49%|████▊     | 3153/6470 [42:42<43:03,  1.28it/s]
1|3154|Loss: 2.046786308288574:  49%|████▊     | 3153/6470 [42:42<43:03,  1.28it/s] 
1|3154|Loss: 2.046786308288574:  49%|████▊     | 3154/6470 [42:43<44:00,  1.26it/s]
1|3155|Loss: 1.5186587572097778:  49%|████▊     | 3154/6470 [42:43<44:00,  1.26it/s]
1|3155|Loss: 1.5186587572097778:  49%|████▉     | 3155/6470 [42:44<43:49,  1.26it/s]
1|3156|Loss: 1.8773488998413086:  49%|████▉     | 3155/6470 [42:44<43:49,  1.26it/s]
1|3156|Loss: 1.8773488998413086:  49%|████▉     | 3156/6470 [42:44<40:43,  1.36it/s]
1|3157|Loss: 1.8847033977508545:  49%|████▉     | 3156/6470 [42:45<40:43,  1.36it/s]
1|3157|Loss: 1.8847033977508545:  49%|████▉     | 3157/6470 [42:45<40:58,  1.35it/s]
1|3158|Loss: 1.8035937547683716:  49%|████▉     | 3157/6470 [42:45<40:58,  1.35it/s]
1|3158|Loss: 1.8035937547683716:  49%|████▉     | 3158/6470 [42:46<43:53,  1.26it/s]
1|3159|Loss: 1.6826809644699097:  49%|████▉     | 3158/6470 [42:46<43:53,  1.26it/s]
1|3159|Loss: 1.6826809644699097:  49%|████▉     | 3159/6470 [42:47<43:05,  1.28it/s]
1|3160|Loss: 1.2795226573944092:  49%|████▉     | 3159/6470 [42:47<43:05,  1.28it/s]
1|3160|Loss: 1.2795226573944092:  49%|████▉     | 3160/6470 [42:47<38:38,  1.43it/s]
1|3161|Loss: 1.3306794166564941:  49%|████▉     | 3160/6470 [42:47<38:38,  1.43it/s]
1|3161|Loss: 1.3306794166564941:  49%|████▉     | 3161/6470 [42:48<37:43,  1.46it/s]
1|3162|Loss: 1.5712947845458984:  49%|████▉     | 3161/6470 [42:48<37:43,  1.46it/s]
1|3162|Loss: 1.5712947845458984:  49%|████▉     | 3162/6470 [42:48<35:07,  1.57it/s]
1|3163|Loss: 1.3450783491134644:  49%|████▉     | 3162/6470 [42:49<35:07,  1.57it/s]
1|3163|Loss: 1.3450783491134644:  49%|████▉     | 3163/6470 [42:49<35:13,  1.56it/s]
1|3164|Loss: 1.4958932399749756:  49%|████▉     | 3163/6470 [42:49<35:13,  1.56it/s]
1|3164|Loss: 1.4958932399749756:  49%|████▉     | 3164/6470 [42:50<39:05,  1.41it/s]
1|3165|Loss: 1.4866197109222412:  49%|████▉     | 3164/6470 [42:50<39:05,  1.41it/s]
1|3165|Loss: 1.4866197109222412:  49%|████▉     | 3165/6470 [42:51<42:06,  1.31it/s]
1|3166|Loss: 3.0243608951568604:  49%|████▉     | 3165/6470 [42:51<42:06,  1.31it/s]
1|3166|Loss: 3.0243608951568604:  49%|████▉     | 3166/6470 [42:52<53:36,  1.03it/s]
1|3167|Loss: 2.8326797485351562:  49%|████▉     | 3166/6470 [42:53<53:36,  1.03it/s]
1|3167|Loss: 2.8326797485351562:  49%|████▉     | 3167/6470 [42:53<55:48,  1.01s/it]
1|3168|Loss: 1.5924286842346191:  49%|████▉     | 3167/6470 [42:54<55:48,  1.01s/it]
1|3168|Loss: 1.5924286842346191:  49%|████▉     | 3168/6470 [42:54<51:27,  1.07it/s]
1|3169|Loss: 1.9698642492294312:  49%|████▉     | 3168/6470 [42:54<51:27,  1.07it/s]
1|3169|Loss: 1.9698642492294312:  49%|████▉     | 3169/6470 [42:55<49:13,  1.12it/s]
1|3170|Loss: 1.7204761505126953:  49%|████▉     | 3169/6470 [42:55<49:13,  1.12it/s]
1|3170|Loss: 1.7204761505126953:  49%|████▉     | 3170/6470 [42:56<47:44,  1.15it/s]
1|3171|Loss: 1.6238586902618408:  49%|████▉     | 3170/6470 [42:56<47:44,  1.15it/s]
1|3171|Loss: 1.6238586902618408:  49%|████▉     | 3171/6470 [42:56<47:31,  1.16it/s]
1|3172|Loss: 1.5018656253814697:  49%|████▉     | 3171/6470 [42:57<47:31,  1.16it/s]
1|3172|Loss: 1.5018656253814697:  49%|████▉     | 3172/6470 [42:57<46:15,  1.19it/s]
1|3173|Loss: 1.8095306158065796:  49%|████▉     | 3172/6470 [42:58<46:15,  1.19it/s]
1|3173|Loss: 1.8095306158065796:  49%|████▉     | 3173/6470 [42:58<46:22,  1.18it/s]
1|3174|Loss: 1.3081681728363037:  49%|████▉     | 3173/6470 [42:58<46:22,  1.18it/s]
1|3174|Loss: 1.3081681728363037:  49%|████▉     | 3174/6470 [42:59<43:07,  1.27it/s]
1|3175|Loss: 1.4741747379302979:  49%|████▉     | 3174/6470 [42:59<43:07,  1.27it/s]
1|3175|Loss: 1.4741747379302979:  49%|████▉     | 3175/6470 [42:59<42:33,  1.29it/s]
1|3176|Loss: 1.2707364559173584:  49%|████▉     | 3175/6470 [43:00<42:33,  1.29it/s]
1|3176|Loss: 1.2707364559173584:  49%|████▉     | 3176/6470 [43:00<43:07,  1.27it/s]
1|3177|Loss: 1.3890495300292969:  49%|████▉     | 3176/6470 [43:01<43:07,  1.27it/s]
1|3177|Loss: 1.3890495300292969:  49%|████▉     | 3177/6470 [43:01<41:55,  1.31it/s]
1|3178|Loss: 1.650659441947937:  49%|████▉     | 3177/6470 [43:01<41:55,  1.31it/s] 
1|3178|Loss: 1.650659441947937:  49%|████▉     | 3178/6470 [43:02<41:42,  1.32it/s]
1|3179|Loss: 1.243265986442566:  49%|████▉     | 3178/6470 [43:02<41:42,  1.32it/s]
1|3179|Loss: 1.243265986442566:  49%|████▉     | 3179/6470 [43:02<40:25,  1.36it/s]
1|3180|Loss: 1.2675251960754395:  49%|████▉     | 3179/6470 [43:03<40:25,  1.36it/s]
1|3180|Loss: 1.2675251960754395:  49%|████▉     | 3180/6470 [43:03<38:27,  1.43it/s]
1|3181|Loss: 1.2406611442565918:  49%|████▉     | 3180/6470 [43:03<38:27,  1.43it/s]
1|3181|Loss: 1.2406611442565918:  49%|████▉     | 3181/6470 [43:04<37:02,  1.48it/s]
1|3182|Loss: 1.3236701488494873:  49%|████▉     | 3181/6470 [43:04<37:02,  1.48it/s]
1|3182|Loss: 1.3236701488494873:  49%|████▉     | 3182/6470 [43:04<36:59,  1.48it/s]
1|3183|Loss: 1.5591341257095337:  49%|████▉     | 3182/6470 [43:05<36:59,  1.48it/s]
1|3183|Loss: 1.5591341257095337:  49%|████▉     | 3183/6470 [43:05<37:38,  1.46it/s]
1|3184|Loss: 1.5131930112838745:  49%|████▉     | 3183/6470 [43:06<37:38,  1.46it/s]
1|3184|Loss: 1.5131930112838745:  49%|████▉     | 3184/6470 [43:06<41:39,  1.31it/s]
1|3185|Loss: 1.3005552291870117:  49%|████▉     | 3184/6470 [43:07<41:39,  1.31it/s]
1|3185|Loss: 1.3005552291870117:  49%|████▉     | 3185/6470 [43:07<44:25,  1.23it/s]
1|3186|Loss: 1.2712582349777222:  49%|████▉     | 3185/6470 [43:07<44:25,  1.23it/s]
1|3186|Loss: 1.2712582349777222:  49%|████▉     | 3186/6470 [43:07<40:25,  1.35it/s]
1|3187|Loss: 1.359938383102417:  49%|████▉     | 3186/6470 [43:08<40:25,  1.35it/s] 
1|3187|Loss: 1.359938383102417:  49%|████▉     | 3187/6470 [43:08<39:05,  1.40it/s]
1|3188|Loss: 1.4216959476470947:  49%|████▉     | 3187/6470 [43:09<39:05,  1.40it/s]
1|3188|Loss: 1.4216959476470947:  49%|████▉     | 3188/6470 [43:09<39:54,  1.37it/s]
1|3189|Loss: 1.2628265619277954:  49%|████▉     | 3188/6470 [43:09<39:54,  1.37it/s]
1|3189|Loss: 1.2628265619277954:  49%|████▉     | 3189/6470 [43:10<39:13,  1.39it/s]
1|3190|Loss: 1.3093204498291016:  49%|████▉     | 3189/6470 [43:10<39:13,  1.39it/s]
1|3190|Loss: 1.3093204498291016:  49%|████▉     | 3190/6470 [43:10<38:55,  1.40it/s]
1|3191|Loss: 1.4101910591125488:  49%|████▉     | 3190/6470 [43:11<38:55,  1.40it/s]
1|3191|Loss: 1.4101910591125488:  49%|████▉     | 3191/6470 [43:11<36:57,  1.48it/s]
1|3192|Loss: 1.433543086051941:  49%|████▉     | 3191/6470 [43:11<36:57,  1.48it/s] 
1|3192|Loss: 1.433543086051941:  49%|████▉     | 3192/6470 [43:12<38:02,  1.44it/s]
1|3193|Loss: 1.442991852760315:  49%|████▉     | 3192/6470 [43:12<38:02,  1.44it/s]
1|3193|Loss: 1.442991852760315:  49%|████▉     | 3193/6470 [43:13<41:23,  1.32it/s]
1|3194|Loss: 1.2720458507537842:  49%|████▉     | 3193/6470 [43:13<41:23,  1.32it/s]
1|3194|Loss: 1.2720458507537842:  49%|████▉     | 3194/6470 [43:13<42:11,  1.29it/s]
1|3195|Loss: 2.3854024410247803:  49%|████▉     | 3194/6470 [43:14<42:11,  1.29it/s]
1|3195|Loss: 2.3854024410247803:  49%|████▉     | 3195/6470 [43:14<46:06,  1.18it/s]
1|3196|Loss: 1.5452648401260376:  49%|████▉     | 3195/6470 [43:15<46:06,  1.18it/s]
1|3196|Loss: 1.5452648401260376:  49%|████▉     | 3196/6470 [43:15<49:01,  1.11it/s]
1|3197|Loss: 1.709315538406372:  49%|████▉     | 3196/6470 [43:16<49:01,  1.11it/s] 
1|3197|Loss: 1.709315538406372:  49%|████▉     | 3197/6470 [43:16<46:50,  1.16it/s]
1|3198|Loss: 1.6944295167922974:  49%|████▉     | 3197/6470 [43:17<46:50,  1.16it/s]
1|3198|Loss: 1.6944295167922974:  49%|████▉     | 3198/6470 [43:17<45:24,  1.20it/s]
1|3199|Loss: 1.5299289226531982:  49%|████▉     | 3198/6470 [43:17<45:24,  1.20it/s]
1|3199|Loss: 1.5299289226531982:  49%|████▉     | 3199/6470 [43:18<42:22,  1.29it/s]
1|3200|Loss: 1.262935996055603:  49%|████▉     | 3199/6470 [43:18<42:22,  1.29it/s] 
1|3200|Loss: 1.262935996055603:  49%|████▉     | 3200/6470 [43:18<41:30,  1.31it/s]
1|3201|Loss: 1.4758281707763672:  49%|████▉     | 3200/6470 [43:19<41:30,  1.31it/s]
1|3201|Loss: 1.4758281707763672:  49%|████▉     | 3201/6470 [43:19<39:53,  1.37it/s]
1|3202|Loss: 1.1961551904678345:  49%|████▉     | 3201/6470 [43:19<39:53,  1.37it/s]
1|3202|Loss: 1.1961551904678345:  49%|████▉     | 3202/6470 [43:20<38:42,  1.41it/s]
1|3203|Loss: 1.6739764213562012:  49%|████▉     | 3202/6470 [43:20<38:42,  1.41it/s]
1|3203|Loss: 1.6739764213562012:  50%|████▉     | 3203/6470 [43:20<37:24,  1.46it/s]
1|3204|Loss: 1.1014825105667114:  50%|████▉     | 3203/6470 [43:21<37:24,  1.46it/s]
1|3204|Loss: 1.1014825105667114:  50%|████▉     | 3204/6470 [43:21<36:11,  1.50it/s]
1|3205|Loss: 1.0033001899719238:  50%|████▉     | 3204/6470 [43:21<36:11,  1.50it/s]
1|3205|Loss: 1.0033001899719238:  50%|████▉     | 3205/6470 [43:22<36:06,  1.51it/s]
1|3206|Loss: 1.546273946762085:  50%|████▉     | 3205/6470 [43:22<36:06,  1.51it/s] 
1|3206|Loss: 1.546273946762085:  50%|████▉     | 3206/6470 [43:22<37:55,  1.43it/s]
1|3207|Loss: 1.3058236837387085:  50%|████▉     | 3206/6470 [43:23<37:55,  1.43it/s]
1|3207|Loss: 1.3058236837387085:  50%|████▉     | 3207/6470 [43:23<36:59,  1.47it/s]
1|3208|Loss: 1.3130850791931152:  50%|████▉     | 3207/6470 [43:23<36:59,  1.47it/s]
1|3208|Loss: 1.3130850791931152:  50%|████▉     | 3208/6470 [43:24<37:48,  1.44it/s]
1|3209|Loss: 1.1870007514953613:  50%|████▉     | 3208/6470 [43:24<37:48,  1.44it/s]
1|3209|Loss: 1.1870007514953613:  50%|████▉     | 3209/6470 [43:24<38:16,  1.42it/s]
1|3210|Loss: 1.3529553413391113:  50%|████▉     | 3209/6470 [43:25<38:16,  1.42it/s]
1|3210|Loss: 1.3529553413391113:  50%|████▉     | 3210/6470 [43:25<39:03,  1.39it/s]
1|3211|Loss: 1.8659948110580444:  50%|████▉     | 3210/6470 [43:26<39:03,  1.39it/s]
1|3211|Loss: 1.8659948110580444:  50%|████▉     | 3211/6470 [43:26<40:15,  1.35it/s]
1|3212|Loss: 1.198627233505249:  50%|████▉     | 3211/6470 [43:26<40:15,  1.35it/s] 
1|3212|Loss: 1.198627233505249:  50%|████▉     | 3212/6470 [43:27<41:13,  1.32it/s]
1|3213|Loss: 1.2617814540863037:  50%|████▉     | 3212/6470 [43:27<41:13,  1.32it/s]
1|3213|Loss: 1.2617814540863037:  50%|████▉     | 3213/6470 [43:28<41:53,  1.30it/s]
1|3214|Loss: 1.6389745473861694:  50%|████▉     | 3213/6470 [43:28<41:53,  1.30it/s]
1|3214|Loss: 1.6389745473861694:  50%|████▉     | 3214/6470 [43:28<41:47,  1.30it/s]
1|3215|Loss: 1.6082115173339844:  50%|████▉     | 3214/6470 [43:29<41:47,  1.30it/s]
1|3215|Loss: 1.6082115173339844:  50%|████▉     | 3215/6470 [43:29<41:36,  1.30it/s]
1|3216|Loss: 1.1454854011535645:  50%|████▉     | 3215/6470 [43:30<41:36,  1.30it/s]
1|3216|Loss: 1.1454854011535645:  50%|████▉     | 3216/6470 [43:30<43:26,  1.25it/s]
1|3217|Loss: 1.5592706203460693:  50%|████▉     | 3216/6470 [43:30<43:26,  1.25it/s]
1|3217|Loss: 1.5592706203460693:  50%|████▉     | 3217/6470 [43:31<43:45,  1.24it/s]
1|3218|Loss: 1.3334577083587646:  50%|████▉     | 3217/6470 [43:31<43:45,  1.24it/s]
1|3218|Loss: 1.3334577083587646:  50%|████▉     | 3218/6470 [43:32<45:15,  1.20it/s]
1|3219|Loss: 1.1886714696884155:  50%|████▉     | 3218/6470 [43:32<45:15,  1.20it/s]
1|3219|Loss: 1.1886714696884155:  50%|████▉     | 3219/6470 [43:32<44:40,  1.21it/s]
1|3220|Loss: 1.5667797327041626:  50%|████▉     | 3219/6470 [43:33<44:40,  1.21it/s]
1|3220|Loss: 1.5667797327041626:  50%|████▉     | 3220/6470 [43:33<43:29,  1.25it/s]
1|3221|Loss: 1.4913219213485718:  50%|████▉     | 3220/6470 [43:34<43:29,  1.25it/s]
1|3221|Loss: 1.4913219213485718:  50%|████▉     | 3221/6470 [43:34<41:03,  1.32it/s]
1|3222|Loss: 1.836476445198059:  50%|████▉     | 3221/6470 [43:34<41:03,  1.32it/s] 
1|3222|Loss: 1.836476445198059:  50%|████▉     | 3222/6470 [43:35<45:06,  1.20it/s]
1|3223|Loss: 1.1670054197311401:  50%|████▉     | 3222/6470 [43:35<45:06,  1.20it/s]
1|3223|Loss: 1.1670054197311401:  50%|████▉     | 3223/6470 [43:36<47:10,  1.15it/s]
1|3224|Loss: 1.2330753803253174:  50%|████▉     | 3223/6470 [43:36<47:10,  1.15it/s]
1|3224|Loss: 1.2330753803253174:  50%|████▉     | 3224/6470 [43:37<48:18,  1.12it/s]
1|3225|Loss: 1.060699224472046:  50%|████▉     | 3224/6470 [43:37<48:18,  1.12it/s] 
1|3225|Loss: 1.060699224472046:  50%|████▉     | 3225/6470 [43:38<45:48,  1.18it/s]
1|3226|Loss: 1.2444902658462524:  50%|████▉     | 3225/6470 [43:38<45:48,  1.18it/s]
1|3226|Loss: 1.2444902658462524:  50%|████▉     | 3226/6470 [43:38<46:15,  1.17it/s]
1|3227|Loss: 1.0354466438293457:  50%|████▉     | 3226/6470 [43:39<46:15,  1.17it/s]
1|3227|Loss: 1.0354466438293457:  50%|████▉     | 3227/6470 [43:39<46:41,  1.16it/s]
1|3228|Loss: 1.278360366821289:  50%|████▉     | 3227/6470 [43:40<46:41,  1.16it/s] 
1|3228|Loss: 1.278360366821289:  50%|████▉     | 3228/6470 [43:40<47:15,  1.14it/s]
1|3229|Loss: 1.5375794172286987:  50%|████▉     | 3228/6470 [43:41<47:15,  1.14it/s]
1|3229|Loss: 1.5375794172286987:  50%|████▉     | 3229/6470 [43:41<47:43,  1.13it/s]
1|3230|Loss: 1.1353133916854858:  50%|████▉     | 3229/6470 [43:41<47:43,  1.13it/s]
1|3230|Loss: 1.1353133916854858:  50%|████▉     | 3230/6470 [43:42<44:05,  1.22it/s]
1|3231|Loss: 1.0805634260177612:  50%|████▉     | 3230/6470 [43:42<44:05,  1.22it/s]
1|3231|Loss: 1.0805634260177612:  50%|████▉     | 3231/6470 [43:42<41:31,  1.30it/s]
1|3232|Loss: 1.3797467947006226:  50%|████▉     | 3231/6470 [43:43<41:31,  1.30it/s]
1|3232|Loss: 1.3797467947006226:  50%|████▉     | 3232/6470 [43:43<41:05,  1.31it/s]
1|3233|Loss: 1.106299638748169:  50%|████▉     | 3232/6470 [43:44<41:05,  1.31it/s] 
1|3233|Loss: 1.106299638748169:  50%|████▉     | 3233/6470 [43:44<43:02,  1.25it/s]
1|3234|Loss: 1.563525915145874:  50%|████▉     | 3233/6470 [43:44<43:02,  1.25it/s]
1|3234|Loss: 1.563525915145874:  50%|████▉     | 3234/6470 [43:45<39:23,  1.37it/s]
1|3235|Loss: 1.3968232870101929:  50%|████▉     | 3234/6470 [43:45<39:23,  1.37it/s]
1|3235|Loss: 1.3968232870101929:  50%|█████     | 3235/6470 [43:45<41:07,  1.31it/s]
1|3236|Loss: 1.2615339756011963:  50%|█████     | 3235/6470 [43:46<41:07,  1.31it/s]
1|3236|Loss: 1.2615339756011963:  50%|█████     | 3236/6470 [43:46<43:42,  1.23it/s]
1|3237|Loss: 1.1530135869979858:  50%|█████     | 3236/6470 [43:47<43:42,  1.23it/s]
1|3237|Loss: 1.1530135869979858:  50%|█████     | 3237/6470 [43:47<43:52,  1.23it/s]
1|3238|Loss: 2.1133782863616943:  50%|█████     | 3237/6470 [43:48<43:52,  1.23it/s]
1|3238|Loss: 2.1133782863616943:  50%|█████     | 3238/6470 [43:48<46:05,  1.17it/s]
1|3239|Loss: 0.9791162610054016:  50%|█████     | 3238/6470 [43:49<46:05,  1.17it/s]
1|3239|Loss: 0.9791162610054016:  50%|█████     | 3239/6470 [43:49<46:00,  1.17it/s]
1|3240|Loss: 1.1486263275146484:  50%|█████     | 3239/6470 [43:49<46:00,  1.17it/s]
1|3240|Loss: 1.1486263275146484:  50%|█████     | 3240/6470 [43:50<45:23,  1.19it/s]
1|3241|Loss: 1.088029146194458:  50%|█████     | 3240/6470 [43:50<45:23,  1.19it/s] 
1|3241|Loss: 1.088029146194458:  50%|█████     | 3241/6470 [43:51<44:12,  1.22it/s]
1|3242|Loss: 1.0679987668991089:  50%|█████     | 3241/6470 [43:51<44:12,  1.22it/s]
1|3242|Loss: 1.0679987668991089:  50%|█████     | 3242/6470 [43:51<43:39,  1.23it/s]
1|3243|Loss: 1.1083470582962036:  50%|█████     | 3242/6470 [43:52<43:39,  1.23it/s]
1|3243|Loss: 1.1083470582962036:  50%|█████     | 3243/6470 [43:52<40:53,  1.32it/s]
1|3244|Loss: 1.6808193922042847:  50%|█████     | 3243/6470 [43:52<40:53,  1.32it/s]
1|3244|Loss: 1.6808193922042847:  50%|█████     | 3244/6470 [43:53<38:57,  1.38it/s]
1|3245|Loss: 1.090735673904419:  50%|█████     | 3244/6470 [43:53<38:57,  1.38it/s] 
1|3245|Loss: 1.090735673904419:  50%|█████     | 3245/6470 [43:53<40:40,  1.32it/s]
1|3246|Loss: 1.086203932762146:  50%|█████     | 3245/6470 [43:54<40:40,  1.32it/s]
1|3246|Loss: 1.086203932762146:  50%|█████     | 3246/6470 [43:54<41:30,  1.29it/s]
1|3247|Loss: 1.1073532104492188:  50%|█████     | 3246/6470 [43:55<41:30,  1.29it/s]
1|3247|Loss: 1.1073532104492188:  50%|█████     | 3247/6470 [43:55<39:46,  1.35it/s]
1|3248|Loss: 1.137322187423706:  50%|█████     | 3247/6470 [43:55<39:46,  1.35it/s] 
1|3248|Loss: 1.137322187423706:  50%|█████     | 3248/6470 [43:56<38:50,  1.38it/s]
1|3249|Loss: 1.1222186088562012:  50%|█████     | 3248/6470 [43:56<38:50,  1.38it/s]
1|3249|Loss: 1.1222186088562012:  50%|█████     | 3249/6470 [43:56<39:59,  1.34it/s]
1|3250|Loss: 1.1870111227035522:  50%|█████     | 3249/6470 [43:57<39:59,  1.34it/s]
1|3250|Loss: 1.1870111227035522:  50%|█████     | 3250/6470 [43:57<39:48,  1.35it/s]
1|3251|Loss: 1.04038667678833:  50%|█████     | 3250/6470 [43:58<39:48,  1.35it/s]  
1|3251|Loss: 1.04038667678833:  50%|█████     | 3251/6470 [43:58<39:15,  1.37it/s]
1|3252|Loss: 1.4704017639160156:  50%|█████     | 3251/6470 [43:58<39:15,  1.37it/s]
1|3252|Loss: 1.4704017639160156:  50%|█████     | 3252/6470 [43:59<38:54,  1.38it/s]
1|3253|Loss: 1.287590742111206:  50%|█████     | 3252/6470 [43:59<38:54,  1.38it/s] 
1|3253|Loss: 1.287590742111206:  50%|█████     | 3253/6470 [43:59<38:37,  1.39it/s]
1|3254|Loss: 1.222133755683899:  50%|█████     | 3253/6470 [44:00<38:37,  1.39it/s]
1|3254|Loss: 1.222133755683899:  50%|█████     | 3254/6470 [44:00<36:10,  1.48it/s]
1|3255|Loss: 1.274307131767273:  50%|█████     | 3254/6470 [44:00<36:10,  1.48it/s]
1|3255|Loss: 1.274307131767273:  50%|█████     | 3255/6470 [44:01<36:08,  1.48it/s]
1|3256|Loss: 1.1169493198394775:  50%|█████     | 3255/6470 [44:01<36:08,  1.48it/s]
1|3256|Loss: 1.1169493198394775:  50%|█████     | 3256/6470 [44:01<38:05,  1.41it/s]
1|3257|Loss: 0.9424256682395935:  50%|█████     | 3256/6470 [44:02<38:05,  1.41it/s]
1|3257|Loss: 0.9424256682395935:  50%|█████     | 3257/6470 [44:02<39:27,  1.36it/s]
1|3258|Loss: 1.0810381174087524:  50%|█████     | 3257/6470 [44:03<39:27,  1.36it/s]
1|3258|Loss: 1.0810381174087524:  50%|█████     | 3258/6470 [44:03<41:58,  1.28it/s]
1|3259|Loss: 1.1183143854141235:  50%|█████     | 3258/6470 [44:03<41:58,  1.28it/s]
1|3259|Loss: 1.1183143854141235:  50%|█████     | 3259/6470 [44:04<41:59,  1.27it/s]
1|3260|Loss: 1.1753758192062378:  50%|█████     | 3259/6470 [44:04<41:59,  1.27it/s]
1|3260|Loss: 1.1753758192062378:  50%|█████     | 3260/6470 [44:05<41:33,  1.29it/s]
1|3261|Loss: 1.0645620822906494:  50%|█████     | 3260/6470 [44:05<41:33,  1.29it/s]
1|3261|Loss: 1.0645620822906494:  50%|█████     | 3261/6470 [44:05<38:01,  1.41it/s]
1|3262|Loss: 1.2175819873809814:  50%|█████     | 3261/6470 [44:06<38:01,  1.41it/s]
1|3262|Loss: 1.2175819873809814:  50%|█████     | 3262/6470 [44:06<40:38,  1.32it/s]
1|3263|Loss: 1.1245644092559814:  50%|█████     | 3262/6470 [44:07<40:38,  1.32it/s]
1|3263|Loss: 1.1245644092559814:  50%|█████     | 3263/6470 [44:07<43:22,  1.23it/s]
1|3264|Loss: 1.0837947130203247:  50%|█████     | 3263/6470 [44:07<43:22,  1.23it/s]
1|3264|Loss: 1.0837947130203247:  50%|█████     | 3264/6470 [44:08<40:16,  1.33it/s]
1|3265|Loss: 1.3506523370742798:  50%|█████     | 3264/6470 [44:08<40:16,  1.33it/s]
1|3265|Loss: 1.3506523370742798:  50%|█████     | 3265/6470 [44:08<40:57,  1.30it/s]
1|3266|Loss: 1.1093103885650635:  50%|█████     | 3265/6470 [44:09<40:57,  1.30it/s]
1|3266|Loss: 1.1093103885650635:  50%|█████     | 3266/6470 [44:09<38:32,  1.39it/s]
1|3267|Loss: 1.0683789253234863:  50%|█████     | 3266/6470 [44:09<38:32,  1.39it/s]
1|3267|Loss: 1.0683789253234863:  50%|█████     | 3267/6470 [44:10<37:12,  1.43it/s]
1|3268|Loss: 1.0923854112625122:  50%|█████     | 3267/6470 [44:10<37:12,  1.43it/s]
1|3268|Loss: 1.0923854112625122:  51%|█████     | 3268/6470 [44:10<36:16,  1.47it/s]
1|3269|Loss: 1.1246415376663208:  51%|█████     | 3268/6470 [44:11<36:16,  1.47it/s]
1|3269|Loss: 1.1246415376663208:  51%|█████     | 3269/6470 [44:11<39:44,  1.34it/s]
1|3270|Loss: 1.184912085533142:  51%|█████     | 3269/6470 [44:12<39:44,  1.34it/s] 
1|3270|Loss: 1.184912085533142:  51%|█████     | 3270/6470 [44:12<39:52,  1.34it/s]
1|3271|Loss: 1.1612440347671509:  51%|█████     | 3270/6470 [44:12<39:52,  1.34it/s]
1|3271|Loss: 1.1612440347671509:  51%|█████     | 3271/6470 [44:13<38:52,  1.37it/s]
1|3272|Loss: 0.9372775554656982:  51%|█████     | 3271/6470 [44:13<38:52,  1.37it/s]
1|3272|Loss: 0.9372775554656982:  51%|█████     | 3272/6470 [44:13<39:44,  1.34it/s]
1|3273|Loss: 1.1861010789871216:  51%|█████     | 3272/6470 [44:14<39:44,  1.34it/s]
1|3273|Loss: 1.1861010789871216:  51%|█████     | 3273/6470 [44:14<42:40,  1.25it/s]
1|3274|Loss: 1.5587211847305298:  51%|█████     | 3273/6470 [44:15<42:40,  1.25it/s]
1|3274|Loss: 1.5587211847305298:  51%|█████     | 3274/6470 [44:15<45:58,  1.16it/s]
1|3275|Loss: 1.0625256299972534:  51%|█████     | 3274/6470 [44:16<45:58,  1.16it/s]
1|3275|Loss: 1.0625256299972534:  51%|█████     | 3275/6470 [44:16<45:11,  1.18it/s]
1|3276|Loss: 0.9299613237380981:  51%|█████     | 3275/6470 [44:17<45:11,  1.18it/s]
1|3276|Loss: 0.9299613237380981:  51%|█████     | 3276/6470 [44:17<44:34,  1.19it/s]
1|3277|Loss: 0.9786674976348877:  51%|█████     | 3276/6470 [44:17<44:34,  1.19it/s]
1|3277|Loss: 0.9786674976348877:  51%|█████     | 3277/6470 [44:17<39:50,  1.34it/s]
1|3278|Loss: 1.155728816986084:  51%|█████     | 3277/6470 [44:18<39:50,  1.34it/s] 
1|3278|Loss: 1.155728816986084:  51%|█████     | 3278/6470 [44:18<39:19,  1.35it/s]
1|3279|Loss: 1.2666420936584473:  51%|█████     | 3278/6470 [44:19<39:19,  1.35it/s]
1|3279|Loss: 1.2666420936584473:  51%|█████     | 3279/6470 [44:19<38:31,  1.38it/s]
1|3280|Loss: 1.182478904724121:  51%|█████     | 3279/6470 [44:19<38:31,  1.38it/s] 
1|3280|Loss: 1.182478904724121:  51%|█████     | 3280/6470 [44:20<40:48,  1.30it/s]
1|3281|Loss: 1.2447630167007446:  51%|█████     | 3280/6470 [44:20<40:48,  1.30it/s]
1|3281|Loss: 1.2447630167007446:  51%|█████     | 3281/6470 [44:20<39:53,  1.33it/s]
1|3282|Loss: 1.0798068046569824:  51%|█████     | 3281/6470 [44:21<39:53,  1.33it/s]
1|3282|Loss: 1.0798068046569824:  51%|█████     | 3282/6470 [44:21<40:09,  1.32it/s]
1|3283|Loss: 1.108009934425354:  51%|█████     | 3282/6470 [44:22<40:09,  1.32it/s] 
1|3283|Loss: 1.108009934425354:  51%|█████     | 3283/6470 [44:22<42:42,  1.24it/s]
1|3284|Loss: 0.8992347717285156:  51%|█████     | 3283/6470 [44:23<42:42,  1.24it/s]
1|3284|Loss: 0.8992347717285156:  51%|█████     | 3284/6470 [44:23<39:36,  1.34it/s]
1|3285|Loss: 1.4631352424621582:  51%|█████     | 3284/6470 [44:23<39:36,  1.34it/s]
1|3285|Loss: 1.4631352424621582:  51%|█████     | 3285/6470 [44:23<38:16,  1.39it/s]
1|3286|Loss: 0.8923519253730774:  51%|█████     | 3285/6470 [44:24<38:16,  1.39it/s]
1|3286|Loss: 0.8923519253730774:  51%|█████     | 3286/6470 [44:24<40:56,  1.30it/s]
1|3287|Loss: 0.9374538660049438:  51%|█████     | 3286/6470 [44:25<40:56,  1.30it/s]
1|3287|Loss: 0.9374538660049438:  51%|█████     | 3287/6470 [44:25<41:19,  1.28it/s]
1|3288|Loss: 1.0893843173980713:  51%|█████     | 3287/6470 [44:25<41:19,  1.28it/s]
1|3288|Loss: 1.0893843173980713:  51%|█████     | 3288/6470 [44:26<36:06,  1.47it/s]
1|3289|Loss: 1.1955430507659912:  51%|█████     | 3288/6470 [44:26<36:06,  1.47it/s]
1|3289|Loss: 1.1955430507659912:  51%|█████     | 3289/6470 [44:26<35:39,  1.49it/s]
1|3290|Loss: 1.3262944221496582:  51%|█████     | 3289/6470 [44:27<35:39,  1.49it/s]
1|3290|Loss: 1.3262944221496582:  51%|█████     | 3290/6470 [44:27<37:00,  1.43it/s]
1|3291|Loss: 1.0413936376571655:  51%|█████     | 3290/6470 [44:27<37:00,  1.43it/s]
1|3291|Loss: 1.0413936376571655:  51%|█████     | 3291/6470 [44:28<36:26,  1.45it/s]
1|3292|Loss: 1.0986099243164062:  51%|█████     | 3291/6470 [44:28<36:26,  1.45it/s]
1|3292|Loss: 1.0986099243164062:  51%|█████     | 3292/6470 [44:28<36:25,  1.45it/s]
1|3293|Loss: 1.069284200668335:  51%|█████     | 3292/6470 [44:29<36:25,  1.45it/s] 
1|3293|Loss: 1.069284200668335:  51%|█████     | 3293/6470 [44:29<39:34,  1.34it/s]
1|3294|Loss: 1.1813322305679321:  51%|█████     | 3293/6470 [44:30<39:34,  1.34it/s]
1|3294|Loss: 1.1813322305679321:  51%|█████     | 3294/6470 [44:30<44:08,  1.20it/s]
1|3295|Loss: 0.9686996340751648:  51%|█████     | 3294/6470 [44:31<44:08,  1.20it/s]
1|3295|Loss: 0.9686996340751648:  51%|█████     | 3295/6470 [44:31<41:49,  1.27it/s]
1|3296|Loss: 1.0919522047042847:  51%|█████     | 3295/6470 [44:31<41:49,  1.27it/s]
1|3296|Loss: 1.0919522047042847:  51%|█████     | 3296/6470 [44:32<38:32,  1.37it/s]
1|3297|Loss: 1.1368718147277832:  51%|█████     | 3296/6470 [44:32<38:32,  1.37it/s]
1|3297|Loss: 1.1368718147277832:  51%|█████     | 3297/6470 [44:32<39:15,  1.35it/s]
1|3298|Loss: 1.1633106470108032:  51%|█████     | 3297/6470 [44:33<39:15,  1.35it/s]
1|3298|Loss: 1.1633106470108032:  51%|█████     | 3298/6470 [44:33<39:45,  1.33it/s]
1|3299|Loss: 1.7044475078582764:  51%|█████     | 3298/6470 [44:33<39:45,  1.33it/s]
1|3299|Loss: 1.7044475078582764:  51%|█████     | 3299/6470 [44:34<36:18,  1.46it/s]
1|3300|Loss: 1.0369606018066406:  51%|█████     | 3299/6470 [44:34<36:18,  1.46it/s]
1|3300|Loss: 1.0369606018066406:  51%|█████     | 3300/6470 [44:34<35:50,  1.47it/s]
1|3301|Loss: 1.1593482494354248:  51%|█████     | 3300/6470 [44:35<35:50,  1.47it/s]
1|3301|Loss: 1.1593482494354248:  51%|█████     | 3301/6470 [44:35<33:25,  1.58it/s]
1|3302|Loss: 1.1180835962295532:  51%|█████     | 3301/6470 [44:35<33:25,  1.58it/s]
1|3302|Loss: 1.1180835962295532:  51%|█████     | 3302/6470 [44:35<34:21,  1.54it/s]
1|3303|Loss: 1.2405682802200317:  51%|█████     | 3302/6470 [44:36<34:21,  1.54it/s]
1|3303|Loss: 1.2405682802200317:  51%|█████     | 3303/6470 [44:36<39:08,  1.35it/s]
1|3304|Loss: 1.440147876739502:  51%|█████     | 3303/6470 [44:37<39:08,  1.35it/s] 
1|3304|Loss: 1.440147876739502:  51%|█████     | 3304/6470 [44:37<38:34,  1.37it/s]
1|3305|Loss: 1.4008145332336426:  51%|█████     | 3304/6470 [44:38<38:34,  1.37it/s]
1|3305|Loss: 1.4008145332336426:  51%|█████     | 3305/6470 [44:38<40:46,  1.29it/s]
1|3306|Loss: 1.1206672191619873:  51%|█████     | 3305/6470 [44:38<40:46,  1.29it/s]
1|3306|Loss: 1.1206672191619873:  51%|█████     | 3306/6470 [44:39<41:35,  1.27it/s]
1|3307|Loss: 0.7786672711372375:  51%|█████     | 3306/6470 [44:39<41:35,  1.27it/s]
1|3307|Loss: 0.7786672711372375:  51%|█████     | 3307/6470 [44:39<38:47,  1.36it/s]
1|3308|Loss: 1.5350080728530884:  51%|█████     | 3307/6470 [44:40<38:47,  1.36it/s]
1|3308|Loss: 1.5350080728530884:  51%|█████     | 3308/6470 [44:40<39:35,  1.33it/s]
1|3309|Loss: 1.0255476236343384:  51%|█████     | 3308/6470 [44:41<39:35,  1.33it/s]
1|3309|Loss: 1.0255476236343384:  51%|█████     | 3309/6470 [44:41<40:24,  1.30it/s]
1|3310|Loss: 0.9682657718658447:  51%|█████     | 3309/6470 [44:42<40:24,  1.30it/s]
1|3310|Loss: 0.9682657718658447:  51%|█████     | 3310/6470 [44:42<42:13,  1.25it/s]
1|3311|Loss: 1.1292271614074707:  51%|█████     | 3310/6470 [44:42<42:13,  1.25it/s]
1|3311|Loss: 1.1292271614074707:  51%|█████     | 3311/6470 [44:43<45:02,  1.17it/s]
1|3312|Loss: 1.146936058998108:  51%|█████     | 3311/6470 [44:43<45:02,  1.17it/s] 
1|3312|Loss: 1.146936058998108:  51%|█████     | 3312/6470 [44:44<44:31,  1.18it/s]
1|3313|Loss: 1.0306789875030518:  51%|█████     | 3312/6470 [44:44<44:31,  1.18it/s]
1|3313|Loss: 1.0306789875030518:  51%|█████     | 3313/6470 [44:45<43:46,  1.20it/s]
1|3314|Loss: 1.1248118877410889:  51%|█████     | 3313/6470 [44:45<43:46,  1.20it/s]
1|3314|Loss: 1.1248118877410889:  51%|█████     | 3314/6470 [44:45<41:08,  1.28it/s]
1|3315|Loss: 1.1393688917160034:  51%|█████     | 3314/6470 [44:46<41:08,  1.28it/s]
1|3315|Loss: 1.1393688917160034:  51%|█████     | 3315/6470 [44:46<40:43,  1.29it/s]
1|3316|Loss: 0.89398193359375:  51%|█████     | 3315/6470 [44:46<40:43,  1.29it/s]  
1|3316|Loss: 0.89398193359375:  51%|█████▏    | 3316/6470 [44:47<38:16,  1.37it/s]
1|3317|Loss: 1.0908398628234863:  51%|█████▏    | 3316/6470 [44:47<38:16,  1.37it/s]
1|3317|Loss: 1.0908398628234863:  51%|█████▏    | 3317/6470 [44:47<37:00,  1.42it/s]
1|3318|Loss: 1.3607664108276367:  51%|█████▏    | 3317/6470 [44:48<37:00,  1.42it/s]
1|3318|Loss: 1.3607664108276367:  51%|█████▏    | 3318/6470 [44:48<38:18,  1.37it/s]
1|3319|Loss: 1.1152385473251343:  51%|█████▏    | 3318/6470 [44:48<38:18,  1.37it/s]
1|3319|Loss: 1.1152385473251343:  51%|█████▏    | 3319/6470 [44:49<35:38,  1.47it/s]
1|3320|Loss: 0.9408291578292847:  51%|█████▏    | 3319/6470 [44:49<35:38,  1.47it/s]
1|3320|Loss: 0.9408291578292847:  51%|█████▏    | 3320/6470 [44:49<37:37,  1.40it/s]
1|3321|Loss: 1.24630606174469:  51%|█████▏    | 3320/6470 [44:50<37:37,  1.40it/s]  
1|3321|Loss: 1.24630606174469:  51%|█████▏    | 3321/6470 [44:50<38:33,  1.36it/s]
1|3322|Loss: 0.9443495869636536:  51%|█████▏    | 3321/6470 [44:51<38:33,  1.36it/s]
1|3322|Loss: 0.9443495869636536:  51%|█████▏    | 3322/6470 [44:51<39:40,  1.32it/s]
1|3323|Loss: 1.0644893646240234:  51%|█████▏    | 3322/6470 [44:51<39:40,  1.32it/s]
1|3323|Loss: 1.0644893646240234:  51%|█████▏    | 3323/6470 [44:52<39:09,  1.34it/s]
1|3324|Loss: 1.1438469886779785:  51%|█████▏    | 3323/6470 [44:52<39:09,  1.34it/s]
1|3324|Loss: 1.1438469886779785:  51%|█████▏    | 3324/6470 [44:53<41:27,  1.26it/s]
1|3325|Loss: 1.4970546960830688:  51%|█████▏    | 3324/6470 [44:53<41:27,  1.26it/s]
1|3325|Loss: 1.4970546960830688:  51%|█████▏    | 3325/6470 [44:53<43:18,  1.21it/s]
1|3326|Loss: 1.1118459701538086:  51%|█████▏    | 3325/6470 [44:54<43:18,  1.21it/s]
1|3326|Loss: 1.1118459701538086:  51%|█████▏    | 3326/6470 [44:54<43:50,  1.20it/s]
1|3327|Loss: 1.010830283164978:  51%|█████▏    | 3326/6470 [44:55<43:50,  1.20it/s] 
1|3327|Loss: 1.010830283164978:  51%|█████▏    | 3327/6470 [44:55<45:19,  1.16it/s]
1|3328|Loss: 0.896756649017334:  51%|█████▏    | 3327/6470 [44:56<45:19,  1.16it/s]
1|3328|Loss: 0.896756649017334:  51%|█████▏    | 3328/6470 [44:56<43:14,  1.21it/s]
1|3329|Loss: 0.9920284152030945:  51%|█████▏    | 3328/6470 [44:56<43:14,  1.21it/s]
1|3329|Loss: 0.9920284152030945:  51%|█████▏    | 3329/6470 [44:57<40:05,  1.31it/s]
1|3330|Loss: 0.9589347839355469:  51%|█████▏    | 3329/6470 [44:57<40:05,  1.31it/s]
1|3330|Loss: 0.9589347839355469:  51%|█████▏    | 3330/6470 [44:57<40:01,  1.31it/s]
1|3331|Loss: 1.0251842737197876:  51%|█████▏    | 3330/6470 [44:58<40:01,  1.31it/s]
1|3331|Loss: 1.0251842737197876:  51%|█████▏    | 3331/6470 [44:58<39:59,  1.31it/s]
1|3332|Loss: 1.1593619585037231:  51%|█████▏    | 3331/6470 [44:59<39:59,  1.31it/s]
1|3332|Loss: 1.1593619585037231:  51%|█████▏    | 3332/6470 [44:59<39:51,  1.31it/s]
1|3333|Loss: 1.1168146133422852:  51%|█████▏    | 3332/6470 [44:59<39:51,  1.31it/s]
1|3333|Loss: 1.1168146133422852:  52%|█████▏    | 3333/6470 [45:00<40:23,  1.29it/s]
1|3334|Loss: 0.875450849533081:  52%|█████▏    | 3333/6470 [45:00<40:23,  1.29it/s] 
1|3334|Loss: 0.875450849533081:  52%|█████▏    | 3334/6470 [45:00<39:10,  1.33it/s]
1|3335|Loss: 1.015572428703308:  52%|█████▏    | 3334/6470 [45:01<39:10,  1.33it/s]
1|3335|Loss: 1.015572428703308:  52%|█████▏    | 3335/6470 [45:01<40:59,  1.27it/s]
1|3336|Loss: 0.9889482855796814:  52%|█████▏    | 3335/6470 [45:02<40:59,  1.27it/s]
1|3336|Loss: 0.9889482855796814:  52%|█████▏    | 3336/6470 [45:02<43:19,  1.21it/s]
1|3337|Loss: 1.0602052211761475:  52%|█████▏    | 3336/6470 [45:03<43:19,  1.21it/s]
1|3337|Loss: 1.0602052211761475:  52%|█████▏    | 3337/6470 [45:03<41:41,  1.25it/s]
1|3338|Loss: 1.2242215871810913:  52%|█████▏    | 3337/6470 [45:03<41:41,  1.25it/s]
1|3338|Loss: 1.2242215871810913:  52%|█████▏    | 3338/6470 [45:04<41:21,  1.26it/s]
1|3339|Loss: 0.971785843372345:  52%|█████▏    | 3338/6470 [45:04<41:21,  1.26it/s] 
1|3339|Loss: 0.971785843372345:  52%|█████▏    | 3339/6470 [45:04<39:03,  1.34it/s]
1|3340|Loss: 0.7802767157554626:  52%|█████▏    | 3339/6470 [45:05<39:03,  1.34it/s]
1|3340|Loss: 0.7802767157554626:  52%|█████▏    | 3340/6470 [45:05<36:06,  1.44it/s]
1|3341|Loss: 0.9827266335487366:  52%|█████▏    | 3340/6470 [45:05<36:06,  1.44it/s]
1|3341|Loss: 0.9827266335487366:  52%|█████▏    | 3341/6470 [45:06<36:47,  1.42it/s]
1|3342|Loss: 0.9534185528755188:  52%|█████▏    | 3341/6470 [45:06<36:47,  1.42it/s]
1|3342|Loss: 0.9534185528755188:  52%|█████▏    | 3342/6470 [45:06<39:00,  1.34it/s]
1|3343|Loss: 1.0529030561447144:  52%|█████▏    | 3342/6470 [45:07<39:00,  1.34it/s]
1|3343|Loss: 1.0529030561447144:  52%|█████▏    | 3343/6470 [45:07<39:56,  1.30it/s]
1|3344|Loss: 0.843206524848938:  52%|█████▏    | 3343/6470 [45:08<39:56,  1.30it/s] 
1|3344|Loss: 0.843206524848938:  52%|█████▏    | 3344/6470 [45:08<38:14,  1.36it/s]
1|3345|Loss: 0.9834542870521545:  52%|█████▏    | 3344/6470 [45:08<38:14,  1.36it/s]
1|3345|Loss: 0.9834542870521545:  52%|█████▏    | 3345/6470 [45:09<38:08,  1.37it/s]
1|3346|Loss: 0.889614999294281:  52%|█████▏    | 3345/6470 [45:09<38:08,  1.37it/s] 
1|3346|Loss: 0.889614999294281:  52%|█████▏    | 3346/6470 [45:09<36:23,  1.43it/s]
1|3347|Loss: 1.2017195224761963:  52%|█████▏    | 3346/6470 [45:10<36:23,  1.43it/s]
1|3347|Loss: 1.2017195224761963:  52%|█████▏    | 3347/6470 [45:10<39:20,  1.32it/s]
1|3348|Loss: 1.2255016565322876:  52%|█████▏    | 3347/6470 [45:11<39:20,  1.32it/s]
1|3348|Loss: 1.2255016565322876:  52%|█████▏    | 3348/6470 [45:11<40:08,  1.30it/s]
1|3349|Loss: 1.0764557123184204:  52%|█████▏    | 3348/6470 [45:11<40:08,  1.30it/s]
1|3349|Loss: 1.0764557123184204:  52%|█████▏    | 3349/6470 [45:12<38:51,  1.34it/s]
1|3350|Loss: 1.0098387002944946:  52%|█████▏    | 3349/6470 [45:12<38:51,  1.34it/s]
1|3350|Loss: 1.0098387002944946:  52%|█████▏    | 3350/6470 [45:13<40:04,  1.30it/s]
1|3351|Loss: 1.0233502388000488:  52%|█████▏    | 3350/6470 [45:13<40:04,  1.30it/s]
1|3351|Loss: 1.0233502388000488:  52%|█████▏    | 3351/6470 [45:13<39:16,  1.32it/s]
1|3352|Loss: 0.9995951056480408:  52%|█████▏    | 3351/6470 [45:14<39:16,  1.32it/s]
1|3352|Loss: 0.9995951056480408:  52%|█████▏    | 3352/6470 [45:14<34:07,  1.52it/s]
1|3353|Loss: 0.9546133279800415:  52%|█████▏    | 3352/6470 [45:14<34:07,  1.52it/s]
1|3353|Loss: 0.9546133279800415:  52%|█████▏    | 3353/6470 [45:14<36:36,  1.42it/s]
1|3354|Loss: 0.9687222242355347:  52%|█████▏    | 3353/6470 [45:15<36:36,  1.42it/s]
1|3354|Loss: 0.9687222242355347:  52%|█████▏    | 3354/6470 [45:15<38:37,  1.34it/s]
1|3355|Loss: 1.1841390132904053:  52%|█████▏    | 3354/6470 [45:16<38:37,  1.34it/s]
1|3355|Loss: 1.1841390132904053:  52%|█████▏    | 3355/6470 [45:16<40:21,  1.29it/s]
1|3356|Loss: 1.1006031036376953:  52%|█████▏    | 3355/6470 [45:17<40:21,  1.29it/s]
1|3356|Loss: 1.1006031036376953:  52%|█████▏    | 3356/6470 [45:17<40:24,  1.28it/s]
1|3357|Loss: 1.2223399877548218:  52%|█████▏    | 3356/6470 [45:17<40:24,  1.28it/s]
1|3357|Loss: 1.2223399877548218:  52%|█████▏    | 3357/6470 [45:18<42:30,  1.22it/s]
1|3358|Loss: 0.9873940944671631:  52%|█████▏    | 3357/6470 [45:18<42:30,  1.22it/s]
1|3358|Loss: 0.9873940944671631:  52%|█████▏    | 3358/6470 [45:19<42:15,  1.23it/s]
1|3359|Loss: 1.0096381902694702:  52%|█████▏    | 3358/6470 [45:19<42:15,  1.23it/s]
1|3359|Loss: 1.0096381902694702:  52%|█████▏    | 3359/6470 [45:19<36:54,  1.40it/s]
1|3360|Loss: 1.042003870010376:  52%|█████▏    | 3359/6470 [45:20<36:54,  1.40it/s] 
1|3360|Loss: 1.042003870010376:  52%|█████▏    | 3360/6470 [45:20<36:26,  1.42it/s]
1|3361|Loss: 0.9601768255233765:  52%|█████▏    | 3360/6470 [45:20<36:26,  1.42it/s]
1|3361|Loss: 0.9601768255233765:  52%|█████▏    | 3361/6470 [45:20<35:08,  1.47it/s]
1|3362|Loss: 0.8536681532859802:  52%|█████▏    | 3361/6470 [45:21<35:08,  1.47it/s]
1|3362|Loss: 0.8536681532859802:  52%|█████▏    | 3362/6470 [45:21<35:27,  1.46it/s]
1|3363|Loss: 1.054939866065979:  52%|█████▏    | 3362/6470 [45:22<35:27,  1.46it/s] 
1|3363|Loss: 1.054939866065979:  52%|█████▏    | 3363/6470 [45:22<36:31,  1.42it/s]
1|3364|Loss: 1.11955726146698:  52%|█████▏    | 3363/6470 [45:22<36:31,  1.42it/s] 
1|3364|Loss: 1.11955726146698:  52%|█████▏    | 3364/6470 [45:23<37:46,  1.37it/s]
1|3365|Loss: 0.8054590821266174:  52%|█████▏    | 3364/6470 [45:23<37:46,  1.37it/s]
1|3365|Loss: 0.8054590821266174:  52%|█████▏    | 3365/6470 [45:23<38:16,  1.35it/s]
1|3366|Loss: 0.9058665037155151:  52%|█████▏    | 3365/6470 [45:24<38:16,  1.35it/s]
1|3366|Loss: 0.9058665037155151:  52%|█████▏    | 3366/6470 [45:24<38:27,  1.35it/s]
1|3367|Loss: 1.1950141191482544:  52%|█████▏    | 3366/6470 [45:25<38:27,  1.35it/s]
1|3367|Loss: 1.1950141191482544:  52%|█████▏    | 3367/6470 [45:25<35:53,  1.44it/s]
1|3368|Loss: 1.0259029865264893:  52%|█████▏    | 3367/6470 [45:25<35:53,  1.44it/s]
1|3368|Loss: 1.0259029865264893:  52%|█████▏    | 3368/6470 [45:25<34:59,  1.48it/s]
1|3369|Loss: 1.1285912990570068:  52%|█████▏    | 3368/6470 [45:26<34:59,  1.48it/s]
1|3369|Loss: 1.1285912990570068:  52%|█████▏    | 3369/6470 [45:26<37:27,  1.38it/s]
1|3370|Loss: 1.217624545097351:  52%|█████▏    | 3369/6470 [45:27<37:27,  1.38it/s] 
1|3370|Loss: 1.217624545097351:  52%|█████▏    | 3370/6470 [45:27<37:05,  1.39it/s]
1|3371|Loss: 0.8453729748725891:  52%|█████▏    | 3370/6470 [45:27<37:05,  1.39it/s]
1|3371|Loss: 0.8453729748725891:  52%|█████▏    | 3371/6470 [45:28<37:47,  1.37it/s]
1|3372|Loss: 0.9361215233802795:  52%|█████▏    | 3371/6470 [45:28<37:47,  1.37it/s]
1|3372|Loss: 0.9361215233802795:  52%|█████▏    | 3372/6470 [45:29<38:55,  1.33it/s]
1|3373|Loss: 0.9993235468864441:  52%|█████▏    | 3372/6470 [45:29<38:55,  1.33it/s]
1|3373|Loss: 0.9993235468864441:  52%|█████▏    | 3373/6470 [45:29<39:08,  1.32it/s]
1|3374|Loss: 0.9962802529335022:  52%|█████▏    | 3373/6470 [45:30<39:08,  1.32it/s]
1|3374|Loss: 0.9962802529335022:  52%|█████▏    | 3374/6470 [45:30<40:30,  1.27it/s]
1|3375|Loss: 1.011627435684204:  52%|█████▏    | 3374/6470 [45:31<40:30,  1.27it/s] 
1|3375|Loss: 1.011627435684204:  52%|█████▏    | 3375/6470 [45:31<40:24,  1.28it/s]
1|3376|Loss: 1.0018641948699951:  52%|█████▏    | 3375/6470 [45:31<40:24,  1.28it/s]
1|3376|Loss: 1.0018641948699951:  52%|█████▏    | 3376/6470 [45:32<38:51,  1.33it/s]
1|3377|Loss: 0.9621796011924744:  52%|█████▏    | 3376/6470 [45:32<38:51,  1.33it/s]
1|3377|Loss: 0.9621796011924744:  52%|█████▏    | 3377/6470 [45:32<35:55,  1.43it/s]
1|3378|Loss: 1.2126916646957397:  52%|█████▏    | 3377/6470 [45:33<35:55,  1.43it/s]
1|3378|Loss: 1.2126916646957397:  52%|█████▏    | 3378/6470 [45:33<37:36,  1.37it/s]
1|3379|Loss: 1.0451644659042358:  52%|█████▏    | 3378/6470 [45:33<37:36,  1.37it/s]
1|3379|Loss: 1.0451644659042358:  52%|█████▏    | 3379/6470 [45:34<38:46,  1.33it/s]
1|3380|Loss: 1.2179750204086304:  52%|█████▏    | 3379/6470 [45:34<38:46,  1.33it/s]
1|3380|Loss: 1.2179750204086304:  52%|█████▏    | 3380/6470 [45:35<39:42,  1.30it/s]
1|3381|Loss: 0.9752113223075867:  52%|█████▏    | 3380/6470 [45:35<39:42,  1.30it/s]
1|3381|Loss: 0.9752113223075867:  52%|█████▏    | 3381/6470 [45:35<39:38,  1.30it/s]
1|3382|Loss: 1.2316535711288452:  52%|█████▏    | 3381/6470 [45:36<39:38,  1.30it/s]
1|3382|Loss: 1.2316535711288452:  52%|█████▏    | 3382/6470 [45:36<39:54,  1.29it/s]
1|3383|Loss: 0.9464130997657776:  52%|█████▏    | 3382/6470 [45:37<39:54,  1.29it/s]
1|3383|Loss: 0.9464130997657776:  52%|█████▏    | 3383/6470 [45:37<40:11,  1.28it/s]
1|3384|Loss: 1.1240075826644897:  52%|█████▏    | 3383/6470 [45:37<40:11,  1.28it/s]
1|3384|Loss: 1.1240075826644897:  52%|█████▏    | 3384/6470 [45:38<41:30,  1.24it/s]
1|3385|Loss: 1.3177539110183716:  52%|█████▏    | 3384/6470 [45:38<41:30,  1.24it/s]
1|3385|Loss: 1.3177539110183716:  52%|█████▏    | 3385/6470 [45:39<43:01,  1.19it/s]
1|3386|Loss: 0.9869725704193115:  52%|█████▏    | 3385/6470 [45:39<43:01,  1.19it/s]
1|3386|Loss: 0.9869725704193115:  52%|█████▏    | 3386/6470 [45:40<44:05,  1.17it/s]
1|3387|Loss: 1.1581964492797852:  52%|█████▏    | 3386/6470 [45:40<44:05,  1.17it/s]
1|3387|Loss: 1.1581964492797852:  52%|█████▏    | 3387/6470 [45:40<41:20,  1.24it/s]
1|3388|Loss: 1.0074373483657837:  52%|█████▏    | 3387/6470 [45:41<41:20,  1.24it/s]
1|3388|Loss: 1.0074373483657837:  52%|█████▏    | 3388/6470 [45:41<40:08,  1.28it/s]
1|3389|Loss: 0.8898530602455139:  52%|█████▏    | 3388/6470 [45:41<40:08,  1.28it/s]
1|3389|Loss: 0.8898530602455139:  52%|█████▏    | 3389/6470 [45:42<38:42,  1.33it/s]
1|3390|Loss: 1.17997145652771:  52%|█████▏    | 3389/6470 [45:42<38:42,  1.33it/s]  
1|3390|Loss: 1.17997145652771:  52%|█████▏    | 3390/6470 [45:43<39:50,  1.29it/s]
1|3391|Loss: 1.1265379190444946:  52%|█████▏    | 3390/6470 [45:43<39:50,  1.29it/s]
1|3391|Loss: 1.1265379190444946:  52%|█████▏    | 3391/6470 [45:43<38:45,  1.32it/s]
1|3392|Loss: 0.9890360832214355:  52%|█████▏    | 3391/6470 [45:44<38:45,  1.32it/s]
1|3392|Loss: 0.9890360832214355:  52%|█████▏    | 3392/6470 [45:44<36:31,  1.40it/s]
1|3393|Loss: 0.9953480958938599:  52%|█████▏    | 3392/6470 [45:44<36:31,  1.40it/s]
1|3393|Loss: 0.9953480958938599:  52%|█████▏    | 3393/6470 [45:45<37:03,  1.38it/s]
1|3394|Loss: 1.1530859470367432:  52%|█████▏    | 3393/6470 [45:45<37:03,  1.38it/s]
1|3394|Loss: 1.1530859470367432:  52%|█████▏    | 3394/6470 [45:45<38:53,  1.32it/s]
1|3395|Loss: 1.3457399606704712:  52%|█████▏    | 3394/6470 [45:46<38:53,  1.32it/s]
1|3395|Loss: 1.3457399606704712:  52%|█████▏    | 3395/6470 [45:46<37:23,  1.37it/s]
1|3396|Loss: 0.9560723304748535:  52%|█████▏    | 3395/6470 [45:46<37:23,  1.37it/s]
1|3396|Loss: 0.9560723304748535:  52%|█████▏    | 3396/6470 [45:47<34:36,  1.48it/s]
1|3397|Loss: 0.9417317509651184:  52%|█████▏    | 3396/6470 [45:47<34:36,  1.48it/s]
1|3397|Loss: 0.9417317509651184:  53%|█████▎    | 3397/6470 [45:48<38:51,  1.32it/s]
1|3398|Loss: 0.9348440766334534:  53%|█████▎    | 3397/6470 [45:48<38:51,  1.32it/s]
1|3398|Loss: 0.9348440766334534:  53%|█████▎    | 3398/6470 [45:48<38:38,  1.32it/s]
1|3399|Loss: 1.0877094268798828:  53%|█████▎    | 3398/6470 [45:49<38:38,  1.32it/s]
1|3399|Loss: 1.0877094268798828:  53%|█████▎    | 3399/6470 [45:49<40:41,  1.26it/s]
1|3400|Loss: 1.044510006904602:  53%|█████▎    | 3399/6470 [45:50<40:41,  1.26it/s] 
1|3400|Loss: 1.044510006904602:  53%|█████▎    | 3400/6470 [45:50<40:37,  1.26it/s]
1|3401|Loss: 1.0065613985061646:  53%|█████▎    | 3400/6470 [45:50<40:37,  1.26it/s]
1|3401|Loss: 1.0065613985061646:  53%|█████▎    | 3401/6470 [45:51<38:53,  1.32it/s]
1|3402|Loss: 1.07614004611969:  53%|█████▎    | 3401/6470 [45:51<38:53,  1.32it/s]  
1|3402|Loss: 1.07614004611969:  53%|█████▎    | 3402/6470 [45:52<41:30,  1.23it/s]
1|3403|Loss: 0.8508577942848206:  53%|█████▎    | 3402/6470 [45:52<41:30,  1.23it/s]
1|3403|Loss: 0.8508577942848206:  53%|█████▎    | 3403/6470 [45:52<39:58,  1.28it/s]
1|3404|Loss: 1.107361078262329:  53%|█████▎    | 3403/6470 [45:53<39:58,  1.28it/s] 
1|3404|Loss: 1.107361078262329:  53%|█████▎    | 3404/6470 [45:53<38:23,  1.33it/s]
1|3405|Loss: 0.9694344997406006:  53%|█████▎    | 3404/6470 [45:53<38:23,  1.33it/s]
1|3405|Loss: 0.9694344997406006:  53%|█████▎    | 3405/6470 [45:54<37:40,  1.36it/s]
1|3406|Loss: 1.0147305727005005:  53%|█████▎    | 3405/6470 [45:54<37:40,  1.36it/s]
1|3406|Loss: 1.0147305727005005:  53%|█████▎    | 3406/6470 [45:55<40:01,  1.28it/s]
1|3407|Loss: 1.048443078994751:  53%|█████▎    | 3406/6470 [45:55<40:01,  1.28it/s] 
1|3407|Loss: 1.048443078994751:  53%|█████▎    | 3407/6470 [45:55<40:04,  1.27it/s]
1|3408|Loss: 0.885757565498352:  53%|█████▎    | 3407/6470 [45:56<40:04,  1.27it/s]
1|3408|Loss: 0.885757565498352:  53%|█████▎    | 3408/6470 [45:56<40:52,  1.25it/s]
1|3409|Loss: 0.9345561861991882:  53%|█████▎    | 3408/6470 [45:57<40:52,  1.25it/s]
1|3409|Loss: 0.9345561861991882:  53%|█████▎    | 3409/6470 [45:57<38:54,  1.31it/s]
1|3410|Loss: 0.9676838517189026:  53%|█████▎    | 3409/6470 [45:57<38:54,  1.31it/s]
1|3410|Loss: 0.9676838517189026:  53%|█████▎    | 3410/6470 [45:58<36:51,  1.38it/s]
1|3411|Loss: 1.0823551416397095:  53%|█████▎    | 3410/6470 [45:58<36:51,  1.38it/s]
1|3411|Loss: 1.0823551416397095:  53%|█████▎    | 3411/6470 [45:58<38:44,  1.32it/s]
1|3412|Loss: 1.2403615713119507:  53%|█████▎    | 3411/6470 [45:59<38:44,  1.32it/s]
1|3412|Loss: 1.2403615713119507:  53%|█████▎    | 3412/6470 [45:59<39:09,  1.30it/s]
1|3413|Loss: 0.9711769819259644:  53%|█████▎    | 3412/6470 [46:00<39:09,  1.30it/s]
1|3413|Loss: 0.9711769819259644:  53%|█████▎    | 3413/6470 [46:00<36:07,  1.41it/s]
1|3414|Loss: 1.0664000511169434:  53%|█████▎    | 3413/6470 [46:00<36:07,  1.41it/s]
1|3414|Loss: 1.0664000511169434:  53%|█████▎    | 3414/6470 [46:01<37:32,  1.36it/s]
1|3415|Loss: 0.9920541048049927:  53%|█████▎    | 3414/6470 [46:01<37:32,  1.36it/s]
1|3415|Loss: 0.9920541048049927:  53%|█████▎    | 3415/6470 [46:02<40:11,  1.27it/s]
1|3416|Loss: 1.186386227607727:  53%|█████▎    | 3415/6470 [46:02<40:11,  1.27it/s] 
1|3416|Loss: 1.186386227607727:  53%|█████▎    | 3416/6470 [46:02<41:21,  1.23it/s]
1|3417|Loss: 1.2265985012054443:  53%|█████▎    | 3416/6470 [46:03<41:21,  1.23it/s]
1|3417|Loss: 1.2265985012054443:  53%|█████▎    | 3417/6470 [46:03<43:23,  1.17it/s]
1|3418|Loss: 0.7978836297988892:  53%|█████▎    | 3417/6470 [46:04<43:23,  1.17it/s]
1|3418|Loss: 0.7978836297988892:  53%|█████▎    | 3418/6470 [46:04<39:39,  1.28it/s]
1|3419|Loss: 1.0285285711288452:  53%|█████▎    | 3418/6470 [46:04<39:39,  1.28it/s]
1|3419|Loss: 1.0285285711288452:  53%|█████▎    | 3419/6470 [46:05<37:44,  1.35it/s]
1|3420|Loss: 0.9340096116065979:  53%|█████▎    | 3419/6470 [46:05<37:44,  1.35it/s]
1|3420|Loss: 0.9340096116065979:  53%|█████▎    | 3420/6470 [46:06<40:47,  1.25it/s]
1|3421|Loss: 1.0442041158676147:  53%|█████▎    | 3420/6470 [46:06<40:47,  1.25it/s]
1|3421|Loss: 1.0442041158676147:  53%|█████▎    | 3421/6470 [46:06<40:07,  1.27it/s]
1|3422|Loss: 1.2017664909362793:  53%|█████▎    | 3421/6470 [46:07<40:07,  1.27it/s]
1|3422|Loss: 1.2017664909362793:  53%|█████▎    | 3422/6470 [46:07<42:00,  1.21it/s]
1|3423|Loss: 1.0698730945587158:  53%|█████▎    | 3422/6470 [46:08<42:00,  1.21it/s]
1|3423|Loss: 1.0698730945587158:  53%|█████▎    | 3423/6470 [46:08<38:15,  1.33it/s]
1|3424|Loss: 0.9135891795158386:  53%|█████▎    | 3423/6470 [46:08<38:15,  1.33it/s]
1|3424|Loss: 0.9135891795158386:  53%|█████▎    | 3424/6470 [46:08<35:36,  1.43it/s]
1|3425|Loss: 0.9412429928779602:  53%|█████▎    | 3424/6470 [46:09<35:36,  1.43it/s]
1|3425|Loss: 0.9412429928779602:  53%|█████▎    | 3425/6470 [46:09<36:32,  1.39it/s]
1|3426|Loss: 0.9793176054954529:  53%|█████▎    | 3425/6470 [46:10<36:32,  1.39it/s]
1|3426|Loss: 0.9793176054954529:  53%|█████▎    | 3426/6470 [46:10<35:16,  1.44it/s]
1|3427|Loss: 0.889146089553833:  53%|█████▎    | 3426/6470 [46:10<35:16,  1.44it/s] 
1|3427|Loss: 0.889146089553833:  53%|█████▎    | 3427/6470 [46:11<36:06,  1.40it/s]
1|3428|Loss: 1.031406283378601:  53%|█████▎    | 3427/6470 [46:11<36:06,  1.40it/s]
1|3428|Loss: 1.031406283378601:  53%|█████▎    | 3428/6470 [46:11<36:48,  1.38it/s]
1|3429|Loss: 1.0466439723968506:  53%|█████▎    | 3428/6470 [46:12<36:48,  1.38it/s]
1|3429|Loss: 1.0466439723968506:  53%|█████▎    | 3429/6470 [46:12<37:25,  1.35it/s]
1|3430|Loss: 1.0242352485656738:  53%|█████▎    | 3429/6470 [46:12<37:25,  1.35it/s]
1|3430|Loss: 1.0242352485656738:  53%|█████▎    | 3430/6470 [46:13<39:01,  1.30it/s]
1|3431|Loss: 1.0241392850875854:  53%|█████▎    | 3430/6470 [46:13<39:01,  1.30it/s]
1|3431|Loss: 1.0241392850875854:  53%|█████▎    | 3431/6470 [46:14<39:31,  1.28it/s]
1|3432|Loss: 1.0112354755401611:  53%|█████▎    | 3431/6470 [46:14<39:31,  1.28it/s]
1|3432|Loss: 1.0112354755401611:  53%|█████▎    | 3432/6470 [46:14<35:50,  1.41it/s]
1|3433|Loss: 0.9498799443244934:  53%|█████▎    | 3432/6470 [46:15<35:50,  1.41it/s]
1|3433|Loss: 0.9498799443244934:  53%|█████▎    | 3433/6470 [46:15<35:27,  1.43it/s]
1|3434|Loss: 1.022722601890564:  53%|█████▎    | 3433/6470 [46:15<35:27,  1.43it/s] 
1|3434|Loss: 1.022722601890564:  53%|█████▎    | 3434/6470 [46:16<35:19,  1.43it/s]
1|3435|Loss: 1.051021933555603:  53%|█████▎    | 3434/6470 [46:16<35:19,  1.43it/s]
1|3435|Loss: 1.051021933555603:  53%|█████▎    | 3435/6470 [46:16<36:32,  1.38it/s]
1|3436|Loss: 0.9432162046432495:  53%|█████▎    | 3435/6470 [46:17<36:32,  1.38it/s]
1|3436|Loss: 0.9432162046432495:  53%|█████▎    | 3436/6470 [46:17<36:29,  1.39it/s]
1|3437|Loss: 0.9207531809806824:  53%|█████▎    | 3436/6470 [46:18<36:29,  1.39it/s]
1|3437|Loss: 0.9207531809806824:  53%|█████▎    | 3437/6470 [46:18<36:16,  1.39it/s]
1|3438|Loss: 0.9170055389404297:  53%|█████▎    | 3437/6470 [46:18<36:16,  1.39it/s]
1|3438|Loss: 0.9170055389404297:  53%|█████▎    | 3438/6470 [46:18<33:12,  1.52it/s]
1|3439|Loss: 0.941419780254364:  53%|█████▎    | 3438/6470 [46:19<33:12,  1.52it/s] 
1|3439|Loss: 0.941419780254364:  53%|█████▎    | 3439/6470 [46:19<32:59,  1.53it/s]
1|3440|Loss: 0.8499591946601868:  53%|█████▎    | 3439/6470 [46:19<32:59,  1.53it/s]
1|3440|Loss: 0.8499591946601868:  53%|█████▎    | 3440/6470 [46:20<36:05,  1.40it/s]
1|3441|Loss: 1.152298092842102:  53%|█████▎    | 3440/6470 [46:20<36:05,  1.40it/s] 
1|3441|Loss: 1.152298092842102:  53%|█████▎    | 3441/6470 [46:21<35:57,  1.40it/s]
1|3442|Loss: 1.0424171686172485:  53%|█████▎    | 3441/6470 [46:21<35:57,  1.40it/s]
1|3442|Loss: 1.0424171686172485:  53%|█████▎    | 3442/6470 [46:21<36:12,  1.39it/s]
1|3443|Loss: 0.8100239038467407:  53%|█████▎    | 3442/6470 [46:22<36:12,  1.39it/s]
1|3443|Loss: 0.8100239038467407:  53%|█████▎    | 3443/6470 [46:22<33:18,  1.51it/s]
1|3444|Loss: 0.9371211528778076:  53%|█████▎    | 3443/6470 [46:22<33:18,  1.51it/s]
1|3444|Loss: 0.9371211528778076:  53%|█████▎    | 3444/6470 [46:23<36:42,  1.37it/s]
1|3445|Loss: 1.0599693059921265:  53%|█████▎    | 3444/6470 [46:23<36:42,  1.37it/s]
1|3445|Loss: 1.0599693059921265:  53%|█████▎    | 3445/6470 [46:24<38:35,  1.31it/s]
1|3446|Loss: 1.0141880512237549:  53%|█████▎    | 3445/6470 [46:24<38:35,  1.31it/s]
1|3446|Loss: 1.0141880512237549:  53%|█████▎    | 3446/6470 [46:24<37:19,  1.35it/s]
1|3447|Loss: 1.049673080444336:  53%|█████▎    | 3446/6470 [46:25<37:19,  1.35it/s] 
1|3447|Loss: 1.049673080444336:  53%|█████▎    | 3447/6470 [46:25<35:50,  1.41it/s]
1|3448|Loss: 0.9289473295211792:  53%|█████▎    | 3447/6470 [46:25<35:50,  1.41it/s]
1|3448|Loss: 0.9289473295211792:  53%|█████▎    | 3448/6470 [46:26<34:59,  1.44it/s]
1|3449|Loss: 0.9140937328338623:  53%|█████▎    | 3448/6470 [46:26<34:59,  1.44it/s]
1|3449|Loss: 0.9140937328338623:  53%|█████▎    | 3449/6470 [46:26<35:52,  1.40it/s]
1|3450|Loss: 1.0247136354446411:  53%|█████▎    | 3449/6470 [46:27<35:52,  1.40it/s]
1|3450|Loss: 1.0247136354446411:  53%|█████▎    | 3450/6470 [46:27<35:32,  1.42it/s]
1|3451|Loss: 1.1509066820144653:  53%|█████▎    | 3450/6470 [46:27<35:32,  1.42it/s]
1|3451|Loss: 1.1509066820144653:  53%|█████▎    | 3451/6470 [46:28<35:14,  1.43it/s]
1|3452|Loss: 1.0692869424819946:  53%|█████▎    | 3451/6470 [46:28<35:14,  1.43it/s]
1|3452|Loss: 1.0692869424819946:  53%|█████▎    | 3452/6470 [46:29<37:50,  1.33it/s]
1|3453|Loss: 0.8885779976844788:  53%|█████▎    | 3452/6470 [46:29<37:50,  1.33it/s]
1|3453|Loss: 0.8885779976844788:  53%|█████▎    | 3453/6470 [46:29<36:54,  1.36it/s]
1|3454|Loss: 0.9086613655090332:  53%|█████▎    | 3453/6470 [46:30<36:54,  1.36it/s]
1|3454|Loss: 0.9086613655090332:  53%|█████▎    | 3454/6470 [46:30<36:09,  1.39it/s]
1|3455|Loss: 0.8907963633537292:  53%|█████▎    | 3454/6470 [46:30<36:09,  1.39it/s]
1|3455|Loss: 0.8907963633537292:  53%|█████▎    | 3455/6470 [46:31<37:01,  1.36it/s]
1|3456|Loss: 0.9757050275802612:  53%|█████▎    | 3455/6470 [46:31<37:01,  1.36it/s]
1|3456|Loss: 0.9757050275802612:  53%|█████▎    | 3456/6470 [46:32<39:03,  1.29it/s]
1|3457|Loss: 0.7553295493125916:  53%|█████▎    | 3456/6470 [46:32<39:03,  1.29it/s]
1|3457|Loss: 0.7553295493125916:  53%|█████▎    | 3457/6470 [46:32<37:36,  1.34it/s]
1|3458|Loss: 0.8917249441146851:  53%|█████▎    | 3457/6470 [46:33<37:36,  1.34it/s]
1|3458|Loss: 0.8917249441146851:  53%|█████▎    | 3458/6470 [46:33<36:33,  1.37it/s]
1|3459|Loss: 1.074318766593933:  53%|█████▎    | 3458/6470 [46:33<36:33,  1.37it/s] 
1|3459|Loss: 1.074318766593933:  53%|█████▎    | 3459/6470 [46:34<37:15,  1.35it/s]
1|3460|Loss: 0.8846654295921326:  53%|█████▎    | 3459/6470 [46:34<37:15,  1.35it/s]
1|3460|Loss: 0.8846654295921326:  53%|█████▎    | 3460/6470 [46:35<40:51,  1.23it/s]
1|3461|Loss: 0.9978212118148804:  53%|█████▎    | 3460/6470 [46:35<40:51,  1.23it/s]
1|3461|Loss: 0.9978212118148804:  53%|█████▎    | 3461/6470 [46:36<42:54,  1.17it/s]
1|3462|Loss: 1.0340290069580078:  53%|█████▎    | 3461/6470 [46:36<42:54,  1.17it/s]
1|3462|Loss: 1.0340290069580078:  54%|█████▎    | 3462/6470 [46:36<42:33,  1.18it/s]
1|3463|Loss: 0.968862771987915:  54%|█████▎    | 3462/6470 [46:37<42:33,  1.18it/s] 
1|3463|Loss: 0.968862771987915:  54%|█████▎    | 3463/6470 [46:37<40:28,  1.24it/s]
1|3464|Loss: 1.0662627220153809:  54%|█████▎    | 3463/6470 [46:38<40:28,  1.24it/s]
1|3464|Loss: 1.0662627220153809:  54%|█████▎    | 3464/6470 [46:38<36:58,  1.36it/s]
1|3465|Loss: 1.0238516330718994:  54%|█████▎    | 3464/6470 [46:38<36:58,  1.36it/s]
1|3465|Loss: 1.0238516330718994:  54%|█████▎    | 3465/6470 [46:38<35:25,  1.41it/s]
1|3466|Loss: 0.9287279844284058:  54%|█████▎    | 3465/6470 [46:39<35:25,  1.41it/s]
1|3466|Loss: 0.9287279844284058:  54%|█████▎    | 3466/6470 [46:39<33:50,  1.48it/s]
1|3467|Loss: 0.9255613684654236:  54%|█████▎    | 3466/6470 [46:39<33:50,  1.48it/s]
1|3467|Loss: 0.9255613684654236:  54%|█████▎    | 3467/6470 [46:40<34:55,  1.43it/s]
1|3468|Loss: 1.0674545764923096:  54%|█████▎    | 3467/6470 [46:40<34:55,  1.43it/s]
1|3468|Loss: 1.0674545764923096:  54%|█████▎    | 3468/6470 [46:41<36:43,  1.36it/s]
1|3469|Loss: 0.9991432428359985:  54%|█████▎    | 3468/6470 [46:41<36:43,  1.36it/s]
1|3469|Loss: 0.9991432428359985:  54%|█████▎    | 3469/6470 [46:41<39:02,  1.28it/s]
1|3470|Loss: 1.0555131435394287:  54%|█████▎    | 3469/6470 [46:42<39:02,  1.28it/s]
1|3470|Loss: 1.0555131435394287:  54%|█████▎    | 3470/6470 [46:42<41:21,  1.21it/s]
1|3471|Loss: 1.0052324533462524:  54%|█████▎    | 3470/6470 [46:43<41:21,  1.21it/s]
1|3471|Loss: 1.0052324533462524:  54%|█████▎    | 3471/6470 [46:43<38:07,  1.31it/s]
1|3472|Loss: 0.6991512775421143:  54%|█████▎    | 3471/6470 [46:43<38:07,  1.31it/s]
1|3472|Loss: 0.6991512775421143:  54%|█████▎    | 3472/6470 [46:44<37:37,  1.33it/s]
1|3473|Loss: 1.0045629739761353:  54%|█████▎    | 3472/6470 [46:44<37:37,  1.33it/s]
1|3473|Loss: 1.0045629739761353:  54%|█████▎    | 3473/6470 [46:45<38:25,  1.30it/s]
1|3474|Loss: 0.9771724939346313:  54%|█████▎    | 3473/6470 [46:45<38:25,  1.30it/s]
1|3474|Loss: 0.9771724939346313:  54%|█████▎    | 3474/6470 [46:45<37:34,  1.33it/s]
1|3475|Loss: 0.9436652064323425:  54%|█████▎    | 3474/6470 [46:46<37:34,  1.33it/s]
1|3475|Loss: 0.9436652064323425:  54%|█████▎    | 3475/6470 [46:46<37:43,  1.32it/s]
1|3476|Loss: 1.0437029600143433:  54%|█████▎    | 3475/6470 [46:46<37:43,  1.32it/s]
1|3476|Loss: 1.0437029600143433:  54%|█████▎    | 3476/6470 [46:47<36:20,  1.37it/s]
1|3477|Loss: 1.2171549797058105:  54%|█████▎    | 3476/6470 [46:47<36:20,  1.37it/s]
1|3477|Loss: 1.2171549797058105:  54%|█████▎    | 3477/6470 [46:47<34:52,  1.43it/s]
1|3478|Loss: 0.9637666344642639:  54%|█████▎    | 3477/6470 [46:48<34:52,  1.43it/s]
1|3478|Loss: 0.9637666344642639:  54%|█████▍    | 3478/6470 [46:48<35:15,  1.41it/s]
1|3479|Loss: 1.6833679676055908:  54%|█████▍    | 3478/6470 [46:48<35:15,  1.41it/s]
1|3479|Loss: 1.6833679676055908:  54%|█████▍    | 3479/6470 [46:49<37:45,  1.32it/s]
1|3480|Loss: 0.81068354845047:  54%|█████▍    | 3479/6470 [46:49<37:45,  1.32it/s]  
1|3480|Loss: 0.81068354845047:  54%|█████▍    | 3480/6470 [46:50<38:08,  1.31it/s]
1|3481|Loss: 0.9723728895187378:  54%|█████▍    | 3480/6470 [46:50<38:08,  1.31it/s]
1|3481|Loss: 0.9723728895187378:  54%|█████▍    | 3481/6470 [46:50<37:15,  1.34it/s]
1|3482|Loss: 0.9082149863243103:  54%|█████▍    | 3481/6470 [46:51<37:15,  1.34it/s]
1|3482|Loss: 0.9082149863243103:  54%|█████▍    | 3482/6470 [46:51<34:34,  1.44it/s]
1|3483|Loss: 0.9869406819343567:  54%|█████▍    | 3482/6470 [46:51<34:34,  1.44it/s]
1|3483|Loss: 0.9869406819343567:  54%|█████▍    | 3483/6470 [46:52<36:52,  1.35it/s]
1|3484|Loss: 0.9765462875366211:  54%|█████▍    | 3483/6470 [46:52<36:52,  1.35it/s]
1|3484|Loss: 0.9765462875366211:  54%|█████▍    | 3484/6470 [46:53<38:13,  1.30it/s]
1|3485|Loss: 1.0711396932601929:  54%|█████▍    | 3484/6470 [46:53<38:13,  1.30it/s]
1|3485|Loss: 1.0711396932601929:  54%|█████▍    | 3485/6470 [46:53<36:01,  1.38it/s]
1|3486|Loss: 1.0004465579986572:  54%|█████▍    | 3485/6470 [46:54<36:01,  1.38it/s]
1|3486|Loss: 1.0004465579986572:  54%|█████▍    | 3486/6470 [46:54<35:47,  1.39it/s]
1|3487|Loss: 0.972743570804596:  54%|█████▍    | 3486/6470 [46:54<35:47,  1.39it/s] 
1|3487|Loss: 0.972743570804596:  54%|█████▍    | 3487/6470 [46:55<34:53,  1.42it/s]
1|3488|Loss: 1.0337952375411987:  54%|█████▍    | 3487/6470 [46:55<34:53,  1.42it/s]
1|3488|Loss: 1.0337952375411987:  54%|█████▍    | 3488/6470 [46:55<35:08,  1.41it/s]
1|3489|Loss: 1.0099231004714966:  54%|█████▍    | 3488/6470 [46:56<35:08,  1.41it/s]
1|3489|Loss: 1.0099231004714966:  54%|█████▍    | 3489/6470 [46:56<37:11,  1.34it/s]
1|3490|Loss: 0.9637564420700073:  54%|█████▍    | 3489/6470 [46:57<37:11,  1.34it/s]
1|3490|Loss: 0.9637564420700073:  54%|█████▍    | 3490/6470 [46:57<37:33,  1.32it/s]
1|3491|Loss: 0.8811147212982178:  54%|█████▍    | 3490/6470 [46:57<37:33,  1.32it/s]
1|3491|Loss: 0.8811147212982178:  54%|█████▍    | 3491/6470 [46:58<37:32,  1.32it/s]
1|3492|Loss: 1.0195777416229248:  54%|█████▍    | 3491/6470 [46:58<37:32,  1.32it/s]
1|3492|Loss: 1.0195777416229248:  54%|█████▍    | 3492/6470 [46:58<36:09,  1.37it/s]
1|3493|Loss: 1.0862255096435547:  54%|█████▍    | 3492/6470 [46:59<36:09,  1.37it/s]
1|3493|Loss: 1.0862255096435547:  54%|█████▍    | 3493/6470 [46:59<36:30,  1.36it/s]
1|3494|Loss: 0.9703920483589172:  54%|█████▍    | 3493/6470 [47:00<36:30,  1.36it/s]
1|3494|Loss: 0.9703920483589172:  54%|█████▍    | 3494/6470 [47:00<34:23,  1.44it/s]
1|3495|Loss: 0.8915467858314514:  54%|█████▍    | 3494/6470 [47:00<34:23,  1.44it/s]
1|3495|Loss: 0.8915467858314514:  54%|█████▍    | 3495/6470 [47:00<34:59,  1.42it/s]
1|3496|Loss: 1.0609734058380127:  54%|█████▍    | 3495/6470 [47:01<34:59,  1.42it/s]
1|3496|Loss: 1.0609734058380127:  54%|█████▍    | 3496/6470 [47:01<35:49,  1.38it/s]
1|3497|Loss: 0.9965957403182983:  54%|█████▍    | 3496/6470 [47:02<35:49,  1.38it/s]
1|3497|Loss: 0.9965957403182983:  54%|█████▍    | 3497/6470 [47:02<36:12,  1.37it/s]
1|3498|Loss: 1.0079431533813477:  54%|█████▍    | 3497/6470 [47:02<36:12,  1.37it/s]
1|3498|Loss: 1.0079431533813477:  54%|█████▍    | 3498/6470 [47:03<34:52,  1.42it/s]
1|3499|Loss: 0.882373571395874:  54%|█████▍    | 3498/6470 [47:03<34:52,  1.42it/s] 
1|3499|Loss: 0.882373571395874:  54%|█████▍    | 3499/6470 [47:03<34:38,  1.43it/s]
1|3500|Loss: 0.8240276575088501:  54%|█████▍    | 3499/6470 [47:04<34:38,  1.43it/s]
1|3500|Loss: 0.8240276575088501:  54%|█████▍    | 3500/6470 [47:04<33:24,  1.48it/s]
1|3501|Loss: 0.9464861750602722:  54%|█████▍    | 3500/6470 [47:04<33:24,  1.48it/s]
1|3501|Loss: 0.9464861750602722:  54%|█████▍    | 3501/6470 [47:05<34:36,  1.43it/s]
1|3502|Loss: 0.9725654721260071:  54%|█████▍    | 3501/6470 [47:05<34:36,  1.43it/s]
1|3502|Loss: 0.9725654721260071:  54%|█████▍    | 3502/6470 [47:05<35:31,  1.39it/s]
1|3503|Loss: 0.877839982509613:  54%|█████▍    | 3502/6470 [47:06<35:31,  1.39it/s] 
1|3503|Loss: 0.877839982509613:  54%|█████▍    | 3503/6470 [47:06<34:06,  1.45it/s]
1|3504|Loss: 0.9432814121246338:  54%|█████▍    | 3503/6470 [47:06<34:06,  1.45it/s]
1|3504|Loss: 0.9432814121246338:  54%|█████▍    | 3504/6470 [47:07<32:54,  1.50it/s]
1|3505|Loss: 0.8998467326164246:  54%|█████▍    | 3504/6470 [47:07<32:54,  1.50it/s]
1|3505|Loss: 0.8998467326164246:  54%|█████▍    | 3505/6470 [47:08<35:36,  1.39it/s]
1|3506|Loss: 0.9691839218139648:  54%|█████▍    | 3505/6470 [47:08<35:36,  1.39it/s]
1|3506|Loss: 0.9691839218139648:  54%|█████▍    | 3506/6470 [47:08<37:16,  1.33it/s]
1|3507|Loss: 1.0311733484268188:  54%|█████▍    | 3506/6470 [47:09<37:16,  1.33it/s]
1|3507|Loss: 1.0311733484268188:  54%|█████▍    | 3507/6470 [47:09<36:49,  1.34it/s]
1|3508|Loss: 0.9548044204711914:  54%|█████▍    | 3507/6470 [47:09<36:49,  1.34it/s]
1|3508|Loss: 0.9548044204711914:  54%|█████▍    | 3508/6470 [47:10<34:34,  1.43it/s]
1|3509|Loss: 1.0867536067962646:  54%|█████▍    | 3508/6470 [47:10<34:34,  1.43it/s]
1|3509|Loss: 1.0867536067962646:  54%|█████▍    | 3509/6470 [47:10<34:03,  1.45it/s]
1|3510|Loss: 1.04092276096344:  54%|█████▍    | 3509/6470 [47:11<34:03,  1.45it/s]  
1|3510|Loss: 1.04092276096344:  54%|█████▍    | 3510/6470 [47:11<30:45,  1.60it/s]
1|3511|Loss: 1.178027868270874:  54%|█████▍    | 3510/6470 [47:11<30:45,  1.60it/s]
1|3511|Loss: 1.178027868270874:  54%|█████▍    | 3511/6470 [47:12<33:04,  1.49it/s]
1|3512|Loss: 1.2216253280639648:  54%|█████▍    | 3511/6470 [47:12<33:04,  1.49it/s]
1|3512|Loss: 1.2216253280639648:  54%|█████▍    | 3512/6470 [47:12<35:06,  1.40it/s]
1|3513|Loss: 0.9286174774169922:  54%|█████▍    | 3512/6470 [47:13<35:06,  1.40it/s]
1|3513|Loss: 0.9286174774169922:  54%|█████▍    | 3513/6470 [47:13<33:44,  1.46it/s]
1|3514|Loss: 0.7822351455688477:  54%|█████▍    | 3513/6470 [47:13<33:44,  1.46it/s]
1|3514|Loss: 0.7822351455688477:  54%|█████▍    | 3514/6470 [47:14<35:24,  1.39it/s]
1|3515|Loss: 1.142807126045227:  54%|█████▍    | 3514/6470 [47:14<35:24,  1.39it/s] 
1|3515|Loss: 1.142807126045227:  54%|█████▍    | 3515/6470 [47:15<36:46,  1.34it/s]
1|3516|Loss: 0.8361405730247498:  54%|█████▍    | 3515/6470 [47:15<36:46,  1.34it/s]
1|3516|Loss: 0.8361405730247498:  54%|█████▍    | 3516/6470 [47:16<38:31,  1.28it/s]
1|3517|Loss: 1.2935690879821777:  54%|█████▍    | 3516/6470 [47:16<38:31,  1.28it/s]
1|3517|Loss: 1.2935690879821777:  54%|█████▍    | 3517/6470 [47:16<37:23,  1.32it/s]
1|3518|Loss: 1.0115584135055542:  54%|█████▍    | 3517/6470 [47:17<37:23,  1.32it/s]
1|3518|Loss: 1.0115584135055542:  54%|█████▍    | 3518/6470 [47:17<35:58,  1.37it/s]
1|3519|Loss: 0.8839772939682007:  54%|█████▍    | 3518/6470 [47:17<35:58,  1.37it/s]
1|3519|Loss: 0.8839772939682007:  54%|█████▍    | 3519/6470 [47:17<33:06,  1.49it/s]
1|3520|Loss: 1.2636398077011108:  54%|█████▍    | 3519/6470 [47:18<33:06,  1.49it/s]
1|3520|Loss: 1.2636398077011108:  54%|█████▍    | 3520/6470 [47:18<34:05,  1.44it/s]
1|3521|Loss: 1.1562020778656006:  54%|█████▍    | 3520/6470 [47:19<34:05,  1.44it/s]
1|3521|Loss: 1.1562020778656006:  54%|█████▍    | 3521/6470 [47:19<37:06,  1.32it/s]
1|3522|Loss: 0.9178339242935181:  54%|█████▍    | 3521/6470 [47:20<37:06,  1.32it/s]
1|3522|Loss: 0.9178339242935181:  54%|█████▍    | 3522/6470 [47:20<37:05,  1.32it/s]
1|3523|Loss: 0.9043560028076172:  54%|█████▍    | 3522/6470 [47:20<37:05,  1.32it/s]
1|3523|Loss: 0.9043560028076172:  54%|█████▍    | 3523/6470 [47:20<35:49,  1.37it/s]
1|3524|Loss: 0.9527362585067749:  54%|█████▍    | 3523/6470 [47:21<35:49,  1.37it/s]
1|3524|Loss: 0.9527362585067749:  54%|█████▍    | 3524/6470 [47:21<34:49,  1.41it/s]
1|3525|Loss: 0.9373708963394165:  54%|█████▍    | 3524/6470 [47:21<34:49,  1.41it/s]
1|3525|Loss: 0.9373708963394165:  54%|█████▍    | 3525/6470 [47:22<33:34,  1.46it/s]
1|3526|Loss: 0.9850876927375793:  54%|█████▍    | 3525/6470 [47:22<33:34,  1.46it/s]
1|3526|Loss: 0.9850876927375793:  54%|█████▍    | 3526/6470 [47:22<33:37,  1.46it/s]
1|3527|Loss: 0.8964616060256958:  54%|█████▍    | 3526/6470 [47:23<33:37,  1.46it/s]
1|3527|Loss: 0.8964616060256958:  55%|█████▍    | 3527/6470 [47:23<34:33,  1.42it/s]
1|3528|Loss: 1.0118824243545532:  55%|█████▍    | 3527/6470 [47:24<34:33,  1.42it/s]
1|3528|Loss: 1.0118824243545532:  55%|█████▍    | 3528/6470 [47:24<34:33,  1.42it/s]
1|3529|Loss: 0.8442233204841614:  55%|█████▍    | 3528/6470 [47:24<34:33,  1.42it/s]
1|3529|Loss: 0.8442233204841614:  55%|█████▍    | 3529/6470 [47:25<35:08,  1.40it/s]
1|3530|Loss: 0.8381364941596985:  55%|█████▍    | 3529/6470 [47:25<35:08,  1.40it/s]
1|3530|Loss: 0.8381364941596985:  55%|█████▍    | 3530/6470 [47:25<36:14,  1.35it/s]
1|3531|Loss: 0.989130437374115:  55%|█████▍    | 3530/6470 [47:26<36:14,  1.35it/s] 
1|3531|Loss: 0.989130437374115:  55%|█████▍    | 3531/6470 [47:26<37:55,  1.29it/s]
1|3532|Loss: 0.959723949432373:  55%|█████▍    | 3531/6470 [47:27<37:55,  1.29it/s]
1|3532|Loss: 0.959723949432373:  55%|█████▍    | 3532/6470 [47:27<35:29,  1.38it/s]
1|3533|Loss: 1.0199267864227295:  55%|█████▍    | 3532/6470 [47:27<35:29,  1.38it/s]
1|3533|Loss: 1.0199267864227295:  55%|█████▍    | 3533/6470 [47:28<33:51,  1.45it/s]
1|3534|Loss: 0.8850882649421692:  55%|█████▍    | 3533/6470 [47:28<33:51,  1.45it/s]
1|3534|Loss: 0.8850882649421692:  55%|█████▍    | 3534/6470 [47:28<33:12,  1.47it/s]
1|3535|Loss: 0.8933812975883484:  55%|█████▍    | 3534/6470 [47:29<33:12,  1.47it/s]
1|3535|Loss: 0.8933812975883484:  55%|█████▍    | 3535/6470 [47:29<33:28,  1.46it/s]
1|3536|Loss: 1.0210593938827515:  55%|█████▍    | 3535/6470 [47:29<33:28,  1.46it/s]
1|3536|Loss: 1.0210593938827515:  55%|█████▍    | 3536/6470 [47:30<34:14,  1.43it/s]
1|3537|Loss: 0.934192955493927:  55%|█████▍    | 3536/6470 [47:30<34:14,  1.43it/s] 
1|3537|Loss: 0.934192955493927:  55%|█████▍    | 3537/6470 [47:30<35:01,  1.40it/s]
1|3538|Loss: 0.9293944835662842:  55%|█████▍    | 3537/6470 [47:31<35:01,  1.40it/s]
1|3538|Loss: 0.9293944835662842:  55%|█████▍    | 3538/6470 [47:31<34:33,  1.41it/s]
1|3539|Loss: 0.805554986000061:  55%|█████▍    | 3538/6470 [47:31<34:33,  1.41it/s] 
1|3539|Loss: 0.805554986000061:  55%|█████▍    | 3539/6470 [47:32<35:42,  1.37it/s]
1|3540|Loss: 1.1636728048324585:  55%|█████▍    | 3539/6470 [47:32<35:42,  1.37it/s]
1|3540|Loss: 1.1636728048324585:  55%|█████▍    | 3540/6470 [47:32<34:21,  1.42it/s]
1|3541|Loss: 0.9333444237709045:  55%|█████▍    | 3540/6470 [47:33<34:21,  1.42it/s]
1|3541|Loss: 0.9333444237709045:  55%|█████▍    | 3541/6470 [47:33<35:02,  1.39it/s]
1|3542|Loss: 0.8462164998054504:  55%|█████▍    | 3541/6470 [47:34<35:02,  1.39it/s]
1|3542|Loss: 0.8462164998054504:  55%|█████▍    | 3542/6470 [47:34<33:11,  1.47it/s]
1|3543|Loss: 0.8853371143341064:  55%|█████▍    | 3542/6470 [47:34<33:11,  1.47it/s]
1|3543|Loss: 0.8853371143341064:  55%|█████▍    | 3543/6470 [47:35<34:02,  1.43it/s]
1|3544|Loss: 1.057289719581604:  55%|█████▍    | 3543/6470 [47:35<34:02,  1.43it/s] 
1|3544|Loss: 1.057289719581604:  55%|█████▍    | 3544/6470 [47:35<36:36,  1.33it/s]
1|3545|Loss: 1.4252312183380127:  55%|█████▍    | 3544/6470 [47:36<36:36,  1.33it/s]
1|3545|Loss: 1.4252312183380127:  55%|█████▍    | 3545/6470 [47:36<37:03,  1.32it/s]
1|3546|Loss: 1.17068350315094:  55%|█████▍    | 3545/6470 [47:37<37:03,  1.32it/s]  
1|3546|Loss: 1.17068350315094:  55%|█████▍    | 3546/6470 [47:37<35:57,  1.36it/s]
1|3547|Loss: 1.1225844621658325:  55%|█████▍    | 3546/6470 [47:37<35:57,  1.36it/s]
1|3547|Loss: 1.1225844621658325:  55%|█████▍    | 3547/6470 [47:38<35:54,  1.36it/s]
1|3548|Loss: 0.9093270897865295:  55%|█████▍    | 3547/6470 [47:38<35:54,  1.36it/s]
1|3548|Loss: 0.9093270897865295:  55%|█████▍    | 3548/6470 [47:38<36:45,  1.33it/s]
1|3549|Loss: 1.0444204807281494:  55%|█████▍    | 3548/6470 [47:39<36:45,  1.33it/s]
1|3549|Loss: 1.0444204807281494:  55%|█████▍    | 3549/6470 [47:39<38:11,  1.27it/s]
1|3550|Loss: 1.005252718925476:  55%|█████▍    | 3549/6470 [47:40<38:11,  1.27it/s] 
1|3550|Loss: 1.005252718925476:  55%|█████▍    | 3550/6470 [47:40<40:05,  1.21it/s]
1|3551|Loss: 1.0731253623962402:  55%|█████▍    | 3550/6470 [47:41<40:05,  1.21it/s]
1|3551|Loss: 1.0731253623962402:  55%|█████▍    | 3551/6470 [47:41<41:58,  1.16it/s]
1|3552|Loss: 0.9197100400924683:  55%|█████▍    | 3551/6470 [47:42<41:58,  1.16it/s]
1|3552|Loss: 0.9197100400924683:  55%|█████▍    | 3552/6470 [47:42<40:32,  1.20it/s]
1|3553|Loss: 0.7441748976707458:  55%|█████▍    | 3552/6470 [47:42<40:32,  1.20it/s]
1|3553|Loss: 0.7441748976707458:  55%|█████▍    | 3553/6470 [47:43<39:13,  1.24it/s]
1|3554|Loss: 1.1287568807601929:  55%|█████▍    | 3553/6470 [47:43<39:13,  1.24it/s]
1|3554|Loss: 1.1287568807601929:  55%|█████▍    | 3554/6470 [47:43<38:27,  1.26it/s]
1|3555|Loss: 1.220703363418579:  55%|█████▍    | 3554/6470 [47:44<38:27,  1.26it/s] 
1|3555|Loss: 1.220703363418579:  55%|█████▍    | 3555/6470 [47:44<39:59,  1.22it/s]
1|3556|Loss: 0.910996675491333:  55%|█████▍    | 3555/6470 [47:45<39:59,  1.22it/s]
1|3556|Loss: 0.910996675491333:  55%|█████▍    | 3556/6470 [47:45<41:21,  1.17it/s]
1|3557|Loss: 0.9917745590209961:  55%|█████▍    | 3556/6470 [47:46<41:21,  1.17it/s]
1|3557|Loss: 0.9917745590209961:  55%|█████▍    | 3557/6470 [47:46<39:18,  1.23it/s]
1|3558|Loss: 0.9708627462387085:  55%|█████▍    | 3557/6470 [47:46<39:18,  1.23it/s]
1|3558|Loss: 0.9708627462387085:  55%|█████▍    | 3558/6470 [47:47<38:37,  1.26it/s]
1|3559|Loss: 0.8092330098152161:  55%|█████▍    | 3558/6470 [47:47<38:37,  1.26it/s]
1|3559|Loss: 0.8092330098152161:  55%|█████▌    | 3559/6470 [47:47<35:39,  1.36it/s]
1|3560|Loss: 0.8955898880958557:  55%|█████▌    | 3559/6470 [47:48<35:39,  1.36it/s]
1|3560|Loss: 0.8955898880958557:  55%|█████▌    | 3560/6470 [47:48<34:29,  1.41it/s]
1|3561|Loss: 0.8051061630249023:  55%|█████▌    | 3560/6470 [47:48<34:29,  1.41it/s]
1|3561|Loss: 0.8051061630249023:  55%|█████▌    | 3561/6470 [47:49<33:17,  1.46it/s]
1|3562|Loss: 1.0605899095535278:  55%|█████▌    | 3561/6470 [47:49<33:17,  1.46it/s]
1|3562|Loss: 1.0605899095535278:  55%|█████▌    | 3562/6470 [47:49<32:40,  1.48it/s]
1|3563|Loss: 0.9537013173103333:  55%|█████▌    | 3562/6470 [47:50<32:40,  1.48it/s]
1|3563|Loss: 0.9537013173103333:  55%|█████▌    | 3563/6470 [47:50<33:01,  1.47it/s]
1|3564|Loss: 1.2788408994674683:  55%|█████▌    | 3563/6470 [47:50<33:01,  1.47it/s]
1|3564|Loss: 1.2788408994674683:  55%|█████▌    | 3564/6470 [47:51<33:16,  1.46it/s]
1|3565|Loss: 0.9017709493637085:  55%|█████▌    | 3564/6470 [47:51<33:16,  1.46it/s]
1|3565|Loss: 0.9017709493637085:  55%|█████▌    | 3565/6470 [47:51<34:06,  1.42it/s]
1|3566|Loss: 0.9919920563697815:  55%|█████▌    | 3565/6470 [47:52<34:06,  1.42it/s]
1|3566|Loss: 0.9919920563697815:  55%|█████▌    | 3566/6470 [47:52<30:23,  1.59it/s]
1|3567|Loss: 0.9333008527755737:  55%|█████▌    | 3566/6470 [47:52<30:23,  1.59it/s]
1|3567|Loss: 0.9333008527755737:  55%|█████▌    | 3567/6470 [47:52<30:37,  1.58it/s]
1|3568|Loss: 0.9553022980690002:  55%|█████▌    | 3567/6470 [47:53<30:37,  1.58it/s]
1|3568|Loss: 0.9553022980690002:  55%|█████▌    | 3568/6470 [47:53<32:42,  1.48it/s]
1|3569|Loss: 1.0260357856750488:  55%|█████▌    | 3568/6470 [47:54<32:42,  1.48it/s]
1|3569|Loss: 1.0260357856750488:  55%|█████▌    | 3569/6470 [47:54<30:49,  1.57it/s]
1|3570|Loss: 1.0033637285232544:  55%|█████▌    | 3569/6470 [47:54<30:49,  1.57it/s]
1|3570|Loss: 1.0033637285232544:  55%|█████▌    | 3570/6470 [47:55<32:16,  1.50it/s]
1|3571|Loss: 1.012266993522644:  55%|█████▌    | 3570/6470 [47:55<32:16,  1.50it/s] 
1|3571|Loss: 1.012266993522644:  55%|█████▌    | 3571/6470 [47:55<32:51,  1.47it/s]
1|3572|Loss: 0.8610144853591919:  55%|█████▌    | 3571/6470 [47:56<32:51,  1.47it/s]
1|3572|Loss: 0.8610144853591919:  55%|█████▌    | 3572/6470 [47:56<33:58,  1.42it/s]
1|3573|Loss: 0.9634402990341187:  55%|█████▌    | 3572/6470 [47:56<33:58,  1.42it/s]
1|3573|Loss: 0.9634402990341187:  55%|█████▌    | 3573/6470 [47:57<33:55,  1.42it/s]
1|3574|Loss: 1.3811123371124268:  55%|█████▌    | 3573/6470 [47:57<33:55,  1.42it/s]
1|3574|Loss: 1.3811123371124268:  55%|█████▌    | 3574/6470 [47:57<34:52,  1.38it/s]
1|3575|Loss: 0.9027950763702393:  55%|█████▌    | 3574/6470 [47:58<34:52,  1.38it/s]
1|3575|Loss: 0.9027950763702393:  55%|█████▌    | 3575/6470 [47:58<33:52,  1.42it/s]
1|3576|Loss: 0.9930237531661987:  55%|█████▌    | 3575/6470 [47:58<33:52,  1.42it/s]
1|3576|Loss: 0.9930237531661987:  55%|█████▌    | 3576/6470 [47:59<32:15,  1.50it/s]
1|3577|Loss: 0.8983516693115234:  55%|█████▌    | 3576/6470 [47:59<32:15,  1.50it/s]
1|3577|Loss: 0.8983516693115234:  55%|█████▌    | 3577/6470 [47:59<31:45,  1.52it/s]
1|3578|Loss: 0.8950557112693787:  55%|█████▌    | 3577/6470 [48:00<31:45,  1.52it/s]
1|3578|Loss: 0.8950557112693787:  55%|█████▌    | 3578/6470 [48:00<34:51,  1.38it/s]
1|3579|Loss: 1.0487103462219238:  55%|█████▌    | 3578/6470 [48:01<34:51,  1.38it/s]
1|3579|Loss: 1.0487103462219238:  55%|█████▌    | 3579/6470 [48:01<35:09,  1.37it/s]
1|3580|Loss: 0.8870379328727722:  55%|█████▌    | 3579/6470 [48:01<35:09,  1.37it/s]
1|3580|Loss: 0.8870379328727722:  55%|█████▌    | 3580/6470 [48:02<35:24,  1.36it/s]
1|3581|Loss: 0.9012823700904846:  55%|█████▌    | 3580/6470 [48:02<35:24,  1.36it/s]
1|3581|Loss: 0.9012823700904846:  55%|█████▌    | 3581/6470 [48:02<36:00,  1.34it/s]
1|3582|Loss: 1.0692399740219116:  55%|█████▌    | 3581/6470 [48:03<36:00,  1.34it/s]
1|3582|Loss: 1.0692399740219116:  55%|█████▌    | 3582/6470 [48:03<34:48,  1.38it/s]
1|3583|Loss: 0.8254903554916382:  55%|█████▌    | 3582/6470 [48:04<34:48,  1.38it/s]
1|3583|Loss: 0.8254903554916382:  55%|█████▌    | 3583/6470 [48:04<34:56,  1.38it/s]
1|3584|Loss: 0.9794284701347351:  55%|█████▌    | 3583/6470 [48:04<34:56,  1.38it/s]
1|3584|Loss: 0.9794284701347351:  55%|█████▌    | 3584/6470 [48:05<34:40,  1.39it/s]
1|3585|Loss: 1.1672300100326538:  55%|█████▌    | 3584/6470 [48:05<34:40,  1.39it/s]
1|3585|Loss: 1.1672300100326538:  55%|█████▌    | 3585/6470 [48:05<35:30,  1.35it/s]
1|3586|Loss: 0.8513259291648865:  55%|█████▌    | 3585/6470 [48:06<35:30,  1.35it/s]
1|3586|Loss: 0.8513259291648865:  55%|█████▌    | 3586/6470 [48:06<36:04,  1.33it/s]
1|3587|Loss: 0.9581397175788879:  55%|█████▌    | 3586/6470 [48:07<36:04,  1.33it/s]
1|3587|Loss: 0.9581397175788879:  55%|█████▌    | 3587/6470 [48:07<37:21,  1.29it/s]
1|3588|Loss: 0.9210395812988281:  55%|█████▌    | 3587/6470 [48:07<37:21,  1.29it/s]
1|3588|Loss: 0.9210395812988281:  55%|█████▌    | 3588/6470 [48:08<37:56,  1.27it/s]
1|3589|Loss: 1.056595802307129:  55%|█████▌    | 3588/6470 [48:08<37:56,  1.27it/s] 
1|3589|Loss: 1.056595802307129:  55%|█████▌    | 3589/6470 [48:08<36:05,  1.33it/s]
1|3590|Loss: 0.945819079875946:  55%|█████▌    | 3589/6470 [48:09<36:05,  1.33it/s]
1|3590|Loss: 0.945819079875946:  55%|█████▌    | 3590/6470 [48:09<33:12,  1.45it/s]
1|3591|Loss: 0.9108200669288635:  55%|█████▌    | 3590/6470 [48:09<33:12,  1.45it/s]
1|3591|Loss: 0.9108200669288635:  56%|█████▌    | 3591/6470 [48:10<32:41,  1.47it/s]
1|3592|Loss: 1.2645297050476074:  56%|█████▌    | 3591/6470 [48:10<32:41,  1.47it/s]
1|3592|Loss: 1.2645297050476074:  56%|█████▌    | 3592/6470 [48:10<34:07,  1.41it/s]
1|3593|Loss: 0.8902086019515991:  56%|█████▌    | 3592/6470 [48:11<34:07,  1.41it/s]
1|3593|Loss: 0.8902086019515991:  56%|█████▌    | 3593/6470 [48:11<33:01,  1.45it/s]
1|3594|Loss: 0.9449619650840759:  56%|█████▌    | 3593/6470 [48:12<33:01,  1.45it/s]
1|3594|Loss: 0.9449619650840759:  56%|█████▌    | 3594/6470 [48:12<35:04,  1.37it/s]
1|3595|Loss: 1.1101157665252686:  56%|█████▌    | 3594/6470 [48:12<35:04,  1.37it/s]
1|3595|Loss: 1.1101157665252686:  56%|█████▌    | 3595/6470 [48:13<36:22,  1.32it/s]
1|3596|Loss: 1.0538419485092163:  56%|█████▌    | 3595/6470 [48:13<36:22,  1.32it/s]
1|3596|Loss: 1.0538419485092163:  56%|█████▌    | 3596/6470 [48:13<35:39,  1.34it/s]
1|3597|Loss: 0.9579835534095764:  56%|█████▌    | 3596/6470 [48:14<35:39,  1.34it/s]
1|3597|Loss: 0.9579835534095764:  56%|█████▌    | 3597/6470 [48:14<35:58,  1.33it/s]
1|3598|Loss: 0.8359822034835815:  56%|█████▌    | 3597/6470 [48:15<35:58,  1.33it/s]
1|3598|Loss: 0.8359822034835815:  56%|█████▌    | 3598/6470 [48:15<34:59,  1.37it/s]
1|3599|Loss: 0.8365006446838379:  56%|█████▌    | 3598/6470 [48:15<34:59,  1.37it/s]
1|3599|Loss: 0.8365006446838379:  56%|█████▌    | 3599/6470 [48:16<33:54,  1.41it/s]
1|3600|Loss: 1.0571976900100708:  56%|█████▌    | 3599/6470 [48:16<33:54,  1.41it/s]
1|3600|Loss: 1.0571976900100708:  56%|█████▌    | 3600/6470 [48:16<32:00,  1.49it/s]
1|3601|Loss: 0.9020429849624634:  56%|█████▌    | 3600/6470 [48:17<32:00,  1.49it/s]
1|3601|Loss: 0.9020429849624634:  56%|█████▌    | 3601/6470 [48:17<33:09,  1.44it/s]
1|3602|Loss: 1.011060118675232:  56%|█████▌    | 3601/6470 [48:17<33:09,  1.44it/s] 
1|3602|Loss: 1.011060118675232:  56%|█████▌    | 3602/6470 [48:18<33:32,  1.43it/s]
1|3603|Loss: 1.348130226135254:  56%|█████▌    | 3602/6470 [48:18<33:32,  1.43it/s]
1|3603|Loss: 1.348130226135254:  56%|█████▌    | 3603/6470 [48:18<35:48,  1.33it/s]
1|3604|Loss: 1.1146862506866455:  56%|█████▌    | 3603/6470 [48:19<35:48,  1.33it/s]
1|3604|Loss: 1.1146862506866455:  56%|█████▌    | 3604/6470 [48:19<35:10,  1.36it/s]
1|3605|Loss: 0.9082343578338623:  56%|█████▌    | 3604/6470 [48:20<35:10,  1.36it/s]
1|3605|Loss: 0.9082343578338623:  56%|█████▌    | 3605/6470 [48:20<34:12,  1.40it/s]
1|3606|Loss: 1.0553358793258667:  56%|█████▌    | 3605/6470 [48:20<34:12,  1.40it/s]
1|3606|Loss: 1.0553358793258667:  56%|█████▌    | 3606/6470 [48:21<36:01,  1.33it/s]
1|3607|Loss: 0.857715904712677:  56%|█████▌    | 3606/6470 [48:21<36:01,  1.33it/s] 
1|3607|Loss: 0.857715904712677:  56%|█████▌    | 3607/6470 [48:21<36:16,  1.32it/s]
1|3608|Loss: 0.9290982484817505:  56%|█████▌    | 3607/6470 [48:22<36:16,  1.32it/s]
1|3608|Loss: 0.9290982484817505:  56%|█████▌    | 3608/6470 [48:22<36:20,  1.31it/s]
1|3609|Loss: 0.8449176549911499:  56%|█████▌    | 3608/6470 [48:23<36:20,  1.31it/s]
1|3609|Loss: 0.8449176549911499:  56%|█████▌    | 3609/6470 [48:23<36:35,  1.30it/s]
1|3610|Loss: 0.927512526512146:  56%|█████▌    | 3609/6470 [48:23<36:35,  1.30it/s] 
1|3610|Loss: 0.927512526512146:  56%|█████▌    | 3610/6470 [48:24<36:52,  1.29it/s]
1|3611|Loss: 0.9049913287162781:  56%|█████▌    | 3610/6470 [48:24<36:52,  1.29it/s]
1|3611|Loss: 0.9049913287162781:  56%|█████▌    | 3611/6470 [48:25<36:57,  1.29it/s]
1|3612|Loss: 0.8576042056083679:  56%|█████▌    | 3611/6470 [48:25<36:57,  1.29it/s]
1|3612|Loss: 0.8576042056083679:  56%|█████▌    | 3612/6470 [48:25<34:50,  1.37it/s]
1|3613|Loss: 0.943196177482605:  56%|█████▌    | 3612/6470 [48:26<34:50,  1.37it/s] 
1|3613|Loss: 0.943196177482605:  56%|█████▌    | 3613/6470 [48:26<36:34,  1.30it/s]
1|3614|Loss: 0.9367166757583618:  56%|█████▌    | 3613/6470 [48:27<36:34,  1.30it/s]
1|3614|Loss: 0.9367166757583618:  56%|█████▌    | 3614/6470 [48:27<36:11,  1.32it/s]
1|3615|Loss: 0.975540816783905:  56%|█████▌    | 3614/6470 [48:27<36:11,  1.32it/s] 
1|3615|Loss: 0.975540816783905:  56%|█████▌    | 3615/6470 [48:27<33:48,  1.41it/s]
1|3616|Loss: 0.9419305920600891:  56%|█████▌    | 3615/6470 [48:28<33:48,  1.41it/s]
1|3616|Loss: 0.9419305920600891:  56%|█████▌    | 3616/6470 [48:28<34:32,  1.38it/s]
1|3617|Loss: 1.1271042823791504:  56%|█████▌    | 3616/6470 [48:29<34:32,  1.38it/s]
1|3617|Loss: 1.1271042823791504:  56%|█████▌    | 3617/6470 [48:29<39:02,  1.22it/s]
1|3618|Loss: 0.9639937877655029:  56%|█████▌    | 3617/6470 [48:30<39:02,  1.22it/s]
1|3618|Loss: 0.9639937877655029:  56%|█████▌    | 3618/6470 [48:30<39:25,  1.21it/s]
1|3619|Loss: 0.8762328624725342:  56%|█████▌    | 3618/6470 [48:30<39:25,  1.21it/s]
1|3619|Loss: 0.8762328624725342:  56%|█████▌    | 3619/6470 [48:31<38:51,  1.22it/s]
1|3620|Loss: 0.9352505803108215:  56%|█████▌    | 3619/6470 [48:31<38:51,  1.22it/s]
1|3620|Loss: 0.9352505803108215:  56%|█████▌    | 3620/6470 [48:32<38:33,  1.23it/s]
1|3621|Loss: 1.05543053150177:  56%|█████▌    | 3620/6470 [48:32<38:33,  1.23it/s]  
1|3621|Loss: 1.05543053150177:  56%|█████▌    | 3621/6470 [48:33<39:56,  1.19it/s]
1|3622|Loss: 0.9341707229614258:  56%|█████▌    | 3621/6470 [48:33<39:56,  1.19it/s]
1|3622|Loss: 0.9341707229614258:  56%|█████▌    | 3622/6470 [48:33<39:52,  1.19it/s]
1|3623|Loss: 0.953100860118866:  56%|█████▌    | 3622/6470 [48:34<39:52,  1.19it/s] 
1|3623|Loss: 0.953100860118866:  56%|█████▌    | 3623/6470 [48:34<41:05,  1.15it/s]
1|3624|Loss: 0.8993235230445862:  56%|█████▌    | 3623/6470 [48:35<41:05,  1.15it/s]
1|3624|Loss: 0.8993235230445862:  56%|█████▌    | 3624/6470 [48:35<40:38,  1.17it/s]
1|3625|Loss: 1.1040643453598022:  56%|█████▌    | 3624/6470 [48:36<40:38,  1.17it/s]
1|3625|Loss: 1.1040643453598022:  56%|█████▌    | 3625/6470 [48:36<39:40,  1.20it/s]
1|3626|Loss: 0.8635125756263733:  56%|█████▌    | 3625/6470 [48:36<39:40,  1.20it/s]
1|3626|Loss: 0.8635125756263733:  56%|█████▌    | 3626/6470 [48:37<37:57,  1.25it/s]
1|3627|Loss: 0.9257370829582214:  56%|█████▌    | 3626/6470 [48:37<37:57,  1.25it/s]
1|3627|Loss: 0.9257370829582214:  56%|█████▌    | 3627/6470 [48:37<36:27,  1.30it/s]
1|3628|Loss: 0.9384757280349731:  56%|█████▌    | 3627/6470 [48:38<36:27,  1.30it/s]
1|3628|Loss: 0.9384757280349731:  56%|█████▌    | 3628/6470 [48:38<35:17,  1.34it/s]
1|3629|Loss: 0.8640023469924927:  56%|█████▌    | 3628/6470 [48:38<35:17,  1.34it/s]
1|3629|Loss: 0.8640023469924927:  56%|█████▌    | 3629/6470 [48:39<35:37,  1.33it/s]
1|3630|Loss: 1.0839940309524536:  56%|█████▌    | 3629/6470 [48:39<35:37,  1.33it/s]
1|3630|Loss: 1.0839940309524536:  56%|█████▌    | 3630/6470 [48:39<34:27,  1.37it/s]
1|3631|Loss: 0.9349679946899414:  56%|█████▌    | 3630/6470 [48:40<34:27,  1.37it/s]
1|3631|Loss: 0.9349679946899414:  56%|█████▌    | 3631/6470 [48:40<33:28,  1.41it/s]
1|3632|Loss: 0.9335314035415649:  56%|█████▌    | 3631/6470 [48:41<33:28,  1.41it/s]
1|3632|Loss: 0.9335314035415649:  56%|█████▌    | 3632/6470 [48:41<34:03,  1.39it/s]
1|3633|Loss: 0.9626356363296509:  56%|█████▌    | 3632/6470 [48:41<34:03,  1.39it/s]
1|3633|Loss: 0.9626356363296509:  56%|█████▌    | 3633/6470 [48:42<33:55,  1.39it/s]
1|3634|Loss: 1.003614068031311:  56%|█████▌    | 3633/6470 [48:42<33:55,  1.39it/s] 
1|3634|Loss: 1.003614068031311:  56%|█████▌    | 3634/6470 [48:42<31:08,  1.52it/s]
1|3635|Loss: 0.9480065703392029:  56%|█████▌    | 3634/6470 [48:42<31:08,  1.52it/s]
1|3635|Loss: 0.9480065703392029:  56%|█████▌    | 3635/6470 [48:43<31:56,  1.48it/s]
1|3636|Loss: 1.084488034248352:  56%|█████▌    | 3635/6470 [48:43<31:56,  1.48it/s] 
1|3636|Loss: 1.084488034248352:  56%|█████▌    | 3636/6470 [48:44<32:14,  1.46it/s]
1|3637|Loss: 0.8552922010421753:  56%|█████▌    | 3636/6470 [48:44<32:14,  1.46it/s]
1|3637|Loss: 0.8552922010421753:  56%|█████▌    | 3637/6470 [48:44<32:00,  1.48it/s]
1|3638|Loss: 0.8914671540260315:  56%|█████▌    | 3637/6470 [48:45<32:00,  1.48it/s]
1|3638|Loss: 0.8914671540260315:  56%|█████▌    | 3638/6470 [48:45<31:54,  1.48it/s]
1|3639|Loss: 0.9385625123977661:  56%|█████▌    | 3638/6470 [48:45<31:54,  1.48it/s]
1|3639|Loss: 0.9385625123977661:  56%|█████▌    | 3639/6470 [48:46<32:58,  1.43it/s]
1|3640|Loss: 0.9394469857215881:  56%|█████▌    | 3639/6470 [48:46<32:58,  1.43it/s]
1|3640|Loss: 0.9394469857215881:  56%|█████▋    | 3640/6470 [48:47<36:07,  1.31it/s]
1|3641|Loss: 0.8304529190063477:  56%|█████▋    | 3640/6470 [48:47<36:07,  1.31it/s]
1|3641|Loss: 0.8304529190063477:  56%|█████▋    | 3641/6470 [48:48<38:39,  1.22it/s]
1|3642|Loss: 0.9408143758773804:  56%|█████▋    | 3641/6470 [48:48<38:39,  1.22it/s]
1|3642|Loss: 0.9408143758773804:  56%|█████▋    | 3642/6470 [48:48<39:58,  1.18it/s]
1|3643|Loss: 0.9955871105194092:  56%|█████▋    | 3642/6470 [48:49<39:58,  1.18it/s]
1|3643|Loss: 0.9955871105194092:  56%|█████▋    | 3643/6470 [48:49<39:30,  1.19it/s]
1|3644|Loss: 0.822425365447998:  56%|█████▋    | 3643/6470 [48:50<39:30,  1.19it/s] 
1|3644|Loss: 0.822425365447998:  56%|█████▋    | 3644/6470 [48:50<37:38,  1.25it/s]
1|3645|Loss: 0.8128989934921265:  56%|█████▋    | 3644/6470 [48:50<37:38,  1.25it/s]
1|3645|Loss: 0.8128989934921265:  56%|█████▋    | 3645/6470 [48:51<35:32,  1.32it/s]
1|3646|Loss: 0.9323334097862244:  56%|█████▋    | 3645/6470 [48:51<35:32,  1.32it/s]
1|3646|Loss: 0.9323334097862244:  56%|█████▋    | 3646/6470 [48:51<37:31,  1.25it/s]
1|3647|Loss: 0.904218852519989:  56%|█████▋    | 3646/6470 [48:52<37:31,  1.25it/s] 
1|3647|Loss: 0.904218852519989:  56%|█████▋    | 3647/6470 [48:52<37:53,  1.24it/s]
1|3648|Loss: 0.9083603024482727:  56%|█████▋    | 3647/6470 [48:53<37:53,  1.24it/s]
1|3648|Loss: 0.9083603024482727:  56%|█████▋    | 3648/6470 [48:53<36:38,  1.28it/s]
1|3649|Loss: 1.0416680574417114:  56%|█████▋    | 3648/6470 [48:53<36:38,  1.28it/s]
1|3649|Loss: 1.0416680574417114:  56%|█████▋    | 3649/6470 [48:54<37:46,  1.24it/s]
1|3650|Loss: 1.0328681468963623:  56%|█████▋    | 3649/6470 [48:54<37:46,  1.24it/s]
1|3650|Loss: 1.0328681468963623:  56%|█████▋    | 3650/6470 [48:55<35:06,  1.34it/s]
1|3651|Loss: 0.9096059203147888:  56%|█████▋    | 3650/6470 [48:55<35:06,  1.34it/s]
1|3651|Loss: 0.9096059203147888:  56%|█████▋    | 3651/6470 [48:55<33:34,  1.40it/s]
1|3652|Loss: 1.0563478469848633:  56%|█████▋    | 3651/6470 [48:56<33:34,  1.40it/s]
1|3652|Loss: 1.0563478469848633:  56%|█████▋    | 3652/6470 [48:56<35:20,  1.33it/s]
1|3653|Loss: 1.0208265781402588:  56%|█████▋    | 3652/6470 [48:56<35:20,  1.33it/s]
1|3653|Loss: 1.0208265781402588:  56%|█████▋    | 3653/6470 [48:57<34:08,  1.37it/s]
1|3654|Loss: 1.0914831161499023:  56%|█████▋    | 3653/6470 [48:57<34:08,  1.37it/s]
1|3654|Loss: 1.0914831161499023:  56%|█████▋    | 3654/6470 [48:57<33:32,  1.40it/s]
1|3655|Loss: 0.8856386542320251:  56%|█████▋    | 3654/6470 [48:58<33:32,  1.40it/s]
1|3655|Loss: 0.8856386542320251:  56%|█████▋    | 3655/6470 [48:58<36:11,  1.30it/s]
1|3656|Loss: 0.7269193530082703:  56%|█████▋    | 3655/6470 [48:59<36:11,  1.30it/s]
1|3656|Loss: 0.7269193530082703:  57%|█████▋    | 3656/6470 [48:59<32:28,  1.44it/s]
1|3657|Loss: 0.7980102896690369:  57%|█████▋    | 3656/6470 [48:59<32:28,  1.44it/s]
1|3657|Loss: 0.7980102896690369:  57%|█████▋    | 3657/6470 [48:59<30:29,  1.54it/s]
1|3658|Loss: 0.9691792726516724:  57%|█████▋    | 3657/6470 [49:00<30:29,  1.54it/s]
1|3658|Loss: 0.9691792726516724:  57%|█████▋    | 3658/6470 [49:00<30:50,  1.52it/s]
1|3659|Loss: 0.8802816271781921:  57%|█████▋    | 3658/6470 [49:00<30:50,  1.52it/s]
1|3659|Loss: 0.8802816271781921:  57%|█████▋    | 3659/6470 [49:01<31:01,  1.51it/s]
1|3660|Loss: 0.8515019416809082:  57%|█████▋    | 3659/6470 [49:01<31:01,  1.51it/s]
1|3660|Loss: 0.8515019416809082:  57%|█████▋    | 3660/6470 [49:01<30:11,  1.55it/s]
1|3661|Loss: 0.8902133107185364:  57%|█████▋    | 3660/6470 [49:02<30:11,  1.55it/s]
1|3661|Loss: 0.8902133107185364:  57%|█████▋    | 3661/6470 [49:02<29:48,  1.57it/s]
1|3662|Loss: 0.8877675533294678:  57%|█████▋    | 3661/6470 [49:02<29:48,  1.57it/s]
1|3662|Loss: 0.8877675533294678:  57%|█████▋    | 3662/6470 [49:03<33:12,  1.41it/s]
1|3663|Loss: 0.7962081432342529:  57%|█████▋    | 3662/6470 [49:03<33:12,  1.41it/s]
1|3663|Loss: 0.7962081432342529:  57%|█████▋    | 3663/6470 [49:04<36:47,  1.27it/s]
1|3664|Loss: 0.9552224278450012:  57%|█████▋    | 3663/6470 [49:04<36:47,  1.27it/s]
1|3664|Loss: 0.9552224278450012:  57%|█████▋    | 3664/6470 [49:05<37:13,  1.26it/s]
1|3665|Loss: 0.9871300458908081:  57%|█████▋    | 3664/6470 [49:05<37:13,  1.26it/s]
1|3665|Loss: 0.9871300458908081:  57%|█████▋    | 3665/6470 [49:05<38:37,  1.21it/s]
1|3666|Loss: 0.960950493812561:  57%|█████▋    | 3665/6470 [49:06<38:37,  1.21it/s] 
1|3666|Loss: 0.960950493812561:  57%|█████▋    | 3666/6470 [49:06<37:03,  1.26it/s]
1|3667|Loss: 0.8893330097198486:  57%|█████▋    | 3666/6470 [49:07<37:03,  1.26it/s]
1|3667|Loss: 0.8893330097198486:  57%|█████▋    | 3667/6470 [49:07<36:19,  1.29it/s]
1|3668|Loss: 0.9793238639831543:  57%|█████▋    | 3667/6470 [49:07<36:19,  1.29it/s]
1|3668|Loss: 0.9793238639831543:  57%|█████▋    | 3668/6470 [49:08<37:10,  1.26it/s]
1|3669|Loss: 0.9953407645225525:  57%|█████▋    | 3668/6470 [49:08<37:10,  1.26it/s]
1|3669|Loss: 0.9953407645225525:  57%|█████▋    | 3669/6470 [49:08<35:26,  1.32it/s]
1|3670|Loss: 1.0909032821655273:  57%|█████▋    | 3669/6470 [49:09<35:26,  1.32it/s]
1|3670|Loss: 1.0909032821655273:  57%|█████▋    | 3670/6470 [49:09<34:04,  1.37it/s]
1|3671|Loss: 1.0224310159683228:  57%|█████▋    | 3670/6470 [49:09<34:04,  1.37it/s]
1|3671|Loss: 1.0224310159683228:  57%|█████▋    | 3671/6470 [49:10<33:32,  1.39it/s]
1|3672|Loss: 1.0664016008377075:  57%|█████▋    | 3671/6470 [49:10<33:32,  1.39it/s]
1|3672|Loss: 1.0664016008377075:  57%|█████▋    | 3672/6470 [49:11<34:05,  1.37it/s]
1|3673|Loss: 0.901595950126648:  57%|█████▋    | 3672/6470 [49:11<34:05,  1.37it/s] 
1|3673|Loss: 0.901595950126648:  57%|█████▋    | 3673/6470 [49:11<36:40,  1.27it/s]
1|3674|Loss: 0.8174978494644165:  57%|█████▋    | 3673/6470 [49:12<36:40,  1.27it/s]
1|3674|Loss: 0.8174978494644165:  57%|█████▋    | 3674/6470 [49:12<36:18,  1.28it/s]
1|3675|Loss: 0.8909446597099304:  57%|█████▋    | 3674/6470 [49:13<36:18,  1.28it/s]
1|3675|Loss: 0.8909446597099304:  57%|█████▋    | 3675/6470 [49:13<36:30,  1.28it/s]
1|3676|Loss: 0.8269182443618774:  57%|█████▋    | 3675/6470 [49:13<36:30,  1.28it/s]
1|3676|Loss: 0.8269182443618774:  57%|█████▋    | 3676/6470 [49:14<37:57,  1.23it/s]
1|3677|Loss: 0.9050975441932678:  57%|█████▋    | 3676/6470 [49:14<37:57,  1.23it/s]
1|3677|Loss: 0.9050975441932678:  57%|█████▋    | 3677/6470 [49:15<36:28,  1.28it/s]
1|3678|Loss: 0.8377975225448608:  57%|█████▋    | 3677/6470 [49:15<36:28,  1.28it/s]
1|3678|Loss: 0.8377975225448608:  57%|█████▋    | 3678/6470 [49:15<35:49,  1.30it/s]
1|3679|Loss: 1.0203584432601929:  57%|█████▋    | 3678/6470 [49:16<35:49,  1.30it/s]
1|3679|Loss: 1.0203584432601929:  57%|█████▋    | 3679/6470 [49:16<35:41,  1.30it/s]
1|3680|Loss: 0.7847543954849243:  57%|█████▋    | 3679/6470 [49:17<35:41,  1.30it/s]
1|3680|Loss: 0.7847543954849243:  57%|█████▋    | 3680/6470 [49:17<36:00,  1.29it/s]
1|3681|Loss: 0.9355826377868652:  57%|█████▋    | 3680/6470 [49:17<36:00,  1.29it/s]
1|3681|Loss: 0.9355826377868652:  57%|█████▋    | 3681/6470 [49:17<33:07,  1.40it/s]
1|3682|Loss: 0.8855370879173279:  57%|█████▋    | 3681/6470 [49:18<33:07,  1.40it/s]
1|3682|Loss: 0.8855370879173279:  57%|█████▋    | 3682/6470 [49:18<32:17,  1.44it/s]
1|3683|Loss: 0.9650264382362366:  57%|█████▋    | 3682/6470 [49:19<32:17,  1.44it/s]
1|3683|Loss: 0.9650264382362366:  57%|█████▋    | 3683/6470 [49:19<33:14,  1.40it/s]
1|3684|Loss: 0.987380862236023:  57%|█████▋    | 3683/6470 [49:19<33:14,  1.40it/s] 
1|3684|Loss: 0.987380862236023:  57%|█████▋    | 3684/6470 [49:20<34:18,  1.35it/s]
1|3685|Loss: 0.9241551756858826:  57%|█████▋    | 3684/6470 [49:20<34:18,  1.35it/s]
1|3685|Loss: 0.9241551756858826:  57%|█████▋    | 3685/6470 [49:21<35:46,  1.30it/s]
1|3686|Loss: 1.0045591592788696:  57%|█████▋    | 3685/6470 [49:21<35:46,  1.30it/s]
1|3686|Loss: 1.0045591592788696:  57%|█████▋    | 3686/6470 [49:21<35:22,  1.31it/s]
1|3687|Loss: 0.9044381976127625:  57%|█████▋    | 3686/6470 [49:22<35:22,  1.31it/s]
1|3687|Loss: 0.9044381976127625:  57%|█████▋    | 3687/6470 [49:22<35:35,  1.30it/s]
1|3688|Loss: 0.9912485480308533:  57%|█████▋    | 3687/6470 [49:23<35:35,  1.30it/s]
1|3688|Loss: 0.9912485480308533:  57%|█████▋    | 3688/6470 [49:23<37:24,  1.24it/s]
1|3689|Loss: 0.9909708499908447:  57%|█████▋    | 3688/6470 [49:23<37:24,  1.24it/s]
1|3689|Loss: 0.9909708499908447:  57%|█████▋    | 3689/6470 [49:24<34:55,  1.33it/s]
1|3690|Loss: 0.9093326926231384:  57%|█████▋    | 3689/6470 [49:24<34:55,  1.33it/s]
1|3690|Loss: 0.9093326926231384:  57%|█████▋    | 3690/6470 [49:24<32:26,  1.43it/s]
1|3691|Loss: 1.0486897230148315:  57%|█████▋    | 3690/6470 [49:25<32:26,  1.43it/s]
1|3691|Loss: 1.0486897230148315:  57%|█████▋    | 3691/6470 [49:25<33:32,  1.38it/s]
1|3692|Loss: 0.9365164041519165:  57%|█████▋    | 3691/6470 [49:25<33:32,  1.38it/s]
1|3692|Loss: 0.9365164041519165:  57%|█████▋    | 3692/6470 [49:25<31:16,  1.48it/s]
1|3693|Loss: 0.8997335433959961:  57%|█████▋    | 3692/6470 [49:26<31:16,  1.48it/s]
1|3693|Loss: 0.8997335433959961:  57%|█████▋    | 3693/6470 [49:26<32:22,  1.43it/s]
1|3694|Loss: 0.8265895843505859:  57%|█████▋    | 3693/6470 [49:27<32:22,  1.43it/s]
1|3694|Loss: 0.8265895843505859:  57%|█████▋    | 3694/6470 [49:27<31:55,  1.45it/s]
1|3695|Loss: 0.8637787699699402:  57%|█████▋    | 3694/6470 [49:27<31:55,  1.45it/s]
1|3695|Loss: 0.8637787699699402:  57%|█████▋    | 3695/6470 [49:28<31:30,  1.47it/s]
1|3696|Loss: 0.9666089415550232:  57%|█████▋    | 3695/6470 [49:28<31:30,  1.47it/s]
1|3696|Loss: 0.9666089415550232:  57%|█████▋    | 3696/6470 [49:28<34:27,  1.34it/s]
1|3697|Loss: 0.8300791382789612:  57%|█████▋    | 3696/6470 [49:29<34:27,  1.34it/s]
1|3697|Loss: 0.8300791382789612:  57%|█████▋    | 3697/6470 [49:29<31:28,  1.47it/s]
1|3698|Loss: 0.9605786204338074:  57%|█████▋    | 3697/6470 [49:29<31:28,  1.47it/s]
1|3698|Loss: 0.9605786204338074:  57%|█████▋    | 3698/6470 [49:30<33:49,  1.37it/s]
1|3699|Loss: 0.9259926080703735:  57%|█████▋    | 3698/6470 [49:30<33:49,  1.37it/s]
1|3699|Loss: 0.9259926080703735:  57%|█████▋    | 3699/6470 [49:31<37:00,  1.25it/s]
1|3700|Loss: 1.1126554012298584:  57%|█████▋    | 3699/6470 [49:31<37:00,  1.25it/s]
1|3700|Loss: 1.1126554012298584:  57%|█████▋    | 3700/6470 [49:32<37:45,  1.22it/s]
1|3701|Loss: 0.8501492142677307:  57%|█████▋    | 3700/6470 [49:32<37:45,  1.22it/s]
1|3701|Loss: 0.8501492142677307:  57%|█████▋    | 3701/6470 [49:33<39:00,  1.18it/s]
1|3702|Loss: 1.0270179510116577:  57%|█████▋    | 3701/6470 [49:33<39:00,  1.18it/s]
1|3702|Loss: 1.0270179510116577:  57%|█████▋    | 3702/6470 [49:33<38:49,  1.19it/s]
1|3703|Loss: 0.8797609806060791:  57%|█████▋    | 3702/6470 [49:34<38:49,  1.19it/s]
1|3703|Loss: 0.8797609806060791:  57%|█████▋    | 3703/6470 [49:34<36:46,  1.25it/s]
1|3704|Loss: 0.9001960158348083:  57%|█████▋    | 3703/6470 [49:34<36:46,  1.25it/s]
1|3704|Loss: 0.9001960158348083:  57%|█████▋    | 3704/6470 [49:35<36:18,  1.27it/s]
1|3705|Loss: 0.9064162373542786:  57%|█████▋    | 3704/6470 [49:35<36:18,  1.27it/s]
1|3705|Loss: 0.9064162373542786:  57%|█████▋    | 3705/6470 [49:36<35:47,  1.29it/s]
1|3706|Loss: 0.8949602842330933:  57%|█████▋    | 3705/6470 [49:36<35:47,  1.29it/s]
1|3706|Loss: 0.8949602842330933:  57%|█████▋    | 3706/6470 [49:36<35:37,  1.29it/s]
1|3707|Loss: 0.7098327279090881:  57%|█████▋    | 3706/6470 [49:37<35:37,  1.29it/s]
1|3707|Loss: 0.7098327279090881:  57%|█████▋    | 3707/6470 [49:37<35:40,  1.29it/s]
1|3708|Loss: 0.9804588556289673:  57%|█████▋    | 3707/6470 [49:38<35:40,  1.29it/s]
1|3708|Loss: 0.9804588556289673:  57%|█████▋    | 3708/6470 [49:38<32:47,  1.40it/s]
1|3709|Loss: 0.8347684741020203:  57%|█████▋    | 3708/6470 [49:38<32:47,  1.40it/s]
1|3709|Loss: 0.8347684741020203:  57%|█████▋    | 3709/6470 [49:38<32:36,  1.41it/s]
1|3710|Loss: 1.2271347045898438:  57%|█████▋    | 3709/6470 [49:39<32:36,  1.41it/s]
1|3710|Loss: 1.2271347045898438:  57%|█████▋    | 3710/6470 [49:39<35:12,  1.31it/s]
1|3711|Loss: 0.7824990153312683:  57%|█████▋    | 3710/6470 [49:40<35:12,  1.31it/s]
1|3711|Loss: 0.7824990153312683:  57%|█████▋    | 3711/6470 [49:40<32:36,  1.41it/s]
1|3712|Loss: 0.8993628025054932:  57%|█████▋    | 3711/6470 [49:40<32:36,  1.41it/s]
1|3712|Loss: 0.8993628025054932:  57%|█████▋    | 3712/6470 [49:41<31:24,  1.46it/s]
1|3713|Loss: 0.9187101125717163:  57%|█████▋    | 3712/6470 [49:41<31:24,  1.46it/s]
1|3713|Loss: 0.9187101125717163:  57%|█████▋    | 3713/6470 [49:41<32:52,  1.40it/s]
1|3714|Loss: 1.2508721351623535:  57%|█████▋    | 3713/6470 [49:42<32:52,  1.40it/s]
1|3714|Loss: 1.2508721351623535:  57%|█████▋    | 3714/6470 [49:42<34:13,  1.34it/s]
1|3715|Loss: 0.8994854688644409:  57%|█████▋    | 3714/6470 [49:43<34:13,  1.34it/s]
1|3715|Loss: 0.8994854688644409:  57%|█████▋    | 3715/6470 [49:43<33:30,  1.37it/s]
1|3716|Loss: 0.9740420579910278:  57%|█████▋    | 3715/6470 [49:43<33:30,  1.37it/s]
1|3716|Loss: 0.9740420579910278:  57%|█████▋    | 3716/6470 [49:44<34:19,  1.34it/s]
1|3717|Loss: 0.9223006963729858:  57%|█████▋    | 3716/6470 [49:44<34:19,  1.34it/s]
1|3717|Loss: 0.9223006963729858:  57%|█████▋    | 3717/6470 [49:44<34:43,  1.32it/s]
1|3718|Loss: 1.0393613576889038:  57%|█████▋    | 3717/6470 [49:45<34:43,  1.32it/s]
1|3718|Loss: 1.0393613576889038:  57%|█████▋    | 3718/6470 [49:45<36:59,  1.24it/s]
1|3719|Loss: 0.8436866402626038:  57%|█████▋    | 3718/6470 [49:46<36:59,  1.24it/s]
1|3719|Loss: 0.8436866402626038:  57%|█████▋    | 3719/6470 [49:46<38:11,  1.20it/s]
1|3720|Loss: 1.066845178604126:  57%|█████▋    | 3719/6470 [49:47<38:11,  1.20it/s] 
1|3720|Loss: 1.066845178604126:  57%|█████▋    | 3720/6470 [49:47<36:58,  1.24it/s]
1|3721|Loss: 0.8269190192222595:  57%|█████▋    | 3720/6470 [49:47<36:58,  1.24it/s]
1|3721|Loss: 0.8269190192222595:  58%|█████▊    | 3721/6470 [49:48<38:05,  1.20it/s]
1|3722|Loss: 0.9084252715110779:  58%|█████▊    | 3721/6470 [49:48<38:05,  1.20it/s]
1|3722|Loss: 0.9084252715110779:  58%|█████▊    | 3722/6470 [49:49<39:26,  1.16it/s]
1|3723|Loss: 0.9800572991371155:  58%|█████▊    | 3722/6470 [49:49<39:26,  1.16it/s]
1|3723|Loss: 0.9800572991371155:  58%|█████▊    | 3723/6470 [49:50<38:41,  1.18it/s]
1|3724|Loss: 1.0091508626937866:  58%|█████▊    | 3723/6470 [49:50<38:41,  1.18it/s]
1|3724|Loss: 1.0091508626937866:  58%|█████▊    | 3724/6470 [49:50<37:31,  1.22it/s]
1|3725|Loss: 0.9124929308891296:  58%|█████▊    | 3724/6470 [49:51<37:31,  1.22it/s]
1|3725|Loss: 0.9124929308891296:  58%|█████▊    | 3725/6470 [49:51<36:38,  1.25it/s]
1|3726|Loss: 0.8208068609237671:  58%|█████▊    | 3725/6470 [49:51<36:38,  1.25it/s]
1|3726|Loss: 0.8208068609237671:  58%|█████▊    | 3726/6470 [49:52<34:52,  1.31it/s]
1|3727|Loss: 0.9233430624008179:  58%|█████▊    | 3726/6470 [49:52<34:52,  1.31it/s]
1|3727|Loss: 0.9233430624008179:  58%|█████▊    | 3727/6470 [49:53<34:50,  1.31it/s]
1|3728|Loss: 1.0550740957260132:  58%|█████▊    | 3727/6470 [49:53<34:50,  1.31it/s]
1|3728|Loss: 1.0550740957260132:  58%|█████▊    | 3728/6470 [49:53<35:20,  1.29it/s]
1|3729|Loss: 0.882418692111969:  58%|█████▊    | 3728/6470 [49:54<35:20,  1.29it/s] 
1|3729|Loss: 0.882418692111969:  58%|█████▊    | 3729/6470 [49:54<34:30,  1.32it/s]
1|3730|Loss: 0.8911910653114319:  58%|█████▊    | 3729/6470 [49:54<34:30,  1.32it/s]
1|3730|Loss: 0.8911910653114319:  58%|█████▊    | 3730/6470 [49:55<35:53,  1.27it/s]
1|3731|Loss: 1.118617057800293:  58%|█████▊    | 3730/6470 [49:55<35:53,  1.27it/s] 
1|3731|Loss: 1.118617057800293:  58%|█████▊    | 3731/6470 [49:56<34:55,  1.31it/s]
1|3732|Loss: 0.9511629343032837:  58%|█████▊    | 3731/6470 [49:56<34:55,  1.31it/s]
1|3732|Loss: 0.9511629343032837:  58%|█████▊    | 3732/6470 [49:56<34:05,  1.34it/s]
1|3733|Loss: 0.8442275524139404:  58%|█████▊    | 3732/6470 [49:57<34:05,  1.34it/s]
1|3733|Loss: 0.8442275524139404:  58%|█████▊    | 3733/6470 [49:57<33:22,  1.37it/s]
1|3734|Loss: 1.0393579006195068:  58%|█████▊    | 3733/6470 [49:57<33:22,  1.37it/s]
1|3734|Loss: 1.0393579006195068:  58%|█████▊    | 3734/6470 [49:58<32:00,  1.42it/s]
1|3735|Loss: 1.0656377077102661:  58%|█████▊    | 3734/6470 [49:58<32:00,  1.42it/s]
1|3735|Loss: 1.0656377077102661:  58%|█████▊    | 3735/6470 [49:58<32:30,  1.40it/s]
1|3736|Loss: 1.2876570224761963:  58%|█████▊    | 3735/6470 [49:59<32:30,  1.40it/s]
1|3736|Loss: 1.2876570224761963:  58%|█████▊    | 3736/6470 [49:59<34:14,  1.33it/s]
1|3737|Loss: 0.8828533887863159:  58%|█████▊    | 3736/6470 [50:00<34:14,  1.33it/s]
1|3737|Loss: 0.8828533887863159:  58%|█████▊    | 3737/6470 [50:00<34:59,  1.30it/s]
1|3738|Loss: 0.8853968381881714:  58%|█████▊    | 3737/6470 [50:00<34:59,  1.30it/s]
1|3738|Loss: 0.8853968381881714:  58%|█████▊    | 3738/6470 [50:01<35:34,  1.28it/s]
1|3739|Loss: 0.8185518980026245:  58%|█████▊    | 3738/6470 [50:01<35:34,  1.28it/s]
1|3739|Loss: 0.8185518980026245:  58%|█████▊    | 3739/6470 [50:02<34:44,  1.31it/s]
1|3740|Loss: 0.9945397973060608:  58%|█████▊    | 3739/6470 [50:02<34:44,  1.31it/s]
1|3740|Loss: 0.9945397973060608:  58%|█████▊    | 3740/6470 [50:02<36:43,  1.24it/s]
1|3741|Loss: 0.9333190321922302:  58%|█████▊    | 3740/6470 [50:03<36:43,  1.24it/s]
1|3741|Loss: 0.9333190321922302:  58%|█████▊    | 3741/6470 [50:03<37:52,  1.20it/s]
1|3742|Loss: 0.9359990954399109:  58%|█████▊    | 3741/6470 [50:04<37:52,  1.20it/s]
1|3742|Loss: 0.9359990954399109:  58%|█████▊    | 3742/6470 [50:04<36:42,  1.24it/s]
1|3743|Loss: 0.8839629292488098:  58%|█████▊    | 3742/6470 [50:05<36:42,  1.24it/s]
1|3743|Loss: 0.8839629292488098:  58%|█████▊    | 3743/6470 [50:05<36:21,  1.25it/s]
1|3744|Loss: 0.9168350696563721:  58%|█████▊    | 3743/6470 [50:05<36:21,  1.25it/s]
1|3744|Loss: 0.9168350696563721:  58%|█████▊    | 3744/6470 [50:06<35:15,  1.29it/s]
1|3745|Loss: 0.9174036979675293:  58%|█████▊    | 3744/6470 [50:06<35:15,  1.29it/s]
1|3745|Loss: 0.9174036979675293:  58%|█████▊    | 3745/6470 [50:06<32:59,  1.38it/s]
1|3746|Loss: 0.9581167697906494:  58%|█████▊    | 3745/6470 [50:07<32:59,  1.38it/s]
1|3746|Loss: 0.9581167697906494:  58%|█████▊    | 3746/6470 [50:07<30:33,  1.49it/s]
1|3747|Loss: 0.8018144369125366:  58%|█████▊    | 3746/6470 [50:07<30:33,  1.49it/s]
1|3747|Loss: 0.8018144369125366:  58%|█████▊    | 3747/6470 [50:07<30:04,  1.51it/s]
1|3748|Loss: 0.8450505137443542:  58%|█████▊    | 3747/6470 [50:08<30:04,  1.51it/s]
1|3748|Loss: 0.8450505137443542:  58%|█████▊    | 3748/6470 [50:08<31:13,  1.45it/s]
1|3749|Loss: 0.9373637437820435:  58%|█████▊    | 3748/6470 [50:09<31:13,  1.45it/s]
1|3749|Loss: 0.9373637437820435:  58%|█████▊    | 3749/6470 [50:09<32:47,  1.38it/s]
1|3750|Loss: 1.1005767583847046:  58%|█████▊    | 3749/6470 [50:09<32:47,  1.38it/s]
1|3750|Loss: 1.1005767583847046:  58%|█████▊    | 3750/6470 [50:10<34:29,  1.31it/s]
1|3751|Loss: 1.0132744312286377:  58%|█████▊    | 3750/6470 [50:10<34:29,  1.31it/s]
1|3751|Loss: 1.0132744312286377:  58%|█████▊    | 3751/6470 [50:10<32:38,  1.39it/s]
1|3752|Loss: 0.976144552230835:  58%|█████▊    | 3751/6470 [50:11<32:38,  1.39it/s] 
1|3752|Loss: 0.976144552230835:  58%|█████▊    | 3752/6470 [50:11<34:08,  1.33it/s]
1|3753|Loss: 0.7679179310798645:  58%|█████▊    | 3752/6470 [50:12<34:08,  1.33it/s]
1|3753|Loss: 0.7679179310798645:  58%|█████▊    | 3753/6470 [50:12<37:16,  1.21it/s]
1|3754|Loss: 0.9274705052375793:  58%|█████▊    | 3753/6470 [50:13<37:16,  1.21it/s]
1|3754|Loss: 0.9274705052375793:  58%|█████▊    | 3754/6470 [50:13<35:18,  1.28it/s]
1|3755|Loss: 0.926669716835022:  58%|█████▊    | 3754/6470 [50:13<35:18,  1.28it/s] 
1|3755|Loss: 0.926669716835022:  58%|█████▊    | 3755/6470 [50:14<33:47,  1.34it/s]
1|3756|Loss: 0.9671576619148254:  58%|█████▊    | 3755/6470 [50:14<33:47,  1.34it/s]
1|3756|Loss: 0.9671576619148254:  58%|█████▊    | 3756/6470 [50:14<33:41,  1.34it/s]
1|3757|Loss: 0.8081997632980347:  58%|█████▊    | 3756/6470 [50:15<33:41,  1.34it/s]
1|3757|Loss: 0.8081997632980347:  58%|█████▊    | 3757/6470 [50:15<33:56,  1.33it/s]
1|3758|Loss: 0.9282544851303101:  58%|█████▊    | 3757/6470 [50:16<33:56,  1.33it/s]
1|3758|Loss: 0.9282544851303101:  58%|█████▊    | 3758/6470 [50:16<34:22,  1.31it/s]
1|3759|Loss: 0.925881028175354:  58%|█████▊    | 3758/6470 [50:16<34:22,  1.31it/s] 
1|3759|Loss: 0.925881028175354:  58%|█████▊    | 3759/6470 [50:17<32:52,  1.37it/s]
1|3760|Loss: 0.9328893423080444:  58%|█████▊    | 3759/6470 [50:17<32:52,  1.37it/s]
1|3760|Loss: 0.9328893423080444:  58%|█████▊    | 3760/6470 [50:17<32:54,  1.37it/s]
1|3761|Loss: 1.0440223217010498:  58%|█████▊    | 3760/6470 [50:18<32:54,  1.37it/s]
1|3761|Loss: 1.0440223217010498:  58%|█████▊    | 3761/6470 [50:18<33:12,  1.36it/s]
1|3762|Loss: 0.9340869784355164:  58%|█████▊    | 3761/6470 [50:18<33:12,  1.36it/s]
1|3762|Loss: 0.9340869784355164:  58%|█████▊    | 3762/6470 [50:19<32:35,  1.38it/s]
1|3763|Loss: 0.9267727732658386:  58%|█████▊    | 3762/6470 [50:19<32:35,  1.38it/s]
1|3763|Loss: 0.9267727732658386:  58%|█████▊    | 3763/6470 [50:19<33:15,  1.36it/s]
1|3764|Loss: 1.061234951019287:  58%|█████▊    | 3763/6470 [50:20<33:15,  1.36it/s] 
1|3764|Loss: 1.061234951019287:  58%|█████▊    | 3764/6470 [50:20<31:31,  1.43it/s]
1|3765|Loss: 0.8033080101013184:  58%|█████▊    | 3764/6470 [50:21<31:31,  1.43it/s]
1|3765|Loss: 0.8033080101013184:  58%|█████▊    | 3765/6470 [50:21<33:58,  1.33it/s]
1|3766|Loss: 0.9340292811393738:  58%|█████▊    | 3765/6470 [50:21<33:58,  1.33it/s]
1|3766|Loss: 0.9340292811393738:  58%|█████▊    | 3766/6470 [50:22<33:43,  1.34it/s]
1|3767|Loss: 0.8497395515441895:  58%|█████▊    | 3766/6470 [50:22<33:43,  1.34it/s]
1|3767|Loss: 0.8497395515441895:  58%|█████▊    | 3767/6470 [50:23<35:41,  1.26it/s]
1|3768|Loss: 0.8379839658737183:  58%|█████▊    | 3767/6470 [50:23<35:41,  1.26it/s]
1|3768|Loss: 0.8379839658737183:  58%|█████▊    | 3768/6470 [50:23<35:05,  1.28it/s]
1|3769|Loss: 1.0149288177490234:  58%|█████▊    | 3768/6470 [50:24<35:05,  1.28it/s]
1|3769|Loss: 1.0149288177490234:  58%|█████▊    | 3769/6470 [50:24<35:59,  1.25it/s]
1|3770|Loss: 0.9384031295776367:  58%|█████▊    | 3769/6470 [50:25<35:59,  1.25it/s]
1|3770|Loss: 0.9384031295776367:  58%|█████▊    | 3770/6470 [50:25<35:42,  1.26it/s]
1|3771|Loss: 0.9688843488693237:  58%|█████▊    | 3770/6470 [50:25<35:42,  1.26it/s]
1|3771|Loss: 0.9688843488693237:  58%|█████▊    | 3771/6470 [50:26<35:34,  1.26it/s]
1|3772|Loss: 0.8756358027458191:  58%|█████▊    | 3771/6470 [50:26<35:34,  1.26it/s]
1|3772|Loss: 0.8756358027458191:  58%|█████▊    | 3772/6470 [50:26<32:58,  1.36it/s]
1|3773|Loss: 0.8676763772964478:  58%|█████▊    | 3772/6470 [50:27<32:58,  1.36it/s]
1|3773|Loss: 0.8676763772964478:  58%|█████▊    | 3773/6470 [50:27<35:30,  1.27it/s]
1|3774|Loss: 1.012533187866211:  58%|█████▊    | 3773/6470 [50:28<35:30,  1.27it/s] 
1|3774|Loss: 1.012533187866211:  58%|█████▊    | 3774/6470 [50:28<35:13,  1.28it/s]
1|3775|Loss: 0.88059401512146:  58%|█████▊    | 3774/6470 [50:28<35:13,  1.28it/s] 
1|3775|Loss: 0.88059401512146:  58%|█████▊    | 3775/6470 [50:29<32:56,  1.36it/s]
1|3776|Loss: 0.9101450443267822:  58%|█████▊    | 3775/6470 [50:29<32:56,  1.36it/s]
1|3776|Loss: 0.9101450443267822:  58%|█████▊    | 3776/6470 [50:29<31:16,  1.44it/s]
1|3777|Loss: 0.8477855920791626:  58%|█████▊    | 3776/6470 [50:30<31:16,  1.44it/s]
1|3777|Loss: 0.8477855920791626:  58%|█████▊    | 3777/6470 [50:30<32:01,  1.40it/s]
1|3778|Loss: 0.952334463596344:  58%|█████▊    | 3777/6470 [50:30<32:01,  1.40it/s] 
1|3778|Loss: 0.952334463596344:  58%|█████▊    | 3778/6470 [50:31<31:47,  1.41it/s]
1|3779|Loss: 0.8691830635070801:  58%|█████▊    | 3778/6470 [50:31<31:47,  1.41it/s]
1|3779|Loss: 0.8691830635070801:  58%|█████▊    | 3779/6470 [50:31<32:07,  1.40it/s]
1|3780|Loss: 0.9939590096473694:  58%|█████▊    | 3779/6470 [50:32<32:07,  1.40it/s]
1|3780|Loss: 0.9939590096473694:  58%|█████▊    | 3780/6470 [50:32<34:24,  1.30it/s]
1|3781|Loss: 0.7879722118377686:  58%|█████▊    | 3780/6470 [50:33<34:24,  1.30it/s]
1|3781|Loss: 0.7879722118377686:  58%|█████▊    | 3781/6470 [50:33<34:13,  1.31it/s]
1|3782|Loss: 0.8266763687133789:  58%|█████▊    | 3781/6470 [50:34<34:13,  1.31it/s]
1|3782|Loss: 0.8266763687133789:  58%|█████▊    | 3782/6470 [50:34<33:02,  1.36it/s]
1|3783|Loss: 0.8560306429862976:  58%|█████▊    | 3782/6470 [50:34<33:02,  1.36it/s]
1|3783|Loss: 0.8560306429862976:  58%|█████▊    | 3783/6470 [50:34<32:28,  1.38it/s]
1|3784|Loss: 0.7750004529953003:  58%|█████▊    | 3783/6470 [50:35<32:28,  1.38it/s]
1|3784|Loss: 0.7750004529953003:  58%|█████▊    | 3784/6470 [50:35<34:45,  1.29it/s]
1|3785|Loss: 0.8535891771316528:  58%|█████▊    | 3784/6470 [50:36<34:45,  1.29it/s]
1|3785|Loss: 0.8535891771316528:  59%|█████▊    | 3785/6470 [50:36<37:07,  1.21it/s]
1|3786|Loss: 0.8803804516792297:  59%|█████▊    | 3785/6470 [50:37<37:07,  1.21it/s]
1|3786|Loss: 0.8803804516792297:  59%|█████▊    | 3786/6470 [50:37<35:58,  1.24it/s]
1|3787|Loss: 0.9648336172103882:  59%|█████▊    | 3786/6470 [50:37<35:58,  1.24it/s]
1|3787|Loss: 0.9648336172103882:  59%|█████▊    | 3787/6470 [50:38<34:16,  1.30it/s]
1|3788|Loss: 0.9448627829551697:  59%|█████▊    | 3787/6470 [50:38<34:16,  1.30it/s]
1|3788|Loss: 0.9448627829551697:  59%|█████▊    | 3788/6470 [50:39<35:25,  1.26it/s]
1|3789|Loss: 0.942155122756958:  59%|█████▊    | 3788/6470 [50:39<35:25,  1.26it/s] 
1|3789|Loss: 0.942155122756958:  59%|█████▊    | 3789/6470 [50:39<35:19,  1.26it/s]
1|3790|Loss: 1.0246403217315674:  59%|█████▊    | 3789/6470 [50:40<35:19,  1.26it/s]
1|3790|Loss: 1.0246403217315674:  59%|█████▊    | 3790/6470 [50:40<36:47,  1.21it/s]
1|3791|Loss: 0.9632646441459656:  59%|█████▊    | 3790/6470 [50:41<36:47,  1.21it/s]
1|3791|Loss: 0.9632646441459656:  59%|█████▊    | 3791/6470 [50:41<36:33,  1.22it/s]
1|3792|Loss: 0.9368442296981812:  59%|█████▊    | 3791/6470 [50:42<36:33,  1.22it/s]
1|3792|Loss: 0.9368442296981812:  59%|█████▊    | 3792/6470 [50:42<35:49,  1.25it/s]
1|3793|Loss: 1.0531445741653442:  59%|█████▊    | 3792/6470 [50:42<35:49,  1.25it/s]
1|3793|Loss: 1.0531445741653442:  59%|█████▊    | 3793/6470 [50:43<37:04,  1.20it/s]
1|3794|Loss: 0.9227837324142456:  59%|█████▊    | 3793/6470 [50:43<37:04,  1.20it/s]
1|3794|Loss: 0.9227837324142456:  59%|█████▊    | 3794/6470 [50:44<38:15,  1.17it/s]
1|3795|Loss: 0.820663332939148:  59%|█████▊    | 3794/6470 [50:44<38:15,  1.17it/s] 
1|3795|Loss: 0.820663332939148:  59%|█████▊    | 3795/6470 [50:44<35:20,  1.26it/s]
1|3796|Loss: 0.9914942383766174:  59%|█████▊    | 3795/6470 [50:45<35:20,  1.26it/s]
1|3796|Loss: 0.9914942383766174:  59%|█████▊    | 3796/6470 [50:45<33:15,  1.34it/s]
1|3797|Loss: 0.8921041488647461:  59%|█████▊    | 3796/6470 [50:45<33:15,  1.34it/s]
1|3797|Loss: 0.8921041488647461:  59%|█████▊    | 3797/6470 [50:46<31:01,  1.44it/s]
1|3798|Loss: 1.0114320516586304:  59%|█████▊    | 3797/6470 [50:46<31:01,  1.44it/s]
1|3798|Loss: 1.0114320516586304:  59%|█████▊    | 3798/6470 [50:46<29:11,  1.53it/s]
1|3799|Loss: 0.8804370164871216:  59%|█████▊    | 3798/6470 [50:46<29:11,  1.53it/s]
1|3799|Loss: 0.8804370164871216:  59%|█████▊    | 3799/6470 [50:47<29:29,  1.51it/s]
1|3800|Loss: 0.7561030983924866:  59%|█████▊    | 3799/6470 [50:47<29:29,  1.51it/s]
1|3800|Loss: 0.7561030983924866:  59%|█████▊    | 3800/6470 [50:47<26:38,  1.67it/s]
1|3801|Loss: 0.932465672492981:  59%|█████▊    | 3800/6470 [50:48<26:38,  1.67it/s] 
1|3801|Loss: 0.932465672492981:  59%|█████▊    | 3801/6470 [50:48<27:42,  1.61it/s]
1|3802|Loss: 0.7792701721191406:  59%|█████▊    | 3801/6470 [50:48<27:42,  1.61it/s]
1|3802|Loss: 0.7792701721191406:  59%|█████▉    | 3802/6470 [50:49<31:03,  1.43it/s]
1|3803|Loss: 0.9001955986022949:  59%|█████▉    | 3802/6470 [50:49<31:03,  1.43it/s]
1|3803|Loss: 0.9001955986022949:  59%|█████▉    | 3803/6470 [50:49<30:45,  1.45it/s]
1|3804|Loss: 0.8444634079933167:  59%|█████▉    | 3803/6470 [50:50<30:45,  1.45it/s]
1|3804|Loss: 0.8444634079933167:  59%|█████▉    | 3804/6470 [50:50<29:46,  1.49it/s]
1|3805|Loss: 1.0140354633331299:  59%|█████▉    | 3804/6470 [50:51<29:46,  1.49it/s]
1|3805|Loss: 1.0140354633331299:  59%|█████▉    | 3805/6470 [50:51<31:53,  1.39it/s]
1|3806|Loss: 0.9087018966674805:  59%|█████▉    | 3805/6470 [50:51<31:53,  1.39it/s]
1|3806|Loss: 0.9087018966674805:  59%|█████▉    | 3806/6470 [50:52<33:56,  1.31it/s]
1|3807|Loss: 1.0004782676696777:  59%|█████▉    | 3806/6470 [50:52<33:56,  1.31it/s]
1|3807|Loss: 1.0004782676696777:  59%|█████▉    | 3807/6470 [50:53<34:30,  1.29it/s]
1|3808|Loss: 0.938775360584259:  59%|█████▉    | 3807/6470 [50:53<34:30,  1.29it/s] 
1|3808|Loss: 0.938775360584259:  59%|█████▉    | 3808/6470 [50:53<33:39,  1.32it/s]
1|3809|Loss: 0.9475668668746948:  59%|█████▉    | 3808/6470 [50:54<33:39,  1.32it/s]
1|3809|Loss: 0.9475668668746948:  59%|█████▉    | 3809/6470 [50:54<31:59,  1.39it/s]
1|3810|Loss: 1.0626955032348633:  59%|█████▉    | 3809/6470 [50:54<31:59,  1.39it/s]
1|3810|Loss: 1.0626955032348633:  59%|█████▉    | 3810/6470 [50:55<32:30,  1.36it/s]
1|3811|Loss: 1.0022871494293213:  59%|█████▉    | 3810/6470 [50:55<32:30,  1.36it/s]
1|3811|Loss: 1.0022871494293213:  59%|█████▉    | 3811/6470 [50:56<33:25,  1.33it/s]
1|3812|Loss: 0.8285140991210938:  59%|█████▉    | 3811/6470 [50:56<33:25,  1.33it/s]
1|3812|Loss: 0.8285140991210938:  59%|█████▉    | 3812/6470 [50:56<31:08,  1.42it/s]
1|3813|Loss: 0.9004397392272949:  59%|█████▉    | 3812/6470 [50:57<31:08,  1.42it/s]
1|3813|Loss: 0.9004397392272949:  59%|█████▉    | 3813/6470 [50:57<33:11,  1.33it/s]
1|3814|Loss: 0.8357166647911072:  59%|█████▉    | 3813/6470 [50:57<33:11,  1.33it/s]
1|3814|Loss: 0.8357166647911072:  59%|█████▉    | 3814/6470 [50:58<33:22,  1.33it/s]
1|3815|Loss: 1.1210592985153198:  59%|█████▉    | 3814/6470 [50:58<33:22,  1.33it/s]
1|3815|Loss: 1.1210592985153198:  59%|█████▉    | 3815/6470 [50:59<35:58,  1.23it/s]
1|3816|Loss: 0.9306356310844421:  59%|█████▉    | 3815/6470 [50:59<35:58,  1.23it/s]
1|3816|Loss: 0.9306356310844421:  59%|█████▉    | 3816/6470 [50:59<34:32,  1.28it/s]
1|3817|Loss: 1.0175724029541016:  59%|█████▉    | 3816/6470 [51:00<34:32,  1.28it/s]
1|3817|Loss: 1.0175724029541016:  59%|█████▉    | 3817/6470 [51:00<35:57,  1.23it/s]
1|3818|Loss: 1.008298397064209:  59%|█████▉    | 3817/6470 [51:01<35:57,  1.23it/s] 
1|3818|Loss: 1.008298397064209:  59%|█████▉    | 3818/6470 [51:01<36:09,  1.22it/s]
1|3819|Loss: 0.7523179650306702:  59%|█████▉    | 3818/6470 [51:01<36:09,  1.22it/s]
1|3819|Loss: 0.7523179650306702:  59%|█████▉    | 3819/6470 [51:02<33:25,  1.32it/s]
1|3820|Loss: 0.8661894798278809:  59%|█████▉    | 3819/6470 [51:02<33:25,  1.32it/s]
1|3820|Loss: 0.8661894798278809:  59%|█████▉    | 3820/6470 [51:02<29:53,  1.48it/s]
1|3821|Loss: 0.9988284111022949:  59%|█████▉    | 3820/6470 [51:03<29:53,  1.48it/s]
1|3821|Loss: 0.9988284111022949:  59%|█████▉    | 3821/6470 [51:03<30:37,  1.44it/s]
1|3822|Loss: 0.8951187133789062:  59%|█████▉    | 3821/6470 [51:03<30:37,  1.44it/s]
1|3822|Loss: 0.8951187133789062:  59%|█████▉    | 3822/6470 [51:04<31:43,  1.39it/s]
1|3823|Loss: 1.0642375946044922:  59%|█████▉    | 3822/6470 [51:04<31:43,  1.39it/s]
1|3823|Loss: 1.0642375946044922:  59%|█████▉    | 3823/6470 [51:04<32:14,  1.37it/s]
1|3824|Loss: 0.8335203528404236:  59%|█████▉    | 3823/6470 [51:05<32:14,  1.37it/s]
1|3824|Loss: 0.8335203528404236:  59%|█████▉    | 3824/6470 [51:05<30:12,  1.46it/s]
1|3825|Loss: 0.8906300663948059:  59%|█████▉    | 3824/6470 [51:05<30:12,  1.46it/s]
1|3825|Loss: 0.8906300663948059:  59%|█████▉    | 3825/6470 [51:06<29:34,  1.49it/s]
1|3826|Loss: 0.8475221395492554:  59%|█████▉    | 3825/6470 [51:06<29:34,  1.49it/s]
1|3826|Loss: 0.8475221395492554:  59%|█████▉    | 3826/6470 [51:06<29:14,  1.51it/s]
1|3827|Loss: 0.9424888491630554:  59%|█████▉    | 3826/6470 [51:07<29:14,  1.51it/s]
1|3827|Loss: 0.9424888491630554:  59%|█████▉    | 3827/6470 [51:07<29:33,  1.49it/s]
1|3828|Loss: 0.8272736072540283:  59%|█████▉    | 3827/6470 [51:07<29:33,  1.49it/s]
1|3828|Loss: 0.8272736072540283:  59%|█████▉    | 3828/6470 [51:08<30:42,  1.43it/s]
1|3829|Loss: 0.9379773139953613:  59%|█████▉    | 3828/6470 [51:08<30:42,  1.43it/s]
1|3829|Loss: 0.9379773139953613:  59%|█████▉    | 3829/6470 [51:08<28:35,  1.54it/s]
1|3830|Loss: 0.8673523664474487:  59%|█████▉    | 3829/6470 [51:09<28:35,  1.54it/s]
1|3830|Loss: 0.8673523664474487:  59%|█████▉    | 3830/6470 [51:09<28:36,  1.54it/s]
1|3831|Loss: 1.6926145553588867:  59%|█████▉    | 3830/6470 [51:09<28:36,  1.54it/s]
1|3831|Loss: 1.6926145553588867:  59%|█████▉    | 3831/6470 [51:10<30:00,  1.47it/s]
1|3832|Loss: 0.8539466857910156:  59%|█████▉    | 3831/6470 [51:10<30:00,  1.47it/s]
1|3832|Loss: 0.8539466857910156:  59%|█████▉    | 3832/6470 [51:11<33:33,  1.31it/s]
1|3833|Loss: 0.8785710334777832:  59%|█████▉    | 3832/6470 [51:11<33:33,  1.31it/s]
1|3833|Loss: 0.8785710334777832:  59%|█████▉    | 3833/6470 [51:11<31:36,  1.39it/s]
1|3834|Loss: 1.1273573637008667:  59%|█████▉    | 3833/6470 [51:12<31:36,  1.39it/s]
1|3834|Loss: 1.1273573637008667:  59%|█████▉    | 3834/6470 [51:12<28:49,  1.52it/s]
1|3835|Loss: 0.9233275651931763:  59%|█████▉    | 3834/6470 [51:12<28:49,  1.52it/s]
1|3835|Loss: 0.9233275651931763:  59%|█████▉    | 3835/6470 [51:12<26:58,  1.63it/s]
1|3836|Loss: 0.868878185749054:  59%|█████▉    | 3835/6470 [51:13<26:58,  1.63it/s] 
1|3836|Loss: 0.868878185749054:  59%|█████▉    | 3836/6470 [51:13<27:30,  1.60it/s]
1|3837|Loss: 0.9515143632888794:  59%|█████▉    | 3836/6470 [51:13<27:30,  1.60it/s]
1|3837|Loss: 0.9515143632888794:  59%|█████▉    | 3837/6470 [51:14<30:24,  1.44it/s]
1|3838|Loss: 0.8648809194564819:  59%|█████▉    | 3837/6470 [51:14<30:24,  1.44it/s]
1|3838|Loss: 0.8648809194564819:  59%|█████▉    | 3838/6470 [51:15<32:10,  1.36it/s]
1|3839|Loss: 1.0782768726348877:  59%|█████▉    | 3838/6470 [51:15<32:10,  1.36it/s]
1|3839|Loss: 1.0782768726348877:  59%|█████▉    | 3839/6470 [51:15<33:18,  1.32it/s]
1|3840|Loss: 0.9264146089553833:  59%|█████▉    | 3839/6470 [51:16<33:18,  1.32it/s]
1|3840|Loss: 0.9264146089553833:  59%|█████▉    | 3840/6470 [51:16<33:52,  1.29it/s]
1|3841|Loss: 0.9392751455307007:  59%|█████▉    | 3840/6470 [51:17<33:52,  1.29it/s]
1|3841|Loss: 0.9392751455307007:  59%|█████▉    | 3841/6470 [51:17<34:58,  1.25it/s]
1|3842|Loss: 0.9207123517990112:  59%|█████▉    | 3841/6470 [51:18<34:58,  1.25it/s]
1|3842|Loss: 0.9207123517990112:  59%|█████▉    | 3842/6470 [51:18<35:21,  1.24it/s]
1|3843|Loss: 0.7865433096885681:  59%|█████▉    | 3842/6470 [51:18<35:21,  1.24it/s]
1|3843|Loss: 0.7865433096885681:  59%|█████▉    | 3843/6470 [51:19<36:36,  1.20it/s]
1|3844|Loss: 1.1528072357177734:  59%|█████▉    | 3843/6470 [51:19<36:36,  1.20it/s]
1|3844|Loss: 1.1528072357177734:  59%|█████▉    | 3844/6470 [51:20<35:10,  1.24it/s]
1|3845|Loss: 1.020386815071106:  59%|█████▉    | 3844/6470 [51:20<35:10,  1.24it/s] 
1|3845|Loss: 1.020386815071106:  59%|█████▉    | 3845/6470 [51:20<33:49,  1.29it/s]
1|3846|Loss: 0.889244019985199:  59%|█████▉    | 3845/6470 [51:21<33:49,  1.29it/s]
1|3846|Loss: 0.889244019985199:  59%|█████▉    | 3846/6470 [51:21<36:11,  1.21it/s]
1|3847|Loss: 0.9956799149513245:  59%|█████▉    | 3846/6470 [51:22<36:11,  1.21it/s]
1|3847|Loss: 0.9956799149513245:  59%|█████▉    | 3847/6470 [51:22<36:10,  1.21it/s]
1|3848|Loss: 0.7962197065353394:  59%|█████▉    | 3847/6470 [51:22<36:10,  1.21it/s]
1|3848|Loss: 0.7962197065353394:  59%|█████▉    | 3848/6470 [51:23<33:00,  1.32it/s]
1|3849|Loss: 1.0054596662521362:  59%|█████▉    | 3848/6470 [51:23<33:00,  1.32it/s]
1|3849|Loss: 1.0054596662521362:  59%|█████▉    | 3849/6470 [51:23<33:53,  1.29it/s]
1|3850|Loss: 0.7128964066505432:  59%|█████▉    | 3849/6470 [51:24<33:53,  1.29it/s]
1|3850|Loss: 0.7128964066505432:  60%|█████▉    | 3850/6470 [51:25<37:12,  1.17it/s]
1|3851|Loss: 0.8153449892997742:  60%|█████▉    | 3850/6470 [51:25<37:12,  1.17it/s]
1|3851|Loss: 0.8153449892997742:  60%|█████▉    | 3851/6470 [51:25<37:17,  1.17it/s]
1|3852|Loss: 1.0236461162567139:  60%|█████▉    | 3851/6470 [51:26<37:17,  1.17it/s]
1|3852|Loss: 1.0236461162567139:  60%|█████▉    | 3852/6470 [51:26<36:19,  1.20it/s]
1|3853|Loss: 0.8384826183319092:  60%|█████▉    | 3852/6470 [51:27<36:19,  1.20it/s]
1|3853|Loss: 0.8384826183319092:  60%|█████▉    | 3853/6470 [51:27<35:59,  1.21it/s]
1|3854|Loss: 0.9222423434257507:  60%|█████▉    | 3853/6470 [51:27<35:59,  1.21it/s]
1|3854|Loss: 0.9222423434257507:  60%|█████▉    | 3854/6470 [51:28<34:41,  1.26it/s]
1|3855|Loss: 0.9074177145957947:  60%|█████▉    | 3854/6470 [51:28<34:41,  1.26it/s]
1|3855|Loss: 0.9074177145957947:  60%|█████▉    | 3855/6470 [51:28<34:28,  1.26it/s]
1|3856|Loss: 0.8180248737335205:  60%|█████▉    | 3855/6470 [51:29<34:28,  1.26it/s]
1|3856|Loss: 0.8180248737335205:  60%|█████▉    | 3856/6470 [51:29<31:49,  1.37it/s]
1|3857|Loss: 0.8749729990959167:  60%|█████▉    | 3856/6470 [51:29<31:49,  1.37it/s]
1|3857|Loss: 0.8749729990959167:  60%|█████▉    | 3857/6470 [51:30<30:54,  1.41it/s]
1|3858|Loss: 0.9403168559074402:  60%|█████▉    | 3857/6470 [51:30<30:54,  1.41it/s]
1|3858|Loss: 0.9403168559074402:  60%|█████▉    | 3858/6470 [51:30<30:07,  1.45it/s]
1|3859|Loss: 0.9509395360946655:  60%|█████▉    | 3858/6470 [51:31<30:07,  1.45it/s]
1|3859|Loss: 0.9509395360946655:  60%|█████▉    | 3859/6470 [51:31<32:29,  1.34it/s]
1|3860|Loss: 0.9773568511009216:  60%|█████▉    | 3859/6470 [51:32<32:29,  1.34it/s]
1|3860|Loss: 0.9773568511009216:  60%|█████▉    | 3860/6470 [51:32<32:19,  1.35it/s]
1|3861|Loss: 1.059417724609375:  60%|█████▉    | 3860/6470 [51:32<32:19,  1.35it/s] 
1|3861|Loss: 1.059417724609375:  60%|█████▉    | 3861/6470 [51:33<31:23,  1.39it/s]
1|3862|Loss: 0.921491265296936:  60%|█████▉    | 3861/6470 [51:33<31:23,  1.39it/s]
1|3862|Loss: 0.921491265296936:  60%|█████▉    | 3862/6470 [51:33<32:10,  1.35it/s]
1|3863|Loss: 0.9430950284004211:  60%|█████▉    | 3862/6470 [51:34<32:10,  1.35it/s]
1|3863|Loss: 0.9430950284004211:  60%|█████▉    | 3863/6470 [51:34<33:01,  1.32it/s]
1|3864|Loss: 0.9796862006187439:  60%|█████▉    | 3863/6470 [51:35<33:01,  1.32it/s]
1|3864|Loss: 0.9796862006187439:  60%|█████▉    | 3864/6470 [51:35<33:44,  1.29it/s]
1|3865|Loss: 1.1499850749969482:  60%|█████▉    | 3864/6470 [51:35<33:44,  1.29it/s]
1|3865|Loss: 1.1499850749969482:  60%|█████▉    | 3865/6470 [51:36<33:48,  1.28it/s]
1|3866|Loss: 1.0011235475540161:  60%|█████▉    | 3865/6470 [51:36<33:48,  1.28it/s]
1|3866|Loss: 1.0011235475540161:  60%|█████▉    | 3866/6470 [51:37<32:22,  1.34it/s]
1|3867|Loss: 0.9631826877593994:  60%|█████▉    | 3866/6470 [51:37<32:22,  1.34it/s]
1|3867|Loss: 0.9631826877593994:  60%|█████▉    | 3867/6470 [51:37<34:09,  1.27it/s]
1|3868|Loss: 0.9124477505683899:  60%|█████▉    | 3867/6470 [51:38<34:09,  1.27it/s]
1|3868|Loss: 0.9124477505683899:  60%|█████▉    | 3868/6470 [51:38<32:31,  1.33it/s]
1|3869|Loss: 0.9719386696815491:  60%|█████▉    | 3868/6470 [51:38<32:31,  1.33it/s]
1|3869|Loss: 0.9719386696815491:  60%|█████▉    | 3869/6470 [51:39<29:56,  1.45it/s]
1|3870|Loss: 1.0287023782730103:  60%|█████▉    | 3869/6470 [51:39<29:56,  1.45it/s]
1|3870|Loss: 1.0287023782730103:  60%|█████▉    | 3870/6470 [51:39<29:45,  1.46it/s]
1|3871|Loss: 1.1380265951156616:  60%|█████▉    | 3870/6470 [51:40<29:45,  1.46it/s]
1|3871|Loss: 1.1380265951156616:  60%|█████▉    | 3871/6470 [51:40<29:28,  1.47it/s]
1|3872|Loss: 1.0190926790237427:  60%|█████▉    | 3871/6470 [51:40<29:28,  1.47it/s]
1|3872|Loss: 1.0190926790237427:  60%|█████▉    | 3872/6470 [51:41<30:21,  1.43it/s]
1|3873|Loss: 0.9196497201919556:  60%|█████▉    | 3872/6470 [51:41<30:21,  1.43it/s]
1|3873|Loss: 0.9196497201919556:  60%|█████▉    | 3873/6470 [51:41<31:08,  1.39it/s]
1|3874|Loss: 0.9549594521522522:  60%|█████▉    | 3873/6470 [51:42<31:08,  1.39it/s]
1|3874|Loss: 0.9549594521522522:  60%|█████▉    | 3874/6470 [51:42<31:38,  1.37it/s]
1|3875|Loss: 0.8739214539527893:  60%|█████▉    | 3874/6470 [51:43<31:38,  1.37it/s]
1|3875|Loss: 0.8739214539527893:  60%|█████▉    | 3875/6470 [51:43<32:21,  1.34it/s]
1|3876|Loss: 1.671799659729004:  60%|█████▉    | 3875/6470 [51:43<32:21,  1.34it/s] 
1|3876|Loss: 1.671799659729004:  60%|█████▉    | 3876/6470 [51:44<33:14,  1.30it/s]
1|3877|Loss: 1.0822361707687378:  60%|█████▉    | 3876/6470 [51:44<33:14,  1.30it/s]
1|3877|Loss: 1.0822361707687378:  60%|█████▉    | 3877/6470 [51:45<33:49,  1.28it/s]
1|3878|Loss: 0.8903716206550598:  60%|█████▉    | 3877/6470 [51:45<33:49,  1.28it/s]
1|3878|Loss: 0.8903716206550598:  60%|█████▉    | 3878/6470 [51:45<32:57,  1.31it/s]
1|3879|Loss: 0.9489037394523621:  60%|█████▉    | 3878/6470 [51:46<32:57,  1.31it/s]
1|3879|Loss: 0.9489037394523621:  60%|█████▉    | 3879/6470 [51:46<31:48,  1.36it/s]
1|3880|Loss: 0.967132568359375:  60%|█████▉    | 3879/6470 [51:46<31:48,  1.36it/s] 
1|3880|Loss: 0.967132568359375:  60%|█████▉    | 3880/6470 [51:47<32:11,  1.34it/s]
1|3881|Loss: 0.8598198294639587:  60%|█████▉    | 3880/6470 [51:47<32:11,  1.34it/s]
1|3881|Loss: 0.8598198294639587:  60%|█████▉    | 3881/6470 [51:48<32:46,  1.32it/s]
1|3882|Loss: 0.8181647062301636:  60%|█████▉    | 3881/6470 [51:48<32:46,  1.32it/s]
1|3882|Loss: 0.8181647062301636:  60%|██████    | 3882/6470 [51:48<32:06,  1.34it/s]
1|3883|Loss: 0.8277676105499268:  60%|██████    | 3882/6470 [51:49<32:06,  1.34it/s]
1|3883|Loss: 0.8277676105499268:  60%|██████    | 3883/6470 [51:49<30:02,  1.43it/s]
1|3884|Loss: 1.0093908309936523:  60%|██████    | 3883/6470 [51:49<30:02,  1.43it/s]
1|3884|Loss: 1.0093908309936523:  60%|██████    | 3884/6470 [51:49<27:29,  1.57it/s]
1|3885|Loss: 0.9253470301628113:  60%|██████    | 3884/6470 [51:50<27:29,  1.57it/s]
1|3885|Loss: 0.9253470301628113:  60%|██████    | 3885/6470 [51:50<29:46,  1.45it/s]
1|3886|Loss: 1.027417540550232:  60%|██████    | 3885/6470 [51:51<29:46,  1.45it/s] 
1|3886|Loss: 1.027417540550232:  60%|██████    | 3886/6470 [51:51<32:37,  1.32it/s]
1|3887|Loss: 0.7989792227745056:  60%|██████    | 3886/6470 [51:52<32:37,  1.32it/s]
1|3887|Loss: 0.7989792227745056:  60%|██████    | 3887/6470 [51:52<32:28,  1.33it/s]
1|3888|Loss: 1.1079102754592896:  60%|██████    | 3887/6470 [51:52<32:28,  1.33it/s]
1|3888|Loss: 1.1079102754592896:  60%|██████    | 3888/6470 [51:53<32:00,  1.34it/s]
1|3889|Loss: 0.9241681098937988:  60%|██████    | 3888/6470 [51:53<32:00,  1.34it/s]
1|3889|Loss: 0.9241681098937988:  60%|██████    | 3889/6470 [51:53<32:32,  1.32it/s]
1|3890|Loss: 0.9179215431213379:  60%|██████    | 3889/6470 [51:54<32:32,  1.32it/s]
1|3890|Loss: 0.9179215431213379:  60%|██████    | 3890/6470 [51:54<31:03,  1.38it/s]
1|3891|Loss: 0.7727581262588501:  60%|██████    | 3890/6470 [51:54<31:03,  1.38it/s]
1|3891|Loss: 0.7727581262588501:  60%|██████    | 3891/6470 [51:55<30:08,  1.43it/s]
1|3892|Loss: 0.6901682615280151:  60%|██████    | 3891/6470 [51:55<30:08,  1.43it/s]
1|3892|Loss: 0.6901682615280151:  60%|██████    | 3892/6470 [51:56<32:14,  1.33it/s]
1|3893|Loss: 0.8875117301940918:  60%|██████    | 3892/6470 [51:56<32:14,  1.33it/s]
1|3893|Loss: 0.8875117301940918:  60%|██████    | 3893/6470 [51:56<34:00,  1.26it/s]
1|3894|Loss: 0.8254960775375366:  60%|██████    | 3893/6470 [51:57<34:00,  1.26it/s]
1|3894|Loss: 0.8254960775375366:  60%|██████    | 3894/6470 [51:57<34:26,  1.25it/s]
1|3895|Loss: 0.8907381296157837:  60%|██████    | 3894/6470 [51:58<34:26,  1.25it/s]
1|3895|Loss: 0.8907381296157837:  60%|██████    | 3895/6470 [51:58<34:05,  1.26it/s]
1|3896|Loss: 0.8554719090461731:  60%|██████    | 3895/6470 [51:58<34:05,  1.26it/s]
1|3896|Loss: 0.8554719090461731:  60%|██████    | 3896/6470 [51:59<30:45,  1.39it/s]
1|3897|Loss: 0.7968893647193909:  60%|██████    | 3896/6470 [51:59<30:45,  1.39it/s]
1|3897|Loss: 0.7968893647193909:  60%|██████    | 3897/6470 [51:59<28:33,  1.50it/s]
1|3898|Loss: 0.9911081790924072:  60%|██████    | 3897/6470 [52:00<28:33,  1.50it/s]
1|3898|Loss: 0.9911081790924072:  60%|██████    | 3898/6470 [52:00<29:45,  1.44it/s]
1|3899|Loss: 0.8061315417289734:  60%|██████    | 3898/6470 [52:00<29:45,  1.44it/s]
1|3899|Loss: 0.8061315417289734:  60%|██████    | 3899/6470 [52:01<31:44,  1.35it/s]
1|3900|Loss: 1.1272672414779663:  60%|██████    | 3899/6470 [52:01<31:44,  1.35it/s]
1|3900|Loss: 1.1272672414779663:  60%|██████    | 3900/6470 [52:02<32:19,  1.32it/s]
1|3901|Loss: 0.9736382961273193:  60%|██████    | 3900/6470 [52:02<32:19,  1.32it/s]
1|3901|Loss: 0.9736382961273193:  60%|██████    | 3901/6470 [52:02<32:43,  1.31it/s]
1|3902|Loss: 0.9520889520645142:  60%|██████    | 3901/6470 [52:03<32:43,  1.31it/s]
1|3902|Loss: 0.9520889520645142:  60%|██████    | 3902/6470 [52:03<32:43,  1.31it/s]
1|3903|Loss: 0.9433965682983398:  60%|██████    | 3902/6470 [52:03<32:43,  1.31it/s]
1|3903|Loss: 0.9433965682983398:  60%|██████    | 3903/6470 [52:04<33:08,  1.29it/s]
1|3904|Loss: 1.0511704683303833:  60%|██████    | 3903/6470 [52:04<33:08,  1.29it/s]
1|3904|Loss: 1.0511704683303833:  60%|██████    | 3904/6470 [52:05<34:42,  1.23it/s]
1|3905|Loss: 0.7219651937484741:  60%|██████    | 3904/6470 [52:05<34:42,  1.23it/s]
1|3905|Loss: 0.7219651937484741:  60%|██████    | 3905/6470 [52:05<32:42,  1.31it/s]
1|3906|Loss: 0.9891541600227356:  60%|██████    | 3905/6470 [52:06<32:42,  1.31it/s]
1|3906|Loss: 0.9891541600227356:  60%|██████    | 3906/6470 [52:06<31:35,  1.35it/s]
1|3907|Loss: 0.9592073559761047:  60%|██████    | 3906/6470 [52:06<31:35,  1.35it/s]
1|3907|Loss: 0.9592073559761047:  60%|██████    | 3907/6470 [52:07<30:54,  1.38it/s]
1|3908|Loss: 1.079991102218628:  60%|██████    | 3907/6470 [52:07<30:54,  1.38it/s] 
1|3908|Loss: 1.079991102218628:  60%|██████    | 3908/6470 [52:08<31:27,  1.36it/s]
1|3909|Loss: 0.8613282442092896:  60%|██████    | 3908/6470 [52:08<31:27,  1.36it/s]
1|3909|Loss: 0.8613282442092896:  60%|██████    | 3909/6470 [52:08<31:13,  1.37it/s]
1|3910|Loss: 0.968396008014679:  60%|██████    | 3909/6470 [52:09<31:13,  1.37it/s] 
1|3910|Loss: 0.968396008014679:  60%|██████    | 3910/6470 [52:09<31:47,  1.34it/s]
1|3911|Loss: 0.8538480997085571:  60%|██████    | 3910/6470 [52:09<31:47,  1.34it/s]
1|3911|Loss: 0.8538480997085571:  60%|██████    | 3911/6470 [52:10<29:49,  1.43it/s]
1|3912|Loss: 1.0725094079971313:  60%|██████    | 3911/6470 [52:10<29:49,  1.43it/s]
1|3912|Loss: 1.0725094079971313:  60%|██████    | 3912/6470 [52:10<28:26,  1.50it/s]
1|3913|Loss: 0.8900106549263:  60%|██████    | 3912/6470 [52:11<28:26,  1.50it/s]   
1|3913|Loss: 0.8900106549263:  60%|██████    | 3913/6470 [52:11<29:35,  1.44it/s]
1|3914|Loss: 0.8399965167045593:  60%|██████    | 3913/6470 [52:11<29:35,  1.44it/s]
1|3914|Loss: 0.8399965167045593:  60%|██████    | 3914/6470 [52:12<31:46,  1.34it/s]
1|3915|Loss: 0.8987916111946106:  60%|██████    | 3914/6470 [52:12<31:46,  1.34it/s]
1|3915|Loss: 0.8987916111946106:  61%|██████    | 3915/6470 [52:13<33:44,  1.26it/s]
1|3916|Loss: 0.9593157172203064:  61%|██████    | 3915/6470 [52:13<33:44,  1.26it/s]
1|3916|Loss: 0.9593157172203064:  61%|██████    | 3916/6470 [52:13<31:58,  1.33it/s]
1|3917|Loss: 0.8638268113136292:  61%|██████    | 3916/6470 [52:14<31:58,  1.33it/s]
1|3917|Loss: 0.8638268113136292:  61%|██████    | 3917/6470 [52:14<31:04,  1.37it/s]
1|3918|Loss: 1.0068880319595337:  61%|██████    | 3917/6470 [52:15<31:04,  1.37it/s]
1|3918|Loss: 1.0068880319595337:  61%|██████    | 3918/6470 [52:15<31:44,  1.34it/s]
1|3919|Loss: 0.9469860792160034:  61%|██████    | 3918/6470 [52:15<31:44,  1.34it/s]
1|3919|Loss: 0.9469860792160034:  61%|██████    | 3919/6470 [52:16<32:18,  1.32it/s]
1|3920|Loss: 0.887421727180481:  61%|██████    | 3919/6470 [52:16<32:18,  1.32it/s] 
1|3920|Loss: 0.887421727180481:  61%|██████    | 3920/6470 [52:17<34:02,  1.25it/s]
1|3921|Loss: 0.8919228315353394:  61%|██████    | 3920/6470 [52:17<34:02,  1.25it/s]
1|3921|Loss: 0.8919228315353394:  61%|██████    | 3921/6470 [52:17<33:04,  1.28it/s]
1|3922|Loss: 0.9983258247375488:  61%|██████    | 3921/6470 [52:18<33:04,  1.28it/s]
1|3922|Loss: 0.9983258247375488:  61%|██████    | 3922/6470 [52:18<33:08,  1.28it/s]
1|3923|Loss: 0.9005763530731201:  61%|██████    | 3922/6470 [52:18<33:08,  1.28it/s]
1|3923|Loss: 0.9005763530731201:  61%|██████    | 3923/6470 [52:19<32:34,  1.30it/s]
1|3924|Loss: 0.980344295501709:  61%|██████    | 3923/6470 [52:19<32:34,  1.30it/s] 
1|3924|Loss: 0.980344295501709:  61%|██████    | 3924/6470 [52:19<31:01,  1.37it/s]
1|3925|Loss: 0.8919045329093933:  61%|██████    | 3924/6470 [52:20<31:01,  1.37it/s]
1|3925|Loss: 0.8919045329093933:  61%|██████    | 3925/6470 [52:20<32:38,  1.30it/s]
1|3926|Loss: 1.0137524604797363:  61%|██████    | 3925/6470 [52:21<32:38,  1.30it/s]
1|3926|Loss: 1.0137524604797363:  61%|██████    | 3926/6470 [52:21<31:58,  1.33it/s]
1|3927|Loss: 1.051658272743225:  61%|██████    | 3926/6470 [52:21<31:58,  1.33it/s] 
1|3927|Loss: 1.051658272743225:  61%|██████    | 3927/6470 [52:22<32:25,  1.31it/s]
1|3928|Loss: 0.9238154888153076:  61%|██████    | 3927/6470 [52:22<32:25,  1.31it/s]
1|3928|Loss: 0.9238154888153076:  61%|██████    | 3928/6470 [52:23<33:07,  1.28it/s]
1|3929|Loss: 0.878669261932373:  61%|██████    | 3928/6470 [52:23<33:07,  1.28it/s] 
1|3929|Loss: 0.878669261932373:  61%|██████    | 3929/6470 [52:23<33:33,  1.26it/s]
1|3930|Loss: 0.7617579698562622:  61%|██████    | 3929/6470 [52:24<33:33,  1.26it/s]
1|3930|Loss: 0.7617579698562622:  61%|██████    | 3930/6470 [52:24<34:48,  1.22it/s]
1|3931|Loss: 1.097921371459961:  61%|██████    | 3930/6470 [52:25<34:48,  1.22it/s] 
1|3931|Loss: 1.097921371459961:  61%|██████    | 3931/6470 [52:25<34:47,  1.22it/s]
1|3932|Loss: 0.8081912994384766:  61%|██████    | 3931/6470 [52:26<34:47,  1.22it/s]
1|3932|Loss: 0.8081912994384766:  61%|██████    | 3932/6470 [52:26<37:07,  1.14it/s]
1|3933|Loss: 1.1222889423370361:  61%|██████    | 3932/6470 [52:27<37:07,  1.14it/s]
1|3933|Loss: 1.1222889423370361:  61%|██████    | 3933/6470 [52:27<37:39,  1.12it/s]
1|3934|Loss: 0.8476619124412537:  61%|██████    | 3933/6470 [52:28<37:39,  1.12it/s]
1|3934|Loss: 0.8476619124412537:  61%|██████    | 3934/6470 [52:28<37:41,  1.12it/s]
1|3935|Loss: 0.9848862886428833:  61%|██████    | 3934/6470 [52:28<37:41,  1.12it/s]
1|3935|Loss: 0.9848862886428833:  61%|██████    | 3935/6470 [52:29<36:44,  1.15it/s]
1|3936|Loss: 0.905927836894989:  61%|██████    | 3935/6470 [52:29<36:44,  1.15it/s] 
1|3936|Loss: 0.905927836894989:  61%|██████    | 3936/6470 [52:30<34:58,  1.21it/s]
1|3937|Loss: 0.9161664247512817:  61%|██████    | 3936/6470 [52:30<34:58,  1.21it/s]
1|3937|Loss: 0.9161664247512817:  61%|██████    | 3937/6470 [52:30<34:33,  1.22it/s]
1|3938|Loss: 1.0039058923721313:  61%|██████    | 3937/6470 [52:31<34:33,  1.22it/s]
1|3938|Loss: 1.0039058923721313:  61%|██████    | 3938/6470 [52:31<35:01,  1.20it/s]
1|3939|Loss: 0.8730595111846924:  61%|██████    | 3938/6470 [52:32<35:01,  1.20it/s]
1|3939|Loss: 0.8730595111846924:  61%|██████    | 3939/6470 [52:32<36:06,  1.17it/s]
1|3940|Loss: 0.9717346429824829:  61%|██████    | 3939/6470 [52:33<36:06,  1.17it/s]
1|3940|Loss: 0.9717346429824829:  61%|██████    | 3940/6470 [52:33<35:52,  1.18it/s]
1|3941|Loss: 1.0110664367675781:  61%|██████    | 3940/6470 [52:33<35:52,  1.18it/s]
1|3941|Loss: 1.0110664367675781:  61%|██████    | 3941/6470 [52:34<34:57,  1.21it/s]
1|3942|Loss: 0.8369297981262207:  61%|██████    | 3941/6470 [52:34<34:57,  1.21it/s]
1|3942|Loss: 0.8369297981262207:  61%|██████    | 3942/6470 [52:35<34:14,  1.23it/s]
1|3943|Loss: 0.930876612663269:  61%|██████    | 3942/6470 [52:35<34:14,  1.23it/s] 
1|3943|Loss: 0.930876612663269:  61%|██████    | 3943/6470 [52:35<33:52,  1.24it/s]
1|3944|Loss: 0.9882223010063171:  61%|██████    | 3943/6470 [52:36<33:52,  1.24it/s]
1|3944|Loss: 0.9882223010063171:  61%|██████    | 3944/6470 [52:36<33:55,  1.24it/s]
1|3945|Loss: 0.8459829688072205:  61%|██████    | 3944/6470 [52:37<33:55,  1.24it/s]
1|3945|Loss: 0.8459829688072205:  61%|██████    | 3945/6470 [52:37<33:33,  1.25it/s]
1|3946|Loss: 0.9000424146652222:  61%|██████    | 3945/6470 [52:37<33:33,  1.25it/s]
1|3946|Loss: 0.9000424146652222:  61%|██████    | 3946/6470 [52:38<31:43,  1.33it/s]
1|3947|Loss: 0.9168359637260437:  61%|██████    | 3946/6470 [52:38<31:43,  1.33it/s]
1|3947|Loss: 0.9168359637260437:  61%|██████    | 3947/6470 [52:38<31:47,  1.32it/s]
1|3948|Loss: 0.8899980783462524:  61%|██████    | 3947/6470 [52:39<31:47,  1.32it/s]
1|3948|Loss: 0.8899980783462524:  61%|██████    | 3948/6470 [52:39<30:23,  1.38it/s]
1|3949|Loss: 1.1098977327346802:  61%|██████    | 3948/6470 [52:39<30:23,  1.38it/s]
1|3949|Loss: 1.1098977327346802:  61%|██████    | 3949/6470 [52:40<29:54,  1.41it/s]
1|3950|Loss: 1.0312392711639404:  61%|██████    | 3949/6470 [52:40<29:54,  1.41it/s]
1|3950|Loss: 1.0312392711639404:  61%|██████    | 3950/6470 [52:40<29:02,  1.45it/s]
1|3951|Loss: 0.8159778118133545:  61%|██████    | 3950/6470 [52:41<29:02,  1.45it/s]
1|3951|Loss: 0.8159778118133545:  61%|██████    | 3951/6470 [52:41<28:32,  1.47it/s]
1|3952|Loss: 1.0989487171173096:  61%|██████    | 3951/6470 [52:41<28:32,  1.47it/s]
1|3952|Loss: 1.0989487171173096:  61%|██████    | 3952/6470 [52:42<28:39,  1.46it/s]
1|3953|Loss: 0.8450597524642944:  61%|██████    | 3952/6470 [52:42<28:39,  1.46it/s]
1|3953|Loss: 0.8450597524642944:  61%|██████    | 3953/6470 [52:42<28:57,  1.45it/s]
1|3954|Loss: 1.0227388143539429:  61%|██████    | 3953/6470 [52:43<28:57,  1.45it/s]
1|3954|Loss: 1.0227388143539429:  61%|██████    | 3954/6470 [52:43<26:47,  1.56it/s]
1|3955|Loss: 0.9664607644081116:  61%|██████    | 3954/6470 [52:43<26:47,  1.56it/s]
1|3955|Loss: 0.9664607644081116:  61%|██████    | 3955/6470 [52:44<27:49,  1.51it/s]
1|3956|Loss: 0.861897349357605:  61%|██████    | 3955/6470 [52:44<27:49,  1.51it/s] 
1|3956|Loss: 0.861897349357605:  61%|██████    | 3956/6470 [52:44<29:10,  1.44it/s]
1|3957|Loss: 0.849425196647644:  61%|██████    | 3956/6470 [52:45<29:10,  1.44it/s]
1|3957|Loss: 0.849425196647644:  61%|██████    | 3957/6470 [52:45<29:56,  1.40it/s]
1|3958|Loss: 0.91718590259552:  61%|██████    | 3957/6470 [52:45<29:56,  1.40it/s] 
1|3958|Loss: 0.91718590259552:  61%|██████    | 3958/6470 [52:46<29:34,  1.42it/s]
1|3959|Loss: 1.1762559413909912:  61%|██████    | 3958/6470 [52:46<29:34,  1.42it/s]
1|3959|Loss: 1.1762559413909912:  61%|██████    | 3959/6470 [52:47<31:25,  1.33it/s]
1|3960|Loss: 1.0039342641830444:  61%|██████    | 3959/6470 [52:47<31:25,  1.33it/s]
1|3960|Loss: 1.0039342641830444:  61%|██████    | 3960/6470 [52:47<29:57,  1.40it/s]
1|3961|Loss: 1.0580817461013794:  61%|██████    | 3960/6470 [52:48<29:57,  1.40it/s]
1|3961|Loss: 1.0580817461013794:  61%|██████    | 3961/6470 [52:48<30:23,  1.38it/s]
1|3962|Loss: 0.8279209733009338:  61%|██████    | 3961/6470 [52:48<30:23,  1.38it/s]
1|3962|Loss: 0.8279209733009338:  61%|██████    | 3962/6470 [52:49<30:48,  1.36it/s]
1|3963|Loss: 0.7551842331886292:  61%|██████    | 3962/6470 [52:49<30:48,  1.36it/s]
1|3963|Loss: 0.7551842331886292:  61%|██████▏   | 3963/6470 [52:50<31:10,  1.34it/s]
1|3964|Loss: 0.7629865407943726:  61%|██████▏   | 3963/6470 [52:50<31:10,  1.34it/s]
1|3964|Loss: 0.7629865407943726:  61%|██████▏   | 3964/6470 [52:50<32:51,  1.27it/s]
1|3965|Loss: 0.837924063205719:  61%|██████▏   | 3964/6470 [52:51<32:51,  1.27it/s] 
1|3965|Loss: 0.837924063205719:  61%|██████▏   | 3965/6470 [52:51<34:41,  1.20it/s]
1|3966|Loss: 0.9359674453735352:  61%|██████▏   | 3965/6470 [52:52<34:41,  1.20it/s]
1|3966|Loss: 0.9359674453735352:  61%|██████▏   | 3966/6470 [52:52<36:06,  1.16it/s]
1|3967|Loss: 0.899137556552887:  61%|██████▏   | 3966/6470 [52:53<36:06,  1.16it/s] 
1|3967|Loss: 0.899137556552887:  61%|██████▏   | 3967/6470 [52:53<36:40,  1.14it/s]
1|3968|Loss: 0.9891893863677979:  61%|██████▏   | 3967/6470 [52:54<36:40,  1.14it/s]
1|3968|Loss: 0.9891893863677979:  61%|██████▏   | 3968/6470 [52:54<35:57,  1.16it/s]
1|3969|Loss: 0.866242527961731:  61%|██████▏   | 3968/6470 [52:54<35:57,  1.16it/s] 
1|3969|Loss: 0.866242527961731:  61%|██████▏   | 3969/6470 [52:55<33:29,  1.24it/s]
1|3970|Loss: 0.9285075068473816:  61%|██████▏   | 3969/6470 [52:55<33:29,  1.24it/s]
1|3970|Loss: 0.9285075068473816:  61%|██████▏   | 3970/6470 [52:56<34:41,  1.20it/s]
1|3971|Loss: 0.86055588722229:  61%|██████▏   | 3970/6470 [52:56<34:41,  1.20it/s]  
1|3971|Loss: 0.86055588722229:  61%|██████▏   | 3971/6470 [52:56<32:47,  1.27it/s]
1|3972|Loss: 0.8301026821136475:  61%|██████▏   | 3971/6470 [52:57<32:47,  1.27it/s]
1|3972|Loss: 0.8301026821136475:  61%|██████▏   | 3972/6470 [52:57<32:29,  1.28it/s]
1|3973|Loss: 0.7418615818023682:  61%|██████▏   | 3972/6470 [52:58<32:29,  1.28it/s]
1|3973|Loss: 0.7418615818023682:  61%|██████▏   | 3973/6470 [52:58<33:20,  1.25it/s]
1|3974|Loss: 0.9339810013771057:  61%|██████▏   | 3973/6470 [52:58<33:20,  1.25it/s]
1|3974|Loss: 0.9339810013771057:  61%|██████▏   | 3974/6470 [52:59<32:01,  1.30it/s]
1|3975|Loss: 1.0277760028839111:  61%|██████▏   | 3974/6470 [52:59<32:01,  1.30it/s]
1|3975|Loss: 1.0277760028839111:  61%|██████▏   | 3975/6470 [52:59<32:00,  1.30it/s]
1|3976|Loss: 1.0162911415100098:  61%|██████▏   | 3975/6470 [53:00<32:00,  1.30it/s]
1|3976|Loss: 1.0162911415100098:  61%|██████▏   | 3976/6470 [53:00<32:27,  1.28it/s]
1|3977|Loss: 1.0495750904083252:  61%|██████▏   | 3976/6470 [53:01<32:27,  1.28it/s]
1|3977|Loss: 1.0495750904083252:  61%|██████▏   | 3977/6470 [53:01<29:39,  1.40it/s]
1|3978|Loss: 0.9344066977500916:  61%|██████▏   | 3977/6470 [53:01<29:39,  1.40it/s]
1|3978|Loss: 0.9344066977500916:  61%|██████▏   | 3978/6470 [53:01<29:20,  1.42it/s]
1|3979|Loss: 1.1794908046722412:  61%|██████▏   | 3978/6470 [53:02<29:20,  1.42it/s]
1|3979|Loss: 1.1794908046722412:  61%|██████▏   | 3979/6470 [53:02<30:15,  1.37it/s]
1|3980|Loss: 0.7841753363609314:  61%|██████▏   | 3979/6470 [53:03<30:15,  1.37it/s]
1|3980|Loss: 0.7841753363609314:  62%|██████▏   | 3980/6470 [53:03<32:23,  1.28it/s]
1|3981|Loss: 0.8792126178741455:  62%|██████▏   | 3980/6470 [53:04<32:23,  1.28it/s]
1|3981|Loss: 0.8792126178741455:  62%|██████▏   | 3981/6470 [53:04<34:13,  1.21it/s]
1|3982|Loss: 0.8555178046226501:  62%|██████▏   | 3981/6470 [53:05<34:13,  1.21it/s]
1|3982|Loss: 0.8555178046226501:  62%|██████▏   | 3982/6470 [53:05<35:39,  1.16it/s]
1|3983|Loss: 1.1475428342819214:  62%|██████▏   | 3982/6470 [53:06<35:39,  1.16it/s]
1|3983|Loss: 1.1475428342819214:  62%|██████▏   | 3983/6470 [53:06<36:36,  1.13it/s]
1|3984|Loss: 0.9554424285888672:  62%|██████▏   | 3983/6470 [53:06<36:36,  1.13it/s]
1|3984|Loss: 0.9554424285888672:  62%|██████▏   | 3984/6470 [53:07<35:52,  1.15it/s]
1|3985|Loss: 0.9771782159805298:  62%|██████▏   | 3984/6470 [53:07<35:52,  1.15it/s]
1|3985|Loss: 0.9771782159805298:  62%|██████▏   | 3985/6470 [53:08<35:00,  1.18it/s]
1|3986|Loss: 0.9889083504676819:  62%|██████▏   | 3985/6470 [53:08<35:00,  1.18it/s]
1|3986|Loss: 0.9889083504676819:  62%|██████▏   | 3986/6470 [53:08<35:09,  1.18it/s]
1|3987|Loss: 0.992775022983551:  62%|██████▏   | 3986/6470 [53:09<35:09,  1.18it/s] 
1|3987|Loss: 0.992775022983551:  62%|██████▏   | 3987/6470 [53:09<31:37,  1.31it/s]
1|3988|Loss: 0.95896977186203:  62%|██████▏   | 3987/6470 [53:09<31:37,  1.31it/s] 
1|3988|Loss: 0.95896977186203:  62%|██████▏   | 3988/6470 [53:10<29:53,  1.38it/s]
1|3989|Loss: 0.9485893845558167:  62%|██████▏   | 3988/6470 [53:10<29:53,  1.38it/s]
1|3989|Loss: 0.9485893845558167:  62%|██████▏   | 3989/6470 [53:10<28:31,  1.45it/s]
1|3990|Loss: 1.0332046747207642:  62%|██████▏   | 3989/6470 [53:11<28:31,  1.45it/s]
1|3990|Loss: 1.0332046747207642:  62%|██████▏   | 3990/6470 [53:11<32:01,  1.29it/s]
1|3991|Loss: 0.8374243974685669:  62%|██████▏   | 3990/6470 [53:12<32:01,  1.29it/s]
1|3991|Loss: 0.8374243974685669:  62%|██████▏   | 3991/6470 [53:12<32:57,  1.25it/s]
1|3992|Loss: 0.9183440804481506:  62%|██████▏   | 3991/6470 [53:12<32:57,  1.25it/s]
1|3992|Loss: 0.9183440804481506:  62%|██████▏   | 3992/6470 [53:13<32:59,  1.25it/s]
1|3993|Loss: 0.846333384513855:  62%|██████▏   | 3992/6470 [53:13<32:59,  1.25it/s] 
1|3993|Loss: 0.846333384513855:  62%|██████▏   | 3993/6470 [53:14<31:50,  1.30it/s]
1|3994|Loss: 0.9552818536758423:  62%|██████▏   | 3993/6470 [53:14<31:50,  1.30it/s]
1|3994|Loss: 0.9552818536758423:  62%|██████▏   | 3994/6470 [53:14<31:50,  1.30it/s]
1|3995|Loss: 0.7512544989585876:  62%|██████▏   | 3994/6470 [53:15<31:50,  1.30it/s]
1|3995|Loss: 0.7512544989585876:  62%|██████▏   | 3995/6470 [53:15<31:43,  1.30it/s]
1|3996|Loss: 0.9936990737915039:  62%|██████▏   | 3995/6470 [53:16<31:43,  1.30it/s]
1|3996|Loss: 0.9936990737915039:  62%|██████▏   | 3996/6470 [53:16<33:18,  1.24it/s]
1|3997|Loss: 0.8450689315795898:  62%|██████▏   | 3996/6470 [53:16<33:18,  1.24it/s]
1|3997|Loss: 0.8450689315795898:  62%|██████▏   | 3997/6470 [53:17<31:52,  1.29it/s]
1|3998|Loss: 0.9530611038208008:  62%|██████▏   | 3997/6470 [53:17<31:52,  1.29it/s]
1|3998|Loss: 0.9530611038208008:  62%|██████▏   | 3998/6470 [53:17<30:40,  1.34it/s]
1|3999|Loss: 1.3522765636444092:  62%|██████▏   | 3998/6470 [53:18<30:40,  1.34it/s]
1|3999|Loss: 1.3522765636444092:  62%|██████▏   | 3999/6470 [53:18<32:14,  1.28it/s]
1|4000|Loss: 0.7331891655921936:  62%|██████▏   | 3999/6470 [53:19<32:14,  1.28it/s]
1|4000|Loss: 0.7331891655921936:  62%|██████▏   | 4000/6470 [53:19<32:34,  1.26it/s]
1|4001|Loss: 0.9270795583724976:  62%|██████▏   | 4000/6470 [53:19<32:34,  1.26it/s]
1|4001|Loss: 0.9270795583724976:  62%|██████▏   | 4001/6470 [53:20<30:50,  1.33it/s]
1|4002|Loss: 0.9860768914222717:  62%|██████▏   | 4001/6470 [53:20<30:50,  1.33it/s]
1|4002|Loss: 0.9860768914222717:  62%|██████▏   | 4002/6470 [53:20<29:59,  1.37it/s]
1|4003|Loss: 0.8016029000282288:  62%|██████▏   | 4002/6470 [53:21<29:59,  1.37it/s]
1|4003|Loss: 0.8016029000282288:  62%|██████▏   | 4003/6470 [53:21<28:12,  1.46it/s]
1|4004|Loss: 0.8447798490524292:  62%|██████▏   | 4003/6470 [53:21<28:12,  1.46it/s]
1|4004|Loss: 0.8447798490524292:  62%|██████▏   | 4004/6470 [53:22<30:02,  1.37it/s]
1|4005|Loss: 0.7653348445892334:  62%|██████▏   | 4004/6470 [53:22<30:02,  1.37it/s]
1|4005|Loss: 0.7653348445892334:  62%|██████▏   | 4005/6470 [53:23<33:39,  1.22it/s]
1|4006|Loss: 0.9918250441551208:  62%|██████▏   | 4005/6470 [53:23<33:39,  1.22it/s]
1|4006|Loss: 0.9918250441551208:  62%|██████▏   | 4006/6470 [53:23<30:57,  1.33it/s]
1|4007|Loss: 0.9758956432342529:  62%|██████▏   | 4006/6470 [53:24<30:57,  1.33it/s]
1|4007|Loss: 0.9758956432342529:  62%|██████▏   | 4007/6470 [53:24<30:34,  1.34it/s]
1|4008|Loss: 0.9105797410011292:  62%|██████▏   | 4007/6470 [53:25<30:34,  1.34it/s]
1|4008|Loss: 0.9105797410011292:  62%|██████▏   | 4008/6470 [53:25<31:26,  1.31it/s]
1|4009|Loss: 0.9230760335922241:  62%|██████▏   | 4008/6470 [53:25<31:26,  1.31it/s]
1|4009|Loss: 0.9230760335922241:  62%|██████▏   | 4009/6470 [53:26<30:22,  1.35it/s]
1|4010|Loss: 0.8742237091064453:  62%|██████▏   | 4009/6470 [53:26<30:22,  1.35it/s]
1|4010|Loss: 0.8742237091064453:  62%|██████▏   | 4010/6470 [53:26<29:41,  1.38it/s]
1|4011|Loss: 0.9094284176826477:  62%|██████▏   | 4010/6470 [53:27<29:41,  1.38it/s]
1|4011|Loss: 0.9094284176826477:  62%|██████▏   | 4011/6470 [53:27<29:14,  1.40it/s]
1|4012|Loss: 0.8545759320259094:  62%|██████▏   | 4011/6470 [53:27<29:14,  1.40it/s]
1|4012|Loss: 0.8545759320259094:  62%|██████▏   | 4012/6470 [53:28<27:38,  1.48it/s]
1|4013|Loss: 0.9695987105369568:  62%|██████▏   | 4012/6470 [53:28<27:38,  1.48it/s]
1|4013|Loss: 0.9695987105369568:  62%|██████▏   | 4013/6470 [53:28<28:41,  1.43it/s]
1|4014|Loss: 0.8298739790916443:  62%|██████▏   | 4013/6470 [53:29<28:41,  1.43it/s]
1|4014|Loss: 0.8298739790916443:  62%|██████▏   | 4014/6470 [53:29<28:39,  1.43it/s]
1|4015|Loss: 0.9309654831886292:  62%|██████▏   | 4014/6470 [53:29<28:39,  1.43it/s]
1|4015|Loss: 0.9309654831886292:  62%|██████▏   | 4015/6470 [53:30<31:07,  1.31it/s]
1|4016|Loss: 0.9600988030433655:  62%|██████▏   | 4015/6470 [53:30<31:07,  1.31it/s]
1|4016|Loss: 0.9600988030433655:  62%|██████▏   | 4016/6470 [53:31<32:16,  1.27it/s]
1|4017|Loss: 0.8299881219863892:  62%|██████▏   | 4016/6470 [53:31<32:16,  1.27it/s]
1|4017|Loss: 0.8299881219863892:  62%|██████▏   | 4017/6470 [53:32<32:20,  1.26it/s]
1|4018|Loss: 0.8380980491638184:  62%|██████▏   | 4017/6470 [53:32<32:20,  1.26it/s]
1|4018|Loss: 0.8380980491638184:  62%|██████▏   | 4018/6470 [53:32<32:57,  1.24it/s]
1|4019|Loss: 0.9055336117744446:  62%|██████▏   | 4018/6470 [53:33<32:57,  1.24it/s]
1|4019|Loss: 0.9055336117744446:  62%|██████▏   | 4019/6470 [53:33<30:56,  1.32it/s]
1|4020|Loss: 0.9314455986022949:  62%|██████▏   | 4019/6470 [53:33<30:56,  1.32it/s]
1|4020|Loss: 0.9314455986022949:  62%|██████▏   | 4020/6470 [53:34<29:55,  1.36it/s]
1|4021|Loss: 0.9189168810844421:  62%|██████▏   | 4020/6470 [53:34<29:55,  1.36it/s]
1|4021|Loss: 0.9189168810844421:  62%|██████▏   | 4021/6470 [53:35<32:04,  1.27it/s]
1|4022|Loss: 0.8925762176513672:  62%|██████▏   | 4021/6470 [53:35<32:04,  1.27it/s]
1|4022|Loss: 0.8925762176513672:  62%|██████▏   | 4022/6470 [53:36<33:42,  1.21it/s]
1|4023|Loss: 0.9554378986358643:  62%|██████▏   | 4022/6470 [53:36<33:42,  1.21it/s]
1|4023|Loss: 0.9554378986358643:  62%|██████▏   | 4023/6470 [53:36<33:41,  1.21it/s]
1|4024|Loss: 0.8968769311904907:  62%|██████▏   | 4023/6470 [53:37<33:41,  1.21it/s]
1|4024|Loss: 0.8968769311904907:  62%|██████▏   | 4024/6470 [53:37<31:29,  1.29it/s]
1|4025|Loss: 0.8444741368293762:  62%|██████▏   | 4024/6470 [53:37<31:29,  1.29it/s]
1|4025|Loss: 0.8444741368293762:  62%|██████▏   | 4025/6470 [53:38<28:21,  1.44it/s]
1|4026|Loss: 0.9249323010444641:  62%|██████▏   | 4025/6470 [53:38<28:21,  1.44it/s]
1|4026|Loss: 0.9249323010444641:  62%|██████▏   | 4026/6470 [53:38<28:08,  1.45it/s]
1|4027|Loss: 1.040160059928894:  62%|██████▏   | 4026/6470 [53:39<28:08,  1.45it/s] 
1|4027|Loss: 1.040160059928894:  62%|██████▏   | 4027/6470 [53:39<30:05,  1.35it/s]
1|4028|Loss: 0.9721673130989075:  62%|██████▏   | 4027/6470 [53:40<30:05,  1.35it/s]
1|4028|Loss: 0.9721673130989075:  62%|██████▏   | 4028/6470 [53:40<29:43,  1.37it/s]
1|4029|Loss: 0.9090638756752014:  62%|██████▏   | 4028/6470 [53:40<29:43,  1.37it/s]
1|4029|Loss: 0.9090638756752014:  62%|██████▏   | 4029/6470 [53:41<29:31,  1.38it/s]
1|4030|Loss: 0.8057060837745667:  62%|██████▏   | 4029/6470 [53:41<29:31,  1.38it/s]
1|4030|Loss: 0.8057060837745667:  62%|██████▏   | 4030/6470 [53:41<30:21,  1.34it/s]
1|4031|Loss: 0.9762191772460938:  62%|██████▏   | 4030/6470 [53:42<30:21,  1.34it/s]
1|4031|Loss: 0.9762191772460938:  62%|██████▏   | 4031/6470 [53:42<29:56,  1.36it/s]
1|4032|Loss: 1.0466017723083496:  62%|██████▏   | 4031/6470 [53:42<29:56,  1.36it/s]
1|4032|Loss: 1.0466017723083496:  62%|██████▏   | 4032/6470 [53:43<28:37,  1.42it/s]
1|4033|Loss: 0.9386772513389587:  62%|██████▏   | 4032/6470 [53:43<28:37,  1.42it/s]
1|4033|Loss: 0.9386772513389587:  62%|██████▏   | 4033/6470 [53:43<30:02,  1.35it/s]
1|4034|Loss: 0.7160006165504456:  62%|██████▏   | 4033/6470 [53:44<30:02,  1.35it/s]
1|4034|Loss: 0.7160006165504456:  62%|██████▏   | 4034/6470 [53:44<31:41,  1.28it/s]
1|4035|Loss: 1.0348061323165894:  62%|██████▏   | 4034/6470 [53:45<31:41,  1.28it/s]
1|4035|Loss: 1.0348061323165894:  62%|██████▏   | 4035/6470 [53:45<34:32,  1.17it/s]
1|4036|Loss: 0.909676730632782:  62%|██████▏   | 4035/6470 [53:46<34:32,  1.17it/s] 
1|4036|Loss: 0.909676730632782:  62%|██████▏   | 4036/6470 [53:46<33:36,  1.21it/s]
1|4037|Loss: 0.9361171126365662:  62%|██████▏   | 4036/6470 [53:47<33:36,  1.21it/s]
1|4037|Loss: 0.9361171126365662:  62%|██████▏   | 4037/6470 [53:47<35:27,  1.14it/s]
1|4038|Loss: 1.0435209274291992:  62%|██████▏   | 4037/6470 [53:48<35:27,  1.14it/s]
1|4038|Loss: 1.0435209274291992:  62%|██████▏   | 4038/6470 [53:48<32:41,  1.24it/s]
1|4039|Loss: 0.9802674651145935:  62%|██████▏   | 4038/6470 [53:48<32:41,  1.24it/s]
1|4039|Loss: 0.9802674651145935:  62%|██████▏   | 4039/6470 [53:48<30:49,  1.31it/s]
1|4040|Loss: 0.867464005947113:  62%|██████▏   | 4039/6470 [53:49<30:49,  1.31it/s] 
1|4040|Loss: 0.867464005947113:  62%|██████▏   | 4040/6470 [53:49<32:34,  1.24it/s]
1|4041|Loss: 0.7904625535011292:  62%|██████▏   | 4040/6470 [53:50<32:34,  1.24it/s]
1|4041|Loss: 0.7904625535011292:  62%|██████▏   | 4041/6470 [53:50<32:43,  1.24it/s]
1|4042|Loss: 0.8983518481254578:  62%|██████▏   | 4041/6470 [53:51<32:43,  1.24it/s]
1|4042|Loss: 0.8983518481254578:  62%|██████▏   | 4042/6470 [53:51<33:10,  1.22it/s]
1|4043|Loss: 0.8426306247711182:  62%|██████▏   | 4042/6470 [53:51<33:10,  1.22it/s]
1|4043|Loss: 0.8426306247711182:  62%|██████▏   | 4043/6470 [53:52<30:07,  1.34it/s]
1|4044|Loss: 0.8257076740264893:  62%|██████▏   | 4043/6470 [53:52<30:07,  1.34it/s]
1|4044|Loss: 0.8257076740264893:  63%|██████▎   | 4044/6470 [53:52<29:45,  1.36it/s]
1|4045|Loss: 0.8139411211013794:  63%|██████▎   | 4044/6470 [53:53<29:45,  1.36it/s]
1|4045|Loss: 0.8139411211013794:  63%|██████▎   | 4045/6470 [53:53<30:27,  1.33it/s]
1|4046|Loss: 0.8986678719520569:  63%|██████▎   | 4045/6470 [53:54<30:27,  1.33it/s]
1|4046|Loss: 0.8986678719520569:  63%|██████▎   | 4046/6470 [53:54<31:49,  1.27it/s]
1|4047|Loss: 1.1501814126968384:  63%|██████▎   | 4046/6470 [53:54<31:49,  1.27it/s]
1|4047|Loss: 1.1501814126968384:  63%|██████▎   | 4047/6470 [53:55<31:09,  1.30it/s]
1|4048|Loss: 0.9771042466163635:  63%|██████▎   | 4047/6470 [53:55<31:09,  1.30it/s]
1|4048|Loss: 0.9771042466163635:  63%|██████▎   | 4048/6470 [53:55<30:18,  1.33it/s]
1|4049|Loss: 0.8402417898178101:  63%|██████▎   | 4048/6470 [53:56<30:18,  1.33it/s]
1|4049|Loss: 0.8402417898178101:  63%|██████▎   | 4049/6470 [53:56<30:44,  1.31it/s]
1|4050|Loss: 0.8180480003356934:  63%|██████▎   | 4049/6470 [53:57<30:44,  1.31it/s]
1|4050|Loss: 0.8180480003356934:  63%|██████▎   | 4050/6470 [53:57<32:00,  1.26it/s]
1|4051|Loss: 0.8776828050613403:  63%|██████▎   | 4050/6470 [53:57<32:00,  1.26it/s]
1|4051|Loss: 0.8776828050613403:  63%|██████▎   | 4051/6470 [53:58<30:46,  1.31it/s]
1|4052|Loss: 0.944750189781189:  63%|██████▎   | 4051/6470 [53:58<30:46,  1.31it/s] 
1|4052|Loss: 0.944750189781189:  63%|██████▎   | 4052/6470 [53:58<29:07,  1.38it/s]
1|4053|Loss: 0.8593961000442505:  63%|██████▎   | 4052/6470 [53:59<29:07,  1.38it/s]
1|4053|Loss: 0.8593961000442505:  63%|██████▎   | 4053/6470 [53:59<27:08,  1.48it/s]
1|4054|Loss: 0.910158097743988:  63%|██████▎   | 4053/6470 [53:59<27:08,  1.48it/s] 
1|4054|Loss: 0.910158097743988:  63%|██████▎   | 4054/6470 [54:00<29:30,  1.36it/s]
1|4055|Loss: 0.9312122464179993:  63%|██████▎   | 4054/6470 [54:00<29:30,  1.36it/s]
1|4055|Loss: 0.9312122464179993:  63%|██████▎   | 4055/6470 [54:00<27:45,  1.45it/s]
1|4056|Loss: 0.7028656601905823:  63%|██████▎   | 4055/6470 [54:01<27:45,  1.45it/s]
1|4056|Loss: 0.7028656601905823:  63%|██████▎   | 4056/6470 [54:01<29:55,  1.34it/s]
1|4057|Loss: 0.8899398446083069:  63%|██████▎   | 4056/6470 [54:02<29:55,  1.34it/s]
1|4057|Loss: 0.8899398446083069:  63%|██████▎   | 4057/6470 [54:02<29:18,  1.37it/s]
1|4058|Loss: 0.9304559230804443:  63%|██████▎   | 4057/6470 [54:02<29:18,  1.37it/s]
1|4058|Loss: 0.9304559230804443:  63%|██████▎   | 4058/6470 [54:03<28:40,  1.40it/s]
1|4059|Loss: 0.8328335285186768:  63%|██████▎   | 4058/6470 [54:03<28:40,  1.40it/s]
1|4059|Loss: 0.8328335285186768:  63%|██████▎   | 4059/6470 [54:03<29:25,  1.37it/s]
1|4060|Loss: 0.943763017654419:  63%|██████▎   | 4059/6470 [54:04<29:25,  1.37it/s] 
1|4060|Loss: 0.943763017654419:  63%|██████▎   | 4060/6470 [54:04<27:14,  1.47it/s]
1|4061|Loss: 0.9219309091567993:  63%|██████▎   | 4060/6470 [54:04<27:14,  1.47it/s]
1|4061|Loss: 0.9219309091567993:  63%|██████▎   | 4061/6470 [54:05<25:56,  1.55it/s]
1|4062|Loss: 0.8584522604942322:  63%|██████▎   | 4061/6470 [54:05<25:56,  1.55it/s]
1|4062|Loss: 0.8584522604942322:  63%|██████▎   | 4062/6470 [54:05<27:15,  1.47it/s]
1|4063|Loss: 0.9402725100517273:  63%|██████▎   | 4062/6470 [54:06<27:15,  1.47it/s]
1|4063|Loss: 0.9402725100517273:  63%|██████▎   | 4063/6470 [54:06<27:39,  1.45it/s]
1|4064|Loss: 1.2985421419143677:  63%|██████▎   | 4063/6470 [54:06<27:39,  1.45it/s]
1|4064|Loss: 1.2985421419143677:  63%|██████▎   | 4064/6470 [54:07<30:04,  1.33it/s]
1|4065|Loss: 0.7806885242462158:  63%|██████▎   | 4064/6470 [54:07<30:04,  1.33it/s]
1|4065|Loss: 0.7806885242462158:  63%|██████▎   | 4065/6470 [54:08<31:20,  1.28it/s]
1|4066|Loss: 1.00863516330719:  63%|██████▎   | 4065/6470 [54:08<31:20,  1.28it/s]  
1|4066|Loss: 1.00863516330719:  63%|██████▎   | 4066/6470 [54:09<31:40,  1.26it/s]
1|4067|Loss: 0.9048922657966614:  63%|██████▎   | 4066/6470 [54:09<31:40,  1.26it/s]
1|4067|Loss: 0.9048922657966614:  63%|██████▎   | 4067/6470 [54:09<31:58,  1.25it/s]
1|4068|Loss: 0.9414706826210022:  63%|██████▎   | 4067/6470 [54:10<31:58,  1.25it/s]
1|4068|Loss: 0.9414706826210022:  63%|██████▎   | 4068/6470 [54:10<32:12,  1.24it/s]
1|4069|Loss: 0.9767603278160095:  63%|██████▎   | 4068/6470 [54:11<32:12,  1.24it/s]
1|4069|Loss: 0.9767603278160095:  63%|██████▎   | 4069/6470 [54:11<31:54,  1.25it/s]
1|4070|Loss: 0.9488563537597656:  63%|██████▎   | 4069/6470 [54:11<31:54,  1.25it/s]
1|4070|Loss: 0.9488563537597656:  63%|██████▎   | 4070/6470 [54:12<32:06,  1.25it/s]
1|4071|Loss: 0.8141298890113831:  63%|██████▎   | 4070/6470 [54:12<32:06,  1.25it/s]
1|4071|Loss: 0.8141298890113831:  63%|██████▎   | 4071/6470 [54:12<30:14,  1.32it/s]
1|4072|Loss: 0.934266209602356:  63%|██████▎   | 4071/6470 [54:13<30:14,  1.32it/s] 
1|4072|Loss: 0.934266209602356:  63%|██████▎   | 4072/6470 [54:13<31:35,  1.26it/s]
1|4073|Loss: 0.9716602563858032:  63%|██████▎   | 4072/6470 [54:14<31:35,  1.26it/s]
1|4073|Loss: 0.9716602563858032:  63%|██████▎   | 4073/6470 [54:14<30:31,  1.31it/s]
1|4074|Loss: 1.0160821676254272:  63%|██████▎   | 4073/6470 [54:14<30:31,  1.31it/s]
1|4074|Loss: 1.0160821676254272:  63%|██████▎   | 4074/6470 [54:15<31:11,  1.28it/s]
1|4075|Loss: 0.9092803597450256:  63%|██████▎   | 4074/6470 [54:15<31:11,  1.28it/s]
1|4075|Loss: 0.9092803597450256:  63%|██████▎   | 4075/6470 [54:16<32:48,  1.22it/s]
1|4076|Loss: 0.9488012194633484:  63%|██████▎   | 4075/6470 [54:16<32:48,  1.22it/s]
1|4076|Loss: 0.9488012194633484:  63%|██████▎   | 4076/6470 [54:17<32:21,  1.23it/s]
1|4077|Loss: 0.917957067489624:  63%|██████▎   | 4076/6470 [54:17<32:21,  1.23it/s] 
1|4077|Loss: 0.917957067489624:  63%|██████▎   | 4077/6470 [54:17<32:07,  1.24it/s]
1|4078|Loss: 0.8580979704856873:  63%|██████▎   | 4077/6470 [54:18<32:07,  1.24it/s]
1|4078|Loss: 0.8580979704856873:  63%|██████▎   | 4078/6470 [54:18<31:43,  1.26it/s]
1|4079|Loss: 0.7918298840522766:  63%|██████▎   | 4078/6470 [54:19<31:43,  1.26it/s]
1|4079|Loss: 0.7918298840522766:  63%|██████▎   | 4079/6470 [54:19<32:56,  1.21it/s]
1|4080|Loss: 0.9820464253425598:  63%|██████▎   | 4079/6470 [54:19<32:56,  1.21it/s]
1|4080|Loss: 0.9820464253425598:  63%|██████▎   | 4080/6470 [54:20<32:54,  1.21it/s]
1|4081|Loss: 1.0336010456085205:  63%|██████▎   | 4080/6470 [54:20<32:54,  1.21it/s]
1|4081|Loss: 1.0336010456085205:  63%|██████▎   | 4081/6470 [54:21<31:18,  1.27it/s]
1|4082|Loss: 1.0001797676086426:  63%|██████▎   | 4081/6470 [54:21<31:18,  1.27it/s]
1|4082|Loss: 1.0001797676086426:  63%|██████▎   | 4082/6470 [54:21<30:44,  1.29it/s]
1|4083|Loss: 0.8972809314727783:  63%|██████▎   | 4082/6470 [54:22<30:44,  1.29it/s]
1|4083|Loss: 0.8972809314727783:  63%|██████▎   | 4083/6470 [54:22<29:44,  1.34it/s]
1|4084|Loss: 0.9064981937408447:  63%|██████▎   | 4083/6470 [54:22<29:44,  1.34it/s]
1|4084|Loss: 0.9064981937408447:  63%|██████▎   | 4084/6470 [54:23<30:44,  1.29it/s]
1|4085|Loss: 0.9433522820472717:  63%|██████▎   | 4084/6470 [54:23<30:44,  1.29it/s]
1|4085|Loss: 0.9433522820472717:  63%|██████▎   | 4085/6470 [54:24<31:48,  1.25it/s]
1|4086|Loss: 1.0112019777297974:  63%|██████▎   | 4085/6470 [54:24<31:48,  1.25it/s]
1|4086|Loss: 1.0112019777297974:  63%|██████▎   | 4086/6470 [54:24<30:21,  1.31it/s]
1|4087|Loss: 0.8812939524650574:  63%|██████▎   | 4086/6470 [54:25<30:21,  1.31it/s]
1|4087|Loss: 0.8812939524650574:  63%|██████▎   | 4087/6470 [54:25<31:16,  1.27it/s]
1|4088|Loss: 0.8737617135047913:  63%|██████▎   | 4087/6470 [54:26<31:16,  1.27it/s]
1|4088|Loss: 0.8737617135047913:  63%|██████▎   | 4088/6470 [54:26<32:58,  1.20it/s]
1|4089|Loss: 0.8456591367721558:  63%|██████▎   | 4088/6470 [54:27<32:58,  1.20it/s]
1|4089|Loss: 0.8456591367721558:  63%|██████▎   | 4089/6470 [54:27<33:07,  1.20it/s]
1|4090|Loss: 0.933939516544342:  63%|██████▎   | 4089/6470 [54:27<33:07,  1.20it/s] 
1|4090|Loss: 0.933939516544342:  63%|██████▎   | 4090/6470 [54:27<29:40,  1.34it/s]
1|4091|Loss: 0.8647730946540833:  63%|██████▎   | 4090/6470 [54:28<29:40,  1.34it/s]
1|4091|Loss: 0.8647730946540833:  63%|██████▎   | 4091/6470 [54:28<26:47,  1.48it/s]
1|4092|Loss: 0.9306544065475464:  63%|██████▎   | 4091/6470 [54:28<26:47,  1.48it/s]
1|4092|Loss: 0.9306544065475464:  63%|██████▎   | 4092/6470 [54:29<26:35,  1.49it/s]
1|4093|Loss: 0.8271898627281189:  63%|██████▎   | 4092/6470 [54:29<26:35,  1.49it/s]
1|4093|Loss: 0.8271898627281189:  63%|██████▎   | 4093/6470 [54:29<27:58,  1.42it/s]
1|4094|Loss: 0.9534545540809631:  63%|██████▎   | 4093/6470 [54:30<27:58,  1.42it/s]
1|4094|Loss: 0.9534545540809631:  63%|██████▎   | 4094/6470 [54:30<27:32,  1.44it/s]
1|4095|Loss: 1.0509878396987915:  63%|██████▎   | 4094/6470 [54:30<27:32,  1.44it/s]
1|4095|Loss: 1.0509878396987915:  63%|██████▎   | 4095/6470 [54:31<26:58,  1.47it/s]
1|4096|Loss: 0.8608142733573914:  63%|██████▎   | 4095/6470 [54:31<26:58,  1.47it/s]
1|4096|Loss: 0.8608142733573914:  63%|██████▎   | 4096/6470 [54:31<26:13,  1.51it/s]
1|4097|Loss: 0.949396550655365:  63%|██████▎   | 4096/6470 [54:32<26:13,  1.51it/s] 
1|4097|Loss: 0.949396550655365:  63%|██████▎   | 4097/6470 [54:32<27:10,  1.46it/s]
1|4098|Loss: 1.0314196348190308:  63%|██████▎   | 4097/6470 [54:32<27:10,  1.46it/s]
1|4098|Loss: 1.0314196348190308:  63%|██████▎   | 4098/6470 [54:33<25:23,  1.56it/s]
1|4099|Loss: 0.8647534251213074:  63%|██████▎   | 4098/6470 [54:33<25:23,  1.56it/s]
1|4099|Loss: 0.8647534251213074:  63%|██████▎   | 4099/6470 [54:33<25:25,  1.55it/s]
1|4100|Loss: 1.4241390228271484:  63%|██████▎   | 4099/6470 [54:34<25:25,  1.55it/s]
1|4100|Loss: 1.4241390228271484:  63%|██████▎   | 4100/6470 [54:34<26:51,  1.47it/s]
1|4101|Loss: 0.6295400261878967:  63%|██████▎   | 4100/6470 [54:35<26:51,  1.47it/s]
1|4101|Loss: 0.6295400261878967:  63%|██████▎   | 4101/6470 [54:35<30:44,  1.28it/s]
1|4102|Loss: 0.9703733325004578:  63%|██████▎   | 4101/6470 [54:36<30:44,  1.28it/s]
1|4102|Loss: 0.9703733325004578:  63%|██████▎   | 4102/6470 [54:36<31:41,  1.25it/s]
1|4103|Loss: 0.7591322660446167:  63%|██████▎   | 4102/6470 [54:36<31:41,  1.25it/s]
1|4103|Loss: 0.7591322660446167:  63%|██████▎   | 4103/6470 [54:37<32:53,  1.20it/s]
1|4104|Loss: 0.7843785881996155:  63%|██████▎   | 4103/6470 [54:37<32:53,  1.20it/s]
1|4104|Loss: 0.7843785881996155:  63%|██████▎   | 4104/6470 [54:38<32:51,  1.20it/s]
1|4105|Loss: 1.0101633071899414:  63%|██████▎   | 4104/6470 [54:38<32:51,  1.20it/s]
1|4105|Loss: 1.0101633071899414:  63%|██████▎   | 4105/6470 [54:39<33:56,  1.16it/s]
1|4106|Loss: 0.9514439105987549:  63%|██████▎   | 4105/6470 [54:39<33:56,  1.16it/s]
1|4106|Loss: 0.9514439105987549:  63%|██████▎   | 4106/6470 [54:39<33:50,  1.16it/s]
1|4107|Loss: 0.8383105993270874:  63%|██████▎   | 4106/6470 [54:40<33:50,  1.16it/s]
1|4107|Loss: 0.8383105993270874:  63%|██████▎   | 4107/6470 [54:40<32:04,  1.23it/s]
1|4108|Loss: 1.0279383659362793:  63%|██████▎   | 4107/6470 [54:41<32:04,  1.23it/s]
1|4108|Loss: 1.0279383659362793:  63%|██████▎   | 4108/6470 [54:41<33:06,  1.19it/s]
1|4109|Loss: 0.9484072923660278:  63%|██████▎   | 4108/6470 [54:41<33:06,  1.19it/s]
1|4109|Loss: 0.9484072923660278:  64%|██████▎   | 4109/6470 [54:42<30:41,  1.28it/s]
1|4110|Loss: 1.0141626596450806:  64%|██████▎   | 4109/6470 [54:42<30:41,  1.28it/s]
1|4110|Loss: 1.0141626596450806:  64%|██████▎   | 4110/6470 [54:42<30:12,  1.30it/s]
1|4111|Loss: 0.9915502071380615:  64%|██████▎   | 4110/6470 [54:43<30:12,  1.30it/s]
1|4111|Loss: 0.9915502071380615:  64%|██████▎   | 4111/6470 [54:43<29:34,  1.33it/s]
1|4112|Loss: 1.0946294069290161:  64%|██████▎   | 4111/6470 [54:44<29:34,  1.33it/s]
1|4112|Loss: 1.0946294069290161:  64%|██████▎   | 4112/6470 [54:44<29:38,  1.33it/s]
1|4113|Loss: 0.9729096293449402:  64%|██████▎   | 4112/6470 [54:44<29:38,  1.33it/s]
1|4113|Loss: 0.9729096293449402:  64%|██████▎   | 4113/6470 [54:45<29:39,  1.32it/s]
1|4114|Loss: 0.8993044495582581:  64%|██████▎   | 4113/6470 [54:45<29:39,  1.32it/s]
1|4114|Loss: 0.8993044495582581:  64%|██████▎   | 4114/6470 [54:46<31:18,  1.25it/s]
1|4115|Loss: 0.723014235496521:  64%|██████▎   | 4114/6470 [54:46<31:18,  1.25it/s] 
1|4115|Loss: 0.723014235496521:  64%|██████▎   | 4115/6470 [54:46<31:17,  1.25it/s]
1|4116|Loss: 0.8982208967208862:  64%|██████▎   | 4115/6470 [54:47<31:17,  1.25it/s]
1|4116|Loss: 0.8982208967208862:  64%|██████▎   | 4116/6470 [54:47<30:19,  1.29it/s]
1|4117|Loss: 0.9119784235954285:  64%|██████▎   | 4116/6470 [54:47<30:19,  1.29it/s]
1|4117|Loss: 0.9119784235954285:  64%|██████▎   | 4117/6470 [54:48<29:35,  1.33it/s]
1|4118|Loss: 0.9588087201118469:  64%|██████▎   | 4117/6470 [54:48<29:35,  1.33it/s]
1|4118|Loss: 0.9588087201118469:  64%|██████▎   | 4118/6470 [54:49<29:49,  1.31it/s]
1|4119|Loss: 0.7981073260307312:  64%|██████▎   | 4118/6470 [54:49<29:49,  1.31it/s]
1|4119|Loss: 0.7981073260307312:  64%|██████▎   | 4119/6470 [54:49<29:14,  1.34it/s]
1|4120|Loss: 0.8097246289253235:  64%|██████▎   | 4119/6470 [54:50<29:14,  1.34it/s]
1|4120|Loss: 0.8097246289253235:  64%|██████▎   | 4120/6470 [54:50<31:00,  1.26it/s]
1|4121|Loss: 0.8300462961196899:  64%|██████▎   | 4120/6470 [54:51<31:00,  1.26it/s]
1|4121|Loss: 0.8300462961196899:  64%|██████▎   | 4121/6470 [54:51<32:32,  1.20it/s]
1|4122|Loss: 0.8328060507774353:  64%|██████▎   | 4121/6470 [54:52<32:32,  1.20it/s]
1|4122|Loss: 0.8328060507774353:  64%|██████▎   | 4122/6470 [54:52<30:05,  1.30it/s]
1|4123|Loss: 0.9163579940795898:  64%|██████▎   | 4122/6470 [54:52<30:05,  1.30it/s]
1|4123|Loss: 0.9163579940795898:  64%|██████▎   | 4123/6470 [54:53<30:36,  1.28it/s]
1|4124|Loss: 0.9551882147789001:  64%|██████▎   | 4123/6470 [54:53<30:36,  1.28it/s]
1|4124|Loss: 0.9551882147789001:  64%|██████▎   | 4124/6470 [54:53<32:23,  1.21it/s]
1|4125|Loss: 0.8527073264122009:  64%|██████▎   | 4124/6470 [54:54<32:23,  1.21it/s]
1|4125|Loss: 0.8527073264122009:  64%|██████▍   | 4125/6470 [54:54<32:09,  1.22it/s]
1|4126|Loss: 0.9710876941680908:  64%|██████▍   | 4125/6470 [54:55<32:09,  1.22it/s]
1|4126|Loss: 0.9710876941680908:  64%|██████▍   | 4126/6470 [54:55<30:30,  1.28it/s]
1|4127|Loss: 0.9446731805801392:  64%|██████▍   | 4126/6470 [54:55<30:30,  1.28it/s]
1|4127|Loss: 0.9446731805801392:  64%|██████▍   | 4127/6470 [54:56<30:15,  1.29it/s]
1|4128|Loss: 0.8882535099983215:  64%|██████▍   | 4127/6470 [54:56<30:15,  1.29it/s]
1|4128|Loss: 0.8882535099983215:  64%|██████▍   | 4128/6470 [54:57<30:41,  1.27it/s]
1|4129|Loss: 0.93006831407547:  64%|██████▍   | 4128/6470 [54:57<30:41,  1.27it/s]  
1|4129|Loss: 0.93006831407547:  64%|██████▍   | 4129/6470 [54:57<31:02,  1.26it/s]
1|4130|Loss: 0.8278741240501404:  64%|██████▍   | 4129/6470 [54:58<31:02,  1.26it/s]
1|4130|Loss: 0.8278741240501404:  64%|██████▍   | 4130/6470 [54:58<28:38,  1.36it/s]
1|4131|Loss: 0.9061394929885864:  64%|██████▍   | 4130/6470 [54:58<28:38,  1.36it/s]
1|4131|Loss: 0.9061394929885864:  64%|██████▍   | 4131/6470 [54:59<27:39,  1.41it/s]
1|4132|Loss: 0.9939230680465698:  64%|██████▍   | 4131/6470 [54:59<27:39,  1.41it/s]
1|4132|Loss: 0.9939230680465698:  64%|██████▍   | 4132/6470 [54:59<27:45,  1.40it/s]
1|4133|Loss: 0.8938347101211548:  64%|██████▍   | 4132/6470 [55:00<27:45,  1.40it/s]
1|4133|Loss: 0.8938347101211548:  64%|██████▍   | 4133/6470 [55:00<29:45,  1.31it/s]
1|4134|Loss: 0.8204975724220276:  64%|██████▍   | 4133/6470 [55:01<29:45,  1.31it/s]
1|4134|Loss: 0.8204975724220276:  64%|██████▍   | 4134/6470 [55:01<28:36,  1.36it/s]
1|4135|Loss: 0.8256805539131165:  64%|██████▍   | 4134/6470 [55:01<28:36,  1.36it/s]
1|4135|Loss: 0.8256805539131165:  64%|██████▍   | 4135/6470 [55:02<28:44,  1.35it/s]
1|4136|Loss: 0.8587620854377747:  64%|██████▍   | 4135/6470 [55:02<28:44,  1.35it/s]
1|4136|Loss: 0.8587620854377747:  64%|██████▍   | 4136/6470 [55:03<30:50,  1.26it/s]
1|4137|Loss: 1.0323594808578491:  64%|██████▍   | 4136/6470 [55:03<30:50,  1.26it/s]
1|4137|Loss: 1.0323594808578491:  64%|██████▍   | 4137/6470 [55:04<33:00,  1.18it/s]
1|4138|Loss: 0.875499963760376:  64%|██████▍   | 4137/6470 [55:04<33:00,  1.18it/s] 
1|4138|Loss: 0.875499963760376:  64%|██████▍   | 4138/6470 [55:04<31:46,  1.22it/s]
1|4139|Loss: 0.8451617956161499:  64%|██████▍   | 4138/6470 [55:05<31:46,  1.22it/s]
1|4139|Loss: 0.8451617956161499:  64%|██████▍   | 4139/6470 [55:05<32:03,  1.21it/s]
1|4140|Loss: 0.7938306331634521:  64%|██████▍   | 4139/6470 [55:06<32:03,  1.21it/s]
1|4140|Loss: 0.7938306331634521:  64%|██████▍   | 4140/6470 [55:06<31:23,  1.24it/s]
1|4141|Loss: 0.8974871635437012:  64%|██████▍   | 4140/6470 [55:06<31:23,  1.24it/s]
1|4141|Loss: 0.8974871635437012:  64%|██████▍   | 4141/6470 [55:07<29:30,  1.32it/s]
1|4142|Loss: 1.0574263334274292:  64%|██████▍   | 4141/6470 [55:07<29:30,  1.32it/s]
1|4142|Loss: 1.0574263334274292:  64%|██████▍   | 4142/6470 [55:07<29:26,  1.32it/s]
1|4143|Loss: 0.7753599882125854:  64%|██████▍   | 4142/6470 [55:08<29:26,  1.32it/s]
1|4143|Loss: 0.7753599882125854:  64%|██████▍   | 4143/6470 [55:08<26:56,  1.44it/s]
1|4144|Loss: 0.8702451586723328:  64%|██████▍   | 4143/6470 [55:08<26:56,  1.44it/s]
1|4144|Loss: 0.8702451586723328:  64%|██████▍   | 4144/6470 [55:09<26:58,  1.44it/s]
1|4145|Loss: 0.8581192493438721:  64%|██████▍   | 4144/6470 [55:09<26:58,  1.44it/s]
1|4145|Loss: 0.8581192493438721:  64%|██████▍   | 4145/6470 [55:09<27:49,  1.39it/s]
1|4146|Loss: 0.8563045859336853:  64%|██████▍   | 4145/6470 [55:10<27:49,  1.39it/s]
1|4146|Loss: 0.8563045859336853:  64%|██████▍   | 4146/6470 [55:10<27:32,  1.41it/s]
1|4147|Loss: 0.8047966361045837:  64%|██████▍   | 4146/6470 [55:10<27:32,  1.41it/s]
1|4147|Loss: 0.8047966361045837:  64%|██████▍   | 4147/6470 [55:11<27:52,  1.39it/s]
1|4148|Loss: 0.9154924750328064:  64%|██████▍   | 4147/6470 [55:11<27:52,  1.39it/s]
1|4148|Loss: 0.9154924750328064:  64%|██████▍   | 4148/6470 [55:11<28:17,  1.37it/s]
1|4149|Loss: 0.8488534688949585:  64%|██████▍   | 4148/6470 [55:12<28:17,  1.37it/s]
1|4149|Loss: 0.8488534688949585:  64%|██████▍   | 4149/6470 [55:12<27:35,  1.40it/s]
1|4150|Loss: 0.977068305015564:  64%|██████▍   | 4149/6470 [55:13<27:35,  1.40it/s] 
1|4150|Loss: 0.977068305015564:  64%|██████▍   | 4150/6470 [55:13<29:38,  1.30it/s]
1|4151|Loss: 0.7746449708938599:  64%|██████▍   | 4150/6470 [55:13<29:38,  1.30it/s]
1|4151|Loss: 0.7746449708938599:  64%|██████▍   | 4151/6470 [55:14<26:55,  1.44it/s]
1|4152|Loss: 0.7888646125793457:  64%|██████▍   | 4151/6470 [55:14<26:55,  1.44it/s]
1|4152|Loss: 0.7888646125793457:  64%|██████▍   | 4152/6470 [55:14<26:13,  1.47it/s]
1|4153|Loss: 0.9934170246124268:  64%|██████▍   | 4152/6470 [55:15<26:13,  1.47it/s]
1|4153|Loss: 0.9934170246124268:  64%|██████▍   | 4153/6470 [55:15<26:37,  1.45it/s]
1|4154|Loss: 0.8122051358222961:  64%|██████▍   | 4153/6470 [55:15<26:37,  1.45it/s]
1|4154|Loss: 0.8122051358222961:  64%|██████▍   | 4154/6470 [55:16<27:49,  1.39it/s]
1|4155|Loss: 0.9161317348480225:  64%|██████▍   | 4154/6470 [55:16<27:49,  1.39it/s]
1|4155|Loss: 0.9161317348480225:  64%|██████▍   | 4155/6470 [55:16<27:38,  1.40it/s]
1|4156|Loss: 0.9593400359153748:  64%|██████▍   | 4155/6470 [55:17<27:38,  1.40it/s]
1|4156|Loss: 0.9593400359153748:  64%|██████▍   | 4156/6470 [55:17<27:04,  1.42it/s]
1|4157|Loss: 0.8930714130401611:  64%|██████▍   | 4156/6470 [55:17<27:04,  1.42it/s]
1|4157|Loss: 0.8930714130401611:  64%|██████▍   | 4157/6470 [55:18<26:42,  1.44it/s]
1|4158|Loss: 0.995752215385437:  64%|██████▍   | 4157/6470 [55:18<26:42,  1.44it/s] 
1|4158|Loss: 0.995752215385437:  64%|██████▍   | 4158/6470 [55:19<27:19,  1.41it/s]
1|4159|Loss: 0.9793825149536133:  64%|██████▍   | 4158/6470 [55:19<27:19,  1.41it/s]
1|4159|Loss: 0.9793825149536133:  64%|██████▍   | 4159/6470 [55:19<25:52,  1.49it/s]
1|4160|Loss: 0.9899545907974243:  64%|██████▍   | 4159/6470 [55:20<25:52,  1.49it/s]
1|4160|Loss: 0.9899545907974243:  64%|██████▍   | 4160/6470 [55:20<27:27,  1.40it/s]
1|4161|Loss: 0.9152376055717468:  64%|██████▍   | 4160/6470 [55:20<27:27,  1.40it/s]
1|4161|Loss: 0.9152376055717468:  64%|██████▍   | 4161/6470 [55:21<27:01,  1.42it/s]
1|4162|Loss: 0.8619571328163147:  64%|██████▍   | 4161/6470 [55:21<27:01,  1.42it/s]
1|4162|Loss: 0.8619571328163147:  64%|██████▍   | 4162/6470 [55:21<24:43,  1.56it/s]
1|4163|Loss: 0.859138011932373:  64%|██████▍   | 4162/6470 [55:21<24:43,  1.56it/s] 
1|4163|Loss: 0.859138011932373:  64%|██████▍   | 4163/6470 [55:22<25:52,  1.49it/s]
1|4164|Loss: 0.8120667338371277:  64%|██████▍   | 4163/6470 [55:22<25:52,  1.49it/s]
1|4164|Loss: 0.8120667338371277:  64%|██████▍   | 4164/6470 [55:23<26:36,  1.44it/s]
1|4165|Loss: 0.8927679657936096:  64%|██████▍   | 4164/6470 [55:23<26:36,  1.44it/s]
1|4165|Loss: 0.8927679657936096:  64%|██████▍   | 4165/6470 [55:23<29:02,  1.32it/s]
1|4166|Loss: 0.8319334983825684:  64%|██████▍   | 4165/6470 [55:24<29:02,  1.32it/s]
1|4166|Loss: 0.8319334983825684:  64%|██████▍   | 4166/6470 [55:25<34:19,  1.12it/s]
1|4167|Loss: 0.9190978407859802:  64%|██████▍   | 4166/6470 [55:25<34:19,  1.12it/s]
1|4167|Loss: 0.9190978407859802:  64%|██████▍   | 4167/6470 [55:26<35:07,  1.09it/s]
1|4168|Loss: 0.8027369379997253:  64%|██████▍   | 4167/6470 [55:26<35:07,  1.09it/s]
1|4168|Loss: 0.8027369379997253:  64%|██████▍   | 4168/6470 [55:26<32:00,  1.20it/s]
1|4169|Loss: 0.9137670993804932:  64%|██████▍   | 4168/6470 [55:27<32:00,  1.20it/s]
1|4169|Loss: 0.9137670993804932:  64%|██████▍   | 4169/6470 [55:27<28:49,  1.33it/s]
1|4170|Loss: 0.8405765891075134:  64%|██████▍   | 4169/6470 [55:27<28:49,  1.33it/s]
1|4170|Loss: 0.8405765891075134:  64%|██████▍   | 4170/6470 [55:28<28:57,  1.32it/s]
1|4171|Loss: 0.7578129768371582:  64%|██████▍   | 4170/6470 [55:28<28:57,  1.32it/s]
1|4171|Loss: 0.7578129768371582:  64%|██████▍   | 4171/6470 [55:28<28:41,  1.34it/s]
1|4172|Loss: 1.0302822589874268:  64%|██████▍   | 4171/6470 [55:29<28:41,  1.34it/s]
1|4172|Loss: 1.0302822589874268:  64%|██████▍   | 4172/6470 [55:29<28:46,  1.33it/s]
1|4173|Loss: 0.819674551486969:  64%|██████▍   | 4172/6470 [55:30<28:46,  1.33it/s] 
1|4173|Loss: 0.819674551486969:  64%|██████▍   | 4173/6470 [55:30<27:59,  1.37it/s]
1|4174|Loss: 0.9127346277236938:  64%|██████▍   | 4173/6470 [55:30<27:59,  1.37it/s]
1|4174|Loss: 0.9127346277236938:  65%|██████▍   | 4174/6470 [55:31<27:32,  1.39it/s]
1|4175|Loss: 0.9294388890266418:  65%|██████▍   | 4174/6470 [55:31<27:32,  1.39it/s]
1|4175|Loss: 0.9294388890266418:  65%|██████▍   | 4175/6470 [55:31<28:18,  1.35it/s]
1|4176|Loss: 0.8671980500221252:  65%|██████▍   | 4175/6470 [55:32<28:18,  1.35it/s]
1|4176|Loss: 0.8671980500221252:  65%|██████▍   | 4176/6470 [55:32<30:00,  1.27it/s]
1|4177|Loss: 0.9533271789550781:  65%|██████▍   | 4176/6470 [55:33<30:00,  1.27it/s]
1|4177|Loss: 0.9533271789550781:  65%|██████▍   | 4177/6470 [55:33<31:00,  1.23it/s]
1|4178|Loss: 0.9161501526832581:  65%|██████▍   | 4177/6470 [55:33<31:00,  1.23it/s]
1|4178|Loss: 0.9161501526832581:  65%|██████▍   | 4178/6470 [55:34<29:37,  1.29it/s]
1|4179|Loss: 0.8441280722618103:  65%|██████▍   | 4178/6470 [55:34<29:37,  1.29it/s]
1|4179|Loss: 0.8441280722618103:  65%|██████▍   | 4179/6470 [55:34<27:50,  1.37it/s]
1|4180|Loss: 0.8545740842819214:  65%|██████▍   | 4179/6470 [55:35<27:50,  1.37it/s]
1|4180|Loss: 0.8545740842819214:  65%|██████▍   | 4180/6470 [55:35<28:53,  1.32it/s]
1|4181|Loss: 0.8381167650222778:  65%|██████▍   | 4180/6470 [55:36<28:53,  1.32it/s]
1|4181|Loss: 0.8381167650222778:  65%|██████▍   | 4181/6470 [55:36<30:26,  1.25it/s]
1|4182|Loss: 0.7177960276603699:  65%|██████▍   | 4181/6470 [55:37<30:26,  1.25it/s]
1|4182|Loss: 0.7177960276603699:  65%|██████▍   | 4182/6470 [55:37<31:54,  1.20it/s]
1|4183|Loss: 0.8478060960769653:  65%|██████▍   | 4182/6470 [55:37<31:54,  1.20it/s]
1|4183|Loss: 0.8478060960769653:  65%|██████▍   | 4183/6470 [55:38<31:00,  1.23it/s]
1|4184|Loss: 0.8275450468063354:  65%|██████▍   | 4183/6470 [55:38<31:00,  1.23it/s]
1|4184|Loss: 0.8275450468063354:  65%|██████▍   | 4184/6470 [55:38<29:44,  1.28it/s]
1|4185|Loss: 0.8339971899986267:  65%|██████▍   | 4184/6470 [55:39<29:44,  1.28it/s]
1|4185|Loss: 0.8339971899986267:  65%|██████▍   | 4185/6470 [55:39<29:37,  1.29it/s]
1|4186|Loss: 1.050081729888916:  65%|██████▍   | 4185/6470 [55:40<29:37,  1.29it/s] 
1|4186|Loss: 1.050081729888916:  65%|██████▍   | 4186/6470 [55:40<28:19,  1.34it/s]
1|4187|Loss: 0.904790997505188:  65%|██████▍   | 4186/6470 [55:40<28:19,  1.34it/s]
1|4187|Loss: 0.904790997505188:  65%|██████▍   | 4187/6470 [55:41<29:49,  1.28it/s]
1|4188|Loss: 0.8937433362007141:  65%|██████▍   | 4187/6470 [55:41<29:49,  1.28it/s]
1|4188|Loss: 0.8937433362007141:  65%|██████▍   | 4188/6470 [55:42<29:13,  1.30it/s]
1|4189|Loss: 0.9415004253387451:  65%|██████▍   | 4188/6470 [55:42<29:13,  1.30it/s]
1|4189|Loss: 0.9415004253387451:  65%|██████▍   | 4189/6470 [55:42<28:24,  1.34it/s]
1|4190|Loss: 0.7914003729820251:  65%|██████▍   | 4189/6470 [55:43<28:24,  1.34it/s]
1|4190|Loss: 0.7914003729820251:  65%|██████▍   | 4190/6470 [55:43<29:41,  1.28it/s]
1|4191|Loss: 0.8332024216651917:  65%|██████▍   | 4190/6470 [55:44<29:41,  1.28it/s]
1|4191|Loss: 0.8332024216651917:  65%|██████▍   | 4191/6470 [55:44<28:55,  1.31it/s]
1|4192|Loss: 0.6913966536521912:  65%|██████▍   | 4191/6470 [55:44<28:55,  1.31it/s]
1|4192|Loss: 0.6913966536521912:  65%|██████▍   | 4192/6470 [55:45<29:03,  1.31it/s]
1|4193|Loss: 0.9358031749725342:  65%|██████▍   | 4192/6470 [55:45<29:03,  1.31it/s]
1|4193|Loss: 0.9358031749725342:  65%|██████▍   | 4193/6470 [55:45<29:04,  1.30it/s]
1|4194|Loss: 0.7464210391044617:  65%|██████▍   | 4193/6470 [55:46<29:04,  1.30it/s]
1|4194|Loss: 0.7464210391044617:  65%|██████▍   | 4194/6470 [55:46<28:21,  1.34it/s]
1|4195|Loss: 0.7882478833198547:  65%|██████▍   | 4194/6470 [55:46<28:21,  1.34it/s]
1|4195|Loss: 0.7882478833198547:  65%|██████▍   | 4195/6470 [55:47<28:30,  1.33it/s]
1|4196|Loss: 0.7222736477851868:  65%|██████▍   | 4195/6470 [55:47<28:30,  1.33it/s]
1|4196|Loss: 0.7222736477851868:  65%|██████▍   | 4196/6470 [55:47<27:10,  1.39it/s]
1|4197|Loss: 0.9546265602111816:  65%|██████▍   | 4196/6470 [55:48<27:10,  1.39it/s]
1|4197|Loss: 0.9546265602111816:  65%|██████▍   | 4197/6470 [55:48<27:47,  1.36it/s]
1|4198|Loss: 1.015395998954773:  65%|██████▍   | 4197/6470 [55:49<27:47,  1.36it/s] 
1|4198|Loss: 1.015395998954773:  65%|██████▍   | 4198/6470 [55:49<30:11,  1.25it/s]
1|4199|Loss: 0.9907517433166504:  65%|██████▍   | 4198/6470 [55:50<30:11,  1.25it/s]
1|4199|Loss: 0.9907517433166504:  65%|██████▍   | 4199/6470 [55:50<29:24,  1.29it/s]
1|4200|Loss: 0.9983502626419067:  65%|██████▍   | 4199/6470 [55:50<29:24,  1.29it/s]
1|4200|Loss: 0.9983502626419067:  65%|██████▍   | 4200/6470 [55:50<27:22,  1.38it/s]
1|4201|Loss: 0.9281461834907532:  65%|██████▍   | 4200/6470 [55:51<27:22,  1.38it/s]
1|4201|Loss: 0.9281461834907532:  65%|██████▍   | 4201/6470 [55:51<27:49,  1.36it/s]
1|4202|Loss: 1.1527068614959717:  65%|██████▍   | 4201/6470 [55:52<27:49,  1.36it/s]
1|4202|Loss: 1.1527068614959717:  65%|██████▍   | 4202/6470 [55:52<28:21,  1.33it/s]
1|4203|Loss: 0.8950429558753967:  65%|██████▍   | 4202/6470 [55:53<28:21,  1.33it/s]
1|4203|Loss: 0.8950429558753967:  65%|██████▍   | 4203/6470 [55:53<29:52,  1.26it/s]
1|4204|Loss: 0.8486813306808472:  65%|██████▍   | 4203/6470 [55:53<29:52,  1.26it/s]
1|4204|Loss: 0.8486813306808472:  65%|██████▍   | 4204/6470 [55:54<30:54,  1.22it/s]
1|4205|Loss: 0.7819591760635376:  65%|██████▍   | 4204/6470 [55:54<30:54,  1.22it/s]
1|4205|Loss: 0.7819591760635376:  65%|██████▍   | 4205/6470 [55:54<28:44,  1.31it/s]
1|4206|Loss: 0.7674899697303772:  65%|██████▍   | 4205/6470 [55:55<28:44,  1.31it/s]
1|4206|Loss: 0.7674899697303772:  65%|██████▌   | 4206/6470 [55:55<28:34,  1.32it/s]
1|4207|Loss: 0.7475408315658569:  65%|██████▌   | 4206/6470 [55:56<28:34,  1.32it/s]
1|4207|Loss: 0.7475408315658569:  65%|██████▌   | 4207/6470 [55:56<29:03,  1.30it/s]
1|4208|Loss: 0.9159073829650879:  65%|██████▌   | 4207/6470 [55:56<29:03,  1.30it/s]
1|4208|Loss: 0.9159073829650879:  65%|██████▌   | 4208/6470 [55:57<29:17,  1.29it/s]
1|4209|Loss: 0.8538767099380493:  65%|██████▌   | 4208/6470 [55:57<29:17,  1.29it/s]
1|4209|Loss: 0.8538767099380493:  65%|██████▌   | 4209/6470 [55:58<30:03,  1.25it/s]
1|4210|Loss: 0.8547419905662537:  65%|██████▌   | 4209/6470 [55:58<30:03,  1.25it/s]
1|4210|Loss: 0.8547419905662537:  65%|██████▌   | 4210/6470 [55:58<28:38,  1.31it/s]
1|4211|Loss: 0.768339991569519:  65%|██████▌   | 4210/6470 [55:59<28:38,  1.31it/s] 
1|4211|Loss: 0.768339991569519:  65%|██████▌   | 4211/6470 [55:59<27:10,  1.39it/s]
1|4212|Loss: 0.8023581504821777:  65%|██████▌   | 4211/6470 [55:59<27:10,  1.39it/s]
1|4212|Loss: 0.8023581504821777:  65%|██████▌   | 4212/6470 [56:00<28:20,  1.33it/s]
1|4213|Loss: 0.9888793230056763:  65%|██████▌   | 4212/6470 [56:00<28:20,  1.33it/s]
1|4213|Loss: 0.9888793230056763:  65%|██████▌   | 4213/6470 [56:01<29:05,  1.29it/s]
1|4214|Loss: 0.8122199773788452:  65%|██████▌   | 4213/6470 [56:01<29:05,  1.29it/s]
1|4214|Loss: 0.8122199773788452:  65%|██████▌   | 4214/6470 [56:01<30:34,  1.23it/s]
1|4215|Loss: 0.7328938245773315:  65%|██████▌   | 4214/6470 [56:02<30:34,  1.23it/s]
1|4215|Loss: 0.7328938245773315:  65%|██████▌   | 4215/6470 [56:02<28:21,  1.33it/s]
1|4216|Loss: 0.9412990212440491:  65%|██████▌   | 4215/6470 [56:03<28:21,  1.33it/s]
1|4216|Loss: 0.9412990212440491:  65%|██████▌   | 4216/6470 [56:03<29:15,  1.28it/s]
1|4217|Loss: 0.7284414768218994:  65%|██████▌   | 4216/6470 [56:03<29:15,  1.28it/s]
1|4217|Loss: 0.7284414768218994:  65%|██████▌   | 4217/6470 [56:04<29:21,  1.28it/s]
1|4218|Loss: 0.8534641265869141:  65%|██████▌   | 4217/6470 [56:04<29:21,  1.28it/s]
1|4218|Loss: 0.8534641265869141:  65%|██████▌   | 4218/6470 [56:05<29:30,  1.27it/s]
1|4219|Loss: 0.8465324640274048:  65%|██████▌   | 4218/6470 [56:05<29:30,  1.27it/s]
1|4219|Loss: 0.8465324640274048:  65%|██████▌   | 4219/6470 [56:05<25:32,  1.47it/s]
1|4220|Loss: 0.9489918947219849:  65%|██████▌   | 4219/6470 [56:05<25:32,  1.47it/s]
1|4220|Loss: 0.9489918947219849:  65%|██████▌   | 4220/6470 [56:06<25:30,  1.47it/s]
1|4221|Loss: 0.9090567231178284:  65%|██████▌   | 4220/6470 [56:06<25:30,  1.47it/s]
1|4221|Loss: 0.9090567231178284:  65%|██████▌   | 4221/6470 [56:06<27:28,  1.36it/s]
1|4222|Loss: 0.9674601554870605:  65%|██████▌   | 4221/6470 [56:07<27:28,  1.36it/s]
1|4222|Loss: 0.9674601554870605:  65%|██████▌   | 4222/6470 [56:07<29:28,  1.27it/s]
1|4223|Loss: 0.7981104850769043:  65%|██████▌   | 4222/6470 [56:08<29:28,  1.27it/s]
1|4223|Loss: 0.7981104850769043:  65%|██████▌   | 4223/6470 [56:08<29:11,  1.28it/s]
1|4224|Loss: 0.8156945705413818:  65%|██████▌   | 4223/6470 [56:09<29:11,  1.28it/s]
1|4224|Loss: 0.8156945705413818:  65%|██████▌   | 4224/6470 [56:09<29:03,  1.29it/s]
1|4225|Loss: 1.0317412614822388:  65%|██████▌   | 4224/6470 [56:09<29:03,  1.29it/s]
1|4225|Loss: 1.0317412614822388:  65%|██████▌   | 4225/6470 [56:10<26:59,  1.39it/s]
1|4226|Loss: 0.9421049356460571:  65%|██████▌   | 4225/6470 [56:10<26:59,  1.39it/s]
1|4226|Loss: 0.9421049356460571:  65%|██████▌   | 4226/6470 [56:10<25:38,  1.46it/s]
1|4227|Loss: 1.0042403936386108:  65%|██████▌   | 4226/6470 [56:11<25:38,  1.46it/s]
1|4227|Loss: 1.0042403936386108:  65%|██████▌   | 4227/6470 [56:11<26:14,  1.42it/s]
1|4228|Loss: 0.9099816679954529:  65%|██████▌   | 4227/6470 [56:11<26:14,  1.42it/s]
1|4228|Loss: 0.9099816679954529:  65%|██████▌   | 4228/6470 [56:12<26:03,  1.43it/s]
1|4229|Loss: 0.9874812960624695:  65%|██████▌   | 4228/6470 [56:12<26:03,  1.43it/s]
1|4229|Loss: 0.9874812960624695:  65%|██████▌   | 4229/6470 [56:12<28:33,  1.31it/s]
1|4230|Loss: 0.8487681150436401:  65%|██████▌   | 4229/6470 [56:13<28:33,  1.31it/s]
1|4230|Loss: 0.8487681150436401:  65%|██████▌   | 4230/6470 [56:13<29:05,  1.28it/s]
1|4231|Loss: 0.8675824403762817:  65%|██████▌   | 4230/6470 [56:14<29:05,  1.28it/s]
1|4231|Loss: 0.8675824403762817:  65%|██████▌   | 4231/6470 [56:14<29:53,  1.25it/s]
1|4232|Loss: 0.9390131235122681:  65%|██████▌   | 4231/6470 [56:15<29:53,  1.25it/s]
1|4232|Loss: 0.9390131235122681:  65%|██████▌   | 4232/6470 [56:15<29:38,  1.26it/s]
1|4233|Loss: 0.8004859685897827:  65%|██████▌   | 4232/6470 [56:15<29:38,  1.26it/s]
1|4233|Loss: 0.8004859685897827:  65%|██████▌   | 4233/6470 [56:16<28:29,  1.31it/s]
1|4234|Loss: 1.0044076442718506:  65%|██████▌   | 4233/6470 [56:16<28:29,  1.31it/s]
1|4234|Loss: 1.0044076442718506:  65%|██████▌   | 4234/6470 [56:16<28:39,  1.30it/s]
1|4235|Loss: 0.8045478463172913:  65%|██████▌   | 4234/6470 [56:17<28:39,  1.30it/s]
1|4235|Loss: 0.8045478463172913:  65%|██████▌   | 4235/6470 [56:17<28:34,  1.30it/s]
1|4236|Loss: 0.9039275050163269:  65%|██████▌   | 4235/6470 [56:18<28:34,  1.30it/s]
1|4236|Loss: 0.9039275050163269:  65%|██████▌   | 4236/6470 [56:18<27:39,  1.35it/s]
1|4237|Loss: 1.0880080461502075:  65%|██████▌   | 4236/6470 [56:18<27:39,  1.35it/s]
1|4237|Loss: 1.0880080461502075:  65%|██████▌   | 4237/6470 [56:18<26:25,  1.41it/s]
1|4238|Loss: 0.7330460548400879:  65%|██████▌   | 4237/6470 [56:19<26:25,  1.41it/s]
1|4238|Loss: 0.7330460548400879:  66%|██████▌   | 4238/6470 [56:19<24:43,  1.50it/s]
1|4239|Loss: 0.8198003172874451:  66%|██████▌   | 4238/6470 [56:19<24:43,  1.50it/s]
1|4239|Loss: 0.8198003172874451:  66%|██████▌   | 4239/6470 [56:20<24:31,  1.52it/s]
1|4240|Loss: 0.8893336057662964:  66%|██████▌   | 4239/6470 [56:20<24:31,  1.52it/s]
1|4240|Loss: 0.8893336057662964:  66%|██████▌   | 4240/6470 [56:20<22:08,  1.68it/s]
1|4241|Loss: 0.9553963541984558:  66%|██████▌   | 4240/6470 [56:20<22:08,  1.68it/s]
1|4241|Loss: 0.9553963541984558:  66%|██████▌   | 4241/6470 [56:21<23:31,  1.58it/s]
1|4242|Loss: 0.8939312100410461:  66%|██████▌   | 4241/6470 [56:21<23:31,  1.58it/s]
1|4242|Loss: 0.8939312100410461:  66%|██████▌   | 4242/6470 [56:22<26:34,  1.40it/s]
1|4243|Loss: 0.7637295126914978:  66%|██████▌   | 4242/6470 [56:22<26:34,  1.40it/s]
1|4243|Loss: 0.7637295126914978:  66%|██████▌   | 4243/6470 [56:23<29:10,  1.27it/s]
1|4244|Loss: 0.8201360106468201:  66%|██████▌   | 4243/6470 [56:23<29:10,  1.27it/s]
1|4244|Loss: 0.8201360106468201:  66%|██████▌   | 4244/6470 [56:23<27:04,  1.37it/s]
1|4245|Loss: 0.824737012386322:  66%|██████▌   | 4244/6470 [56:24<27:04,  1.37it/s] 
1|4245|Loss: 0.824737012386322:  66%|██████▌   | 4245/6470 [56:24<28:26,  1.30it/s]
1|4246|Loss: 0.924019455909729:  66%|██████▌   | 4245/6470 [56:25<28:26,  1.30it/s]
1|4246|Loss: 0.924019455909729:  66%|██████▌   | 4246/6470 [56:25<30:29,  1.22it/s]
1|4247|Loss: 0.9977735877037048:  66%|██████▌   | 4246/6470 [56:26<30:29,  1.22it/s]
1|4247|Loss: 0.9977735877037048:  66%|██████▌   | 4247/6470 [56:26<30:32,  1.21it/s]
1|4248|Loss: 0.8947252035140991:  66%|██████▌   | 4247/6470 [56:26<30:32,  1.21it/s]
1|4248|Loss: 0.8947252035140991:  66%|██████▌   | 4248/6470 [56:27<29:24,  1.26it/s]
1|4249|Loss: 0.827734112739563:  66%|██████▌   | 4248/6470 [56:27<29:24,  1.26it/s] 
1|4249|Loss: 0.827734112739563:  66%|██████▌   | 4249/6470 [56:27<28:21,  1.31it/s]
1|4250|Loss: 1.0335222482681274:  66%|██████▌   | 4249/6470 [56:28<28:21,  1.31it/s]
1|4250|Loss: 1.0335222482681274:  66%|██████▌   | 4250/6470 [56:28<27:02,  1.37it/s]
1|4251|Loss: 0.785524845123291:  66%|██████▌   | 4250/6470 [56:28<27:02,  1.37it/s] 
1|4251|Loss: 0.785524845123291:  66%|██████▌   | 4251/6470 [56:29<26:23,  1.40it/s]
1|4252|Loss: 0.9976775050163269:  66%|██████▌   | 4251/6470 [56:29<26:23,  1.40it/s]
1|4252|Loss: 0.9976775050163269:  66%|██████▌   | 4252/6470 [56:29<26:43,  1.38it/s]
1|4253|Loss: 1.0138431787490845:  66%|██████▌   | 4252/6470 [56:30<26:43,  1.38it/s]
1|4253|Loss: 1.0138431787490845:  66%|██████▌   | 4253/6470 [56:30<27:39,  1.34it/s]
1|4254|Loss: 0.9128733277320862:  66%|██████▌   | 4253/6470 [56:31<27:39,  1.34it/s]
1|4254|Loss: 0.9128733277320862:  66%|██████▌   | 4254/6470 [56:31<27:19,  1.35it/s]
1|4255|Loss: 0.904364287853241:  66%|██████▌   | 4254/6470 [56:31<27:19,  1.35it/s] 
1|4255|Loss: 0.904364287853241:  66%|██████▌   | 4255/6470 [56:32<28:52,  1.28it/s]
1|4256|Loss: 1.1153672933578491:  66%|██████▌   | 4255/6470 [56:32<28:52,  1.28it/s]
1|4256|Loss: 1.1153672933578491:  66%|██████▌   | 4256/6470 [56:33<28:30,  1.29it/s]
1|4257|Loss: 1.2656503915786743:  66%|██████▌   | 4256/6470 [56:33<28:30,  1.29it/s]
1|4257|Loss: 1.2656503915786743:  66%|██████▌   | 4257/6470 [56:33<28:48,  1.28it/s]
1|4258|Loss: 0.8104686141014099:  66%|██████▌   | 4257/6470 [56:34<28:48,  1.28it/s]
1|4258|Loss: 0.8104686141014099:  66%|██████▌   | 4258/6470 [56:34<28:01,  1.32it/s]
1|4259|Loss: 0.9391173124313354:  66%|██████▌   | 4258/6470 [56:34<28:01,  1.32it/s]
1|4259|Loss: 0.9391173124313354:  66%|██████▌   | 4259/6470 [56:35<27:51,  1.32it/s]
1|4260|Loss: 0.9437094330787659:  66%|██████▌   | 4259/6470 [56:35<27:51,  1.32it/s]
1|4260|Loss: 0.9437094330787659:  66%|██████▌   | 4260/6470 [56:36<28:21,  1.30it/s]
1|4261|Loss: 0.939415693283081:  66%|██████▌   | 4260/6470 [56:36<28:21,  1.30it/s] 
1|4261|Loss: 0.939415693283081:  66%|██████▌   | 4261/6470 [56:36<27:57,  1.32it/s]
1|4262|Loss: 0.9782179594039917:  66%|██████▌   | 4261/6470 [56:37<27:57,  1.32it/s]
1|4262|Loss: 0.9782179594039917:  66%|██████▌   | 4262/6470 [56:37<28:17,  1.30it/s]
1|4263|Loss: 0.9182416200637817:  66%|██████▌   | 4262/6470 [56:38<28:17,  1.30it/s]
1|4263|Loss: 0.9182416200637817:  66%|██████▌   | 4263/6470 [56:38<26:28,  1.39it/s]
1|4264|Loss: 0.9599043726921082:  66%|██████▌   | 4263/6470 [56:38<26:28,  1.39it/s]
1|4264|Loss: 0.9599043726921082:  66%|██████▌   | 4264/6470 [56:38<25:11,  1.46it/s]
1|4265|Loss: 0.8408383131027222:  66%|██████▌   | 4264/6470 [56:39<25:11,  1.46it/s]
1|4265|Loss: 0.8408383131027222:  66%|██████▌   | 4265/6470 [56:39<27:36,  1.33it/s]
1|4266|Loss: 0.9202356338500977:  66%|██████▌   | 4265/6470 [56:40<27:36,  1.33it/s]
1|4266|Loss: 0.9202356338500977:  66%|██████▌   | 4266/6470 [56:40<26:26,  1.39it/s]
1|4267|Loss: 0.9446585774421692:  66%|██████▌   | 4266/6470 [56:40<26:26,  1.39it/s]
1|4267|Loss: 0.9446585774421692:  66%|██████▌   | 4267/6470 [56:41<27:24,  1.34it/s]
1|4268|Loss: 0.8863527178764343:  66%|██████▌   | 4267/6470 [56:41<27:24,  1.34it/s]
1|4268|Loss: 0.8863527178764343:  66%|██████▌   | 4268/6470 [56:42<27:49,  1.32it/s]
1|4269|Loss: 0.8784053921699524:  66%|██████▌   | 4268/6470 [56:42<27:49,  1.32it/s]
1|4269|Loss: 0.8784053921699524:  66%|██████▌   | 4269/6470 [56:42<28:27,  1.29it/s]
1|4270|Loss: 0.7231305241584778:  66%|██████▌   | 4269/6470 [56:43<28:27,  1.29it/s]
1|4270|Loss: 0.7231305241584778:  66%|██████▌   | 4270/6470 [56:43<28:00,  1.31it/s]
1|4271|Loss: 0.7565957903862:  66%|██████▌   | 4270/6470 [56:43<28:00,  1.31it/s]   
1|4271|Loss: 0.7565957903862:  66%|██████▌   | 4271/6470 [56:44<26:37,  1.38it/s]
1|4272|Loss: 0.9362698793411255:  66%|██████▌   | 4271/6470 [56:44<26:37,  1.38it/s]
1|4272|Loss: 0.9362698793411255:  66%|██████▌   | 4272/6470 [56:44<26:57,  1.36it/s]
1|4273|Loss: 0.8343586921691895:  66%|██████▌   | 4272/6470 [56:45<26:57,  1.36it/s]
1|4273|Loss: 0.8343586921691895:  66%|██████▌   | 4273/6470 [56:45<28:56,  1.27it/s]
1|4274|Loss: 0.9809892773628235:  66%|██████▌   | 4273/6470 [56:46<28:56,  1.27it/s]
1|4274|Loss: 0.9809892773628235:  66%|██████▌   | 4274/6470 [56:46<29:28,  1.24it/s]
1|4275|Loss: 0.8575882315635681:  66%|██████▌   | 4274/6470 [56:47<29:28,  1.24it/s]
1|4275|Loss: 0.8575882315635681:  66%|██████▌   | 4275/6470 [56:47<29:09,  1.25it/s]
1|4276|Loss: 0.9675351977348328:  66%|██████▌   | 4275/6470 [56:47<29:09,  1.25it/s]
1|4276|Loss: 0.9675351977348328:  66%|██████▌   | 4276/6470 [56:48<26:47,  1.36it/s]
1|4277|Loss: 0.8405864238739014:  66%|██████▌   | 4276/6470 [56:48<26:47,  1.36it/s]
1|4277|Loss: 0.8405864238739014:  66%|██████▌   | 4277/6470 [56:48<26:06,  1.40it/s]
1|4278|Loss: 0.8308351039886475:  66%|██████▌   | 4277/6470 [56:49<26:06,  1.40it/s]
1|4278|Loss: 0.8308351039886475:  66%|██████▌   | 4278/6470 [56:49<27:52,  1.31it/s]
1|4279|Loss: 0.9584197998046875:  66%|██████▌   | 4278/6470 [56:50<27:52,  1.31it/s]
1|4279|Loss: 0.9584197998046875:  66%|██████▌   | 4279/6470 [56:50<28:24,  1.29it/s]
1|4280|Loss: 0.7426838874816895:  66%|██████▌   | 4279/6470 [56:50<28:24,  1.29it/s]
1|4280|Loss: 0.7426838874816895:  66%|██████▌   | 4280/6470 [56:51<28:35,  1.28it/s]
1|4281|Loss: 0.7980133295059204:  66%|██████▌   | 4280/6470 [56:51<28:35,  1.28it/s]
1|4281|Loss: 0.7980133295059204:  66%|██████▌   | 4281/6470 [56:51<26:43,  1.37it/s]
1|4282|Loss: 0.8462158441543579:  66%|██████▌   | 4281/6470 [56:52<26:43,  1.37it/s]
1|4282|Loss: 0.8462158441543579:  66%|██████▌   | 4282/6470 [56:52<26:30,  1.38it/s]
1|4283|Loss: 0.8037396669387817:  66%|██████▌   | 4282/6470 [56:53<26:30,  1.38it/s]
1|4283|Loss: 0.8037396669387817:  66%|██████▌   | 4283/6470 [56:53<26:56,  1.35it/s]
1|4284|Loss: 0.8526927828788757:  66%|██████▌   | 4283/6470 [56:53<26:56,  1.35it/s]
1|4284|Loss: 0.8526927828788757:  66%|██████▌   | 4284/6470 [56:54<28:07,  1.30it/s]
1|4285|Loss: 0.759401798248291:  66%|██████▌   | 4284/6470 [56:54<28:07,  1.30it/s] 
1|4285|Loss: 0.759401798248291:  66%|██████▌   | 4285/6470 [56:55<29:33,  1.23it/s]
1|4286|Loss: 0.8639055490493774:  66%|██████▌   | 4285/6470 [56:55<29:33,  1.23it/s]
1|4286|Loss: 0.8639055490493774:  66%|██████▌   | 4286/6470 [56:55<27:48,  1.31it/s]
1|4287|Loss: 1.0375651121139526:  66%|██████▌   | 4286/6470 [56:56<27:48,  1.31it/s]
1|4287|Loss: 1.0375651121139526:  66%|██████▋   | 4287/6470 [56:56<27:45,  1.31it/s]
1|4288|Loss: 0.927003800868988:  66%|██████▋   | 4287/6470 [56:56<27:45,  1.31it/s] 
1|4288|Loss: 0.927003800868988:  66%|██████▋   | 4288/6470 [56:57<27:01,  1.35it/s]
1|4289|Loss: 1.0423338413238525:  66%|██████▋   | 4288/6470 [56:57<27:01,  1.35it/s]
1|4289|Loss: 1.0423338413238525:  66%|██████▋   | 4289/6470 [56:57<23:58,  1.52it/s]
1|4290|Loss: 0.9238585233688354:  66%|██████▋   | 4289/6470 [56:58<23:58,  1.52it/s]
1|4290|Loss: 0.9238585233688354:  66%|██████▋   | 4290/6470 [56:58<24:25,  1.49it/s]
1|4291|Loss: 0.799858033657074:  66%|██████▋   | 4290/6470 [56:58<24:25,  1.49it/s] 
1|4291|Loss: 0.799858033657074:  66%|██████▋   | 4291/6470 [56:59<25:24,  1.43it/s]
1|4292|Loss: 1.0041677951812744:  66%|██████▋   | 4291/6470 [56:59<25:24,  1.43it/s]
1|4292|Loss: 1.0041677951812744:  66%|██████▋   | 4292/6470 [56:59<25:14,  1.44it/s]
1|4293|Loss: 1.0047986507415771:  66%|██████▋   | 4292/6470 [57:00<25:14,  1.44it/s]
1|4293|Loss: 1.0047986507415771:  66%|██████▋   | 4293/6470 [57:00<24:58,  1.45it/s]
1|4294|Loss: 0.8862112164497375:  66%|██████▋   | 4293/6470 [57:00<24:58,  1.45it/s]
1|4294|Loss: 0.8862112164497375:  66%|██████▋   | 4294/6470 [57:01<25:00,  1.45it/s]
1|4295|Loss: 0.9121088981628418:  66%|██████▋   | 4294/6470 [57:01<25:00,  1.45it/s]
1|4295|Loss: 0.9121088981628418:  66%|██████▋   | 4295/6470 [57:01<25:12,  1.44it/s]
1|4296|Loss: 0.8822507262229919:  66%|██████▋   | 4295/6470 [57:02<25:12,  1.44it/s]
1|4296|Loss: 0.8822507262229919:  66%|██████▋   | 4296/6470 [57:02<23:50,  1.52it/s]
1|4297|Loss: 1.052771806716919:  66%|██████▋   | 4296/6470 [57:02<23:50,  1.52it/s] 
1|4297|Loss: 1.052771806716919:  66%|██████▋   | 4297/6470 [57:03<25:57,  1.40it/s]
1|4298|Loss: 0.9543188214302063:  66%|██████▋   | 4297/6470 [57:03<25:57,  1.40it/s]
1|4298|Loss: 0.9543188214302063:  66%|██████▋   | 4298/6470 [57:04<25:42,  1.41it/s]
1|4299|Loss: 0.8263126611709595:  66%|██████▋   | 4298/6470 [57:04<25:42,  1.41it/s]
1|4299|Loss: 0.8263126611709595:  66%|██████▋   | 4299/6470 [57:04<23:31,  1.54it/s]
1|4300|Loss: 0.9654718041419983:  66%|██████▋   | 4299/6470 [57:04<23:31,  1.54it/s]
1|4300|Loss: 0.9654718041419983:  66%|██████▋   | 4300/6470 [57:05<25:35,  1.41it/s]
1|4301|Loss: 0.7297854423522949:  66%|██████▋   | 4300/6470 [57:05<25:35,  1.41it/s]
1|4301|Loss: 0.7297854423522949:  66%|██████▋   | 4301/6470 [57:06<27:59,  1.29it/s]
1|4302|Loss: 1.0039016008377075:  66%|██████▋   | 4301/6470 [57:06<27:59,  1.29it/s]
1|4302|Loss: 1.0039016008377075:  66%|██████▋   | 4302/6470 [57:07<28:07,  1.28it/s]
1|4303|Loss: 1.087876796722412:  66%|██████▋   | 4302/6470 [57:07<28:07,  1.28it/s] 
1|4303|Loss: 1.087876796722412:  67%|██████▋   | 4303/6470 [57:07<27:51,  1.30it/s]
1|4304|Loss: 0.7858574986457825:  67%|██████▋   | 4303/6470 [57:08<27:51,  1.30it/s]
1|4304|Loss: 0.7858574986457825:  67%|██████▋   | 4304/6470 [57:08<28:55,  1.25it/s]
1|4305|Loss: 0.902279794216156:  67%|██████▋   | 4304/6470 [57:09<28:55,  1.25it/s] 
1|4305|Loss: 0.902279794216156:  67%|██████▋   | 4305/6470 [57:09<28:59,  1.24it/s]
1|4306|Loss: 0.9146353602409363:  67%|██████▋   | 4305/6470 [57:09<28:59,  1.24it/s]
1|4306|Loss: 0.9146353602409363:  67%|██████▋   | 4306/6470 [57:10<28:33,  1.26it/s]
1|4307|Loss: 0.8798062801361084:  67%|██████▋   | 4306/6470 [57:10<28:33,  1.26it/s]
1|4307|Loss: 0.8798062801361084:  67%|██████▋   | 4307/6470 [57:11<28:15,  1.28it/s]
1|4308|Loss: 0.8901844620704651:  67%|██████▋   | 4307/6470 [57:11<28:15,  1.28it/s]
1|4308|Loss: 0.8901844620704651:  67%|██████▋   | 4308/6470 [57:11<28:11,  1.28it/s]
1|4309|Loss: 0.7793639302253723:  67%|██████▋   | 4308/6470 [57:12<28:11,  1.28it/s]
1|4309|Loss: 0.7793639302253723:  67%|██████▋   | 4309/6470 [57:12<27:11,  1.32it/s]
1|4310|Loss: 0.9887045621871948:  67%|██████▋   | 4309/6470 [57:12<27:11,  1.32it/s]
1|4310|Loss: 0.9887045621871948:  67%|██████▋   | 4310/6470 [57:13<27:01,  1.33it/s]
1|4311|Loss: 0.8121066093444824:  67%|██████▋   | 4310/6470 [57:13<27:01,  1.33it/s]
1|4311|Loss: 0.8121066093444824:  67%|██████▋   | 4311/6470 [57:14<27:00,  1.33it/s]
1|4312|Loss: 0.9456813931465149:  67%|██████▋   | 4311/6470 [57:14<27:00,  1.33it/s]
1|4312|Loss: 0.9456813931465149:  67%|██████▋   | 4312/6470 [57:14<27:32,  1.31it/s]
1|4313|Loss: 1.1342077255249023:  67%|██████▋   | 4312/6470 [57:15<27:32,  1.31it/s]
1|4313|Loss: 1.1342077255249023:  67%|██████▋   | 4313/6470 [57:15<28:59,  1.24it/s]
1|4314|Loss: 0.7982510328292847:  67%|██████▋   | 4313/6470 [57:16<28:59,  1.24it/s]
1|4314|Loss: 0.7982510328292847:  67%|██████▋   | 4314/6470 [57:16<28:50,  1.25it/s]
1|4315|Loss: 0.8051280379295349:  67%|██████▋   | 4314/6470 [57:16<28:50,  1.25it/s]
1|4315|Loss: 0.8051280379295349:  67%|██████▋   | 4315/6470 [57:17<29:53,  1.20it/s]
1|4316|Loss: 0.8621048331260681:  67%|██████▋   | 4315/6470 [57:17<29:53,  1.20it/s]
1|4316|Loss: 0.8621048331260681:  67%|██████▋   | 4316/6470 [57:18<28:52,  1.24it/s]
1|4317|Loss: 0.8149372935295105:  67%|██████▋   | 4316/6470 [57:18<28:52,  1.24it/s]
1|4317|Loss: 0.8149372935295105:  67%|██████▋   | 4317/6470 [57:18<28:28,  1.26it/s]
1|4318|Loss: 0.9341018199920654:  67%|██████▋   | 4317/6470 [57:19<28:28,  1.26it/s]
1|4318|Loss: 0.9341018199920654:  67%|██████▋   | 4318/6470 [57:19<28:43,  1.25it/s]
1|4319|Loss: 1.0044589042663574:  67%|██████▋   | 4318/6470 [57:20<28:43,  1.25it/s]
1|4319|Loss: 1.0044589042663574:  67%|██████▋   | 4319/6470 [57:20<27:32,  1.30it/s]
1|4320|Loss: 1.001914381980896:  67%|██████▋   | 4319/6470 [57:20<27:32,  1.30it/s] 
1|4320|Loss: 1.001914381980896:  67%|██████▋   | 4320/6470 [57:21<28:28,  1.26it/s]
1|4321|Loss: 1.0265512466430664:  67%|██████▋   | 4320/6470 [57:21<28:28,  1.26it/s]
1|4321|Loss: 1.0265512466430664:  67%|██████▋   | 4321/6470 [57:21<26:57,  1.33it/s]
1|4322|Loss: 0.8783669471740723:  67%|██████▋   | 4321/6470 [57:22<26:57,  1.33it/s]
1|4322|Loss: 0.8783669471740723:  67%|██████▋   | 4322/6470 [57:22<28:03,  1.28it/s]
1|4323|Loss: 0.9425203204154968:  67%|██████▋   | 4322/6470 [57:23<28:03,  1.28it/s]
1|4323|Loss: 0.9425203204154968:  67%|██████▋   | 4323/6470 [57:23<29:20,  1.22it/s]
1|4324|Loss: 0.9150805473327637:  67%|██████▋   | 4323/6470 [57:24<29:20,  1.22it/s]
1|4324|Loss: 0.9150805473327637:  67%|██████▋   | 4324/6470 [57:24<29:52,  1.20it/s]
1|4325|Loss: 0.8219400644302368:  67%|██████▋   | 4324/6470 [57:24<29:52,  1.20it/s]
1|4325|Loss: 0.8219400644302368:  67%|██████▋   | 4325/6470 [57:25<28:33,  1.25it/s]
1|4326|Loss: 0.8382006883621216:  67%|██████▋   | 4325/6470 [57:25<28:33,  1.25it/s]
1|4326|Loss: 0.8382006883621216:  67%|██████▋   | 4326/6470 [57:26<28:57,  1.23it/s]
1|4327|Loss: 0.9107811450958252:  67%|██████▋   | 4326/6470 [57:26<28:57,  1.23it/s]
1|4327|Loss: 0.9107811450958252:  67%|██████▋   | 4327/6470 [57:26<28:15,  1.26it/s]
1|4328|Loss: 0.7869083881378174:  67%|██████▋   | 4327/6470 [57:27<28:15,  1.26it/s]
1|4328|Loss: 0.7869083881378174:  67%|██████▋   | 4328/6470 [57:27<25:25,  1.40it/s]
1|4329|Loss: 0.8527197241783142:  67%|██████▋   | 4328/6470 [57:27<25:25,  1.40it/s]
1|4329|Loss: 0.8527197241783142:  67%|██████▋   | 4329/6470 [57:28<25:41,  1.39it/s]
1|4330|Loss: 0.9683701992034912:  67%|██████▋   | 4329/6470 [57:28<25:41,  1.39it/s]
1|4330|Loss: 0.9683701992034912:  67%|██████▋   | 4330/6470 [57:28<24:42,  1.44it/s]
1|4331|Loss: 0.794077455997467:  67%|██████▋   | 4330/6470 [57:29<24:42,  1.44it/s] 
1|4331|Loss: 0.794077455997467:  67%|██████▋   | 4331/6470 [57:29<24:28,  1.46it/s]
1|4332|Loss: 0.7955448031425476:  67%|██████▋   | 4331/6470 [57:29<24:28,  1.46it/s]
1|4332|Loss: 0.7955448031425476:  67%|██████▋   | 4332/6470 [57:30<26:41,  1.33it/s]
1|4333|Loss: 0.9293400049209595:  67%|██████▋   | 4332/6470 [57:30<26:41,  1.33it/s]
1|4333|Loss: 0.9293400049209595:  67%|██████▋   | 4333/6470 [57:31<27:16,  1.31it/s]
1|4334|Loss: 0.90534508228302:  67%|██████▋   | 4333/6470 [57:31<27:16,  1.31it/s]  
1|4334|Loss: 0.90534508228302:  67%|██████▋   | 4334/6470 [57:31<27:15,  1.31it/s]
1|4335|Loss: 0.8828068375587463:  67%|██████▋   | 4334/6470 [57:32<27:15,  1.31it/s]
1|4335|Loss: 0.8828068375587463:  67%|██████▋   | 4335/6470 [57:32<27:36,  1.29it/s]
1|4336|Loss: 0.8019915223121643:  67%|██████▋   | 4335/6470 [57:33<27:36,  1.29it/s]
1|4336|Loss: 0.8019915223121643:  67%|██████▋   | 4336/6470 [57:33<26:47,  1.33it/s]
1|4337|Loss: 0.8662658333778381:  67%|██████▋   | 4336/6470 [57:33<26:47,  1.33it/s]
1|4337|Loss: 0.8662658333778381:  67%|██████▋   | 4337/6470 [57:34<26:48,  1.33it/s]
1|4338|Loss: 0.8468796014785767:  67%|██████▋   | 4337/6470 [57:34<26:48,  1.33it/s]
1|4338|Loss: 0.8468796014785767:  67%|██████▋   | 4338/6470 [57:35<29:05,  1.22it/s]
1|4339|Loss: 0.8541321754455566:  67%|██████▋   | 4338/6470 [57:35<29:05,  1.22it/s]
1|4339|Loss: 0.8541321754455566:  67%|██████▋   | 4339/6470 [57:35<29:04,  1.22it/s]
1|4340|Loss: 0.8714529871940613:  67%|██████▋   | 4339/6470 [57:36<29:04,  1.22it/s]
1|4340|Loss: 0.8714529871940613:  67%|██████▋   | 4340/6470 [57:36<27:56,  1.27it/s]
1|4341|Loss: 0.9565649032592773:  67%|██████▋   | 4340/6470 [57:37<27:56,  1.27it/s]
1|4341|Loss: 0.9565649032592773:  67%|██████▋   | 4341/6470 [57:37<26:40,  1.33it/s]
1|4342|Loss: 0.8509341478347778:  67%|██████▋   | 4341/6470 [57:37<26:40,  1.33it/s]
1|4342|Loss: 0.8509341478347778:  67%|██████▋   | 4342/6470 [57:38<27:57,  1.27it/s]
1|4343|Loss: 1.0484273433685303:  67%|██████▋   | 4342/6470 [57:38<27:57,  1.27it/s]
1|4343|Loss: 1.0484273433685303:  67%|██████▋   | 4343/6470 [57:38<26:36,  1.33it/s]
1|4344|Loss: 0.8876605033874512:  67%|██████▋   | 4343/6470 [57:39<26:36,  1.33it/s]
1|4344|Loss: 0.8876605033874512:  67%|██████▋   | 4344/6470 [57:39<26:22,  1.34it/s]
1|4345|Loss: 1.0042393207550049:  67%|██████▋   | 4344/6470 [57:40<26:22,  1.34it/s]
1|4345|Loss: 1.0042393207550049:  67%|██████▋   | 4345/6470 [57:40<26:29,  1.34it/s]
1|4346|Loss: 0.9551719427108765:  67%|██████▋   | 4345/6470 [57:40<26:29,  1.34it/s]
1|4346|Loss: 0.9551719427108765:  67%|██████▋   | 4346/6470 [57:40<24:13,  1.46it/s]
1|4347|Loss: 1.1378036737442017:  67%|██████▋   | 4346/6470 [57:41<24:13,  1.46it/s]
1|4347|Loss: 1.1378036737442017:  67%|██████▋   | 4347/6470 [57:41<22:51,  1.55it/s]
1|4348|Loss: 0.8540055751800537:  67%|██████▋   | 4347/6470 [57:41<22:51,  1.55it/s]
1|4348|Loss: 0.8540055751800537:  67%|██████▋   | 4348/6470 [57:42<25:10,  1.41it/s]
1|4349|Loss: 1.149838924407959:  67%|██████▋   | 4348/6470 [57:42<25:10,  1.41it/s] 
1|4349|Loss: 1.149838924407959:  67%|██████▋   | 4349/6470 [57:43<27:22,  1.29it/s]
1|4350|Loss: 0.8349698185920715:  67%|██████▋   | 4349/6470 [57:43<27:22,  1.29it/s]
1|4350|Loss: 0.8349698185920715:  67%|██████▋   | 4350/6470 [57:44<27:57,  1.26it/s]
1|4351|Loss: 0.8491516709327698:  67%|██████▋   | 4350/6470 [57:44<27:57,  1.26it/s]
1|4351|Loss: 0.8491516709327698:  67%|██████▋   | 4351/6470 [57:44<25:25,  1.39it/s]
1|4352|Loss: 0.8987016677856445:  67%|██████▋   | 4351/6470 [57:45<25:25,  1.39it/s]
1|4352|Loss: 0.8987016677856445:  67%|██████▋   | 4352/6470 [57:45<26:42,  1.32it/s]
1|4353|Loss: 0.8995802998542786:  67%|██████▋   | 4352/6470 [57:45<26:42,  1.32it/s]
1|4353|Loss: 0.8995802998542786:  67%|██████▋   | 4353/6470 [57:46<25:25,  1.39it/s]
1|4354|Loss: 0.9888470768928528:  67%|██████▋   | 4353/6470 [57:46<25:25,  1.39it/s]
1|4354|Loss: 0.9888470768928528:  67%|██████▋   | 4354/6470 [57:46<24:56,  1.41it/s]
1|4355|Loss: 0.9388163685798645:  67%|██████▋   | 4354/6470 [57:47<24:56,  1.41it/s]
1|4355|Loss: 0.9388163685798645:  67%|██████▋   | 4355/6470 [57:47<26:33,  1.33it/s]
1|4356|Loss: 0.9546360969543457:  67%|██████▋   | 4355/6470 [57:48<26:33,  1.33it/s]
1|4356|Loss: 0.9546360969543457:  67%|██████▋   | 4356/6470 [57:48<26:25,  1.33it/s]
1|4357|Loss: 0.890365481376648:  67%|██████▋   | 4356/6470 [57:48<26:25,  1.33it/s] 
1|4357|Loss: 0.890365481376648:  67%|██████▋   | 4357/6470 [57:49<26:53,  1.31it/s]
1|4358|Loss: 0.8624590635299683:  67%|██████▋   | 4357/6470 [57:49<26:53,  1.31it/s]
1|4358|Loss: 0.8624590635299683:  67%|██████▋   | 4358/6470 [57:49<25:49,  1.36it/s]
1|4359|Loss: 0.8471165895462036:  67%|██████▋   | 4358/6470 [57:50<25:49,  1.36it/s]
1|4359|Loss: 0.8471165895462036:  67%|██████▋   | 4359/6470 [57:50<25:46,  1.37it/s]
1|4360|Loss: 0.7475916743278503:  67%|██████▋   | 4359/6470 [57:51<25:46,  1.37it/s]
1|4360|Loss: 0.7475916743278503:  67%|██████▋   | 4360/6470 [57:51<28:22,  1.24it/s]
1|4361|Loss: 0.9518504738807678:  67%|██████▋   | 4360/6470 [57:52<28:22,  1.24it/s]
1|4361|Loss: 0.9518504738807678:  67%|██████▋   | 4361/6470 [57:52<29:26,  1.19it/s]
1|4362|Loss: 0.8719292879104614:  67%|██████▋   | 4361/6470 [57:52<29:26,  1.19it/s]
1|4362|Loss: 0.8719292879104614:  67%|██████▋   | 4362/6470 [57:53<27:29,  1.28it/s]
1|4363|Loss: 0.8098893165588379:  67%|██████▋   | 4362/6470 [57:53<27:29,  1.28it/s]
1|4363|Loss: 0.8098893165588379:  67%|██████▋   | 4363/6470 [57:53<27:17,  1.29it/s]
1|4364|Loss: 0.892124354839325:  67%|██████▋   | 4363/6470 [57:54<27:17,  1.29it/s] 
1|4364|Loss: 0.892124354839325:  67%|██████▋   | 4364/6470 [57:54<28:15,  1.24it/s]
1|4365|Loss: 0.9024878144264221:  67%|██████▋   | 4364/6470 [57:55<28:15,  1.24it/s]
1|4365|Loss: 0.9024878144264221:  67%|██████▋   | 4365/6470 [57:55<26:38,  1.32it/s]
1|4366|Loss: 0.7972885966300964:  67%|██████▋   | 4365/6470 [57:55<26:38,  1.32it/s]
1|4366|Loss: 0.7972885966300964:  67%|██████▋   | 4366/6470 [57:56<26:24,  1.33it/s]
1|4367|Loss: 0.9539869427680969:  67%|██████▋   | 4366/6470 [57:56<26:24,  1.33it/s]
1|4367|Loss: 0.9539869427680969:  67%|██████▋   | 4367/6470 [57:56<25:27,  1.38it/s]
1|4368|Loss: 0.7266536951065063:  67%|██████▋   | 4367/6470 [57:57<25:27,  1.38it/s]
1|4368|Loss: 0.7266536951065063:  68%|██████▊   | 4368/6470 [57:57<25:53,  1.35it/s]
1|4369|Loss: 0.7552905082702637:  68%|██████▊   | 4368/6470 [57:57<25:53,  1.35it/s]
1|4369|Loss: 0.7552905082702637:  68%|██████▊   | 4369/6470 [57:58<25:22,  1.38it/s]
1|4370|Loss: 0.9255257248878479:  68%|██████▊   | 4369/6470 [57:58<25:22,  1.38it/s]
1|4370|Loss: 0.9255257248878479:  68%|██████▊   | 4370/6470 [57:58<25:11,  1.39it/s]
1|4371|Loss: 1.0264580249786377:  68%|██████▊   | 4370/6470 [57:59<25:11,  1.39it/s]
1|4371|Loss: 1.0264580249786377:  68%|██████▊   | 4371/6470 [57:59<27:35,  1.27it/s]
1|4372|Loss: 0.8335109949111938:  68%|██████▊   | 4371/6470 [58:00<27:35,  1.27it/s]
1|4372|Loss: 0.8335109949111938:  68%|██████▊   | 4372/6470 [58:00<27:46,  1.26it/s]
1|4373|Loss: 0.8719707727432251:  68%|██████▊   | 4372/6470 [58:01<27:46,  1.26it/s]
1|4373|Loss: 0.8719707727432251:  68%|██████▊   | 4373/6470 [58:01<26:14,  1.33it/s]
1|4374|Loss: 0.8836075067520142:  68%|██████▊   | 4373/6470 [58:01<26:14,  1.33it/s]
1|4374|Loss: 0.8836075067520142:  68%|██████▊   | 4374/6470 [58:01<24:42,  1.41it/s]
1|4375|Loss: 1.0553914308547974:  68%|██████▊   | 4374/6470 [58:02<24:42,  1.41it/s]
1|4375|Loss: 1.0553914308547974:  68%|██████▊   | 4375/6470 [58:02<24:31,  1.42it/s]
1|4376|Loss: 0.8319791555404663:  68%|██████▊   | 4375/6470 [58:03<24:31,  1.42it/s]
1|4376|Loss: 0.8319791555404663:  68%|██████▊   | 4376/6470 [58:03<25:21,  1.38it/s]
1|4377|Loss: 0.8847908973693848:  68%|██████▊   | 4376/6470 [58:03<25:21,  1.38it/s]
1|4377|Loss: 0.8847908973693848:  68%|██████▊   | 4377/6470 [58:04<25:18,  1.38it/s]
1|4378|Loss: 0.9139692783355713:  68%|██████▊   | 4377/6470 [58:04<25:18,  1.38it/s]
1|4378|Loss: 0.9139692783355713:  68%|██████▊   | 4378/6470 [58:04<24:47,  1.41it/s]
1|4379|Loss: 0.854766309261322:  68%|██████▊   | 4378/6470 [58:05<24:47,  1.41it/s] 
1|4379|Loss: 0.854766309261322:  68%|██████▊   | 4379/6470 [58:05<25:09,  1.38it/s]
1|4380|Loss: 1.0673178434371948:  68%|██████▊   | 4379/6470 [58:06<25:09,  1.38it/s]
1|4380|Loss: 1.0673178434371948:  68%|██████▊   | 4380/6470 [58:06<25:15,  1.38it/s]
1|4381|Loss: 0.9510870575904846:  68%|██████▊   | 4380/6470 [58:06<25:15,  1.38it/s]
1|4381|Loss: 0.9510870575904846:  68%|██████▊   | 4381/6470 [58:07<25:59,  1.34it/s]
1|4382|Loss: 0.8980503678321838:  68%|██████▊   | 4381/6470 [58:07<25:59,  1.34it/s]
1|4382|Loss: 0.8980503678321838:  68%|██████▊   | 4382/6470 [58:08<28:02,  1.24it/s]
1|4383|Loss: 0.9352355003356934:  68%|██████▊   | 4382/6470 [58:08<28:02,  1.24it/s]
1|4383|Loss: 0.9352355003356934:  68%|██████▊   | 4383/6470 [58:08<28:17,  1.23it/s]
1|4384|Loss: 0.9341253638267517:  68%|██████▊   | 4383/6470 [58:09<28:17,  1.23it/s]
1|4384|Loss: 0.9341253638267517:  68%|██████▊   | 4384/6470 [58:09<27:06,  1.28it/s]
1|4385|Loss: 0.8835224509239197:  68%|██████▊   | 4384/6470 [58:10<27:06,  1.28it/s]
1|4385|Loss: 0.8835224509239197:  68%|██████▊   | 4385/6470 [58:10<27:57,  1.24it/s]
1|4386|Loss: 0.9295051693916321:  68%|██████▊   | 4385/6470 [58:10<27:57,  1.24it/s]
1|4386|Loss: 0.9295051693916321:  68%|██████▊   | 4386/6470 [58:11<28:05,  1.24it/s]
1|4387|Loss: 0.9130629897117615:  68%|██████▊   | 4386/6470 [58:11<28:05,  1.24it/s]
1|4387|Loss: 0.9130629897117615:  68%|██████▊   | 4387/6470 [58:12<28:40,  1.21it/s]
1|4388|Loss: 0.9328312873840332:  68%|██████▊   | 4387/6470 [58:12<28:40,  1.21it/s]
1|4388|Loss: 0.9328312873840332:  68%|██████▊   | 4388/6470 [58:13<29:45,  1.17it/s]
1|4389|Loss: 0.8085506558418274:  68%|██████▊   | 4388/6470 [58:13<29:45,  1.17it/s]
1|4389|Loss: 0.8085506558418274:  68%|██████▊   | 4389/6470 [58:13<29:29,  1.18it/s]
1|4390|Loss: 0.9414894580841064:  68%|██████▊   | 4389/6470 [58:14<29:29,  1.18it/s]
1|4390|Loss: 0.9414894580841064:  68%|██████▊   | 4390/6470 [58:14<26:22,  1.31it/s]
1|4391|Loss: 0.9606907367706299:  68%|██████▊   | 4390/6470 [58:14<26:22,  1.31it/s]
1|4391|Loss: 0.9606907367706299:  68%|██████▊   | 4391/6470 [58:15<24:21,  1.42it/s]
1|4392|Loss: 0.8811222910881042:  68%|██████▊   | 4391/6470 [58:15<24:21,  1.42it/s]
1|4392|Loss: 0.8811222910881042:  68%|██████▊   | 4392/6470 [58:15<25:50,  1.34it/s]
1|4393|Loss: 0.8492789268493652:  68%|██████▊   | 4392/6470 [58:16<25:50,  1.34it/s]
1|4393|Loss: 0.8492789268493652:  68%|██████▊   | 4393/6470 [58:16<27:25,  1.26it/s]
1|4394|Loss: 0.727429986000061:  68%|██████▊   | 4393/6470 [58:17<27:25,  1.26it/s] 
1|4394|Loss: 0.727429986000061:  68%|██████▊   | 4394/6470 [58:17<27:27,  1.26it/s]
1|4395|Loss: 0.9694334268569946:  68%|██████▊   | 4394/6470 [58:18<27:27,  1.26it/s]
1|4395|Loss: 0.9694334268569946:  68%|██████▊   | 4395/6470 [58:18<27:16,  1.27it/s]
1|4396|Loss: 0.8723775744438171:  68%|██████▊   | 4395/6470 [58:18<27:16,  1.27it/s]
1|4396|Loss: 0.8723775744438171:  68%|██████▊   | 4396/6470 [58:19<27:18,  1.27it/s]
1|4397|Loss: 0.9988549947738647:  68%|██████▊   | 4396/6470 [58:19<27:18,  1.27it/s]
1|4397|Loss: 0.9988549947738647:  68%|██████▊   | 4397/6470 [58:20<28:18,  1.22it/s]
1|4398|Loss: 0.9400876760482788:  68%|██████▊   | 4397/6470 [58:20<28:18,  1.22it/s]
1|4398|Loss: 0.9400876760482788:  68%|██████▊   | 4398/6470 [58:20<29:14,  1.18it/s]
1|4399|Loss: 0.9389798641204834:  68%|██████▊   | 4398/6470 [58:21<29:14,  1.18it/s]
1|4399|Loss: 0.9389798641204834:  68%|██████▊   | 4399/6470 [58:21<27:56,  1.24it/s]
1|4400|Loss: 0.8499029278755188:  68%|██████▊   | 4399/6470 [58:22<27:56,  1.24it/s]
1|4400|Loss: 0.8499029278755188:  68%|██████▊   | 4400/6470 [58:22<26:18,  1.31it/s]
1|4401|Loss: 0.8448396325111389:  68%|██████▊   | 4400/6470 [58:22<26:18,  1.31it/s]
1|4401|Loss: 0.8448396325111389:  68%|██████▊   | 4401/6470 [58:23<25:35,  1.35it/s]
1|4402|Loss: 1.0260000228881836:  68%|██████▊   | 4401/6470 [58:23<25:35,  1.35it/s]
1|4402|Loss: 1.0260000228881836:  68%|██████▊   | 4402/6470 [58:23<27:14,  1.27it/s]
1|4403|Loss: 0.8476064801216125:  68%|██████▊   | 4402/6470 [58:24<27:14,  1.27it/s]
1|4403|Loss: 0.8476064801216125:  68%|██████▊   | 4403/6470 [58:24<27:29,  1.25it/s]
1|4404|Loss: 0.8282102346420288:  68%|██████▊   | 4403/6470 [58:25<27:29,  1.25it/s]
1|4404|Loss: 0.8282102346420288:  68%|██████▊   | 4404/6470 [58:25<26:37,  1.29it/s]
1|4405|Loss: 0.8163012266159058:  68%|██████▊   | 4404/6470 [58:25<26:37,  1.29it/s]
1|4405|Loss: 0.8163012266159058:  68%|██████▊   | 4405/6470 [58:26<27:45,  1.24it/s]
1|4406|Loss: 0.9310711622238159:  68%|██████▊   | 4405/6470 [58:26<27:45,  1.24it/s]
1|4406|Loss: 0.9310711622238159:  68%|██████▊   | 4406/6470 [58:27<29:12,  1.18it/s]
1|4407|Loss: 0.8966768383979797:  68%|██████▊   | 4406/6470 [58:27<29:12,  1.18it/s]
1|4407|Loss: 0.8966768383979797:  68%|██████▊   | 4407/6470 [58:28<27:50,  1.24it/s]
1|4408|Loss: 0.8462614417076111:  68%|██████▊   | 4407/6470 [58:28<27:50,  1.24it/s]
1|4408|Loss: 0.8462614417076111:  68%|██████▊   | 4408/6470 [58:28<26:28,  1.30it/s]
1|4409|Loss: 0.8746739625930786:  68%|██████▊   | 4408/6470 [58:29<26:28,  1.30it/s]
1|4409|Loss: 0.8746739625930786:  68%|██████▊   | 4409/6470 [58:29<26:17,  1.31it/s]
1|4410|Loss: 1.108635425567627:  68%|██████▊   | 4409/6470 [58:29<26:17,  1.31it/s] 
1|4410|Loss: 1.108635425567627:  68%|██████▊   | 4410/6470 [58:30<28:20,  1.21it/s]
1|4411|Loss: 0.9720357060432434:  68%|██████▊   | 4410/6470 [58:30<28:20,  1.21it/s]
1|4411|Loss: 0.9720357060432434:  68%|██████▊   | 4411/6470 [58:31<27:48,  1.23it/s]
1|4412|Loss: 0.7981202006340027:  68%|██████▊   | 4411/6470 [58:31<27:48,  1.23it/s]
1|4412|Loss: 0.7981202006340027:  68%|██████▊   | 4412/6470 [58:31<27:21,  1.25it/s]
1|4413|Loss: 0.8839179873466492:  68%|██████▊   | 4412/6470 [58:32<27:21,  1.25it/s]
1|4413|Loss: 0.8839179873466492:  68%|██████▊   | 4413/6470 [58:32<26:31,  1.29it/s]
1|4414|Loss: 1.0993924140930176:  68%|██████▊   | 4413/6470 [58:33<26:31,  1.29it/s]
1|4414|Loss: 1.0993924140930176:  68%|██████▊   | 4414/6470 [58:33<24:16,  1.41it/s]
1|4415|Loss: 0.9153919816017151:  68%|██████▊   | 4414/6470 [58:33<24:16,  1.41it/s]
1|4415|Loss: 0.9153919816017151:  68%|██████▊   | 4415/6470 [58:33<24:18,  1.41it/s]
1|4416|Loss: 0.9061483144760132:  68%|██████▊   | 4415/6470 [58:34<24:18,  1.41it/s]
1|4416|Loss: 0.9061483144760132:  68%|██████▊   | 4416/6470 [58:34<22:31,  1.52it/s]
1|4417|Loss: 0.8710660934448242:  68%|██████▊   | 4416/6470 [58:34<22:31,  1.52it/s]
1|4417|Loss: 0.8710660934448242:  68%|██████▊   | 4417/6470 [58:35<22:22,  1.53it/s]
1|4418|Loss: 1.004014492034912:  68%|██████▊   | 4417/6470 [58:35<22:22,  1.53it/s] 
1|4418|Loss: 1.004014492034912:  68%|██████▊   | 4418/6470 [58:35<24:23,  1.40it/s]
1|4419|Loss: 0.8079137206077576:  68%|██████▊   | 4418/6470 [58:36<24:23,  1.40it/s]
1|4419|Loss: 0.8079137206077576:  68%|██████▊   | 4419/6470 [58:36<21:34,  1.58it/s]
1|4420|Loss: 1.720619797706604:  68%|██████▊   | 4419/6470 [58:36<21:34,  1.58it/s] 
1|4420|Loss: 1.720619797706604:  68%|██████▊   | 4420/6470 [58:37<23:45,  1.44it/s]
1|4421|Loss: 1.1304570436477661:  68%|██████▊   | 4420/6470 [58:37<23:45,  1.44it/s]
1|4421|Loss: 1.1304570436477661:  68%|██████▊   | 4421/6470 [58:38<25:04,  1.36it/s]
1|4422|Loss: 0.8417790532112122:  68%|██████▊   | 4421/6470 [58:38<25:04,  1.36it/s]
1|4422|Loss: 0.8417790532112122:  68%|██████▊   | 4422/6470 [58:38<25:46,  1.32it/s]
1|4423|Loss: 0.8595453500747681:  68%|██████▊   | 4422/6470 [58:39<25:46,  1.32it/s]
1|4423|Loss: 0.8595453500747681:  68%|██████▊   | 4423/6470 [58:39<24:29,  1.39it/s]
1|4424|Loss: 0.8445979952812195:  68%|██████▊   | 4423/6470 [58:39<24:29,  1.39it/s]
1|4424|Loss: 0.8445979952812195:  68%|██████▊   | 4424/6470 [58:40<24:11,  1.41it/s]
1|4425|Loss: 0.8707705736160278:  68%|██████▊   | 4424/6470 [58:40<24:11,  1.41it/s]
1|4425|Loss: 0.8707705736160278:  68%|██████▊   | 4425/6470 [58:40<25:02,  1.36it/s]
1|4426|Loss: 0.8062523603439331:  68%|██████▊   | 4425/6470 [58:41<25:02,  1.36it/s]
1|4426|Loss: 0.8062523603439331:  68%|██████▊   | 4426/6470 [58:41<25:29,  1.34it/s]
1|4427|Loss: 0.7772365212440491:  68%|██████▊   | 4426/6470 [58:42<25:29,  1.34it/s]
1|4427|Loss: 0.7772365212440491:  68%|██████▊   | 4427/6470 [58:42<25:30,  1.33it/s]
1|4428|Loss: 0.8393471837043762:  68%|██████▊   | 4427/6470 [58:42<25:30,  1.33it/s]
1|4428|Loss: 0.8393471837043762:  68%|██████▊   | 4428/6470 [58:43<26:40,  1.28it/s]
1|4429|Loss: 0.8772874474525452:  68%|██████▊   | 4428/6470 [58:43<26:40,  1.28it/s]
1|4429|Loss: 0.8772874474525452:  68%|██████▊   | 4429/6470 [58:44<29:08,  1.17it/s]
1|4430|Loss: 1.7957580089569092:  68%|██████▊   | 4429/6470 [58:45<29:08,  1.17it/s]
1|4430|Loss: 1.7957580089569092:  68%|██████▊   | 4430/6470 [58:45<31:22,  1.08it/s]
1|4431|Loss: 0.8925502300262451:  68%|██████▊   | 4430/6470 [58:45<31:22,  1.08it/s]
1|4431|Loss: 0.8925502300262451:  68%|██████▊   | 4431/6470 [58:46<29:56,  1.13it/s]
1|4432|Loss: 0.9423800706863403:  68%|██████▊   | 4431/6470 [58:46<29:56,  1.13it/s]
1|4432|Loss: 0.9423800706863403:  69%|██████▊   | 4432/6470 [58:46<27:16,  1.25it/s]
1|4433|Loss: 0.9067227244377136:  69%|██████▊   | 4432/6470 [58:47<27:16,  1.25it/s]
1|4433|Loss: 0.9067227244377136:  69%|██████▊   | 4433/6470 [58:47<25:23,  1.34it/s]
1|4434|Loss: 0.9132121205329895:  69%|██████▊   | 4433/6470 [58:47<25:23,  1.34it/s]
1|4434|Loss: 0.9132121205329895:  69%|██████▊   | 4434/6470 [58:48<25:28,  1.33it/s]
1|4435|Loss: 0.9213489294052124:  69%|██████▊   | 4434/6470 [58:48<25:28,  1.33it/s]
1|4435|Loss: 0.9213489294052124:  69%|██████▊   | 4435/6470 [58:49<25:43,  1.32it/s]
1|4436|Loss: 0.9008101224899292:  69%|██████▊   | 4435/6470 [58:49<25:43,  1.32it/s]
1|4436|Loss: 0.9008101224899292:  69%|██████▊   | 4436/6470 [58:49<24:02,  1.41it/s]
1|4437|Loss: 0.6232351660728455:  69%|██████▊   | 4436/6470 [58:49<24:02,  1.41it/s]
1|4437|Loss: 0.6232351660728455:  69%|██████▊   | 4437/6470 [58:50<22:21,  1.52it/s]
1|4438|Loss: 0.9403469562530518:  69%|██████▊   | 4437/6470 [58:50<22:21,  1.52it/s]
1|4438|Loss: 0.9403469562530518:  69%|██████▊   | 4438/6470 [58:50<22:09,  1.53it/s]
1|4439|Loss: 0.9445433020591736:  69%|██████▊   | 4438/6470 [58:51<22:09,  1.53it/s]
1|4439|Loss: 0.9445433020591736:  69%|██████▊   | 4439/6470 [58:51<23:33,  1.44it/s]
1|4440|Loss: 0.8349220156669617:  69%|██████▊   | 4439/6470 [58:52<23:33,  1.44it/s]
1|4440|Loss: 0.8349220156669617:  69%|██████▊   | 4440/6470 [58:52<23:04,  1.47it/s]
1|4441|Loss: 0.8365061283111572:  69%|██████▊   | 4440/6470 [58:52<23:04,  1.47it/s]
1|4441|Loss: 0.8365061283111572:  69%|██████▊   | 4441/6470 [58:52<22:20,  1.51it/s]
1|4442|Loss: 0.824743390083313:  69%|██████▊   | 4441/6470 [58:53<22:20,  1.51it/s] 
1|4442|Loss: 0.824743390083313:  69%|██████▊   | 4442/6470 [58:53<23:59,  1.41it/s]
1|4443|Loss: 0.8606209754943848:  69%|██████▊   | 4442/6470 [58:54<23:59,  1.41it/s]
1|4443|Loss: 0.8606209754943848:  69%|██████▊   | 4443/6470 [58:54<24:39,  1.37it/s]
1|4444|Loss: 1.0349382162094116:  69%|██████▊   | 4443/6470 [58:54<24:39,  1.37it/s]
1|4444|Loss: 1.0349382162094116:  69%|██████▊   | 4444/6470 [58:55<24:59,  1.35it/s]
1|4445|Loss: 0.7290744781494141:  69%|██████▊   | 4444/6470 [58:55<24:59,  1.35it/s]
1|4445|Loss: 0.7290744781494141:  69%|██████▊   | 4445/6470 [58:56<25:11,  1.34it/s]
1|4446|Loss: 0.9424552321434021:  69%|██████▊   | 4445/6470 [58:56<25:11,  1.34it/s]
1|4446|Loss: 0.9424552321434021:  69%|██████▊   | 4446/6470 [58:56<25:37,  1.32it/s]
1|4447|Loss: 0.7738755345344543:  69%|██████▊   | 4446/6470 [58:57<25:37,  1.32it/s]
1|4447|Loss: 0.7738755345344543:  69%|██████▊   | 4447/6470 [58:57<24:40,  1.37it/s]
1|4448|Loss: 0.9734915494918823:  69%|██████▊   | 4447/6470 [58:57<24:40,  1.37it/s]
1|4448|Loss: 0.9734915494918823:  69%|██████▊   | 4448/6470 [58:58<23:53,  1.41it/s]
1|4449|Loss: 0.8435407280921936:  69%|██████▊   | 4448/6470 [58:58<23:53,  1.41it/s]
1|4449|Loss: 0.8435407280921936:  69%|██████▉   | 4449/6470 [58:58<24:08,  1.40it/s]
1|4450|Loss: 0.9090843200683594:  69%|██████▉   | 4449/6470 [58:59<24:08,  1.40it/s]
1|4450|Loss: 0.9090843200683594:  69%|██████▉   | 4450/6470 [58:59<23:17,  1.45it/s]
1|4451|Loss: 1.0349669456481934:  69%|██████▉   | 4450/6470 [58:59<23:17,  1.45it/s]
1|4451|Loss: 1.0349669456481934:  69%|██████▉   | 4451/6470 [59:00<24:58,  1.35it/s]
1|4452|Loss: 0.788664698600769:  69%|██████▉   | 4451/6470 [59:00<24:58,  1.35it/s] 
1|4452|Loss: 0.788664698600769:  69%|██████▉   | 4452/6470 [59:00<23:53,  1.41it/s]
1|4453|Loss: 0.8966048955917358:  69%|██████▉   | 4452/6470 [59:01<23:53,  1.41it/s]
1|4453|Loss: 0.8966048955917358:  69%|██████▉   | 4453/6470 [59:01<24:03,  1.40it/s]
1|4454|Loss: 0.9365147948265076:  69%|██████▉   | 4453/6470 [59:02<24:03,  1.40it/s]
1|4454|Loss: 0.9365147948265076:  69%|██████▉   | 4454/6470 [59:02<24:01,  1.40it/s]
1|4455|Loss: 0.9008603692054749:  69%|██████▉   | 4454/6470 [59:02<24:01,  1.40it/s]
1|4455|Loss: 0.9008603692054749:  69%|██████▉   | 4455/6470 [59:03<26:53,  1.25it/s]
1|4456|Loss: 0.7789275050163269:  69%|██████▉   | 4455/6470 [59:03<26:53,  1.25it/s]
1|4456|Loss: 0.7789275050163269:  69%|██████▉   | 4456/6470 [59:04<26:04,  1.29it/s]
1|4457|Loss: 0.779265284538269:  69%|██████▉   | 4456/6470 [59:04<26:04,  1.29it/s] 
1|4457|Loss: 0.779265284538269:  69%|██████▉   | 4457/6470 [59:05<27:11,  1.23it/s]
1|4458|Loss: 0.9933740496635437:  69%|██████▉   | 4457/6470 [59:05<27:11,  1.23it/s]
1|4458|Loss: 0.9933740496635437:  69%|██████▉   | 4458/6470 [59:05<26:33,  1.26it/s]
1|4459|Loss: 1.030846357345581:  69%|██████▉   | 4458/6470 [59:06<26:33,  1.26it/s] 
1|4459|Loss: 1.030846357345581:  69%|██████▉   | 4459/6470 [59:06<26:25,  1.27it/s]
1|4460|Loss: 0.9644593596458435:  69%|██████▉   | 4459/6470 [59:06<26:25,  1.27it/s]
1|4460|Loss: 0.9644593596458435:  69%|██████▉   | 4460/6470 [59:07<25:07,  1.33it/s]
1|4461|Loss: 1.174111247062683:  69%|██████▉   | 4460/6470 [59:07<25:07,  1.33it/s] 
1|4461|Loss: 1.174111247062683:  69%|██████▉   | 4461/6470 [59:07<25:01,  1.34it/s]
1|4462|Loss: 0.7947567105293274:  69%|██████▉   | 4461/6470 [59:08<25:01,  1.34it/s]
1|4462|Loss: 0.7947567105293274:  69%|██████▉   | 4462/6470 [59:08<24:24,  1.37it/s]
1|4463|Loss: 0.8347525000572205:  69%|██████▉   | 4462/6470 [59:09<24:24,  1.37it/s]
1|4463|Loss: 0.8347525000572205:  69%|██████▉   | 4463/6470 [59:09<26:12,  1.28it/s]
1|4464|Loss: 0.9194284677505493:  69%|██████▉   | 4463/6470 [59:10<26:12,  1.28it/s]
1|4464|Loss: 0.9194284677505493:  69%|██████▉   | 4464/6470 [59:10<25:37,  1.31it/s]
1|4465|Loss: 0.8142064809799194:  69%|██████▉   | 4464/6470 [59:10<25:37,  1.31it/s]
1|4465|Loss: 0.8142064809799194:  69%|██████▉   | 4465/6470 [59:11<25:35,  1.31it/s]
1|4466|Loss: 0.7506866455078125:  69%|██████▉   | 4465/6470 [59:11<25:35,  1.31it/s]
1|4466|Loss: 0.7506866455078125:  69%|██████▉   | 4466/6470 [59:11<25:07,  1.33it/s]
1|4467|Loss: 0.9057362079620361:  69%|██████▉   | 4466/6470 [59:12<25:07,  1.33it/s]
1|4467|Loss: 0.9057362079620361:  69%|██████▉   | 4467/6470 [59:12<25:05,  1.33it/s]
1|4468|Loss: 0.8784798979759216:  69%|██████▉   | 4467/6470 [59:12<25:05,  1.33it/s]
1|4468|Loss: 0.8784798979759216:  69%|██████▉   | 4468/6470 [59:13<25:23,  1.31it/s]
1|4469|Loss: 0.7652875185012817:  69%|██████▉   | 4468/6470 [59:13<25:23,  1.31it/s]
1|4469|Loss: 0.7652875185012817:  69%|██████▉   | 4469/6470 [59:14<25:41,  1.30it/s]
1|4470|Loss: 0.9766051173210144:  69%|██████▉   | 4469/6470 [59:14<25:41,  1.30it/s]
1|4470|Loss: 0.9766051173210144:  69%|██████▉   | 4470/6470 [59:15<27:01,  1.23it/s]
1|4471|Loss: 0.8547931909561157:  69%|██████▉   | 4470/6470 [59:15<27:01,  1.23it/s]
1|4471|Loss: 0.8547931909561157:  69%|██████▉   | 4471/6470 [59:15<28:04,  1.19it/s]
1|4472|Loss: 0.90690016746521:  69%|██████▉   | 4471/6470 [59:16<28:04,  1.19it/s]  
1|4472|Loss: 0.90690016746521:  69%|██████▉   | 4472/6470 [59:16<26:55,  1.24it/s]
1|4473|Loss: 0.901500403881073:  69%|██████▉   | 4472/6470 [59:17<26:55,  1.24it/s]
1|4473|Loss: 0.901500403881073:  69%|██████▉   | 4473/6470 [59:17<25:45,  1.29it/s]
1|4474|Loss: 0.8351022005081177:  69%|██████▉   | 4473/6470 [59:17<25:45,  1.29it/s]
1|4474|Loss: 0.8351022005081177:  69%|██████▉   | 4474/6470 [59:18<25:03,  1.33it/s]
1|4475|Loss: 0.8066339492797852:  69%|██████▉   | 4474/6470 [59:18<25:03,  1.33it/s]
1|4475|Loss: 0.8066339492797852:  69%|██████▉   | 4475/6470 [59:18<24:39,  1.35it/s]
1|4476|Loss: 0.9098360538482666:  69%|██████▉   | 4475/6470 [59:19<24:39,  1.35it/s]
1|4476|Loss: 0.9098360538482666:  69%|██████▉   | 4476/6470 [59:19<23:51,  1.39it/s]
1|4477|Loss: 0.9532524347305298:  69%|██████▉   | 4476/6470 [59:19<23:51,  1.39it/s]
1|4477|Loss: 0.9532524347305298:  69%|██████▉   | 4477/6470 [59:20<23:34,  1.41it/s]
1|4478|Loss: 0.6531124711036682:  69%|██████▉   | 4477/6470 [59:20<23:34,  1.41it/s]
1|4478|Loss: 0.6531124711036682:  69%|██████▉   | 4478/6470 [59:20<23:14,  1.43it/s]
1|4479|Loss: 0.7972128987312317:  69%|██████▉   | 4478/6470 [59:21<23:14,  1.43it/s]
1|4479|Loss: 0.7972128987312317:  69%|██████▉   | 4479/6470 [59:21<24:54,  1.33it/s]
1|4480|Loss: 0.8078901171684265:  69%|██████▉   | 4479/6470 [59:22<24:54,  1.33it/s]
1|4480|Loss: 0.8078901171684265:  69%|██████▉   | 4480/6470 [59:22<25:19,  1.31it/s]
1|4481|Loss: 0.8791570663452148:  69%|██████▉   | 4480/6470 [59:22<25:19,  1.31it/s]
1|4481|Loss: 0.8791570663452148:  69%|██████▉   | 4481/6470 [59:23<25:19,  1.31it/s]
1|4482|Loss: 0.8161900043487549:  69%|██████▉   | 4481/6470 [59:23<25:19,  1.31it/s]
1|4482|Loss: 0.8161900043487549:  69%|██████▉   | 4482/6470 [59:23<25:16,  1.31it/s]
1|4483|Loss: 0.9536676406860352:  69%|██████▉   | 4482/6470 [59:24<25:16,  1.31it/s]
1|4483|Loss: 0.9536676406860352:  69%|██████▉   | 4483/6470 [59:24<26:16,  1.26it/s]
1|4484|Loss: 0.772389829158783:  69%|██████▉   | 4483/6470 [59:25<26:16,  1.26it/s] 
1|4484|Loss: 0.772389829158783:  69%|██████▉   | 4484/6470 [59:25<26:57,  1.23it/s]
1|4485|Loss: 0.8254699110984802:  69%|██████▉   | 4484/6470 [59:26<26:57,  1.23it/s]
1|4485|Loss: 0.8254699110984802:  69%|██████▉   | 4485/6470 [59:26<24:28,  1.35it/s]
1|4486|Loss: 1.0329136848449707:  69%|██████▉   | 4485/6470 [59:26<24:28,  1.35it/s]
1|4486|Loss: 1.0329136848449707:  69%|██████▉   | 4486/6470 [59:27<25:30,  1.30it/s]
1|4487|Loss: 0.9624212384223938:  69%|██████▉   | 4486/6470 [59:27<25:30,  1.30it/s]
1|4487|Loss: 0.9624212384223938:  69%|██████▉   | 4487/6470 [59:27<24:59,  1.32it/s]
1|4488|Loss: 0.8897190690040588:  69%|██████▉   | 4487/6470 [59:28<24:59,  1.32it/s]
1|4488|Loss: 0.8897190690040588:  69%|██████▉   | 4488/6470 [59:28<25:42,  1.28it/s]
1|4489|Loss: 0.924630880355835:  69%|██████▉   | 4488/6470 [59:29<25:42,  1.28it/s] 
1|4489|Loss: 0.924630880355835:  69%|██████▉   | 4489/6470 [59:29<25:10,  1.31it/s]
1|4490|Loss: 0.8662236332893372:  69%|██████▉   | 4489/6470 [59:29<25:10,  1.31it/s]
1|4490|Loss: 0.8662236332893372:  69%|██████▉   | 4490/6470 [59:29<23:15,  1.42it/s]
1|4491|Loss: 0.9206291437149048:  69%|██████▉   | 4490/6470 [59:30<23:15,  1.42it/s]
1|4491|Loss: 0.9206291437149048:  69%|██████▉   | 4491/6470 [59:30<24:09,  1.37it/s]
1|4492|Loss: 0.9252732396125793:  69%|██████▉   | 4491/6470 [59:31<24:09,  1.37it/s]
1|4492|Loss: 0.9252732396125793:  69%|██████▉   | 4492/6470 [59:31<22:43,  1.45it/s]
1|4493|Loss: 0.8903766870498657:  69%|██████▉   | 4492/6470 [59:31<22:43,  1.45it/s]
1|4493|Loss: 0.8903766870498657:  69%|██████▉   | 4493/6470 [59:32<22:50,  1.44it/s]
1|4494|Loss: 0.9664890170097351:  69%|██████▉   | 4493/6470 [59:32<22:50,  1.44it/s]
1|4494|Loss: 0.9664890170097351:  69%|██████▉   | 4494/6470 [59:32<23:08,  1.42it/s]
1|4495|Loss: 0.8274011015892029:  69%|██████▉   | 4494/6470 [59:33<23:08,  1.42it/s]
1|4495|Loss: 0.8274011015892029:  69%|██████▉   | 4495/6470 [59:33<22:32,  1.46it/s]
1|4496|Loss: 0.924835741519928:  69%|██████▉   | 4495/6470 [59:33<22:32,  1.46it/s] 
1|4496|Loss: 0.924835741519928:  69%|██████▉   | 4496/6470 [59:33<21:05,  1.56it/s]
1|4497|Loss: 1.2034852504730225:  69%|██████▉   | 4496/6470 [59:34<21:05,  1.56it/s]
1|4497|Loss: 1.2034852504730225:  70%|██████▉   | 4497/6470 [59:34<21:03,  1.56it/s]
1|4498|Loss: 0.969815194606781:  70%|██████▉   | 4497/6470 [59:34<21:03,  1.56it/s] 
1|4498|Loss: 0.969815194606781:  70%|██████▉   | 4498/6470 [59:35<21:04,  1.56it/s]
1|4499|Loss: 0.9990705251693726:  70%|██████▉   | 4498/6470 [59:35<21:04,  1.56it/s]
1|4499|Loss: 0.9990705251693726:  70%|██████▉   | 4499/6470 [59:36<23:10,  1.42it/s]
1|4500|Loss: 0.9102162718772888:  70%|██████▉   | 4499/6470 [59:36<23:10,  1.42it/s]
1|4500|Loss: 0.9102162718772888:  70%|██████▉   | 4500/6470 [59:36<23:48,  1.38it/s]
1|4501|Loss: 0.9160748720169067:  70%|██████▉   | 4500/6470 [59:37<23:48,  1.38it/s]
1|4501|Loss: 0.9160748720169067:  70%|██████▉   | 4501/6470 [59:37<24:24,  1.34it/s]
1|4502|Loss: 0.8441430926322937:  70%|██████▉   | 4501/6470 [59:38<24:24,  1.34it/s]
1|4502|Loss: 0.8441430926322937:  70%|██████▉   | 4502/6470 [59:38<24:35,  1.33it/s]
1|4503|Loss: 0.801400363445282:  70%|██████▉   | 4502/6470 [59:38<24:35,  1.33it/s] 
1|4503|Loss: 0.801400363445282:  70%|██████▉   | 4503/6470 [59:39<24:41,  1.33it/s]
1|4504|Loss: 0.9930095672607422:  70%|██████▉   | 4503/6470 [59:39<24:41,  1.33it/s]
1|4504|Loss: 0.9930095672607422:  70%|██████▉   | 4504/6470 [59:39<25:11,  1.30it/s]
1|4505|Loss: 0.8932580351829529:  70%|██████▉   | 4504/6470 [59:40<25:11,  1.30it/s]
1|4505|Loss: 0.8932580351829529:  70%|██████▉   | 4505/6470 [59:40<26:37,  1.23it/s]
1|4506|Loss: 0.9468784332275391:  70%|██████▉   | 4505/6470 [59:41<26:37,  1.23it/s]
1|4506|Loss: 0.9468784332275391:  70%|██████▉   | 4506/6470 [59:41<25:33,  1.28it/s]
1|4507|Loss: 0.9383569359779358:  70%|██████▉   | 4506/6470 [59:42<25:33,  1.28it/s]
1|4507|Loss: 0.9383569359779358:  70%|██████▉   | 4507/6470 [59:42<26:44,  1.22it/s]
1|4508|Loss: 1.111812710762024:  70%|██████▉   | 4507/6470 [59:42<26:44,  1.22it/s] 
1|4508|Loss: 1.111812710762024:  70%|██████▉   | 4508/6470 [59:43<26:12,  1.25it/s]
1|4509|Loss: 1.1747688055038452:  70%|██████▉   | 4508/6470 [59:43<26:12,  1.25it/s]
1|4509|Loss: 1.1747688055038452:  70%|██████▉   | 4509/6470 [59:43<25:15,  1.29it/s]
1|4510|Loss: 1.0067819356918335:  70%|██████▉   | 4509/6470 [59:44<25:15,  1.29it/s]
1|4510|Loss: 1.0067819356918335:  70%|██████▉   | 4510/6470 [59:44<25:18,  1.29it/s]
1|4511|Loss: 0.8863958120346069:  70%|██████▉   | 4510/6470 [59:45<25:18,  1.29it/s]
1|4511|Loss: 0.8863958120346069:  70%|██████▉   | 4511/6470 [59:45<24:34,  1.33it/s]
1|4512|Loss: 0.8273289799690247:  70%|██████▉   | 4511/6470 [59:45<24:34,  1.33it/s]
1|4512|Loss: 0.8273289799690247:  70%|██████▉   | 4512/6470 [59:46<26:13,  1.24it/s]
1|4513|Loss: 0.8913013935089111:  70%|██████▉   | 4512/6470 [59:46<26:13,  1.24it/s]
1|4513|Loss: 0.8913013935089111:  70%|██████▉   | 4513/6470 [59:47<25:24,  1.28it/s]
1|4514|Loss: 0.8806312680244446:  70%|██████▉   | 4513/6470 [59:47<25:24,  1.28it/s]
1|4514|Loss: 0.8806312680244446:  70%|██████▉   | 4514/6470 [59:47<25:13,  1.29it/s]
1|4515|Loss: 0.9021621346473694:  70%|██████▉   | 4514/6470 [59:48<25:13,  1.29it/s]
1|4515|Loss: 0.9021621346473694:  70%|██████▉   | 4515/6470 [59:48<24:26,  1.33it/s]
1|4516|Loss: 0.8319721221923828:  70%|██████▉   | 4515/6470 [59:49<24:26,  1.33it/s]
1|4516|Loss: 0.8319721221923828:  70%|██████▉   | 4516/6470 [59:49<25:25,  1.28it/s]
1|4517|Loss: 0.9734238982200623:  70%|██████▉   | 4516/6470 [59:49<25:25,  1.28it/s]
1|4517|Loss: 0.9734238982200623:  70%|██████▉   | 4517/6470 [59:50<24:49,  1.31it/s]
1|4518|Loss: 0.9598197340965271:  70%|██████▉   | 4517/6470 [59:50<24:49,  1.31it/s]
1|4518|Loss: 0.9598197340965271:  70%|██████▉   | 4518/6470 [59:50<23:57,  1.36it/s]
1|4519|Loss: 0.8138534426689148:  70%|██████▉   | 4518/6470 [59:51<23:57,  1.36it/s]
1|4519|Loss: 0.8138534426689148:  70%|██████▉   | 4519/6470 [59:51<23:18,  1.40it/s]
1|4520|Loss: 0.9975121021270752:  70%|██████▉   | 4519/6470 [59:51<23:18,  1.40it/s]
1|4520|Loss: 0.9975121021270752:  70%|██████▉   | 4520/6470 [59:52<23:57,  1.36it/s]
1|4521|Loss: 0.8135800957679749:  70%|██████▉   | 4520/6470 [59:52<23:57,  1.36it/s]
1|4521|Loss: 0.8135800957679749:  70%|██████▉   | 4521/6470 [59:52<23:40,  1.37it/s]
1|4522|Loss: 0.9511386752128601:  70%|██████▉   | 4521/6470 [59:53<23:40,  1.37it/s]
1|4522|Loss: 0.9511386752128601:  70%|██████▉   | 4522/6470 [59:53<23:24,  1.39it/s]
1|4523|Loss: 0.8404196500778198:  70%|██████▉   | 4522/6470 [59:54<23:24,  1.39it/s]
1|4523|Loss: 0.8404196500778198:  70%|██████▉   | 4523/6470 [59:54<23:06,  1.40it/s]
1|4524|Loss: 0.8662917017936707:  70%|██████▉   | 4523/6470 [59:54<23:06,  1.40it/s]
1|4524|Loss: 0.8662917017936707:  70%|██████▉   | 4524/6470 [59:55<22:46,  1.42it/s]
1|4525|Loss: 0.9113300442695618:  70%|██████▉   | 4524/6470 [59:55<22:46,  1.42it/s]
1|4525|Loss: 0.9113300442695618:  70%|██████▉   | 4525/6470 [59:55<22:44,  1.43it/s]
1|4526|Loss: 0.7393923997879028:  70%|██████▉   | 4525/6470 [59:56<22:44,  1.43it/s]
1|4526|Loss: 0.7393923997879028:  70%|██████▉   | 4526/6470 [59:56<23:21,  1.39it/s]
1|4527|Loss: 0.7665167450904846:  70%|██████▉   | 4526/6470 [59:56<23:21,  1.39it/s]
1|4527|Loss: 0.7665167450904846:  70%|██████▉   | 4527/6470 [59:57<23:23,  1.38it/s]
1|4528|Loss: 0.9359430074691772:  70%|██████▉   | 4527/6470 [59:57<23:23,  1.38it/s]
1|4528|Loss: 0.9359430074691772:  70%|██████▉   | 4528/6470 [59:57<23:35,  1.37it/s]
1|4529|Loss: 0.8752247095108032:  70%|██████▉   | 4528/6470 [59:58<23:35,  1.37it/s]
1|4529|Loss: 0.8752247095108032:  70%|███████   | 4529/6470 [59:58<23:28,  1.38it/s]
1|4530|Loss: 0.805120587348938:  70%|███████   | 4529/6470 [59:59<23:28,  1.38it/s] 
1|4530|Loss: 0.805120587348938:  70%|███████   | 4530/6470 [59:59<22:39,  1.43it/s]
1|4531|Loss: 0.7819761037826538:  70%|███████   | 4530/6470 [59:59<22:39,  1.43it/s]
1|4531|Loss: 0.7819761037826538:  70%|███████   | 4531/6470 [59:59<21:46,  1.48it/s]
1|4532|Loss: 0.8896626830101013:  70%|███████   | 4531/6470 [1:00:00<21:46,  1.48it/s]
1|4532|Loss: 0.8896626830101013:  70%|███████   | 4532/6470 [1:00:00<22:36,  1.43it/s]
1|4533|Loss: 0.9543259739875793:  70%|███████   | 4532/6470 [1:00:01<22:36,  1.43it/s]
1|4533|Loss: 0.9543259739875793:  70%|███████   | 4533/6470 [1:00:01<23:31,  1.37it/s]
1|4534|Loss: 1.011884331703186:  70%|███████   | 4533/6470 [1:00:01<23:31,  1.37it/s] 
1|4534|Loss: 1.011884331703186:  70%|███████   | 4534/6470 [1:00:02<24:11,  1.33it/s]
1|4535|Loss: 0.981852114200592:  70%|███████   | 4534/6470 [1:00:02<24:11,  1.33it/s]
1|4535|Loss: 0.981852114200592:  70%|███████   | 4535/6470 [1:00:03<23:56,  1.35it/s]
1|4536|Loss: 0.9141939282417297:  70%|███████   | 4535/6470 [1:00:03<23:56,  1.35it/s]
1|4536|Loss: 0.9141939282417297:  70%|███████   | 4536/6470 [1:00:03<24:16,  1.33it/s]
1|4537|Loss: 0.9832415580749512:  70%|███████   | 4536/6470 [1:00:04<24:16,  1.33it/s]
1|4537|Loss: 0.9832415580749512:  70%|███████   | 4537/6470 [1:00:04<24:23,  1.32it/s]
1|4538|Loss: 0.8094784617424011:  70%|███████   | 4537/6470 [1:00:04<24:23,  1.32it/s]
1|4538|Loss: 0.8094784617424011:  70%|███████   | 4538/6470 [1:00:05<23:43,  1.36it/s]
1|4539|Loss: 0.780612051486969:  70%|███████   | 4538/6470 [1:00:05<23:43,  1.36it/s] 
1|4539|Loss: 0.780612051486969:  70%|███████   | 4539/6470 [1:00:06<24:03,  1.34it/s]
1|4540|Loss: 0.9792660474777222:  70%|███████   | 4539/6470 [1:00:06<24:03,  1.34it/s]
1|4540|Loss: 0.9792660474777222:  70%|███████   | 4540/6470 [1:00:06<23:34,  1.36it/s]
1|4541|Loss: 0.8743467330932617:  70%|███████   | 4540/6470 [1:00:07<23:34,  1.36it/s]
1|4541|Loss: 0.8743467330932617:  70%|███████   | 4541/6470 [1:00:07<22:48,  1.41it/s]
1|4542|Loss: 0.9691916108131409:  70%|███████   | 4541/6470 [1:00:07<22:48,  1.41it/s]
1|4542|Loss: 0.9691916108131409:  70%|███████   | 4542/6470 [1:00:08<23:15,  1.38it/s]
1|4543|Loss: 0.7671164870262146:  70%|███████   | 4542/6470 [1:00:08<23:15,  1.38it/s]
1|4543|Loss: 0.7671164870262146:  70%|███████   | 4543/6470 [1:00:08<22:22,  1.44it/s]
1|4544|Loss: 0.7682836055755615:  70%|███████   | 4543/6470 [1:00:09<22:22,  1.44it/s]
1|4544|Loss: 0.7682836055755615:  70%|███████   | 4544/6470 [1:00:09<23:35,  1.36it/s]
1|4545|Loss: 0.7546120882034302:  70%|███████   | 4544/6470 [1:00:10<23:35,  1.36it/s]
1|4545|Loss: 0.7546120882034302:  70%|███████   | 4545/6470 [1:00:10<25:28,  1.26it/s]
1|4546|Loss: 0.8415734767913818:  70%|███████   | 4545/6470 [1:00:11<25:28,  1.26it/s]
1|4546|Loss: 0.8415734767913818:  70%|███████   | 4546/6470 [1:00:11<27:37,  1.16it/s]
1|4547|Loss: 0.966115415096283:  70%|███████   | 4546/6470 [1:00:12<27:37,  1.16it/s] 
1|4547|Loss: 0.966115415096283:  70%|███████   | 4547/6470 [1:00:12<27:59,  1.15it/s]
1|4548|Loss: 0.7947496175765991:  70%|███████   | 4547/6470 [1:00:12<27:59,  1.15it/s]
1|4548|Loss: 0.7947496175765991:  70%|███████   | 4548/6470 [1:00:13<27:07,  1.18it/s]
1|4549|Loss: 0.8960415124893188:  70%|███████   | 4548/6470 [1:00:13<27:07,  1.18it/s]
1|4549|Loss: 0.8960415124893188:  70%|███████   | 4549/6470 [1:00:14<27:39,  1.16it/s]
1|4550|Loss: 0.8937448263168335:  70%|███████   | 4549/6470 [1:00:14<27:39,  1.16it/s]
1|4550|Loss: 0.8937448263168335:  70%|███████   | 4550/6470 [1:00:14<26:49,  1.19it/s]
1|4551|Loss: 1.0960513353347778:  70%|███████   | 4550/6470 [1:00:15<26:49,  1.19it/s]
1|4551|Loss: 1.0960513353347778:  70%|███████   | 4551/6470 [1:00:15<26:07,  1.22it/s]
1|4552|Loss: 0.8542225360870361:  70%|███████   | 4551/6470 [1:00:16<26:07,  1.22it/s]
1|4552|Loss: 0.8542225360870361:  70%|███████   | 4552/6470 [1:00:16<25:36,  1.25it/s]
1|4553|Loss: 0.9144379496574402:  70%|███████   | 4552/6470 [1:00:16<25:36,  1.25it/s]
1|4553|Loss: 0.9144379496574402:  70%|███████   | 4553/6470 [1:00:17<25:24,  1.26it/s]
1|4554|Loss: 0.8014478087425232:  70%|███████   | 4553/6470 [1:00:17<25:24,  1.26it/s]
1|4554|Loss: 0.8014478087425232:  70%|███████   | 4554/6470 [1:00:18<25:38,  1.25it/s]
1|4555|Loss: 1.1679699420928955:  70%|███████   | 4554/6470 [1:00:18<25:38,  1.25it/s]
1|4555|Loss: 1.1679699420928955:  70%|███████   | 4555/6470 [1:00:18<26:33,  1.20it/s]
1|4556|Loss: 0.9428251385688782:  70%|███████   | 4555/6470 [1:00:19<26:33,  1.20it/s]
1|4556|Loss: 0.9428251385688782:  70%|███████   | 4556/6470 [1:00:19<27:23,  1.16it/s]
1|4557|Loss: 0.9589935541152954:  70%|███████   | 4556/6470 [1:00:20<27:23,  1.16it/s]
1|4557|Loss: 0.9589935541152954:  70%|███████   | 4557/6470 [1:00:20<25:15,  1.26it/s]
1|4558|Loss: 0.8997437357902527:  70%|███████   | 4557/6470 [1:00:20<25:15,  1.26it/s]
1|4558|Loss: 0.8997437357902527:  70%|███████   | 4558/6470 [1:00:21<23:25,  1.36it/s]
1|4559|Loss: 0.8161609768867493:  70%|███████   | 4558/6470 [1:00:21<23:25,  1.36it/s]
1|4559|Loss: 0.8161609768867493:  70%|███████   | 4559/6470 [1:00:21<23:47,  1.34it/s]
1|4560|Loss: 0.7682852149009705:  70%|███████   | 4559/6470 [1:00:22<23:47,  1.34it/s]
1|4560|Loss: 0.7682852149009705:  70%|███████   | 4560/6470 [1:00:22<25:53,  1.23it/s]
1|4561|Loss: 0.7417614459991455:  70%|███████   | 4560/6470 [1:00:23<25:53,  1.23it/s]
1|4561|Loss: 0.7417614459991455:  70%|███████   | 4561/6470 [1:00:23<27:21,  1.16it/s]
1|4562|Loss: 0.9510672688484192:  70%|███████   | 4561/6470 [1:00:24<27:21,  1.16it/s]
1|4562|Loss: 0.9510672688484192:  71%|███████   | 4562/6470 [1:00:24<26:48,  1.19it/s]
1|4563|Loss: 0.8297606110572815:  71%|███████   | 4562/6470 [1:00:25<26:48,  1.19it/s]
1|4563|Loss: 0.8297606110572815:  71%|███████   | 4563/6470 [1:00:25<26:01,  1.22it/s]
1|4564|Loss: 1.0153363943099976:  71%|███████   | 4563/6470 [1:00:25<26:01,  1.22it/s]
1|4564|Loss: 1.0153363943099976:  71%|███████   | 4564/6470 [1:00:26<25:50,  1.23it/s]
1|4565|Loss: 0.840717613697052:  71%|███████   | 4564/6470 [1:00:26<25:50,  1.23it/s] 
1|4565|Loss: 0.840717613697052:  71%|███████   | 4565/6470 [1:00:26<25:25,  1.25it/s]
1|4566|Loss: 0.9888120293617249:  71%|███████   | 4565/6470 [1:00:27<25:25,  1.25it/s]
1|4566|Loss: 0.9888120293617249:  71%|███████   | 4566/6470 [1:00:27<25:16,  1.26it/s]
1|4567|Loss: 0.7664074897766113:  71%|███████   | 4566/6470 [1:00:28<25:16,  1.26it/s]
1|4567|Loss: 0.7664074897766113:  71%|███████   | 4567/6470 [1:00:28<25:11,  1.26it/s]
1|4568|Loss: 1.1870795488357544:  71%|███████   | 4567/6470 [1:00:28<25:11,  1.26it/s]
1|4568|Loss: 1.1870795488357544:  71%|███████   | 4568/6470 [1:00:29<24:16,  1.31it/s]
1|4569|Loss: 0.8060992360115051:  71%|███████   | 4568/6470 [1:00:29<24:16,  1.31it/s]
1|4569|Loss: 0.8060992360115051:  71%|███████   | 4569/6470 [1:00:30<25:09,  1.26it/s]
1|4570|Loss: 0.7817975282669067:  71%|███████   | 4569/6470 [1:00:30<25:09,  1.26it/s]
1|4570|Loss: 0.7817975282669067:  71%|███████   | 4570/6470 [1:00:30<24:03,  1.32it/s]
1|4571|Loss: 0.6804349422454834:  71%|███████   | 4570/6470 [1:00:31<24:03,  1.32it/s]
1|4571|Loss: 0.6804349422454834:  71%|███████   | 4571/6470 [1:00:31<24:07,  1.31it/s]
1|4572|Loss: 0.9945403933525085:  71%|███████   | 4571/6470 [1:00:32<24:07,  1.31it/s]
1|4572|Loss: 0.9945403933525085:  71%|███████   | 4572/6470 [1:00:32<24:24,  1.30it/s]
1|4573|Loss: 0.9209710955619812:  71%|███████   | 4572/6470 [1:00:32<24:24,  1.30it/s]
1|4573|Loss: 0.9209710955619812:  71%|███████   | 4573/6470 [1:00:33<24:44,  1.28it/s]
1|4574|Loss: 1.0662879943847656:  71%|███████   | 4573/6470 [1:00:33<24:44,  1.28it/s]
1|4574|Loss: 1.0662879943847656:  71%|███████   | 4574/6470 [1:00:33<24:58,  1.27it/s]
1|4575|Loss: 0.9147723913192749:  71%|███████   | 4574/6470 [1:00:34<24:58,  1.27it/s]
1|4575|Loss: 0.9147723913192749:  71%|███████   | 4575/6470 [1:00:34<24:44,  1.28it/s]
1|4576|Loss: 0.9515188932418823:  71%|███████   | 4575/6470 [1:00:35<24:44,  1.28it/s]
1|4576|Loss: 0.9515188932418823:  71%|███████   | 4576/6470 [1:00:35<25:50,  1.22it/s]
1|4577|Loss: 0.9229596853256226:  71%|███████   | 4576/6470 [1:00:36<25:50,  1.22it/s]
1|4577|Loss: 0.9229596853256226:  71%|███████   | 4577/6470 [1:00:36<26:52,  1.17it/s]
1|4578|Loss: 1.0965352058410645:  71%|███████   | 4577/6470 [1:00:37<26:52,  1.17it/s]
1|4578|Loss: 1.0965352058410645:  71%|███████   | 4578/6470 [1:00:37<27:39,  1.14it/s]
1|4579|Loss: 0.7681949734687805:  71%|███████   | 4578/6470 [1:00:37<27:39,  1.14it/s]
1|4579|Loss: 0.7681949734687805:  71%|███████   | 4579/6470 [1:00:38<27:01,  1.17it/s]
1|4580|Loss: 0.9413789510726929:  71%|███████   | 4579/6470 [1:00:38<27:01,  1.17it/s]
1|4580|Loss: 0.9413789510726929:  71%|███████   | 4580/6470 [1:00:39<25:28,  1.24it/s]
1|4581|Loss: 0.8310537338256836:  71%|███████   | 4580/6470 [1:00:39<25:28,  1.24it/s]
1|4581|Loss: 0.8310537338256836:  71%|███████   | 4581/6470 [1:00:39<24:26,  1.29it/s]
1|4582|Loss: 0.8217836618423462:  71%|███████   | 4581/6470 [1:00:40<24:26,  1.29it/s]
1|4582|Loss: 0.8217836618423462:  71%|███████   | 4582/6470 [1:00:40<21:19,  1.48it/s]
1|4583|Loss: 0.9131913185119629:  71%|███████   | 4582/6470 [1:00:40<21:19,  1.48it/s]
1|4583|Loss: 0.9131913185119629:  71%|███████   | 4583/6470 [1:00:40<20:23,  1.54it/s]
1|4584|Loss: 0.7987286448478699:  71%|███████   | 4583/6470 [1:00:41<20:23,  1.54it/s]
1|4584|Loss: 0.7987286448478699:  71%|███████   | 4584/6470 [1:00:41<21:10,  1.48it/s]
1|4585|Loss: 0.9068362712860107:  71%|███████   | 4584/6470 [1:00:41<21:10,  1.48it/s]
1|4585|Loss: 0.9068362712860107:  71%|███████   | 4585/6470 [1:00:42<21:16,  1.48it/s]
1|4586|Loss: 0.9290961027145386:  71%|███████   | 4585/6470 [1:00:42<21:16,  1.48it/s]
1|4586|Loss: 0.9290961027145386:  71%|███████   | 4586/6470 [1:00:42<21:54,  1.43it/s]
1|4587|Loss: 1.039287805557251:  71%|███████   | 4586/6470 [1:00:43<21:54,  1.43it/s] 
1|4587|Loss: 1.039287805557251:  71%|███████   | 4587/6470 [1:00:43<22:29,  1.40it/s]
1|4588|Loss: 0.8465538620948792:  71%|███████   | 4587/6470 [1:00:43<22:29,  1.40it/s]
1|4588|Loss: 0.8465538620948792:  71%|███████   | 4588/6470 [1:00:44<20:00,  1.57it/s]
1|4589|Loss: 0.7730756998062134:  71%|███████   | 4588/6470 [1:00:44<20:00,  1.57it/s]
1|4589|Loss: 0.7730756998062134:  71%|███████   | 4589/6470 [1:00:44<19:44,  1.59it/s]
1|4590|Loss: 0.930477499961853:  71%|███████   | 4589/6470 [1:00:45<19:44,  1.59it/s] 
1|4590|Loss: 0.930477499961853:  71%|███████   | 4590/6470 [1:00:45<22:36,  1.39it/s]
1|4591|Loss: 0.8789259195327759:  71%|███████   | 4590/6470 [1:00:46<22:36,  1.39it/s]
1|4591|Loss: 0.8789259195327759:  71%|███████   | 4591/6470 [1:00:46<22:09,  1.41it/s]
1|4592|Loss: 0.8260313868522644:  71%|███████   | 4591/6470 [1:00:46<22:09,  1.41it/s]
1|4592|Loss: 0.8260313868522644:  71%|███████   | 4592/6470 [1:00:47<22:52,  1.37it/s]
1|4593|Loss: 0.9260979294776917:  71%|███████   | 4592/6470 [1:00:47<22:52,  1.37it/s]
1|4593|Loss: 0.9260979294776917:  71%|███████   | 4593/6470 [1:00:47<22:20,  1.40it/s]
1|4594|Loss: 0.7869224548339844:  71%|███████   | 4593/6470 [1:00:48<22:20,  1.40it/s]
1|4594|Loss: 0.7869224548339844:  71%|███████   | 4594/6470 [1:00:48<24:07,  1.30it/s]
1|4595|Loss: 1.0802950859069824:  71%|███████   | 4594/6470 [1:00:49<24:07,  1.30it/s]
1|4595|Loss: 1.0802950859069824:  71%|███████   | 4595/6470 [1:00:49<24:49,  1.26it/s]
1|4596|Loss: 0.9206400513648987:  71%|███████   | 4595/6470 [1:00:49<24:49,  1.26it/s]
1|4596|Loss: 0.9206400513648987:  71%|███████   | 4596/6470 [1:00:50<22:57,  1.36it/s]
1|4597|Loss: 1.0092988014221191:  71%|███████   | 4596/6470 [1:00:50<22:57,  1.36it/s]
1|4597|Loss: 1.0092988014221191:  71%|███████   | 4597/6470 [1:00:51<24:43,  1.26it/s]
1|4598|Loss: 0.9236162900924683:  71%|███████   | 4597/6470 [1:00:51<24:43,  1.26it/s]
1|4598|Loss: 0.9236162900924683:  71%|███████   | 4598/6470 [1:00:51<22:16,  1.40it/s]
1|4599|Loss: 0.9025315642356873:  71%|███████   | 4598/6470 [1:00:51<22:16,  1.40it/s]
1|4599|Loss: 0.9025315642356873:  71%|███████   | 4599/6470 [1:00:52<21:03,  1.48it/s]
1|4600|Loss: 0.9696776866912842:  71%|███████   | 4599/6470 [1:00:52<21:03,  1.48it/s]
1|4600|Loss: 0.9696776866912842:  71%|███████   | 4600/6470 [1:00:52<22:01,  1.42it/s]
1|4601|Loss: 0.8937916159629822:  71%|███████   | 4600/6470 [1:00:53<22:01,  1.42it/s]
1|4601|Loss: 0.8937916159629822:  71%|███████   | 4601/6470 [1:00:53<21:48,  1.43it/s]
1|4602|Loss: 0.9317120909690857:  71%|███████   | 4601/6470 [1:00:54<21:48,  1.43it/s]
1|4602|Loss: 0.9317120909690857:  71%|███████   | 4602/6470 [1:00:54<23:25,  1.33it/s]
1|4603|Loss: 0.9596226215362549:  71%|███████   | 4602/6470 [1:00:54<23:25,  1.33it/s]
1|4603|Loss: 0.9596226215362549:  71%|███████   | 4603/6470 [1:00:55<23:50,  1.30it/s]
1|4604|Loss: 0.93863844871521:  71%|███████   | 4603/6470 [1:00:55<23:50,  1.30it/s]  
1|4604|Loss: 0.93863844871521:  71%|███████   | 4604/6470 [1:00:56<24:15,  1.28it/s]
1|4605|Loss: 0.9431029558181763:  71%|███████   | 4604/6470 [1:00:56<24:15,  1.28it/s]
1|4605|Loss: 0.9431029558181763:  71%|███████   | 4605/6470 [1:00:56<24:40,  1.26it/s]
1|4606|Loss: 1.0687280893325806:  71%|███████   | 4605/6470 [1:00:57<24:40,  1.26it/s]
1|4606|Loss: 1.0687280893325806:  71%|███████   | 4606/6470 [1:00:57<26:14,  1.18it/s]
1|4607|Loss: 1.046834111213684:  71%|███████   | 4606/6470 [1:00:58<26:14,  1.18it/s] 
1|4607|Loss: 1.046834111213684:  71%|███████   | 4607/6470 [1:00:58<25:12,  1.23it/s]
1|4608|Loss: 0.9561402201652527:  71%|███████   | 4607/6470 [1:00:59<25:12,  1.23it/s]
1|4608|Loss: 0.9561402201652527:  71%|███████   | 4608/6470 [1:00:59<24:53,  1.25it/s]
1|4609|Loss: 0.8609941601753235:  71%|███████   | 4608/6470 [1:00:59<24:53,  1.25it/s]
1|4609|Loss: 0.8609941601753235:  71%|███████   | 4609/6470 [1:01:00<24:35,  1.26it/s]
1|4610|Loss: 0.8590851426124573:  71%|███████   | 4609/6470 [1:01:00<24:35,  1.26it/s]
1|4610|Loss: 0.8590851426124573:  71%|███████▏  | 4610/6470 [1:01:00<23:52,  1.30it/s]
1|4611|Loss: 0.9115216732025146:  71%|███████▏  | 4610/6470 [1:01:01<23:52,  1.30it/s]
1|4611|Loss: 0.9115216732025146:  71%|███████▏  | 4611/6470 [1:01:01<23:13,  1.33it/s]
1|4612|Loss: 1.1508041620254517:  71%|███████▏  | 4611/6470 [1:01:02<23:13,  1.33it/s]
1|4612|Loss: 1.1508041620254517:  71%|███████▏  | 4612/6470 [1:01:02<23:31,  1.32it/s]
1|4613|Loss: 0.866762101650238:  71%|███████▏  | 4612/6470 [1:01:02<23:31,  1.32it/s] 
1|4613|Loss: 0.866762101650238:  71%|███████▏  | 4613/6470 [1:01:03<24:28,  1.26it/s]
1|4614|Loss: 0.7153278589248657:  71%|███████▏  | 4613/6470 [1:01:03<24:28,  1.26it/s]
1|4614|Loss: 0.7153278589248657:  71%|███████▏  | 4614/6470 [1:01:04<24:45,  1.25it/s]
1|4615|Loss: 0.806059718132019:  71%|███████▏  | 4614/6470 [1:01:04<24:45,  1.25it/s] 
1|4615|Loss: 0.806059718132019:  71%|███████▏  | 4615/6470 [1:01:04<25:18,  1.22it/s]
1|4616|Loss: 0.8772978782653809:  71%|███████▏  | 4615/6470 [1:01:05<25:18,  1.22it/s]
1|4616|Loss: 0.8772978782653809:  71%|███████▏  | 4616/6470 [1:01:05<25:10,  1.23it/s]
1|4617|Loss: 0.8393006920814514:  71%|███████▏  | 4616/6470 [1:01:06<25:10,  1.23it/s]
1|4617|Loss: 0.8393006920814514:  71%|███████▏  | 4617/6470 [1:01:06<25:10,  1.23it/s]
1|4618|Loss: 0.7568981051445007:  71%|███████▏  | 4617/6470 [1:01:07<25:10,  1.23it/s]
1|4618|Loss: 0.7568981051445007:  71%|███████▏  | 4618/6470 [1:01:07<24:46,  1.25it/s]
1|4619|Loss: 0.9159495830535889:  71%|███████▏  | 4618/6470 [1:01:07<24:46,  1.25it/s]
1|4619|Loss: 0.9159495830535889:  71%|███████▏  | 4619/6470 [1:01:08<24:49,  1.24it/s]
1|4620|Loss: 0.9551123380661011:  71%|███████▏  | 4619/6470 [1:01:08<24:49,  1.24it/s]
1|4620|Loss: 0.9551123380661011:  71%|███████▏  | 4620/6470 [1:01:08<24:48,  1.24it/s]
1|4621|Loss: 0.9226081967353821:  71%|███████▏  | 4620/6470 [1:01:09<24:48,  1.24it/s]
1|4621|Loss: 0.9226081967353821:  71%|███████▏  | 4621/6470 [1:01:09<25:27,  1.21it/s]
1|4622|Loss: 0.842030942440033:  71%|███████▏  | 4621/6470 [1:01:10<25:27,  1.21it/s] 
1|4622|Loss: 0.842030942440033:  71%|███████▏  | 4622/6470 [1:01:10<24:41,  1.25it/s]
1|4623|Loss: 0.9147740006446838:  71%|███████▏  | 4622/6470 [1:01:10<24:41,  1.25it/s]
1|4623|Loss: 0.9147740006446838:  71%|███████▏  | 4623/6470 [1:01:11<23:37,  1.30it/s]
1|4624|Loss: 0.8779622316360474:  71%|███████▏  | 4623/6470 [1:01:11<23:37,  1.30it/s]
1|4624|Loss: 0.8779622316360474:  71%|███████▏  | 4624/6470 [1:01:11<22:27,  1.37it/s]
1|4625|Loss: 0.7562403678894043:  71%|███████▏  | 4624/6470 [1:01:12<22:27,  1.37it/s]
1|4625|Loss: 0.7562403678894043:  71%|███████▏  | 4625/6470 [1:01:12<21:29,  1.43it/s]
1|4626|Loss: 1.05879545211792:  71%|███████▏  | 4625/6470 [1:01:12<21:29,  1.43it/s]  
1|4626|Loss: 1.05879545211792:  71%|███████▏  | 4626/6470 [1:01:13<22:13,  1.38it/s]
1|4627|Loss: 0.8648571968078613:  71%|███████▏  | 4626/6470 [1:01:13<22:13,  1.38it/s]
1|4627|Loss: 0.8648571968078613:  72%|███████▏  | 4627/6470 [1:01:14<22:52,  1.34it/s]
1|4628|Loss: 0.8248451948165894:  72%|███████▏  | 4627/6470 [1:01:14<22:52,  1.34it/s]
1|4628|Loss: 0.8248451948165894:  72%|███████▏  | 4628/6470 [1:01:14<23:08,  1.33it/s]
1|4629|Loss: 0.9438134431838989:  72%|███████▏  | 4628/6470 [1:01:15<23:08,  1.33it/s]
1|4629|Loss: 0.9438134431838989:  72%|███████▏  | 4629/6470 [1:01:15<23:19,  1.32it/s]
1|4630|Loss: 0.9579639434814453:  72%|███████▏  | 4629/6470 [1:01:16<23:19,  1.32it/s]
1|4630|Loss: 0.9579639434814453:  72%|███████▏  | 4630/6470 [1:01:16<24:50,  1.23it/s]
1|4631|Loss: 0.9081906676292419:  72%|███████▏  | 4630/6470 [1:01:17<24:50,  1.23it/s]
1|4631|Loss: 0.9081906676292419:  72%|███████▏  | 4631/6470 [1:01:17<23:18,  1.32it/s]
1|4632|Loss: 0.9482631683349609:  72%|███████▏  | 4631/6470 [1:01:17<23:18,  1.32it/s]
1|4632|Loss: 0.9482631683349609:  72%|███████▏  | 4632/6470 [1:01:17<23:12,  1.32it/s]
1|4633|Loss: 0.704003095626831:  72%|███████▏  | 4632/6470 [1:01:18<23:12,  1.32it/s] 
1|4633|Loss: 0.704003095626831:  72%|███████▏  | 4633/6470 [1:01:18<23:37,  1.30it/s]
1|4634|Loss: 0.8180751800537109:  72%|███████▏  | 4633/6470 [1:01:19<23:37,  1.30it/s]
1|4634|Loss: 0.8180751800537109:  72%|███████▏  | 4634/6470 [1:01:19<23:17,  1.31it/s]
1|4635|Loss: 0.9850669503211975:  72%|███████▏  | 4634/6470 [1:01:19<23:17,  1.31it/s]
1|4635|Loss: 0.9850669503211975:  72%|███████▏  | 4635/6470 [1:01:20<22:36,  1.35it/s]
1|4636|Loss: 0.965503454208374:  72%|███████▏  | 4635/6470 [1:01:20<22:36,  1.35it/s] 
1|4636|Loss: 0.965503454208374:  72%|███████▏  | 4636/6470 [1:01:21<22:57,  1.33it/s]
1|4637|Loss: 0.9874600768089294:  72%|███████▏  | 4636/6470 [1:01:21<22:57,  1.33it/s]
1|4637|Loss: 0.9874600768089294:  72%|███████▏  | 4637/6470 [1:01:21<22:43,  1.34it/s]
1|4638|Loss: 0.9558121562004089:  72%|███████▏  | 4637/6470 [1:01:22<22:43,  1.34it/s]
1|4638|Loss: 0.9558121562004089:  72%|███████▏  | 4638/6470 [1:01:22<23:00,  1.33it/s]
1|4639|Loss: 0.9200505614280701:  72%|███████▏  | 4638/6470 [1:01:22<23:00,  1.33it/s]
1|4639|Loss: 0.9200505614280701:  72%|███████▏  | 4639/6470 [1:01:23<22:27,  1.36it/s]
1|4640|Loss: 0.8757731914520264:  72%|███████▏  | 4639/6470 [1:01:23<22:27,  1.36it/s]
1|4640|Loss: 0.8757731914520264:  72%|███████▏  | 4640/6470 [1:01:23<21:51,  1.39it/s]
1|4641|Loss: 0.8451789021492004:  72%|███████▏  | 4640/6470 [1:01:24<21:51,  1.39it/s]
1|4641|Loss: 0.8451789021492004:  72%|███████▏  | 4641/6470 [1:01:24<22:18,  1.37it/s]
1|4642|Loss: 1.0838900804519653:  72%|███████▏  | 4641/6470 [1:01:25<22:18,  1.37it/s]
1|4642|Loss: 1.0838900804519653:  72%|███████▏  | 4642/6470 [1:01:25<21:56,  1.39it/s]
1|4643|Loss: 1.1502050161361694:  72%|███████▏  | 4642/6470 [1:01:25<21:56,  1.39it/s]
1|4643|Loss: 1.1502050161361694:  72%|███████▏  | 4643/6470 [1:01:26<22:31,  1.35it/s]
1|4644|Loss: 0.8724865317344666:  72%|███████▏  | 4643/6470 [1:01:26<22:31,  1.35it/s]
1|4644|Loss: 0.8724865317344666:  72%|███████▏  | 4644/6470 [1:01:26<23:03,  1.32it/s]
1|4645|Loss: 0.9918994307518005:  72%|███████▏  | 4644/6470 [1:01:27<23:03,  1.32it/s]
1|4645|Loss: 0.9918994307518005:  72%|███████▏  | 4645/6470 [1:01:27<23:38,  1.29it/s]
1|4646|Loss: 0.9423730969429016:  72%|███████▏  | 4645/6470 [1:01:28<23:38,  1.29it/s]
1|4646|Loss: 0.9423730969429016:  72%|███████▏  | 4646/6470 [1:01:28<24:27,  1.24it/s]
1|4647|Loss: 1.036432147026062:  72%|███████▏  | 4646/6470 [1:01:29<24:27,  1.24it/s] 
1|4647|Loss: 1.036432147026062:  72%|███████▏  | 4647/6470 [1:01:29<23:49,  1.28it/s]
1|4648|Loss: 0.9652411341667175:  72%|███████▏  | 4647/6470 [1:01:29<23:49,  1.28it/s]
1|4648|Loss: 0.9652411341667175:  72%|███████▏  | 4648/6470 [1:01:30<23:07,  1.31it/s]
1|4649|Loss: 0.9184377193450928:  72%|███████▏  | 4648/6470 [1:01:30<23:07,  1.31it/s]
1|4649|Loss: 0.9184377193450928:  72%|███████▏  | 4649/6470 [1:01:30<22:40,  1.34it/s]
1|4650|Loss: 0.8648855090141296:  72%|███████▏  | 4649/6470 [1:01:31<22:40,  1.34it/s]
1|4650|Loss: 0.8648855090141296:  72%|███████▏  | 4650/6470 [1:01:31<22:38,  1.34it/s]
1|4651|Loss: 0.9474939703941345:  72%|███████▏  | 4650/6470 [1:01:31<22:38,  1.34it/s]
1|4651|Loss: 0.9474939703941345:  72%|███████▏  | 4651/6470 [1:01:32<22:42,  1.34it/s]
1|4652|Loss: 0.8748576641082764:  72%|███████▏  | 4651/6470 [1:01:32<22:42,  1.34it/s]
1|4652|Loss: 0.8748576641082764:  72%|███████▏  | 4652/6470 [1:01:33<22:45,  1.33it/s]
1|4653|Loss: 0.8451358675956726:  72%|███████▏  | 4652/6470 [1:01:33<22:45,  1.33it/s]
1|4653|Loss: 0.8451358675956726:  72%|███████▏  | 4653/6470 [1:01:33<22:13,  1.36it/s]
1|4654|Loss: 0.9123393297195435:  72%|███████▏  | 4653/6470 [1:01:34<22:13,  1.36it/s]
1|4654|Loss: 0.9123393297195435:  72%|███████▏  | 4654/6470 [1:01:34<22:31,  1.34it/s]
1|4655|Loss: 0.9854786396026611:  72%|███████▏  | 4654/6470 [1:01:34<22:31,  1.34it/s]
1|4655|Loss: 0.9854786396026611:  72%|███████▏  | 4655/6470 [1:01:35<21:00,  1.44it/s]
1|4656|Loss: 0.7336680293083191:  72%|███████▏  | 4655/6470 [1:01:35<21:00,  1.44it/s]
1|4656|Loss: 0.7336680293083191:  72%|███████▏  | 4656/6470 [1:01:35<18:32,  1.63it/s]
1|4657|Loss: 0.7646710276603699:  72%|███████▏  | 4656/6470 [1:01:35<18:32,  1.63it/s]
1|4657|Loss: 0.7646710276603699:  72%|███████▏  | 4657/6470 [1:01:36<18:52,  1.60it/s]
1|4658|Loss: 0.9468273520469666:  72%|███████▏  | 4657/6470 [1:01:36<18:52,  1.60it/s]
1|4658|Loss: 0.9468273520469666:  72%|███████▏  | 4658/6470 [1:01:36<20:19,  1.49it/s]
1|4659|Loss: 0.8331917524337769:  72%|███████▏  | 4658/6470 [1:01:37<20:19,  1.49it/s]
1|4659|Loss: 0.8331917524337769:  72%|███████▏  | 4659/6470 [1:01:37<21:20,  1.41it/s]
1|4660|Loss: 0.8285741806030273:  72%|███████▏  | 4659/6470 [1:01:38<21:20,  1.41it/s]
1|4660|Loss: 0.8285741806030273:  72%|███████▏  | 4660/6470 [1:01:38<22:35,  1.34it/s]
1|4661|Loss: 0.7321094870567322:  72%|███████▏  | 4660/6470 [1:01:38<22:35,  1.34it/s]
1|4661|Loss: 0.7321094870567322:  72%|███████▏  | 4661/6470 [1:01:39<20:08,  1.50it/s]
1|4662|Loss: 0.9057908654212952:  72%|███████▏  | 4661/6470 [1:01:39<20:08,  1.50it/s]
1|4662|Loss: 0.9057908654212952:  72%|███████▏  | 4662/6470 [1:01:39<19:33,  1.54it/s]
1|4663|Loss: 0.93292236328125:  72%|███████▏  | 4662/6470 [1:01:40<19:33,  1.54it/s]  
1|4663|Loss: 0.93292236328125:  72%|███████▏  | 4663/6470 [1:01:40<20:24,  1.48it/s]
1|4664|Loss: 0.7216550707817078:  72%|███████▏  | 4663/6470 [1:01:40<20:24,  1.48it/s]
1|4664|Loss: 0.7216550707817078:  72%|███████▏  | 4664/6470 [1:01:41<21:47,  1.38it/s]
1|4665|Loss: 0.9648690223693848:  72%|███████▏  | 4664/6470 [1:01:41<21:47,  1.38it/s]
1|4665|Loss: 0.9648690223693848:  72%|███████▏  | 4665/6470 [1:01:42<22:40,  1.33it/s]
1|4666|Loss: 0.8776514530181885:  72%|███████▏  | 4665/6470 [1:01:42<22:40,  1.33it/s]
1|4666|Loss: 0.8776514530181885:  72%|███████▏  | 4666/6470 [1:01:43<25:01,  1.20it/s]
1|4667|Loss: 0.8449546694755554:  72%|███████▏  | 4666/6470 [1:01:43<25:01,  1.20it/s]
1|4667|Loss: 0.8449546694755554:  72%|███████▏  | 4667/6470 [1:01:44<26:02,  1.15it/s]
1|4668|Loss: 0.9408643841743469:  72%|███████▏  | 4667/6470 [1:01:44<26:02,  1.15it/s]
1|4668|Loss: 0.9408643841743469:  72%|███████▏  | 4668/6470 [1:01:44<24:40,  1.22it/s]
1|4669|Loss: 0.6211317181587219:  72%|███████▏  | 4668/6470 [1:01:45<24:40,  1.22it/s]
1|4669|Loss: 0.6211317181587219:  72%|███████▏  | 4669/6470 [1:01:45<25:39,  1.17it/s]
1|4670|Loss: 0.8988585472106934:  72%|███████▏  | 4669/6470 [1:01:46<25:39,  1.17it/s]
1|4670|Loss: 0.8988585472106934:  72%|███████▏  | 4670/6470 [1:01:46<25:30,  1.18it/s]
1|4671|Loss: 1.021777629852295:  72%|███████▏  | 4670/6470 [1:01:46<25:30,  1.18it/s] 
1|4671|Loss: 1.021777629852295:  72%|███████▏  | 4671/6470 [1:01:47<25:05,  1.19it/s]
1|4672|Loss: 0.8533706665039062:  72%|███████▏  | 4671/6470 [1:01:47<25:05,  1.19it/s]
1|4672|Loss: 0.8533706665039062:  72%|███████▏  | 4672/6470 [1:01:48<24:46,  1.21it/s]
1|4673|Loss: 0.7291080951690674:  72%|███████▏  | 4672/6470 [1:01:48<24:46,  1.21it/s]
1|4673|Loss: 0.7291080951690674:  72%|███████▏  | 4673/6470 [1:01:48<22:26,  1.33it/s]
1|4674|Loss: 0.7674439549446106:  72%|███████▏  | 4673/6470 [1:01:49<22:26,  1.33it/s]
1|4674|Loss: 0.7674439549446106:  72%|███████▏  | 4674/6470 [1:01:49<24:02,  1.25it/s]
1|4675|Loss: 0.8508004546165466:  72%|███████▏  | 4674/6470 [1:01:49<24:02,  1.25it/s]
1|4675|Loss: 0.8508004546165466:  72%|███████▏  | 4675/6470 [1:01:50<22:05,  1.35it/s]
1|4676|Loss: 0.9216582775115967:  72%|███████▏  | 4675/6470 [1:01:50<22:05,  1.35it/s]
1|4676|Loss: 0.9216582775115967:  72%|███████▏  | 4676/6470 [1:01:50<22:34,  1.32it/s]
1|4677|Loss: 0.793282687664032:  72%|███████▏  | 4676/6470 [1:01:51<22:34,  1.32it/s] 
1|4677|Loss: 0.793282687664032:  72%|███████▏  | 4677/6470 [1:01:51<23:47,  1.26it/s]
1|4678|Loss: 0.9246667623519897:  72%|███████▏  | 4677/6470 [1:01:52<23:47,  1.26it/s]
1|4678|Loss: 0.9246667623519897:  72%|███████▏  | 4678/6470 [1:01:52<23:46,  1.26it/s]
1|4679|Loss: 0.7416560053825378:  72%|███████▏  | 4678/6470 [1:01:53<23:46,  1.26it/s]
1|4679|Loss: 0.7416560053825378:  72%|███████▏  | 4679/6470 [1:01:53<23:54,  1.25it/s]
1|4680|Loss: 0.9712092876434326:  72%|███████▏  | 4679/6470 [1:01:53<23:54,  1.25it/s]
1|4680|Loss: 0.9712092876434326:  72%|███████▏  | 4680/6470 [1:01:54<22:59,  1.30it/s]
1|4681|Loss: 1.0909357070922852:  72%|███████▏  | 4680/6470 [1:01:54<22:59,  1.30it/s]
1|4681|Loss: 1.0909357070922852:  72%|███████▏  | 4681/6470 [1:01:55<24:05,  1.24it/s]
1|4682|Loss: 0.7588034272193909:  72%|███████▏  | 4681/6470 [1:01:55<24:05,  1.24it/s]
1|4682|Loss: 0.7588034272193909:  72%|███████▏  | 4682/6470 [1:01:55<23:10,  1.29it/s]
1|4683|Loss: 1.06008780002594:  72%|███████▏  | 4682/6470 [1:01:56<23:10,  1.29it/s]  
1|4683|Loss: 1.06008780002594:  72%|███████▏  | 4683/6470 [1:01:56<22:48,  1.31it/s]
1|4684|Loss: 0.8420150279998779:  72%|███████▏  | 4683/6470 [1:01:56<22:48,  1.31it/s]
1|4684|Loss: 0.8420150279998779:  72%|███████▏  | 4684/6470 [1:01:57<23:02,  1.29it/s]
1|4685|Loss: 0.940974771976471:  72%|███████▏  | 4684/6470 [1:01:57<23:02,  1.29it/s] 
1|4685|Loss: 0.940974771976471:  72%|███████▏  | 4685/6470 [1:01:58<22:42,  1.31it/s]
1|4686|Loss: 0.8714484572410583:  72%|███████▏  | 4685/6470 [1:01:58<22:42,  1.31it/s]
1|4686|Loss: 0.8714484572410583:  72%|███████▏  | 4686/6470 [1:01:58<23:42,  1.25it/s]
1|4687|Loss: 0.9162261486053467:  72%|███████▏  | 4686/6470 [1:01:59<23:42,  1.25it/s]
1|4687|Loss: 0.9162261486053467:  72%|███████▏  | 4687/6470 [1:01:59<24:35,  1.21it/s]
1|4688|Loss: 0.8782810568809509:  72%|███████▏  | 4687/6470 [1:02:00<24:35,  1.21it/s]
1|4688|Loss: 0.8782810568809509:  72%|███████▏  | 4688/6470 [1:02:00<24:27,  1.21it/s]
1|4689|Loss: 0.8632545471191406:  72%|███████▏  | 4688/6470 [1:02:01<24:27,  1.21it/s]
1|4689|Loss: 0.8632545471191406:  72%|███████▏  | 4689/6470 [1:02:01<27:42,  1.07it/s]
1|4690|Loss: 0.888774037361145:  72%|███████▏  | 4689/6470 [1:02:02<27:42,  1.07it/s] 
1|4690|Loss: 0.888774037361145:  72%|███████▏  | 4690/6470 [1:02:02<28:19,  1.05it/s]
1|4691|Loss: 0.892122209072113:  72%|███████▏  | 4690/6470 [1:02:03<28:19,  1.05it/s]
1|4691|Loss: 0.892122209072113:  73%|███████▎  | 4691/6470 [1:02:03<26:01,  1.14it/s]
1|4692|Loss: 0.6233496069908142:  73%|███████▎  | 4691/6470 [1:02:03<26:01,  1.14it/s]
1|4692|Loss: 0.6233496069908142:  73%|███████▎  | 4692/6470 [1:02:04<22:41,  1.31it/s]
1|4693|Loss: 0.8922440409660339:  73%|███████▎  | 4692/6470 [1:02:04<22:41,  1.31it/s]
1|4693|Loss: 0.8922440409660339:  73%|███████▎  | 4693/6470 [1:02:04<22:24,  1.32it/s]
1|4694|Loss: 1.0546468496322632:  73%|███████▎  | 4693/6470 [1:02:05<22:24,  1.32it/s]
1|4694|Loss: 1.0546468496322632:  73%|███████▎  | 4694/6470 [1:02:05<21:13,  1.39it/s]
1|4695|Loss: 0.9573445320129395:  73%|███████▎  | 4694/6470 [1:02:05<21:13,  1.39it/s]
1|4695|Loss: 0.9573445320129395:  73%|███████▎  | 4695/6470 [1:02:05<19:14,  1.54it/s]
1|4696|Loss: 0.7863197326660156:  73%|███████▎  | 4695/6470 [1:02:06<19:14,  1.54it/s]
1|4696|Loss: 0.7863197326660156:  73%|███████▎  | 4696/6470 [1:02:06<18:27,  1.60it/s]
1|4697|Loss: 0.8244757652282715:  73%|███████▎  | 4696/6470 [1:02:06<18:27,  1.60it/s]
1|4697|Loss: 0.8244757652282715:  73%|███████▎  | 4697/6470 [1:02:07<18:07,  1.63it/s]
1|4698|Loss: 1.0629788637161255:  73%|███████▎  | 4697/6470 [1:02:07<18:07,  1.63it/s]
1|4698|Loss: 1.0629788637161255:  73%|███████▎  | 4698/6470 [1:02:07<18:33,  1.59it/s]
1|4699|Loss: 0.8815245032310486:  73%|███████▎  | 4698/6470 [1:02:08<18:33,  1.59it/s]
1|4699|Loss: 0.8815245032310486:  73%|███████▎  | 4699/6470 [1:02:08<20:25,  1.45it/s]
1|4700|Loss: 1.0105575323104858:  73%|███████▎  | 4699/6470 [1:02:08<20:25,  1.45it/s]
1|4700|Loss: 1.0105575323104858:  73%|███████▎  | 4700/6470 [1:02:09<19:27,  1.52it/s]
1|4701|Loss: 0.623037576675415:  73%|███████▎  | 4700/6470 [1:02:09<19:27,  1.52it/s] 
1|4701|Loss: 0.623037576675415:  73%|███████▎  | 4701/6470 [1:02:10<21:17,  1.39it/s]
1|4702|Loss: 0.969784140586853:  73%|███████▎  | 4701/6470 [1:02:10<21:17,  1.39it/s]
1|4702|Loss: 0.969784140586853:  73%|███████▎  | 4702/6470 [1:02:10<21:18,  1.38it/s]
1|4703|Loss: 0.8785611391067505:  73%|███████▎  | 4702/6470 [1:02:11<21:18,  1.38it/s]
1|4703|Loss: 0.8785611391067505:  73%|███████▎  | 4703/6470 [1:02:11<21:31,  1.37it/s]
1|4704|Loss: 0.8053315877914429:  73%|███████▎  | 4703/6470 [1:02:11<21:31,  1.37it/s]
1|4704|Loss: 0.8053315877914429:  73%|███████▎  | 4704/6470 [1:02:12<21:54,  1.34it/s]
1|4705|Loss: 0.7415048480033875:  73%|███████▎  | 4704/6470 [1:02:12<21:54,  1.34it/s]
1|4705|Loss: 0.7415048480033875:  73%|███████▎  | 4705/6470 [1:02:13<23:16,  1.26it/s]
1|4706|Loss: 0.7316737174987793:  73%|███████▎  | 4705/6470 [1:02:13<23:16,  1.26it/s]
1|4706|Loss: 0.7316737174987793:  73%|███████▎  | 4706/6470 [1:02:13<22:21,  1.31it/s]
1|4707|Loss: 1.011286735534668:  73%|███████▎  | 4706/6470 [1:02:14<22:21,  1.31it/s] 
1|4707|Loss: 1.011286735534668:  73%|███████▎  | 4707/6470 [1:02:14<22:09,  1.33it/s]
1|4708|Loss: 1.168019413948059:  73%|███████▎  | 4707/6470 [1:02:15<22:09,  1.33it/s]
1|4708|Loss: 1.168019413948059:  73%|███████▎  | 4708/6470 [1:02:15<23:03,  1.27it/s]
1|4709|Loss: 0.8503198027610779:  73%|███████▎  | 4708/6470 [1:02:15<23:03,  1.27it/s]
1|4709|Loss: 0.8503198027610779:  73%|███████▎  | 4709/6470 [1:02:16<22:27,  1.31it/s]
1|4710|Loss: 0.7766174077987671:  73%|███████▎  | 4709/6470 [1:02:16<22:27,  1.31it/s]
1|4710|Loss: 0.7766174077987671:  73%|███████▎  | 4710/6470 [1:02:16<20:53,  1.40it/s]
1|4711|Loss: 0.8357046246528625:  73%|███████▎  | 4710/6470 [1:02:17<20:53,  1.40it/s]
1|4711|Loss: 0.8357046246528625:  73%|███████▎  | 4711/6470 [1:02:17<21:04,  1.39it/s]
1|4712|Loss: 0.9060911536216736:  73%|███████▎  | 4711/6470 [1:02:17<21:04,  1.39it/s]
1|4712|Loss: 0.9060911536216736:  73%|███████▎  | 4712/6470 [1:02:18<21:23,  1.37it/s]
1|4713|Loss: 0.9220131635665894:  73%|███████▎  | 4712/6470 [1:02:18<21:23,  1.37it/s]
1|4713|Loss: 0.9220131635665894:  73%|███████▎  | 4713/6470 [1:02:19<22:23,  1.31it/s]
1|4714|Loss: 0.9970057606697083:  73%|███████▎  | 4713/6470 [1:02:19<22:23,  1.31it/s]
1|4714|Loss: 0.9970057606697083:  73%|███████▎  | 4714/6470 [1:02:19<22:10,  1.32it/s]
1|4715|Loss: 0.8584011197090149:  73%|███████▎  | 4714/6470 [1:02:20<22:10,  1.32it/s]
1|4715|Loss: 0.8584011197090149:  73%|███████▎  | 4715/6470 [1:02:20<21:31,  1.36it/s]
1|4716|Loss: 0.862817645072937:  73%|███████▎  | 4715/6470 [1:02:20<21:31,  1.36it/s] 
1|4716|Loss: 0.862817645072937:  73%|███████▎  | 4716/6470 [1:02:21<21:39,  1.35it/s]
1|4717|Loss: 0.8104007244110107:  73%|███████▎  | 4716/6470 [1:02:21<21:39,  1.35it/s]
1|4717|Loss: 0.8104007244110107:  73%|███████▎  | 4717/6470 [1:02:22<22:54,  1.28it/s]
1|4718|Loss: 0.8356910347938538:  73%|███████▎  | 4717/6470 [1:02:22<22:54,  1.28it/s]
1|4718|Loss: 0.8356910347938538:  73%|███████▎  | 4718/6470 [1:02:22<21:28,  1.36it/s]
1|4719|Loss: 0.8799659609794617:  73%|███████▎  | 4718/6470 [1:02:23<21:28,  1.36it/s]
1|4719|Loss: 0.8799659609794617:  73%|███████▎  | 4719/6470 [1:02:23<20:52,  1.40it/s]
1|4720|Loss: 0.7676395177841187:  73%|███████▎  | 4719/6470 [1:02:23<20:52,  1.40it/s]
1|4720|Loss: 0.7676395177841187:  73%|███████▎  | 4720/6470 [1:02:24<20:43,  1.41it/s]
1|4721|Loss: 0.7133724093437195:  73%|███████▎  | 4720/6470 [1:02:24<20:43,  1.41it/s]
1|4721|Loss: 0.7133724093437195:  73%|███████▎  | 4721/6470 [1:02:24<20:51,  1.40it/s]
1|4722|Loss: 0.957784116268158:  73%|███████▎  | 4721/6470 [1:02:25<20:51,  1.40it/s] 
1|4722|Loss: 0.957784116268158:  73%|███████▎  | 4722/6470 [1:02:25<20:51,  1.40it/s]
1|4723|Loss: 0.8967645764350891:  73%|███████▎  | 4722/6470 [1:02:25<20:51,  1.40it/s]
1|4723|Loss: 0.8967645764350891:  73%|███████▎  | 4723/6470 [1:02:26<20:17,  1.43it/s]
1|4724|Loss: 0.7980889081954956:  73%|███████▎  | 4723/6470 [1:02:26<20:17,  1.43it/s]
1|4724|Loss: 0.7980889081954956:  73%|███████▎  | 4724/6470 [1:02:26<19:00,  1.53it/s]
1|4725|Loss: 0.9459694027900696:  73%|███████▎  | 4724/6470 [1:02:27<19:00,  1.53it/s]
1|4725|Loss: 0.9459694027900696:  73%|███████▎  | 4725/6470 [1:02:27<17:03,  1.70it/s]
1|4726|Loss: 0.9307056069374084:  73%|███████▎  | 4725/6470 [1:02:27<17:03,  1.70it/s]
1|4726|Loss: 0.9307056069374084:  73%|███████▎  | 4726/6470 [1:02:27<16:42,  1.74it/s]
1|4727|Loss: 0.9377790093421936:  73%|███████▎  | 4726/6470 [1:02:28<16:42,  1.74it/s]
1|4727|Loss: 0.9377790093421936:  73%|███████▎  | 4727/6470 [1:02:28<16:01,  1.81it/s]
1|4728|Loss: 0.9163475632667542:  73%|███████▎  | 4727/6470 [1:02:28<16:01,  1.81it/s]
1|4728|Loss: 0.9163475632667542:  73%|███████▎  | 4728/6470 [1:02:28<17:38,  1.65it/s]
1|4729|Loss: 0.98381507396698:  73%|███████▎  | 4728/6470 [1:02:29<17:38,  1.65it/s]  
1|4729|Loss: 0.98381507396698:  73%|███████▎  | 4729/6470 [1:02:29<19:17,  1.50it/s]
1|4730|Loss: 0.967950701713562:  73%|███████▎  | 4729/6470 [1:02:30<19:17,  1.50it/s]
1|4730|Loss: 0.967950701713562:  73%|███████▎  | 4730/6470 [1:02:30<19:35,  1.48it/s]
1|4731|Loss: 0.5525951385498047:  73%|███████▎  | 4730/6470 [1:02:30<19:35,  1.48it/s]
1|4731|Loss: 0.5525951385498047:  73%|███████▎  | 4731/6470 [1:02:31<21:45,  1.33it/s]
1|4732|Loss: 0.809691846370697:  73%|███████▎  | 4731/6470 [1:02:31<21:45,  1.33it/s] 
1|4732|Loss: 0.809691846370697:  73%|███████▎  | 4732/6470 [1:02:31<19:40,  1.47it/s]
1|4733|Loss: 0.8167073726654053:  73%|███████▎  | 4732/6470 [1:02:32<19:40,  1.47it/s]
1|4733|Loss: 0.8167073726654053:  73%|███████▎  | 4733/6470 [1:02:32<20:05,  1.44it/s]
1|4734|Loss: 1.0423147678375244:  73%|███████▎  | 4733/6470 [1:02:33<20:05,  1.44it/s]
1|4734|Loss: 1.0423147678375244:  73%|███████▎  | 4734/6470 [1:02:33<21:59,  1.32it/s]
1|4735|Loss: 0.909006655216217:  73%|███████▎  | 4734/6470 [1:02:34<21:59,  1.32it/s] 
1|4735|Loss: 0.909006655216217:  73%|███████▎  | 4735/6470 [1:02:34<22:23,  1.29it/s]
1|4736|Loss: 0.864820122718811:  73%|███████▎  | 4735/6470 [1:02:34<22:23,  1.29it/s]
1|4736|Loss: 0.864820122718811:  73%|███████▎  | 4736/6470 [1:02:35<21:37,  1.34it/s]
1|4737|Loss: 0.9163773059844971:  73%|███████▎  | 4736/6470 [1:02:35<21:37,  1.34it/s]
1|4737|Loss: 0.9163773059844971:  73%|███████▎  | 4737/6470 [1:02:35<18:56,  1.52it/s]
1|4738|Loss: 1.1691914796829224:  73%|███████▎  | 4737/6470 [1:02:35<18:56,  1.52it/s]
1|4738|Loss: 1.1691914796829224:  73%|███████▎  | 4738/6470 [1:02:36<19:38,  1.47it/s]
1|4739|Loss: 1.0173453092575073:  73%|███████▎  | 4738/6470 [1:02:36<19:38,  1.47it/s]
1|4739|Loss: 1.0173453092575073:  73%|███████▎  | 4739/6470 [1:02:36<19:28,  1.48it/s]
1|4740|Loss: 0.7874479293823242:  73%|███████▎  | 4739/6470 [1:02:37<19:28,  1.48it/s]
1|4740|Loss: 0.7874479293823242:  73%|███████▎  | 4740/6470 [1:02:37<19:27,  1.48it/s]
1|4741|Loss: 0.9335674047470093:  73%|███████▎  | 4740/6470 [1:02:38<19:27,  1.48it/s]
1|4741|Loss: 0.9335674047470093:  73%|███████▎  | 4741/6470 [1:02:38<20:16,  1.42it/s]
1|4742|Loss: 0.8524072170257568:  73%|███████▎  | 4741/6470 [1:02:38<20:16,  1.42it/s]
1|4742|Loss: 0.8524072170257568:  73%|███████▎  | 4742/6470 [1:02:39<20:23,  1.41it/s]
1|4743|Loss: 0.8441609740257263:  73%|███████▎  | 4742/6470 [1:02:39<20:23,  1.41it/s]
1|4743|Loss: 0.8441609740257263:  73%|███████▎  | 4743/6470 [1:02:39<19:09,  1.50it/s]
1|4744|Loss: 0.9865898489952087:  73%|███████▎  | 4743/6470 [1:02:40<19:09,  1.50it/s]
1|4744|Loss: 0.9865898489952087:  73%|███████▎  | 4744/6470 [1:02:40<19:34,  1.47it/s]
1|4745|Loss: 0.7884328961372375:  73%|███████▎  | 4744/6470 [1:02:40<19:34,  1.47it/s]
1|4745|Loss: 0.7884328961372375:  73%|███████▎  | 4745/6470 [1:02:41<20:18,  1.42it/s]
1|4746|Loss: 0.9246227145195007:  73%|███████▎  | 4745/6470 [1:02:41<20:18,  1.42it/s]
1|4746|Loss: 0.9246227145195007:  73%|███████▎  | 4746/6470 [1:02:42<22:05,  1.30it/s]
1|4747|Loss: 0.7916280627250671:  73%|███████▎  | 4746/6470 [1:02:42<22:05,  1.30it/s]
1|4747|Loss: 0.7916280627250671:  73%|███████▎  | 4747/6470 [1:02:42<23:05,  1.24it/s]
1|4748|Loss: 0.9318082928657532:  73%|███████▎  | 4747/6470 [1:02:43<23:05,  1.24it/s]
1|4748|Loss: 0.9318082928657532:  73%|███████▎  | 4748/6470 [1:02:43<22:54,  1.25it/s]
1|4749|Loss: 0.763028621673584:  73%|███████▎  | 4748/6470 [1:02:44<22:54,  1.25it/s] 
1|4749|Loss: 0.763028621673584:  73%|███████▎  | 4749/6470 [1:02:44<20:06,  1.43it/s]
1|4750|Loss: 0.921635627746582:  73%|███████▎  | 4749/6470 [1:02:44<20:06,  1.43it/s]
1|4750|Loss: 0.921635627746582:  73%|███████▎  | 4750/6470 [1:02:44<19:40,  1.46it/s]
1|4751|Loss: 0.8275659084320068:  73%|███████▎  | 4750/6470 [1:02:45<19:40,  1.46it/s]
1|4751|Loss: 0.8275659084320068:  73%|███████▎  | 4751/6470 [1:02:45<18:44,  1.53it/s]
1|4752|Loss: 0.8354511857032776:  73%|███████▎  | 4751/6470 [1:02:45<18:44,  1.53it/s]
1|4752|Loss: 0.8354511857032776:  73%|███████▎  | 4752/6470 [1:02:45<17:20,  1.65it/s]
1|4753|Loss: 0.8010510206222534:  73%|███████▎  | 4752/6470 [1:02:46<17:20,  1.65it/s]
1|4753|Loss: 0.8010510206222534:  73%|███████▎  | 4753/6470 [1:02:46<16:58,  1.69it/s]
1|4754|Loss: 0.8805860280990601:  73%|███████▎  | 4753/6470 [1:02:46<16:58,  1.69it/s]
1|4754|Loss: 0.8805860280990601:  73%|███████▎  | 4754/6470 [1:02:47<19:17,  1.48it/s]
1|4755|Loss: 0.7534781694412231:  73%|███████▎  | 4754/6470 [1:02:47<19:17,  1.48it/s]
1|4755|Loss: 0.7534781694412231:  73%|███████▎  | 4755/6470 [1:02:48<19:51,  1.44it/s]
1|4756|Loss: 0.9659939408302307:  73%|███████▎  | 4755/6470 [1:02:48<19:51,  1.44it/s]
1|4756|Loss: 0.9659939408302307:  74%|███████▎  | 4756/6470 [1:02:48<21:20,  1.34it/s]
1|4757|Loss: 0.9876418113708496:  74%|███████▎  | 4756/6470 [1:02:49<21:20,  1.34it/s]
1|4757|Loss: 0.9876418113708496:  74%|███████▎  | 4757/6470 [1:02:49<20:45,  1.37it/s]
1|4758|Loss: 0.818640947341919:  74%|███████▎  | 4757/6470 [1:02:50<20:45,  1.37it/s] 
1|4758|Loss: 0.818640947341919:  74%|███████▎  | 4758/6470 [1:02:50<21:32,  1.32it/s]
1|4759|Loss: 0.691762387752533:  74%|███████▎  | 4758/6470 [1:02:50<21:32,  1.32it/s]
1|4759|Loss: 0.691762387752533:  74%|███████▎  | 4759/6470 [1:02:51<21:24,  1.33it/s]
1|4760|Loss: 0.8435907959938049:  74%|███████▎  | 4759/6470 [1:02:51<21:24,  1.33it/s]
1|4760|Loss: 0.8435907959938049:  74%|███████▎  | 4760/6470 [1:02:52<22:13,  1.28it/s]
1|4761|Loss: 0.7960374355316162:  74%|███████▎  | 4760/6470 [1:02:52<22:13,  1.28it/s]
1|4761|Loss: 0.7960374355316162:  74%|███████▎  | 4761/6470 [1:02:52<21:52,  1.30it/s]
1|4762|Loss: 0.9665193557739258:  74%|███████▎  | 4761/6470 [1:02:53<21:52,  1.30it/s]
1|4762|Loss: 0.9665193557739258:  74%|███████▎  | 4762/6470 [1:02:53<21:01,  1.35it/s]
1|4763|Loss: 0.7219425439834595:  74%|███████▎  | 4762/6470 [1:02:53<21:01,  1.35it/s]
1|4763|Loss: 0.7219425439834595:  74%|███████▎  | 4763/6470 [1:02:54<20:08,  1.41it/s]
1|4764|Loss: 0.8517963886260986:  74%|███████▎  | 4763/6470 [1:02:54<20:08,  1.41it/s]
1|4764|Loss: 0.8517963886260986:  74%|███████▎  | 4764/6470 [1:02:54<19:45,  1.44it/s]
1|4765|Loss: 0.9301342964172363:  74%|███████▎  | 4764/6470 [1:02:55<19:45,  1.44it/s]
1|4765|Loss: 0.9301342964172363:  74%|███████▎  | 4765/6470 [1:02:55<21:24,  1.33it/s]
1|4766|Loss: 0.7506961822509766:  74%|███████▎  | 4765/6470 [1:02:56<21:24,  1.33it/s]
1|4766|Loss: 0.7506961822509766:  74%|███████▎  | 4766/6470 [1:02:56<22:10,  1.28it/s]
1|4767|Loss: 0.9121041893959045:  74%|███████▎  | 4766/6470 [1:02:56<22:10,  1.28it/s]
1|4767|Loss: 0.9121041893959045:  74%|███████▎  | 4767/6470 [1:02:57<23:17,  1.22it/s]
1|4768|Loss: 0.7852881550788879:  74%|███████▎  | 4767/6470 [1:02:57<23:17,  1.22it/s]
1|4768|Loss: 0.7852881550788879:  74%|███████▎  | 4768/6470 [1:02:58<24:07,  1.18it/s]
1|4769|Loss: 0.8658342957496643:  74%|███████▎  | 4768/6470 [1:02:58<24:07,  1.18it/s]
1|4769|Loss: 0.8658342957496643:  74%|███████▎  | 4769/6470 [1:02:59<24:50,  1.14it/s]
1|4770|Loss: 0.9620676040649414:  74%|███████▎  | 4769/6470 [1:02:59<24:50,  1.14it/s]
1|4770|Loss: 0.9620676040649414:  74%|███████▎  | 4770/6470 [1:03:00<24:32,  1.15it/s]
1|4771|Loss: 0.9965639114379883:  74%|███████▎  | 4770/6470 [1:03:00<24:32,  1.15it/s]
1|4771|Loss: 0.9965639114379883:  74%|███████▎  | 4771/6470 [1:03:00<24:48,  1.14it/s]
1|4772|Loss: 0.9271113276481628:  74%|███████▎  | 4771/6470 [1:03:01<24:48,  1.14it/s]
1|4772|Loss: 0.9271113276481628:  74%|███████▍  | 4772/6470 [1:03:01<24:56,  1.13it/s]
1|4773|Loss: 0.8922519683837891:  74%|███████▍  | 4772/6470 [1:03:02<24:56,  1.13it/s]
1|4773|Loss: 0.8922519683837891:  74%|███████▍  | 4773/6470 [1:03:02<24:21,  1.16it/s]
1|4774|Loss: 0.8246642351150513:  74%|███████▍  | 4773/6470 [1:03:03<24:21,  1.16it/s]
1|4774|Loss: 0.8246642351150513:  74%|███████▍  | 4774/6470 [1:03:03<22:04,  1.28it/s]
1|4775|Loss: 0.7737565040588379:  74%|███████▍  | 4774/6470 [1:03:03<22:04,  1.28it/s]
1|4775|Loss: 0.7737565040588379:  74%|███████▍  | 4775/6470 [1:03:03<21:01,  1.34it/s]
1|4776|Loss: 0.9279261827468872:  74%|███████▍  | 4775/6470 [1:03:04<21:01,  1.34it/s]
1|4776|Loss: 0.9279261827468872:  74%|███████▍  | 4776/6470 [1:03:04<21:28,  1.31it/s]
1|4777|Loss: 0.7258335947990417:  74%|███████▍  | 4776/6470 [1:03:05<21:28,  1.31it/s]
1|4777|Loss: 0.7258335947990417:  74%|███████▍  | 4777/6470 [1:03:05<21:51,  1.29it/s]
1|4778|Loss: 0.9623145461082458:  74%|███████▍  | 4777/6470 [1:03:05<21:51,  1.29it/s]
1|4778|Loss: 0.9623145461082458:  74%|███████▍  | 4778/6470 [1:03:06<21:01,  1.34it/s]
1|4779|Loss: 0.875508189201355:  74%|███████▍  | 4778/6470 [1:03:06<21:01,  1.34it/s] 
1|4779|Loss: 0.875508189201355:  74%|███████▍  | 4779/6470 [1:03:07<22:02,  1.28it/s]
1|4780|Loss: 0.8793631196022034:  74%|███████▍  | 4779/6470 [1:03:07<22:02,  1.28it/s]
1|4780|Loss: 0.8793631196022034:  74%|███████▍  | 4780/6470 [1:03:07<20:57,  1.34it/s]
1|4781|Loss: 0.7952059507369995:  74%|███████▍  | 4780/6470 [1:03:08<20:57,  1.34it/s]
1|4781|Loss: 0.7952059507369995:  74%|███████▍  | 4781/6470 [1:03:08<21:01,  1.34it/s]
1|4782|Loss: 0.9210051894187927:  74%|███████▍  | 4781/6470 [1:03:08<21:01,  1.34it/s]
1|4782|Loss: 0.9210051894187927:  74%|███████▍  | 4782/6470 [1:03:09<19:42,  1.43it/s]
1|4783|Loss: 0.869053840637207:  74%|███████▍  | 4782/6470 [1:03:09<19:42,  1.43it/s] 
1|4783|Loss: 0.869053840637207:  74%|███████▍  | 4783/6470 [1:03:09<18:08,  1.55it/s]
1|4784|Loss: 0.9580467939376831:  74%|███████▍  | 4783/6470 [1:03:09<18:08,  1.55it/s]
1|4784|Loss: 0.9580467939376831:  74%|███████▍  | 4784/6470 [1:03:10<18:40,  1.50it/s]
1|4785|Loss: 0.8671725392341614:  74%|███████▍  | 4784/6470 [1:03:10<18:40,  1.50it/s]
1|4785|Loss: 0.8671725392341614:  74%|███████▍  | 4785/6470 [1:03:11<18:53,  1.49it/s]
1|4786|Loss: 0.9145443439483643:  74%|███████▍  | 4785/6470 [1:03:11<18:53,  1.49it/s]
1|4786|Loss: 0.9145443439483643:  74%|███████▍  | 4786/6470 [1:03:11<19:02,  1.47it/s]
1|4787|Loss: 0.8903305530548096:  74%|███████▍  | 4786/6470 [1:03:12<19:02,  1.47it/s]
1|4787|Loss: 0.8903305530548096:  74%|███████▍  | 4787/6470 [1:03:12<19:37,  1.43it/s]
1|4788|Loss: 0.8756012916564941:  74%|███████▍  | 4787/6470 [1:03:12<19:37,  1.43it/s]
1|4788|Loss: 0.8756012916564941:  74%|███████▍  | 4788/6470 [1:03:13<20:23,  1.37it/s]
1|4789|Loss: 0.9997808337211609:  74%|███████▍  | 4788/6470 [1:03:13<20:23,  1.37it/s]
1|4789|Loss: 0.9997808337211609:  74%|███████▍  | 4789/6470 [1:03:13<20:20,  1.38it/s]
1|4790|Loss: 1.006955862045288:  74%|███████▍  | 4789/6470 [1:03:14<20:20,  1.38it/s] 
1|4790|Loss: 1.006955862045288:  74%|███████▍  | 4790/6470 [1:03:14<20:31,  1.36it/s]
1|4791|Loss: 1.0267212390899658:  74%|███████▍  | 4790/6470 [1:03:15<20:31,  1.36it/s]
1|4791|Loss: 1.0267212390899658:  74%|███████▍  | 4791/6470 [1:03:15<21:10,  1.32it/s]
1|4792|Loss: 0.7622640132904053:  74%|███████▍  | 4791/6470 [1:03:15<21:10,  1.32it/s]
1|4792|Loss: 0.7622640132904053:  74%|███████▍  | 4792/6470 [1:03:16<19:31,  1.43it/s]
1|4793|Loss: 0.9094119668006897:  74%|███████▍  | 4792/6470 [1:03:16<19:31,  1.43it/s]
1|4793|Loss: 0.9094119668006897:  74%|███████▍  | 4793/6470 [1:03:16<20:34,  1.36it/s]
1|4794|Loss: 0.8481568694114685:  74%|███████▍  | 4793/6470 [1:03:17<20:34,  1.36it/s]
1|4794|Loss: 0.8481568694114685:  74%|███████▍  | 4794/6470 [1:03:17<21:13,  1.32it/s]
1|4795|Loss: 0.9219872951507568:  74%|███████▍  | 4794/6470 [1:03:18<21:13,  1.32it/s]
1|4795|Loss: 0.9219872951507568:  74%|███████▍  | 4795/6470 [1:03:18<19:53,  1.40it/s]
1|4796|Loss: 0.8981173038482666:  74%|███████▍  | 4795/6470 [1:03:18<19:53,  1.40it/s]
1|4796|Loss: 0.8981173038482666:  74%|███████▍  | 4796/6470 [1:03:18<19:00,  1.47it/s]
1|4797|Loss: 0.664757490158081:  74%|███████▍  | 4796/6470 [1:03:19<19:00,  1.47it/s] 
1|4797|Loss: 0.664757490158081:  74%|███████▍  | 4797/6470 [1:03:19<21:32,  1.29it/s]
1|4798|Loss: 0.8089836835861206:  74%|███████▍  | 4797/6470 [1:03:20<21:32,  1.29it/s]
1|4798|Loss: 0.8089836835861206:  74%|███████▍  | 4798/6470 [1:03:20<22:11,  1.26it/s]
1|4799|Loss: 0.8934476375579834:  74%|███████▍  | 4798/6470 [1:03:21<22:11,  1.26it/s]
1|4799|Loss: 0.8934476375579834:  74%|███████▍  | 4799/6470 [1:03:21<23:04,  1.21it/s]
1|4800|Loss: 0.9703327417373657:  74%|███████▍  | 4799/6470 [1:03:22<23:04,  1.21it/s]
1|4800|Loss: 0.9703327417373657:  74%|███████▍  | 4800/6470 [1:03:22<22:50,  1.22it/s]
1|4801|Loss: 0.9353123307228088:  74%|███████▍  | 4800/6470 [1:03:22<22:50,  1.22it/s]
1|4801|Loss: 0.9353123307228088:  74%|███████▍  | 4801/6470 [1:03:23<22:21,  1.24it/s]
1|4802|Loss: 0.9268773794174194:  74%|███████▍  | 4801/6470 [1:03:23<22:21,  1.24it/s]
1|4802|Loss: 0.9268773794174194:  74%|███████▍  | 4802/6470 [1:03:24<23:05,  1.20it/s]
1|4803|Loss: 1.0188970565795898:  74%|███████▍  | 4802/6470 [1:03:24<23:05,  1.20it/s]
1|4803|Loss: 1.0188970565795898:  74%|███████▍  | 4803/6470 [1:03:25<24:01,  1.16it/s]
1|4804|Loss: 0.8998173475265503:  74%|███████▍  | 4803/6470 [1:03:25<24:01,  1.16it/s]
1|4804|Loss: 0.8998173475265503:  74%|███████▍  | 4804/6470 [1:03:25<23:15,  1.19it/s]
1|4805|Loss: 0.9137745499610901:  74%|███████▍  | 4804/6470 [1:03:26<23:15,  1.19it/s]
1|4805|Loss: 0.9137745499610901:  74%|███████▍  | 4805/6470 [1:03:26<23:47,  1.17it/s]
1|4806|Loss: 0.8477553725242615:  74%|███████▍  | 4805/6470 [1:03:27<23:47,  1.17it/s]
1|4806|Loss: 0.8477553725242615:  74%|███████▍  | 4806/6470 [1:03:27<22:52,  1.21it/s]
1|4807|Loss: 0.7929492592811584:  74%|███████▍  | 4806/6470 [1:03:27<22:52,  1.21it/s]
1|4807|Loss: 0.7929492592811584:  74%|███████▍  | 4807/6470 [1:03:28<21:51,  1.27it/s]
1|4808|Loss: 0.7410060167312622:  74%|███████▍  | 4807/6470 [1:03:28<21:51,  1.27it/s]
1|4808|Loss: 0.7410060167312622:  74%|███████▍  | 4808/6470 [1:03:28<20:34,  1.35it/s]
1|4809|Loss: 0.7130149602890015:  74%|███████▍  | 4808/6470 [1:03:29<20:34,  1.35it/s]
1|4809|Loss: 0.7130149602890015:  74%|███████▍  | 4809/6470 [1:03:29<20:01,  1.38it/s]
1|4810|Loss: 0.8202757835388184:  74%|███████▍  | 4809/6470 [1:03:29<20:01,  1.38it/s]
1|4810|Loss: 0.8202757835388184:  74%|███████▍  | 4810/6470 [1:03:30<18:46,  1.47it/s]
1|4811|Loss: 0.8261769413948059:  74%|███████▍  | 4810/6470 [1:03:30<18:46,  1.47it/s]
1|4811|Loss: 0.8261769413948059:  74%|███████▍  | 4811/6470 [1:03:30<19:09,  1.44it/s]
1|4812|Loss: 0.9490304589271545:  74%|███████▍  | 4811/6470 [1:03:31<19:09,  1.44it/s]
1|4812|Loss: 0.9490304589271545:  74%|███████▍  | 4812/6470 [1:03:31<19:54,  1.39it/s]
1|4813|Loss: 0.9197131395339966:  74%|███████▍  | 4812/6470 [1:03:32<19:54,  1.39it/s]
1|4813|Loss: 0.9197131395339966:  74%|███████▍  | 4813/6470 [1:03:32<20:21,  1.36it/s]
1|4814|Loss: 0.858514666557312:  74%|███████▍  | 4813/6470 [1:03:32<20:21,  1.36it/s] 
1|4814|Loss: 0.858514666557312:  74%|███████▍  | 4814/6470 [1:03:33<20:33,  1.34it/s]
1|4815|Loss: 0.7331864237785339:  74%|███████▍  | 4814/6470 [1:03:33<20:33,  1.34it/s]
1|4815|Loss: 0.7331864237785339:  74%|███████▍  | 4815/6470 [1:03:33<21:01,  1.31it/s]
1|4816|Loss: 0.8003904223442078:  74%|███████▍  | 4815/6470 [1:03:34<21:01,  1.31it/s]
1|4816|Loss: 0.8003904223442078:  74%|███████▍  | 4816/6470 [1:03:34<21:30,  1.28it/s]
1|4817|Loss: 1.032602310180664:  74%|███████▍  | 4816/6470 [1:03:35<21:30,  1.28it/s] 
1|4817|Loss: 1.032602310180664:  74%|███████▍  | 4817/6470 [1:03:35<21:27,  1.28it/s]
1|4818|Loss: 0.8268275260925293:  74%|███████▍  | 4817/6470 [1:03:35<21:27,  1.28it/s]
1|4818|Loss: 0.8268275260925293:  74%|███████▍  | 4818/6470 [1:03:36<20:42,  1.33it/s]
1|4819|Loss: 0.9089710116386414:  74%|███████▍  | 4818/6470 [1:03:36<20:42,  1.33it/s]
1|4819|Loss: 0.9089710116386414:  74%|███████▍  | 4819/6470 [1:03:37<21:00,  1.31it/s]
1|4820|Loss: 0.92879319190979:  74%|███████▍  | 4819/6470 [1:03:37<21:00,  1.31it/s]  
1|4820|Loss: 0.92879319190979:  74%|███████▍  | 4820/6470 [1:03:37<20:13,  1.36it/s]
1|4821|Loss: 0.9241719245910645:  74%|███████▍  | 4820/6470 [1:03:38<20:13,  1.36it/s]
1|4821|Loss: 0.9241719245910645:  75%|███████▍  | 4821/6470 [1:03:38<18:42,  1.47it/s]
1|4822|Loss: 1.0741310119628906:  75%|███████▍  | 4821/6470 [1:03:38<18:42,  1.47it/s]
1|4822|Loss: 1.0741310119628906:  75%|███████▍  | 4822/6470 [1:03:39<19:27,  1.41it/s]
1|4823|Loss: 0.8284844756126404:  75%|███████▍  | 4822/6470 [1:03:39<19:27,  1.41it/s]
1|4823|Loss: 0.8284844756126404:  75%|███████▍  | 4823/6470 [1:03:39<19:41,  1.39it/s]
1|4824|Loss: 0.9914674162864685:  75%|███████▍  | 4823/6470 [1:03:40<19:41,  1.39it/s]
1|4824|Loss: 0.9914674162864685:  75%|███████▍  | 4824/6470 [1:03:40<18:51,  1.45it/s]
1|4825|Loss: 0.8966578841209412:  75%|███████▍  | 4824/6470 [1:03:40<18:51,  1.45it/s]
1|4825|Loss: 0.8966578841209412:  75%|███████▍  | 4825/6470 [1:03:41<19:27,  1.41it/s]
1|4826|Loss: 0.8448653221130371:  75%|███████▍  | 4825/6470 [1:03:41<19:27,  1.41it/s]
1|4826|Loss: 0.8448653221130371:  75%|███████▍  | 4826/6470 [1:03:41<20:20,  1.35it/s]
1|4827|Loss: 0.8405113816261292:  75%|███████▍  | 4826/6470 [1:03:42<20:20,  1.35it/s]
1|4827|Loss: 0.8405113816261292:  75%|███████▍  | 4827/6470 [1:03:42<20:35,  1.33it/s]
1|4828|Loss: 0.9969813823699951:  75%|███████▍  | 4827/6470 [1:03:43<20:35,  1.33it/s]
1|4828|Loss: 0.9969813823699951:  75%|███████▍  | 4828/6470 [1:03:43<21:47,  1.26it/s]
1|4829|Loss: 0.895618736743927:  75%|███████▍  | 4828/6470 [1:03:44<21:47,  1.26it/s] 
1|4829|Loss: 0.895618736743927:  75%|███████▍  | 4829/6470 [1:03:44<22:37,  1.21it/s]
1|4830|Loss: 0.8785750865936279:  75%|███████▍  | 4829/6470 [1:03:45<22:37,  1.21it/s]
1|4830|Loss: 0.8785750865936279:  75%|███████▍  | 4830/6470 [1:03:45<23:03,  1.19it/s]
1|4831|Loss: 0.7763929963111877:  75%|███████▍  | 4830/6470 [1:03:45<23:03,  1.19it/s]
1|4831|Loss: 0.7763929963111877:  75%|███████▍  | 4831/6470 [1:03:46<21:59,  1.24it/s]
1|4832|Loss: 0.9532941579818726:  75%|███████▍  | 4831/6470 [1:03:46<21:59,  1.24it/s]
1|4832|Loss: 0.9532941579818726:  75%|███████▍  | 4832/6470 [1:03:46<21:48,  1.25it/s]
1|4833|Loss: 0.7434414029121399:  75%|███████▍  | 4832/6470 [1:03:47<21:48,  1.25it/s]
1|4833|Loss: 0.7434414029121399:  75%|███████▍  | 4833/6470 [1:03:47<19:50,  1.38it/s]
1|4834|Loss: 0.9013895392417908:  75%|███████▍  | 4833/6470 [1:03:47<19:50,  1.38it/s]
1|4834|Loss: 0.9013895392417908:  75%|███████▍  | 4834/6470 [1:03:48<18:39,  1.46it/s]
1|4835|Loss: 0.8438463807106018:  75%|███████▍  | 4834/6470 [1:03:48<18:39,  1.46it/s]
1|4835|Loss: 0.8438463807106018:  75%|███████▍  | 4835/6470 [1:03:48<17:15,  1.58it/s]
1|4836|Loss: 0.9411871433258057:  75%|███████▍  | 4835/6470 [1:03:49<17:15,  1.58it/s]
1|4836|Loss: 0.9411871433258057:  75%|███████▍  | 4836/6470 [1:03:49<19:27,  1.40it/s]
1|4837|Loss: 0.7331146597862244:  75%|███████▍  | 4836/6470 [1:03:49<19:27,  1.40it/s]
1|4837|Loss: 0.7331146597862244:  75%|███████▍  | 4837/6470 [1:03:50<18:58,  1.43it/s]
1|4838|Loss: 0.7546744346618652:  75%|███████▍  | 4837/6470 [1:03:50<18:58,  1.43it/s]
1|4838|Loss: 0.7546744346618652:  75%|███████▍  | 4838/6470 [1:03:50<19:20,  1.41it/s]
1|4839|Loss: 0.9584754705429077:  75%|███████▍  | 4838/6470 [1:03:51<19:20,  1.41it/s]
1|4839|Loss: 0.9584754705429077:  75%|███████▍  | 4839/6470 [1:03:51<20:30,  1.33it/s]
1|4840|Loss: 0.831722617149353:  75%|███████▍  | 4839/6470 [1:03:52<20:30,  1.33it/s] 
1|4840|Loss: 0.831722617149353:  75%|███████▍  | 4840/6470 [1:03:52<21:01,  1.29it/s]
1|4841|Loss: 0.8944880962371826:  75%|███████▍  | 4840/6470 [1:03:52<21:01,  1.29it/s]
1|4841|Loss: 0.8944880962371826:  75%|███████▍  | 4841/6470 [1:03:53<20:05,  1.35it/s]
1|4842|Loss: 0.8513412475585938:  75%|███████▍  | 4841/6470 [1:03:53<20:05,  1.35it/s]
1|4842|Loss: 0.8513412475585938:  75%|███████▍  | 4842/6470 [1:03:53<19:14,  1.41it/s]
1|4843|Loss: 0.8997793793678284:  75%|███████▍  | 4842/6470 [1:03:54<19:14,  1.41it/s]
1|4843|Loss: 0.8997793793678284:  75%|███████▍  | 4843/6470 [1:03:54<18:11,  1.49it/s]
1|4844|Loss: 0.9354234933853149:  75%|███████▍  | 4843/6470 [1:03:54<18:11,  1.49it/s]
1|4844|Loss: 0.9354234933853149:  75%|███████▍  | 4844/6470 [1:03:55<19:19,  1.40it/s]
1|4845|Loss: 0.8530158996582031:  75%|███████▍  | 4844/6470 [1:03:55<19:19,  1.40it/s]
1|4845|Loss: 0.8530158996582031:  75%|███████▍  | 4845/6470 [1:03:56<20:31,  1.32it/s]
1|4846|Loss: 0.9006532430648804:  75%|███████▍  | 4845/6470 [1:03:56<20:31,  1.32it/s]
1|4846|Loss: 0.9006532430648804:  75%|███████▍  | 4846/6470 [1:03:57<21:26,  1.26it/s]
1|4847|Loss: 0.9715957641601562:  75%|███████▍  | 4846/6470 [1:03:57<21:26,  1.26it/s]
1|4847|Loss: 0.9715957641601562:  75%|███████▍  | 4847/6470 [1:03:57<20:46,  1.30it/s]
1|4848|Loss: 0.7898637056350708:  75%|███████▍  | 4847/6470 [1:03:58<20:46,  1.30it/s]
1|4848|Loss: 0.7898637056350708:  75%|███████▍  | 4848/6470 [1:03:58<20:37,  1.31it/s]
1|4849|Loss: 1.0151219367980957:  75%|███████▍  | 4848/6470 [1:03:58<20:37,  1.31it/s]
1|4849|Loss: 1.0151219367980957:  75%|███████▍  | 4849/6470 [1:03:59<20:53,  1.29it/s]
1|4850|Loss: 0.9505484700202942:  75%|███████▍  | 4849/6470 [1:03:59<20:53,  1.29it/s]
1|4850|Loss: 0.9505484700202942:  75%|███████▍  | 4850/6470 [1:04:00<21:04,  1.28it/s]
1|4851|Loss: 0.8239732980728149:  75%|███████▍  | 4850/6470 [1:04:00<21:04,  1.28it/s]
1|4851|Loss: 0.8239732980728149:  75%|███████▍  | 4851/6470 [1:04:00<19:18,  1.40it/s]
1|4852|Loss: 0.8206872344017029:  75%|███████▍  | 4851/6470 [1:04:01<19:18,  1.40it/s]
1|4852|Loss: 0.8206872344017029:  75%|███████▍  | 4852/6470 [1:04:01<20:12,  1.33it/s]
1|4853|Loss: 0.8251014947891235:  75%|███████▍  | 4852/6470 [1:04:01<20:12,  1.33it/s]
1|4853|Loss: 0.8251014947891235:  75%|███████▌  | 4853/6470 [1:04:02<20:51,  1.29it/s]
1|4854|Loss: 0.8234773874282837:  75%|███████▌  | 4853/6470 [1:04:02<20:51,  1.29it/s]
1|4854|Loss: 0.8234773874282837:  75%|███████▌  | 4854/6470 [1:04:02<20:21,  1.32it/s]
1|4855|Loss: 1.0739703178405762:  75%|███████▌  | 4854/6470 [1:04:03<20:21,  1.32it/s]
1|4855|Loss: 1.0739703178405762:  75%|███████▌  | 4855/6470 [1:04:03<21:17,  1.26it/s]
1|4856|Loss: 0.9509574174880981:  75%|███████▌  | 4855/6470 [1:04:04<21:17,  1.26it/s]
1|4856|Loss: 0.9509574174880981:  75%|███████▌  | 4856/6470 [1:04:04<21:22,  1.26it/s]
1|4857|Loss: 0.8442111015319824:  75%|███████▌  | 4856/6470 [1:04:05<21:22,  1.26it/s]
1|4857|Loss: 0.8442111015319824:  75%|███████▌  | 4857/6470 [1:04:05<20:48,  1.29it/s]
1|4858|Loss: 0.7625715136528015:  75%|███████▌  | 4857/6470 [1:04:05<20:48,  1.29it/s]
1|4858|Loss: 0.7625715136528015:  75%|███████▌  | 4858/6470 [1:04:06<20:18,  1.32it/s]
1|4859|Loss: 0.7989235520362854:  75%|███████▌  | 4858/6470 [1:04:06<20:18,  1.32it/s]
1|4859|Loss: 0.7989235520362854:  75%|███████▌  | 4859/6470 [1:04:06<19:23,  1.38it/s]
1|4860|Loss: 0.9677790403366089:  75%|███████▌  | 4859/6470 [1:04:07<19:23,  1.38it/s]
1|4860|Loss: 0.9677790403366089:  75%|███████▌  | 4860/6470 [1:04:07<19:36,  1.37it/s]
1|4861|Loss: 0.7796117067337036:  75%|███████▌  | 4860/6470 [1:04:08<19:36,  1.37it/s]
1|4861|Loss: 0.7796117067337036:  75%|███████▌  | 4861/6470 [1:04:08<21:43,  1.23it/s]
1|4862|Loss: 0.6398534178733826:  75%|███████▌  | 4861/6470 [1:04:09<21:43,  1.23it/s]
1|4862|Loss: 0.6398534178733826:  75%|███████▌  | 4862/6470 [1:04:09<22:07,  1.21it/s]
1|4863|Loss: 0.8728083968162537:  75%|███████▌  | 4862/6470 [1:04:09<22:07,  1.21it/s]
1|4863|Loss: 0.8728083968162537:  75%|███████▌  | 4863/6470 [1:04:10<21:16,  1.26it/s]
1|4864|Loss: 1.069100260734558:  75%|███████▌  | 4863/6470 [1:04:10<21:16,  1.26it/s] 
1|4864|Loss: 1.069100260734558:  75%|███████▌  | 4864/6470 [1:04:10<20:16,  1.32it/s]
1|4865|Loss: 0.8725147247314453:  75%|███████▌  | 4864/6470 [1:04:11<20:16,  1.32it/s]
1|4865|Loss: 0.8725147247314453:  75%|███████▌  | 4865/6470 [1:04:11<18:33,  1.44it/s]
1|4866|Loss: 0.7687810063362122:  75%|███████▌  | 4865/6470 [1:04:11<18:33,  1.44it/s]
1|4866|Loss: 0.7687810063362122:  75%|███████▌  | 4866/6470 [1:04:11<17:36,  1.52it/s]
1|4867|Loss: 0.7897396087646484:  75%|███████▌  | 4866/6470 [1:04:12<17:36,  1.52it/s]
1|4867|Loss: 0.7897396087646484:  75%|███████▌  | 4867/6470 [1:04:12<17:03,  1.57it/s]
1|4868|Loss: 0.9788995385169983:  75%|███████▌  | 4867/6470 [1:04:12<17:03,  1.57it/s]
1|4868|Loss: 0.9788995385169983:  75%|███████▌  | 4868/6470 [1:04:13<17:28,  1.53it/s]
1|4869|Loss: 0.7859313488006592:  75%|███████▌  | 4868/6470 [1:04:13<17:28,  1.53it/s]
1|4869|Loss: 0.7859313488006592:  75%|███████▌  | 4869/6470 [1:04:13<17:34,  1.52it/s]
1|4870|Loss: 0.7792352437973022:  75%|███████▌  | 4869/6470 [1:04:14<17:34,  1.52it/s]
1|4870|Loss: 0.7792352437973022:  75%|███████▌  | 4870/6470 [1:04:14<15:46,  1.69it/s]
1|4871|Loss: 1.004117727279663:  75%|███████▌  | 4870/6470 [1:04:14<15:46,  1.69it/s] 
1|4871|Loss: 1.004117727279663:  75%|███████▌  | 4871/6470 [1:04:14<16:08,  1.65it/s]
1|4872|Loss: 0.8542848825454712:  75%|███████▌  | 4871/6470 [1:04:15<16:08,  1.65it/s]
1|4872|Loss: 0.8542848825454712:  75%|███████▌  | 4872/6470 [1:04:15<16:57,  1.57it/s]
1|4873|Loss: 0.853794515132904:  75%|███████▌  | 4872/6470 [1:04:16<16:57,  1.57it/s] 
1|4873|Loss: 0.853794515132904:  75%|███████▌  | 4873/6470 [1:04:16<18:06,  1.47it/s]
1|4874|Loss: 0.9448482990264893:  75%|███████▌  | 4873/6470 [1:04:16<18:06,  1.47it/s]
1|4874|Loss: 0.9448482990264893:  75%|███████▌  | 4874/6470 [1:04:17<18:49,  1.41it/s]
1|4875|Loss: 0.8006751537322998:  75%|███████▌  | 4874/6470 [1:04:17<18:49,  1.41it/s]
1|4875|Loss: 0.8006751537322998:  75%|███████▌  | 4875/6470 [1:04:17<19:12,  1.38it/s]
1|4876|Loss: 0.8031524419784546:  75%|███████▌  | 4875/6470 [1:04:18<19:12,  1.38it/s]
1|4876|Loss: 0.8031524419784546:  75%|███████▌  | 4876/6470 [1:04:18<19:52,  1.34it/s]
1|4877|Loss: 1.1228041648864746:  75%|███████▌  | 4876/6470 [1:04:19<19:52,  1.34it/s]
1|4877|Loss: 1.1228041648864746:  75%|███████▌  | 4877/6470 [1:04:19<19:39,  1.35it/s]
1|4878|Loss: 0.9969116449356079:  75%|███████▌  | 4877/6470 [1:04:19<19:39,  1.35it/s]
1|4878|Loss: 0.9969116449356079:  75%|███████▌  | 4878/6470 [1:04:20<19:53,  1.33it/s]
1|4879|Loss: 0.7286118268966675:  75%|███████▌  | 4878/6470 [1:04:20<19:53,  1.33it/s]
1|4879|Loss: 0.7286118268966675:  75%|███████▌  | 4879/6470 [1:04:21<20:26,  1.30it/s]
1|4880|Loss: 0.6765038967132568:  75%|███████▌  | 4879/6470 [1:04:21<20:26,  1.30it/s]
1|4880|Loss: 0.6765038967132568:  75%|███████▌  | 4880/6470 [1:04:21<21:15,  1.25it/s]
1|4881|Loss: 0.9352266788482666:  75%|███████▌  | 4880/6470 [1:04:22<21:15,  1.25it/s]
1|4881|Loss: 0.9352266788482666:  75%|███████▌  | 4881/6470 [1:04:22<20:36,  1.28it/s]
1|4882|Loss: 0.8386085033416748:  75%|███████▌  | 4881/6470 [1:04:23<20:36,  1.28it/s]
1|4882|Loss: 0.8386085033416748:  75%|███████▌  | 4882/6470 [1:04:23<19:54,  1.33it/s]
1|4883|Loss: 0.9245119094848633:  75%|███████▌  | 4882/6470 [1:04:23<19:54,  1.33it/s]
1|4883|Loss: 0.9245119094848633:  75%|███████▌  | 4883/6470 [1:04:24<19:49,  1.33it/s]
1|4884|Loss: 0.8809292316436768:  75%|███████▌  | 4883/6470 [1:04:24<19:49,  1.33it/s]
1|4884|Loss: 0.8809292316436768:  75%|███████▌  | 4884/6470 [1:04:24<19:20,  1.37it/s]
1|4885|Loss: 0.6788759231567383:  75%|███████▌  | 4884/6470 [1:04:25<19:20,  1.37it/s]
1|4885|Loss: 0.6788759231567383:  76%|███████▌  | 4885/6470 [1:04:25<21:20,  1.24it/s]
1|4886|Loss: 0.9254605770111084:  76%|███████▌  | 4885/6470 [1:04:26<21:20,  1.24it/s]
1|4886|Loss: 0.9254605770111084:  76%|███████▌  | 4886/6470 [1:04:26<21:30,  1.23it/s]
1|4887|Loss: 1.045116662979126:  76%|███████▌  | 4886/6470 [1:04:27<21:30,  1.23it/s] 
1|4887|Loss: 1.045116662979126:  76%|███████▌  | 4887/6470 [1:04:27<22:17,  1.18it/s]
1|4888|Loss: 0.9809648394584656:  76%|███████▌  | 4887/6470 [1:04:28<22:17,  1.18it/s]
1|4888|Loss: 0.9809648394584656:  76%|███████▌  | 4888/6470 [1:04:28<22:57,  1.15it/s]
1|4889|Loss: 0.876785933971405:  76%|███████▌  | 4888/6470 [1:04:28<22:57,  1.15it/s] 
1|4889|Loss: 0.876785933971405:  76%|███████▌  | 4889/6470 [1:04:29<22:20,  1.18it/s]
1|4890|Loss: 1.0710922479629517:  76%|███████▌  | 4889/6470 [1:04:29<22:20,  1.18it/s]
1|4890|Loss: 1.0710922479629517:  76%|███████▌  | 4890/6470 [1:04:30<22:01,  1.20it/s]
1|4891|Loss: 0.838104248046875:  76%|███████▌  | 4890/6470 [1:04:30<22:01,  1.20it/s] 
1|4891|Loss: 0.838104248046875:  76%|███████▌  | 4891/6470 [1:04:30<21:47,  1.21it/s]
1|4892|Loss: 0.9190201759338379:  76%|███████▌  | 4891/6470 [1:04:31<21:47,  1.21it/s]
1|4892|Loss: 0.9190201759338379:  76%|███████▌  | 4892/6470 [1:04:31<21:17,  1.23it/s]
1|4893|Loss: 0.9421881437301636:  76%|███████▌  | 4892/6470 [1:04:31<21:17,  1.23it/s]
1|4893|Loss: 0.9421881437301636:  76%|███████▌  | 4893/6470 [1:04:32<19:30,  1.35it/s]
1|4894|Loss: 0.8728183507919312:  76%|███████▌  | 4893/6470 [1:04:32<19:30,  1.35it/s]
1|4894|Loss: 0.8728183507919312:  76%|███████▌  | 4894/6470 [1:04:32<18:56,  1.39it/s]
1|4895|Loss: 0.8418364524841309:  76%|███████▌  | 4894/6470 [1:04:33<18:56,  1.39it/s]
1|4895|Loss: 0.8418364524841309:  76%|███████▌  | 4895/6470 [1:04:33<19:12,  1.37it/s]
1|4896|Loss: 0.9031417369842529:  76%|███████▌  | 4895/6470 [1:04:34<19:12,  1.37it/s]
1|4896|Loss: 0.9031417369842529:  76%|███████▌  | 4896/6470 [1:04:34<19:18,  1.36it/s]
1|4897|Loss: 0.8380694389343262:  76%|███████▌  | 4896/6470 [1:04:34<19:18,  1.36it/s]
1|4897|Loss: 0.8380694389343262:  76%|███████▌  | 4897/6470 [1:04:35<19:30,  1.34it/s]
1|4898|Loss: 0.8295358419418335:  76%|███████▌  | 4897/6470 [1:04:35<19:30,  1.34it/s]
1|4898|Loss: 0.8295358419418335:  76%|███████▌  | 4898/6470 [1:04:35<19:10,  1.37it/s]
1|4899|Loss: 0.7977231740951538:  76%|███████▌  | 4898/6470 [1:04:36<19:10,  1.37it/s]
1|4899|Loss: 0.7977231740951538:  76%|███████▌  | 4899/6470 [1:04:36<21:11,  1.24it/s]
1|4900|Loss: 0.9870439171791077:  76%|███████▌  | 4899/6470 [1:04:37<21:11,  1.24it/s]
1|4900|Loss: 0.9870439171791077:  76%|███████▌  | 4900/6470 [1:04:37<21:36,  1.21it/s]
1|4901|Loss: 0.7577665448188782:  76%|███████▌  | 4900/6470 [1:04:38<21:36,  1.21it/s]
1|4901|Loss: 0.7577665448188782:  76%|███████▌  | 4901/6470 [1:04:38<20:20,  1.29it/s]
1|4902|Loss: 1.0160521268844604:  76%|███████▌  | 4901/6470 [1:04:38<20:20,  1.29it/s]
1|4902|Loss: 1.0160521268844604:  76%|███████▌  | 4902/6470 [1:04:39<20:41,  1.26it/s]
1|4903|Loss: 0.8470178246498108:  76%|███████▌  | 4902/6470 [1:04:39<20:41,  1.26it/s]
1|4903|Loss: 0.8470178246498108:  76%|███████▌  | 4903/6470 [1:04:39<20:15,  1.29it/s]
1|4904|Loss: 0.7731354832649231:  76%|███████▌  | 4903/6470 [1:04:40<20:15,  1.29it/s]
1|4904|Loss: 0.7731354832649231:  76%|███████▌  | 4904/6470 [1:04:40<19:35,  1.33it/s]
1|4905|Loss: 0.7811411023139954:  76%|███████▌  | 4904/6470 [1:04:41<19:35,  1.33it/s]
1|4905|Loss: 0.7811411023139954:  76%|███████▌  | 4905/6470 [1:04:41<19:37,  1.33it/s]
1|4906|Loss: 0.8535302877426147:  76%|███████▌  | 4905/6470 [1:04:41<19:37,  1.33it/s]
1|4906|Loss: 0.8535302877426147:  76%|███████▌  | 4906/6470 [1:04:42<19:56,  1.31it/s]
1|4907|Loss: 0.8800733685493469:  76%|███████▌  | 4906/6470 [1:04:42<19:56,  1.31it/s]
1|4907|Loss: 0.8800733685493469:  76%|███████▌  | 4907/6470 [1:04:43<21:55,  1.19it/s]
1|4908|Loss: 0.7067440152168274:  76%|███████▌  | 4907/6470 [1:04:44<21:55,  1.19it/s]
1|4908|Loss: 0.7067440152168274:  76%|███████▌  | 4908/6470 [1:04:44<28:17,  1.09s/it]
1|4909|Loss: 0.7839676141738892:  76%|███████▌  | 4908/6470 [1:04:45<28:17,  1.09s/it]
1|4909|Loss: 0.7839676141738892:  76%|███████▌  | 4909/6470 [1:04:45<25:30,  1.02it/s]
1|4910|Loss: 0.8898439407348633:  76%|███████▌  | 4909/6470 [1:04:45<25:30,  1.02it/s]
1|4910|Loss: 0.8898439407348633:  76%|███████▌  | 4910/6470 [1:04:46<23:00,  1.13it/s]
1|4911|Loss: 1.09328031539917:  76%|███████▌  | 4910/6470 [1:04:46<23:00,  1.13it/s]  
1|4911|Loss: 1.09328031539917:  76%|███████▌  | 4911/6470 [1:04:47<22:26,  1.16it/s]
1|4912|Loss: 0.9683075547218323:  76%|███████▌  | 4911/6470 [1:04:47<22:26,  1.16it/s]
1|4912|Loss: 0.9683075547218323:  76%|███████▌  | 4912/6470 [1:04:47<21:53,  1.19it/s]
1|4913|Loss: 1.0485587120056152:  76%|███████▌  | 4912/6470 [1:04:48<21:53,  1.19it/s]
1|4913|Loss: 1.0485587120056152:  76%|███████▌  | 4913/6470 [1:04:48<21:18,  1.22it/s]
1|4914|Loss: 0.8649803996086121:  76%|███████▌  | 4913/6470 [1:04:48<21:18,  1.22it/s]
1|4914|Loss: 0.8649803996086121:  76%|███████▌  | 4914/6470 [1:04:49<19:37,  1.32it/s]
1|4915|Loss: 0.8976413607597351:  76%|███████▌  | 4914/6470 [1:04:49<19:37,  1.32it/s]
1|4915|Loss: 0.8976413607597351:  76%|███████▌  | 4915/6470 [1:04:49<18:39,  1.39it/s]
1|4916|Loss: 0.8523496389389038:  76%|███████▌  | 4915/6470 [1:04:50<18:39,  1.39it/s]
1|4916|Loss: 0.8523496389389038:  76%|███████▌  | 4916/6470 [1:04:50<19:11,  1.35it/s]
1|4917|Loss: 1.118294596672058:  76%|███████▌  | 4916/6470 [1:04:51<19:11,  1.35it/s] 
1|4917|Loss: 1.118294596672058:  76%|███████▌  | 4917/6470 [1:04:51<19:08,  1.35it/s]
1|4918|Loss: 0.8410184383392334:  76%|███████▌  | 4917/6470 [1:04:51<19:08,  1.35it/s]
1|4918|Loss: 0.8410184383392334:  76%|███████▌  | 4918/6470 [1:04:52<18:40,  1.38it/s]
1|4919|Loss: 0.9064019918441772:  76%|███████▌  | 4918/6470 [1:04:52<18:40,  1.38it/s]
1|4919|Loss: 0.9064019918441772:  76%|███████▌  | 4919/6470 [1:04:52<18:17,  1.41it/s]
1|4920|Loss: 0.7847240567207336:  76%|███████▌  | 4919/6470 [1:04:53<18:17,  1.41it/s]
1|4920|Loss: 0.7847240567207336:  76%|███████▌  | 4920/6470 [1:04:53<19:26,  1.33it/s]
1|4921|Loss: 0.7847894430160522:  76%|███████▌  | 4920/6470 [1:04:54<19:26,  1.33it/s]
1|4921|Loss: 0.7847894430160522:  76%|███████▌  | 4921/6470 [1:04:54<19:15,  1.34it/s]
1|4922|Loss: 0.88093501329422:  76%|███████▌  | 4921/6470 [1:04:54<19:15,  1.34it/s]  
1|4922|Loss: 0.88093501329422:  76%|███████▌  | 4922/6470 [1:04:55<20:30,  1.26it/s]
1|4923|Loss: 0.843921422958374:  76%|███████▌  | 4922/6470 [1:04:55<20:30,  1.26it/s]
1|4923|Loss: 0.843921422958374:  76%|███████▌  | 4923/6470 [1:04:56<21:59,  1.17it/s]
1|4924|Loss: 0.7981958389282227:  76%|███████▌  | 4923/6470 [1:04:56<21:59,  1.17it/s]
1|4924|Loss: 0.7981958389282227:  76%|███████▌  | 4924/6470 [1:04:56<20:25,  1.26it/s]
1|4925|Loss: 0.9159983396530151:  76%|███████▌  | 4924/6470 [1:04:57<20:25,  1.26it/s]
1|4925|Loss: 0.9159983396530151:  76%|███████▌  | 4925/6470 [1:04:57<19:27,  1.32it/s]
1|4926|Loss: 0.8946194052696228:  76%|███████▌  | 4925/6470 [1:04:58<19:27,  1.32it/s]
1|4926|Loss: 0.8946194052696228:  76%|███████▌  | 4926/6470 [1:04:58<21:19,  1.21it/s]
1|4927|Loss: 0.8818144798278809:  76%|███████▌  | 4926/6470 [1:04:59<21:19,  1.21it/s]
1|4927|Loss: 0.8818144798278809:  76%|███████▌  | 4927/6470 [1:04:59<21:29,  1.20it/s]
1|4928|Loss: 0.9604916572570801:  76%|███████▌  | 4927/6470 [1:04:59<21:29,  1.20it/s]
1|4928|Loss: 0.9604916572570801:  76%|███████▌  | 4928/6470 [1:05:00<21:14,  1.21it/s]
1|4929|Loss: 0.784970223903656:  76%|███████▌  | 4928/6470 [1:05:00<21:14,  1.21it/s] 
1|4929|Loss: 0.784970223903656:  76%|███████▌  | 4929/6470 [1:05:00<19:43,  1.30it/s]
1|4930|Loss: 0.8269395232200623:  76%|███████▌  | 4929/6470 [1:05:01<19:43,  1.30it/s]
1|4930|Loss: 0.8269395232200623:  76%|███████▌  | 4930/6470 [1:05:01<17:41,  1.45it/s]
1|4931|Loss: 1.2943183183670044:  76%|███████▌  | 4930/6470 [1:05:01<17:41,  1.45it/s]
1|4931|Loss: 1.2943183183670044:  76%|███████▌  | 4931/6470 [1:05:02<18:52,  1.36it/s]
1|4932|Loss: 1.1626999378204346:  76%|███████▌  | 4931/6470 [1:05:02<18:52,  1.36it/s]
1|4932|Loss: 1.1626999378204346:  76%|███████▌  | 4932/6470 [1:05:02<18:58,  1.35it/s]
1|4933|Loss: 0.7887593507766724:  76%|███████▌  | 4932/6470 [1:05:03<18:58,  1.35it/s]
1|4933|Loss: 0.7887593507766724:  76%|███████▌  | 4933/6470 [1:05:03<19:16,  1.33it/s]
1|4934|Loss: 0.9791676998138428:  76%|███████▌  | 4933/6470 [1:05:04<19:16,  1.33it/s]
1|4934|Loss: 0.9791676998138428:  76%|███████▋  | 4934/6470 [1:05:04<19:25,  1.32it/s]
1|4935|Loss: 0.7517383098602295:  76%|███████▋  | 4934/6470 [1:05:04<19:25,  1.32it/s]
1|4935|Loss: 0.7517383098602295:  76%|███████▋  | 4935/6470 [1:05:05<19:23,  1.32it/s]
1|4936|Loss: 1.044905185699463:  76%|███████▋  | 4935/6470 [1:05:05<19:23,  1.32it/s] 
1|4936|Loss: 1.044905185699463:  76%|███████▋  | 4936/6470 [1:05:05<18:05,  1.41it/s]
1|4937|Loss: 0.840587317943573:  76%|███████▋  | 4936/6470 [1:05:06<18:05,  1.41it/s]
1|4937|Loss: 0.840587317943573:  76%|███████▋  | 4937/6470 [1:05:06<18:04,  1.41it/s]
1|4938|Loss: 0.9297879338264465:  76%|███████▋  | 4937/6470 [1:05:07<18:04,  1.41it/s]
1|4938|Loss: 0.9297879338264465:  76%|███████▋  | 4938/6470 [1:05:07<19:30,  1.31it/s]
1|4939|Loss: 0.7499263882637024:  76%|███████▋  | 4938/6470 [1:05:07<19:30,  1.31it/s]
1|4939|Loss: 0.7499263882637024:  76%|███████▋  | 4939/6470 [1:05:08<18:38,  1.37it/s]
1|4940|Loss: 0.7603749632835388:  76%|███████▋  | 4939/6470 [1:05:08<18:38,  1.37it/s]
1|4940|Loss: 0.7603749632835388:  76%|███████▋  | 4940/6470 [1:05:08<18:35,  1.37it/s]
1|4941|Loss: 0.9101819396018982:  76%|███████▋  | 4940/6470 [1:05:09<18:35,  1.37it/s]
1|4941|Loss: 0.9101819396018982:  76%|███████▋  | 4941/6470 [1:05:09<17:20,  1.47it/s]
1|4942|Loss: 0.8496848940849304:  76%|███████▋  | 4941/6470 [1:05:09<17:20,  1.47it/s]
1|4942|Loss: 0.8496848940849304:  76%|███████▋  | 4942/6470 [1:05:10<16:57,  1.50it/s]
1|4943|Loss: 0.8427149653434753:  76%|███████▋  | 4942/6470 [1:05:10<16:57,  1.50it/s]
1|4943|Loss: 0.8427149653434753:  76%|███████▋  | 4943/6470 [1:05:10<16:16,  1.56it/s]
1|4944|Loss: 0.964034378528595:  76%|███████▋  | 4943/6470 [1:05:10<16:16,  1.56it/s] 
1|4944|Loss: 0.964034378528595:  76%|███████▋  | 4944/6470 [1:05:11<17:07,  1.48it/s]
1|4945|Loss: 0.8527365326881409:  76%|███████▋  | 4944/6470 [1:05:11<17:07,  1.48it/s]
1|4945|Loss: 0.8527365326881409:  76%|███████▋  | 4945/6470 [1:05:12<17:47,  1.43it/s]
1|4946|Loss: 0.8763355612754822:  76%|███████▋  | 4945/6470 [1:05:12<17:47,  1.43it/s]
1|4946|Loss: 0.8763355612754822:  76%|███████▋  | 4946/6470 [1:05:12<17:41,  1.44it/s]
1|4947|Loss: 0.8664232492446899:  76%|███████▋  | 4946/6470 [1:05:13<17:41,  1.44it/s]
1|4947|Loss: 0.8664232492446899:  76%|███████▋  | 4947/6470 [1:05:13<17:23,  1.46it/s]
1|4948|Loss: 1.1500886678695679:  76%|███████▋  | 4947/6470 [1:05:13<17:23,  1.46it/s]
1|4948|Loss: 1.1500886678695679:  76%|███████▋  | 4948/6470 [1:05:14<19:08,  1.33it/s]
1|4949|Loss: 0.8603236079216003:  76%|███████▋  | 4948/6470 [1:05:14<19:08,  1.33it/s]
1|4949|Loss: 0.8603236079216003:  76%|███████▋  | 4949/6470 [1:05:15<19:37,  1.29it/s]
1|4950|Loss: 0.837657630443573:  76%|███████▋  | 4949/6470 [1:05:15<19:37,  1.29it/s] 
1|4950|Loss: 0.837657630443573:  77%|███████▋  | 4950/6470 [1:05:15<19:46,  1.28it/s]
1|4951|Loss: 0.7907255291938782:  77%|███████▋  | 4950/6470 [1:05:16<19:46,  1.28it/s]
1|4951|Loss: 0.7907255291938782:  77%|███████▋  | 4951/6470 [1:05:16<19:36,  1.29it/s]
1|4952|Loss: 0.8629618883132935:  77%|███████▋  | 4951/6470 [1:05:17<19:36,  1.29it/s]
1|4952|Loss: 0.8629618883132935:  77%|███████▋  | 4952/6470 [1:05:17<18:17,  1.38it/s]
1|4953|Loss: 0.8983281254768372:  77%|███████▋  | 4952/6470 [1:05:17<18:17,  1.38it/s]
1|4953|Loss: 0.8983281254768372:  77%|███████▋  | 4953/6470 [1:05:17<16:32,  1.53it/s]
1|4954|Loss: 1.070707082748413:  77%|███████▋  | 4953/6470 [1:05:18<16:32,  1.53it/s] 
1|4954|Loss: 1.070707082748413:  77%|███████▋  | 4954/6470 [1:05:18<16:04,  1.57it/s]
1|4955|Loss: 0.9562970399856567:  77%|███████▋  | 4954/6470 [1:05:18<16:04,  1.57it/s]
1|4955|Loss: 0.9562970399856567:  77%|███████▋  | 4955/6470 [1:05:18<15:04,  1.67it/s]
1|4956|Loss: 0.9613823890686035:  77%|███████▋  | 4955/6470 [1:05:19<15:04,  1.67it/s]
1|4956|Loss: 0.9613823890686035:  77%|███████▋  | 4956/6470 [1:05:19<14:37,  1.72it/s]
1|4957|Loss: 0.7882011532783508:  77%|███████▋  | 4956/6470 [1:05:19<14:37,  1.72it/s]
1|4957|Loss: 0.7882011532783508:  77%|███████▋  | 4957/6470 [1:05:20<14:34,  1.73it/s]
1|4958|Loss: 0.9036752581596375:  77%|███████▋  | 4957/6470 [1:05:20<14:34,  1.73it/s]
1|4958|Loss: 0.9036752581596375:  77%|███████▋  | 4958/6470 [1:05:20<15:15,  1.65it/s]
1|4959|Loss: 0.9234218001365662:  77%|███████▋  | 4958/6470 [1:05:21<15:15,  1.65it/s]
1|4959|Loss: 0.9234218001365662:  77%|███████▋  | 4959/6470 [1:05:21<15:44,  1.60it/s]
1|4960|Loss: 0.9112405776977539:  77%|███████▋  | 4959/6470 [1:05:21<15:44,  1.60it/s]
1|4960|Loss: 0.9112405776977539:  77%|███████▋  | 4960/6470 [1:05:22<16:39,  1.51it/s]
1|4961|Loss: 0.9721131920814514:  77%|███████▋  | 4960/6470 [1:05:22<16:39,  1.51it/s]
1|4961|Loss: 0.9721131920814514:  77%|███████▋  | 4961/6470 [1:05:22<17:21,  1.45it/s]
1|4962|Loss: 0.9167512059211731:  77%|███████▋  | 4961/6470 [1:05:23<17:21,  1.45it/s]
1|4962|Loss: 0.9167512059211731:  77%|███████▋  | 4962/6470 [1:05:23<18:32,  1.36it/s]
1|4963|Loss: 1.0283433198928833:  77%|███████▋  | 4962/6470 [1:05:24<18:32,  1.36it/s]
1|4963|Loss: 1.0283433198928833:  77%|███████▋  | 4963/6470 [1:05:24<19:38,  1.28it/s]
1|4964|Loss: 0.8815109133720398:  77%|███████▋  | 4963/6470 [1:05:25<19:38,  1.28it/s]
1|4964|Loss: 0.8815109133720398:  77%|███████▋  | 4964/6470 [1:05:25<19:09,  1.31it/s]
1|4965|Loss: 1.2279088497161865:  77%|███████▋  | 4964/6470 [1:05:25<19:09,  1.31it/s]
1|4965|Loss: 1.2279088497161865:  77%|███████▋  | 4965/6470 [1:05:26<20:06,  1.25it/s]
1|4966|Loss: 0.800216555595398:  77%|███████▋  | 4965/6470 [1:05:26<20:06,  1.25it/s] 
1|4966|Loss: 0.800216555595398:  77%|███████▋  | 4966/6470 [1:05:26<18:29,  1.36it/s]
1|4967|Loss: 0.9380747079849243:  77%|███████▋  | 4966/6470 [1:05:27<18:29,  1.36it/s]
1|4967|Loss: 0.9380747079849243:  77%|███████▋  | 4967/6470 [1:05:27<17:18,  1.45it/s]
1|4968|Loss: 0.9903587102890015:  77%|███████▋  | 4967/6470 [1:05:27<17:18,  1.45it/s]
1|4968|Loss: 0.9903587102890015:  77%|███████▋  | 4968/6470 [1:05:27<15:28,  1.62it/s]
1|4969|Loss: 0.7964779734611511:  77%|███████▋  | 4968/6470 [1:05:28<15:28,  1.62it/s]
1|4969|Loss: 0.7964779734611511:  77%|███████▋  | 4969/6470 [1:05:28<15:31,  1.61it/s]
1|4970|Loss: 0.8794932961463928:  77%|███████▋  | 4969/6470 [1:05:28<15:31,  1.61it/s]
1|4970|Loss: 0.8794932961463928:  77%|███████▋  | 4970/6470 [1:05:29<15:35,  1.60it/s]
1|4971|Loss: 0.9332400560379028:  77%|███████▋  | 4970/6470 [1:05:29<15:35,  1.60it/s]
1|4971|Loss: 0.9332400560379028:  77%|███████▋  | 4971/6470 [1:05:29<15:57,  1.57it/s]
1|4972|Loss: 0.8113815784454346:  77%|███████▋  | 4971/6470 [1:05:30<15:57,  1.57it/s]
1|4972|Loss: 0.8113815784454346:  77%|███████▋  | 4972/6470 [1:05:30<15:27,  1.62it/s]
1|4973|Loss: 0.8179500699043274:  77%|███████▋  | 4972/6470 [1:05:30<15:27,  1.62it/s]
1|4973|Loss: 0.8179500699043274:  77%|███████▋  | 4973/6470 [1:05:31<16:23,  1.52it/s]
1|4974|Loss: 0.864877462387085:  77%|███████▋  | 4973/6470 [1:05:31<16:23,  1.52it/s] 
1|4974|Loss: 0.864877462387085:  77%|███████▋  | 4974/6470 [1:05:31<17:21,  1.44it/s]
1|4975|Loss: 1.0521951913833618:  77%|███████▋  | 4974/6470 [1:05:32<17:21,  1.44it/s]
1|4975|Loss: 1.0521951913833618:  77%|███████▋  | 4975/6470 [1:05:32<17:59,  1.39it/s]
1|4976|Loss: 0.7657327651977539:  77%|███████▋  | 4975/6470 [1:05:33<17:59,  1.39it/s]
1|4976|Loss: 0.7657327651977539:  77%|███████▋  | 4976/6470 [1:05:33<18:33,  1.34it/s]
1|4977|Loss: 0.9835583567619324:  77%|███████▋  | 4976/6470 [1:05:33<18:33,  1.34it/s]
1|4977|Loss: 0.9835583567619324:  77%|███████▋  | 4977/6470 [1:05:34<18:48,  1.32it/s]
1|4978|Loss: 0.8183174133300781:  77%|███████▋  | 4977/6470 [1:05:34<18:48,  1.32it/s]
1|4978|Loss: 0.8183174133300781:  77%|███████▋  | 4978/6470 [1:05:35<18:52,  1.32it/s]
1|4979|Loss: 0.8434858918190002:  77%|███████▋  | 4978/6470 [1:05:35<18:52,  1.32it/s]
1|4979|Loss: 0.8434858918190002:  77%|███████▋  | 4979/6470 [1:05:35<17:40,  1.41it/s]
1|4980|Loss: 0.7957358956336975:  77%|███████▋  | 4979/6470 [1:05:36<17:40,  1.41it/s]
1|4980|Loss: 0.7957358956336975:  77%|███████▋  | 4980/6470 [1:05:36<17:54,  1.39it/s]
1|4981|Loss: 0.9049745798110962:  77%|███████▋  | 4980/6470 [1:05:36<17:54,  1.39it/s]
1|4981|Loss: 0.9049745798110962:  77%|███████▋  | 4981/6470 [1:05:37<17:28,  1.42it/s]
1|4982|Loss: 0.8561213612556458:  77%|███████▋  | 4981/6470 [1:05:37<17:28,  1.42it/s]
1|4982|Loss: 0.8561213612556458:  77%|███████▋  | 4982/6470 [1:05:37<16:28,  1.50it/s]
1|4983|Loss: 0.8553910255432129:  77%|███████▋  | 4982/6470 [1:05:37<16:28,  1.50it/s]
1|4983|Loss: 0.8553910255432129:  77%|███████▋  | 4983/6470 [1:05:38<16:39,  1.49it/s]
1|4984|Loss: 0.8259779810905457:  77%|███████▋  | 4983/6470 [1:05:38<16:39,  1.49it/s]
1|4984|Loss: 0.8259779810905457:  77%|███████▋  | 4984/6470 [1:05:39<16:57,  1.46it/s]
1|4985|Loss: 0.8927890062332153:  77%|███████▋  | 4984/6470 [1:05:39<16:57,  1.46it/s]
1|4985|Loss: 0.8927890062332153:  77%|███████▋  | 4985/6470 [1:05:39<17:07,  1.45it/s]
1|4986|Loss: 0.816735565662384:  77%|███████▋  | 4985/6470 [1:05:40<17:07,  1.45it/s] 
1|4986|Loss: 0.816735565662384:  77%|███████▋  | 4986/6470 [1:05:40<17:12,  1.44it/s]
1|4987|Loss: 1.0039353370666504:  77%|███████▋  | 4986/6470 [1:05:40<17:12,  1.44it/s]
1|4987|Loss: 1.0039353370666504:  77%|███████▋  | 4987/6470 [1:05:41<18:42,  1.32it/s]
1|4988|Loss: 0.8351978659629822:  77%|███████▋  | 4987/6470 [1:05:41<18:42,  1.32it/s]
1|4988|Loss: 0.8351978659629822:  77%|███████▋  | 4988/6470 [1:05:41<17:28,  1.41it/s]
1|4989|Loss: 0.7926459312438965:  77%|███████▋  | 4988/6470 [1:05:42<17:28,  1.41it/s]
1|4989|Loss: 0.7926459312438965:  77%|███████▋  | 4989/6470 [1:05:42<15:10,  1.63it/s]
1|4990|Loss: 0.8206308484077454:  77%|███████▋  | 4989/6470 [1:05:42<15:10,  1.63it/s]
1|4990|Loss: 0.8206308484077454:  77%|███████▋  | 4990/6470 [1:05:43<17:10,  1.44it/s]
1|4991|Loss: 0.821758508682251:  77%|███████▋  | 4990/6470 [1:05:43<17:10,  1.44it/s] 
1|4991|Loss: 0.821758508682251:  77%|███████▋  | 4991/6470 [1:05:44<18:22,  1.34it/s]
1|4992|Loss: 0.8071931004524231:  77%|███████▋  | 4991/6470 [1:05:44<18:22,  1.34it/s]
1|4992|Loss: 0.8071931004524231:  77%|███████▋  | 4992/6470 [1:05:45<20:41,  1.19it/s]
1|4993|Loss: 0.7976776957511902:  77%|███████▋  | 4992/6470 [1:05:45<20:41,  1.19it/s]
1|4993|Loss: 0.7976776957511902:  77%|███████▋  | 4993/6470 [1:05:45<19:47,  1.24it/s]
1|4994|Loss: 0.7911866307258606:  77%|███████▋  | 4993/6470 [1:05:46<19:47,  1.24it/s]
1|4994|Loss: 0.7911866307258606:  77%|███████▋  | 4994/6470 [1:05:46<17:51,  1.38it/s]
1|4995|Loss: 0.7455927133560181:  77%|███████▋  | 4994/6470 [1:05:46<17:51,  1.38it/s]
1|4995|Loss: 0.7455927133560181:  77%|███████▋  | 4995/6470 [1:05:47<18:03,  1.36it/s]
1|4996|Loss: 0.913435161113739:  77%|███████▋  | 4995/6470 [1:05:47<18:03,  1.36it/s] 
1|4996|Loss: 0.913435161113739:  77%|███████▋  | 4996/6470 [1:05:48<19:25,  1.26it/s]
1|4997|Loss: 0.8275983929634094:  77%|███████▋  | 4996/6470 [1:05:48<19:25,  1.26it/s]
1|4997|Loss: 0.8275983929634094:  77%|███████▋  | 4997/6470 [1:05:48<19:39,  1.25it/s]
1|4998|Loss: 0.8253591060638428:  77%|███████▋  | 4997/6470 [1:05:49<19:39,  1.25it/s]
1|4998|Loss: 0.8253591060638428:  77%|███████▋  | 4998/6470 [1:05:49<19:14,  1.28it/s]
1|4999|Loss: 0.9607335925102234:  77%|███████▋  | 4998/6470 [1:05:50<19:14,  1.28it/s]
1|4999|Loss: 0.9607335925102234:  77%|███████▋  | 4999/6470 [1:05:50<19:58,  1.23it/s]
1|5000|Loss: 0.8386166095733643:  77%|███████▋  | 4999/6470 [1:05:50<19:58,  1.23it/s]
1|5000|Loss: 0.8386166095733643:  77%|███████▋  | 5000/6470 [1:05:51<19:52,  1.23it/s]
1|5001|Loss: 1.252733588218689:  77%|███████▋  | 5000/6470 [1:05:51<19:52,  1.23it/s] 
1|5001|Loss: 1.252733588218689:  77%|███████▋  | 5001/6470 [1:05:52<22:12,  1.10it/s]
1|5002|Loss: 0.9146406650543213:  77%|███████▋  | 5001/6470 [1:05:52<22:12,  1.10it/s]
1|5002|Loss: 0.9146406650543213:  77%|███████▋  | 5002/6470 [1:05:53<22:05,  1.11it/s]
1|5003|Loss: 0.7974905371665955:  77%|███████▋  | 5002/6470 [1:05:53<22:05,  1.11it/s]
1|5003|Loss: 0.7974905371665955:  77%|███████▋  | 5003/6470 [1:05:54<21:26,  1.14it/s]
1|5004|Loss: 0.911365807056427:  77%|███████▋  | 5003/6470 [1:05:54<21:26,  1.14it/s] 
1|5004|Loss: 0.911365807056427:  77%|███████▋  | 5004/6470 [1:05:54<20:58,  1.16it/s]
1|5005|Loss: 0.9884850978851318:  77%|███████▋  | 5004/6470 [1:05:55<20:58,  1.16it/s]
1|5005|Loss: 0.9884850978851318:  77%|███████▋  | 5005/6470 [1:05:55<20:37,  1.18it/s]
1|5006|Loss: 0.9437991976737976:  77%|███████▋  | 5005/6470 [1:05:56<20:37,  1.18it/s]
1|5006|Loss: 0.9437991976737976:  77%|███████▋  | 5006/6470 [1:05:56<18:52,  1.29it/s]
1|5007|Loss: 0.810161828994751:  77%|███████▋  | 5006/6470 [1:05:56<18:52,  1.29it/s] 
1|5007|Loss: 0.810161828994751:  77%|███████▋  | 5007/6470 [1:05:57<18:38,  1.31it/s]
1|5008|Loss: 1.0087201595306396:  77%|███████▋  | 5007/6470 [1:05:57<18:38,  1.31it/s]
1|5008|Loss: 1.0087201595306396:  77%|███████▋  | 5008/6470 [1:05:57<17:33,  1.39it/s]
1|5009|Loss: 0.9513658285140991:  77%|███████▋  | 5008/6470 [1:05:58<17:33,  1.39it/s]
1|5009|Loss: 0.9513658285140991:  77%|███████▋  | 5009/6470 [1:05:58<17:47,  1.37it/s]
1|5010|Loss: 1.0479601621627808:  77%|███████▋  | 5009/6470 [1:05:58<17:47,  1.37it/s]
1|5010|Loss: 1.0479601621627808:  77%|███████▋  | 5010/6470 [1:05:59<19:06,  1.27it/s]
1|5011|Loss: 0.9880726933479309:  77%|███████▋  | 5010/6470 [1:05:59<19:06,  1.27it/s]
1|5011|Loss: 0.9880726933479309:  77%|███████▋  | 5011/6470 [1:06:00<18:13,  1.33it/s]
1|5012|Loss: 0.6603971123695374:  77%|███████▋  | 5011/6470 [1:06:00<18:13,  1.33it/s]
1|5012|Loss: 0.6603971123695374:  77%|███████▋  | 5012/6470 [1:06:00<19:01,  1.28it/s]
1|5013|Loss: 0.7632628083229065:  77%|███████▋  | 5012/6470 [1:06:01<19:01,  1.28it/s]
1|5013|Loss: 0.7632628083229065:  77%|███████▋  | 5013/6470 [1:06:01<17:49,  1.36it/s]
1|5014|Loss: 0.9627700448036194:  77%|███████▋  | 5013/6470 [1:06:01<17:49,  1.36it/s]
1|5014|Loss: 0.9627700448036194:  77%|███████▋  | 5014/6470 [1:06:02<17:04,  1.42it/s]
1|5015|Loss: 0.9104211330413818:  77%|███████▋  | 5014/6470 [1:06:02<17:04,  1.42it/s]
1|5015|Loss: 0.9104211330413818:  78%|███████▊  | 5015/6470 [1:06:03<18:21,  1.32it/s]
1|5016|Loss: 0.7838485240936279:  78%|███████▊  | 5015/6470 [1:06:03<18:21,  1.32it/s]
1|5016|Loss: 0.7838485240936279:  78%|███████▊  | 5016/6470 [1:06:03<19:19,  1.25it/s]
1|5017|Loss: 0.9548308849334717:  78%|███████▊  | 5016/6470 [1:06:04<19:19,  1.25it/s]
1|5017|Loss: 0.9548308849334717:  78%|███████▊  | 5017/6470 [1:06:04<19:54,  1.22it/s]
1|5018|Loss: 1.0829317569732666:  78%|███████▊  | 5017/6470 [1:06:05<19:54,  1.22it/s]
1|5018|Loss: 1.0829317569732666:  78%|███████▊  | 5018/6470 [1:06:05<20:33,  1.18it/s]
1|5019|Loss: 1.0757863521575928:  78%|███████▊  | 5018/6470 [1:06:06<20:33,  1.18it/s]
1|5019|Loss: 1.0757863521575928:  78%|███████▊  | 5019/6470 [1:06:06<20:42,  1.17it/s]
1|5020|Loss: 0.9620296955108643:  78%|███████▊  | 5019/6470 [1:06:07<20:42,  1.17it/s]
1|5020|Loss: 0.9620296955108643:  78%|███████▊  | 5020/6470 [1:06:07<20:13,  1.19it/s]
1|5021|Loss: 0.9092370271682739:  78%|███████▊  | 5020/6470 [1:06:07<20:13,  1.19it/s]
1|5021|Loss: 0.9092370271682739:  78%|███████▊  | 5021/6470 [1:06:08<20:08,  1.20it/s]
1|5022|Loss: 0.9049818515777588:  78%|███████▊  | 5021/6470 [1:06:08<20:08,  1.20it/s]
1|5022|Loss: 0.9049818515777588:  78%|███████▊  | 5022/6470 [1:06:08<19:20,  1.25it/s]
1|5023|Loss: 1.0915782451629639:  78%|███████▊  | 5022/6470 [1:06:09<19:20,  1.25it/s]
1|5023|Loss: 1.0915782451629639:  78%|███████▊  | 5023/6470 [1:06:09<19:42,  1.22it/s]
1|5024|Loss: 0.8616726398468018:  78%|███████▊  | 5023/6470 [1:06:10<19:42,  1.22it/s]
1|5024|Loss: 0.8616726398468018:  78%|███████▊  | 5024/6470 [1:06:10<20:35,  1.17it/s]
1|5025|Loss: 1.1295082569122314:  78%|███████▊  | 5024/6470 [1:06:11<20:35,  1.17it/s]
1|5025|Loss: 1.1295082569122314:  78%|███████▊  | 5025/6470 [1:06:11<19:50,  1.21it/s]
1|5026|Loss: 0.847256600856781:  78%|███████▊  | 5025/6470 [1:06:11<19:50,  1.21it/s] 
1|5026|Loss: 0.847256600856781:  78%|███████▊  | 5026/6470 [1:06:12<18:15,  1.32it/s]
1|5027|Loss: 0.8160256147384644:  78%|███████▊  | 5026/6470 [1:06:12<18:15,  1.32it/s]
1|5027|Loss: 0.8160256147384644:  78%|███████▊  | 5027/6470 [1:06:12<16:44,  1.44it/s]
1|5028|Loss: 0.8424581289291382:  78%|███████▊  | 5027/6470 [1:06:13<16:44,  1.44it/s]
1|5028|Loss: 0.8424581289291382:  78%|███████▊  | 5028/6470 [1:06:13<17:05,  1.41it/s]
1|5029|Loss: 0.8439522385597229:  78%|███████▊  | 5028/6470 [1:06:13<17:05,  1.41it/s]
1|5029|Loss: 0.8439522385597229:  78%|███████▊  | 5029/6470 [1:06:14<18:55,  1.27it/s]
1|5030|Loss: 0.8894782066345215:  78%|███████▊  | 5029/6470 [1:06:14<18:55,  1.27it/s]
1|5030|Loss: 0.8894782066345215:  78%|███████▊  | 5030/6470 [1:06:15<17:40,  1.36it/s]
1|5031|Loss: 0.770709216594696:  78%|███████▊  | 5030/6470 [1:06:15<17:40,  1.36it/s] 
1|5031|Loss: 0.770709216594696:  78%|███████▊  | 5031/6470 [1:06:15<18:10,  1.32it/s]
1|5032|Loss: 0.8618451356887817:  78%|███████▊  | 5031/6470 [1:06:16<18:10,  1.32it/s]
1|5032|Loss: 0.8618451356887817:  78%|███████▊  | 5032/6470 [1:06:16<18:55,  1.27it/s]
1|5033|Loss: 0.8571817874908447:  78%|███████▊  | 5032/6470 [1:06:17<18:55,  1.27it/s]
1|5033|Loss: 0.8571817874908447:  78%|███████▊  | 5033/6470 [1:06:17<18:49,  1.27it/s]
1|5034|Loss: 0.789166271686554:  78%|███████▊  | 5033/6470 [1:06:17<18:49,  1.27it/s] 
1|5034|Loss: 0.789166271686554:  78%|███████▊  | 5034/6470 [1:06:18<18:41,  1.28it/s]
1|5035|Loss: 0.9234098792076111:  78%|███████▊  | 5034/6470 [1:06:18<18:41,  1.28it/s]
1|5035|Loss: 0.9234098792076111:  78%|███████▊  | 5035/6470 [1:06:19<18:42,  1.28it/s]
1|5036|Loss: 0.8048816919326782:  78%|███████▊  | 5035/6470 [1:06:19<18:42,  1.28it/s]
1|5036|Loss: 0.8048816919326782:  78%|███████▊  | 5036/6470 [1:06:19<18:01,  1.33it/s]
1|5037|Loss: 0.7635276913642883:  78%|███████▊  | 5036/6470 [1:06:20<18:01,  1.33it/s]
1|5037|Loss: 0.7635276913642883:  78%|███████▊  | 5037/6470 [1:06:20<18:34,  1.29it/s]
1|5038|Loss: 0.8235569000244141:  78%|███████▊  | 5037/6470 [1:06:21<18:34,  1.29it/s]
1|5038|Loss: 0.8235569000244141:  78%|███████▊  | 5038/6470 [1:06:21<19:27,  1.23it/s]
1|5039|Loss: 0.9208157062530518:  78%|███████▊  | 5038/6470 [1:06:21<19:27,  1.23it/s]
1|5039|Loss: 0.9208157062530518:  78%|███████▊  | 5039/6470 [1:06:22<19:12,  1.24it/s]
1|5040|Loss: 0.7424399256706238:  78%|███████▊  | 5039/6470 [1:06:22<19:12,  1.24it/s]
1|5040|Loss: 0.7424399256706238:  78%|███████▊  | 5040/6470 [1:06:23<19:58,  1.19it/s]
1|5041|Loss: 0.8517431020736694:  78%|███████▊  | 5040/6470 [1:06:23<19:58,  1.19it/s]
1|5041|Loss: 0.8517431020736694:  78%|███████▊  | 5041/6470 [1:06:23<19:14,  1.24it/s]
1|5042|Loss: 0.8404746055603027:  78%|███████▊  | 5041/6470 [1:06:24<19:14,  1.24it/s]
1|5042|Loss: 0.8404746055603027:  78%|███████▊  | 5042/6470 [1:06:24<19:01,  1.25it/s]
1|5043|Loss: 0.9661846160888672:  78%|███████▊  | 5042/6470 [1:06:25<19:01,  1.25it/s]
1|5043|Loss: 0.9661846160888672:  78%|███████▊  | 5043/6470 [1:06:25<18:29,  1.29it/s]
1|5044|Loss: 0.8047559857368469:  78%|███████▊  | 5043/6470 [1:06:25<18:29,  1.29it/s]
1|5044|Loss: 0.8047559857368469:  78%|███████▊  | 5044/6470 [1:06:26<18:31,  1.28it/s]
1|5045|Loss: 0.967104434967041:  78%|███████▊  | 5044/6470 [1:06:26<18:31,  1.28it/s] 
1|5045|Loss: 0.967104434967041:  78%|███████▊  | 5045/6470 [1:06:26<18:38,  1.27it/s]
1|5046|Loss: 0.9145390391349792:  78%|███████▊  | 5045/6470 [1:06:27<18:38,  1.27it/s]
1|5046|Loss: 0.9145390391349792:  78%|███████▊  | 5046/6470 [1:06:27<19:21,  1.23it/s]
1|5047|Loss: 1.0470807552337646:  78%|███████▊  | 5046/6470 [1:06:28<19:21,  1.23it/s]
1|5047|Loss: 1.0470807552337646:  78%|███████▊  | 5047/6470 [1:06:28<20:54,  1.13it/s]
1|5048|Loss: 0.7627099752426147:  78%|███████▊  | 5047/6470 [1:06:29<20:54,  1.13it/s]
1|5048|Loss: 0.7627099752426147:  78%|███████▊  | 5048/6470 [1:06:29<21:30,  1.10it/s]
1|5049|Loss: 0.9019052386283875:  78%|███████▊  | 5048/6470 [1:06:30<21:30,  1.10it/s]
1|5049|Loss: 0.9019052386283875:  78%|███████▊  | 5049/6470 [1:06:30<20:55,  1.13it/s]
1|5050|Loss: 0.7929617166519165:  78%|███████▊  | 5049/6470 [1:06:31<20:55,  1.13it/s]
1|5050|Loss: 0.7929617166519165:  78%|███████▊  | 5050/6470 [1:06:31<19:45,  1.20it/s]
1|5051|Loss: 0.9374847412109375:  78%|███████▊  | 5050/6470 [1:06:31<19:45,  1.20it/s]
1|5051|Loss: 0.9374847412109375:  78%|███████▊  | 5051/6470 [1:06:32<18:40,  1.27it/s]
1|5052|Loss: 0.8505716323852539:  78%|███████▊  | 5051/6470 [1:06:32<18:40,  1.27it/s]
1|5052|Loss: 0.8505716323852539:  78%|███████▊  | 5052/6470 [1:06:32<19:10,  1.23it/s]
1|5053|Loss: 0.9139442443847656:  78%|███████▊  | 5052/6470 [1:06:33<19:10,  1.23it/s]
1|5053|Loss: 0.9139442443847656:  78%|███████▊  | 5053/6470 [1:06:33<19:19,  1.22it/s]
1|5054|Loss: 0.8150538206100464:  78%|███████▊  | 5053/6470 [1:06:34<19:19,  1.22it/s]
1|5054|Loss: 0.8150538206100464:  78%|███████▊  | 5054/6470 [1:06:34<19:59,  1.18it/s]
1|5055|Loss: 1.0237349271774292:  78%|███████▊  | 5054/6470 [1:06:35<19:59,  1.18it/s]
1|5055|Loss: 1.0237349271774292:  78%|███████▊  | 5055/6470 [1:06:35<18:31,  1.27it/s]
1|5056|Loss: 0.8960394263267517:  78%|███████▊  | 5055/6470 [1:06:35<18:31,  1.27it/s]
1|5056|Loss: 0.8960394263267517:  78%|███████▊  | 5056/6470 [1:06:36<18:58,  1.24it/s]
1|5057|Loss: 0.8093310594558716:  78%|███████▊  | 5056/6470 [1:06:36<18:58,  1.24it/s]
1|5057|Loss: 0.8093310594558716:  78%|███████▊  | 5057/6470 [1:06:36<18:27,  1.28it/s]
1|5058|Loss: 0.8119123578071594:  78%|███████▊  | 5057/6470 [1:06:37<18:27,  1.28it/s]
1|5058|Loss: 0.8119123578071594:  78%|███████▊  | 5058/6470 [1:06:37<19:07,  1.23it/s]
1|5059|Loss: 0.9848735332489014:  78%|███████▊  | 5058/6470 [1:06:38<19:07,  1.23it/s]
1|5059|Loss: 0.9848735332489014:  78%|███████▊  | 5059/6470 [1:06:38<19:35,  1.20it/s]
1|5060|Loss: 0.8414142727851868:  78%|███████▊  | 5059/6470 [1:06:39<19:35,  1.20it/s]
1|5060|Loss: 0.8414142727851868:  78%|███████▊  | 5060/6470 [1:06:39<19:28,  1.21it/s]
1|5061|Loss: 1.0158323049545288:  78%|███████▊  | 5060/6470 [1:06:39<19:28,  1.21it/s]
1|5061|Loss: 1.0158323049545288:  78%|███████▊  | 5061/6470 [1:06:40<18:35,  1.26it/s]
1|5062|Loss: 0.8319340944290161:  78%|███████▊  | 5061/6470 [1:06:40<18:35,  1.26it/s]
1|5062|Loss: 0.8319340944290161:  78%|███████▊  | 5062/6470 [1:06:40<18:16,  1.28it/s]
1|5063|Loss: 0.7685695290565491:  78%|███████▊  | 5062/6470 [1:06:41<18:16,  1.28it/s]
1|5063|Loss: 0.7685695290565491:  78%|███████▊  | 5063/6470 [1:06:41<18:07,  1.29it/s]
1|5064|Loss: 0.8719387650489807:  78%|███████▊  | 5063/6470 [1:06:42<18:07,  1.29it/s]
1|5064|Loss: 0.8719387650489807:  78%|███████▊  | 5064/6470 [1:06:42<18:49,  1.24it/s]
1|5065|Loss: 0.9186953902244568:  78%|███████▊  | 5064/6470 [1:06:43<18:49,  1.24it/s]
1|5065|Loss: 0.9186953902244568:  78%|███████▊  | 5065/6470 [1:06:43<17:43,  1.32it/s]
1|5066|Loss: 0.9240028858184814:  78%|███████▊  | 5065/6470 [1:06:43<17:43,  1.32it/s]
1|5066|Loss: 0.9240028858184814:  78%|███████▊  | 5066/6470 [1:06:43<17:01,  1.37it/s]
1|5067|Loss: 0.9507094621658325:  78%|███████▊  | 5066/6470 [1:06:44<17:01,  1.37it/s]
1|5067|Loss: 0.9507094621658325:  78%|███████▊  | 5067/6470 [1:06:44<17:18,  1.35it/s]
1|5068|Loss: 0.8371349573135376:  78%|███████▊  | 5067/6470 [1:06:45<17:18,  1.35it/s]
1|5068|Loss: 0.8371349573135376:  78%|███████▊  | 5068/6470 [1:06:45<18:15,  1.28it/s]
1|5069|Loss: 0.8879727721214294:  78%|███████▊  | 5068/6470 [1:06:46<18:15,  1.28it/s]
1|5069|Loss: 0.8879727721214294:  78%|███████▊  | 5069/6470 [1:06:46<18:55,  1.23it/s]
1|5070|Loss: 0.785417377948761:  78%|███████▊  | 5069/6470 [1:06:46<18:55,  1.23it/s] 
1|5070|Loss: 0.785417377948761:  78%|███████▊  | 5070/6470 [1:06:47<18:18,  1.27it/s]
1|5071|Loss: 0.743783712387085:  78%|███████▊  | 5070/6470 [1:06:47<18:18,  1.27it/s]
1|5071|Loss: 0.743783712387085:  78%|███████▊  | 5071/6470 [1:06:48<19:18,  1.21it/s]
1|5072|Loss: 0.8932504653930664:  78%|███████▊  | 5071/6470 [1:06:48<19:18,  1.21it/s]
1|5072|Loss: 0.8932504653930664:  78%|███████▊  | 5072/6470 [1:06:48<18:47,  1.24it/s]
1|5073|Loss: 0.870437502861023:  78%|███████▊  | 5072/6470 [1:06:49<18:47,  1.24it/s] 
1|5073|Loss: 0.870437502861023:  78%|███████▊  | 5073/6470 [1:06:49<18:38,  1.25it/s]
1|5074|Loss: 0.8549367189407349:  78%|███████▊  | 5073/6470 [1:06:50<18:38,  1.25it/s]
1|5074|Loss: 0.8549367189407349:  78%|███████▊  | 5074/6470 [1:06:50<17:46,  1.31it/s]
1|5075|Loss: 0.8918957114219666:  78%|███████▊  | 5074/6470 [1:06:50<17:46,  1.31it/s]
1|5075|Loss: 0.8918957114219666:  78%|███████▊  | 5075/6470 [1:06:50<15:32,  1.50it/s]
1|5076|Loss: 1.1429766416549683:  78%|███████▊  | 5075/6470 [1:06:51<15:32,  1.50it/s]
1|5076|Loss: 1.1429766416549683:  78%|███████▊  | 5076/6470 [1:06:51<16:36,  1.40it/s]
1|5077|Loss: 0.6827009916305542:  78%|███████▊  | 5076/6470 [1:06:51<16:36,  1.40it/s]
1|5077|Loss: 0.6827009916305542:  78%|███████▊  | 5077/6470 [1:06:52<16:00,  1.45it/s]
1|5078|Loss: 0.7877994179725647:  78%|███████▊  | 5077/6470 [1:06:52<16:00,  1.45it/s]
1|5078|Loss: 0.7877994179725647:  78%|███████▊  | 5078/6470 [1:06:52<15:48,  1.47it/s]
1|5079|Loss: 0.9542247653007507:  78%|███████▊  | 5078/6470 [1:06:53<15:48,  1.47it/s]
1|5079|Loss: 0.9542247653007507:  79%|███████▊  | 5079/6470 [1:06:53<16:35,  1.40it/s]
1|5080|Loss: 0.9979314208030701:  79%|███████▊  | 5079/6470 [1:06:54<16:35,  1.40it/s]
1|5080|Loss: 0.9979314208030701:  79%|███████▊  | 5080/6470 [1:06:54<18:06,  1.28it/s]
1|5081|Loss: 0.8105769157409668:  79%|███████▊  | 5080/6470 [1:06:55<18:06,  1.28it/s]
1|5081|Loss: 0.8105769157409668:  79%|███████▊  | 5081/6470 [1:06:55<19:14,  1.20it/s]
1|5082|Loss: 0.8744645118713379:  79%|███████▊  | 5081/6470 [1:06:55<19:14,  1.20it/s]
1|5082|Loss: 0.8744645118713379:  79%|███████▊  | 5082/6470 [1:06:56<17:58,  1.29it/s]
1|5083|Loss: 0.7775725722312927:  79%|███████▊  | 5082/6470 [1:06:56<17:58,  1.29it/s]
1|5083|Loss: 0.7775725722312927:  79%|███████▊  | 5083/6470 [1:06:56<17:34,  1.32it/s]
1|5084|Loss: 0.9428866505622864:  79%|███████▊  | 5083/6470 [1:06:57<17:34,  1.32it/s]
1|5084|Loss: 0.9428866505622864:  79%|███████▊  | 5084/6470 [1:06:57<17:43,  1.30it/s]
1|5085|Loss: 0.8379241824150085:  79%|███████▊  | 5084/6470 [1:06:58<17:43,  1.30it/s]
1|5085|Loss: 0.8379241824150085:  79%|███████▊  | 5085/6470 [1:06:58<16:11,  1.43it/s]
1|5086|Loss: 0.7936772704124451:  79%|███████▊  | 5085/6470 [1:06:58<16:11,  1.43it/s]
1|5086|Loss: 0.7936772704124451:  79%|███████▊  | 5086/6470 [1:06:58<15:51,  1.46it/s]
1|5087|Loss: 0.873807430267334:  79%|███████▊  | 5086/6470 [1:06:59<15:51,  1.46it/s] 
1|5087|Loss: 0.873807430267334:  79%|███████▊  | 5087/6470 [1:06:59<16:31,  1.39it/s]
1|5088|Loss: 0.9302265644073486:  79%|███████▊  | 5087/6470 [1:07:00<16:31,  1.39it/s]
1|5088|Loss: 0.9302265644073486:  79%|███████▊  | 5088/6470 [1:07:00<14:53,  1.55it/s]
1|5089|Loss: 0.8690744638442993:  79%|███████▊  | 5088/6470 [1:07:00<14:53,  1.55it/s]
1|5089|Loss: 0.8690744638442993:  79%|███████▊  | 5089/6470 [1:07:00<14:56,  1.54it/s]
1|5090|Loss: 0.9883974194526672:  79%|███████▊  | 5089/6470 [1:07:01<14:56,  1.54it/s]
1|5090|Loss: 0.9883974194526672:  79%|███████▊  | 5090/6470 [1:07:01<15:14,  1.51it/s]
1|5091|Loss: 0.8938308954238892:  79%|███████▊  | 5090/6470 [1:07:01<15:14,  1.51it/s]
1|5091|Loss: 0.8938308954238892:  79%|███████▊  | 5091/6470 [1:07:02<14:37,  1.57it/s]
1|5092|Loss: 0.9659115672111511:  79%|███████▊  | 5091/6470 [1:07:02<14:37,  1.57it/s]
1|5092|Loss: 0.9659115672111511:  79%|███████▊  | 5092/6470 [1:07:02<15:23,  1.49it/s]
1|5093|Loss: 0.7890642881393433:  79%|███████▊  | 5092/6470 [1:07:03<15:23,  1.49it/s]
1|5093|Loss: 0.7890642881393433:  79%|███████▊  | 5093/6470 [1:07:03<15:44,  1.46it/s]
1|5094|Loss: 0.9562427997589111:  79%|███████▊  | 5093/6470 [1:07:03<15:44,  1.46it/s]
1|5094|Loss: 0.9562427997589111:  79%|███████▊  | 5094/6470 [1:07:04<15:28,  1.48it/s]
1|5095|Loss: 0.9186531901359558:  79%|███████▊  | 5094/6470 [1:07:04<15:28,  1.48it/s]
1|5095|Loss: 0.9186531901359558:  79%|███████▊  | 5095/6470 [1:07:05<16:29,  1.39it/s]
1|5096|Loss: 1.0218678712844849:  79%|███████▊  | 5095/6470 [1:07:05<16:29,  1.39it/s]
1|5096|Loss: 1.0218678712844849:  79%|███████▉  | 5096/6470 [1:07:05<17:54,  1.28it/s]
1|5097|Loss: 0.9457414150238037:  79%|███████▉  | 5096/6470 [1:07:06<17:54,  1.28it/s]
1|5097|Loss: 0.9457414150238037:  79%|███████▉  | 5097/6470 [1:07:06<18:26,  1.24it/s]
1|5098|Loss: 0.7798791527748108:  79%|███████▉  | 5097/6470 [1:07:07<18:26,  1.24it/s]
1|5098|Loss: 0.7798791527748108:  79%|███████▉  | 5098/6470 [1:07:07<18:47,  1.22it/s]
1|5099|Loss: 0.8199386596679688:  79%|███████▉  | 5098/6470 [1:07:08<18:47,  1.22it/s]
1|5099|Loss: 0.8199386596679688:  79%|███████▉  | 5099/6470 [1:07:08<18:27,  1.24it/s]
1|5100|Loss: 0.8269620537757874:  79%|███████▉  | 5099/6470 [1:07:08<18:27,  1.24it/s]
1|5100|Loss: 0.8269620537757874:  79%|███████▉  | 5100/6470 [1:07:09<18:19,  1.25it/s]
1|5101|Loss: 0.7854345440864563:  79%|███████▉  | 5100/6470 [1:07:09<18:19,  1.25it/s]
1|5101|Loss: 0.7854345440864563:  79%|███████▉  | 5101/6470 [1:07:09<17:39,  1.29it/s]
1|5102|Loss: 0.8082156181335449:  79%|███████▉  | 5101/6470 [1:07:10<17:39,  1.29it/s]
1|5102|Loss: 0.8082156181335449:  79%|███████▉  | 5102/6470 [1:07:10<16:41,  1.37it/s]
1|5103|Loss: 0.7910040616989136:  79%|███████▉  | 5102/6470 [1:07:10<16:41,  1.37it/s]
1|5103|Loss: 0.7910040616989136:  79%|███████▉  | 5103/6470 [1:07:11<15:55,  1.43it/s]
1|5104|Loss: 0.7461574673652649:  79%|███████▉  | 5103/6470 [1:07:11<15:55,  1.43it/s]
1|5104|Loss: 0.7461574673652649:  79%|███████▉  | 5104/6470 [1:07:11<15:27,  1.47it/s]
1|5105|Loss: 0.7684369087219238:  79%|███████▉  | 5104/6470 [1:07:12<15:27,  1.47it/s]
1|5105|Loss: 0.7684369087219238:  79%|███████▉  | 5105/6470 [1:07:12<17:13,  1.32it/s]
1|5106|Loss: 0.7590283751487732:  79%|███████▉  | 5105/6470 [1:07:13<17:13,  1.32it/s]
1|5106|Loss: 0.7590283751487732:  79%|███████▉  | 5106/6470 [1:07:13<17:48,  1.28it/s]
1|5107|Loss: 0.8597098588943481:  79%|███████▉  | 5106/6470 [1:07:14<17:48,  1.28it/s]
1|5107|Loss: 0.8597098588943481:  79%|███████▉  | 5107/6470 [1:07:14<17:17,  1.31it/s]
1|5108|Loss: 0.9463657140731812:  79%|███████▉  | 5107/6470 [1:07:14<17:17,  1.31it/s]
1|5108|Loss: 0.9463657140731812:  79%|███████▉  | 5108/6470 [1:07:15<18:06,  1.25it/s]
1|5109|Loss: 0.7613109946250916:  79%|███████▉  | 5108/6470 [1:07:15<18:06,  1.25it/s]
1|5109|Loss: 0.7613109946250916:  79%|███████▉  | 5109/6470 [1:07:16<18:05,  1.25it/s]
1|5110|Loss: 0.8267747163772583:  79%|███████▉  | 5109/6470 [1:07:16<18:05,  1.25it/s]
1|5110|Loss: 0.8267747163772583:  79%|███████▉  | 5110/6470 [1:07:16<18:00,  1.26it/s]
1|5111|Loss: 0.9555250406265259:  79%|███████▉  | 5110/6470 [1:07:17<18:00,  1.26it/s]
1|5111|Loss: 0.9555250406265259:  79%|███████▉  | 5111/6470 [1:07:17<17:19,  1.31it/s]
1|5112|Loss: 0.704640805721283:  79%|███████▉  | 5111/6470 [1:07:18<17:19,  1.31it/s] 
1|5112|Loss: 0.704640805721283:  79%|███████▉  | 5112/6470 [1:07:18<18:51,  1.20it/s]
1|5113|Loss: 0.8704589009284973:  79%|███████▉  | 5112/6470 [1:07:19<18:51,  1.20it/s]
1|5113|Loss: 0.8704589009284973:  79%|███████▉  | 5113/6470 [1:07:19<18:55,  1.20it/s]
1|5114|Loss: 0.9636632800102234:  79%|███████▉  | 5113/6470 [1:07:19<18:55,  1.20it/s]
1|5114|Loss: 0.9636632800102234:  79%|███████▉  | 5114/6470 [1:07:20<17:58,  1.26it/s]
1|5115|Loss: 0.8459604382514954:  79%|███████▉  | 5114/6470 [1:07:20<17:58,  1.26it/s]
1|5115|Loss: 0.8459604382514954:  79%|███████▉  | 5115/6470 [1:07:20<17:35,  1.28it/s]
1|5116|Loss: 1.0131855010986328:  79%|███████▉  | 5115/6470 [1:07:21<17:35,  1.28it/s]
1|5116|Loss: 1.0131855010986328:  79%|███████▉  | 5116/6470 [1:07:21<17:42,  1.27it/s]
1|5117|Loss: 0.8080419898033142:  79%|███████▉  | 5116/6470 [1:07:22<17:42,  1.27it/s]
1|5117|Loss: 0.8080419898033142:  79%|███████▉  | 5117/6470 [1:07:22<17:19,  1.30it/s]
1|5118|Loss: 1.0455015897750854:  79%|███████▉  | 5117/6470 [1:07:22<17:19,  1.30it/s]
1|5118|Loss: 1.0455015897750854:  79%|███████▉  | 5118/6470 [1:07:23<16:54,  1.33it/s]
1|5119|Loss: 0.7598644495010376:  79%|███████▉  | 5118/6470 [1:07:23<16:54,  1.33it/s]
1|5119|Loss: 0.7598644495010376:  79%|███████▉  | 5119/6470 [1:07:23<16:29,  1.36it/s]
1|5120|Loss: 0.8267554640769958:  79%|███████▉  | 5119/6470 [1:07:24<16:29,  1.36it/s]
1|5120|Loss: 0.8267554640769958:  79%|███████▉  | 5120/6470 [1:07:24<16:32,  1.36it/s]
1|5121|Loss: 0.9935958385467529:  79%|███████▉  | 5120/6470 [1:07:24<16:32,  1.36it/s]
1|5121|Loss: 0.9935958385467529:  79%|███████▉  | 5121/6470 [1:07:25<16:59,  1.32it/s]
1|5122|Loss: 0.912011444568634:  79%|███████▉  | 5121/6470 [1:07:25<16:59,  1.32it/s] 
1|5122|Loss: 0.912011444568634:  79%|███████▉  | 5122/6470 [1:07:26<16:51,  1.33it/s]
1|5123|Loss: 0.850570797920227:  79%|███████▉  | 5122/6470 [1:07:26<16:51,  1.33it/s]
1|5123|Loss: 0.850570797920227:  79%|███████▉  | 5123/6470 [1:07:26<17:54,  1.25it/s]
1|5124|Loss: 0.8382074236869812:  79%|███████▉  | 5123/6470 [1:07:27<17:54,  1.25it/s]
1|5124|Loss: 0.8382074236869812:  79%|███████▉  | 5124/6470 [1:07:27<18:11,  1.23it/s]
1|5125|Loss: 0.7449687123298645:  79%|███████▉  | 5124/6470 [1:07:28<18:11,  1.23it/s]
1|5125|Loss: 0.7449687123298645:  79%|███████▉  | 5125/6470 [1:07:28<18:36,  1.21it/s]
1|5126|Loss: 0.8669016361236572:  79%|███████▉  | 5125/6470 [1:07:28<18:36,  1.21it/s]
1|5126|Loss: 0.8669016361236572:  79%|███████▉  | 5126/6470 [1:07:29<16:47,  1.33it/s]
1|5127|Loss: 0.8549411296844482:  79%|███████▉  | 5126/6470 [1:07:29<16:47,  1.33it/s]
1|5127|Loss: 0.8549411296844482:  79%|███████▉  | 5127/6470 [1:07:29<14:56,  1.50it/s]
1|5128|Loss: 0.9403815865516663:  79%|███████▉  | 5127/6470 [1:07:30<14:56,  1.50it/s]
1|5128|Loss: 0.9403815865516663:  79%|███████▉  | 5128/6470 [1:07:30<14:47,  1.51it/s]
1|5129|Loss: 0.9692167043685913:  79%|███████▉  | 5128/6470 [1:07:30<14:47,  1.51it/s]
1|5129|Loss: 0.9692167043685913:  79%|███████▉  | 5129/6470 [1:07:31<15:33,  1.44it/s]
1|5130|Loss: 0.7891005277633667:  79%|███████▉  | 5129/6470 [1:07:31<15:33,  1.44it/s]
1|5130|Loss: 0.7891005277633667:  79%|███████▉  | 5130/6470 [1:07:31<16:19,  1.37it/s]
1|5131|Loss: 0.9255873560905457:  79%|███████▉  | 5130/6470 [1:07:32<16:19,  1.37it/s]
1|5131|Loss: 0.9255873560905457:  79%|███████▉  | 5131/6470 [1:07:32<17:24,  1.28it/s]
1|5132|Loss: 0.7925580739974976:  79%|███████▉  | 5131/6470 [1:07:33<17:24,  1.28it/s]
1|5132|Loss: 0.7925580739974976:  79%|███████▉  | 5132/6470 [1:07:33<15:56,  1.40it/s]
1|5133|Loss: 0.8645461797714233:  79%|███████▉  | 5132/6470 [1:07:33<15:56,  1.40it/s]
1|5133|Loss: 0.8645461797714233:  79%|███████▉  | 5133/6470 [1:07:34<16:11,  1.38it/s]
1|5134|Loss: 0.775493323802948:  79%|███████▉  | 5133/6470 [1:07:34<16:11,  1.38it/s] 
1|5134|Loss: 0.775493323802948:  79%|███████▉  | 5134/6470 [1:07:35<17:30,  1.27it/s]
1|5135|Loss: 0.9205470085144043:  79%|███████▉  | 5134/6470 [1:07:35<17:30,  1.27it/s]
1|5135|Loss: 0.9205470085144043:  79%|███████▉  | 5135/6470 [1:07:35<18:22,  1.21it/s]
1|5136|Loss: 0.9162256121635437:  79%|███████▉  | 5135/6470 [1:07:36<18:22,  1.21it/s]
1|5136|Loss: 0.9162256121635437:  79%|███████▉  | 5136/6470 [1:07:36<17:15,  1.29it/s]
1|5137|Loss: 0.8248718976974487:  79%|███████▉  | 5136/6470 [1:07:36<17:15,  1.29it/s]
1|5137|Loss: 0.8248718976974487:  79%|███████▉  | 5137/6470 [1:07:37<17:02,  1.30it/s]
1|5138|Loss: 0.9980998635292053:  79%|███████▉  | 5137/6470 [1:07:37<17:02,  1.30it/s]
1|5138|Loss: 0.9980998635292053:  79%|███████▉  | 5138/6470 [1:07:38<17:18,  1.28it/s]
1|5139|Loss: 0.9816108345985413:  79%|███████▉  | 5138/6470 [1:07:38<17:18,  1.28it/s]
1|5139|Loss: 0.9816108345985413:  79%|███████▉  | 5139/6470 [1:07:38<17:21,  1.28it/s]
1|5140|Loss: 0.7846729159355164:  79%|███████▉  | 5139/6470 [1:07:39<17:21,  1.28it/s]
1|5140|Loss: 0.7846729159355164:  79%|███████▉  | 5140/6470 [1:07:39<17:21,  1.28it/s]
1|5141|Loss: 0.9124801158905029:  79%|███████▉  | 5140/6470 [1:07:40<17:21,  1.28it/s]
1|5141|Loss: 0.9124801158905029:  79%|███████▉  | 5141/6470 [1:07:40<17:15,  1.28it/s]
1|5142|Loss: 0.7705086469650269:  79%|███████▉  | 5141/6470 [1:07:40<17:15,  1.28it/s]
1|5142|Loss: 0.7705086469650269:  79%|███████▉  | 5142/6470 [1:07:41<16:46,  1.32it/s]
1|5143|Loss: 0.6940284967422485:  79%|███████▉  | 5142/6470 [1:07:41<16:46,  1.32it/s]
1|5143|Loss: 0.6940284967422485:  79%|███████▉  | 5143/6470 [1:07:41<16:18,  1.36it/s]
1|5144|Loss: 0.9180521368980408:  79%|███████▉  | 5143/6470 [1:07:42<16:18,  1.36it/s]
1|5144|Loss: 0.9180521368980408:  80%|███████▉  | 5144/6470 [1:07:42<15:47,  1.40it/s]
1|5145|Loss: 0.7919101119041443:  80%|███████▉  | 5144/6470 [1:07:42<15:47,  1.40it/s]
1|5145|Loss: 0.7919101119041443:  80%|███████▉  | 5145/6470 [1:07:43<16:06,  1.37it/s]
1|5146|Loss: 0.7691469192504883:  80%|███████▉  | 5145/6470 [1:07:43<16:06,  1.37it/s]
1|5146|Loss: 0.7691469192504883:  80%|███████▉  | 5146/6470 [1:07:44<17:58,  1.23it/s]
1|5147|Loss: 0.8639565110206604:  80%|███████▉  | 5146/6470 [1:07:44<17:58,  1.23it/s]
1|5147|Loss: 0.8639565110206604:  80%|███████▉  | 5147/6470 [1:07:44<16:37,  1.33it/s]
1|5148|Loss: 0.9274290800094604:  80%|███████▉  | 5147/6470 [1:07:45<16:37,  1.33it/s]
1|5148|Loss: 0.9274290800094604:  80%|███████▉  | 5148/6470 [1:07:45<15:27,  1.43it/s]
1|5149|Loss: 0.6750837564468384:  80%|███████▉  | 5148/6470 [1:07:45<15:27,  1.43it/s]
1|5149|Loss: 0.6750837564468384:  80%|███████▉  | 5149/6470 [1:07:46<15:49,  1.39it/s]
1|5150|Loss: 0.7710200548171997:  80%|███████▉  | 5149/6470 [1:07:46<15:49,  1.39it/s]
1|5150|Loss: 0.7710200548171997:  80%|███████▉  | 5150/6470 [1:07:46<14:05,  1.56it/s]
1|5151|Loss: 0.9075611233711243:  80%|███████▉  | 5150/6470 [1:07:47<14:05,  1.56it/s]
1|5151|Loss: 0.9075611233711243:  80%|███████▉  | 5151/6470 [1:07:47<15:18,  1.44it/s]
1|5152|Loss: 0.7464255690574646:  80%|███████▉  | 5151/6470 [1:07:48<15:18,  1.44it/s]
1|5152|Loss: 0.7464255690574646:  80%|███████▉  | 5152/6470 [1:07:48<16:51,  1.30it/s]
1|5153|Loss: 0.8896470665931702:  80%|███████▉  | 5152/6470 [1:07:48<16:51,  1.30it/s]
1|5153|Loss: 0.8896470665931702:  80%|███████▉  | 5153/6470 [1:07:49<17:11,  1.28it/s]
1|5154|Loss: 0.9123218655586243:  80%|███████▉  | 5153/6470 [1:07:49<17:11,  1.28it/s]
1|5154|Loss: 0.9123218655586243:  80%|███████▉  | 5154/6470 [1:07:50<17:54,  1.22it/s]
1|5155|Loss: 0.7511256337165833:  80%|███████▉  | 5154/6470 [1:07:50<17:54,  1.22it/s]
1|5155|Loss: 0.7511256337165833:  80%|███████▉  | 5155/6470 [1:07:50<16:42,  1.31it/s]
1|5156|Loss: 0.9076292514801025:  80%|███████▉  | 5155/6470 [1:07:51<16:42,  1.31it/s]
1|5156|Loss: 0.9076292514801025:  80%|███████▉  | 5156/6470 [1:07:51<16:36,  1.32it/s]
1|5157|Loss: 0.9852892160415649:  80%|███████▉  | 5156/6470 [1:07:52<16:36,  1.32it/s]
1|5157|Loss: 0.9852892160415649:  80%|███████▉  | 5157/6470 [1:07:52<16:48,  1.30it/s]
1|5158|Loss: 0.9153406023979187:  80%|███████▉  | 5157/6470 [1:07:52<16:48,  1.30it/s]
1|5158|Loss: 0.9153406023979187:  80%|███████▉  | 5158/6470 [1:07:53<16:30,  1.32it/s]
1|5159|Loss: 0.8489702939987183:  80%|███████▉  | 5158/6470 [1:07:53<16:30,  1.32it/s]
1|5159|Loss: 0.8489702939987183:  80%|███████▉  | 5159/6470 [1:07:53<16:14,  1.35it/s]
1|5160|Loss: 0.9246718883514404:  80%|███████▉  | 5159/6470 [1:07:54<16:14,  1.35it/s]
1|5160|Loss: 0.9246718883514404:  80%|███████▉  | 5160/6470 [1:07:54<16:28,  1.33it/s]
1|5161|Loss: 0.9694258570671082:  80%|███████▉  | 5160/6470 [1:07:55<16:28,  1.33it/s]
1|5161|Loss: 0.9694258570671082:  80%|███████▉  | 5161/6470 [1:07:55<16:16,  1.34it/s]
1|5162|Loss: 0.8544161915779114:  80%|███████▉  | 5161/6470 [1:07:55<16:16,  1.34it/s]
1|5162|Loss: 0.8544161915779114:  80%|███████▉  | 5162/6470 [1:07:56<16:30,  1.32it/s]
1|5163|Loss: 0.9604761004447937:  80%|███████▉  | 5162/6470 [1:07:56<16:30,  1.32it/s]
1|5163|Loss: 0.9604761004447937:  80%|███████▉  | 5163/6470 [1:07:57<17:30,  1.24it/s]
1|5164|Loss: 0.7677170634269714:  80%|███████▉  | 5163/6470 [1:07:57<17:30,  1.24it/s]
1|5164|Loss: 0.7677170634269714:  80%|███████▉  | 5164/6470 [1:07:57<17:09,  1.27it/s]
1|5165|Loss: 1.0559897422790527:  80%|███████▉  | 5164/6470 [1:07:58<17:09,  1.27it/s]
1|5165|Loss: 1.0559897422790527:  80%|███████▉  | 5165/6470 [1:07:58<16:58,  1.28it/s]
1|5166|Loss: 0.9147849678993225:  80%|███████▉  | 5165/6470 [1:07:59<16:58,  1.28it/s]
1|5166|Loss: 0.9147849678993225:  80%|███████▉  | 5166/6470 [1:07:59<17:41,  1.23it/s]
1|5167|Loss: 0.9848647713661194:  80%|███████▉  | 5166/6470 [1:07:59<17:41,  1.23it/s]
1|5167|Loss: 0.9848647713661194:  80%|███████▉  | 5167/6470 [1:08:00<17:15,  1.26it/s]
1|5168|Loss: 0.7203430533409119:  80%|███████▉  | 5167/6470 [1:08:00<17:15,  1.26it/s]
1|5168|Loss: 0.7203430533409119:  80%|███████▉  | 5168/6470 [1:08:01<17:43,  1.22it/s]
1|5169|Loss: 0.7992417812347412:  80%|███████▉  | 5168/6470 [1:08:01<17:43,  1.22it/s]
1|5169|Loss: 0.7992417812347412:  80%|███████▉  | 5169/6470 [1:08:01<17:45,  1.22it/s]
1|5170|Loss: 0.8385487794876099:  80%|███████▉  | 5169/6470 [1:08:02<17:45,  1.22it/s]
1|5170|Loss: 0.8385487794876099:  80%|███████▉  | 5170/6470 [1:08:02<17:26,  1.24it/s]
1|5171|Loss: 0.8201665282249451:  80%|███████▉  | 5170/6470 [1:08:03<17:26,  1.24it/s]
1|5171|Loss: 0.8201665282249451:  80%|███████▉  | 5171/6470 [1:08:03<15:52,  1.36it/s]
1|5172|Loss: 0.8695555329322815:  80%|███████▉  | 5171/6470 [1:08:03<15:52,  1.36it/s]
1|5172|Loss: 0.8695555329322815:  80%|███████▉  | 5172/6470 [1:08:03<14:59,  1.44it/s]
1|5173|Loss: 0.814780056476593:  80%|███████▉  | 5172/6470 [1:08:04<14:59,  1.44it/s] 
1|5173|Loss: 0.814780056476593:  80%|███████▉  | 5173/6470 [1:08:04<14:49,  1.46it/s]
1|5174|Loss: 0.9407117962837219:  80%|███████▉  | 5173/6470 [1:08:04<14:49,  1.46it/s]
1|5174|Loss: 0.9407117962837219:  80%|███████▉  | 5174/6470 [1:08:05<16:05,  1.34it/s]
1|5175|Loss: 0.8571832180023193:  80%|███████▉  | 5174/6470 [1:08:05<16:05,  1.34it/s]
1|5175|Loss: 0.8571832180023193:  80%|███████▉  | 5175/6470 [1:08:06<17:23,  1.24it/s]
1|5176|Loss: 0.900502622127533:  80%|███████▉  | 5175/6470 [1:08:06<17:23,  1.24it/s] 
1|5176|Loss: 0.900502622127533:  80%|████████  | 5176/6470 [1:08:07<17:26,  1.24it/s]
1|5177|Loss: 0.8435408473014832:  80%|████████  | 5176/6470 [1:08:07<17:26,  1.24it/s]
1|5177|Loss: 0.8435408473014832:  80%|████████  | 5177/6470 [1:08:07<17:10,  1.25it/s]
1|5178|Loss: 0.996819794178009:  80%|████████  | 5177/6470 [1:08:08<17:10,  1.25it/s] 
1|5178|Loss: 0.996819794178009:  80%|████████  | 5178/6470 [1:08:08<17:11,  1.25it/s]
1|5179|Loss: 0.8447100520133972:  80%|████████  | 5178/6470 [1:08:09<17:11,  1.25it/s]
1|5179|Loss: 0.8447100520133972:  80%|████████  | 5179/6470 [1:08:09<15:09,  1.42it/s]
1|5180|Loss: 0.771523654460907:  80%|████████  | 5179/6470 [1:08:09<15:09,  1.42it/s] 
1|5180|Loss: 0.771523654460907:  80%|████████  | 5180/6470 [1:08:09<15:09,  1.42it/s]
1|5181|Loss: 0.7398990392684937:  80%|████████  | 5180/6470 [1:08:10<15:09,  1.42it/s]
1|5181|Loss: 0.7398990392684937:  80%|████████  | 5181/6470 [1:08:10<14:51,  1.45it/s]
1|5182|Loss: 0.871056854724884:  80%|████████  | 5181/6470 [1:08:11<14:51,  1.45it/s] 
1|5182|Loss: 0.871056854724884:  80%|████████  | 5182/6470 [1:08:11<16:26,  1.31it/s]
1|5183|Loss: 1.0466725826263428:  80%|████████  | 5182/6470 [1:08:11<16:26,  1.31it/s]
1|5183|Loss: 1.0466725826263428:  80%|████████  | 5183/6470 [1:08:12<16:57,  1.26it/s]
1|5184|Loss: 0.9417297840118408:  80%|████████  | 5183/6470 [1:08:12<16:57,  1.26it/s]
1|5184|Loss: 0.9417297840118408:  80%|████████  | 5184/6470 [1:08:13<16:30,  1.30it/s]
1|5185|Loss: 0.8772433400154114:  80%|████████  | 5184/6470 [1:08:13<16:30,  1.30it/s]
1|5185|Loss: 0.8772433400154114:  80%|████████  | 5185/6470 [1:08:13<16:25,  1.30it/s]
1|5186|Loss: 0.8768804669380188:  80%|████████  | 5185/6470 [1:08:14<16:25,  1.30it/s]
1|5186|Loss: 0.8768804669380188:  80%|████████  | 5186/6470 [1:08:14<17:25,  1.23it/s]
1|5187|Loss: 0.790459394454956:  80%|████████  | 5186/6470 [1:08:15<17:25,  1.23it/s] 
1|5187|Loss: 0.790459394454956:  80%|████████  | 5187/6470 [1:08:15<16:24,  1.30it/s]
1|5188|Loss: 1.038503646850586:  80%|████████  | 5187/6470 [1:08:15<16:24,  1.30it/s]
1|5188|Loss: 1.038503646850586:  80%|████████  | 5188/6470 [1:08:16<16:57,  1.26it/s]
1|5189|Loss: 0.9050816893577576:  80%|████████  | 5188/6470 [1:08:16<16:57,  1.26it/s]
1|5189|Loss: 0.9050816893577576:  80%|████████  | 5189/6470 [1:08:17<17:04,  1.25it/s]
1|5190|Loss: 0.9228916764259338:  80%|████████  | 5189/6470 [1:08:17<17:04,  1.25it/s]
1|5190|Loss: 0.9228916764259338:  80%|████████  | 5190/6470 [1:08:17<17:10,  1.24it/s]
1|5191|Loss: 0.6953293681144714:  80%|████████  | 5190/6470 [1:08:18<17:10,  1.24it/s]
1|5191|Loss: 0.6953293681144714:  80%|████████  | 5191/6470 [1:08:18<15:35,  1.37it/s]
1|5192|Loss: 0.6480080485343933:  80%|████████  | 5191/6470 [1:08:18<15:35,  1.37it/s]
1|5192|Loss: 0.6480080485343933:  80%|████████  | 5192/6470 [1:08:19<14:45,  1.44it/s]
1|5193|Loss: 0.8655216693878174:  80%|████████  | 5192/6470 [1:08:19<14:45,  1.44it/s]
1|5193|Loss: 0.8655216693878174:  80%|████████  | 5193/6470 [1:08:19<14:16,  1.49it/s]
1|5194|Loss: 0.9159857630729675:  80%|████████  | 5193/6470 [1:08:20<14:16,  1.49it/s]
1|5194|Loss: 0.9159857630729675:  80%|████████  | 5194/6470 [1:08:20<14:13,  1.49it/s]
1|5195|Loss: 0.9651509523391724:  80%|████████  | 5194/6470 [1:08:20<14:13,  1.49it/s]
1|5195|Loss: 0.9651509523391724:  80%|████████  | 5195/6470 [1:08:21<14:46,  1.44it/s]
1|5196|Loss: 1.0365486145019531:  80%|████████  | 5195/6470 [1:08:21<14:46,  1.44it/s]
1|5196|Loss: 1.0365486145019531:  80%|████████  | 5196/6470 [1:08:21<15:22,  1.38it/s]
1|5197|Loss: 0.8687664270401001:  80%|████████  | 5196/6470 [1:08:22<15:22,  1.38it/s]
1|5197|Loss: 0.8687664270401001:  80%|████████  | 5197/6470 [1:08:22<15:19,  1.38it/s]
1|5198|Loss: 0.8599228262901306:  80%|████████  | 5197/6470 [1:08:23<15:19,  1.38it/s]
1|5198|Loss: 0.8599228262901306:  80%|████████  | 5198/6470 [1:08:23<15:28,  1.37it/s]
1|5199|Loss: 0.8805241584777832:  80%|████████  | 5198/6470 [1:08:23<15:28,  1.37it/s]
1|5199|Loss: 0.8805241584777832:  80%|████████  | 5199/6470 [1:08:24<15:21,  1.38it/s]
1|5200|Loss: 1.0199297666549683:  80%|████████  | 5199/6470 [1:08:24<15:21,  1.38it/s]
1|5200|Loss: 1.0199297666549683:  80%|████████  | 5200/6470 [1:08:24<15:31,  1.36it/s]
1|5201|Loss: 0.8108766078948975:  80%|████████  | 5200/6470 [1:08:25<15:31,  1.36it/s]
1|5201|Loss: 0.8108766078948975:  80%|████████  | 5201/6470 [1:08:25<16:26,  1.29it/s]
1|5202|Loss: 0.828486442565918:  80%|████████  | 5201/6470 [1:08:26<16:26,  1.29it/s] 
1|5202|Loss: 0.828486442565918:  80%|████████  | 5202/6470 [1:08:26<16:05,  1.31it/s]
1|5203|Loss: 0.9428360462188721:  80%|████████  | 5202/6470 [1:08:26<16:05,  1.31it/s]
1|5203|Loss: 0.9428360462188721:  80%|████████  | 5203/6470 [1:08:27<16:37,  1.27it/s]
1|5204|Loss: 0.810683012008667:  80%|████████  | 5203/6470 [1:08:27<16:37,  1.27it/s] 
1|5204|Loss: 0.810683012008667:  80%|████████  | 5204/6470 [1:08:28<16:16,  1.30it/s]
1|5205|Loss: 0.8636502027511597:  80%|████████  | 5204/6470 [1:08:28<16:16,  1.30it/s]
1|5205|Loss: 0.8636502027511597:  80%|████████  | 5205/6470 [1:08:28<16:10,  1.30it/s]
1|5206|Loss: 0.8995293974876404:  80%|████████  | 5205/6470 [1:08:29<16:10,  1.30it/s]
1|5206|Loss: 0.8995293974876404:  80%|████████  | 5206/6470 [1:08:29<15:30,  1.36it/s]
1|5207|Loss: 0.8553844094276428:  80%|████████  | 5206/6470 [1:08:29<15:30,  1.36it/s]
1|5207|Loss: 0.8553844094276428:  80%|████████  | 5207/6470 [1:08:30<16:17,  1.29it/s]
1|5208|Loss: 0.8788695335388184:  80%|████████  | 5207/6470 [1:08:30<16:17,  1.29it/s]
1|5208|Loss: 0.8788695335388184:  80%|████████  | 5208/6470 [1:08:31<16:05,  1.31it/s]
1|5209|Loss: 0.8303744792938232:  80%|████████  | 5208/6470 [1:08:31<16:05,  1.31it/s]
1|5209|Loss: 0.8303744792938232:  81%|████████  | 5209/6470 [1:08:31<16:59,  1.24it/s]
1|5210|Loss: 0.8650162816047668:  81%|████████  | 5209/6470 [1:08:32<16:59,  1.24it/s]
1|5210|Loss: 0.8650162816047668:  81%|████████  | 5210/6470 [1:08:32<18:06,  1.16it/s]
1|5211|Loss: 0.6703609824180603:  81%|████████  | 5210/6470 [1:08:33<18:06,  1.16it/s]
1|5211|Loss: 0.6703609824180603:  81%|████████  | 5211/6470 [1:08:33<17:50,  1.18it/s]
1|5212|Loss: 0.7435529232025146:  81%|████████  | 5211/6470 [1:08:34<17:50,  1.18it/s]
1|5212|Loss: 0.7435529232025146:  81%|████████  | 5212/6470 [1:08:34<15:46,  1.33it/s]
1|5213|Loss: 0.7557863593101501:  81%|████████  | 5212/6470 [1:08:34<15:46,  1.33it/s]
1|5213|Loss: 0.7557863593101501:  81%|████████  | 5213/6470 [1:08:34<15:22,  1.36it/s]
1|5214|Loss: 0.8648890852928162:  81%|████████  | 5213/6470 [1:08:35<15:22,  1.36it/s]
1|5214|Loss: 0.8648890852928162:  81%|████████  | 5214/6470 [1:08:35<15:46,  1.33it/s]
1|5215|Loss: 1.0117202997207642:  81%|████████  | 5214/6470 [1:08:36<15:46,  1.33it/s]
1|5215|Loss: 1.0117202997207642:  81%|████████  | 5215/6470 [1:08:36<16:40,  1.25it/s]
1|5216|Loss: 0.9419193267822266:  81%|████████  | 5215/6470 [1:08:37<16:40,  1.25it/s]
1|5216|Loss: 0.9419193267822266:  81%|████████  | 5216/6470 [1:08:37<16:08,  1.29it/s]
1|5217|Loss: 0.8667523860931396:  81%|████████  | 5216/6470 [1:08:37<16:08,  1.29it/s]
1|5217|Loss: 0.8667523860931396:  81%|████████  | 5217/6470 [1:08:38<15:32,  1.34it/s]
1|5218|Loss: 0.8116310238838196:  81%|████████  | 5217/6470 [1:08:38<15:32,  1.34it/s]
1|5218|Loss: 0.8116310238838196:  81%|████████  | 5218/6470 [1:08:38<15:19,  1.36it/s]
1|5219|Loss: 0.8476574420928955:  81%|████████  | 5218/6470 [1:08:39<15:19,  1.36it/s]
1|5219|Loss: 0.8476574420928955:  81%|████████  | 5219/6470 [1:08:39<15:58,  1.31it/s]
1|5220|Loss: 0.8148462772369385:  81%|████████  | 5219/6470 [1:08:39<15:58,  1.31it/s]
1|5220|Loss: 0.8148462772369385:  81%|████████  | 5220/6470 [1:08:40<14:40,  1.42it/s]
1|5221|Loss: 0.8291490077972412:  81%|████████  | 5220/6470 [1:08:40<14:40,  1.42it/s]
1|5221|Loss: 0.8291490077972412:  81%|████████  | 5221/6470 [1:08:40<14:34,  1.43it/s]
1|5222|Loss: 0.8001422882080078:  81%|████████  | 5221/6470 [1:08:41<14:34,  1.43it/s]
1|5222|Loss: 0.8001422882080078:  81%|████████  | 5222/6470 [1:08:41<14:45,  1.41it/s]
1|5223|Loss: 0.9353586435317993:  81%|████████  | 5222/6470 [1:08:42<14:45,  1.41it/s]
1|5223|Loss: 0.9353586435317993:  81%|████████  | 5223/6470 [1:08:42<15:14,  1.36it/s]
1|5224|Loss: 0.8756230473518372:  81%|████████  | 5223/6470 [1:08:42<15:14,  1.36it/s]
1|5224|Loss: 0.8756230473518372:  81%|████████  | 5224/6470 [1:08:43<16:08,  1.29it/s]
1|5225|Loss: 1.004510521888733:  81%|████████  | 5224/6470 [1:08:43<16:08,  1.29it/s] 
1|5225|Loss: 1.004510521888733:  81%|████████  | 5225/6470 [1:08:43<15:12,  1.36it/s]
1|5226|Loss: 0.9113612771034241:  81%|████████  | 5225/6470 [1:08:44<15:12,  1.36it/s]
1|5226|Loss: 0.9113612771034241:  81%|████████  | 5226/6470 [1:08:44<14:43,  1.41it/s]
1|5227|Loss: 0.9018161296844482:  81%|████████  | 5226/6470 [1:08:45<14:43,  1.41it/s]
1|5227|Loss: 0.9018161296844482:  81%|████████  | 5227/6470 [1:08:45<15:17,  1.36it/s]
1|5228|Loss: 0.9342138767242432:  81%|████████  | 5227/6470 [1:08:45<15:17,  1.36it/s]
1|5228|Loss: 0.9342138767242432:  81%|████████  | 5228/6470 [1:08:46<15:44,  1.31it/s]
1|5229|Loss: 0.8306779265403748:  81%|████████  | 5228/6470 [1:08:46<15:44,  1.31it/s]
1|5229|Loss: 0.8306779265403748:  81%|████████  | 5229/6470 [1:08:46<15:11,  1.36it/s]
1|5230|Loss: 0.7718245387077332:  81%|████████  | 5229/6470 [1:08:47<15:11,  1.36it/s]
1|5230|Loss: 0.7718245387077332:  81%|████████  | 5230/6470 [1:08:47<15:10,  1.36it/s]
1|5231|Loss: 0.9087124466896057:  81%|████████  | 5230/6470 [1:08:48<15:10,  1.36it/s]
1|5231|Loss: 0.9087124466896057:  81%|████████  | 5231/6470 [1:08:48<15:03,  1.37it/s]
1|5232|Loss: 0.8244397044181824:  81%|████████  | 5231/6470 [1:08:48<15:03,  1.37it/s]
1|5232|Loss: 0.8244397044181824:  81%|████████  | 5232/6470 [1:08:49<15:22,  1.34it/s]
1|5233|Loss: 0.8750362396240234:  81%|████████  | 5232/6470 [1:08:49<15:22,  1.34it/s]
1|5233|Loss: 0.8750362396240234:  81%|████████  | 5233/6470 [1:08:49<15:27,  1.33it/s]
1|5234|Loss: 0.9126536250114441:  81%|████████  | 5233/6470 [1:08:50<15:27,  1.33it/s]
1|5234|Loss: 0.9126536250114441:  81%|████████  | 5234/6470 [1:08:50<15:04,  1.37it/s]
1|5235|Loss: 0.8850086331367493:  81%|████████  | 5234/6470 [1:08:50<15:04,  1.37it/s]
1|5235|Loss: 0.8850086331367493:  81%|████████  | 5235/6470 [1:08:51<15:11,  1.36it/s]
1|5236|Loss: 0.7808501720428467:  81%|████████  | 5235/6470 [1:08:51<15:11,  1.36it/s]
1|5236|Loss: 0.7808501720428467:  81%|████████  | 5236/6470 [1:08:52<16:14,  1.27it/s]
1|5237|Loss: 0.8967646360397339:  81%|████████  | 5236/6470 [1:08:52<16:14,  1.27it/s]
1|5237|Loss: 0.8967646360397339:  81%|████████  | 5237/6470 [1:08:53<17:10,  1.20it/s]
1|5238|Loss: 0.967110812664032:  81%|████████  | 5237/6470 [1:08:53<17:10,  1.20it/s] 
1|5238|Loss: 0.967110812664032:  81%|████████  | 5238/6470 [1:08:53<17:14,  1.19it/s]
1|5239|Loss: 0.8885419964790344:  81%|████████  | 5238/6470 [1:08:54<17:14,  1.19it/s]
1|5239|Loss: 0.8885419964790344:  81%|████████  | 5239/6470 [1:08:54<15:30,  1.32it/s]
1|5240|Loss: 0.9373490810394287:  81%|████████  | 5239/6470 [1:08:54<15:30,  1.32it/s]
1|5240|Loss: 0.9373490810394287:  81%|████████  | 5240/6470 [1:08:55<15:59,  1.28it/s]
1|5241|Loss: 0.8022365570068359:  81%|████████  | 5240/6470 [1:08:55<15:59,  1.28it/s]
1|5241|Loss: 0.8022365570068359:  81%|████████  | 5241/6470 [1:08:56<17:05,  1.20it/s]
1|5242|Loss: 0.8649418950080872:  81%|████████  | 5241/6470 [1:08:56<17:05,  1.20it/s]
1|5242|Loss: 0.8649418950080872:  81%|████████  | 5242/6470 [1:08:57<16:02,  1.28it/s]
1|5243|Loss: 0.924619734287262:  81%|████████  | 5242/6470 [1:08:57<16:02,  1.28it/s] 
1|5243|Loss: 0.924619734287262:  81%|████████  | 5243/6470 [1:08:57<15:35,  1.31it/s]
1|5244|Loss: 0.7916487455368042:  81%|████████  | 5243/6470 [1:08:58<15:35,  1.31it/s]
1|5244|Loss: 0.7916487455368042:  81%|████████  | 5244/6470 [1:08:58<16:27,  1.24it/s]
1|5245|Loss: 0.8965269327163696:  81%|████████  | 5244/6470 [1:08:59<16:27,  1.24it/s]
1|5245|Loss: 0.8965269327163696:  81%|████████  | 5245/6470 [1:08:59<16:15,  1.26it/s]
1|5246|Loss: 0.9852496385574341:  81%|████████  | 5245/6470 [1:08:59<16:15,  1.26it/s]
1|5246|Loss: 0.9852496385574341:  81%|████████  | 5246/6470 [1:09:00<15:20,  1.33it/s]
1|5247|Loss: 1.0201653242111206:  81%|████████  | 5246/6470 [1:09:00<15:20,  1.33it/s]
1|5247|Loss: 1.0201653242111206:  81%|████████  | 5247/6470 [1:09:00<14:57,  1.36it/s]
1|5248|Loss: 0.8678673505783081:  81%|████████  | 5247/6470 [1:09:01<14:57,  1.36it/s]
1|5248|Loss: 0.8678673505783081:  81%|████████  | 5248/6470 [1:09:01<14:20,  1.42it/s]
1|5249|Loss: 0.7170832753181458:  81%|████████  | 5248/6470 [1:09:01<14:20,  1.42it/s]
1|5249|Loss: 0.7170832753181458:  81%|████████  | 5249/6470 [1:09:01<13:23,  1.52it/s]
1|5250|Loss: 0.8057971000671387:  81%|████████  | 5249/6470 [1:09:02<13:23,  1.52it/s]
1|5250|Loss: 0.8057971000671387:  81%|████████  | 5250/6470 [1:09:02<13:59,  1.45it/s]
1|5251|Loss: 0.905027449131012:  81%|████████  | 5250/6470 [1:09:03<13:59,  1.45it/s] 
1|5251|Loss: 0.905027449131012:  81%|████████  | 5251/6470 [1:09:03<14:14,  1.43it/s]
1|5252|Loss: 0.9405236840248108:  81%|████████  | 5251/6470 [1:09:03<14:14,  1.43it/s]
1|5252|Loss: 0.9405236840248108:  81%|████████  | 5252/6470 [1:09:04<14:35,  1.39it/s]
1|5253|Loss: 0.988449215888977:  81%|████████  | 5252/6470 [1:09:04<14:35,  1.39it/s] 
1|5253|Loss: 0.988449215888977:  81%|████████  | 5253/6470 [1:09:05<15:23,  1.32it/s]
1|5254|Loss: 0.867858350276947:  81%|████████  | 5253/6470 [1:09:05<15:23,  1.32it/s]
1|5254|Loss: 0.867858350276947:  81%|████████  | 5254/6470 [1:09:05<15:13,  1.33it/s]
1|5255|Loss: 1.1662691831588745:  81%|████████  | 5254/6470 [1:09:06<15:13,  1.33it/s]
1|5255|Loss: 1.1662691831588745:  81%|████████  | 5255/6470 [1:09:06<15:28,  1.31it/s]
1|5256|Loss: 0.8192205429077148:  81%|████████  | 5255/6470 [1:09:07<15:28,  1.31it/s]
1|5256|Loss: 0.8192205429077148:  81%|████████  | 5256/6470 [1:09:07<15:41,  1.29it/s]
1|5257|Loss: 0.8638482689857483:  81%|████████  | 5256/6470 [1:09:07<15:41,  1.29it/s]
1|5257|Loss: 0.8638482689857483:  81%|████████▏ | 5257/6470 [1:09:07<14:17,  1.41it/s]
1|5258|Loss: 0.7869527339935303:  81%|████████▏ | 5257/6470 [1:09:08<14:17,  1.41it/s]
1|5258|Loss: 0.7869527339935303:  81%|████████▏ | 5258/6470 [1:09:08<14:25,  1.40it/s]
1|5259|Loss: 0.8122567534446716:  81%|████████▏ | 5258/6470 [1:09:09<14:25,  1.40it/s]
1|5259|Loss: 0.8122567534446716:  81%|████████▏ | 5259/6470 [1:09:09<14:44,  1.37it/s]
1|5260|Loss: 0.9680584073066711:  81%|████████▏ | 5259/6470 [1:09:09<14:44,  1.37it/s]
1|5260|Loss: 0.9680584073066711:  81%|████████▏ | 5260/6470 [1:09:09<13:37,  1.48it/s]
1|5261|Loss: 0.8893827795982361:  81%|████████▏ | 5260/6470 [1:09:10<13:37,  1.48it/s]
1|5261|Loss: 0.8893827795982361:  81%|████████▏ | 5261/6470 [1:09:10<12:45,  1.58it/s]
1|5262|Loss: 0.8755135536193848:  81%|████████▏ | 5261/6470 [1:09:10<12:45,  1.58it/s]
1|5262|Loss: 0.8755135536193848:  81%|████████▏ | 5262/6470 [1:09:11<13:49,  1.46it/s]
1|5263|Loss: 0.8363342881202698:  81%|████████▏ | 5262/6470 [1:09:11<13:49,  1.46it/s]
1|5263|Loss: 0.8363342881202698:  81%|████████▏ | 5263/6470 [1:09:12<14:10,  1.42it/s]
1|5264|Loss: 0.7553258538246155:  81%|████████▏ | 5263/6470 [1:09:12<14:10,  1.42it/s]
1|5264|Loss: 0.7553258538246155:  81%|████████▏ | 5264/6470 [1:09:12<14:09,  1.42it/s]
1|5265|Loss: 0.8053720593452454:  81%|████████▏ | 5264/6470 [1:09:13<14:09,  1.42it/s]
1|5265|Loss: 0.8053720593452454:  81%|████████▏ | 5265/6470 [1:09:13<14:15,  1.41it/s]
1|5266|Loss: 0.8844562768936157:  81%|████████▏ | 5265/6470 [1:09:13<14:15,  1.41it/s]
1|5266|Loss: 0.8844562768936157:  81%|████████▏ | 5266/6470 [1:09:14<14:05,  1.42it/s]
1|5267|Loss: 0.9785496592521667:  81%|████████▏ | 5266/6470 [1:09:14<14:05,  1.42it/s]
1|5267|Loss: 0.9785496592521667:  81%|████████▏ | 5267/6470 [1:09:14<14:16,  1.41it/s]
1|5268|Loss: 1.0131416320800781:  81%|████████▏ | 5267/6470 [1:09:15<14:16,  1.41it/s]
1|5268|Loss: 1.0131416320800781:  81%|████████▏ | 5268/6470 [1:09:15<15:25,  1.30it/s]
1|5269|Loss: 0.9514188170433044:  81%|████████▏ | 5268/6470 [1:09:16<15:25,  1.30it/s]
1|5269|Loss: 0.9514188170433044:  81%|████████▏ | 5269/6470 [1:09:16<15:54,  1.26it/s]
1|5270|Loss: 0.9648615717887878:  81%|████████▏ | 5269/6470 [1:09:17<15:54,  1.26it/s]
1|5270|Loss: 0.9648615717887878:  81%|████████▏ | 5270/6470 [1:09:17<17:21,  1.15it/s]
1|5271|Loss: 0.7713364958763123:  81%|████████▏ | 5270/6470 [1:09:18<17:21,  1.15it/s]
1|5271|Loss: 0.7713364958763123:  81%|████████▏ | 5271/6470 [1:09:18<19:02,  1.05it/s]
1|5272|Loss: 0.8244472146034241:  81%|████████▏ | 5271/6470 [1:09:19<19:02,  1.05it/s]
1|5272|Loss: 0.8244472146034241:  81%|████████▏ | 5272/6470 [1:09:19<18:48,  1.06it/s]
1|5273|Loss: 1.0924129486083984:  81%|████████▏ | 5272/6470 [1:09:20<18:48,  1.06it/s]
1|5273|Loss: 1.0924129486083984:  81%|████████▏ | 5273/6470 [1:09:20<18:21,  1.09it/s]
1|5274|Loss: 0.97037273645401:  81%|████████▏ | 5273/6470 [1:09:21<18:21,  1.09it/s]  
1|5274|Loss: 0.97037273645401:  82%|████████▏ | 5274/6470 [1:09:21<18:01,  1.11it/s]
1|5275|Loss: 0.9193810224533081:  82%|████████▏ | 5274/6470 [1:09:21<18:01,  1.11it/s]
1|5275|Loss: 0.9193810224533081:  82%|████████▏ | 5275/6470 [1:09:22<16:58,  1.17it/s]
1|5276|Loss: 0.9490131735801697:  82%|████████▏ | 5275/6470 [1:09:22<16:58,  1.17it/s]
1|5276|Loss: 0.9490131735801697:  82%|████████▏ | 5276/6470 [1:09:23<16:31,  1.20it/s]
1|5277|Loss: 0.8723136782646179:  82%|████████▏ | 5276/6470 [1:09:23<16:31,  1.20it/s]
1|5277|Loss: 0.8723136782646179:  82%|████████▏ | 5277/6470 [1:09:23<15:59,  1.24it/s]
1|5278|Loss: 0.7923665642738342:  82%|████████▏ | 5277/6470 [1:09:24<15:59,  1.24it/s]
1|5278|Loss: 0.7923665642738342:  82%|████████▏ | 5278/6470 [1:09:24<15:03,  1.32it/s]
1|5279|Loss: 0.8137055039405823:  82%|████████▏ | 5278/6470 [1:09:24<15:03,  1.32it/s]
1|5279|Loss: 0.8137055039405823:  82%|████████▏ | 5279/6470 [1:09:25<14:36,  1.36it/s]
1|5280|Loss: 0.8021954298019409:  82%|████████▏ | 5279/6470 [1:09:25<14:36,  1.36it/s]
1|5280|Loss: 0.8021954298019409:  82%|████████▏ | 5280/6470 [1:09:25<14:49,  1.34it/s]
1|5281|Loss: 0.8333112001419067:  82%|████████▏ | 5280/6470 [1:09:26<14:49,  1.34it/s]
1|5281|Loss: 0.8333112001419067:  82%|████████▏ | 5281/6470 [1:09:26<14:23,  1.38it/s]
1|5282|Loss: 0.918979287147522:  82%|████████▏ | 5281/6470 [1:09:26<14:23,  1.38it/s] 
1|5282|Loss: 0.918979287147522:  82%|████████▏ | 5282/6470 [1:09:27<15:25,  1.28it/s]
1|5283|Loss: 0.7638434171676636:  82%|████████▏ | 5282/6470 [1:09:27<15:25,  1.28it/s]
1|5283|Loss: 0.7638434171676636:  82%|████████▏ | 5283/6470 [1:09:28<16:32,  1.20it/s]
1|5284|Loss: 0.7531243562698364:  82%|████████▏ | 5283/6470 [1:09:28<16:32,  1.20it/s]
1|5284|Loss: 0.7531243562698364:  82%|████████▏ | 5284/6470 [1:09:29<17:06,  1.16it/s]
1|5285|Loss: 0.9473716020584106:  82%|████████▏ | 5284/6470 [1:09:29<17:06,  1.16it/s]
1|5285|Loss: 0.9473716020584106:  82%|████████▏ | 5285/6470 [1:09:30<16:48,  1.17it/s]
1|5286|Loss: 1.1045966148376465:  82%|████████▏ | 5285/6470 [1:09:30<16:48,  1.17it/s]
1|5286|Loss: 1.1045966148376465:  82%|████████▏ | 5286/6470 [1:09:31<16:59,  1.16it/s]
1|5287|Loss: 0.8934965133666992:  82%|████████▏ | 5286/6470 [1:09:31<16:59,  1.16it/s]
1|5287|Loss: 0.8934965133666992:  82%|████████▏ | 5287/6470 [1:09:31<16:16,  1.21it/s]
1|5288|Loss: 0.904439389705658:  82%|████████▏ | 5287/6470 [1:09:32<16:16,  1.21it/s] 
1|5288|Loss: 0.904439389705658:  82%|████████▏ | 5288/6470 [1:09:32<15:20,  1.28it/s]
1|5289|Loss: 0.9845700263977051:  82%|████████▏ | 5288/6470 [1:09:32<15:20,  1.28it/s]
1|5289|Loss: 0.9845700263977051:  82%|████████▏ | 5289/6470 [1:09:33<14:28,  1.36it/s]
1|5290|Loss: 0.7487937808036804:  82%|████████▏ | 5289/6470 [1:09:33<14:28,  1.36it/s]
1|5290|Loss: 0.7487937808036804:  82%|████████▏ | 5290/6470 [1:09:33<15:24,  1.28it/s]
1|5291|Loss: 0.8190223574638367:  82%|████████▏ | 5290/6470 [1:09:34<15:24,  1.28it/s]
1|5291|Loss: 0.8190223574638367:  82%|████████▏ | 5291/6470 [1:09:34<15:09,  1.30it/s]
1|5292|Loss: 0.8742387890815735:  82%|████████▏ | 5291/6470 [1:09:35<15:09,  1.30it/s]
1|5292|Loss: 0.8742387890815735:  82%|████████▏ | 5292/6470 [1:09:35<15:28,  1.27it/s]
1|5293|Loss: 0.7972341179847717:  82%|████████▏ | 5292/6470 [1:09:36<15:28,  1.27it/s]
1|5293|Loss: 0.7972341179847717:  82%|████████▏ | 5293/6470 [1:09:36<15:38,  1.25it/s]
1|5294|Loss: 0.8792734742164612:  82%|████████▏ | 5293/6470 [1:09:36<15:38,  1.25it/s]
1|5294|Loss: 0.8792734742164612:  82%|████████▏ | 5294/6470 [1:09:37<15:02,  1.30it/s]
1|5295|Loss: 0.7800885438919067:  82%|████████▏ | 5294/6470 [1:09:37<15:02,  1.30it/s]
1|5295|Loss: 0.7800885438919067:  82%|████████▏ | 5295/6470 [1:09:37<15:02,  1.30it/s]
1|5296|Loss: 0.8976691365242004:  82%|████████▏ | 5295/6470 [1:09:38<15:02,  1.30it/s]
1|5296|Loss: 0.8976691365242004:  82%|████████▏ | 5296/6470 [1:09:38<15:31,  1.26it/s]
1|5297|Loss: 0.9277328848838806:  82%|████████▏ | 5296/6470 [1:09:39<15:31,  1.26it/s]
1|5297|Loss: 0.9277328848838806:  82%|████████▏ | 5297/6470 [1:09:39<15:03,  1.30it/s]
1|5298|Loss: 0.9184175133705139:  82%|████████▏ | 5297/6470 [1:09:39<15:03,  1.30it/s]
1|5298|Loss: 0.9184175133705139:  82%|████████▏ | 5298/6470 [1:09:40<15:22,  1.27it/s]
1|5299|Loss: 0.8773094415664673:  82%|████████▏ | 5298/6470 [1:09:40<15:22,  1.27it/s]
1|5299|Loss: 0.8773094415664673:  82%|████████▏ | 5299/6470 [1:09:41<16:07,  1.21it/s]
1|5300|Loss: 0.8815957903862:  82%|████████▏ | 5299/6470 [1:09:41<16:07,  1.21it/s]   
1|5300|Loss: 0.8815957903862:  82%|████████▏ | 5300/6470 [1:09:41<15:16,  1.28it/s]
1|5301|Loss: 0.8922151327133179:  82%|████████▏ | 5300/6470 [1:09:42<15:16,  1.28it/s]
1|5301|Loss: 0.8922151327133179:  82%|████████▏ | 5301/6470 [1:09:42<15:42,  1.24it/s]
1|5302|Loss: 0.824919581413269:  82%|████████▏ | 5301/6470 [1:09:43<15:42,  1.24it/s] 
1|5302|Loss: 0.824919581413269:  82%|████████▏ | 5302/6470 [1:09:43<15:49,  1.23it/s]
1|5303|Loss: 0.8293721079826355:  82%|████████▏ | 5302/6470 [1:09:44<15:49,  1.23it/s]
1|5303|Loss: 0.8293721079826355:  82%|████████▏ | 5303/6470 [1:09:44<16:22,  1.19it/s]
1|5304|Loss: 0.8769119381904602:  82%|████████▏ | 5303/6470 [1:09:44<16:22,  1.19it/s]
1|5304|Loss: 0.8769119381904602:  82%|████████▏ | 5304/6470 [1:09:45<15:41,  1.24it/s]
1|5305|Loss: 0.9669424891471863:  82%|████████▏ | 5304/6470 [1:09:45<15:41,  1.24it/s]
1|5305|Loss: 0.9669424891471863:  82%|████████▏ | 5305/6470 [1:09:46<15:48,  1.23it/s]
1|5306|Loss: 0.7485801577568054:  82%|████████▏ | 5305/6470 [1:09:46<15:48,  1.23it/s]
1|5306|Loss: 0.7485801577568054:  82%|████████▏ | 5306/6470 [1:09:46<13:55,  1.39it/s]
1|5307|Loss: 1.0426039695739746:  82%|████████▏ | 5306/6470 [1:09:46<13:55,  1.39it/s]
1|5307|Loss: 1.0426039695739746:  82%|████████▏ | 5307/6470 [1:09:47<13:17,  1.46it/s]
1|5308|Loss: 0.7775202989578247:  82%|████████▏ | 5307/6470 [1:09:47<13:17,  1.46it/s]
1|5308|Loss: 0.7775202989578247:  82%|████████▏ | 5308/6470 [1:09:47<14:24,  1.34it/s]
1|5309|Loss: 0.9281003475189209:  82%|████████▏ | 5308/6470 [1:09:48<14:24,  1.34it/s]
1|5309|Loss: 0.9281003475189209:  82%|████████▏ | 5309/6470 [1:09:48<14:49,  1.31it/s]
1|5310|Loss: 0.8794443011283875:  82%|████████▏ | 5309/6470 [1:09:49<14:49,  1.31it/s]
1|5310|Loss: 0.8794443011283875:  82%|████████▏ | 5310/6470 [1:09:49<15:01,  1.29it/s]
1|5311|Loss: 0.87168949842453:  82%|████████▏ | 5310/6470 [1:09:49<15:01,  1.29it/s]  
1|5311|Loss: 0.87168949842453:  82%|████████▏ | 5311/6470 [1:09:50<13:57,  1.38it/s]
1|5312|Loss: 0.8287601470947266:  82%|████████▏ | 5311/6470 [1:09:50<13:57,  1.38it/s]
1|5312|Loss: 0.8287601470947266:  82%|████████▏ | 5312/6470 [1:09:50<13:34,  1.42it/s]
1|5313|Loss: 0.8058863282203674:  82%|████████▏ | 5312/6470 [1:09:51<13:34,  1.42it/s]
1|5313|Loss: 0.8058863282203674:  82%|████████▏ | 5313/6470 [1:09:51<13:02,  1.48it/s]
1|5314|Loss: 0.78229820728302:  82%|████████▏ | 5313/6470 [1:09:51<13:02,  1.48it/s]  
1|5314|Loss: 0.78229820728302:  82%|████████▏ | 5314/6470 [1:09:52<13:26,  1.43it/s]
1|5315|Loss: 0.8065968155860901:  82%|████████▏ | 5314/6470 [1:09:52<13:26,  1.43it/s]
1|5315|Loss: 0.8065968155860901:  82%|████████▏ | 5315/6470 [1:09:53<14:13,  1.35it/s]
1|5316|Loss: 0.8172630071640015:  82%|████████▏ | 5315/6470 [1:09:53<14:13,  1.35it/s]
1|5316|Loss: 0.8172630071640015:  82%|████████▏ | 5316/6470 [1:09:53<13:42,  1.40it/s]
1|5317|Loss: 0.9280391931533813:  82%|████████▏ | 5316/6470 [1:09:54<13:42,  1.40it/s]
1|5317|Loss: 0.9280391931533813:  82%|████████▏ | 5317/6470 [1:09:54<13:08,  1.46it/s]
1|5318|Loss: 0.8681495785713196:  82%|████████▏ | 5317/6470 [1:09:54<13:08,  1.46it/s]
1|5318|Loss: 0.8681495785713196:  82%|████████▏ | 5318/6470 [1:09:55<13:34,  1.41it/s]
1|5319|Loss: 0.9031743407249451:  82%|████████▏ | 5318/6470 [1:09:55<13:34,  1.41it/s]
1|5319|Loss: 0.9031743407249451:  82%|████████▏ | 5319/6470 [1:09:55<14:05,  1.36it/s]
1|5320|Loss: 0.8847025036811829:  82%|████████▏ | 5319/6470 [1:09:56<14:05,  1.36it/s]
1|5320|Loss: 0.8847025036811829:  82%|████████▏ | 5320/6470 [1:09:56<14:24,  1.33it/s]
1|5321|Loss: 0.8784646391868591:  82%|████████▏ | 5320/6470 [1:09:57<14:24,  1.33it/s]
1|5321|Loss: 0.8784646391868591:  82%|████████▏ | 5321/6470 [1:09:57<14:31,  1.32it/s]
1|5322|Loss: 0.773987352848053:  82%|████████▏ | 5321/6470 [1:09:57<14:31,  1.32it/s] 
1|5322|Loss: 0.773987352848053:  82%|████████▏ | 5322/6470 [1:09:58<13:42,  1.39it/s]
1|5323|Loss: 0.8478739261627197:  82%|████████▏ | 5322/6470 [1:09:58<13:42,  1.39it/s]
1|5323|Loss: 0.8478739261627197:  82%|████████▏ | 5323/6470 [1:09:58<14:23,  1.33it/s]
1|5324|Loss: 0.8754122257232666:  82%|████████▏ | 5323/6470 [1:09:59<14:23,  1.33it/s]
1|5324|Loss: 0.8754122257232666:  82%|████████▏ | 5324/6470 [1:09:59<14:38,  1.30it/s]
1|5325|Loss: 0.8687705993652344:  82%|████████▏ | 5324/6470 [1:10:00<14:38,  1.30it/s]
1|5325|Loss: 0.8687705993652344:  82%|████████▏ | 5325/6470 [1:10:00<14:48,  1.29it/s]
1|5326|Loss: 0.838065505027771:  82%|████████▏ | 5325/6470 [1:10:00<14:48,  1.29it/s] 
1|5326|Loss: 0.838065505027771:  82%|████████▏ | 5326/6470 [1:10:01<14:24,  1.32it/s]
1|5327|Loss: 0.811103880405426:  82%|████████▏ | 5326/6470 [1:10:01<14:24,  1.32it/s]
1|5327|Loss: 0.811103880405426:  82%|████████▏ | 5327/6470 [1:10:02<15:41,  1.21it/s]
1|5328|Loss: 0.987297534942627:  82%|████████▏ | 5327/6470 [1:10:02<15:41,  1.21it/s]
1|5328|Loss: 0.987297534942627:  82%|████████▏ | 5328/6470 [1:10:02<15:10,  1.25it/s]
1|5329|Loss: 0.8330116271972656:  82%|████████▏ | 5328/6470 [1:10:03<15:10,  1.25it/s]
1|5329|Loss: 0.8330116271972656:  82%|████████▏ | 5329/6470 [1:10:03<14:31,  1.31it/s]
1|5330|Loss: 0.9277471303939819:  82%|████████▏ | 5329/6470 [1:10:04<14:31,  1.31it/s]
1|5330|Loss: 0.9277471303939819:  82%|████████▏ | 5330/6470 [1:10:04<15:01,  1.26it/s]
1|5331|Loss: 0.9793849587440491:  82%|████████▏ | 5330/6470 [1:10:04<15:01,  1.26it/s]
1|5331|Loss: 0.9793849587440491:  82%|████████▏ | 5331/6470 [1:10:05<15:47,  1.20it/s]
1|5332|Loss: 0.8225154876708984:  82%|████████▏ | 5331/6470 [1:10:05<15:47,  1.20it/s]
1|5332|Loss: 0.8225154876708984:  82%|████████▏ | 5332/6470 [1:10:06<14:59,  1.26it/s]
1|5333|Loss: 0.8718815445899963:  82%|████████▏ | 5332/6470 [1:10:06<14:59,  1.26it/s]
1|5333|Loss: 0.8718815445899963:  82%|████████▏ | 5333/6470 [1:10:06<14:21,  1.32it/s]
1|5334|Loss: 0.9719109535217285:  82%|████████▏ | 5333/6470 [1:10:07<14:21,  1.32it/s]
1|5334|Loss: 0.9719109535217285:  82%|████████▏ | 5334/6470 [1:10:07<14:29,  1.31it/s]
1|5335|Loss: 0.9544883966445923:  82%|████████▏ | 5334/6470 [1:10:07<14:29,  1.31it/s]
1|5335|Loss: 0.9544883966445923:  82%|████████▏ | 5335/6470 [1:10:08<14:31,  1.30it/s]
1|5336|Loss: 0.8719912171363831:  82%|████████▏ | 5335/6470 [1:10:08<14:31,  1.30it/s]
1|5336|Loss: 0.8719912171363831:  82%|████████▏ | 5336/6470 [1:10:09<14:40,  1.29it/s]
1|5337|Loss: 0.9375913739204407:  82%|████████▏ | 5336/6470 [1:10:09<14:40,  1.29it/s]
1|5337|Loss: 0.9375913739204407:  82%|████████▏ | 5337/6470 [1:10:09<15:03,  1.25it/s]
1|5338|Loss: 0.8575717806816101:  82%|████████▏ | 5337/6470 [1:10:10<15:03,  1.25it/s]
1|5338|Loss: 0.8575717806816101:  83%|████████▎ | 5338/6470 [1:10:10<15:44,  1.20it/s]
1|5339|Loss: 1.1412560939788818:  83%|████████▎ | 5338/6470 [1:10:11<15:44,  1.20it/s]
1|5339|Loss: 1.1412560939788818:  83%|████████▎ | 5339/6470 [1:10:11<16:10,  1.16it/s]
1|5340|Loss: 0.7994298338890076:  83%|████████▎ | 5339/6470 [1:10:12<16:10,  1.16it/s]
1|5340|Loss: 0.7994298338890076:  83%|████████▎ | 5340/6470 [1:10:12<16:38,  1.13it/s]
1|5341|Loss: 0.7917237281799316:  83%|████████▎ | 5340/6470 [1:10:13<16:38,  1.13it/s]
1|5341|Loss: 0.7917237281799316:  83%|████████▎ | 5341/6470 [1:10:13<16:43,  1.12it/s]
1|5342|Loss: 1.0095947980880737:  83%|████████▎ | 5341/6470 [1:10:14<16:43,  1.12it/s]
1|5342|Loss: 1.0095947980880737:  83%|████████▎ | 5342/6470 [1:10:14<16:18,  1.15it/s]
1|5343|Loss: 0.819168746471405:  83%|████████▎ | 5342/6470 [1:10:14<16:18,  1.15it/s] 
1|5343|Loss: 0.819168746471405:  83%|████████▎ | 5343/6470 [1:10:15<15:15,  1.23it/s]
1|5344|Loss: 0.8556262254714966:  83%|████████▎ | 5343/6470 [1:10:15<15:15,  1.23it/s]
1|5344|Loss: 0.8556262254714966:  83%|████████▎ | 5344/6470 [1:10:15<14:56,  1.26it/s]
1|5345|Loss: 0.8590719699859619:  83%|████████▎ | 5344/6470 [1:10:16<14:56,  1.26it/s]
1|5345|Loss: 0.8590719699859619:  83%|████████▎ | 5345/6470 [1:10:16<14:17,  1.31it/s]
1|5346|Loss: 0.9253590703010559:  83%|████████▎ | 5345/6470 [1:10:16<14:17,  1.31it/s]
1|5346|Loss: 0.9253590703010559:  83%|████████▎ | 5346/6470 [1:10:17<13:52,  1.35it/s]
1|5347|Loss: 0.939295768737793:  83%|████████▎ | 5346/6470 [1:10:17<13:52,  1.35it/s] 
1|5347|Loss: 0.939295768737793:  83%|████████▎ | 5347/6470 [1:10:17<13:09,  1.42it/s]
1|5348|Loss: 0.8579788208007812:  83%|████████▎ | 5347/6470 [1:10:18<13:09,  1.42it/s]
1|5348|Loss: 0.8579788208007812:  83%|████████▎ | 5348/6470 [1:10:18<13:24,  1.39it/s]
1|5349|Loss: 0.8106712698936462:  83%|████████▎ | 5348/6470 [1:10:19<13:24,  1.39it/s]
1|5349|Loss: 0.8106712698936462:  83%|████████▎ | 5349/6470 [1:10:19<13:50,  1.35it/s]
1|5350|Loss: 0.9479983448982239:  83%|████████▎ | 5349/6470 [1:10:19<13:50,  1.35it/s]
1|5350|Loss: 0.9479983448982239:  83%|████████▎ | 5350/6470 [1:10:20<13:36,  1.37it/s]
1|5351|Loss: 0.9810875654220581:  83%|████████▎ | 5350/6470 [1:10:20<13:36,  1.37it/s]
1|5351|Loss: 0.9810875654220581:  83%|████████▎ | 5351/6470 [1:10:21<14:21,  1.30it/s]
1|5352|Loss: 0.8913571238517761:  83%|████████▎ | 5351/6470 [1:10:21<14:21,  1.30it/s]
1|5352|Loss: 0.8913571238517761:  83%|████████▎ | 5352/6470 [1:10:21<14:41,  1.27it/s]
1|5353|Loss: 0.7872316837310791:  83%|████████▎ | 5352/6470 [1:10:22<14:41,  1.27it/s]
1|5353|Loss: 0.7872316837310791:  83%|████████▎ | 5353/6470 [1:10:22<14:34,  1.28it/s]
1|5354|Loss: 0.9236286878585815:  83%|████████▎ | 5353/6470 [1:10:23<14:34,  1.28it/s]
1|5354|Loss: 0.9236286878585815:  83%|████████▎ | 5354/6470 [1:10:23<14:10,  1.31it/s]
1|5355|Loss: 0.8068490028381348:  83%|████████▎ | 5354/6470 [1:10:23<14:10,  1.31it/s]
1|5355|Loss: 0.8068490028381348:  83%|████████▎ | 5355/6470 [1:10:23<12:22,  1.50it/s]
1|5356|Loss: 0.845740020275116:  83%|████████▎ | 5355/6470 [1:10:24<12:22,  1.50it/s] 
1|5356|Loss: 0.845740020275116:  83%|████████▎ | 5356/6470 [1:10:24<12:28,  1.49it/s]
1|5357|Loss: 0.8307832479476929:  83%|████████▎ | 5356/6470 [1:10:24<12:28,  1.49it/s]
1|5357|Loss: 0.8307832479476929:  83%|████████▎ | 5357/6470 [1:10:25<12:26,  1.49it/s]
1|5358|Loss: 0.8566277027130127:  83%|████████▎ | 5357/6470 [1:10:25<12:26,  1.49it/s]
1|5358|Loss: 0.8566277027130127:  83%|████████▎ | 5358/6470 [1:10:25<12:27,  1.49it/s]
1|5359|Loss: 0.812820553779602:  83%|████████▎ | 5358/6470 [1:10:26<12:27,  1.49it/s] 
1|5359|Loss: 0.812820553779602:  83%|████████▎ | 5359/6470 [1:10:26<13:04,  1.42it/s]
1|5360|Loss: 0.7555649876594543:  83%|████████▎ | 5359/6470 [1:10:27<13:04,  1.42it/s]
1|5360|Loss: 0.7555649876594543:  83%|████████▎ | 5360/6470 [1:10:27<13:22,  1.38it/s]
1|5361|Loss: 0.8880910873413086:  83%|████████▎ | 5360/6470 [1:10:27<13:22,  1.38it/s]
1|5361|Loss: 0.8880910873413086:  83%|████████▎ | 5361/6470 [1:10:28<14:13,  1.30it/s]
1|5362|Loss: 0.8417511582374573:  83%|████████▎ | 5361/6470 [1:10:28<14:13,  1.30it/s]
1|5362|Loss: 0.8417511582374573:  83%|████████▎ | 5362/6470 [1:10:29<14:26,  1.28it/s]
1|5363|Loss: 0.809752881526947:  83%|████████▎ | 5362/6470 [1:10:29<14:26,  1.28it/s] 
1|5363|Loss: 0.809752881526947:  83%|████████▎ | 5363/6470 [1:10:29<14:44,  1.25it/s]
1|5364|Loss: 0.8586544394493103:  83%|████████▎ | 5363/6470 [1:10:30<14:44,  1.25it/s]
1|5364|Loss: 0.8586544394493103:  83%|████████▎ | 5364/6470 [1:10:30<15:24,  1.20it/s]
1|5365|Loss: 0.9607928395271301:  83%|████████▎ | 5364/6470 [1:10:31<15:24,  1.20it/s]
1|5365|Loss: 0.9607928395271301:  83%|████████▎ | 5365/6470 [1:10:31<15:15,  1.21it/s]
1|5366|Loss: 0.9018934369087219:  83%|████████▎ | 5365/6470 [1:10:31<15:15,  1.21it/s]
1|5366|Loss: 0.9018934369087219:  83%|████████▎ | 5366/6470 [1:10:32<14:15,  1.29it/s]
1|5367|Loss: 0.7524189352989197:  83%|████████▎ | 5366/6470 [1:10:32<14:15,  1.29it/s]
1|5367|Loss: 0.7524189352989197:  83%|████████▎ | 5367/6470 [1:10:32<13:40,  1.34it/s]
1|5368|Loss: 1.0356781482696533:  83%|████████▎ | 5367/6470 [1:10:33<13:40,  1.34it/s]
1|5368|Loss: 1.0356781482696533:  83%|████████▎ | 5368/6470 [1:10:33<14:20,  1.28it/s]
1|5369|Loss: 1.2736765146255493:  83%|████████▎ | 5368/6470 [1:10:34<14:20,  1.28it/s]
1|5369|Loss: 1.2736765146255493:  83%|████████▎ | 5369/6470 [1:10:34<15:03,  1.22it/s]
1|5370|Loss: 0.8449526429176331:  83%|████████▎ | 5369/6470 [1:10:35<15:03,  1.22it/s]
1|5370|Loss: 0.8449526429176331:  83%|████████▎ | 5370/6470 [1:10:35<15:57,  1.15it/s]
1|5371|Loss: 0.8794536590576172:  83%|████████▎ | 5370/6470 [1:10:36<15:57,  1.15it/s]
1|5371|Loss: 0.8794536590576172:  83%|████████▎ | 5371/6470 [1:10:36<14:53,  1.23it/s]
1|5372|Loss: 0.8535794019699097:  83%|████████▎ | 5371/6470 [1:10:36<14:53,  1.23it/s]
1|5372|Loss: 0.8535794019699097:  83%|████████▎ | 5372/6470 [1:10:37<14:38,  1.25it/s]
1|5373|Loss: 0.8985505104064941:  83%|████████▎ | 5372/6470 [1:10:37<14:38,  1.25it/s]
1|5373|Loss: 0.8985505104064941:  83%|████████▎ | 5373/6470 [1:10:37<14:39,  1.25it/s]
1|5374|Loss: 1.2003594636917114:  83%|████████▎ | 5373/6470 [1:10:38<14:39,  1.25it/s]
1|5374|Loss: 1.2003594636917114:  83%|████████▎ | 5374/6470 [1:10:38<14:53,  1.23it/s]
1|5375|Loss: 0.9137147665023804:  83%|████████▎ | 5374/6470 [1:10:39<14:53,  1.23it/s]
1|5375|Loss: 0.9137147665023804:  83%|████████▎ | 5375/6470 [1:10:39<16:40,  1.09it/s]
1|5376|Loss: 0.9563918709754944:  83%|████████▎ | 5375/6470 [1:10:40<16:40,  1.09it/s]
1|5376|Loss: 0.9563918709754944:  83%|████████▎ | 5376/6470 [1:10:40<15:40,  1.16it/s]
1|5377|Loss: 0.7452318072319031:  83%|████████▎ | 5376/6470 [1:10:41<15:40,  1.16it/s]
1|5377|Loss: 0.7452318072319031:  83%|████████▎ | 5377/6470 [1:10:41<13:57,  1.30it/s]
1|5378|Loss: 0.7371922135353088:  83%|████████▎ | 5377/6470 [1:10:41<13:57,  1.30it/s]
1|5378|Loss: 0.7371922135353088:  83%|████████▎ | 5378/6470 [1:10:41<13:13,  1.38it/s]
1|5379|Loss: 0.8970484137535095:  83%|████████▎ | 5378/6470 [1:10:42<13:13,  1.38it/s]
1|5379|Loss: 0.8970484137535095:  83%|████████▎ | 5379/6470 [1:10:42<12:42,  1.43it/s]
1|5380|Loss: 0.8318656086921692:  83%|████████▎ | 5379/6470 [1:10:42<12:42,  1.43it/s]
1|5380|Loss: 0.8318656086921692:  83%|████████▎ | 5380/6470 [1:10:43<11:39,  1.56it/s]
1|5381|Loss: 0.8377184271812439:  83%|████████▎ | 5380/6470 [1:10:43<11:39,  1.56it/s]
1|5381|Loss: 0.8377184271812439:  83%|████████▎ | 5381/6470 [1:10:43<11:28,  1.58it/s]
1|5382|Loss: 0.7636523842811584:  83%|████████▎ | 5381/6470 [1:10:44<11:28,  1.58it/s]
1|5382|Loss: 0.7636523842811584:  83%|████████▎ | 5382/6470 [1:10:44<12:14,  1.48it/s]
1|5383|Loss: 0.907107412815094:  83%|████████▎ | 5382/6470 [1:10:44<12:14,  1.48it/s] 
1|5383|Loss: 0.907107412815094:  83%|████████▎ | 5383/6470 [1:10:45<12:58,  1.40it/s]
1|5384|Loss: 0.8938639760017395:  83%|████████▎ | 5383/6470 [1:10:45<12:58,  1.40it/s]
1|5384|Loss: 0.8938639760017395:  83%|████████▎ | 5384/6470 [1:10:45<13:15,  1.36it/s]
1|5385|Loss: 0.7637244462966919:  83%|████████▎ | 5384/6470 [1:10:46<13:15,  1.36it/s]
1|5385|Loss: 0.7637244462966919:  83%|████████▎ | 5385/6470 [1:10:46<12:55,  1.40it/s]
1|5386|Loss: 1.1413487195968628:  83%|████████▎ | 5385/6470 [1:10:46<12:55,  1.40it/s]
1|5386|Loss: 1.1413487195968628:  83%|████████▎ | 5386/6470 [1:10:47<12:07,  1.49it/s]
1|5387|Loss: 0.7995052337646484:  83%|████████▎ | 5386/6470 [1:10:47<12:07,  1.49it/s]
1|5387|Loss: 0.7995052337646484:  83%|████████▎ | 5387/6470 [1:10:47<12:07,  1.49it/s]
1|5388|Loss: 0.902301013469696:  83%|████████▎ | 5387/6470 [1:10:48<12:07,  1.49it/s] 
1|5388|Loss: 0.902301013469696:  83%|████████▎ | 5388/6470 [1:10:48<12:46,  1.41it/s]
1|5389|Loss: 1.080883264541626:  83%|████████▎ | 5388/6470 [1:10:49<12:46,  1.41it/s]
1|5389|Loss: 1.080883264541626:  83%|████████▎ | 5389/6470 [1:10:49<13:16,  1.36it/s]
1|5390|Loss: 0.7852138876914978:  83%|████████▎ | 5389/6470 [1:10:49<13:16,  1.36it/s]
1|5390|Loss: 0.7852138876914978:  83%|████████▎ | 5390/6470 [1:10:50<13:57,  1.29it/s]
1|5391|Loss: 0.8662527203559875:  83%|████████▎ | 5390/6470 [1:10:50<13:57,  1.29it/s]
1|5391|Loss: 0.8662527203559875:  83%|████████▎ | 5391/6470 [1:10:51<14:27,  1.24it/s]
1|5392|Loss: 0.8089566230773926:  83%|████████▎ | 5391/6470 [1:10:51<14:27,  1.24it/s]
1|5392|Loss: 0.8089566230773926:  83%|████████▎ | 5392/6470 [1:10:52<14:48,  1.21it/s]
1|5393|Loss: 0.9251179695129395:  83%|████████▎ | 5392/6470 [1:10:52<14:48,  1.21it/s]
1|5393|Loss: 0.9251179695129395:  83%|████████▎ | 5393/6470 [1:10:52<15:04,  1.19it/s]
1|5394|Loss: 0.9473901391029358:  83%|████████▎ | 5393/6470 [1:10:53<15:04,  1.19it/s]
1|5394|Loss: 0.9473901391029358:  83%|████████▎ | 5394/6470 [1:10:53<14:55,  1.20it/s]
1|5395|Loss: 0.988586962223053:  83%|████████▎ | 5394/6470 [1:10:54<14:55,  1.20it/s] 
1|5395|Loss: 0.988586962223053:  83%|████████▎ | 5395/6470 [1:10:54<13:40,  1.31it/s]
1|5396|Loss: 0.8599041104316711:  83%|████████▎ | 5395/6470 [1:10:54<13:40,  1.31it/s]
1|5396|Loss: 0.8599041104316711:  83%|████████▎ | 5396/6470 [1:10:55<13:27,  1.33it/s]
1|5397|Loss: 0.8483597040176392:  83%|████████▎ | 5396/6470 [1:10:55<13:27,  1.33it/s]
1|5397|Loss: 0.8483597040176392:  83%|████████▎ | 5397/6470 [1:10:55<13:40,  1.31it/s]
1|5398|Loss: 0.7800795435905457:  83%|████████▎ | 5397/6470 [1:10:56<13:40,  1.31it/s]
1|5398|Loss: 0.7800795435905457:  83%|████████▎ | 5398/6470 [1:10:56<13:53,  1.29it/s]
1|5399|Loss: 0.9420464634895325:  83%|████████▎ | 5398/6470 [1:10:57<13:53,  1.29it/s]
1|5399|Loss: 0.9420464634895325:  83%|████████▎ | 5399/6470 [1:10:57<14:03,  1.27it/s]
1|5400|Loss: 0.8199998140335083:  83%|████████▎ | 5399/6470 [1:10:57<14:03,  1.27it/s]
1|5400|Loss: 0.8199998140335083:  83%|████████▎ | 5400/6470 [1:10:58<13:35,  1.31it/s]
1|5401|Loss: 0.8570684194564819:  83%|████████▎ | 5400/6470 [1:10:58<13:35,  1.31it/s]
1|5401|Loss: 0.8570684194564819:  83%|████████▎ | 5401/6470 [1:10:58<13:05,  1.36it/s]
1|5402|Loss: 0.879666805267334:  83%|████████▎ | 5401/6470 [1:10:59<13:05,  1.36it/s] 
1|5402|Loss: 0.879666805267334:  83%|████████▎ | 5402/6470 [1:10:59<13:13,  1.35it/s]
1|5403|Loss: 0.9145851135253906:  83%|████████▎ | 5402/6470 [1:11:00<13:13,  1.35it/s]
1|5403|Loss: 0.9145851135253906:  84%|████████▎ | 5403/6470 [1:11:00<13:05,  1.36it/s]
1|5404|Loss: 0.8472785949707031:  84%|████████▎ | 5403/6470 [1:11:00<13:05,  1.36it/s]
1|5404|Loss: 0.8472785949707031:  84%|████████▎ | 5404/6470 [1:11:01<13:09,  1.35it/s]
1|5405|Loss: 0.8842324018478394:  84%|████████▎ | 5404/6470 [1:11:01<13:09,  1.35it/s]
1|5405|Loss: 0.8842324018478394:  84%|████████▎ | 5405/6470 [1:11:01<12:30,  1.42it/s]
1|5406|Loss: 0.7271913886070251:  84%|████████▎ | 5405/6470 [1:11:02<12:30,  1.42it/s]
1|5406|Loss: 0.7271913886070251:  84%|████████▎ | 5406/6470 [1:11:02<13:17,  1.33it/s]
1|5407|Loss: 1.000969648361206:  84%|████████▎ | 5406/6470 [1:11:03<13:17,  1.33it/s] 
1|5407|Loss: 1.000969648361206:  84%|████████▎ | 5407/6470 [1:11:03<14:17,  1.24it/s]
1|5408|Loss: 1.009751796722412:  84%|████████▎ | 5407/6470 [1:11:04<14:17,  1.24it/s]
1|5408|Loss: 1.009751796722412:  84%|████████▎ | 5408/6470 [1:11:04<14:28,  1.22it/s]
1|5409|Loss: 0.8947207927703857:  84%|████████▎ | 5408/6470 [1:11:04<14:28,  1.22it/s]
1|5409|Loss: 0.8947207927703857:  84%|████████▎ | 5409/6470 [1:11:05<13:33,  1.30it/s]
1|5410|Loss: 0.788722813129425:  84%|████████▎ | 5409/6470 [1:11:05<13:33,  1.30it/s] 
1|5410|Loss: 0.788722813129425:  84%|████████▎ | 5410/6470 [1:11:05<13:02,  1.35it/s]
1|5411|Loss: 0.8634777665138245:  84%|████████▎ | 5410/6470 [1:11:06<13:02,  1.35it/s]
1|5411|Loss: 0.8634777665138245:  84%|████████▎ | 5411/6470 [1:11:06<13:15,  1.33it/s]
1|5412|Loss: 0.880622923374176:  84%|████████▎ | 5411/6470 [1:11:06<13:15,  1.33it/s] 
1|5412|Loss: 0.880622923374176:  84%|████████▎ | 5412/6470 [1:11:07<13:32,  1.30it/s]
1|5413|Loss: 0.8984039425849915:  84%|████████▎ | 5412/6470 [1:11:07<13:32,  1.30it/s]
1|5413|Loss: 0.8984039425849915:  84%|████████▎ | 5413/6470 [1:11:07<13:09,  1.34it/s]
1|5414|Loss: 0.8703120946884155:  84%|████████▎ | 5413/6470 [1:11:08<13:09,  1.34it/s]
1|5414|Loss: 0.8703120946884155:  84%|████████▎ | 5414/6470 [1:11:08<12:51,  1.37it/s]
1|5415|Loss: 0.8524529337882996:  84%|████████▎ | 5414/6470 [1:11:09<12:51,  1.37it/s]
1|5415|Loss: 0.8524529337882996:  84%|████████▎ | 5415/6470 [1:11:09<12:01,  1.46it/s]
1|5416|Loss: 0.9358071684837341:  84%|████████▎ | 5415/6470 [1:11:09<12:01,  1.46it/s]
1|5416|Loss: 0.9358071684837341:  84%|████████▎ | 5416/6470 [1:11:09<10:42,  1.64it/s]
1|5417|Loss: 0.8619231581687927:  84%|████████▎ | 5416/6470 [1:11:10<10:42,  1.64it/s]
1|5417|Loss: 0.8619231581687927:  84%|████████▎ | 5417/6470 [1:11:10<10:48,  1.62it/s]
1|5418|Loss: 0.8576627373695374:  84%|████████▎ | 5417/6470 [1:11:10<10:48,  1.62it/s]
1|5418|Loss: 0.8576627373695374:  84%|████████▎ | 5418/6470 [1:11:11<11:55,  1.47it/s]
1|5419|Loss: 0.7481188774108887:  84%|████████▎ | 5418/6470 [1:11:11<11:55,  1.47it/s]
1|5419|Loss: 0.7481188774108887:  84%|████████▍ | 5419/6470 [1:11:11<12:06,  1.45it/s]
1|5420|Loss: 0.8379709720611572:  84%|████████▍ | 5419/6470 [1:11:12<12:06,  1.45it/s]
1|5420|Loss: 0.8379709720611572:  84%|████████▍ | 5420/6470 [1:11:12<12:37,  1.39it/s]
1|5421|Loss: 0.9107820987701416:  84%|████████▍ | 5420/6470 [1:11:13<12:37,  1.39it/s]
1|5421|Loss: 0.9107820987701416:  84%|████████▍ | 5421/6470 [1:11:13<12:33,  1.39it/s]
1|5422|Loss: 0.9964560270309448:  84%|████████▍ | 5421/6470 [1:11:13<12:33,  1.39it/s]
1|5422|Loss: 0.9964560270309448:  84%|████████▍ | 5422/6470 [1:11:14<12:26,  1.40it/s]
1|5423|Loss: 0.8386426568031311:  84%|████████▍ | 5422/6470 [1:11:14<12:26,  1.40it/s]
1|5423|Loss: 0.8386426568031311:  84%|████████▍ | 5423/6470 [1:11:14<12:36,  1.38it/s]
1|5424|Loss: 0.8796403408050537:  84%|████████▍ | 5423/6470 [1:11:15<12:36,  1.38it/s]
1|5424|Loss: 0.8796403408050537:  84%|████████▍ | 5424/6470 [1:11:15<12:52,  1.35it/s]
1|5425|Loss: 0.8711639642715454:  84%|████████▍ | 5424/6470 [1:11:15<12:52,  1.35it/s]
1|5425|Loss: 0.8711639642715454:  84%|████████▍ | 5425/6470 [1:11:16<12:18,  1.41it/s]
1|5426|Loss: 0.9306992888450623:  84%|████████▍ | 5425/6470 [1:11:16<12:18,  1.41it/s]
1|5426|Loss: 0.9306992888450623:  84%|████████▍ | 5426/6470 [1:11:16<11:40,  1.49it/s]
1|5427|Loss: 0.8887587785720825:  84%|████████▍ | 5426/6470 [1:11:17<11:40,  1.49it/s]
1|5427|Loss: 0.8887587785720825:  84%|████████▍ | 5427/6470 [1:11:17<11:56,  1.45it/s]
1|5428|Loss: 0.909343421459198:  84%|████████▍ | 5427/6470 [1:11:17<11:56,  1.45it/s] 
1|5428|Loss: 0.909343421459198:  84%|████████▍ | 5428/6470 [1:11:18<11:23,  1.52it/s]
1|5429|Loss: 0.8305225968360901:  84%|████████▍ | 5428/6470 [1:11:18<11:23,  1.52it/s]
1|5429|Loss: 0.8305225968360901:  84%|████████▍ | 5429/6470 [1:11:18<11:39,  1.49it/s]
1|5430|Loss: 0.8293273448944092:  84%|████████▍ | 5429/6470 [1:11:19<11:39,  1.49it/s]
1|5430|Loss: 0.8293273448944092:  84%|████████▍ | 5430/6470 [1:11:19<12:28,  1.39it/s]
1|5431|Loss: 0.8025233745574951:  84%|████████▍ | 5430/6470 [1:11:20<12:28,  1.39it/s]
1|5431|Loss: 0.8025233745574951:  84%|████████▍ | 5431/6470 [1:11:20<12:34,  1.38it/s]
1|5432|Loss: 1.046668529510498:  84%|████████▍ | 5431/6470 [1:11:20<12:34,  1.38it/s] 
1|5432|Loss: 1.046668529510498:  84%|████████▍ | 5432/6470 [1:11:21<13:23,  1.29it/s]
1|5433|Loss: 0.9154739379882812:  84%|████████▍ | 5432/6470 [1:11:21<13:23,  1.29it/s]
1|5433|Loss: 0.9154739379882812:  84%|████████▍ | 5433/6470 [1:11:22<14:22,  1.20it/s]
1|5434|Loss: 0.8358121514320374:  84%|████████▍ | 5433/6470 [1:11:23<14:22,  1.20it/s]
1|5434|Loss: 0.8358121514320374:  84%|████████▍ | 5434/6470 [1:11:23<15:43,  1.10it/s]
1|5435|Loss: 0.9120758771896362:  84%|████████▍ | 5434/6470 [1:11:23<15:43,  1.10it/s]
1|5435|Loss: 0.9120758771896362:  84%|████████▍ | 5435/6470 [1:11:24<16:01,  1.08it/s]
1|5436|Loss: 0.8457356095314026:  84%|████████▍ | 5435/6470 [1:11:24<16:01,  1.08it/s]
1|5436|Loss: 0.8457356095314026:  84%|████████▍ | 5436/6470 [1:11:25<14:56,  1.15it/s]
1|5437|Loss: 0.9111368060112:  84%|████████▍ | 5436/6470 [1:11:25<14:56,  1.15it/s]   
1|5437|Loss: 0.9111368060112:  84%|████████▍ | 5437/6470 [1:11:25<14:24,  1.19it/s]
1|5438|Loss: 0.905819833278656:  84%|████████▍ | 5437/6470 [1:11:26<14:24,  1.19it/s]
1|5438|Loss: 0.905819833278656:  84%|████████▍ | 5438/6470 [1:11:26<14:03,  1.22it/s]
1|5439|Loss: 0.8931819200515747:  84%|████████▍ | 5438/6470 [1:11:26<14:03,  1.22it/s]
1|5439|Loss: 0.8931819200515747:  84%|████████▍ | 5439/6470 [1:11:27<13:24,  1.28it/s]
1|5440|Loss: 0.9250189065933228:  84%|████████▍ | 5439/6470 [1:11:27<13:24,  1.28it/s]
1|5440|Loss: 0.9250189065933228:  84%|████████▍ | 5440/6470 [1:11:28<13:18,  1.29it/s]
1|5441|Loss: 0.959670901298523:  84%|████████▍ | 5440/6470 [1:11:28<13:18,  1.29it/s] 
1|5441|Loss: 0.959670901298523:  84%|████████▍ | 5441/6470 [1:11:28<13:22,  1.28it/s]
1|5442|Loss: 0.9892453551292419:  84%|████████▍ | 5441/6470 [1:11:29<13:22,  1.28it/s]
1|5442|Loss: 0.9892453551292419:  84%|████████▍ | 5442/6470 [1:11:29<13:48,  1.24it/s]
1|5443|Loss: 0.7983338832855225:  84%|████████▍ | 5442/6470 [1:11:30<13:48,  1.24it/s]
1|5443|Loss: 0.7983338832855225:  84%|████████▍ | 5443/6470 [1:11:30<14:16,  1.20it/s]
1|5444|Loss: 0.9242207407951355:  84%|████████▍ | 5443/6470 [1:11:31<14:16,  1.20it/s]
1|5444|Loss: 0.9242207407951355:  84%|████████▍ | 5444/6470 [1:11:31<13:37,  1.25it/s]
1|5445|Loss: 0.8519962430000305:  84%|████████▍ | 5444/6470 [1:11:31<13:37,  1.25it/s]
1|5445|Loss: 0.8519962430000305:  84%|████████▍ | 5445/6470 [1:11:32<13:52,  1.23it/s]
1|5446|Loss: 0.8881790041923523:  84%|████████▍ | 5445/6470 [1:11:32<13:52,  1.23it/s]
1|5446|Loss: 0.8881790041923523:  84%|████████▍ | 5446/6470 [1:11:33<14:18,  1.19it/s]
1|5447|Loss: 0.8994175791740417:  84%|████████▍ | 5446/6470 [1:11:33<14:18,  1.19it/s]
1|5447|Loss: 0.8994175791740417:  84%|████████▍ | 5447/6470 [1:11:33<14:14,  1.20it/s]
1|5448|Loss: 0.9776973724365234:  84%|████████▍ | 5447/6470 [1:11:34<14:14,  1.20it/s]
1|5448|Loss: 0.9776973724365234:  84%|████████▍ | 5448/6470 [1:11:34<14:31,  1.17it/s]
1|5449|Loss: 0.8129236102104187:  84%|████████▍ | 5448/6470 [1:11:35<14:31,  1.17it/s]
1|5449|Loss: 0.8129236102104187:  84%|████████▍ | 5449/6470 [1:11:35<15:12,  1.12it/s]
1|5450|Loss: 0.8044427633285522:  84%|████████▍ | 5449/6470 [1:11:36<15:12,  1.12it/s]
1|5450|Loss: 0.8044427633285522:  84%|████████▍ | 5450/6470 [1:11:36<15:33,  1.09it/s]
1|5451|Loss: 0.8774897456169128:  84%|████████▍ | 5450/6470 [1:11:37<15:33,  1.09it/s]
1|5451|Loss: 0.8774897456169128:  84%|████████▍ | 5451/6470 [1:11:37<15:09,  1.12it/s]
1|5452|Loss: 0.8358446955680847:  84%|████████▍ | 5451/6470 [1:11:37<15:09,  1.12it/s]
1|5452|Loss: 0.8358446955680847:  84%|████████▍ | 5452/6470 [1:11:38<14:00,  1.21it/s]
1|5453|Loss: 0.7620759606361389:  84%|████████▍ | 5452/6470 [1:11:38<14:00,  1.21it/s]
1|5453|Loss: 0.7620759606361389:  84%|████████▍ | 5453/6470 [1:11:38<13:14,  1.28it/s]
1|5454|Loss: 1.111046552658081:  84%|████████▍ | 5453/6470 [1:11:39<13:14,  1.28it/s] 
1|5454|Loss: 1.111046552658081:  84%|████████▍ | 5454/6470 [1:11:39<12:32,  1.35it/s]
1|5455|Loss: 0.9256471395492554:  84%|████████▍ | 5454/6470 [1:11:39<12:32,  1.35it/s]
1|5455|Loss: 0.9256471395492554:  84%|████████▍ | 5455/6470 [1:11:40<12:34,  1.34it/s]
1|5456|Loss: 0.8488385677337646:  84%|████████▍ | 5455/6470 [1:11:40<12:34,  1.34it/s]
1|5456|Loss: 0.8488385677337646:  84%|████████▍ | 5456/6470 [1:11:41<12:24,  1.36it/s]
1|5457|Loss: 0.6854804754257202:  84%|████████▍ | 5456/6470 [1:11:41<12:24,  1.36it/s]
1|5457|Loss: 0.6854804754257202:  84%|████████▍ | 5457/6470 [1:11:41<12:06,  1.39it/s]
1|5458|Loss: 0.8635321259498596:  84%|████████▍ | 5457/6470 [1:11:42<12:06,  1.39it/s]
1|5458|Loss: 0.8635321259498596:  84%|████████▍ | 5458/6470 [1:11:42<11:51,  1.42it/s]
1|5459|Loss: 0.8761082887649536:  84%|████████▍ | 5458/6470 [1:11:42<11:51,  1.42it/s]
1|5459|Loss: 0.8761082887649536:  84%|████████▍ | 5459/6470 [1:11:43<12:39,  1.33it/s]
1|5460|Loss: 0.9323242902755737:  84%|████████▍ | 5459/6470 [1:11:43<12:39,  1.33it/s]
1|5460|Loss: 0.9323242902755737:  84%|████████▍ | 5460/6470 [1:11:44<13:01,  1.29it/s]
1|5461|Loss: 0.936939537525177:  84%|████████▍ | 5460/6470 [1:11:44<13:01,  1.29it/s] 
1|5461|Loss: 0.936939537525177:  84%|████████▍ | 5461/6470 [1:11:44<12:46,  1.32it/s]
1|5462|Loss: 0.7942996025085449:  84%|████████▍ | 5461/6470 [1:11:45<12:46,  1.32it/s]
1|5462|Loss: 0.7942996025085449:  84%|████████▍ | 5462/6470 [1:11:45<12:23,  1.36it/s]
1|5463|Loss: 0.8011060357093811:  84%|████████▍ | 5462/6470 [1:11:45<12:23,  1.36it/s]
1|5463|Loss: 0.8011060357093811:  84%|████████▍ | 5463/6470 [1:11:46<12:52,  1.30it/s]
1|5464|Loss: 0.9341004490852356:  84%|████████▍ | 5463/6470 [1:11:46<12:52,  1.30it/s]
1|5464|Loss: 0.9341004490852356:  84%|████████▍ | 5464/6470 [1:11:46<12:09,  1.38it/s]
1|5465|Loss: 0.7455598711967468:  84%|████████▍ | 5464/6470 [1:11:47<12:09,  1.38it/s]
1|5465|Loss: 0.7455598711967468:  84%|████████▍ | 5465/6470 [1:11:47<12:47,  1.31it/s]
1|5466|Loss: 0.9759229421615601:  84%|████████▍ | 5465/6470 [1:11:48<12:47,  1.31it/s]
1|5466|Loss: 0.9759229421615601:  84%|████████▍ | 5466/6470 [1:11:48<12:08,  1.38it/s]
1|5467|Loss: 0.8051775097846985:  84%|████████▍ | 5466/6470 [1:11:48<12:08,  1.38it/s]
1|5467|Loss: 0.8051775097846985:  84%|████████▍ | 5467/6470 [1:11:49<13:02,  1.28it/s]
1|5468|Loss: 0.8678513169288635:  84%|████████▍ | 5467/6470 [1:11:49<13:02,  1.28it/s]
1|5468|Loss: 0.8678513169288635:  85%|████████▍ | 5468/6470 [1:11:50<13:07,  1.27it/s]
1|5469|Loss: 0.8352121114730835:  85%|████████▍ | 5468/6470 [1:11:50<13:07,  1.27it/s]
1|5469|Loss: 0.8352121114730835:  85%|████████▍ | 5469/6470 [1:11:50<11:59,  1.39it/s]
1|5470|Loss: 0.8437804579734802:  85%|████████▍ | 5469/6470 [1:11:51<11:59,  1.39it/s]
1|5470|Loss: 0.8437804579734802:  85%|████████▍ | 5470/6470 [1:11:51<12:45,  1.31it/s]
1|5471|Loss: 0.8654099702835083:  85%|████████▍ | 5470/6470 [1:11:52<12:45,  1.31it/s]
1|5471|Loss: 0.8654099702835083:  85%|████████▍ | 5471/6470 [1:11:52<13:43,  1.21it/s]
1|5472|Loss: 0.7894133925437927:  85%|████████▍ | 5471/6470 [1:11:53<13:43,  1.21it/s]
1|5472|Loss: 0.7894133925437927:  85%|████████▍ | 5472/6470 [1:11:53<14:02,  1.18it/s]
1|5473|Loss: 0.8962121605873108:  85%|████████▍ | 5472/6470 [1:11:53<14:02,  1.18it/s]
1|5473|Loss: 0.8962121605873108:  85%|████████▍ | 5473/6470 [1:11:53<12:37,  1.32it/s]
1|5474|Loss: 0.782756507396698:  85%|████████▍ | 5473/6470 [1:11:54<12:37,  1.32it/s] 
1|5474|Loss: 0.782756507396698:  85%|████████▍ | 5474/6470 [1:11:54<12:26,  1.33it/s]
1|5475|Loss: 0.8038970232009888:  85%|████████▍ | 5474/6470 [1:11:55<12:26,  1.33it/s]
1|5475|Loss: 0.8038970232009888:  85%|████████▍ | 5475/6470 [1:11:55<12:12,  1.36it/s]
1|5476|Loss: 0.806877851486206:  85%|████████▍ | 5475/6470 [1:11:55<12:12,  1.36it/s] 
1|5476|Loss: 0.806877851486206:  85%|████████▍ | 5476/6470 [1:11:56<12:13,  1.36it/s]
1|5477|Loss: 0.8415216207504272:  85%|████████▍ | 5476/6470 [1:11:56<12:13,  1.36it/s]
1|5477|Loss: 0.8415216207504272:  85%|████████▍ | 5477/6470 [1:11:57<12:51,  1.29it/s]
1|5478|Loss: 0.9157041907310486:  85%|████████▍ | 5477/6470 [1:11:57<12:51,  1.29it/s]
1|5478|Loss: 0.9157041907310486:  85%|████████▍ | 5478/6470 [1:11:57<13:23,  1.23it/s]
1|5479|Loss: 0.7771689891815186:  85%|████████▍ | 5478/6470 [1:11:58<13:23,  1.23it/s]
1|5479|Loss: 0.7771689891815186:  85%|████████▍ | 5479/6470 [1:11:58<13:02,  1.27it/s]
1|5480|Loss: 0.7957745790481567:  85%|████████▍ | 5479/6470 [1:11:59<13:02,  1.27it/s]
1|5480|Loss: 0.7957745790481567:  85%|████████▍ | 5480/6470 [1:11:59<13:17,  1.24it/s]
1|5481|Loss: 0.9352145791053772:  85%|████████▍ | 5480/6470 [1:11:59<13:17,  1.24it/s]
1|5481|Loss: 0.9352145791053772:  85%|████████▍ | 5481/6470 [1:12:00<13:51,  1.19it/s]
1|5482|Loss: 0.8637161254882812:  85%|████████▍ | 5481/6470 [1:12:00<13:51,  1.19it/s]
1|5482|Loss: 0.8637161254882812:  85%|████████▍ | 5482/6470 [1:12:01<13:48,  1.19it/s]
1|5483|Loss: 0.8383587002754211:  85%|████████▍ | 5482/6470 [1:12:01<13:48,  1.19it/s]
1|5483|Loss: 0.8383587002754211:  85%|████████▍ | 5483/6470 [1:12:01<12:59,  1.27it/s]
1|5484|Loss: 0.8440967202186584:  85%|████████▍ | 5483/6470 [1:12:02<12:59,  1.27it/s]
1|5484|Loss: 0.8440967202186584:  85%|████████▍ | 5484/6470 [1:12:02<13:17,  1.24it/s]
1|5485|Loss: 0.8852744698524475:  85%|████████▍ | 5484/6470 [1:12:03<13:17,  1.24it/s]
1|5485|Loss: 0.8852744698524475:  85%|████████▍ | 5485/6470 [1:12:03<13:35,  1.21it/s]
1|5486|Loss: 1.0105270147323608:  85%|████████▍ | 5485/6470 [1:12:04<13:35,  1.21it/s]
1|5486|Loss: 1.0105270147323608:  85%|████████▍ | 5486/6470 [1:12:04<13:28,  1.22it/s]
1|5487|Loss: 0.9432365894317627:  85%|████████▍ | 5486/6470 [1:12:04<13:28,  1.22it/s]
1|5487|Loss: 0.9432365894317627:  85%|████████▍ | 5487/6470 [1:12:05<12:50,  1.28it/s]
1|5488|Loss: 0.9163663983345032:  85%|████████▍ | 5487/6470 [1:12:05<12:50,  1.28it/s]
1|5488|Loss: 0.9163663983345032:  85%|████████▍ | 5488/6470 [1:12:06<13:15,  1.23it/s]
1|5489|Loss: 0.7372843027114868:  85%|████████▍ | 5488/6470 [1:12:06<13:15,  1.23it/s]
1|5489|Loss: 0.7372843027114868:  85%|████████▍ | 5489/6470 [1:12:06<12:55,  1.26it/s]
1|5490|Loss: 0.8432009816169739:  85%|████████▍ | 5489/6470 [1:12:07<12:55,  1.26it/s]
1|5490|Loss: 0.8432009816169739:  85%|████████▍ | 5490/6470 [1:12:07<12:06,  1.35it/s]
1|5491|Loss: 0.9325051307678223:  85%|████████▍ | 5490/6470 [1:12:07<12:06,  1.35it/s]
1|5491|Loss: 0.9325051307678223:  85%|████████▍ | 5491/6470 [1:12:08<12:13,  1.34it/s]
1|5492|Loss: 0.7461386919021606:  85%|████████▍ | 5491/6470 [1:12:08<12:13,  1.34it/s]
1|5492|Loss: 0.7461386919021606:  85%|████████▍ | 5492/6470 [1:12:08<12:16,  1.33it/s]
1|5493|Loss: 0.8617374300956726:  85%|████████▍ | 5492/6470 [1:12:09<12:16,  1.33it/s]
1|5493|Loss: 0.8617374300956726:  85%|████████▍ | 5493/6470 [1:12:09<12:28,  1.31it/s]
1|5494|Loss: 0.862246036529541:  85%|████████▍ | 5493/6470 [1:12:10<12:28,  1.31it/s] 
1|5494|Loss: 0.862246036529541:  85%|████████▍ | 5494/6470 [1:12:10<12:54,  1.26it/s]
1|5495|Loss: 0.8369269967079163:  85%|████████▍ | 5494/6470 [1:12:11<12:54,  1.26it/s]
1|5495|Loss: 0.8369269967079163:  85%|████████▍ | 5495/6470 [1:12:11<13:37,  1.19it/s]
1|5496|Loss: 0.8179840445518494:  85%|████████▍ | 5495/6470 [1:12:12<13:37,  1.19it/s]
1|5496|Loss: 0.8179840445518494:  85%|████████▍ | 5496/6470 [1:12:12<13:28,  1.20it/s]
1|5497|Loss: 0.887161910533905:  85%|████████▍ | 5496/6470 [1:12:12<13:28,  1.20it/s] 
1|5497|Loss: 0.887161910533905:  85%|████████▍ | 5497/6470 [1:12:13<13:17,  1.22it/s]
1|5498|Loss: 0.8076021671295166:  85%|████████▍ | 5497/6470 [1:12:13<13:17,  1.22it/s]
1|5498|Loss: 0.8076021671295166:  85%|████████▍ | 5498/6470 [1:12:14<13:47,  1.17it/s]
1|5499|Loss: 0.8294059038162231:  85%|████████▍ | 5498/6470 [1:12:14<13:47,  1.17it/s]
1|5499|Loss: 0.8294059038162231:  85%|████████▍ | 5499/6470 [1:12:14<13:18,  1.22it/s]
1|5500|Loss: 0.9149951934814453:  85%|████████▍ | 5499/6470 [1:12:15<13:18,  1.22it/s]
1|5500|Loss: 0.9149951934814453:  85%|████████▌ | 5500/6470 [1:12:15<13:01,  1.24it/s]
1|5501|Loss: 0.7388956546783447:  85%|████████▌ | 5500/6470 [1:12:15<13:01,  1.24it/s]
1|5501|Loss: 0.7388956546783447:  85%|████████▌ | 5501/6470 [1:12:16<12:25,  1.30it/s]
1|5502|Loss: 0.8905501365661621:  85%|████████▌ | 5501/6470 [1:12:16<12:25,  1.30it/s]
1|5502|Loss: 0.8905501365661621:  85%|████████▌ | 5502/6470 [1:12:17<12:27,  1.30it/s]
1|5503|Loss: 0.769505500793457:  85%|████████▌ | 5502/6470 [1:12:17<12:27,  1.30it/s] 
1|5503|Loss: 0.769505500793457:  85%|████████▌ | 5503/6470 [1:12:17<12:12,  1.32it/s]
1|5504|Loss: 0.8623037338256836:  85%|████████▌ | 5503/6470 [1:12:18<12:12,  1.32it/s]
1|5504|Loss: 0.8623037338256836:  85%|████████▌ | 5504/6470 [1:12:18<11:19,  1.42it/s]
1|5505|Loss: 0.846464991569519:  85%|████████▌ | 5504/6470 [1:12:18<11:19,  1.42it/s] 
1|5505|Loss: 0.846464991569519:  85%|████████▌ | 5505/6470 [1:12:18<10:44,  1.50it/s]
1|5506|Loss: 0.9761942028999329:  85%|████████▌ | 5505/6470 [1:12:19<10:44,  1.50it/s]
1|5506|Loss: 0.9761942028999329:  85%|████████▌ | 5506/6470 [1:12:19<11:28,  1.40it/s]
1|5507|Loss: 0.8473222851753235:  85%|████████▌ | 5506/6470 [1:12:20<11:28,  1.40it/s]
1|5507|Loss: 0.8473222851753235:  85%|████████▌ | 5507/6470 [1:12:20<11:26,  1.40it/s]
1|5508|Loss: 0.8508597016334534:  85%|████████▌ | 5507/6470 [1:12:20<11:26,  1.40it/s]
1|5508|Loss: 0.8508597016334534:  85%|████████▌ | 5508/6470 [1:12:21<11:15,  1.42it/s]
1|5509|Loss: 0.9426620006561279:  85%|████████▌ | 5508/6470 [1:12:21<11:15,  1.42it/s]
1|5509|Loss: 0.9426620006561279:  85%|████████▌ | 5509/6470 [1:12:21<11:31,  1.39it/s]
1|5510|Loss: 0.680711030960083:  85%|████████▌ | 5509/6470 [1:12:22<11:31,  1.39it/s] 
1|5510|Loss: 0.680711030960083:  85%|████████▌ | 5510/6470 [1:12:22<11:29,  1.39it/s]
1|5511|Loss: 0.8892189860343933:  85%|████████▌ | 5510/6470 [1:12:22<11:29,  1.39it/s]
1|5511|Loss: 0.8892189860343933:  85%|████████▌ | 5511/6470 [1:12:23<11:06,  1.44it/s]
1|5512|Loss: 0.8753167986869812:  85%|████████▌ | 5511/6470 [1:12:23<11:06,  1.44it/s]
1|5512|Loss: 0.8753167986869812:  85%|████████▌ | 5512/6470 [1:12:23<10:57,  1.46it/s]
1|5513|Loss: 0.8232877850532532:  85%|████████▌ | 5512/6470 [1:12:24<10:57,  1.46it/s]
1|5513|Loss: 0.8232877850532532:  85%|████████▌ | 5513/6470 [1:12:24<12:03,  1.32it/s]
1|5514|Loss: 0.9139967560768127:  85%|████████▌ | 5513/6470 [1:12:25<12:03,  1.32it/s]
1|5514|Loss: 0.9139967560768127:  85%|████████▌ | 5514/6470 [1:12:25<11:48,  1.35it/s]
1|5515|Loss: 0.98423171043396:  85%|████████▌ | 5514/6470 [1:12:25<11:48,  1.35it/s]  
1|5515|Loss: 0.98423171043396:  85%|████████▌ | 5515/6470 [1:12:26<11:53,  1.34it/s]
1|5516|Loss: 0.8917070031166077:  85%|████████▌ | 5515/6470 [1:12:26<11:53,  1.34it/s]
1|5516|Loss: 0.8917070031166077:  85%|████████▌ | 5516/6470 [1:12:27<12:07,  1.31it/s]
1|5517|Loss: 0.909608781337738:  85%|████████▌ | 5516/6470 [1:12:27<12:07,  1.31it/s] 
1|5517|Loss: 0.909608781337738:  85%|████████▌ | 5517/6470 [1:12:27<12:23,  1.28it/s]
1|5518|Loss: 0.8439871072769165:  85%|████████▌ | 5517/6470 [1:12:28<12:23,  1.28it/s]
1|5518|Loss: 0.8439871072769165:  85%|████████▌ | 5518/6470 [1:12:28<12:07,  1.31it/s]
1|5519|Loss: 0.7375944256782532:  85%|████████▌ | 5518/6470 [1:12:29<12:07,  1.31it/s]
1|5519|Loss: 0.7375944256782532:  85%|████████▌ | 5519/6470 [1:12:29<11:59,  1.32it/s]
1|5520|Loss: 0.904020369052887:  85%|████████▌ | 5519/6470 [1:12:29<11:59,  1.32it/s] 
1|5520|Loss: 0.904020369052887:  85%|████████▌ | 5520/6470 [1:12:30<11:45,  1.35it/s]
1|5521|Loss: 0.9054554104804993:  85%|████████▌ | 5520/6470 [1:12:30<11:45,  1.35it/s]
1|5521|Loss: 0.9054554104804993:  85%|████████▌ | 5521/6470 [1:12:30<11:52,  1.33it/s]
1|5522|Loss: 0.8947963118553162:  85%|████████▌ | 5521/6470 [1:12:31<11:52,  1.33it/s]
1|5522|Loss: 0.8947963118553162:  85%|████████▌ | 5522/6470 [1:12:31<12:01,  1.31it/s]
1|5523|Loss: 0.9453197121620178:  85%|████████▌ | 5522/6470 [1:12:32<12:01,  1.31it/s]
1|5523|Loss: 0.9453197121620178:  85%|████████▌ | 5523/6470 [1:12:32<11:15,  1.40it/s]
1|5524|Loss: 0.9715076684951782:  85%|████████▌ | 5523/6470 [1:12:32<11:15,  1.40it/s]
1|5524|Loss: 0.9715076684951782:  85%|████████▌ | 5524/6470 [1:12:32<10:53,  1.45it/s]
1|5525|Loss: 0.8476406335830688:  85%|████████▌ | 5524/6470 [1:12:33<10:53,  1.45it/s]
1|5525|Loss: 0.8476406335830688:  85%|████████▌ | 5525/6470 [1:12:33<11:48,  1.33it/s]
1|5526|Loss: 0.9897890090942383:  85%|████████▌ | 5525/6470 [1:12:34<11:48,  1.33it/s]
1|5526|Loss: 0.9897890090942383:  85%|████████▌ | 5526/6470 [1:12:34<11:46,  1.34it/s]
1|5527|Loss: 0.9164679050445557:  85%|████████▌ | 5526/6470 [1:12:34<11:46,  1.34it/s]
1|5527|Loss: 0.9164679050445557:  85%|████████▌ | 5527/6470 [1:12:35<11:22,  1.38it/s]
1|5528|Loss: 0.9472041130065918:  85%|████████▌ | 5527/6470 [1:12:35<11:22,  1.38it/s]
1|5528|Loss: 0.9472041130065918:  85%|████████▌ | 5528/6470 [1:12:35<10:57,  1.43it/s]
1|5529|Loss: 0.780663788318634:  85%|████████▌ | 5528/6470 [1:12:36<10:57,  1.43it/s] 
1|5529|Loss: 0.780663788318634:  85%|████████▌ | 5529/6470 [1:12:36<11:02,  1.42it/s]
1|5530|Loss: 0.8607966899871826:  85%|████████▌ | 5529/6470 [1:12:37<11:02,  1.42it/s]
1|5530|Loss: 0.8607966899871826:  85%|████████▌ | 5530/6470 [1:12:37<11:51,  1.32it/s]
1|5531|Loss: 0.9407817721366882:  85%|████████▌ | 5530/6470 [1:12:37<11:51,  1.32it/s]
1|5531|Loss: 0.9407817721366882:  85%|████████▌ | 5531/6470 [1:12:38<12:31,  1.25it/s]
1|5532|Loss: 0.8221818804740906:  85%|████████▌ | 5531/6470 [1:12:38<12:31,  1.25it/s]
1|5532|Loss: 0.8221818804740906:  86%|████████▌ | 5532/6470 [1:12:39<13:01,  1.20it/s]
1|5533|Loss: 0.8937116861343384:  86%|████████▌ | 5532/6470 [1:12:39<13:01,  1.20it/s]
1|5533|Loss: 0.8937116861343384:  86%|████████▌ | 5533/6470 [1:12:39<12:31,  1.25it/s]
1|5534|Loss: 0.8069761395454407:  86%|████████▌ | 5533/6470 [1:12:40<12:31,  1.25it/s]
1|5534|Loss: 0.8069761395454407:  86%|████████▌ | 5534/6470 [1:12:40<11:23,  1.37it/s]
1|5535|Loss: 0.8565782904624939:  86%|████████▌ | 5534/6470 [1:12:40<11:23,  1.37it/s]
1|5535|Loss: 0.8565782904624939:  86%|████████▌ | 5535/6470 [1:12:41<12:16,  1.27it/s]
1|5536|Loss: 0.9664142727851868:  86%|████████▌ | 5535/6470 [1:12:41<12:16,  1.27it/s]
1|5536|Loss: 0.9664142727851868:  86%|████████▌ | 5536/6470 [1:12:42<12:59,  1.20it/s]
1|5537|Loss: 0.9566453099250793:  86%|████████▌ | 5536/6470 [1:12:42<12:59,  1.20it/s]
1|5537|Loss: 0.9566453099250793:  86%|████████▌ | 5537/6470 [1:12:43<12:29,  1.24it/s]
1|5538|Loss: 0.9924410581588745:  86%|████████▌ | 5537/6470 [1:12:43<12:29,  1.24it/s]
1|5538|Loss: 0.9924410581588745:  86%|████████▌ | 5538/6470 [1:12:43<12:12,  1.27it/s]
1|5539|Loss: 0.8518762588500977:  86%|████████▌ | 5538/6470 [1:12:44<12:12,  1.27it/s]
1|5539|Loss: 0.8518762588500977:  86%|████████▌ | 5539/6470 [1:12:44<11:19,  1.37it/s]
1|5540|Loss: 0.7735096216201782:  86%|████████▌ | 5539/6470 [1:12:44<11:19,  1.37it/s]
1|5540|Loss: 0.7735096216201782:  86%|████████▌ | 5540/6470 [1:12:45<10:41,  1.45it/s]
1|5541|Loss: 0.8170477151870728:  86%|████████▌ | 5540/6470 [1:12:45<10:41,  1.45it/s]
1|5541|Loss: 0.8170477151870728:  86%|████████▌ | 5541/6470 [1:12:45<11:00,  1.41it/s]
1|5542|Loss: 0.8997663855552673:  86%|████████▌ | 5541/6470 [1:12:46<11:00,  1.41it/s]
1|5542|Loss: 0.8997663855552673:  86%|████████▌ | 5542/6470 [1:12:46<11:17,  1.37it/s]
1|5543|Loss: 0.792681872844696:  86%|████████▌ | 5542/6470 [1:12:47<11:17,  1.37it/s] 
1|5543|Loss: 0.792681872844696:  86%|████████▌ | 5543/6470 [1:12:47<11:58,  1.29it/s]
1|5544|Loss: 0.8923200368881226:  86%|████████▌ | 5543/6470 [1:12:47<11:58,  1.29it/s]
1|5544|Loss: 0.8923200368881226:  86%|████████▌ | 5544/6470 [1:12:48<11:17,  1.37it/s]
1|5545|Loss: 0.9131126403808594:  86%|████████▌ | 5544/6470 [1:12:48<11:17,  1.37it/s]
1|5545|Loss: 0.9131126403808594:  86%|████████▌ | 5545/6470 [1:12:48<11:20,  1.36it/s]
1|5546|Loss: 0.8723099827766418:  86%|████████▌ | 5545/6470 [1:12:49<11:20,  1.36it/s]
1|5546|Loss: 0.8723099827766418:  86%|████████▌ | 5546/6470 [1:12:49<11:51,  1.30it/s]
1|5547|Loss: 0.8531631231307983:  86%|████████▌ | 5546/6470 [1:12:50<11:51,  1.30it/s]
1|5547|Loss: 0.8531631231307983:  86%|████████▌ | 5547/6470 [1:12:50<11:46,  1.31it/s]
1|5548|Loss: 0.8458021879196167:  86%|████████▌ | 5547/6470 [1:12:50<11:46,  1.31it/s]
1|5548|Loss: 0.8458021879196167:  86%|████████▌ | 5548/6470 [1:12:51<12:28,  1.23it/s]
1|5549|Loss: 0.761760413646698:  86%|████████▌ | 5548/6470 [1:12:51<12:28,  1.23it/s] 
1|5549|Loss: 0.761760413646698:  86%|████████▌ | 5549/6470 [1:12:52<12:05,  1.27it/s]
1|5550|Loss: 0.7671003341674805:  86%|████████▌ | 5549/6470 [1:12:52<12:05,  1.27it/s]
1|5550|Loss: 0.7671003341674805:  86%|████████▌ | 5550/6470 [1:12:52<11:32,  1.33it/s]
1|5551|Loss: 0.9482471942901611:  86%|████████▌ | 5550/6470 [1:12:53<11:32,  1.33it/s]
1|5551|Loss: 0.9482471942901611:  86%|████████▌ | 5551/6470 [1:12:53<11:30,  1.33it/s]
1|5552|Loss: 0.8484253287315369:  86%|████████▌ | 5551/6470 [1:12:53<11:30,  1.33it/s]
1|5552|Loss: 0.8484253287315369:  86%|████████▌ | 5552/6470 [1:12:54<10:30,  1.46it/s]
1|5553|Loss: 0.9625436663627625:  86%|████████▌ | 5552/6470 [1:12:54<10:30,  1.46it/s]
1|5553|Loss: 0.9625436663627625:  86%|████████▌ | 5553/6470 [1:12:54<10:18,  1.48it/s]
1|5554|Loss: 0.7489818334579468:  86%|████████▌ | 5553/6470 [1:12:55<10:18,  1.48it/s]
1|5554|Loss: 0.7489818334579468:  86%|████████▌ | 5554/6470 [1:12:55<09:20,  1.64it/s]
1|5555|Loss: 0.727035403251648:  86%|████████▌ | 5554/6470 [1:12:55<09:20,  1.64it/s] 
1|5555|Loss: 0.727035403251648:  86%|████████▌ | 5555/6470 [1:12:55<09:44,  1.56it/s]
1|5556|Loss: 0.9175565242767334:  86%|████████▌ | 5555/6470 [1:12:56<09:44,  1.56it/s]
1|5556|Loss: 0.9175565242767334:  86%|████████▌ | 5556/6470 [1:12:56<10:31,  1.45it/s]
1|5557|Loss: 0.8949916362762451:  86%|████████▌ | 5556/6470 [1:12:57<10:31,  1.45it/s]
1|5557|Loss: 0.8949916362762451:  86%|████████▌ | 5557/6470 [1:12:57<10:41,  1.42it/s]
1|5558|Loss: 0.7929645776748657:  86%|████████▌ | 5557/6470 [1:12:57<10:41,  1.42it/s]
1|5558|Loss: 0.7929645776748657:  86%|████████▌ | 5558/6470 [1:12:58<10:12,  1.49it/s]
1|5559|Loss: 0.9395873546600342:  86%|████████▌ | 5558/6470 [1:12:58<10:12,  1.49it/s]
1|5559|Loss: 0.9395873546600342:  86%|████████▌ | 5559/6470 [1:12:58<10:56,  1.39it/s]
1|5560|Loss: 0.8473960161209106:  86%|████████▌ | 5559/6470 [1:12:59<10:56,  1.39it/s]
1|5560|Loss: 0.8473960161209106:  86%|████████▌ | 5560/6470 [1:12:59<11:09,  1.36it/s]
1|5561|Loss: 0.8187946677207947:  86%|████████▌ | 5560/6470 [1:13:00<11:09,  1.36it/s]
1|5561|Loss: 0.8187946677207947:  86%|████████▌ | 5561/6470 [1:13:00<11:28,  1.32it/s]
1|5562|Loss: 0.8930100202560425:  86%|████████▌ | 5561/6470 [1:13:00<11:28,  1.32it/s]
1|5562|Loss: 0.8930100202560425:  86%|████████▌ | 5562/6470 [1:13:01<11:34,  1.31it/s]
1|5563|Loss: 0.8944491147994995:  86%|████████▌ | 5562/6470 [1:13:01<11:34,  1.31it/s]
1|5563|Loss: 0.8944491147994995:  86%|████████▌ | 5563/6470 [1:13:01<11:02,  1.37it/s]
1|5564|Loss: 0.8892362117767334:  86%|████████▌ | 5563/6470 [1:13:02<11:02,  1.37it/s]
1|5564|Loss: 0.8892362117767334:  86%|████████▌ | 5564/6470 [1:13:02<11:56,  1.26it/s]
1|5565|Loss: 0.8219900131225586:  86%|████████▌ | 5564/6470 [1:13:03<11:56,  1.26it/s]
1|5565|Loss: 0.8219900131225586:  86%|████████▌ | 5565/6470 [1:13:03<11:47,  1.28it/s]
1|5566|Loss: 0.9742031693458557:  86%|████████▌ | 5565/6470 [1:13:04<11:47,  1.28it/s]
1|5566|Loss: 0.9742031693458557:  86%|████████▌ | 5566/6470 [1:13:04<12:06,  1.24it/s]
1|5567|Loss: 0.8107303380966187:  86%|████████▌ | 5566/6470 [1:13:04<12:06,  1.24it/s]
1|5567|Loss: 0.8107303380966187:  86%|████████▌ | 5567/6470 [1:13:05<11:11,  1.35it/s]
1|5568|Loss: 0.914523720741272:  86%|████████▌ | 5567/6470 [1:13:05<11:11,  1.35it/s] 
1|5568|Loss: 0.914523720741272:  86%|████████▌ | 5568/6470 [1:13:05<10:44,  1.40it/s]
1|5569|Loss: 1.019808292388916:  86%|████████▌ | 5568/6470 [1:13:06<10:44,  1.40it/s]
1|5569|Loss: 1.019808292388916:  86%|████████▌ | 5569/6470 [1:13:06<10:53,  1.38it/s]
1|5570|Loss: 0.9558780789375305:  86%|████████▌ | 5569/6470 [1:13:06<10:53,  1.38it/s]
1|5570|Loss: 0.9558780789375305:  86%|████████▌ | 5570/6470 [1:13:07<10:31,  1.43it/s]
1|5571|Loss: 0.8127010464668274:  86%|████████▌ | 5570/6470 [1:13:07<10:31,  1.43it/s]
1|5571|Loss: 0.8127010464668274:  86%|████████▌ | 5571/6470 [1:13:07<10:45,  1.39it/s]
1|5572|Loss: 0.8264105319976807:  86%|████████▌ | 5571/6470 [1:13:08<10:45,  1.39it/s]
1|5572|Loss: 0.8264105319976807:  86%|████████▌ | 5572/6470 [1:13:08<11:02,  1.36it/s]
1|5573|Loss: 0.9494814276695251:  86%|████████▌ | 5572/6470 [1:13:09<11:02,  1.36it/s]
1|5573|Loss: 0.9494814276695251:  86%|████████▌ | 5573/6470 [1:13:09<10:56,  1.37it/s]
1|5574|Loss: 0.883554220199585:  86%|████████▌ | 5573/6470 [1:13:09<10:56,  1.37it/s] 
1|5574|Loss: 0.883554220199585:  86%|████████▌ | 5574/6470 [1:13:10<10:59,  1.36it/s]
1|5575|Loss: 0.8109546899795532:  86%|████████▌ | 5574/6470 [1:13:10<10:59,  1.36it/s]
1|5575|Loss: 0.8109546899795532:  86%|████████▌ | 5575/6470 [1:13:10<10:19,  1.44it/s]
1|5576|Loss: 0.8988090753555298:  86%|████████▌ | 5575/6470 [1:13:11<10:19,  1.44it/s]
1|5576|Loss: 0.8988090753555298:  86%|████████▌ | 5576/6470 [1:13:11<10:08,  1.47it/s]
1|5577|Loss: 0.973100483417511:  86%|████████▌ | 5576/6470 [1:13:11<10:08,  1.47it/s] 
1|5577|Loss: 0.973100483417511:  86%|████████▌ | 5577/6470 [1:13:11<09:46,  1.52it/s]
1|5578|Loss: 0.6566060781478882:  86%|████████▌ | 5577/6470 [1:13:12<09:46,  1.52it/s]
1|5578|Loss: 0.6566060781478882:  86%|████████▌ | 5578/6470 [1:13:12<09:44,  1.52it/s]
1|5579|Loss: 0.8958589434623718:  86%|████████▌ | 5578/6470 [1:13:13<09:44,  1.52it/s]
1|5579|Loss: 0.8958589434623718:  86%|████████▌ | 5579/6470 [1:13:13<10:19,  1.44it/s]
1|5580|Loss: 0.7041375637054443:  86%|████████▌ | 5579/6470 [1:13:13<10:19,  1.44it/s]
1|5580|Loss: 0.7041375637054443:  86%|████████▌ | 5580/6470 [1:13:14<10:35,  1.40it/s]
1|5581|Loss: 0.9185398817062378:  86%|████████▌ | 5580/6470 [1:13:14<10:35,  1.40it/s]
1|5581|Loss: 0.9185398817062378:  86%|████████▋ | 5581/6470 [1:13:14<10:32,  1.41it/s]
1|5582|Loss: 0.8514097929000854:  86%|████████▋ | 5581/6470 [1:13:15<10:32,  1.41it/s]
1|5582|Loss: 0.8514097929000854:  86%|████████▋ | 5582/6470 [1:13:15<10:44,  1.38it/s]
1|5583|Loss: 0.7744755744934082:  86%|████████▋ | 5582/6470 [1:13:16<10:44,  1.38it/s]
1|5583|Loss: 0.7744755744934082:  86%|████████▋ | 5583/6470 [1:13:16<11:22,  1.30it/s]
1|5584|Loss: 0.8879681825637817:  86%|████████▋ | 5583/6470 [1:13:16<11:22,  1.30it/s]
1|5584|Loss: 0.8879681825637817:  86%|████████▋ | 5584/6470 [1:13:17<12:03,  1.22it/s]
1|5585|Loss: 0.9131718873977661:  86%|████████▋ | 5584/6470 [1:13:17<12:03,  1.22it/s]
1|5585|Loss: 0.9131718873977661:  86%|████████▋ | 5585/6470 [1:13:18<12:09,  1.21it/s]
1|5586|Loss: 0.9280622005462646:  86%|████████▋ | 5585/6470 [1:13:18<12:09,  1.21it/s]
1|5586|Loss: 0.9280622005462646:  86%|████████▋ | 5586/6470 [1:13:18<11:39,  1.26it/s]
1|5587|Loss: 0.8329289555549622:  86%|████████▋ | 5586/6470 [1:13:19<11:39,  1.26it/s]
1|5587|Loss: 0.8329289555549622:  86%|████████▋ | 5587/6470 [1:13:19<10:27,  1.41it/s]
1|5588|Loss: 0.9882126450538635:  86%|████████▋ | 5587/6470 [1:13:19<10:27,  1.41it/s]
1|5588|Loss: 0.9882126450538635:  86%|████████▋ | 5588/6470 [1:13:20<10:22,  1.42it/s]
1|5589|Loss: 0.8614208102226257:  86%|████████▋ | 5588/6470 [1:13:20<10:22,  1.42it/s]
1|5589|Loss: 0.8614208102226257:  86%|████████▋ | 5589/6470 [1:13:20<10:24,  1.41it/s]
1|5590|Loss: 0.8485100865364075:  86%|████████▋ | 5589/6470 [1:13:21<10:24,  1.41it/s]
1|5590|Loss: 0.8485100865364075:  86%|████████▋ | 5590/6470 [1:13:21<09:46,  1.50it/s]
1|5591|Loss: 1.055262804031372:  86%|████████▋ | 5590/6470 [1:13:21<09:46,  1.50it/s] 
1|5591|Loss: 1.055262804031372:  86%|████████▋ | 5591/6470 [1:13:22<09:18,  1.57it/s]
1|5592|Loss: 0.8502174615859985:  86%|████████▋ | 5591/6470 [1:13:22<09:18,  1.57it/s]
1|5592|Loss: 0.8502174615859985:  86%|████████▋ | 5592/6470 [1:13:22<08:52,  1.65it/s]
1|5593|Loss: 0.7717310190200806:  86%|████████▋ | 5592/6470 [1:13:22<08:52,  1.65it/s]
1|5593|Loss: 0.7717310190200806:  86%|████████▋ | 5593/6470 [1:13:23<09:18,  1.57it/s]
1|5594|Loss: 0.9011735320091248:  86%|████████▋ | 5593/6470 [1:13:23<09:18,  1.57it/s]
1|5594|Loss: 0.9011735320091248:  86%|████████▋ | 5594/6470 [1:13:23<09:21,  1.56it/s]
1|5595|Loss: 0.8203802704811096:  86%|████████▋ | 5594/6470 [1:13:24<09:21,  1.56it/s]
1|5595|Loss: 0.8203802704811096:  86%|████████▋ | 5595/6470 [1:13:24<09:31,  1.53it/s]
1|5596|Loss: 0.8086079359054565:  86%|████████▋ | 5595/6470 [1:13:24<09:31,  1.53it/s]
1|5596|Loss: 0.8086079359054565:  86%|████████▋ | 5596/6470 [1:13:25<09:44,  1.50it/s]
1|5597|Loss: 1.0421584844589233:  86%|████████▋ | 5596/6470 [1:13:25<09:44,  1.50it/s]
1|5597|Loss: 1.0421584844589233:  87%|████████▋ | 5597/6470 [1:13:26<09:53,  1.47it/s]
1|5598|Loss: 0.8598430156707764:  87%|████████▋ | 5597/6470 [1:13:26<09:53,  1.47it/s]
1|5598|Loss: 0.8598430156707764:  87%|████████▋ | 5598/6470 [1:13:26<09:52,  1.47it/s]
1|5599|Loss: 0.8240752220153809:  87%|████████▋ | 5598/6470 [1:13:27<09:52,  1.47it/s]
1|5599|Loss: 0.8240752220153809:  87%|████████▋ | 5599/6470 [1:13:27<10:00,  1.45it/s]
1|5600|Loss: 0.7913119196891785:  87%|████████▋ | 5599/6470 [1:13:27<10:00,  1.45it/s]
1|5600|Loss: 0.7913119196891785:  87%|████████▋ | 5600/6470 [1:13:28<10:25,  1.39it/s]
1|5601|Loss: 0.6401288509368896:  87%|████████▋ | 5600/6470 [1:13:28<10:25,  1.39it/s]
1|5601|Loss: 0.6401288509368896:  87%|████████▋ | 5601/6470 [1:13:28<09:19,  1.55it/s]
1|5602|Loss: 0.8663948178291321:  87%|████████▋ | 5601/6470 [1:13:28<09:19,  1.55it/s]
1|5602|Loss: 0.8663948178291321:  87%|████████▋ | 5602/6470 [1:13:29<09:16,  1.56it/s]
1|5603|Loss: 0.6988924145698547:  87%|████████▋ | 5602/6470 [1:13:29<09:16,  1.56it/s]
1|5603|Loss: 0.6988924145698547:  87%|████████▋ | 5603/6470 [1:13:30<09:48,  1.47it/s]
1|5604|Loss: 0.8972736597061157:  87%|████████▋ | 5603/6470 [1:13:30<09:48,  1.47it/s]
1|5604|Loss: 0.8972736597061157:  87%|████████▋ | 5604/6470 [1:13:30<10:13,  1.41it/s]
1|5605|Loss: 0.8850847482681274:  87%|████████▋ | 5604/6470 [1:13:31<10:13,  1.41it/s]
1|5605|Loss: 0.8850847482681274:  87%|████████▋ | 5605/6470 [1:13:31<10:13,  1.41it/s]
1|5606|Loss: 0.7402374148368835:  87%|████████▋ | 5605/6470 [1:13:31<10:13,  1.41it/s]
1|5606|Loss: 0.7402374148368835:  87%|████████▋ | 5606/6470 [1:13:32<09:47,  1.47it/s]
1|5607|Loss: 0.9184921383857727:  87%|████████▋ | 5606/6470 [1:13:32<09:47,  1.47it/s]
1|5607|Loss: 0.9184921383857727:  87%|████████▋ | 5607/6470 [1:13:32<10:03,  1.43it/s]
1|5608|Loss: 0.7591509222984314:  87%|████████▋ | 5607/6470 [1:13:33<10:03,  1.43it/s]
1|5608|Loss: 0.7591509222984314:  87%|████████▋ | 5608/6470 [1:13:33<10:24,  1.38it/s]
1|5609|Loss: 0.9767581224441528:  87%|████████▋ | 5608/6470 [1:13:34<10:24,  1.38it/s]
1|5609|Loss: 0.9767581224441528:  87%|████████▋ | 5609/6470 [1:13:34<10:33,  1.36it/s]
1|5610|Loss: 0.9270857572555542:  87%|████████▋ | 5609/6470 [1:13:34<10:33,  1.36it/s]
1|5610|Loss: 0.9270857572555542:  87%|████████▋ | 5610/6470 [1:13:35<10:45,  1.33it/s]
1|5611|Loss: 0.8841982483863831:  87%|████████▋ | 5610/6470 [1:13:35<10:45,  1.33it/s]
1|5611|Loss: 0.8841982483863831:  87%|████████▋ | 5611/6470 [1:13:35<10:32,  1.36it/s]
1|5612|Loss: 0.8554994463920593:  87%|████████▋ | 5611/6470 [1:13:36<10:32,  1.36it/s]
1|5612|Loss: 0.8554994463920593:  87%|████████▋ | 5612/6470 [1:13:36<09:59,  1.43it/s]
1|5613|Loss: 0.9545459747314453:  87%|████████▋ | 5612/6470 [1:13:36<09:59,  1.43it/s]
1|5613|Loss: 0.9545459747314453:  87%|████████▋ | 5613/6470 [1:13:37<10:35,  1.35it/s]
1|5614|Loss: 0.9559344053268433:  87%|████████▋ | 5613/6470 [1:13:37<10:35,  1.35it/s]
1|5614|Loss: 0.9559344053268433:  87%|████████▋ | 5614/6470 [1:13:38<10:28,  1.36it/s]
1|5615|Loss: 0.8345370292663574:  87%|████████▋ | 5614/6470 [1:13:38<10:28,  1.36it/s]
1|5615|Loss: 0.8345370292663574:  87%|████████▋ | 5615/6470 [1:13:38<10:19,  1.38it/s]
1|5616|Loss: 0.7804015278816223:  87%|████████▋ | 5615/6470 [1:13:39<10:19,  1.38it/s]
1|5616|Loss: 0.7804015278816223:  87%|████████▋ | 5616/6470 [1:13:39<09:57,  1.43it/s]
1|5617|Loss: 0.90227210521698:  87%|████████▋ | 5616/6470 [1:13:39<09:57,  1.43it/s]  
1|5617|Loss: 0.90227210521698:  87%|████████▋ | 5617/6470 [1:13:40<10:13,  1.39it/s]
1|5618|Loss: 0.8926650881767273:  87%|████████▋ | 5617/6470 [1:13:40<10:13,  1.39it/s]
1|5618|Loss: 0.8926650881767273:  87%|████████▋ | 5618/6470 [1:13:40<10:05,  1.41it/s]
1|5619|Loss: 0.9629364609718323:  87%|████████▋ | 5618/6470 [1:13:41<10:05,  1.41it/s]
1|5619|Loss: 0.9629364609718323:  87%|████████▋ | 5619/6470 [1:13:41<10:44,  1.32it/s]
1|5620|Loss: 0.9954590201377869:  87%|████████▋ | 5619/6470 [1:13:42<10:44,  1.32it/s]
1|5620|Loss: 0.9954590201377869:  87%|████████▋ | 5620/6470 [1:13:42<10:07,  1.40it/s]
1|5621|Loss: 0.9241665601730347:  87%|████████▋ | 5620/6470 [1:13:42<10:07,  1.40it/s]
1|5621|Loss: 0.9241665601730347:  87%|████████▋ | 5621/6470 [1:13:43<10:06,  1.40it/s]
1|5622|Loss: 0.8213045597076416:  87%|████████▋ | 5621/6470 [1:13:43<10:06,  1.40it/s]
1|5622|Loss: 0.8213045597076416:  87%|████████▋ | 5622/6470 [1:13:43<10:05,  1.40it/s]
1|5623|Loss: 0.891753077507019:  87%|████████▋ | 5622/6470 [1:13:44<10:05,  1.40it/s] 
1|5623|Loss: 0.891753077507019:  87%|████████▋ | 5623/6470 [1:13:44<09:55,  1.42it/s]
1|5624|Loss: 0.8768589496612549:  87%|████████▋ | 5623/6470 [1:13:44<09:55,  1.42it/s]
1|5624|Loss: 0.8768589496612549:  87%|████████▋ | 5624/6470 [1:13:45<09:36,  1.47it/s]
1|5625|Loss: 0.8149693012237549:  87%|████████▋ | 5624/6470 [1:13:45<09:36,  1.47it/s]
1|5625|Loss: 0.8149693012237549:  87%|████████▋ | 5625/6470 [1:13:45<09:54,  1.42it/s]
1|5626|Loss: 0.9705308675765991:  87%|████████▋ | 5625/6470 [1:13:46<09:54,  1.42it/s]
1|5626|Loss: 0.9705308675765991:  87%|████████▋ | 5626/6470 [1:13:46<10:18,  1.37it/s]
1|5627|Loss: 0.9206366539001465:  87%|████████▋ | 5626/6470 [1:13:47<10:18,  1.37it/s]
1|5627|Loss: 0.9206366539001465:  87%|████████▋ | 5627/6470 [1:13:47<09:52,  1.42it/s]
1|5628|Loss: 0.7739946246147156:  87%|████████▋ | 5627/6470 [1:13:47<09:52,  1.42it/s]
1|5628|Loss: 0.7739946246147156:  87%|████████▋ | 5628/6470 [1:13:47<09:43,  1.44it/s]
1|5629|Loss: 0.8887765407562256:  87%|████████▋ | 5628/6470 [1:13:48<09:43,  1.44it/s]
1|5629|Loss: 0.8887765407562256:  87%|████████▋ | 5629/6470 [1:13:48<09:38,  1.45it/s]
1|5630|Loss: 0.8862066268920898:  87%|████████▋ | 5629/6470 [1:13:49<09:38,  1.45it/s]
1|5630|Loss: 0.8862066268920898:  87%|████████▋ | 5630/6470 [1:13:49<09:11,  1.52it/s]
1|5631|Loss: 0.8328207731246948:  87%|████████▋ | 5630/6470 [1:13:49<09:11,  1.52it/s]
1|5631|Loss: 0.8328207731246948:  87%|████████▋ | 5631/6470 [1:13:49<09:14,  1.51it/s]
1|5632|Loss: 0.8230380415916443:  87%|████████▋ | 5631/6470 [1:13:50<09:14,  1.51it/s]
1|5632|Loss: 0.8230380415916443:  87%|████████▋ | 5632/6470 [1:13:50<09:43,  1.44it/s]
1|5633|Loss: 0.8597632646560669:  87%|████████▋ | 5632/6470 [1:13:51<09:43,  1.44it/s]
1|5633|Loss: 0.8597632646560669:  87%|████████▋ | 5633/6470 [1:13:51<09:37,  1.45it/s]
1|5634|Loss: 0.8680469393730164:  87%|████████▋ | 5633/6470 [1:13:51<09:37,  1.45it/s]
1|5634|Loss: 0.8680469393730164:  87%|████████▋ | 5634/6470 [1:13:51<09:19,  1.49it/s]
1|5635|Loss: 0.8012005090713501:  87%|████████▋ | 5634/6470 [1:13:52<09:19,  1.49it/s]
1|5635|Loss: 0.8012005090713501:  87%|████████▋ | 5635/6470 [1:13:52<09:35,  1.45it/s]
1|5636|Loss: 0.7322055101394653:  87%|████████▋ | 5635/6470 [1:13:53<09:35,  1.45it/s]
1|5636|Loss: 0.7322055101394653:  87%|████████▋ | 5636/6470 [1:13:53<10:35,  1.31it/s]
1|5637|Loss: 0.8103189468383789:  87%|████████▋ | 5636/6470 [1:13:54<10:35,  1.31it/s]
1|5637|Loss: 0.8103189468383789:  87%|████████▋ | 5637/6470 [1:13:54<10:44,  1.29it/s]
1|5638|Loss: 0.7904236316680908:  87%|████████▋ | 5637/6470 [1:13:54<10:44,  1.29it/s]
1|5638|Loss: 0.7904236316680908:  87%|████████▋ | 5638/6470 [1:13:54<09:22,  1.48it/s]
1|5639|Loss: 0.7790703773498535:  87%|████████▋ | 5638/6470 [1:13:55<09:22,  1.48it/s]
1|5639|Loss: 0.7790703773498535:  87%|████████▋ | 5639/6470 [1:13:55<09:29,  1.46it/s]
1|5640|Loss: 0.754112184047699:  87%|████████▋ | 5639/6470 [1:13:56<09:29,  1.46it/s] 
1|5640|Loss: 0.754112184047699:  87%|████████▋ | 5640/6470 [1:13:57<14:25,  1.04s/it]
1|5641|Loss: 0.77617347240448:  87%|████████▋ | 5640/6470 [1:13:58<14:25,  1.04s/it] 
1|5641|Loss: 0.77617347240448:  87%|████████▋ | 5641/6470 [1:13:58<13:25,  1.03it/s]
1|5642|Loss: 1.0091081857681274:  87%|████████▋ | 5641/6470 [1:13:58<13:25,  1.03it/s]
1|5642|Loss: 1.0091081857681274:  87%|████████▋ | 5642/6470 [1:13:59<12:23,  1.11it/s]
1|5643|Loss: 0.6193006038665771:  87%|████████▋ | 5642/6470 [1:13:59<12:23,  1.11it/s]
1|5643|Loss: 0.6193006038665771:  87%|████████▋ | 5643/6470 [1:13:59<12:11,  1.13it/s]
1|5644|Loss: 0.8441025018692017:  87%|████████▋ | 5643/6470 [1:14:00<12:11,  1.13it/s]
1|5644|Loss: 0.8441025018692017:  87%|████████▋ | 5644/6470 [1:14:00<12:13,  1.13it/s]
1|5645|Loss: 0.9970039129257202:  87%|████████▋ | 5644/6470 [1:14:01<12:13,  1.13it/s]
1|5645|Loss: 0.9970039129257202:  87%|████████▋ | 5645/6470 [1:14:01<12:16,  1.12it/s]
1|5646|Loss: 0.823360800743103:  87%|████████▋ | 5645/6470 [1:14:02<12:16,  1.12it/s] 
1|5646|Loss: 0.823360800743103:  87%|████████▋ | 5646/6470 [1:14:02<12:17,  1.12it/s]
1|5647|Loss: 0.8564836978912354:  87%|████████▋ | 5646/6470 [1:14:03<12:17,  1.12it/s]
1|5647|Loss: 0.8564836978912354:  87%|████████▋ | 5647/6470 [1:14:03<11:50,  1.16it/s]
1|5648|Loss: 0.8005690574645996:  87%|████████▋ | 5647/6470 [1:14:03<11:50,  1.16it/s]
1|5648|Loss: 0.8005690574645996:  87%|████████▋ | 5648/6470 [1:14:03<10:46,  1.27it/s]
1|5649|Loss: 0.993865966796875:  87%|████████▋ | 5648/6470 [1:14:04<10:46,  1.27it/s] 
1|5649|Loss: 0.993865966796875:  87%|████████▋ | 5649/6470 [1:14:04<10:35,  1.29it/s]
1|5650|Loss: 0.7667181491851807:  87%|████████▋ | 5649/6470 [1:14:05<10:35,  1.29it/s]
1|5650|Loss: 0.7667181491851807:  87%|████████▋ | 5650/6470 [1:14:05<10:56,  1.25it/s]
1|5651|Loss: 0.896321177482605:  87%|████████▋ | 5650/6470 [1:14:05<10:56,  1.25it/s] 
1|5651|Loss: 0.896321177482605:  87%|████████▋ | 5651/6470 [1:14:06<10:24,  1.31it/s]
1|5652|Loss: 0.9244539141654968:  87%|████████▋ | 5651/6470 [1:14:06<10:24,  1.31it/s]
1|5652|Loss: 0.9244539141654968:  87%|████████▋ | 5652/6470 [1:14:06<09:40,  1.41it/s]
1|5653|Loss: 0.7820959091186523:  87%|████████▋ | 5652/6470 [1:14:07<09:40,  1.41it/s]
1|5653|Loss: 0.7820959091186523:  87%|████████▋ | 5653/6470 [1:14:07<09:00,  1.51it/s]
1|5654|Loss: 0.9052266478538513:  87%|████████▋ | 5653/6470 [1:14:07<09:00,  1.51it/s]
1|5654|Loss: 0.9052266478538513:  87%|████████▋ | 5654/6470 [1:14:07<07:58,  1.71it/s]
1|5655|Loss: 0.9349240064620972:  87%|████████▋ | 5654/6470 [1:14:08<07:58,  1.71it/s]
1|5655|Loss: 0.9349240064620972:  87%|████████▋ | 5655/6470 [1:14:08<08:19,  1.63it/s]
1|5656|Loss: 0.7281810641288757:  87%|████████▋ | 5655/6470 [1:14:08<08:19,  1.63it/s]
1|5656|Loss: 0.7281810641288757:  87%|████████▋ | 5656/6470 [1:14:09<09:01,  1.50it/s]
1|5657|Loss: 0.9911127090454102:  87%|████████▋ | 5656/6470 [1:14:09<09:01,  1.50it/s]
1|5657|Loss: 0.9911127090454102:  87%|████████▋ | 5657/6470 [1:14:09<08:59,  1.51it/s]
1|5658|Loss: 1.031555414199829:  87%|████████▋ | 5657/6470 [1:14:10<08:59,  1.51it/s] 
1|5658|Loss: 1.031555414199829:  87%|████████▋ | 5658/6470 [1:14:10<09:39,  1.40it/s]
1|5659|Loss: 0.6944851279258728:  87%|████████▋ | 5658/6470 [1:14:11<09:39,  1.40it/s]
1|5659|Loss: 0.6944851279258728:  87%|████████▋ | 5659/6470 [1:14:11<09:43,  1.39it/s]
1|5660|Loss: 0.7783238887786865:  87%|████████▋ | 5659/6470 [1:14:11<09:43,  1.39it/s]
1|5660|Loss: 0.7783238887786865:  87%|████████▋ | 5660/6470 [1:14:12<09:51,  1.37it/s]
1|5661|Loss: 0.7450425028800964:  87%|████████▋ | 5660/6470 [1:14:12<09:51,  1.37it/s]
1|5661|Loss: 0.7450425028800964:  87%|████████▋ | 5661/6470 [1:14:12<09:09,  1.47it/s]
1|5662|Loss: 0.8263552188873291:  87%|████████▋ | 5661/6470 [1:14:13<09:09,  1.47it/s]
1|5662|Loss: 0.8263552188873291:  88%|████████▊ | 5662/6470 [1:14:13<08:37,  1.56it/s]
1|5663|Loss: 0.8767510056495667:  88%|████████▊ | 5662/6470 [1:14:13<08:37,  1.56it/s]
1|5663|Loss: 0.8767510056495667:  88%|████████▊ | 5663/6470 [1:14:14<09:01,  1.49it/s]
1|5664|Loss: 0.7700511813163757:  88%|████████▊ | 5663/6470 [1:14:14<09:01,  1.49it/s]
1|5664|Loss: 0.7700511813163757:  88%|████████▊ | 5664/6470 [1:14:14<09:22,  1.43it/s]
1|5665|Loss: 0.8612694144248962:  88%|████████▊ | 5664/6470 [1:14:15<09:22,  1.43it/s]
1|5665|Loss: 0.8612694144248962:  88%|████████▊ | 5665/6470 [1:14:15<08:40,  1.55it/s]
1|5666|Loss: 0.8502280116081238:  88%|████████▊ | 5665/6470 [1:14:15<08:40,  1.55it/s]
1|5666|Loss: 0.8502280116081238:  88%|████████▊ | 5666/6470 [1:14:16<08:56,  1.50it/s]
1|5667|Loss: 0.796651303768158:  88%|████████▊ | 5666/6470 [1:14:16<08:56,  1.50it/s] 
1|5667|Loss: 0.796651303768158:  88%|████████▊ | 5667/6470 [1:14:17<10:17,  1.30it/s]
1|5668|Loss: 0.813200056552887:  88%|████████▊ | 5667/6470 [1:14:17<10:17,  1.30it/s]
1|5668|Loss: 0.813200056552887:  88%|████████▊ | 5668/6470 [1:14:17<10:20,  1.29it/s]
1|5669|Loss: 0.8716307282447815:  88%|████████▊ | 5668/6470 [1:14:18<10:20,  1.29it/s]
1|5669|Loss: 0.8716307282447815:  88%|████████▊ | 5669/6470 [1:14:18<10:35,  1.26it/s]
1|5670|Loss: 0.8538714051246643:  88%|████████▊ | 5669/6470 [1:14:19<10:35,  1.26it/s]
1|5670|Loss: 0.8538714051246643:  88%|████████▊ | 5670/6470 [1:14:19<10:16,  1.30it/s]
1|5671|Loss: 0.7309629321098328:  88%|████████▊ | 5670/6470 [1:14:19<10:16,  1.30it/s]
1|5671|Loss: 0.7309629321098328:  88%|████████▊ | 5671/6470 [1:14:19<08:53,  1.50it/s]
1|5672|Loss: 0.7938732504844666:  88%|████████▊ | 5671/6470 [1:14:20<08:53,  1.50it/s]
1|5672|Loss: 0.7938732504844666:  88%|████████▊ | 5672/6470 [1:14:20<08:35,  1.55it/s]
1|5673|Loss: 0.8770080804824829:  88%|████████▊ | 5672/6470 [1:14:20<08:35,  1.55it/s]
1|5673|Loss: 0.8770080804824829:  88%|████████▊ | 5673/6470 [1:14:21<08:59,  1.48it/s]
1|5674|Loss: 0.8617148995399475:  88%|████████▊ | 5673/6470 [1:14:21<08:59,  1.48it/s]
1|5674|Loss: 0.8617148995399475:  88%|████████▊ | 5674/6470 [1:14:21<09:00,  1.47it/s]
1|5675|Loss: 1.0260993242263794:  88%|████████▊ | 5674/6470 [1:14:22<09:00,  1.47it/s]
1|5675|Loss: 1.0260993242263794:  88%|████████▊ | 5675/6470 [1:14:22<09:35,  1.38it/s]
1|5676|Loss: 0.8231130242347717:  88%|████████▊ | 5675/6470 [1:14:23<09:35,  1.38it/s]
1|5676|Loss: 0.8231130242347717:  88%|████████▊ | 5676/6470 [1:14:23<09:38,  1.37it/s]
1|5677|Loss: 0.9220542907714844:  88%|████████▊ | 5676/6470 [1:14:23<09:38,  1.37it/s]
1|5677|Loss: 0.9220542907714844:  88%|████████▊ | 5677/6470 [1:14:24<09:22,  1.41it/s]
1|5678|Loss: 0.7744653820991516:  88%|████████▊ | 5677/6470 [1:14:24<09:22,  1.41it/s]
1|5678|Loss: 0.7744653820991516:  88%|████████▊ | 5678/6470 [1:14:24<09:09,  1.44it/s]
1|5679|Loss: 0.8451060056686401:  88%|████████▊ | 5678/6470 [1:14:25<09:09,  1.44it/s]
1|5679|Loss: 0.8451060056686401:  88%|████████▊ | 5679/6470 [1:14:25<08:37,  1.53it/s]
1|5680|Loss: 0.873336672782898:  88%|████████▊ | 5679/6470 [1:14:25<08:37,  1.53it/s] 
1|5680|Loss: 0.873336672782898:  88%|████████▊ | 5680/6470 [1:14:26<09:18,  1.42it/s]
1|5681|Loss: 0.7710058093070984:  88%|████████▊ | 5680/6470 [1:14:26<09:18,  1.42it/s]
1|5681|Loss: 0.7710058093070984:  88%|████████▊ | 5681/6470 [1:14:26<08:59,  1.46it/s]
1|5682|Loss: 0.9577001929283142:  88%|████████▊ | 5681/6470 [1:14:27<08:59,  1.46it/s]
1|5682|Loss: 0.9577001929283142:  88%|████████▊ | 5682/6470 [1:14:27<09:25,  1.39it/s]
1|5683|Loss: 0.8123655915260315:  88%|████████▊ | 5682/6470 [1:14:27<09:25,  1.39it/s]
1|5683|Loss: 0.8123655915260315:  88%|████████▊ | 5683/6470 [1:14:28<08:48,  1.49it/s]
1|5684|Loss: 0.853783905506134:  88%|████████▊ | 5683/6470 [1:14:28<08:48,  1.49it/s] 
1|5684|Loss: 0.853783905506134:  88%|████████▊ | 5684/6470 [1:14:28<08:58,  1.46it/s]
1|5685|Loss: 0.8158000707626343:  88%|████████▊ | 5684/6470 [1:14:29<08:58,  1.46it/s]
1|5685|Loss: 0.8158000707626343:  88%|████████▊ | 5685/6470 [1:14:29<09:25,  1.39it/s]
1|5686|Loss: 0.7330767512321472:  88%|████████▊ | 5685/6470 [1:14:30<09:25,  1.39it/s]
1|5686|Loss: 0.7330767512321472:  88%|████████▊ | 5686/6470 [1:14:30<09:23,  1.39it/s]
1|5687|Loss: 0.8045737147331238:  88%|████████▊ | 5686/6470 [1:14:30<09:23,  1.39it/s]
1|5687|Loss: 0.8045737147331238:  88%|████████▊ | 5687/6470 [1:14:31<09:26,  1.38it/s]
1|5688|Loss: 0.9235644340515137:  88%|████████▊ | 5687/6470 [1:14:31<09:26,  1.38it/s]
1|5688|Loss: 0.9235644340515137:  88%|████████▊ | 5688/6470 [1:14:31<09:13,  1.41it/s]
1|5689|Loss: 0.918499767780304:  88%|████████▊ | 5688/6470 [1:14:32<09:13,  1.41it/s] 
1|5689|Loss: 0.918499767780304:  88%|████████▊ | 5689/6470 [1:14:32<09:19,  1.40it/s]
1|5690|Loss: 0.678783118724823:  88%|████████▊ | 5689/6470 [1:14:33<09:19,  1.40it/s]
1|5690|Loss: 0.678783118724823:  88%|████████▊ | 5690/6470 [1:14:33<10:22,  1.25it/s]
1|5691|Loss: 0.8623872399330139:  88%|████████▊ | 5690/6470 [1:14:34<10:22,  1.25it/s]
1|5691|Loss: 0.8623872399330139:  88%|████████▊ | 5691/6470 [1:14:34<10:35,  1.23it/s]
1|5692|Loss: 0.9254781007766724:  88%|████████▊ | 5691/6470 [1:14:34<10:35,  1.23it/s]
1|5692|Loss: 0.9254781007766724:  88%|████████▊ | 5692/6470 [1:14:35<10:03,  1.29it/s]
1|5693|Loss: 0.6856092810630798:  88%|████████▊ | 5692/6470 [1:14:35<10:03,  1.29it/s]
1|5693|Loss: 0.6856092810630798:  88%|████████▊ | 5693/6470 [1:14:35<09:05,  1.43it/s]
1|5694|Loss: 0.8744679093360901:  88%|████████▊ | 5693/6470 [1:14:36<09:05,  1.43it/s]
1|5694|Loss: 0.8744679093360901:  88%|████████▊ | 5694/6470 [1:14:36<09:29,  1.36it/s]
1|5695|Loss: 0.6921281218528748:  88%|████████▊ | 5694/6470 [1:14:36<09:29,  1.36it/s]
1|5695|Loss: 0.6921281218528748:  88%|████████▊ | 5695/6470 [1:14:37<10:10,  1.27it/s]
1|5696|Loss: 0.784365713596344:  88%|████████▊ | 5695/6470 [1:14:37<10:10,  1.27it/s] 
1|5696|Loss: 0.784365713596344:  88%|████████▊ | 5696/6470 [1:14:38<09:48,  1.32it/s]
1|5697|Loss: 0.8966224193572998:  88%|████████▊ | 5696/6470 [1:14:38<09:48,  1.32it/s]
1|5697|Loss: 0.8966224193572998:  88%|████████▊ | 5697/6470 [1:14:38<09:48,  1.31it/s]
1|5698|Loss: 0.8889980316162109:  88%|████████▊ | 5697/6470 [1:14:39<09:48,  1.31it/s]
1|5698|Loss: 0.8889980316162109:  88%|████████▊ | 5698/6470 [1:14:39<09:24,  1.37it/s]
1|5699|Loss: 0.9007048010826111:  88%|████████▊ | 5698/6470 [1:14:39<09:24,  1.37it/s]
1|5699|Loss: 0.9007048010826111:  88%|████████▊ | 5699/6470 [1:14:40<09:26,  1.36it/s]
1|5700|Loss: 0.7943685054779053:  88%|████████▊ | 5699/6470 [1:14:40<09:26,  1.36it/s]
1|5700|Loss: 0.7943685054779053:  88%|████████▊ | 5700/6470 [1:14:40<09:01,  1.42it/s]
1|5701|Loss: 0.9396979212760925:  88%|████████▊ | 5700/6470 [1:14:41<09:01,  1.42it/s]
1|5701|Loss: 0.9396979212760925:  88%|████████▊ | 5701/6470 [1:14:41<09:05,  1.41it/s]
1|5702|Loss: 0.9353282451629639:  88%|████████▊ | 5701/6470 [1:14:42<09:05,  1.41it/s]
1|5702|Loss: 0.9353282451629639:  88%|████████▊ | 5702/6470 [1:14:42<09:40,  1.32it/s]
1|5703|Loss: 0.8635769486427307:  88%|████████▊ | 5702/6470 [1:14:42<09:40,  1.32it/s]
1|5703|Loss: 0.8635769486427307:  88%|████████▊ | 5703/6470 [1:14:43<09:33,  1.34it/s]
1|5704|Loss: 0.793027400970459:  88%|████████▊ | 5703/6470 [1:14:43<09:33,  1.34it/s] 
1|5704|Loss: 0.793027400970459:  88%|████████▊ | 5704/6470 [1:14:43<09:30,  1.34it/s]
1|5705|Loss: 0.8070052266120911:  88%|████████▊ | 5704/6470 [1:14:44<09:30,  1.34it/s]
1|5705|Loss: 0.8070052266120911:  88%|████████▊ | 5705/6470 [1:14:44<08:52,  1.44it/s]
1|5706|Loss: 0.6831871867179871:  88%|████████▊ | 5705/6470 [1:14:44<08:52,  1.44it/s]
1|5706|Loss: 0.6831871867179871:  88%|████████▊ | 5706/6470 [1:14:45<08:55,  1.43it/s]
1|5707|Loss: 0.9239130616188049:  88%|████████▊ | 5706/6470 [1:14:45<08:55,  1.43it/s]
1|5707|Loss: 0.9239130616188049:  88%|████████▊ | 5707/6470 [1:14:46<09:30,  1.34it/s]
1|5708|Loss: 0.8950989246368408:  88%|████████▊ | 5707/6470 [1:14:46<09:30,  1.34it/s]
1|5708|Loss: 0.8950989246368408:  88%|████████▊ | 5708/6470 [1:14:46<08:53,  1.43it/s]
1|5709|Loss: 0.7929378747940063:  88%|████████▊ | 5708/6470 [1:14:46<08:53,  1.43it/s]
1|5709|Loss: 0.7929378747940063:  88%|████████▊ | 5709/6470 [1:14:47<08:51,  1.43it/s]
1|5710|Loss: 0.7931191325187683:  88%|████████▊ | 5709/6470 [1:14:47<08:51,  1.43it/s]
1|5710|Loss: 0.7931191325187683:  88%|████████▊ | 5710/6470 [1:14:48<09:13,  1.37it/s]
1|5711|Loss: 0.7972752451896667:  88%|████████▊ | 5710/6470 [1:14:48<09:13,  1.37it/s]
1|5711|Loss: 0.7972752451896667:  88%|████████▊ | 5711/6470 [1:14:48<09:26,  1.34it/s]
1|5712|Loss: 0.8116082549095154:  88%|████████▊ | 5711/6470 [1:14:49<09:26,  1.34it/s]
1|5712|Loss: 0.8116082549095154:  88%|████████▊ | 5712/6470 [1:14:49<10:18,  1.22it/s]
1|5713|Loss: 0.806035041809082:  88%|████████▊ | 5712/6470 [1:14:50<10:18,  1.22it/s] 
1|5713|Loss: 0.806035041809082:  88%|████████▊ | 5713/6470 [1:14:50<10:26,  1.21it/s]
1|5714|Loss: 0.8774305582046509:  88%|████████▊ | 5713/6470 [1:14:51<10:26,  1.21it/s]
1|5714|Loss: 0.8774305582046509:  88%|████████▊ | 5714/6470 [1:14:51<09:57,  1.27it/s]
1|5715|Loss: 0.7494691610336304:  88%|████████▊ | 5714/6470 [1:14:51<09:57,  1.27it/s]
1|5715|Loss: 0.7494691610336304:  88%|████████▊ | 5715/6470 [1:14:52<09:52,  1.27it/s]
1|5716|Loss: 0.9397862553596497:  88%|████████▊ | 5715/6470 [1:14:52<09:52,  1.27it/s]
1|5716|Loss: 0.9397862553596497:  88%|████████▊ | 5716/6470 [1:14:52<09:38,  1.30it/s]
1|5717|Loss: 0.9129104614257812:  88%|████████▊ | 5716/6470 [1:14:53<09:38,  1.30it/s]
1|5717|Loss: 0.9129104614257812:  88%|████████▊ | 5717/6470 [1:14:53<09:24,  1.33it/s]
1|5718|Loss: 0.9465002417564392:  88%|████████▊ | 5717/6470 [1:14:54<09:24,  1.33it/s]
1|5718|Loss: 0.9465002417564392:  88%|████████▊ | 5718/6470 [1:14:54<09:29,  1.32it/s]
1|5719|Loss: 0.8712924122810364:  88%|████████▊ | 5718/6470 [1:14:54<09:29,  1.32it/s]
1|5719|Loss: 0.8712924122810364:  88%|████████▊ | 5719/6470 [1:14:55<09:38,  1.30it/s]
1|5720|Loss: 0.9701938033103943:  88%|████████▊ | 5719/6470 [1:14:55<09:38,  1.30it/s]
1|5720|Loss: 0.9701938033103943:  88%|████████▊ | 5720/6470 [1:14:55<08:59,  1.39it/s]
1|5721|Loss: 0.7656447291374207:  88%|████████▊ | 5720/6470 [1:14:56<08:59,  1.39it/s]
1|5721|Loss: 0.7656447291374207:  88%|████████▊ | 5721/6470 [1:14:56<08:21,  1.49it/s]
1|5722|Loss: 0.8069915175437927:  88%|████████▊ | 5721/6470 [1:14:56<08:21,  1.49it/s]
1|5722|Loss: 0.8069915175437927:  88%|████████▊ | 5722/6470 [1:14:56<07:46,  1.60it/s]
1|5723|Loss: 0.8836396336555481:  88%|████████▊ | 5722/6470 [1:14:57<07:46,  1.60it/s]
1|5723|Loss: 0.8836396336555481:  88%|████████▊ | 5723/6470 [1:14:57<08:13,  1.51it/s]
1|5724|Loss: 0.8149863481521606:  88%|████████▊ | 5723/6470 [1:14:58<08:13,  1.51it/s]
1|5724|Loss: 0.8149863481521606:  88%|████████▊ | 5724/6470 [1:14:58<08:23,  1.48it/s]
1|5725|Loss: 0.9561963677406311:  88%|████████▊ | 5724/6470 [1:14:58<08:23,  1.48it/s]
1|5725|Loss: 0.9561963677406311:  88%|████████▊ | 5725/6470 [1:14:59<09:09,  1.36it/s]
1|5726|Loss: 0.8780927062034607:  88%|████████▊ | 5725/6470 [1:14:59<09:09,  1.36it/s]
1|5726|Loss: 0.8780927062034607:  89%|████████▊ | 5726/6470 [1:15:00<09:30,  1.30it/s]
1|5727|Loss: 1.0099412202835083:  89%|████████▊ | 5726/6470 [1:15:00<09:30,  1.30it/s]
1|5727|Loss: 1.0099412202835083:  89%|████████▊ | 5727/6470 [1:15:00<10:00,  1.24it/s]
1|5728|Loss: 0.8465752005577087:  89%|████████▊ | 5727/6470 [1:15:01<10:00,  1.24it/s]
1|5728|Loss: 0.8465752005577087:  89%|████████▊ | 5728/6470 [1:15:01<09:58,  1.24it/s]
1|5729|Loss: 0.8418816924095154:  89%|████████▊ | 5728/6470 [1:15:02<09:58,  1.24it/s]
1|5729|Loss: 0.8418816924095154:  89%|████████▊ | 5729/6470 [1:15:02<09:54,  1.25it/s]
1|5730|Loss: 0.7773919701576233:  89%|████████▊ | 5729/6470 [1:15:03<09:54,  1.25it/s]
1|5730|Loss: 0.7773919701576233:  89%|████████▊ | 5730/6470 [1:15:03<10:10,  1.21it/s]
1|5731|Loss: 0.8301395773887634:  89%|████████▊ | 5730/6470 [1:15:03<10:10,  1.21it/s]
1|5731|Loss: 0.8301395773887634:  89%|████████▊ | 5731/6470 [1:15:04<09:22,  1.31it/s]
1|5732|Loss: 0.7868022918701172:  89%|████████▊ | 5731/6470 [1:15:04<09:22,  1.31it/s]
1|5732|Loss: 0.7868022918701172:  89%|████████▊ | 5732/6470 [1:15:04<09:08,  1.34it/s]
1|5733|Loss: 0.9743304252624512:  89%|████████▊ | 5732/6470 [1:15:05<09:08,  1.34it/s]
1|5733|Loss: 0.9743304252624512:  89%|████████▊ | 5733/6470 [1:15:05<09:38,  1.27it/s]
1|5734|Loss: 0.8458847999572754:  89%|████████▊ | 5733/6470 [1:15:06<09:38,  1.27it/s]
1|5734|Loss: 0.8458847999572754:  89%|████████▊ | 5734/6470 [1:15:06<10:01,  1.22it/s]
1|5735|Loss: 0.7381836175918579:  89%|████████▊ | 5734/6470 [1:15:06<10:01,  1.22it/s]
1|5735|Loss: 0.7381836175918579:  89%|████████▊ | 5735/6470 [1:15:06<08:44,  1.40it/s]
1|5736|Loss: 0.7662800550460815:  89%|████████▊ | 5735/6470 [1:15:07<08:44,  1.40it/s]
1|5736|Loss: 0.7662800550460815:  89%|████████▊ | 5736/6470 [1:15:07<08:14,  1.48it/s]
1|5737|Loss: 0.7691729068756104:  89%|████████▊ | 5736/6470 [1:15:07<08:14,  1.48it/s]
1|5737|Loss: 0.7691729068756104:  89%|████████▊ | 5737/6470 [1:15:08<08:33,  1.43it/s]
1|5738|Loss: 0.7527190446853638:  89%|████████▊ | 5737/6470 [1:15:08<08:33,  1.43it/s]
1|5738|Loss: 0.7527190446853638:  89%|████████▊ | 5738/6470 [1:15:09<08:51,  1.38it/s]
1|5739|Loss: 0.8948615193367004:  89%|████████▊ | 5738/6470 [1:15:09<08:51,  1.38it/s]
1|5739|Loss: 0.8948615193367004:  89%|████████▊ | 5739/6470 [1:15:09<08:49,  1.38it/s]
1|5740|Loss: 0.9482530355453491:  89%|████████▊ | 5739/6470 [1:15:10<08:49,  1.38it/s]
1|5740|Loss: 0.9482530355453491:  89%|████████▊ | 5740/6470 [1:15:10<08:58,  1.36it/s]
1|5741|Loss: 0.9539194703102112:  89%|████████▊ | 5740/6470 [1:15:11<08:58,  1.36it/s]
1|5741|Loss: 0.9539194703102112:  89%|████████▊ | 5741/6470 [1:15:11<08:52,  1.37it/s]
1|5742|Loss: 0.8445263504981995:  89%|████████▊ | 5741/6470 [1:15:11<08:52,  1.37it/s]
1|5742|Loss: 0.8445263504981995:  89%|████████▊ | 5742/6470 [1:15:12<08:54,  1.36it/s]
1|5743|Loss: 0.7757671475410461:  89%|████████▊ | 5742/6470 [1:15:12<08:54,  1.36it/s]
1|5743|Loss: 0.7757671475410461:  89%|████████▉ | 5743/6470 [1:15:13<09:39,  1.25it/s]
1|5744|Loss: 0.8773707747459412:  89%|████████▉ | 5743/6470 [1:15:13<09:39,  1.25it/s]
1|5744|Loss: 0.8773707747459412:  89%|████████▉ | 5744/6470 [1:15:13<09:41,  1.25it/s]
1|5745|Loss: 1.009668231010437:  89%|████████▉ | 5744/6470 [1:15:14<09:41,  1.25it/s] 
1|5745|Loss: 1.009668231010437:  89%|████████▉ | 5745/6470 [1:15:14<09:49,  1.23it/s]
1|5746|Loss: 0.8324875235557556:  89%|████████▉ | 5745/6470 [1:15:15<09:49,  1.23it/s]
1|5746|Loss: 0.8324875235557556:  89%|████████▉ | 5746/6470 [1:15:15<10:01,  1.20it/s]
1|5747|Loss: 0.8524630665779114:  89%|████████▉ | 5746/6470 [1:15:15<10:01,  1.20it/s]
1|5747|Loss: 0.8524630665779114:  89%|████████▉ | 5747/6470 [1:15:16<09:23,  1.28it/s]
1|5748|Loss: 0.8718035817146301:  89%|████████▉ | 5747/6470 [1:15:16<09:23,  1.28it/s]
1|5748|Loss: 0.8718035817146301:  89%|████████▉ | 5748/6470 [1:15:16<08:42,  1.38it/s]
1|5749|Loss: 1.16208016872406:  89%|████████▉ | 5748/6470 [1:15:17<08:42,  1.38it/s]  
1|5749|Loss: 1.16208016872406:  89%|████████▉ | 5749/6470 [1:15:17<08:50,  1.36it/s]
1|5750|Loss: 0.8180347681045532:  89%|████████▉ | 5749/6470 [1:15:18<08:50,  1.36it/s]
1|5750|Loss: 0.8180347681045532:  89%|████████▉ | 5750/6470 [1:15:18<09:04,  1.32it/s]
1|5751|Loss: 0.9171590209007263:  89%|████████▉ | 5750/6470 [1:15:18<09:04,  1.32it/s]
1|5751|Loss: 0.9171590209007263:  89%|████████▉ | 5751/6470 [1:15:19<08:52,  1.35it/s]
1|5752|Loss: 0.850857675075531:  89%|████████▉ | 5751/6470 [1:15:19<08:52,  1.35it/s] 
1|5752|Loss: 0.850857675075531:  89%|████████▉ | 5752/6470 [1:15:19<09:15,  1.29it/s]
1|5753|Loss: 0.9273973107337952:  89%|████████▉ | 5752/6470 [1:15:20<09:15,  1.29it/s]
1|5753|Loss: 0.9273973107337952:  89%|████████▉ | 5753/6470 [1:15:20<09:00,  1.33it/s]
1|5754|Loss: 0.8465468287467957:  89%|████████▉ | 5753/6470 [1:15:20<09:00,  1.33it/s]
1|5754|Loss: 0.8465468287467957:  89%|████████▉ | 5754/6470 [1:15:21<08:32,  1.40it/s]
1|5755|Loss: 0.9012833833694458:  89%|████████▉ | 5754/6470 [1:15:21<08:32,  1.40it/s]
1|5755|Loss: 0.9012833833694458:  89%|████████▉ | 5755/6470 [1:15:22<08:43,  1.37it/s]
1|5756|Loss: 0.6741446852684021:  89%|████████▉ | 5755/6470 [1:15:22<08:43,  1.37it/s]
1|5756|Loss: 0.6741446852684021:  89%|████████▉ | 5756/6470 [1:15:22<08:55,  1.33it/s]
1|5757|Loss: 0.7660147547721863:  89%|████████▉ | 5756/6470 [1:15:23<08:55,  1.33it/s]
1|5757|Loss: 0.7660147547721863:  89%|████████▉ | 5757/6470 [1:15:23<09:21,  1.27it/s]
1|5758|Loss: 1.0868589878082275:  89%|████████▉ | 5757/6470 [1:15:24<09:21,  1.27it/s]
1|5758|Loss: 1.0868589878082275:  89%|████████▉ | 5758/6470 [1:15:24<10:01,  1.18it/s]
1|5759|Loss: 0.8840858340263367:  89%|████████▉ | 5758/6470 [1:15:25<10:01,  1.18it/s]
1|5759|Loss: 0.8840858340263367:  89%|████████▉ | 5759/6470 [1:15:25<09:58,  1.19it/s]
1|5760|Loss: 0.838570773601532:  89%|████████▉ | 5759/6470 [1:15:25<09:58,  1.19it/s] 
1|5760|Loss: 0.838570773601532:  89%|████████▉ | 5760/6470 [1:15:26<09:48,  1.21it/s]
1|5761|Loss: 0.72914719581604:  89%|████████▉ | 5760/6470 [1:15:26<09:48,  1.21it/s] 
1|5761|Loss: 0.72914719581604:  89%|████████▉ | 5761/6470 [1:15:27<09:37,  1.23it/s]
1|5762|Loss: 0.8929657936096191:  89%|████████▉ | 5761/6470 [1:15:27<09:37,  1.23it/s]
1|5762|Loss: 0.8929657936096191:  89%|████████▉ | 5762/6470 [1:15:27<08:26,  1.40it/s]
1|5763|Loss: 0.7792234420776367:  89%|████████▉ | 5762/6470 [1:15:27<08:26,  1.40it/s]
1|5763|Loss: 0.7792234420776367:  89%|████████▉ | 5763/6470 [1:15:28<08:10,  1.44it/s]
1|5764|Loss: 0.8638060092926025:  89%|████████▉ | 5763/6470 [1:15:28<08:10,  1.44it/s]
1|5764|Loss: 0.8638060092926025:  89%|████████▉ | 5764/6470 [1:15:29<08:43,  1.35it/s]
1|5765|Loss: 0.8742608428001404:  89%|████████▉ | 5764/6470 [1:15:29<08:43,  1.35it/s]
1|5765|Loss: 0.8742608428001404:  89%|████████▉ | 5765/6470 [1:15:29<08:30,  1.38it/s]
1|5766|Loss: 0.7444993853569031:  89%|████████▉ | 5765/6470 [1:15:30<08:30,  1.38it/s]
1|5766|Loss: 0.7444993853569031:  89%|████████▉ | 5766/6470 [1:15:30<08:34,  1.37it/s]
1|5767|Loss: 0.685327410697937:  89%|████████▉ | 5766/6470 [1:15:30<08:34,  1.37it/s] 
1|5767|Loss: 0.685327410697937:  89%|████████▉ | 5767/6470 [1:15:31<08:04,  1.45it/s]
1|5768|Loss: 1.0302222967147827:  89%|████████▉ | 5767/6470 [1:15:31<08:04,  1.45it/s]
1|5768|Loss: 1.0302222967147827:  89%|████████▉ | 5768/6470 [1:15:31<07:50,  1.49it/s]
1|5769|Loss: 0.9042221307754517:  89%|████████▉ | 5768/6470 [1:15:32<07:50,  1.49it/s]
1|5769|Loss: 0.9042221307754517:  89%|████████▉ | 5769/6470 [1:15:32<07:29,  1.56it/s]
1|5770|Loss: 0.9325756430625916:  89%|████████▉ | 5769/6470 [1:15:32<07:29,  1.56it/s]
1|5770|Loss: 0.9325756430625916:  89%|████████▉ | 5770/6470 [1:15:33<08:15,  1.41it/s]
1|5771|Loss: 0.7671958208084106:  89%|████████▉ | 5770/6470 [1:15:33<08:15,  1.41it/s]
1|5771|Loss: 0.7671958208084106:  89%|████████▉ | 5771/6470 [1:15:33<08:41,  1.34it/s]
1|5772|Loss: 0.8643025755882263:  89%|████████▉ | 5771/6470 [1:15:34<08:41,  1.34it/s]
1|5772|Loss: 0.8643025755882263:  89%|████████▉ | 5772/6470 [1:15:34<08:51,  1.31it/s]
1|5773|Loss: 0.9370565414428711:  89%|████████▉ | 5772/6470 [1:15:35<08:51,  1.31it/s]
1|5773|Loss: 0.9370565414428711:  89%|████████▉ | 5773/6470 [1:15:35<08:16,  1.40it/s]
1|5774|Loss: 0.8188199996948242:  89%|████████▉ | 5773/6470 [1:15:35<08:16,  1.40it/s]
1|5774|Loss: 0.8188199996948242:  89%|████████▉ | 5774/6470 [1:15:36<08:14,  1.41it/s]
1|5775|Loss: 0.9757674336433411:  89%|████████▉ | 5774/6470 [1:15:36<08:14,  1.41it/s]
1|5775|Loss: 0.9757674336433411:  89%|████████▉ | 5775/6470 [1:15:36<08:29,  1.36it/s]
1|5776|Loss: 0.9250898361206055:  89%|████████▉ | 5775/6470 [1:15:37<08:29,  1.36it/s]
1|5776|Loss: 0.9250898361206055:  89%|████████▉ | 5776/6470 [1:15:37<08:54,  1.30it/s]
1|5777|Loss: 0.9309516549110413:  89%|████████▉ | 5776/6470 [1:15:38<08:54,  1.30it/s]
1|5777|Loss: 0.9309516549110413:  89%|████████▉ | 5777/6470 [1:15:38<09:00,  1.28it/s]
1|5778|Loss: 0.8253268599510193:  89%|████████▉ | 5777/6470 [1:15:38<09:00,  1.28it/s]
1|5778|Loss: 0.8253268599510193:  89%|████████▉ | 5778/6470 [1:15:39<09:03,  1.27it/s]
1|5779|Loss: 0.9305999279022217:  89%|████████▉ | 5778/6470 [1:15:39<09:03,  1.27it/s]
1|5779|Loss: 0.9305999279022217:  89%|████████▉ | 5779/6470 [1:15:40<08:45,  1.32it/s]
1|5780|Loss: 0.8681981563568115:  89%|████████▉ | 5779/6470 [1:15:40<08:45,  1.32it/s]
1|5780|Loss: 0.8681981563568115:  89%|████████▉ | 5780/6470 [1:15:40<08:47,  1.31it/s]
1|5781|Loss: 0.7973318099975586:  89%|████████▉ | 5780/6470 [1:15:41<08:47,  1.31it/s]
1|5781|Loss: 0.7973318099975586:  89%|████████▉ | 5781/6470 [1:15:41<09:14,  1.24it/s]
1|5782|Loss: 0.9554230570793152:  89%|████████▉ | 5781/6470 [1:15:42<09:14,  1.24it/s]
1|5782|Loss: 0.9554230570793152:  89%|████████▉ | 5782/6470 [1:15:42<09:17,  1.23it/s]
1|5783|Loss: 0.8612968325614929:  89%|████████▉ | 5782/6470 [1:15:42<09:17,  1.23it/s]
1|5783|Loss: 0.8612968325614929:  89%|████████▉ | 5783/6470 [1:15:43<09:27,  1.21it/s]
1|5784|Loss: 0.9434207677841187:  89%|████████▉ | 5783/6470 [1:15:43<09:27,  1.21it/s]
1|5784|Loss: 0.9434207677841187:  89%|████████▉ | 5784/6470 [1:15:44<09:24,  1.22it/s]
1|5785|Loss: 0.8267375230789185:  89%|████████▉ | 5784/6470 [1:15:44<09:24,  1.22it/s]
1|5785|Loss: 0.8267375230789185:  89%|████████▉ | 5785/6470 [1:15:44<08:55,  1.28it/s]
1|5786|Loss: 0.9052594900131226:  89%|████████▉ | 5785/6470 [1:15:45<08:55,  1.28it/s]
1|5786|Loss: 0.9052594900131226:  89%|████████▉ | 5786/6470 [1:15:45<08:54,  1.28it/s]
1|5787|Loss: 0.8809213042259216:  89%|████████▉ | 5786/6470 [1:15:46<08:54,  1.28it/s]
1|5787|Loss: 0.8809213042259216:  89%|████████▉ | 5787/6470 [1:15:46<09:19,  1.22it/s]
1|5788|Loss: 0.9137255549430847:  89%|████████▉ | 5787/6470 [1:15:47<09:19,  1.22it/s]
1|5788|Loss: 0.9137255549430847:  89%|████████▉ | 5788/6470 [1:15:47<09:19,  1.22it/s]
1|5789|Loss: 0.7937315702438354:  89%|████████▉ | 5788/6470 [1:15:47<09:19,  1.22it/s]
1|5789|Loss: 0.7937315702438354:  89%|████████▉ | 5789/6470 [1:15:48<08:41,  1.31it/s]
1|5790|Loss: 0.8245205879211426:  89%|████████▉ | 5789/6470 [1:15:48<08:41,  1.31it/s]
1|5790|Loss: 0.8245205879211426:  89%|████████▉ | 5790/6470 [1:15:48<08:34,  1.32it/s]
1|5791|Loss: 0.7700432538986206:  89%|████████▉ | 5790/6470 [1:15:49<08:34,  1.32it/s]
1|5791|Loss: 0.7700432538986206:  90%|████████▉ | 5791/6470 [1:15:49<09:04,  1.25it/s]
1|5792|Loss: 0.8486312031745911:  90%|████████▉ | 5791/6470 [1:15:50<09:04,  1.25it/s]
1|5792|Loss: 0.8486312031745911:  90%|████████▉ | 5792/6470 [1:15:50<09:12,  1.23it/s]
1|5793|Loss: 0.8983123898506165:  90%|████████▉ | 5792/6470 [1:15:50<09:12,  1.23it/s]
1|5793|Loss: 0.8983123898506165:  90%|████████▉ | 5793/6470 [1:15:51<09:07,  1.24it/s]
1|5794|Loss: 0.9622118473052979:  90%|████████▉ | 5793/6470 [1:15:51<09:07,  1.24it/s]
1|5794|Loss: 0.9622118473052979:  90%|████████▉ | 5794/6470 [1:15:51<08:30,  1.32it/s]
1|5795|Loss: 0.8411421775817871:  90%|████████▉ | 5794/6470 [1:15:52<08:30,  1.32it/s]
1|5795|Loss: 0.8411421775817871:  90%|████████▉ | 5795/6470 [1:15:52<08:27,  1.33it/s]
1|5796|Loss: 0.7417920231819153:  90%|████████▉ | 5795/6470 [1:15:53<08:27,  1.33it/s]
1|5796|Loss: 0.7417920231819153:  90%|████████▉ | 5796/6470 [1:15:53<08:37,  1.30it/s]
1|5797|Loss: 0.9190413951873779:  90%|████████▉ | 5796/6470 [1:15:53<08:37,  1.30it/s]
1|5797|Loss: 0.9190413951873779:  90%|████████▉ | 5797/6470 [1:15:54<08:12,  1.37it/s]
1|5798|Loss: 0.7682648301124573:  90%|████████▉ | 5797/6470 [1:15:54<08:12,  1.37it/s]
1|5798|Loss: 0.7682648301124573:  90%|████████▉ | 5798/6470 [1:15:54<08:25,  1.33it/s]
1|5799|Loss: 0.832568347454071:  90%|████████▉ | 5798/6470 [1:15:55<08:25,  1.33it/s] 
1|5799|Loss: 0.832568347454071:  90%|████████▉ | 5799/6470 [1:15:55<08:55,  1.25it/s]
1|5800|Loss: 0.7678343653678894:  90%|████████▉ | 5799/6470 [1:15:56<08:55,  1.25it/s]
1|5800|Loss: 0.7678343653678894:  90%|████████▉ | 5800/6470 [1:15:56<09:16,  1.20it/s]
1|5801|Loss: 0.9114795327186584:  90%|████████▉ | 5800/6470 [1:15:57<09:16,  1.20it/s]
1|5801|Loss: 0.9114795327186584:  90%|████████▉ | 5801/6470 [1:15:57<09:14,  1.21it/s]
1|5802|Loss: 0.7714158892631531:  90%|████████▉ | 5801/6470 [1:15:57<09:14,  1.21it/s]
1|5802|Loss: 0.7714158892631531:  90%|████████▉ | 5802/6470 [1:15:58<08:52,  1.25it/s]
1|5803|Loss: 0.9350976943969727:  90%|████████▉ | 5802/6470 [1:15:58<08:52,  1.25it/s]
1|5803|Loss: 0.9350976943969727:  90%|████████▉ | 5803/6470 [1:15:59<09:04,  1.22it/s]
1|5804|Loss: 0.8291222453117371:  90%|████████▉ | 5803/6470 [1:15:59<09:04,  1.22it/s]
1|5804|Loss: 0.8291222453117371:  90%|████████▉ | 5804/6470 [1:15:59<08:57,  1.24it/s]
1|5805|Loss: 0.7302212715148926:  90%|████████▉ | 5804/6470 [1:16:00<08:57,  1.24it/s]
1|5805|Loss: 0.7302212715148926:  90%|████████▉ | 5805/6470 [1:16:00<08:32,  1.30it/s]
1|5806|Loss: 0.8672254085540771:  90%|████████▉ | 5805/6470 [1:16:01<08:32,  1.30it/s]
1|5806|Loss: 0.8672254085540771:  90%|████████▉ | 5806/6470 [1:16:01<08:34,  1.29it/s]
1|5807|Loss: 0.8650968074798584:  90%|████████▉ | 5806/6470 [1:16:01<08:34,  1.29it/s]
1|5807|Loss: 0.8650968074798584:  90%|████████▉ | 5807/6470 [1:16:02<08:11,  1.35it/s]
1|5808|Loss: 0.9831881523132324:  90%|████████▉ | 5807/6470 [1:16:02<08:11,  1.35it/s]
1|5808|Loss: 0.9831881523132324:  90%|████████▉ | 5808/6470 [1:16:02<08:11,  1.35it/s]
1|5809|Loss: 0.7840163707733154:  90%|████████▉ | 5808/6470 [1:16:03<08:11,  1.35it/s]
1|5809|Loss: 0.7840163707733154:  90%|████████▉ | 5809/6470 [1:16:03<08:13,  1.34it/s]
1|5810|Loss: 0.8792219161987305:  90%|████████▉ | 5809/6470 [1:16:03<08:13,  1.34it/s]
1|5810|Loss: 0.8792219161987305:  90%|████████▉ | 5810/6470 [1:16:04<07:59,  1.38it/s]
1|5811|Loss: 0.8333450555801392:  90%|████████▉ | 5810/6470 [1:16:04<07:59,  1.38it/s]
1|5811|Loss: 0.8333450555801392:  90%|████████▉ | 5811/6470 [1:16:05<08:05,  1.36it/s]
1|5812|Loss: 1.0060133934020996:  90%|████████▉ | 5811/6470 [1:16:05<08:05,  1.36it/s]
1|5812|Loss: 1.0060133934020996:  90%|████████▉ | 5812/6470 [1:16:05<08:10,  1.34it/s]
1|5813|Loss: 0.9020411372184753:  90%|████████▉ | 5812/6470 [1:16:06<08:10,  1.34it/s]
1|5813|Loss: 0.9020411372184753:  90%|████████▉ | 5813/6470 [1:16:06<08:03,  1.36it/s]
1|5814|Loss: 0.8156672120094299:  90%|████████▉ | 5813/6470 [1:16:06<08:03,  1.36it/s]
1|5814|Loss: 0.8156672120094299:  90%|████████▉ | 5814/6470 [1:16:07<08:04,  1.35it/s]
1|5815|Loss: 0.8185800909996033:  90%|████████▉ | 5814/6470 [1:16:07<08:04,  1.35it/s]
1|5815|Loss: 0.8185800909996033:  90%|████████▉ | 5815/6470 [1:16:08<08:08,  1.34it/s]
1|5816|Loss: 0.7806525826454163:  90%|████████▉ | 5815/6470 [1:16:08<08:08,  1.34it/s]
1|5816|Loss: 0.7806525826454163:  90%|████████▉ | 5816/6470 [1:16:08<08:08,  1.34it/s]
1|5817|Loss: 0.7913146615028381:  90%|████████▉ | 5816/6470 [1:16:09<08:08,  1.34it/s]
1|5817|Loss: 0.7913146615028381:  90%|████████▉ | 5817/6470 [1:16:09<07:38,  1.42it/s]
1|5818|Loss: 0.8603215217590332:  90%|████████▉ | 5817/6470 [1:16:09<07:38,  1.42it/s]
1|5818|Loss: 0.8603215217590332:  90%|████████▉ | 5818/6470 [1:16:10<07:45,  1.40it/s]
1|5819|Loss: 0.9342116713523865:  90%|████████▉ | 5818/6470 [1:16:10<07:45,  1.40it/s]
1|5819|Loss: 0.9342116713523865:  90%|████████▉ | 5819/6470 [1:16:10<07:46,  1.40it/s]
1|5820|Loss: 0.7808275818824768:  90%|████████▉ | 5819/6470 [1:16:11<07:46,  1.40it/s]
1|5820|Loss: 0.7808275818824768:  90%|████████▉ | 5820/6470 [1:16:11<07:57,  1.36it/s]
1|5821|Loss: 0.909591019153595:  90%|████████▉ | 5820/6470 [1:16:11<07:57,  1.36it/s] 
1|5821|Loss: 0.909591019153595:  90%|████████▉ | 5821/6470 [1:16:12<07:47,  1.39it/s]
1|5822|Loss: 1.0322608947753906:  90%|████████▉ | 5821/6470 [1:16:12<07:47,  1.39it/s]
1|5822|Loss: 1.0322608947753906:  90%|████████▉ | 5822/6470 [1:16:12<06:56,  1.56it/s]
1|5823|Loss: 0.9245640635490417:  90%|████████▉ | 5822/6470 [1:16:13<06:56,  1.56it/s]
1|5823|Loss: 0.9245640635490417:  90%|█████████ | 5823/6470 [1:16:13<06:34,  1.64it/s]
1|5824|Loss: 0.9097111225128174:  90%|█████████ | 5823/6470 [1:16:13<06:34,  1.64it/s]
1|5824|Loss: 0.9097111225128174:  90%|█████████ | 5824/6470 [1:16:13<06:24,  1.68it/s]
1|5825|Loss: 0.649116039276123:  90%|█████████ | 5824/6470 [1:16:14<06:24,  1.68it/s] 
1|5825|Loss: 0.649116039276123:  90%|█████████ | 5825/6470 [1:16:14<06:51,  1.57it/s]
1|5826|Loss: 0.8940687775611877:  90%|█████████ | 5825/6470 [1:16:14<06:51,  1.57it/s]
1|5826|Loss: 0.8940687775611877:  90%|█████████ | 5826/6470 [1:16:15<06:53,  1.56it/s]
1|5827|Loss: 0.8833751678466797:  90%|█████████ | 5826/6470 [1:16:15<06:53,  1.56it/s]
1|5827|Loss: 0.8833751678466797:  90%|█████████ | 5827/6470 [1:16:16<07:35,  1.41it/s]
1|5828|Loss: 0.8701044321060181:  90%|█████████ | 5827/6470 [1:16:16<07:35,  1.41it/s]
1|5828|Loss: 0.8701044321060181:  90%|█████████ | 5828/6470 [1:16:17<08:25,  1.27it/s]
1|5829|Loss: 0.7464767694473267:  90%|█████████ | 5828/6470 [1:16:17<08:25,  1.27it/s]
1|5829|Loss: 0.7464767694473267:  90%|█████████ | 5829/6470 [1:16:17<08:33,  1.25it/s]
1|5830|Loss: 0.824059784412384:  90%|█████████ | 5829/6470 [1:16:18<08:33,  1.25it/s] 
1|5830|Loss: 0.824059784412384:  90%|█████████ | 5830/6470 [1:16:18<08:16,  1.29it/s]
1|5831|Loss: 0.8355421423912048:  90%|█████████ | 5830/6470 [1:16:19<08:16,  1.29it/s]
1|5831|Loss: 0.8355421423912048:  90%|█████████ | 5831/6470 [1:16:19<08:12,  1.30it/s]
1|5832|Loss: 0.7605729103088379:  90%|█████████ | 5831/6470 [1:16:19<08:12,  1.30it/s]
1|5832|Loss: 0.7605729103088379:  90%|█████████ | 5832/6470 [1:16:20<08:39,  1.23it/s]
1|5833|Loss: 0.7248526215553284:  90%|█████████ | 5832/6470 [1:16:20<08:39,  1.23it/s]
1|5833|Loss: 0.7248526215553284:  90%|█████████ | 5833/6470 [1:16:21<08:29,  1.25it/s]
1|5834|Loss: 0.7823484539985657:  90%|█████████ | 5833/6470 [1:16:21<08:29,  1.25it/s]
1|5834|Loss: 0.7823484539985657:  90%|█████████ | 5834/6470 [1:16:21<08:21,  1.27it/s]
1|5835|Loss: 0.7617514133453369:  90%|█████████ | 5834/6470 [1:16:22<08:21,  1.27it/s]
1|5835|Loss: 0.7617514133453369:  90%|█████████ | 5835/6470 [1:16:22<08:17,  1.28it/s]
1|5836|Loss: 0.9859733581542969:  90%|█████████ | 5835/6470 [1:16:22<08:17,  1.28it/s]
1|5836|Loss: 0.9859733581542969:  90%|█████████ | 5836/6470 [1:16:23<08:02,  1.32it/s]
1|5837|Loss: 1.0483616590499878:  90%|█████████ | 5836/6470 [1:16:23<08:02,  1.32it/s]
1|5837|Loss: 1.0483616590499878:  90%|█████████ | 5837/6470 [1:16:24<08:00,  1.32it/s]
1|5838|Loss: 0.8767179846763611:  90%|█████████ | 5837/6470 [1:16:24<08:00,  1.32it/s]
1|5838|Loss: 0.8767179846763611:  90%|█████████ | 5838/6470 [1:16:24<07:48,  1.35it/s]
1|5839|Loss: 0.8680474162101746:  90%|█████████ | 5838/6470 [1:16:25<07:48,  1.35it/s]
1|5839|Loss: 0.8680474162101746:  90%|█████████ | 5839/6470 [1:16:25<07:28,  1.41it/s]
1|5840|Loss: 0.928680419921875:  90%|█████████ | 5839/6470 [1:16:25<07:28,  1.41it/s] 
1|5840|Loss: 0.928680419921875:  90%|█████████ | 5840/6470 [1:16:25<06:50,  1.53it/s]
1|5841|Loss: 0.8697108030319214:  90%|█████████ | 5840/6470 [1:16:26<06:50,  1.53it/s]
1|5841|Loss: 0.8697108030319214:  90%|█████████ | 5841/6470 [1:16:26<06:51,  1.53it/s]
1|5842|Loss: 0.7635952830314636:  90%|█████████ | 5841/6470 [1:16:26<06:51,  1.53it/s]
1|5842|Loss: 0.7635952830314636:  90%|█████████ | 5842/6470 [1:16:27<06:28,  1.62it/s]
1|5843|Loss: 0.9090703129768372:  90%|█████████ | 5842/6470 [1:16:27<06:28,  1.62it/s]
1|5843|Loss: 0.9090703129768372:  90%|█████████ | 5843/6470 [1:16:27<07:04,  1.48it/s]
1|5844|Loss: 0.882337749004364:  90%|█████████ | 5843/6470 [1:16:28<07:04,  1.48it/s] 
1|5844|Loss: 0.882337749004364:  90%|█████████ | 5844/6470 [1:16:28<07:10,  1.46it/s]
1|5845|Loss: 0.9914150238037109:  90%|█████████ | 5844/6470 [1:16:29<07:10,  1.46it/s]
1|5845|Loss: 0.9914150238037109:  90%|█████████ | 5845/6470 [1:16:29<07:24,  1.40it/s]
1|5846|Loss: 0.8783608675003052:  90%|█████████ | 5845/6470 [1:16:29<07:24,  1.40it/s]
1|5846|Loss: 0.8783608675003052:  90%|█████████ | 5846/6470 [1:16:30<07:24,  1.40it/s]
1|5847|Loss: 0.7950462102890015:  90%|█████████ | 5846/6470 [1:16:30<07:24,  1.40it/s]
1|5847|Loss: 0.7950462102890015:  90%|█████████ | 5847/6470 [1:16:30<07:16,  1.43it/s]
1|5848|Loss: 0.9910175204277039:  90%|█████████ | 5847/6470 [1:16:31<07:16,  1.43it/s]
1|5848|Loss: 0.9910175204277039:  90%|█████████ | 5848/6470 [1:16:31<07:28,  1.39it/s]
1|5849|Loss: 0.7337063550949097:  90%|█████████ | 5848/6470 [1:16:31<07:28,  1.39it/s]
1|5849|Loss: 0.7337063550949097:  90%|█████████ | 5849/6470 [1:16:32<07:34,  1.37it/s]
1|5850|Loss: 0.7973525524139404:  90%|█████████ | 5849/6470 [1:16:32<07:34,  1.37it/s]
1|5850|Loss: 0.7973525524139404:  90%|█████████ | 5850/6470 [1:16:32<07:13,  1.43it/s]
1|5851|Loss: 0.8256067633628845:  90%|█████████ | 5850/6470 [1:16:33<07:13,  1.43it/s]
1|5851|Loss: 0.8256067633628845:  90%|█████████ | 5851/6470 [1:16:33<07:21,  1.40it/s]
1|5852|Loss: 0.8879972100257874:  90%|█████████ | 5851/6470 [1:16:34<07:21,  1.40it/s]
1|5852|Loss: 0.8879972100257874:  90%|█████████ | 5852/6470 [1:16:34<07:18,  1.41it/s]
1|5853|Loss: 0.7635014057159424:  90%|█████████ | 5852/6470 [1:16:34<07:18,  1.41it/s]
1|5853|Loss: 0.7635014057159424:  90%|█████████ | 5853/6470 [1:16:35<07:30,  1.37it/s]
1|5854|Loss: 0.8094008564949036:  90%|█████████ | 5853/6470 [1:16:35<07:30,  1.37it/s]
1|5854|Loss: 0.8094008564949036:  90%|█████████ | 5854/6470 [1:16:35<07:12,  1.42it/s]
1|5855|Loss: 0.8634156584739685:  90%|█████████ | 5854/6470 [1:16:36<07:12,  1.42it/s]
1|5855|Loss: 0.8634156584739685:  90%|█████████ | 5855/6470 [1:16:36<06:46,  1.51it/s]
1|5856|Loss: 0.7991099953651428:  90%|█████████ | 5855/6470 [1:16:36<06:46,  1.51it/s]
1|5856|Loss: 0.7991099953651428:  91%|█████████ | 5856/6470 [1:16:37<07:21,  1.39it/s]
1|5857|Loss: 0.8594799041748047:  91%|█████████ | 5856/6470 [1:16:37<07:21,  1.39it/s]
1|5857|Loss: 0.8594799041748047:  91%|█████████ | 5857/6470 [1:16:37<06:55,  1.47it/s]
1|5858|Loss: 0.8289088606834412:  91%|█████████ | 5857/6470 [1:16:38<06:55,  1.47it/s]
1|5858|Loss: 0.8289088606834412:  91%|█████████ | 5858/6470 [1:16:38<06:36,  1.54it/s]
1|5859|Loss: 0.8865272998809814:  91%|█████████ | 5858/6470 [1:16:38<06:36,  1.54it/s]
1|5859|Loss: 0.8865272998809814:  91%|█████████ | 5859/6470 [1:16:39<06:51,  1.48it/s]
1|5860|Loss: 0.8617473840713501:  91%|█████████ | 5859/6470 [1:16:39<06:51,  1.48it/s]
1|5860|Loss: 0.8617473840713501:  91%|█████████ | 5860/6470 [1:16:39<06:56,  1.46it/s]
1|5861|Loss: 0.8504225611686707:  91%|█████████ | 5860/6470 [1:16:40<06:56,  1.46it/s]
1|5861|Loss: 0.8504225611686707:  91%|█████████ | 5861/6470 [1:16:40<07:13,  1.40it/s]
1|5862|Loss: 0.8323831558227539:  91%|█████████ | 5861/6470 [1:16:40<07:13,  1.40it/s]
1|5862|Loss: 0.8323831558227539:  91%|█████████ | 5862/6470 [1:16:41<07:04,  1.43it/s]
1|5863|Loss: 0.8247143030166626:  91%|█████████ | 5862/6470 [1:16:41<07:04,  1.43it/s]
1|5863|Loss: 0.8247143030166626:  91%|█████████ | 5863/6470 [1:16:41<06:54,  1.47it/s]
1|5864|Loss: 0.8126332759857178:  91%|█████████ | 5863/6470 [1:16:42<06:54,  1.47it/s]
1|5864|Loss: 0.8126332759857178:  91%|█████████ | 5864/6470 [1:16:42<06:38,  1.52it/s]
1|5865|Loss: 0.884032666683197:  91%|█████████ | 5864/6470 [1:16:42<06:38,  1.52it/s] 
1|5865|Loss: 0.884032666683197:  91%|█████████ | 5865/6470 [1:16:43<06:55,  1.45it/s]
1|5866|Loss: 0.948577880859375:  91%|█████████ | 5865/6470 [1:16:43<06:55,  1.45it/s]
1|5866|Loss: 0.948577880859375:  91%|█████████ | 5866/6470 [1:16:44<07:25,  1.36it/s]
1|5867|Loss: 0.8714430928230286:  91%|█████████ | 5866/6470 [1:16:44<07:25,  1.36it/s]
1|5867|Loss: 0.8714430928230286:  91%|█████████ | 5867/6470 [1:16:44<07:36,  1.32it/s]
1|5868|Loss: 0.8756483793258667:  91%|█████████ | 5867/6470 [1:16:45<07:36,  1.32it/s]
1|5868|Loss: 0.8756483793258667:  91%|█████████ | 5868/6470 [1:16:45<07:06,  1.41it/s]
1|5869|Loss: 1.0228735208511353:  91%|█████████ | 5868/6470 [1:16:45<07:06,  1.41it/s]
1|5869|Loss: 1.0228735208511353:  91%|█████████ | 5869/6470 [1:16:46<07:10,  1.39it/s]
1|5870|Loss: 0.7207528948783875:  91%|█████████ | 5869/6470 [1:16:46<07:10,  1.39it/s]
1|5870|Loss: 0.7207528948783875:  91%|█████████ | 5870/6470 [1:16:47<07:42,  1.30it/s]
1|5871|Loss: 0.7626439929008484:  91%|█████████ | 5870/6470 [1:16:47<07:42,  1.30it/s]
1|5871|Loss: 0.7626439929008484:  91%|█████████ | 5871/6470 [1:16:47<07:53,  1.27it/s]
1|5872|Loss: 0.8980310559272766:  91%|█████████ | 5871/6470 [1:16:48<07:53,  1.27it/s]
1|5872|Loss: 0.8980310559272766:  91%|█████████ | 5872/6470 [1:16:48<07:35,  1.31it/s]
1|5873|Loss: 0.6518862247467041:  91%|█████████ | 5872/6470 [1:16:49<07:35,  1.31it/s]
1|5873|Loss: 0.6518862247467041:  91%|█████████ | 5873/6470 [1:16:49<07:22,  1.35it/s]
1|5874|Loss: 0.7864282727241516:  91%|█████████ | 5873/6470 [1:16:49<07:22,  1.35it/s]
1|5874|Loss: 0.7864282727241516:  91%|█████████ | 5874/6470 [1:16:49<06:25,  1.55it/s]
1|5875|Loss: 0.7771909832954407:  91%|█████████ | 5874/6470 [1:16:50<06:25,  1.55it/s]
1|5875|Loss: 0.7771909832954407:  91%|█████████ | 5875/6470 [1:16:50<05:49,  1.70it/s]
1|5876|Loss: 1.0123844146728516:  91%|█████████ | 5875/6470 [1:16:50<05:49,  1.70it/s]
1|5876|Loss: 1.0123844146728516:  91%|█████████ | 5876/6470 [1:16:50<05:42,  1.73it/s]
1|5877|Loss: 0.7619529366493225:  91%|█████████ | 5876/6470 [1:16:51<05:42,  1.73it/s]
1|5877|Loss: 0.7619529366493225:  91%|█████████ | 5877/6470 [1:16:51<06:28,  1.53it/s]
1|5878|Loss: 1.0063241720199585:  91%|█████████ | 5877/6470 [1:16:52<06:28,  1.53it/s]
1|5878|Loss: 1.0063241720199585:  91%|█████████ | 5878/6470 [1:16:52<07:11,  1.37it/s]
1|5879|Loss: 0.8384533524513245:  91%|█████████ | 5878/6470 [1:16:53<07:11,  1.37it/s]
1|5879|Loss: 0.8384533524513245:  91%|█████████ | 5879/6470 [1:16:53<07:52,  1.25it/s]
1|5880|Loss: 0.892590343952179:  91%|█████████ | 5879/6470 [1:16:53<07:52,  1.25it/s] 
1|5880|Loss: 0.892590343952179:  91%|█████████ | 5880/6470 [1:16:54<07:46,  1.26it/s]
1|5881|Loss: 0.8918951749801636:  91%|█████████ | 5880/6470 [1:16:54<07:46,  1.26it/s]
1|5881|Loss: 0.8918951749801636:  91%|█████████ | 5881/6470 [1:16:54<07:19,  1.34it/s]
1|5882|Loss: 0.9898390173912048:  91%|█████████ | 5881/6470 [1:16:55<07:19,  1.34it/s]
1|5882|Loss: 0.9898390173912048:  91%|█████████ | 5882/6470 [1:16:55<07:13,  1.36it/s]
1|5883|Loss: 0.9496036767959595:  91%|█████████ | 5882/6470 [1:16:56<07:13,  1.36it/s]
1|5883|Loss: 0.9496036767959595:  91%|█████████ | 5883/6470 [1:16:56<06:56,  1.41it/s]
1|5884|Loss: 0.7560629844665527:  91%|█████████ | 5883/6470 [1:16:56<06:56,  1.41it/s]
1|5884|Loss: 0.7560629844665527:  91%|█████████ | 5884/6470 [1:16:57<07:13,  1.35it/s]
1|5885|Loss: 0.8446776270866394:  91%|█████████ | 5884/6470 [1:16:57<07:13,  1.35it/s]
1|5885|Loss: 0.8446776270866394:  91%|█████████ | 5885/6470 [1:16:57<07:11,  1.36it/s]
1|5886|Loss: 0.8729792833328247:  91%|█████████ | 5885/6470 [1:16:58<07:11,  1.36it/s]
1|5886|Loss: 0.8729792833328247:  91%|█████████ | 5886/6470 [1:16:58<06:53,  1.41it/s]
1|5887|Loss: 0.788726270198822:  91%|█████████ | 5886/6470 [1:16:58<06:53,  1.41it/s] 
1|5887|Loss: 0.788726270198822:  91%|█████████ | 5887/6470 [1:16:59<06:56,  1.40it/s]
1|5888|Loss: 0.8962967991828918:  91%|█████████ | 5887/6470 [1:16:59<06:56,  1.40it/s]
1|5888|Loss: 0.8962967991828918:  91%|█████████ | 5888/6470 [1:16:59<06:49,  1.42it/s]
1|5889|Loss: 0.7444254159927368:  91%|█████████ | 5888/6470 [1:17:00<06:49,  1.42it/s]
1|5889|Loss: 0.7444254159927368:  91%|█████████ | 5889/6470 [1:17:00<06:39,  1.45it/s]
1|5890|Loss: 0.928393542766571:  91%|█████████ | 5889/6470 [1:17:00<06:39,  1.45it/s] 
1|5890|Loss: 0.928393542766571:  91%|█████████ | 5890/6470 [1:17:01<06:09,  1.57it/s]
1|5891|Loss: 0.8899985551834106:  91%|█████████ | 5890/6470 [1:17:01<06:09,  1.57it/s]
1|5891|Loss: 0.8899985551834106:  91%|█████████ | 5891/6470 [1:17:01<06:23,  1.51it/s]
1|5892|Loss: 0.858139157295227:  91%|█████████ | 5891/6470 [1:17:02<06:23,  1.51it/s] 
1|5892|Loss: 0.858139157295227:  91%|█████████ | 5892/6470 [1:17:02<06:46,  1.42it/s]
1|5893|Loss: 0.86008620262146:  91%|█████████ | 5892/6470 [1:17:03<06:46,  1.42it/s] 
1|5893|Loss: 0.86008620262146:  91%|█████████ | 5893/6470 [1:17:03<07:05,  1.36it/s]
1|5894|Loss: 0.9549915790557861:  91%|█████████ | 5893/6470 [1:17:03<07:05,  1.36it/s]
1|5894|Loss: 0.9549915790557861:  91%|█████████ | 5894/6470 [1:17:04<06:52,  1.40it/s]
1|5895|Loss: 0.9497367739677429:  91%|█████████ | 5894/6470 [1:17:04<06:52,  1.40it/s]
1|5895|Loss: 0.9497367739677429:  91%|█████████ | 5895/6470 [1:17:04<07:08,  1.34it/s]
1|5896|Loss: 0.7974372506141663:  91%|█████████ | 5895/6470 [1:17:05<07:08,  1.34it/s]
1|5896|Loss: 0.7974372506141663:  91%|█████████ | 5896/6470 [1:17:05<06:35,  1.45it/s]
1|5897|Loss: 0.907787561416626:  91%|█████████ | 5896/6470 [1:17:05<06:35,  1.45it/s] 
1|5897|Loss: 0.907787561416626:  91%|█████████ | 5897/6470 [1:17:06<07:03,  1.35it/s]
1|5898|Loss: 0.7081711888313293:  91%|█████████ | 5897/6470 [1:17:06<07:03,  1.35it/s]
1|5898|Loss: 0.7081711888313293:  91%|█████████ | 5898/6470 [1:17:07<07:21,  1.29it/s]
1|5899|Loss: 0.8930748701095581:  91%|█████████ | 5898/6470 [1:17:07<07:21,  1.29it/s]
1|5899|Loss: 0.8930748701095581:  91%|█████████ | 5899/6470 [1:17:07<07:23,  1.29it/s]
1|5900|Loss: 0.7284416556358337:  91%|█████████ | 5899/6470 [1:17:08<07:23,  1.29it/s]
1|5900|Loss: 0.7284416556358337:  91%|█████████ | 5900/6470 [1:17:08<07:12,  1.32it/s]
1|5901|Loss: 0.7699201703071594:  91%|█████████ | 5900/6470 [1:17:08<07:12,  1.32it/s]
1|5901|Loss: 0.7699201703071594:  91%|█████████ | 5901/6470 [1:17:09<06:30,  1.46it/s]
1|5902|Loss: 1.0003859996795654:  91%|█████████ | 5901/6470 [1:17:09<06:30,  1.46it/s]
1|5902|Loss: 1.0003859996795654:  91%|█████████ | 5902/6470 [1:17:09<06:15,  1.51it/s]
1|5903|Loss: 0.7943329215049744:  91%|█████████ | 5902/6470 [1:17:10<06:15,  1.51it/s]
1|5903|Loss: 0.7943329215049744:  91%|█████████ | 5903/6470 [1:17:10<06:29,  1.45it/s]
1|5904|Loss: 0.8545525670051575:  91%|█████████ | 5903/6470 [1:17:10<06:29,  1.45it/s]
1|5904|Loss: 0.8545525670051575:  91%|█████████▏| 5904/6470 [1:17:11<06:15,  1.51it/s]
1|5905|Loss: 0.8295934200286865:  91%|█████████▏| 5904/6470 [1:17:11<06:15,  1.51it/s]
1|5905|Loss: 0.8295934200286865:  91%|█████████▏| 5905/6470 [1:17:11<06:15,  1.50it/s]
1|5906|Loss: 1.0700448751449585:  91%|█████████▏| 5905/6470 [1:17:12<06:15,  1.50it/s]
1|5906|Loss: 1.0700448751449585:  91%|█████████▏| 5906/6470 [1:17:12<06:00,  1.57it/s]
1|5907|Loss: 0.7597384452819824:  91%|█████████▏| 5906/6470 [1:17:12<06:00,  1.57it/s]
1|5907|Loss: 0.7597384452819824:  91%|█████████▏| 5907/6470 [1:17:13<06:10,  1.52it/s]
1|5908|Loss: 0.9741597175598145:  91%|█████████▏| 5907/6470 [1:17:13<06:10,  1.52it/s]
1|5908|Loss: 0.9741597175598145:  91%|█████████▏| 5908/6470 [1:17:13<06:15,  1.50it/s]
1|5909|Loss: 0.8460286259651184:  91%|█████████▏| 5908/6470 [1:17:14<06:15,  1.50it/s]
1|5909|Loss: 0.8460286259651184:  91%|█████████▏| 5909/6470 [1:17:14<06:15,  1.50it/s]
1|5910|Loss: 0.8262393474578857:  91%|█████████▏| 5909/6470 [1:17:14<06:15,  1.50it/s]
1|5910|Loss: 0.8262393474578857:  91%|█████████▏| 5910/6470 [1:17:15<06:31,  1.43it/s]
1|5911|Loss: 0.8326653242111206:  91%|█████████▏| 5910/6470 [1:17:15<06:31,  1.43it/s]
1|5911|Loss: 0.8326653242111206:  91%|█████████▏| 5911/6470 [1:17:15<06:29,  1.44it/s]
1|5912|Loss: 0.966110348701477:  91%|█████████▏| 5911/6470 [1:17:16<06:29,  1.44it/s] 
1|5912|Loss: 0.966110348701477:  91%|█████████▏| 5912/6470 [1:17:16<06:21,  1.46it/s]
1|5913|Loss: 0.7355669140815735:  91%|█████████▏| 5912/6470 [1:17:16<06:21,  1.46it/s]
1|5913|Loss: 0.7355669140815735:  91%|█████████▏| 5913/6470 [1:17:17<06:44,  1.38it/s]
1|5914|Loss: 0.9710442423820496:  91%|█████████▏| 5913/6470 [1:17:17<06:44,  1.38it/s]
1|5914|Loss: 0.9710442423820496:  91%|█████████▏| 5914/6470 [1:17:17<06:21,  1.46it/s]
1|5915|Loss: 0.7673066854476929:  91%|█████████▏| 5914/6470 [1:17:18<06:21,  1.46it/s]
1|5915|Loss: 0.7673066854476929:  91%|█████████▏| 5915/6470 [1:17:18<06:49,  1.36it/s]
1|5916|Loss: 0.8647335767745972:  91%|█████████▏| 5915/6470 [1:17:19<06:49,  1.36it/s]
1|5916|Loss: 0.8647335767745972:  91%|█████████▏| 5916/6470 [1:17:19<07:11,  1.28it/s]
1|5917|Loss: 0.9371808171272278:  91%|█████████▏| 5916/6470 [1:17:20<07:11,  1.28it/s]
1|5917|Loss: 0.9371808171272278:  91%|█████████▏| 5917/6470 [1:17:20<07:15,  1.27it/s]
1|5918|Loss: 0.9847491979598999:  91%|█████████▏| 5917/6470 [1:17:20<07:15,  1.27it/s]
1|5918|Loss: 0.9847491979598999:  91%|█████████▏| 5918/6470 [1:17:21<06:52,  1.34it/s]
1|5919|Loss: 0.872088611125946:  91%|█████████▏| 5918/6470 [1:17:21<06:52,  1.34it/s] 
1|5919|Loss: 0.872088611125946:  91%|█████████▏| 5919/6470 [1:17:22<07:11,  1.28it/s]
1|5920|Loss: 0.9206966757774353:  91%|█████████▏| 5919/6470 [1:17:22<07:11,  1.28it/s]
1|5920|Loss: 0.9206966757774353:  91%|█████████▏| 5920/6470 [1:17:22<06:54,  1.33it/s]
1|5921|Loss: 0.9450567960739136:  91%|█████████▏| 5920/6470 [1:17:23<06:54,  1.33it/s]
1|5921|Loss: 0.9450567960739136:  92%|█████████▏| 5921/6470 [1:17:23<06:53,  1.33it/s]
1|5922|Loss: 0.9397026300430298:  92%|█████████▏| 5921/6470 [1:17:23<06:53,  1.33it/s]
1|5922|Loss: 0.9397026300430298:  92%|█████████▏| 5922/6470 [1:17:24<06:34,  1.39it/s]
1|5923|Loss: 0.8762537837028503:  92%|█████████▏| 5922/6470 [1:17:24<06:34,  1.39it/s]
1|5923|Loss: 0.8762537837028503:  92%|█████████▏| 5923/6470 [1:17:24<06:40,  1.37it/s]
1|5924|Loss: 0.8973625302314758:  92%|█████████▏| 5923/6470 [1:17:25<06:40,  1.37it/s]
1|5924|Loss: 0.8973625302314758:  92%|█████████▏| 5924/6470 [1:17:25<07:13,  1.26it/s]
1|5925|Loss: 0.9077414870262146:  92%|█████████▏| 5924/6470 [1:17:26<07:13,  1.26it/s]
1|5925|Loss: 0.9077414870262146:  92%|█████████▏| 5925/6470 [1:17:26<07:01,  1.29it/s]
1|5926|Loss: 0.6962621212005615:  92%|█████████▏| 5925/6470 [1:17:26<07:01,  1.29it/s]
1|5926|Loss: 0.6962621212005615:  92%|█████████▏| 5926/6470 [1:17:26<06:06,  1.49it/s]
1|5927|Loss: 0.760435938835144:  92%|█████████▏| 5926/6470 [1:17:27<06:06,  1.49it/s] 
1|5927|Loss: 0.760435938835144:  92%|█████████▏| 5927/6470 [1:17:27<06:30,  1.39it/s]
1|5928|Loss: 0.9430351257324219:  92%|█████████▏| 5927/6470 [1:17:28<06:30,  1.39it/s]
1|5928|Loss: 0.9430351257324219:  92%|█████████▏| 5928/6470 [1:17:28<06:24,  1.41it/s]
1|5929|Loss: 0.854793131351471:  92%|█████████▏| 5928/6470 [1:17:28<06:24,  1.41it/s] 
1|5929|Loss: 0.854793131351471:  92%|█████████▏| 5929/6470 [1:17:29<06:08,  1.47it/s]
1|5930|Loss: 0.7320075631141663:  92%|█████████▏| 5929/6470 [1:17:29<06:08,  1.47it/s]
1|5930|Loss: 0.7320075631141663:  92%|█████████▏| 5930/6470 [1:17:29<06:16,  1.44it/s]
1|5931|Loss: 0.9455121755599976:  92%|█████████▏| 5930/6470 [1:17:30<06:16,  1.44it/s]
1|5931|Loss: 0.9455121755599976:  92%|█████████▏| 5931/6470 [1:17:30<06:25,  1.40it/s]
1|5932|Loss: 0.7373564839363098:  92%|█████████▏| 5931/6470 [1:17:30<06:25,  1.40it/s]
1|5932|Loss: 0.7373564839363098:  92%|█████████▏| 5932/6470 [1:17:31<06:24,  1.40it/s]
1|5933|Loss: 0.9524312019348145:  92%|█████████▏| 5932/6470 [1:17:31<06:24,  1.40it/s]
1|5933|Loss: 0.9524312019348145:  92%|█████████▏| 5933/6470 [1:17:32<06:33,  1.37it/s]
1|5934|Loss: 0.846403181552887:  92%|█████████▏| 5933/6470 [1:17:32<06:33,  1.37it/s] 
1|5934|Loss: 0.846403181552887:  92%|█████████▏| 5934/6470 [1:17:32<06:52,  1.30it/s]
1|5935|Loss: 0.8713288903236389:  92%|█████████▏| 5934/6470 [1:17:33<06:52,  1.30it/s]
1|5935|Loss: 0.8713288903236389:  92%|█████████▏| 5935/6470 [1:17:33<07:07,  1.25it/s]
1|5936|Loss: 0.9271063208580017:  92%|█████████▏| 5935/6470 [1:17:34<07:07,  1.25it/s]
1|5936|Loss: 0.9271063208580017:  92%|█████████▏| 5936/6470 [1:17:34<07:25,  1.20it/s]
1|5937|Loss: 0.8906912803649902:  92%|█████████▏| 5936/6470 [1:17:35<07:25,  1.20it/s]
1|5937|Loss: 0.8906912803649902:  92%|█████████▏| 5937/6470 [1:17:35<07:10,  1.24it/s]
1|5938|Loss: 0.7400718927383423:  92%|█████████▏| 5937/6470 [1:17:35<07:10,  1.24it/s]
1|5938|Loss: 0.7400718927383423:  92%|█████████▏| 5938/6470 [1:17:36<06:59,  1.27it/s]
1|5939|Loss: 0.7735608220100403:  92%|█████████▏| 5938/6470 [1:17:36<06:59,  1.27it/s]
1|5939|Loss: 0.7735608220100403:  92%|█████████▏| 5939/6470 [1:17:36<06:55,  1.28it/s]
1|5940|Loss: 0.9064026474952698:  92%|█████████▏| 5939/6470 [1:17:37<06:55,  1.28it/s]
1|5940|Loss: 0.9064026474952698:  92%|█████████▏| 5940/6470 [1:17:37<06:25,  1.38it/s]
1|5941|Loss: 0.7515851855278015:  92%|█████████▏| 5940/6470 [1:17:37<06:25,  1.38it/s]
1|5941|Loss: 0.7515851855278015:  92%|█████████▏| 5941/6470 [1:17:38<06:15,  1.41it/s]
1|5942|Loss: 0.7840641140937805:  92%|█████████▏| 5941/6470 [1:17:38<06:15,  1.41it/s]
1|5942|Loss: 0.7840641140937805:  92%|█████████▏| 5942/6470 [1:17:39<06:36,  1.33it/s]
1|5943|Loss: 0.8919757008552551:  92%|█████████▏| 5942/6470 [1:17:39<06:36,  1.33it/s]
1|5943|Loss: 0.8919757008552551:  92%|█████████▏| 5943/6470 [1:17:39<06:45,  1.30it/s]
1|5944|Loss: 0.8765877485275269:  92%|█████████▏| 5943/6470 [1:17:40<06:45,  1.30it/s]
1|5944|Loss: 0.8765877485275269:  92%|█████████▏| 5944/6470 [1:17:40<06:49,  1.29it/s]
1|5945|Loss: 0.8107249736785889:  92%|█████████▏| 5944/6470 [1:17:41<06:49,  1.29it/s]
1|5945|Loss: 0.8107249736785889:  92%|█████████▏| 5945/6470 [1:17:41<06:38,  1.32it/s]
1|5946|Loss: 0.7989747524261475:  92%|█████████▏| 5945/6470 [1:17:41<06:38,  1.32it/s]
1|5946|Loss: 0.7989747524261475:  92%|█████████▏| 5946/6470 [1:17:42<06:19,  1.38it/s]
1|5947|Loss: 0.9742769598960876:  92%|█████████▏| 5946/6470 [1:17:42<06:19,  1.38it/s]
1|5947|Loss: 0.9742769598960876:  92%|█████████▏| 5947/6470 [1:17:42<06:09,  1.42it/s]
1|5948|Loss: 0.8633131384849548:  92%|█████████▏| 5947/6470 [1:17:43<06:09,  1.42it/s]
1|5948|Loss: 0.8633131384849548:  92%|█████████▏| 5948/6470 [1:17:43<05:48,  1.50it/s]
1|5949|Loss: 0.8490304350852966:  92%|█████████▏| 5948/6470 [1:17:43<05:48,  1.50it/s]
1|5949|Loss: 0.8490304350852966:  92%|█████████▏| 5949/6470 [1:17:43<05:43,  1.52it/s]
1|5950|Loss: 0.9566829204559326:  92%|█████████▏| 5949/6470 [1:17:44<05:43,  1.52it/s]
1|5950|Loss: 0.9566829204559326:  92%|█████████▏| 5950/6470 [1:17:44<05:59,  1.44it/s]
1|5951|Loss: 0.7573673129081726:  92%|█████████▏| 5950/6470 [1:17:45<05:59,  1.44it/s]
1|5951|Loss: 0.7573673129081726:  92%|█████████▏| 5951/6470 [1:17:45<06:02,  1.43it/s]
1|5952|Loss: 0.9188523292541504:  92%|█████████▏| 5951/6470 [1:17:45<06:02,  1.43it/s]
1|5952|Loss: 0.9188523292541504:  92%|█████████▏| 5952/6470 [1:17:45<05:46,  1.49it/s]
1|5953|Loss: 0.9414405226707458:  92%|█████████▏| 5952/6470 [1:17:46<05:46,  1.49it/s]
1|5953|Loss: 0.9414405226707458:  92%|█████████▏| 5953/6470 [1:17:46<05:59,  1.44it/s]
1|5954|Loss: 0.8955214023590088:  92%|█████████▏| 5953/6470 [1:17:47<05:59,  1.44it/s]
1|5954|Loss: 0.8955214023590088:  92%|█████████▏| 5954/6470 [1:17:47<06:02,  1.42it/s]
1|5955|Loss: 0.8639774918556213:  92%|█████████▏| 5954/6470 [1:17:47<06:02,  1.42it/s]
1|5955|Loss: 0.8639774918556213:  92%|█████████▏| 5955/6470 [1:17:48<06:14,  1.38it/s]
1|5956|Loss: 1.0528326034545898:  92%|█████████▏| 5955/6470 [1:17:48<06:14,  1.38it/s]
1|5956|Loss: 1.0528326034545898:  92%|█████████▏| 5956/6470 [1:17:49<06:32,  1.31it/s]
1|5957|Loss: 0.852365255355835:  92%|█████████▏| 5956/6470 [1:17:49<06:32,  1.31it/s] 
1|5957|Loss: 0.852365255355835:  92%|█████████▏| 5957/6470 [1:17:49<06:50,  1.25it/s]
1|5958|Loss: 0.810962438583374:  92%|█████████▏| 5957/6470 [1:17:50<06:50,  1.25it/s]
1|5958|Loss: 0.810962438583374:  92%|█████████▏| 5958/6470 [1:17:50<06:27,  1.32it/s]
1|5959|Loss: 0.9288948774337769:  92%|█████████▏| 5958/6470 [1:17:50<06:27,  1.32it/s]
1|5959|Loss: 0.9288948774337769:  92%|█████████▏| 5959/6470 [1:17:51<06:09,  1.38it/s]
1|5960|Loss: 0.8795284032821655:  92%|█████████▏| 5959/6470 [1:17:51<06:09,  1.38it/s]
1|5960|Loss: 0.8795284032821655:  92%|█████████▏| 5960/6470 [1:17:52<06:11,  1.37it/s]
1|5961|Loss: 0.7805283069610596:  92%|█████████▏| 5960/6470 [1:17:52<06:11,  1.37it/s]
1|5961|Loss: 0.7805283069610596:  92%|█████████▏| 5961/6470 [1:17:52<06:33,  1.29it/s]
1|5962|Loss: 0.8230545520782471:  92%|█████████▏| 5961/6470 [1:17:53<06:33,  1.29it/s]
1|5962|Loss: 0.8230545520782471:  92%|█████████▏| 5962/6470 [1:17:53<06:10,  1.37it/s]
1|5963|Loss: 0.8797998428344727:  92%|█████████▏| 5962/6470 [1:17:53<06:10,  1.37it/s]
1|5963|Loss: 0.8797998428344727:  92%|█████████▏| 5963/6470 [1:17:54<06:09,  1.37it/s]
1|5964|Loss: 0.7791150808334351:  92%|█████████▏| 5963/6470 [1:17:54<06:09,  1.37it/s]
1|5964|Loss: 0.7791150808334351:  92%|█████████▏| 5964/6470 [1:17:55<06:19,  1.33it/s]
1|5965|Loss: 0.9130702614784241:  92%|█████████▏| 5964/6470 [1:17:55<06:19,  1.33it/s]
1|5965|Loss: 0.9130702614784241:  92%|█████████▏| 5965/6470 [1:17:55<06:22,  1.32it/s]
1|5966|Loss: 1.031459927558899:  92%|█████████▏| 5965/6470 [1:17:56<06:22,  1.32it/s] 
1|5966|Loss: 1.031459927558899:  92%|█████████▏| 5966/6470 [1:17:56<06:10,  1.36it/s]
1|5967|Loss: 0.792508065700531:  92%|█████████▏| 5966/6470 [1:17:56<06:10,  1.36it/s]
1|5967|Loss: 0.792508065700531:  92%|█████████▏| 5967/6470 [1:17:57<06:36,  1.27it/s]
1|5968|Loss: 0.8916041254997253:  92%|█████████▏| 5967/6470 [1:17:57<06:36,  1.27it/s]
1|5968|Loss: 0.8916041254997253:  92%|█████████▏| 5968/6470 [1:17:58<06:44,  1.24it/s]
1|5969|Loss: 0.9405539631843567:  92%|█████████▏| 5968/6470 [1:17:58<06:44,  1.24it/s]
1|5969|Loss: 0.9405539631843567:  92%|█████████▏| 5969/6470 [1:17:58<06:31,  1.28it/s]
1|5970|Loss: 0.8762734532356262:  92%|█████████▏| 5969/6470 [1:17:59<06:31,  1.28it/s]
1|5970|Loss: 0.8762734532356262:  92%|█████████▏| 5970/6470 [1:17:59<06:30,  1.28it/s]
1|5971|Loss: 0.9691317081451416:  92%|█████████▏| 5970/6470 [1:18:00<06:30,  1.28it/s]
1|5971|Loss: 0.9691317081451416:  92%|█████████▏| 5971/6470 [1:18:00<06:27,  1.29it/s]
1|5972|Loss: 0.8906785249710083:  92%|█████████▏| 5971/6470 [1:18:00<06:27,  1.29it/s]
1|5972|Loss: 0.8906785249710083:  92%|█████████▏| 5972/6470 [1:18:01<06:22,  1.30it/s]
1|5973|Loss: 0.8654335737228394:  92%|█████████▏| 5972/6470 [1:18:01<06:22,  1.30it/s]
1|5973|Loss: 0.8654335737228394:  92%|█████████▏| 5973/6470 [1:18:02<06:13,  1.33it/s]
1|5974|Loss: 0.7694661021232605:  92%|█████████▏| 5973/6470 [1:18:02<06:13,  1.33it/s]
1|5974|Loss: 0.7694661021232605:  92%|█████████▏| 5974/6470 [1:18:02<06:04,  1.36it/s]
1|5975|Loss: 0.8152377605438232:  92%|█████████▏| 5974/6470 [1:18:03<06:04,  1.36it/s]
1|5975|Loss: 0.8152377605438232:  92%|█████████▏| 5975/6470 [1:18:03<05:41,  1.45it/s]
1|5976|Loss: 0.8966571092605591:  92%|█████████▏| 5975/6470 [1:18:03<05:41,  1.45it/s]
1|5976|Loss: 0.8966571092605591:  92%|█████████▏| 5976/6470 [1:18:04<05:46,  1.43it/s]
1|5977|Loss: 1.0540869235992432:  92%|█████████▏| 5976/6470 [1:18:04<05:46,  1.43it/s]
1|5977|Loss: 1.0540869235992432:  92%|█████████▏| 5977/6470 [1:18:04<05:57,  1.38it/s]
1|5978|Loss: 1.0039118528366089:  92%|█████████▏| 5977/6470 [1:18:05<05:57,  1.38it/s]
1|5978|Loss: 1.0039118528366089:  92%|█████████▏| 5978/6470 [1:18:05<05:55,  1.38it/s]
1|5979|Loss: 0.8830751776695251:  92%|█████████▏| 5978/6470 [1:18:05<05:55,  1.38it/s]
1|5979|Loss: 0.8830751776695251:  92%|█████████▏| 5979/6470 [1:18:06<06:13,  1.31it/s]
1|5980|Loss: 0.8479746580123901:  92%|█████████▏| 5979/6470 [1:18:06<06:13,  1.31it/s]
1|5980|Loss: 0.8479746580123901:  92%|█████████▏| 5980/6470 [1:18:07<06:04,  1.34it/s]
1|5981|Loss: 0.8827539086341858:  92%|█████████▏| 5980/6470 [1:18:07<06:04,  1.34it/s]
1|5981|Loss: 0.8827539086341858:  92%|█████████▏| 5981/6470 [1:18:07<05:57,  1.37it/s]
1|5982|Loss: 0.83082515001297:  92%|█████████▏| 5981/6470 [1:18:08<05:57,  1.37it/s]  
1|5982|Loss: 0.83082515001297:  92%|█████████▏| 5982/6470 [1:18:08<05:35,  1.45it/s]
1|5983|Loss: 0.9517503976821899:  92%|█████████▏| 5982/6470 [1:18:08<05:35,  1.45it/s]
1|5983|Loss: 0.9517503976821899:  92%|█████████▏| 5983/6470 [1:18:09<05:45,  1.41it/s]
1|5984|Loss: 0.8000304102897644:  92%|█████████▏| 5983/6470 [1:18:09<05:45,  1.41it/s]
1|5984|Loss: 0.8000304102897644:  92%|█████████▏| 5984/6470 [1:18:09<06:04,  1.33it/s]
1|5985|Loss: 0.9585669636726379:  92%|█████████▏| 5984/6470 [1:18:10<06:04,  1.33it/s]
1|5985|Loss: 0.9585669636726379:  93%|█████████▎| 5985/6470 [1:18:10<05:41,  1.42it/s]
1|5986|Loss: 0.9490153193473816:  93%|█████████▎| 5985/6470 [1:18:10<05:41,  1.42it/s]
1|5986|Loss: 0.9490153193473816:  93%|█████████▎| 5986/6470 [1:18:11<05:09,  1.57it/s]
1|5987|Loss: 0.8506717681884766:  93%|█████████▎| 5986/6470 [1:18:11<05:09,  1.57it/s]
1|5987|Loss: 0.8506717681884766:  93%|█████████▎| 5987/6470 [1:18:11<05:19,  1.51it/s]
1|5988|Loss: 0.9744048118591309:  93%|█████████▎| 5987/6470 [1:18:12<05:19,  1.51it/s]
1|5988|Loss: 0.9744048118591309:  93%|█████████▎| 5988/6470 [1:18:12<05:24,  1.49it/s]
1|5989|Loss: 0.7431555390357971:  93%|█████████▎| 5988/6470 [1:18:12<05:24,  1.49it/s]
1|5989|Loss: 0.7431555390357971:  93%|█████████▎| 5989/6470 [1:18:13<06:08,  1.31it/s]
1|5990|Loss: 0.9657328724861145:  93%|█████████▎| 5989/6470 [1:18:13<06:08,  1.31it/s]
1|5990|Loss: 0.9657328724861145:  93%|█████████▎| 5990/6470 [1:18:14<05:59,  1.33it/s]
1|5991|Loss: 0.9323961138725281:  93%|█████████▎| 5990/6470 [1:18:14<05:59,  1.33it/s]
1|5991|Loss: 0.9323961138725281:  93%|█████████▎| 5991/6470 [1:18:15<06:19,  1.26it/s]
1|5992|Loss: 0.9020738005638123:  93%|█████████▎| 5991/6470 [1:18:15<06:19,  1.26it/s]
1|5992|Loss: 0.9020738005638123:  93%|█████████▎| 5992/6470 [1:18:15<06:22,  1.25it/s]
1|5993|Loss: 0.9308141469955444:  93%|█████████▎| 5992/6470 [1:18:16<06:22,  1.25it/s]
1|5993|Loss: 0.9308141469955444:  93%|█████████▎| 5993/6470 [1:18:16<06:32,  1.21it/s]
1|5994|Loss: 0.7430205345153809:  93%|█████████▎| 5993/6470 [1:18:17<06:32,  1.21it/s]
1|5994|Loss: 0.7430205345153809:  93%|█████████▎| 5994/6470 [1:18:17<05:43,  1.39it/s]
1|5995|Loss: 0.8974160552024841:  93%|█████████▎| 5994/6470 [1:18:17<05:43,  1.39it/s]
1|5995|Loss: 0.8974160552024841:  93%|█████████▎| 5995/6470 [1:18:17<05:36,  1.41it/s]
1|5996|Loss: 0.9298938512802124:  93%|█████████▎| 5995/6470 [1:18:18<05:36,  1.41it/s]
1|5996|Loss: 0.9298938512802124:  93%|█████████▎| 5996/6470 [1:18:18<05:37,  1.40it/s]
1|5997|Loss: 0.9507887363433838:  93%|█████████▎| 5996/6470 [1:18:19<05:37,  1.40it/s]
1|5997|Loss: 0.9507887363433838:  93%|█████████▎| 5997/6470 [1:18:19<05:44,  1.37it/s]
1|5998|Loss: 0.8341497182846069:  93%|█████████▎| 5997/6470 [1:18:19<05:44,  1.37it/s]
1|5998|Loss: 0.8341497182846069:  93%|█████████▎| 5998/6470 [1:18:20<06:23,  1.23it/s]
1|5999|Loss: 1.0359838008880615:  93%|█████████▎| 5998/6470 [1:18:20<06:23,  1.23it/s]
1|5999|Loss: 1.0359838008880615:  93%|█████████▎| 5999/6470 [1:18:21<06:46,  1.16it/s]
1|6000|Loss: 0.786693274974823:  93%|█████████▎| 5999/6470 [1:18:21<06:46,  1.16it/s] 
1|6000|Loss: 0.786693274974823:  93%|█████████▎| 6000/6470 [1:18:22<06:39,  1.18it/s]
1|6001|Loss: 0.7778567671775818:  93%|█████████▎| 6000/6470 [1:18:22<06:39,  1.18it/s]
1|6001|Loss: 0.7778567671775818:  93%|█████████▎| 6001/6470 [1:18:22<06:17,  1.24it/s]
1|6002|Loss: 0.8815389275550842:  93%|█████████▎| 6001/6470 [1:18:23<06:17,  1.24it/s]
1|6002|Loss: 0.8815389275550842:  93%|█████████▎| 6002/6470 [1:18:23<06:15,  1.24it/s]
1|6003|Loss: 0.8677546381950378:  93%|█████████▎| 6002/6470 [1:18:24<06:15,  1.24it/s]
1|6003|Loss: 0.8677546381950378:  93%|█████████▎| 6003/6470 [1:18:24<06:17,  1.24it/s]
1|6004|Loss: 0.6963716745376587:  93%|█████████▎| 6003/6470 [1:18:24<06:17,  1.24it/s]
1|6004|Loss: 0.6963716745376587:  93%|█████████▎| 6004/6470 [1:18:25<06:15,  1.24it/s]
1|6005|Loss: 0.9299285411834717:  93%|█████████▎| 6004/6470 [1:18:25<06:15,  1.24it/s]
1|6005|Loss: 0.9299285411834717:  93%|█████████▎| 6005/6470 [1:18:25<05:53,  1.31it/s]
1|6006|Loss: 0.9472132325172424:  93%|█████████▎| 6005/6470 [1:18:26<05:53,  1.31it/s]
1|6006|Loss: 0.9472132325172424:  93%|█████████▎| 6006/6470 [1:18:26<05:27,  1.42it/s]
1|6007|Loss: 0.8188056349754333:  93%|█████████▎| 6006/6470 [1:18:26<05:27,  1.42it/s]
1|6007|Loss: 0.8188056349754333:  93%|█████████▎| 6007/6470 [1:18:27<05:18,  1.45it/s]
1|6008|Loss: 0.8733257055282593:  93%|█████████▎| 6007/6470 [1:18:27<05:18,  1.45it/s]
1|6008|Loss: 0.8733257055282593:  93%|█████████▎| 6008/6470 [1:18:27<05:16,  1.46it/s]
1|6009|Loss: 1.0130735635757446:  93%|█████████▎| 6008/6470 [1:18:28<05:16,  1.46it/s]
1|6009|Loss: 1.0130735635757446:  93%|█████████▎| 6009/6470 [1:18:28<05:07,  1.50it/s]
1|6010|Loss: 1.0234709978103638:  93%|█████████▎| 6009/6470 [1:18:28<05:07,  1.50it/s]
1|6010|Loss: 1.0234709978103638:  93%|█████████▎| 6010/6470 [1:18:29<05:09,  1.48it/s]
1|6011|Loss: 0.7577646970748901:  93%|█████████▎| 6010/6470 [1:18:29<05:09,  1.48it/s]
1|6011|Loss: 0.7577646970748901:  93%|█████████▎| 6011/6470 [1:18:29<05:26,  1.41it/s]
1|6012|Loss: 1.0504904985427856:  93%|█████████▎| 6011/6470 [1:18:30<05:26,  1.41it/s]
1|6012|Loss: 1.0504904985427856:  93%|█████████▎| 6012/6470 [1:18:30<05:28,  1.40it/s]
1|6013|Loss: 0.9742423892021179:  93%|█████████▎| 6012/6470 [1:18:31<05:28,  1.40it/s]
1|6013|Loss: 0.9742423892021179:  93%|█████████▎| 6013/6470 [1:18:31<05:26,  1.40it/s]
1|6014|Loss: 0.9380942583084106:  93%|█████████▎| 6013/6470 [1:18:31<05:26,  1.40it/s]
1|6014|Loss: 0.9380942583084106:  93%|█████████▎| 6014/6470 [1:18:32<05:24,  1.41it/s]
1|6015|Loss: 0.9511027336120605:  93%|█████████▎| 6014/6470 [1:18:32<05:24,  1.41it/s]
1|6015|Loss: 0.9511027336120605:  93%|█████████▎| 6015/6470 [1:18:32<05:23,  1.41it/s]
1|6016|Loss: 0.7305280566215515:  93%|█████████▎| 6015/6470 [1:18:33<05:23,  1.41it/s]
1|6016|Loss: 0.7305280566215515:  93%|█████████▎| 6016/6470 [1:18:33<05:44,  1.32it/s]
1|6017|Loss: 0.9216884970664978:  93%|█████████▎| 6016/6470 [1:18:34<05:44,  1.32it/s]
1|6017|Loss: 0.9216884970664978:  93%|█████████▎| 6017/6470 [1:18:34<06:07,  1.23it/s]
1|6018|Loss: 0.9116005301475525:  93%|█████████▎| 6017/6470 [1:18:35<06:07,  1.23it/s]
1|6018|Loss: 0.9116005301475525:  93%|█████████▎| 6018/6470 [1:18:35<06:19,  1.19it/s]
1|6019|Loss: 0.7896071672439575:  93%|█████████▎| 6018/6470 [1:18:35<06:19,  1.19it/s]
1|6019|Loss: 0.7896071672439575:  93%|█████████▎| 6019/6470 [1:18:36<06:10,  1.22it/s]
1|6020|Loss: 0.8696098327636719:  93%|█████████▎| 6019/6470 [1:18:36<06:10,  1.22it/s]
1|6020|Loss: 0.8696098327636719:  93%|█████████▎| 6020/6470 [1:18:37<06:06,  1.23it/s]
1|6021|Loss: 0.8860988020896912:  93%|█████████▎| 6020/6470 [1:18:37<06:06,  1.23it/s]
1|6021|Loss: 0.8860988020896912:  93%|█████████▎| 6021/6470 [1:18:37<05:51,  1.28it/s]
1|6022|Loss: 0.930422842502594:  93%|█████████▎| 6021/6470 [1:18:38<05:51,  1.28it/s] 
1|6022|Loss: 0.930422842502594:  93%|█████████▎| 6022/6470 [1:18:38<06:00,  1.24it/s]
1|6023|Loss: 0.917729914188385:  93%|█████████▎| 6022/6470 [1:18:39<06:00,  1.24it/s]
1|6023|Loss: 0.917729914188385:  93%|█████████▎| 6023/6470 [1:18:39<06:09,  1.21it/s]
1|6024|Loss: 0.8898798823356628:  93%|█████████▎| 6023/6470 [1:18:40<06:09,  1.21it/s]
1|6024|Loss: 0.8898798823356628:  93%|█████████▎| 6024/6470 [1:18:40<06:07,  1.21it/s]
1|6025|Loss: 0.8608548641204834:  93%|█████████▎| 6024/6470 [1:18:40<06:07,  1.21it/s]
1|6025|Loss: 0.8608548641204834:  93%|█████████▎| 6025/6470 [1:18:41<05:49,  1.27it/s]
1|6026|Loss: 0.900919497013092:  93%|█████████▎| 6025/6470 [1:18:41<05:49,  1.27it/s] 
1|6026|Loss: 0.900919497013092:  93%|█████████▎| 6026/6470 [1:18:41<05:42,  1.29it/s]
1|6027|Loss: 0.7114568948745728:  93%|█████████▎| 6026/6470 [1:18:42<05:42,  1.29it/s]
1|6027|Loss: 0.7114568948745728:  93%|█████████▎| 6027/6470 [1:18:42<05:20,  1.38it/s]
1|6028|Loss: 0.7608945965766907:  93%|█████████▎| 6027/6470 [1:18:42<05:20,  1.38it/s]
1|6028|Loss: 0.7608945965766907:  93%|█████████▎| 6028/6470 [1:18:43<05:41,  1.29it/s]
1|6029|Loss: 0.8951324820518494:  93%|█████████▎| 6028/6470 [1:18:43<05:41,  1.29it/s]
1|6029|Loss: 0.8951324820518494:  93%|█████████▎| 6029/6470 [1:18:44<05:40,  1.30it/s]
1|6030|Loss: 0.7992609739303589:  93%|█████████▎| 6029/6470 [1:18:44<05:40,  1.30it/s]
1|6030|Loss: 0.7992609739303589:  93%|█████████▎| 6030/6470 [1:18:44<05:29,  1.33it/s]
1|6031|Loss: 0.9398577809333801:  93%|█████████▎| 6030/6470 [1:18:45<05:29,  1.33it/s]
1|6031|Loss: 0.9398577809333801:  93%|█████████▎| 6031/6470 [1:18:45<05:33,  1.32it/s]
1|6032|Loss: 0.867364227771759:  93%|█████████▎| 6031/6470 [1:18:46<05:33,  1.32it/s] 
1|6032|Loss: 0.867364227771759:  93%|█████████▎| 6032/6470 [1:18:46<05:37,  1.30it/s]
1|6033|Loss: 0.7604028582572937:  93%|█████████▎| 6032/6470 [1:18:46<05:37,  1.30it/s]
1|6033|Loss: 0.7604028582572937:  93%|█████████▎| 6033/6470 [1:18:47<05:37,  1.29it/s]
1|6034|Loss: 0.8996480703353882:  93%|█████████▎| 6033/6470 [1:18:47<05:37,  1.29it/s]
1|6034|Loss: 0.8996480703353882:  93%|█████████▎| 6034/6470 [1:18:47<05:27,  1.33it/s]
1|6035|Loss: 0.7869009375572205:  93%|█████████▎| 6034/6470 [1:18:48<05:27,  1.33it/s]
1|6035|Loss: 0.7869009375572205:  93%|█████████▎| 6035/6470 [1:18:48<05:09,  1.40it/s]
1|6036|Loss: 0.8114565014839172:  93%|█████████▎| 6035/6470 [1:18:48<05:09,  1.40it/s]
1|6036|Loss: 0.8114565014839172:  93%|█████████▎| 6036/6470 [1:18:48<04:40,  1.55it/s]
1|6037|Loss: 0.8521748185157776:  93%|█████████▎| 6036/6470 [1:18:49<04:40,  1.55it/s]
1|6037|Loss: 0.8521748185157776:  93%|█████████▎| 6037/6470 [1:18:49<04:45,  1.51it/s]
1|6038|Loss: 0.8013742566108704:  93%|█████████▎| 6037/6470 [1:18:50<04:45,  1.51it/s]
1|6038|Loss: 0.8013742566108704:  93%|█████████▎| 6038/6470 [1:18:50<04:58,  1.45it/s]
1|6039|Loss: 0.9678013920783997:  93%|█████████▎| 6038/6470 [1:18:50<04:58,  1.45it/s]
1|6039|Loss: 0.9678013920783997:  93%|█████████▎| 6039/6470 [1:18:51<05:06,  1.41it/s]
1|6040|Loss: 0.8442339897155762:  93%|█████████▎| 6039/6470 [1:18:51<05:06,  1.41it/s]
1|6040|Loss: 0.8442339897155762:  93%|█████████▎| 6040/6470 [1:18:51<04:49,  1.49it/s]
1|6041|Loss: 0.8096632361412048:  93%|█████████▎| 6040/6470 [1:18:52<04:49,  1.49it/s]
1|6041|Loss: 0.8096632361412048:  93%|█████████▎| 6041/6470 [1:18:52<04:39,  1.53it/s]
1|6042|Loss: 0.8404425978660583:  93%|█████████▎| 6041/6470 [1:18:52<04:39,  1.53it/s]
1|6042|Loss: 0.8404425978660583:  93%|█████████▎| 6042/6470 [1:18:53<04:54,  1.45it/s]
1|6043|Loss: 1.0677928924560547:  93%|█████████▎| 6042/6470 [1:18:53<04:54,  1.45it/s]
1|6043|Loss: 1.0677928924560547:  93%|█████████▎| 6043/6470 [1:18:53<05:09,  1.38it/s]
1|6044|Loss: 0.8656976819038391:  93%|█████████▎| 6043/6470 [1:18:54<05:09,  1.38it/s]
1|6044|Loss: 0.8656976819038391:  93%|█████████▎| 6044/6470 [1:18:54<05:24,  1.31it/s]
1|6045|Loss: 0.7930483222007751:  93%|█████████▎| 6044/6470 [1:18:55<05:24,  1.31it/s]
1|6045|Loss: 0.7930483222007751:  93%|█████████▎| 6045/6470 [1:18:55<05:29,  1.29it/s]
1|6046|Loss: 0.8657336831092834:  93%|█████████▎| 6045/6470 [1:18:55<05:29,  1.29it/s]
1|6046|Loss: 0.8657336831092834:  93%|█████████▎| 6046/6470 [1:18:56<05:20,  1.32it/s]
1|6047|Loss: 0.9546496868133545:  93%|█████████▎| 6046/6470 [1:18:56<05:20,  1.32it/s]
1|6047|Loss: 0.9546496868133545:  93%|█████████▎| 6047/6470 [1:18:57<05:20,  1.32it/s]
1|6048|Loss: 0.8177930116653442:  93%|█████████▎| 6047/6470 [1:18:57<05:20,  1.32it/s]
1|6048|Loss: 0.8177930116653442:  93%|█████████▎| 6048/6470 [1:18:57<05:39,  1.24it/s]
1|6049|Loss: 0.8199959993362427:  93%|█████████▎| 6048/6470 [1:18:58<05:39,  1.24it/s]
1|6049|Loss: 0.8199959993362427:  93%|█████████▎| 6049/6470 [1:18:58<05:44,  1.22it/s]
1|6050|Loss: 0.8612276911735535:  93%|█████████▎| 6049/6470 [1:18:59<05:44,  1.22it/s]
1|6050|Loss: 0.8612276911735535:  94%|█████████▎| 6050/6470 [1:18:59<05:38,  1.24it/s]
1|6051|Loss: 0.6664574146270752:  94%|█████████▎| 6050/6470 [1:18:59<05:38,  1.24it/s]
1|6051|Loss: 0.6664574146270752:  94%|█████████▎| 6051/6470 [1:19:00<05:02,  1.38it/s]
1|6052|Loss: 0.8740201592445374:  94%|█████████▎| 6051/6470 [1:19:00<05:02,  1.38it/s]
1|6052|Loss: 0.8740201592445374:  94%|█████████▎| 6052/6470 [1:19:00<04:44,  1.47it/s]
1|6053|Loss: 0.9421470761299133:  94%|█████████▎| 6052/6470 [1:19:01<04:44,  1.47it/s]
1|6053|Loss: 0.9421470761299133:  94%|█████████▎| 6053/6470 [1:19:01<04:42,  1.47it/s]
1|6054|Loss: 0.8029496073722839:  94%|█████████▎| 6053/6470 [1:19:01<04:42,  1.47it/s]
1|6054|Loss: 0.8029496073722839:  94%|█████████▎| 6054/6470 [1:19:02<04:43,  1.47it/s]
1|6055|Loss: 0.8199415802955627:  94%|█████████▎| 6054/6470 [1:19:02<04:43,  1.47it/s]
1|6055|Loss: 0.8199415802955627:  94%|█████████▎| 6055/6470 [1:19:02<04:42,  1.47it/s]
1|6056|Loss: 0.9645059108734131:  94%|█████████▎| 6055/6470 [1:19:03<04:42,  1.47it/s]
1|6056|Loss: 0.9645059108734131:  94%|█████████▎| 6056/6470 [1:19:03<04:48,  1.43it/s]
1|6057|Loss: 0.8168430924415588:  94%|█████████▎| 6056/6470 [1:19:03<04:48,  1.43it/s]
1|6057|Loss: 0.8168430924415588:  94%|█████████▎| 6057/6470 [1:19:04<04:58,  1.38it/s]
1|6058|Loss: 0.731397807598114:  94%|█████████▎| 6057/6470 [1:19:04<04:58,  1.38it/s] 
1|6058|Loss: 0.731397807598114:  94%|█████████▎| 6058/6470 [1:19:04<04:54,  1.40it/s]
1|6059|Loss: 0.8897804021835327:  94%|█████████▎| 6058/6470 [1:19:05<04:54,  1.40it/s]
1|6059|Loss: 0.8897804021835327:  94%|█████████▎| 6059/6470 [1:19:05<04:56,  1.39it/s]
1|6060|Loss: 0.8899324536323547:  94%|█████████▎| 6059/6470 [1:19:06<04:56,  1.39it/s]
1|6060|Loss: 0.8899324536323547:  94%|█████████▎| 6060/6470 [1:19:06<05:16,  1.29it/s]
1|6061|Loss: 0.8066737651824951:  94%|█████████▎| 6060/6470 [1:19:06<05:16,  1.29it/s]
1|6061|Loss: 0.8066737651824951:  94%|█████████▎| 6061/6470 [1:19:07<04:43,  1.44it/s]
1|6062|Loss: 0.9285870790481567:  94%|█████████▎| 6061/6470 [1:19:07<04:43,  1.44it/s]
1|6062|Loss: 0.9285870790481567:  94%|█████████▎| 6062/6470 [1:19:07<04:50,  1.40it/s]
1|6063|Loss: 0.8907107710838318:  94%|█████████▎| 6062/6470 [1:19:08<04:50,  1.40it/s]
1|6063|Loss: 0.8907107710838318:  94%|█████████▎| 6063/6470 [1:19:08<04:32,  1.49it/s]
1|6064|Loss: 0.8067746162414551:  94%|█████████▎| 6063/6470 [1:19:08<04:32,  1.49it/s]
1|6064|Loss: 0.8067746162414551:  94%|█████████▎| 6064/6470 [1:19:09<04:26,  1.52it/s]
1|6065|Loss: 0.9639850854873657:  94%|█████████▎| 6064/6470 [1:19:09<04:26,  1.52it/s]
1|6065|Loss: 0.9639850854873657:  94%|█████████▎| 6065/6470 [1:19:09<04:37,  1.46it/s]
1|6066|Loss: 0.8335893750190735:  94%|█████████▎| 6065/6470 [1:19:10<04:37,  1.46it/s]
1|6066|Loss: 0.8335893750190735:  94%|█████████▍| 6066/6470 [1:19:10<04:37,  1.46it/s]
1|6067|Loss: 0.8068066835403442:  94%|█████████▍| 6066/6470 [1:19:10<04:37,  1.46it/s]
1|6067|Loss: 0.8068066835403442:  94%|█████████▍| 6067/6470 [1:19:11<04:34,  1.47it/s]
1|6068|Loss: 0.7717719078063965:  94%|█████████▍| 6067/6470 [1:19:11<04:34,  1.47it/s]
1|6068|Loss: 0.7717719078063965:  94%|█████████▍| 6068/6470 [1:19:11<04:40,  1.43it/s]
1|6069|Loss: 0.9334148168563843:  94%|█████████▍| 6068/6470 [1:19:12<04:40,  1.43it/s]
1|6069|Loss: 0.9334148168563843:  94%|█████████▍| 6069/6470 [1:19:12<04:42,  1.42it/s]
1|6070|Loss: 0.7224275469779968:  94%|█████████▍| 6069/6470 [1:19:13<04:42,  1.42it/s]
1|6070|Loss: 0.7224275469779968:  94%|█████████▍| 6070/6470 [1:19:13<04:34,  1.46it/s]
1|6071|Loss: 0.7870627045631409:  94%|█████████▍| 6070/6470 [1:19:13<04:34,  1.46it/s]
1|6071|Loss: 0.7870627045631409:  94%|█████████▍| 6071/6470 [1:19:13<04:19,  1.54it/s]
1|6072|Loss: 0.8145871162414551:  94%|█████████▍| 6071/6470 [1:19:14<04:19,  1.54it/s]
1|6072|Loss: 0.8145871162414551:  94%|█████████▍| 6072/6470 [1:19:14<04:23,  1.51it/s]
1|6073|Loss: 0.9204061031341553:  94%|█████████▍| 6072/6470 [1:19:14<04:23,  1.51it/s]
1|6073|Loss: 0.9204061031341553:  94%|█████████▍| 6073/6470 [1:19:15<04:17,  1.54it/s]
1|6074|Loss: 0.9666913151741028:  94%|█████████▍| 6073/6470 [1:19:15<04:17,  1.54it/s]
1|6074|Loss: 0.9666913151741028:  94%|█████████▍| 6074/6470 [1:19:15<04:12,  1.57it/s]
1|6075|Loss: 0.7762704491615295:  94%|█████████▍| 6074/6470 [1:19:16<04:12,  1.57it/s]
1|6075|Loss: 0.7762704491615295:  94%|█████████▍| 6075/6470 [1:19:16<04:09,  1.58it/s]
1|6076|Loss: 0.8796738982200623:  94%|█████████▍| 6075/6470 [1:19:16<04:09,  1.58it/s]
1|6076|Loss: 0.8796738982200623:  94%|█████████▍| 6076/6470 [1:19:17<04:35,  1.43it/s]
1|6077|Loss: 0.9264734387397766:  94%|█████████▍| 6076/6470 [1:19:17<04:35,  1.43it/s]
1|6077|Loss: 0.9264734387397766:  94%|█████████▍| 6077/6470 [1:19:17<04:41,  1.40it/s]
1|6078|Loss: 0.8911202549934387:  94%|█████████▍| 6077/6470 [1:19:18<04:41,  1.40it/s]
1|6078|Loss: 0.8911202549934387:  94%|█████████▍| 6078/6470 [1:19:18<04:44,  1.38it/s]
1|6079|Loss: 0.8847642540931702:  94%|█████████▍| 6078/6470 [1:19:19<04:44,  1.38it/s]
1|6079|Loss: 0.8847642540931702:  94%|█████████▍| 6079/6470 [1:19:19<04:51,  1.34it/s]
1|6080|Loss: 0.8522941470146179:  94%|█████████▍| 6079/6470 [1:19:19<04:51,  1.34it/s]
1|6080|Loss: 0.8522941470146179:  94%|█████████▍| 6080/6470 [1:19:20<04:43,  1.37it/s]
1|6081|Loss: 0.8030132055282593:  94%|█████████▍| 6080/6470 [1:19:20<04:43,  1.37it/s]
1|6081|Loss: 0.8030132055282593:  94%|█████████▍| 6081/6470 [1:19:20<04:46,  1.36it/s]
1|6082|Loss: 0.7940400242805481:  94%|█████████▍| 6081/6470 [1:19:21<04:46,  1.36it/s]
1|6082|Loss: 0.7940400242805481:  94%|█████████▍| 6082/6470 [1:19:21<04:40,  1.38it/s]
1|6083|Loss: 0.8900743722915649:  94%|█████████▍| 6082/6470 [1:19:22<04:40,  1.38it/s]
1|6083|Loss: 0.8900743722915649:  94%|█████████▍| 6083/6470 [1:19:22<04:51,  1.33it/s]
1|6084|Loss: 0.8059564232826233:  94%|█████████▍| 6083/6470 [1:19:22<04:51,  1.33it/s]
1|6084|Loss: 0.8059564232826233:  94%|█████████▍| 6084/6470 [1:19:23<05:09,  1.25it/s]
1|6085|Loss: 0.8553717136383057:  94%|█████████▍| 6084/6470 [1:19:23<05:09,  1.25it/s]
1|6085|Loss: 0.8553717136383057:  94%|█████████▍| 6085/6470 [1:19:24<05:26,  1.18it/s]
1|6086|Loss: 0.8518174290657043:  94%|█████████▍| 6085/6470 [1:19:24<05:26,  1.18it/s]
1|6086|Loss: 0.8518174290657043:  94%|█████████▍| 6086/6470 [1:19:25<05:20,  1.20it/s]
1|6087|Loss: 0.9443119764328003:  94%|█████████▍| 6086/6470 [1:19:25<05:20,  1.20it/s]
1|6087|Loss: 0.9443119764328003:  94%|█████████▍| 6087/6470 [1:19:25<05:11,  1.23it/s]
1|6088|Loss: 0.8287606239318848:  94%|█████████▍| 6087/6470 [1:19:26<05:11,  1.23it/s]
1|6088|Loss: 0.8287606239318848:  94%|█████████▍| 6088/6470 [1:19:26<04:45,  1.34it/s]
1|6089|Loss: 0.8035192489624023:  94%|█████████▍| 6088/6470 [1:19:26<04:45,  1.34it/s]
1|6089|Loss: 0.8035192489624023:  94%|█████████▍| 6089/6470 [1:19:27<05:00,  1.27it/s]
1|6090|Loss: 0.9272462129592896:  94%|█████████▍| 6089/6470 [1:19:27<05:00,  1.27it/s]
1|6090|Loss: 0.9272462129592896:  94%|█████████▍| 6090/6470 [1:19:28<04:55,  1.28it/s]
1|6091|Loss: 0.9492820501327515:  94%|█████████▍| 6090/6470 [1:19:28<04:55,  1.28it/s]
1|6091|Loss: 0.9492820501327515:  94%|█████████▍| 6091/6470 [1:19:28<04:53,  1.29it/s]
1|6092|Loss: 0.8570132851600647:  94%|█████████▍| 6091/6470 [1:19:29<04:53,  1.29it/s]
1|6092|Loss: 0.8570132851600647:  94%|█████████▍| 6092/6470 [1:19:29<04:44,  1.33it/s]
1|6093|Loss: 0.9130606651306152:  94%|█████████▍| 6092/6470 [1:19:29<04:44,  1.33it/s]
1|6093|Loss: 0.9130606651306152:  94%|█████████▍| 6093/6470 [1:19:30<04:30,  1.39it/s]
1|6094|Loss: 0.9055965542793274:  94%|█████████▍| 6093/6470 [1:19:30<04:30,  1.39it/s]
1|6094|Loss: 0.9055965542793274:  94%|█████████▍| 6094/6470 [1:19:31<04:35,  1.36it/s]
1|6095|Loss: 0.8775931596755981:  94%|█████████▍| 6094/6470 [1:19:31<04:35,  1.36it/s]
1|6095|Loss: 0.8775931596755981:  94%|█████████▍| 6095/6470 [1:19:31<04:39,  1.34it/s]
1|6096|Loss: 0.8754098415374756:  94%|█████████▍| 6095/6470 [1:19:32<04:39,  1.34it/s]
1|6096|Loss: 0.8754098415374756:  94%|█████████▍| 6096/6470 [1:19:32<04:24,  1.41it/s]
1|6097|Loss: 0.8456804156303406:  94%|█████████▍| 6096/6470 [1:19:32<04:24,  1.41it/s]
1|6097|Loss: 0.8456804156303406:  94%|█████████▍| 6097/6470 [1:19:33<04:19,  1.44it/s]
1|6098|Loss: 0.7740748524665833:  94%|█████████▍| 6097/6470 [1:19:33<04:19,  1.44it/s]
1|6098|Loss: 0.7740748524665833:  94%|█████████▍| 6098/6470 [1:19:33<04:24,  1.40it/s]
1|6099|Loss: 0.8140228390693665:  94%|█████████▍| 6098/6470 [1:19:34<04:24,  1.40it/s]
1|6099|Loss: 0.8140228390693665:  94%|█████████▍| 6099/6470 [1:19:34<04:28,  1.38it/s]
1|6100|Loss: 0.9041417837142944:  94%|█████████▍| 6099/6470 [1:19:34<04:28,  1.38it/s]
1|6100|Loss: 0.9041417837142944:  94%|█████████▍| 6100/6470 [1:19:35<04:13,  1.46it/s]
1|6101|Loss: 0.9250243902206421:  94%|█████████▍| 6100/6470 [1:19:35<04:13,  1.46it/s]
1|6101|Loss: 0.9250243902206421:  94%|█████████▍| 6101/6470 [1:19:35<04:16,  1.44it/s]
1|6102|Loss: 0.8049137592315674:  94%|█████████▍| 6101/6470 [1:19:36<04:16,  1.44it/s]
1|6102|Loss: 0.8049137592315674:  94%|█████████▍| 6102/6470 [1:19:36<04:27,  1.38it/s]
1|6103|Loss: 0.9906845688819885:  94%|█████████▍| 6102/6470 [1:19:37<04:27,  1.38it/s]
1|6103|Loss: 0.9906845688819885:  94%|█████████▍| 6103/6470 [1:19:37<04:18,  1.42it/s]
1|6104|Loss: 0.9802946448326111:  94%|█████████▍| 6103/6470 [1:19:37<04:18,  1.42it/s]
1|6104|Loss: 0.9802946448326111:  94%|█████████▍| 6104/6470 [1:19:37<04:09,  1.47it/s]
1|6105|Loss: 0.8182365298271179:  94%|█████████▍| 6104/6470 [1:19:38<04:09,  1.47it/s]
1|6105|Loss: 0.8182365298271179:  94%|█████████▍| 6105/6470 [1:19:38<04:29,  1.36it/s]
1|6106|Loss: 0.9230344295501709:  94%|█████████▍| 6105/6470 [1:19:39<04:29,  1.36it/s]
1|6106|Loss: 0.9230344295501709:  94%|█████████▍| 6106/6470 [1:19:39<04:34,  1.32it/s]
1|6107|Loss: 0.9648343920707703:  94%|█████████▍| 6106/6470 [1:19:39<04:34,  1.32it/s]
1|6107|Loss: 0.9648343920707703:  94%|█████████▍| 6107/6470 [1:19:40<04:09,  1.46it/s]
1|6108|Loss: 0.7652389407157898:  94%|█████████▍| 6107/6470 [1:19:40<04:09,  1.46it/s]
1|6108|Loss: 0.7652389407157898:  94%|█████████▍| 6108/6470 [1:19:40<04:19,  1.40it/s]
1|6109|Loss: 0.9361345171928406:  94%|█████████▍| 6108/6470 [1:19:41<04:19,  1.40it/s]
1|6109|Loss: 0.9361345171928406:  94%|█████████▍| 6109/6470 [1:19:41<04:16,  1.41it/s]
1|6110|Loss: 0.9172989726066589:  94%|█████████▍| 6109/6470 [1:19:42<04:16,  1.41it/s]
1|6110|Loss: 0.9172989726066589:  94%|█████████▍| 6110/6470 [1:19:42<04:11,  1.43it/s]
1|6111|Loss: 0.8321548104286194:  94%|█████████▍| 6110/6470 [1:19:42<04:11,  1.43it/s]
1|6111|Loss: 0.8321548104286194:  94%|█████████▍| 6111/6470 [1:19:42<04:06,  1.45it/s]
1|6112|Loss: 0.9230448603630066:  94%|█████████▍| 6111/6470 [1:19:43<04:06,  1.45it/s]
1|6112|Loss: 0.9230448603630066:  94%|█████████▍| 6112/6470 [1:19:43<04:21,  1.37it/s]
1|6113|Loss: 0.6374751925468445:  94%|█████████▍| 6112/6470 [1:19:44<04:21,  1.37it/s]
1|6113|Loss: 0.6374751925468445:  94%|█████████▍| 6113/6470 [1:19:44<04:09,  1.43it/s]
1|6114|Loss: 0.7938874959945679:  94%|█████████▍| 6113/6470 [1:19:44<04:09,  1.43it/s]
1|6114|Loss: 0.7938874959945679:  94%|█████████▍| 6114/6470 [1:19:45<04:31,  1.31it/s]
1|6115|Loss: 0.8895284533500671:  94%|█████████▍| 6114/6470 [1:19:45<04:31,  1.31it/s]
1|6115|Loss: 0.8895284533500671:  95%|█████████▍| 6115/6470 [1:19:46<04:51,  1.22it/s]
1|6116|Loss: 0.8426374197006226:  95%|█████████▍| 6115/6470 [1:19:46<04:51,  1.22it/s]
1|6116|Loss: 0.8426374197006226:  95%|█████████▍| 6116/6470 [1:19:47<04:38,  1.27it/s]
1|6117|Loss: 0.828030526638031:  95%|█████████▍| 6116/6470 [1:19:47<04:38,  1.27it/s] 
1|6117|Loss: 0.828030526638031:  95%|█████████▍| 6117/6470 [1:19:47<04:53,  1.20it/s]
1|6118|Loss: 0.967818021774292:  95%|█████████▍| 6117/6470 [1:19:48<04:53,  1.20it/s]
1|6118|Loss: 0.967818021774292:  95%|█████████▍| 6118/6470 [1:19:48<04:53,  1.20it/s]
1|6119|Loss: 0.8130002021789551:  95%|█████████▍| 6118/6470 [1:19:49<04:53,  1.20it/s]
1|6119|Loss: 0.8130002021789551:  95%|█████████▍| 6119/6470 [1:19:49<04:50,  1.21it/s]
1|6120|Loss: 0.7541795372962952:  95%|█████████▍| 6119/6470 [1:19:50<04:50,  1.21it/s]
1|6120|Loss: 0.7541795372962952:  95%|█████████▍| 6120/6470 [1:19:50<04:38,  1.25it/s]
1|6121|Loss: 0.6981000900268555:  95%|█████████▍| 6120/6470 [1:19:50<04:38,  1.25it/s]
1|6121|Loss: 0.6981000900268555:  95%|█████████▍| 6121/6470 [1:19:50<04:09,  1.40it/s]
1|6122|Loss: 0.7944573760032654:  95%|█████████▍| 6121/6470 [1:19:51<04:09,  1.40it/s]
1|6122|Loss: 0.7944573760032654:  95%|█████████▍| 6122/6470 [1:19:51<03:59,  1.45it/s]
1|6123|Loss: 0.8188761472702026:  95%|█████████▍| 6122/6470 [1:19:51<03:59,  1.45it/s]
1|6123|Loss: 0.8188761472702026:  95%|█████████▍| 6123/6470 [1:19:52<04:08,  1.40it/s]
1|6124|Loss: 0.9153571128845215:  95%|█████████▍| 6123/6470 [1:19:52<04:08,  1.40it/s]
1|6124|Loss: 0.9153571128845215:  95%|█████████▍| 6124/6470 [1:19:53<04:27,  1.29it/s]
1|6125|Loss: 0.8060866594314575:  95%|█████████▍| 6124/6470 [1:19:53<04:27,  1.29it/s]
1|6125|Loss: 0.8060866594314575:  95%|█████████▍| 6125/6470 [1:19:53<04:29,  1.28it/s]
1|6126|Loss: 0.7737741470336914:  95%|█████████▍| 6125/6470 [1:19:54<04:29,  1.28it/s]
1|6126|Loss: 0.7737741470336914:  95%|█████████▍| 6126/6470 [1:19:54<04:35,  1.25it/s]
1|6127|Loss: 0.815136194229126:  95%|█████████▍| 6126/6470 [1:19:55<04:35,  1.25it/s] 
1|6127|Loss: 0.815136194229126:  95%|█████████▍| 6127/6470 [1:19:55<04:28,  1.28it/s]
1|6128|Loss: 0.921165943145752:  95%|█████████▍| 6127/6470 [1:19:55<04:28,  1.28it/s]
1|6128|Loss: 0.921165943145752:  95%|█████████▍| 6128/6470 [1:19:56<04:18,  1.32it/s]
1|6129|Loss: 0.85970139503479:  95%|█████████▍| 6128/6470 [1:19:56<04:18,  1.32it/s] 
1|6129|Loss: 0.85970139503479:  95%|█████████▍| 6129/6470 [1:19:56<03:59,  1.42it/s]
1|6130|Loss: 0.9671984910964966:  95%|█████████▍| 6129/6470 [1:19:57<03:59,  1.42it/s]
1|6130|Loss: 0.9671984910964966:  95%|█████████▍| 6130/6470 [1:19:57<04:16,  1.33it/s]
1|6131|Loss: 0.9482869505882263:  95%|█████████▍| 6130/6470 [1:19:58<04:16,  1.33it/s]
1|6131|Loss: 0.9482869505882263:  95%|█████████▍| 6131/6470 [1:19:58<04:33,  1.24it/s]
1|6132|Loss: 0.8500941395759583:  95%|█████████▍| 6131/6470 [1:19:59<04:33,  1.24it/s]
1|6132|Loss: 0.8500941395759583:  95%|█████████▍| 6132/6470 [1:19:59<04:31,  1.24it/s]
1|6133|Loss: 0.9340818524360657:  95%|█████████▍| 6132/6470 [1:19:59<04:31,  1.24it/s]
1|6133|Loss: 0.9340818524360657:  95%|█████████▍| 6133/6470 [1:20:00<04:41,  1.20it/s]
1|6134|Loss: 0.742398738861084:  95%|█████████▍| 6133/6470 [1:20:00<04:41,  1.20it/s] 
1|6134|Loss: 0.742398738861084:  95%|█████████▍| 6134/6470 [1:20:01<04:51,  1.15it/s]
1|6135|Loss: 0.8361626863479614:  95%|█████████▍| 6134/6470 [1:20:01<04:51,  1.15it/s]
1|6135|Loss: 0.8361626863479614:  95%|█████████▍| 6135/6470 [1:20:02<04:44,  1.18it/s]
1|6136|Loss: 0.7606020569801331:  95%|█████████▍| 6135/6470 [1:20:02<04:44,  1.18it/s]
1|6136|Loss: 0.7606020569801331:  95%|█████████▍| 6136/6470 [1:20:02<04:38,  1.20it/s]
1|6137|Loss: 0.8468437194824219:  95%|█████████▍| 6136/6470 [1:20:03<04:38,  1.20it/s]
1|6137|Loss: 0.8468437194824219:  95%|█████████▍| 6137/6470 [1:20:03<04:45,  1.17it/s]
1|6138|Loss: 0.863903820514679:  95%|█████████▍| 6137/6470 [1:20:04<04:45,  1.17it/s] 
1|6138|Loss: 0.863903820514679:  95%|█████████▍| 6138/6470 [1:20:04<04:28,  1.24it/s]
1|6139|Loss: 0.830720067024231:  95%|█████████▍| 6138/6470 [1:20:04<04:28,  1.24it/s]
1|6139|Loss: 0.830720067024231:  95%|█████████▍| 6139/6470 [1:20:05<04:35,  1.20it/s]
1|6140|Loss: 0.8568742871284485:  95%|█████████▍| 6139/6470 [1:20:05<04:35,  1.20it/s]
1|6140|Loss: 0.8568742871284485:  95%|█████████▍| 6140/6470 [1:20:06<04:31,  1.22it/s]
1|6141|Loss: 0.8301260471343994:  95%|█████████▍| 6140/6470 [1:20:06<04:31,  1.22it/s]
1|6141|Loss: 0.8301260471343994:  95%|█████████▍| 6141/6470 [1:20:07<04:44,  1.16it/s]
1|6142|Loss: 0.8269612789154053:  95%|█████████▍| 6141/6470 [1:20:07<04:44,  1.16it/s]
1|6142|Loss: 0.8269612789154053:  95%|█████████▍| 6142/6470 [1:20:07<04:35,  1.19it/s]
1|6143|Loss: 0.9758504629135132:  95%|█████████▍| 6142/6470 [1:20:08<04:35,  1.19it/s]
1|6143|Loss: 0.9758504629135132:  95%|█████████▍| 6143/6470 [1:20:08<04:28,  1.22it/s]
1|6144|Loss: 0.8562703132629395:  95%|█████████▍| 6143/6470 [1:20:09<04:28,  1.22it/s]
1|6144|Loss: 0.8562703132629395:  95%|█████████▍| 6144/6470 [1:20:09<04:22,  1.24it/s]
1|6145|Loss: 0.9667564630508423:  95%|█████████▍| 6144/6470 [1:20:09<04:22,  1.24it/s]
1|6145|Loss: 0.9667564630508423:  95%|█████████▍| 6145/6470 [1:20:10<04:00,  1.35it/s]
1|6146|Loss: 0.8239200115203857:  95%|█████████▍| 6145/6470 [1:20:10<04:00,  1.35it/s]
1|6146|Loss: 0.8239200115203857:  95%|█████████▍| 6146/6470 [1:20:10<03:49,  1.41it/s]
1|6147|Loss: 1.1825298070907593:  95%|█████████▍| 6146/6470 [1:20:11<03:49,  1.41it/s]
1|6147|Loss: 1.1825298070907593:  95%|█████████▌| 6147/6470 [1:20:11<03:33,  1.51it/s]
1|6148|Loss: 0.7761172652244568:  95%|█████████▌| 6147/6470 [1:20:11<03:33,  1.51it/s]
1|6148|Loss: 0.7761172652244568:  95%|█████████▌| 6148/6470 [1:20:11<03:26,  1.56it/s]
1|6149|Loss: 0.7385208010673523:  95%|█████████▌| 6148/6470 [1:20:12<03:26,  1.56it/s]
1|6149|Loss: 0.7385208010673523:  95%|█████████▌| 6149/6470 [1:20:12<03:41,  1.45it/s]
1|6150|Loss: 0.8552089929580688:  95%|█████████▌| 6149/6470 [1:20:12<03:41,  1.45it/s]
1|6150|Loss: 0.8552089929580688:  95%|█████████▌| 6150/6470 [1:20:13<03:19,  1.60it/s]
1|6151|Loss: 0.7811774015426636:  95%|█████████▌| 6150/6470 [1:20:13<03:19,  1.60it/s]
1|6151|Loss: 0.7811774015426636:  95%|█████████▌| 6151/6470 [1:20:13<03:27,  1.54it/s]
1|6152|Loss: 0.8732529878616333:  95%|█████████▌| 6151/6470 [1:20:14<03:27,  1.54it/s]
1|6152|Loss: 0.8732529878616333:  95%|█████████▌| 6152/6470 [1:20:14<03:38,  1.46it/s]
1|6153|Loss: 0.8174307346343994:  95%|█████████▌| 6152/6470 [1:20:15<03:38,  1.46it/s]
1|6153|Loss: 0.8174307346343994:  95%|█████████▌| 6153/6470 [1:20:15<03:47,  1.39it/s]
1|6154|Loss: 0.84471595287323:  95%|█████████▌| 6153/6470 [1:20:15<03:47,  1.39it/s]  
1|6154|Loss: 0.84471595287323:  95%|█████████▌| 6154/6470 [1:20:16<03:46,  1.39it/s]
1|6155|Loss: 0.8181833028793335:  95%|█████████▌| 6154/6470 [1:20:16<03:46,  1.39it/s]
1|6155|Loss: 0.8181833028793335:  95%|█████████▌| 6155/6470 [1:20:16<03:49,  1.37it/s]
1|6156|Loss: 0.9149969220161438:  95%|█████████▌| 6155/6470 [1:20:17<03:49,  1.37it/s]
1|6156|Loss: 0.9149969220161438:  95%|█████████▌| 6156/6470 [1:20:17<03:55,  1.33it/s]
1|6157|Loss: 0.8299562335014343:  95%|█████████▌| 6156/6470 [1:20:18<03:55,  1.33it/s]
1|6157|Loss: 0.8299562335014343:  95%|█████████▌| 6157/6470 [1:20:18<04:08,  1.26it/s]
1|6158|Loss: 1.247642159461975:  95%|█████████▌| 6157/6470 [1:20:19<04:08,  1.26it/s] 
1|6158|Loss: 1.247642159461975:  95%|█████████▌| 6158/6470 [1:20:19<04:26,  1.17it/s]
1|6159|Loss: 0.9010798335075378:  95%|█████████▌| 6158/6470 [1:20:20<04:26,  1.17it/s]
1|6159|Loss: 0.9010798335075378:  95%|█████████▌| 6159/6470 [1:20:20<04:18,  1.20it/s]
1|6160|Loss: 0.9269248843193054:  95%|█████████▌| 6159/6470 [1:20:20<04:18,  1.20it/s]
1|6160|Loss: 0.9269248843193054:  95%|█████████▌| 6160/6470 [1:20:21<04:14,  1.22it/s]
1|6161|Loss: 0.939083456993103:  95%|█████████▌| 6160/6470 [1:20:21<04:14,  1.22it/s] 
1|6161|Loss: 0.939083456993103:  95%|█████████▌| 6161/6470 [1:20:21<04:05,  1.26it/s]
1|6162|Loss: 0.8673751950263977:  95%|█████████▌| 6161/6470 [1:20:22<04:05,  1.26it/s]
1|6162|Loss: 0.8673751950263977:  95%|█████████▌| 6162/6470 [1:20:22<04:03,  1.27it/s]
1|6163|Loss: 0.9780818223953247:  95%|█████████▌| 6162/6470 [1:20:22<04:03,  1.27it/s]
1|6163|Loss: 0.9780818223953247:  95%|█████████▌| 6163/6470 [1:20:23<03:39,  1.40it/s]
1|6164|Loss: 0.7776886820793152:  95%|█████████▌| 6163/6470 [1:20:23<03:39,  1.40it/s]
1|6164|Loss: 0.7776886820793152:  95%|█████████▌| 6164/6470 [1:20:23<03:41,  1.38it/s]
1|6165|Loss: 0.818753182888031:  95%|█████████▌| 6164/6470 [1:20:24<03:41,  1.38it/s] 
1|6165|Loss: 0.818753182888031:  95%|█████████▌| 6165/6470 [1:20:24<03:36,  1.41it/s]
1|6166|Loss: 0.9258757829666138:  95%|█████████▌| 6165/6470 [1:20:24<03:36,  1.41it/s]
1|6166|Loss: 0.9258757829666138:  95%|█████████▌| 6166/6470 [1:20:25<03:16,  1.55it/s]
1|6167|Loss: 0.8314639329910278:  95%|█████████▌| 6166/6470 [1:20:25<03:16,  1.55it/s]
1|6167|Loss: 0.8314639329910278:  95%|█████████▌| 6167/6470 [1:20:25<03:29,  1.44it/s]
1|6168|Loss: 0.8702837824821472:  95%|█████████▌| 6167/6470 [1:20:26<03:29,  1.44it/s]
1|6168|Loss: 0.8702837824821472:  95%|█████████▌| 6168/6470 [1:20:26<03:21,  1.50it/s]
1|6169|Loss: 0.7204045057296753:  95%|█████████▌| 6168/6470 [1:20:26<03:21,  1.50it/s]
1|6169|Loss: 0.7204045057296753:  95%|█████████▌| 6169/6470 [1:20:27<03:07,  1.61it/s]
1|6170|Loss: 0.8030838370323181:  95%|█████████▌| 6169/6470 [1:20:27<03:07,  1.61it/s]
1|6170|Loss: 0.8030838370323181:  95%|█████████▌| 6170/6470 [1:20:27<03:04,  1.63it/s]
1|6171|Loss: 0.748137354850769:  95%|█████████▌| 6170/6470 [1:20:28<03:04,  1.63it/s] 
1|6171|Loss: 0.748137354850769:  95%|█████████▌| 6171/6470 [1:20:28<03:18,  1.51it/s]
1|6172|Loss: 0.9245545864105225:  95%|█████████▌| 6171/6470 [1:20:28<03:18,  1.51it/s]
1|6172|Loss: 0.9245545864105225:  95%|█████████▌| 6172/6470 [1:20:29<03:21,  1.48it/s]
1|6173|Loss: 0.9092815518379211:  95%|█████████▌| 6172/6470 [1:20:29<03:21,  1.48it/s]
1|6173|Loss: 0.9092815518379211:  95%|█████████▌| 6173/6470 [1:20:29<03:15,  1.52it/s]
1|6174|Loss: 0.86765056848526:  95%|█████████▌| 6173/6470 [1:20:30<03:15,  1.52it/s]  
1|6174|Loss: 0.86765056848526:  95%|█████████▌| 6174/6470 [1:20:30<03:21,  1.47it/s]
1|6175|Loss: 0.8373206853866577:  95%|█████████▌| 6174/6470 [1:20:30<03:21,  1.47it/s]
1|6175|Loss: 0.8373206853866577:  95%|█████████▌| 6175/6470 [1:20:31<03:18,  1.49it/s]
1|6176|Loss: 0.8375898003578186:  95%|█████████▌| 6175/6470 [1:20:31<03:18,  1.49it/s]
1|6176|Loss: 0.8375898003578186:  95%|█████████▌| 6176/6470 [1:20:31<03:29,  1.40it/s]
1|6177|Loss: 0.9303144812583923:  95%|█████████▌| 6176/6470 [1:20:32<03:29,  1.40it/s]
1|6177|Loss: 0.9303144812583923:  95%|█████████▌| 6177/6470 [1:20:32<03:28,  1.41it/s]
1|6178|Loss: 0.8993736505508423:  95%|█████████▌| 6177/6470 [1:20:32<03:28,  1.41it/s]
1|6178|Loss: 0.8993736505508423:  95%|█████████▌| 6178/6470 [1:20:33<03:14,  1.50it/s]
1|6179|Loss: 0.965430736541748:  95%|█████████▌| 6178/6470 [1:20:33<03:14,  1.50it/s] 
1|6179|Loss: 0.965430736541748:  96%|█████████▌| 6179/6470 [1:20:34<03:27,  1.40it/s]
1|6180|Loss: 0.9628137350082397:  96%|█████████▌| 6179/6470 [1:20:34<03:27,  1.40it/s]
1|6180|Loss: 0.9628137350082397:  96%|█████████▌| 6180/6470 [1:20:34<03:32,  1.36it/s]
1|6181|Loss: 0.8404120206832886:  96%|█████████▌| 6180/6470 [1:20:35<03:32,  1.36it/s]
1|6181|Loss: 0.8404120206832886:  96%|█████████▌| 6181/6470 [1:20:35<03:28,  1.39it/s]
1|6182|Loss: 0.8652042150497437:  96%|█████████▌| 6181/6470 [1:20:35<03:28,  1.39it/s]
1|6182|Loss: 0.8652042150497437:  96%|█████████▌| 6182/6470 [1:20:36<03:23,  1.42it/s]
1|6183|Loss: 0.7110296487808228:  96%|█████████▌| 6182/6470 [1:20:36<03:23,  1.42it/s]
1|6183|Loss: 0.7110296487808228:  96%|█████████▌| 6183/6470 [1:20:36<03:19,  1.44it/s]
1|6184|Loss: 0.887440025806427:  96%|█████████▌| 6183/6470 [1:20:37<03:19,  1.44it/s] 
1|6184|Loss: 0.887440025806427:  96%|█████████▌| 6184/6470 [1:20:37<03:25,  1.39it/s]
1|6185|Loss: 0.8453119397163391:  96%|█████████▌| 6184/6470 [1:20:38<03:25,  1.39it/s]
1|6185|Loss: 0.8453119397163391:  96%|█████████▌| 6185/6470 [1:20:38<03:41,  1.29it/s]
1|6186|Loss: 0.9193336963653564:  96%|█████████▌| 6185/6470 [1:20:38<03:41,  1.29it/s]
1|6186|Loss: 0.9193336963653564:  96%|█████████▌| 6186/6470 [1:20:39<03:46,  1.26it/s]
1|6187|Loss: 0.865315854549408:  96%|█████████▌| 6186/6470 [1:20:39<03:46,  1.26it/s] 
1|6187|Loss: 0.865315854549408:  96%|█████████▌| 6187/6470 [1:20:40<03:45,  1.26it/s]
1|6188|Loss: 0.969097375869751:  96%|█████████▌| 6187/6470 [1:20:40<03:45,  1.26it/s]
1|6188|Loss: 0.969097375869751:  96%|█████████▌| 6188/6470 [1:20:40<03:16,  1.44it/s]
1|6189|Loss: 0.9743995070457458:  96%|█████████▌| 6188/6470 [1:20:40<03:16,  1.44it/s]
1|6189|Loss: 0.9743995070457458:  96%|█████████▌| 6189/6470 [1:20:41<03:11,  1.47it/s]
1|6190|Loss: 0.8571740984916687:  96%|█████████▌| 6189/6470 [1:20:41<03:11,  1.47it/s]
1|6190|Loss: 0.8571740984916687:  96%|█████████▌| 6190/6470 [1:20:42<03:39,  1.28it/s]
1|6191|Loss: 0.8870528340339661:  96%|█████████▌| 6190/6470 [1:20:42<03:39,  1.28it/s]
1|6191|Loss: 0.8870528340339661:  96%|█████████▌| 6191/6470 [1:20:43<03:44,  1.24it/s]
1|6192|Loss: 0.933405339717865:  96%|█████████▌| 6191/6470 [1:20:43<03:44,  1.24it/s] 
1|6192|Loss: 0.933405339717865:  96%|█████████▌| 6192/6470 [1:20:43<03:43,  1.24it/s]
1|6193|Loss: 0.9266785979270935:  96%|█████████▌| 6192/6470 [1:20:44<03:43,  1.24it/s]
1|6193|Loss: 0.9266785979270935:  96%|█████████▌| 6193/6470 [1:20:44<03:35,  1.29it/s]
1|6194|Loss: 0.7325912714004517:  96%|█████████▌| 6193/6470 [1:20:45<03:35,  1.29it/s]
1|6194|Loss: 0.7325912714004517:  96%|█████████▌| 6194/6470 [1:20:45<03:19,  1.39it/s]
1|6195|Loss: 0.8806334733963013:  96%|█████████▌| 6194/6470 [1:20:45<03:19,  1.39it/s]
1|6195|Loss: 0.8806334733963013:  96%|█████████▌| 6195/6470 [1:20:45<03:17,  1.39it/s]
1|6196|Loss: 0.8954592943191528:  96%|█████████▌| 6195/6470 [1:20:46<03:17,  1.39it/s]
1|6196|Loss: 0.8954592943191528:  96%|█████████▌| 6196/6470 [1:20:46<03:21,  1.36it/s]
1|6197|Loss: 0.9429002404212952:  96%|█████████▌| 6196/6470 [1:20:47<03:21,  1.36it/s]
1|6197|Loss: 0.9429002404212952:  96%|█████████▌| 6197/6470 [1:20:47<03:31,  1.29it/s]
1|6198|Loss: 0.8477960824966431:  96%|█████████▌| 6197/6470 [1:20:48<03:31,  1.29it/s]
1|6198|Loss: 0.8477960824966431:  96%|█████████▌| 6198/6470 [1:20:48<03:31,  1.29it/s]
1|6199|Loss: 0.929368257522583:  96%|█████████▌| 6198/6470 [1:20:48<03:31,  1.29it/s] 
1|6199|Loss: 0.929368257522583:  96%|█████████▌| 6199/6470 [1:20:49<03:30,  1.29it/s]
1|6200|Loss: 0.9824523329734802:  96%|█████████▌| 6199/6470 [1:20:49<03:30,  1.29it/s]
1|6200|Loss: 0.9824523329734802:  96%|█████████▌| 6200/6470 [1:20:49<03:35,  1.25it/s]
1|6201|Loss: 0.8969451785087585:  96%|█████████▌| 6200/6470 [1:20:50<03:35,  1.25it/s]
1|6201|Loss: 0.8969451785087585:  96%|█████████▌| 6201/6470 [1:20:50<03:32,  1.26it/s]
1|6202|Loss: 0.8541879653930664:  96%|█████████▌| 6201/6470 [1:20:51<03:32,  1.26it/s]
1|6202|Loss: 0.8541879653930664:  96%|█████████▌| 6202/6470 [1:20:51<03:26,  1.30it/s]
1|6203|Loss: 0.668477475643158:  96%|█████████▌| 6202/6470 [1:20:51<03:26,  1.30it/s] 
1|6203|Loss: 0.668477475643158:  96%|█████████▌| 6203/6470 [1:20:52<03:16,  1.36it/s]
1|6204|Loss: 0.8428736329078674:  96%|█████████▌| 6203/6470 [1:20:52<03:16,  1.36it/s]
1|6204|Loss: 0.8428736329078674:  96%|█████████▌| 6204/6470 [1:20:52<03:07,  1.42it/s]
1|6205|Loss: 0.9429754614830017:  96%|█████████▌| 6204/6470 [1:20:53<03:07,  1.42it/s]
1|6205|Loss: 0.9429754614830017:  96%|█████████▌| 6205/6470 [1:20:53<03:16,  1.35it/s]
1|6206|Loss: 0.6829063892364502:  96%|█████████▌| 6205/6470 [1:20:53<03:16,  1.35it/s]
1|6206|Loss: 0.6829063892364502:  96%|█████████▌| 6206/6470 [1:20:54<02:55,  1.51it/s]
1|6207|Loss: 0.7919871211051941:  96%|█████████▌| 6206/6470 [1:20:54<02:55,  1.51it/s]
1|6207|Loss: 0.7919871211051941:  96%|█████████▌| 6207/6470 [1:20:54<02:58,  1.47it/s]
1|6208|Loss: 0.89723801612854:  96%|█████████▌| 6207/6470 [1:20:55<02:58,  1.47it/s]  
1|6208|Loss: 0.89723801612854:  96%|█████████▌| 6208/6470 [1:20:55<03:02,  1.44it/s]
1|6209|Loss: 0.8729605078697205:  96%|█████████▌| 6208/6470 [1:20:55<03:02,  1.44it/s]
1|6209|Loss: 0.8729605078697205:  96%|█████████▌| 6209/6470 [1:20:56<03:06,  1.40it/s]
1|6210|Loss: 0.7503730058670044:  96%|█████████▌| 6209/6470 [1:20:56<03:06,  1.40it/s]
1|6210|Loss: 0.7503730058670044:  96%|█████████▌| 6210/6470 [1:20:56<03:03,  1.42it/s]
1|6211|Loss: 0.9242728352546692:  96%|█████████▌| 6210/6470 [1:20:57<03:03,  1.42it/s]
1|6211|Loss: 0.9242728352546692:  96%|█████████▌| 6211/6470 [1:20:57<03:05,  1.40it/s]
1|6212|Loss: 0.8499609231948853:  96%|█████████▌| 6211/6470 [1:20:58<03:05,  1.40it/s]
1|6212|Loss: 0.8499609231948853:  96%|█████████▌| 6212/6470 [1:20:58<03:00,  1.43it/s]
1|6213|Loss: 0.7104197144508362:  96%|█████████▌| 6212/6470 [1:20:58<03:00,  1.43it/s]
1|6213|Loss: 0.7104197144508362:  96%|█████████▌| 6213/6470 [1:20:59<03:13,  1.33it/s]
1|6214|Loss: 0.8564960956573486:  96%|█████████▌| 6213/6470 [1:20:59<03:13,  1.33it/s]
1|6214|Loss: 0.8564960956573486:  96%|█████████▌| 6214/6470 [1:20:59<02:59,  1.43it/s]
1|6215|Loss: 0.9601337313652039:  96%|█████████▌| 6214/6470 [1:21:00<02:59,  1.43it/s]
1|6215|Loss: 0.9601337313652039:  96%|█████████▌| 6215/6470 [1:21:00<02:54,  1.46it/s]
1|6216|Loss: 0.9411094188690186:  96%|█████████▌| 6215/6470 [1:21:00<02:54,  1.46it/s]
1|6216|Loss: 0.9411094188690186:  96%|█████████▌| 6216/6470 [1:21:01<03:08,  1.35it/s]
1|6217|Loss: 0.9040451645851135:  96%|█████████▌| 6216/6470 [1:21:01<03:08,  1.35it/s]
1|6217|Loss: 0.9040451645851135:  96%|█████████▌| 6217/6470 [1:21:02<03:17,  1.28it/s]
1|6218|Loss: 0.9270201921463013:  96%|█████████▌| 6217/6470 [1:21:02<03:17,  1.28it/s]
1|6218|Loss: 0.9270201921463013:  96%|█████████▌| 6218/6470 [1:21:03<03:29,  1.20it/s]
1|6219|Loss: 0.8409244418144226:  96%|█████████▌| 6218/6470 [1:21:03<03:29,  1.20it/s]
1|6219|Loss: 0.8409244418144226:  96%|█████████▌| 6219/6470 [1:21:04<03:29,  1.20it/s]
1|6220|Loss: 0.8750547170639038:  96%|█████████▌| 6219/6470 [1:21:04<03:29,  1.20it/s]
1|6220|Loss: 0.8750547170639038:  96%|█████████▌| 6220/6470 [1:21:04<03:34,  1.17it/s]
1|6221|Loss: 0.6912496089935303:  96%|█████████▌| 6220/6470 [1:21:05<03:34,  1.17it/s]
1|6221|Loss: 0.6912496089935303:  96%|█████████▌| 6221/6470 [1:21:05<03:36,  1.15it/s]
1|6222|Loss: 0.9582141637802124:  96%|█████████▌| 6221/6470 [1:21:06<03:36,  1.15it/s]
1|6222|Loss: 0.9582141637802124:  96%|█████████▌| 6222/6470 [1:21:06<03:25,  1.21it/s]
1|6223|Loss: 0.8924494981765747:  96%|█████████▌| 6222/6470 [1:21:06<03:25,  1.21it/s]
1|6223|Loss: 0.8924494981765747:  96%|█████████▌| 6223/6470 [1:21:07<03:10,  1.29it/s]
1|6224|Loss: 0.8634909391403198:  96%|█████████▌| 6223/6470 [1:21:07<03:10,  1.29it/s]
1|6224|Loss: 0.8634909391403198:  96%|█████████▌| 6224/6470 [1:21:08<03:20,  1.23it/s]
1|6225|Loss: 0.8323058485984802:  96%|█████████▌| 6224/6470 [1:21:08<03:20,  1.23it/s]
1|6225|Loss: 0.8323058485984802:  96%|█████████▌| 6225/6470 [1:21:08<03:14,  1.26it/s]
1|6226|Loss: 0.7523061037063599:  96%|█████████▌| 6225/6470 [1:21:09<03:14,  1.26it/s]
1|6226|Loss: 0.7523061037063599:  96%|█████████▌| 6226/6470 [1:21:09<03:20,  1.22it/s]
1|6227|Loss: 1.0280002355575562:  96%|█████████▌| 6226/6470 [1:21:10<03:20,  1.22it/s]
1|6227|Loss: 1.0280002355575562:  96%|█████████▌| 6227/6470 [1:21:10<03:32,  1.14it/s]
1|6228|Loss: 0.9284749031066895:  96%|█████████▌| 6227/6470 [1:21:11<03:32,  1.14it/s]
1|6228|Loss: 0.9284749031066895:  96%|█████████▋| 6228/6470 [1:21:11<03:29,  1.16it/s]
1|6229|Loss: 0.8827570676803589:  96%|█████████▋| 6228/6470 [1:21:12<03:29,  1.16it/s]
1|6229|Loss: 0.8827570676803589:  96%|█████████▋| 6229/6470 [1:21:12<03:24,  1.18it/s]
1|6230|Loss: 0.8480127453804016:  96%|█████████▋| 6229/6470 [1:21:12<03:24,  1.18it/s]
1|6230|Loss: 0.8480127453804016:  96%|█████████▋| 6230/6470 [1:21:13<03:09,  1.27it/s]
1|6231|Loss: 0.6343719959259033:  96%|█████████▋| 6230/6470 [1:21:13<03:09,  1.27it/s]
1|6231|Loss: 0.6343719959259033:  96%|█████████▋| 6231/6470 [1:21:13<03:06,  1.28it/s]
1|6232|Loss: 0.9650198817253113:  96%|█████████▋| 6231/6470 [1:21:14<03:06,  1.28it/s]
1|6232|Loss: 0.9650198817253113:  96%|█████████▋| 6232/6470 [1:21:14<02:57,  1.34it/s]
1|6233|Loss: 0.7463557124137878:  96%|█████████▋| 6232/6470 [1:21:14<02:57,  1.34it/s]
1|6233|Loss: 0.7463557124137878:  96%|█████████▋| 6233/6470 [1:21:15<02:51,  1.38it/s]
1|6234|Loss: 0.8884797096252441:  96%|█████████▋| 6233/6470 [1:21:15<02:51,  1.38it/s]
1|6234|Loss: 0.8884797096252441:  96%|█████████▋| 6234/6470 [1:21:15<02:52,  1.37it/s]
1|6235|Loss: 0.7668923139572144:  96%|█████████▋| 6234/6470 [1:21:16<02:52,  1.37it/s]
1|6235|Loss: 0.7668923139572144:  96%|█████████▋| 6235/6470 [1:21:16<02:56,  1.33it/s]
1|6236|Loss: 0.9118770360946655:  96%|█████████▋| 6235/6470 [1:21:17<02:56,  1.33it/s]
1|6236|Loss: 0.9118770360946655:  96%|█████████▋| 6236/6470 [1:21:17<02:57,  1.32it/s]
1|6237|Loss: 0.982734203338623:  96%|█████████▋| 6236/6470 [1:21:17<02:57,  1.32it/s] 
1|6237|Loss: 0.982734203338623:  96%|█████████▋| 6237/6470 [1:21:18<03:00,  1.29it/s]
1|6238|Loss: 0.76131272315979:  96%|█████████▋| 6237/6470 [1:21:18<03:00,  1.29it/s] 
1|6238|Loss: 0.76131272315979:  96%|█████████▋| 6238/6470 [1:21:18<02:52,  1.35it/s]
1|6239|Loss: 0.8551524877548218:  96%|█████████▋| 6238/6470 [1:21:19<02:52,  1.35it/s]
1|6239|Loss: 0.8551524877548218:  96%|█████████▋| 6239/6470 [1:21:19<02:47,  1.38it/s]
1|6240|Loss: 0.7502575516700745:  96%|█████████▋| 6239/6470 [1:21:20<02:47,  1.38it/s]
1|6240|Loss: 0.7502575516700745:  96%|█████████▋| 6240/6470 [1:21:20<02:50,  1.35it/s]
1|6241|Loss: 0.729079008102417:  96%|█████████▋| 6240/6470 [1:21:20<02:50,  1.35it/s] 
1|6241|Loss: 0.729079008102417:  96%|█████████▋| 6241/6470 [1:21:21<02:57,  1.29it/s]
1|6242|Loss: 0.865292489528656:  96%|█████████▋| 6241/6470 [1:21:21<02:57,  1.29it/s]
1|6242|Loss: 0.865292489528656:  96%|█████████▋| 6242/6470 [1:21:22<02:58,  1.27it/s]
1|6243|Loss: 0.9308212399482727:  96%|█████████▋| 6242/6470 [1:21:22<02:58,  1.27it/s]
1|6243|Loss: 0.9308212399482727:  96%|█████████▋| 6243/6470 [1:21:22<02:59,  1.26it/s]
1|6244|Loss: 0.758152961730957:  96%|█████████▋| 6243/6470 [1:21:23<02:59,  1.26it/s] 
1|6244|Loss: 0.758152961730957:  97%|█████████▋| 6244/6470 [1:21:23<02:50,  1.33it/s]
1|6245|Loss: 0.835261881351471:  97%|█████████▋| 6244/6470 [1:21:23<02:50,  1.33it/s]
1|6245|Loss: 0.835261881351471:  97%|█████████▋| 6245/6470 [1:21:24<02:48,  1.34it/s]
1|6246|Loss: 0.8102425932884216:  97%|█████████▋| 6245/6470 [1:21:24<02:48,  1.34it/s]
1|6246|Loss: 0.8102425932884216:  97%|█████████▋| 6246/6470 [1:21:25<02:48,  1.33it/s]
1|6247|Loss: 1.015748381614685:  97%|█████████▋| 6246/6470 [1:21:25<02:48,  1.33it/s] 
1|6247|Loss: 1.015748381614685:  97%|█████████▋| 6247/6470 [1:21:25<02:49,  1.31it/s]
1|6248|Loss: 0.8798350691795349:  97%|█████████▋| 6247/6470 [1:21:26<02:49,  1.31it/s]
1|6248|Loss: 0.8798350691795349:  97%|█████████▋| 6248/6470 [1:21:26<02:50,  1.30it/s]
1|6249|Loss: 0.8576912879943848:  97%|█████████▋| 6248/6470 [1:21:27<02:50,  1.30it/s]
1|6249|Loss: 0.8576912879943848:  97%|█████████▋| 6249/6470 [1:21:27<02:52,  1.28it/s]
1|6250|Loss: 0.7746590971946716:  97%|█████████▋| 6249/6470 [1:21:27<02:52,  1.28it/s]
1|6250|Loss: 0.7746590971946716:  97%|█████████▋| 6250/6470 [1:21:28<02:56,  1.24it/s]
1|6251|Loss: 0.8055184483528137:  97%|█████████▋| 6250/6470 [1:21:28<02:56,  1.24it/s]
1|6251|Loss: 0.8055184483528137:  97%|█████████▋| 6251/6470 [1:21:28<02:41,  1.35it/s]
1|6252|Loss: 0.8826130628585815:  97%|█████████▋| 6251/6470 [1:21:29<02:41,  1.35it/s]
1|6252|Loss: 0.8826130628585815:  97%|█████████▋| 6252/6470 [1:21:29<02:34,  1.41it/s]
1|6253|Loss: 0.8555240631103516:  97%|█████████▋| 6252/6470 [1:21:29<02:34,  1.41it/s]
1|6253|Loss: 0.8555240631103516:  97%|█████████▋| 6253/6470 [1:21:30<02:26,  1.48it/s]
1|6254|Loss: 0.8138952255249023:  97%|█████████▋| 6253/6470 [1:21:30<02:26,  1.48it/s]
1|6254|Loss: 0.8138952255249023:  97%|█████████▋| 6254/6470 [1:21:30<02:30,  1.43it/s]
1|6255|Loss: 0.9109194278717041:  97%|█████████▋| 6254/6470 [1:21:31<02:30,  1.43it/s]
1|6255|Loss: 0.9109194278717041:  97%|█████████▋| 6255/6470 [1:21:31<02:42,  1.32it/s]
1|6256|Loss: 0.7992584109306335:  97%|█████████▋| 6255/6470 [1:21:32<02:42,  1.32it/s]
1|6256|Loss: 0.7992584109306335:  97%|█████████▋| 6256/6470 [1:21:32<02:49,  1.26it/s]
1|6257|Loss: 1.0181940793991089:  97%|█████████▋| 6256/6470 [1:21:33<02:49,  1.26it/s]
1|6257|Loss: 1.0181940793991089:  97%|█████████▋| 6257/6470 [1:21:33<02:44,  1.30it/s]
1|6258|Loss: 0.8010616898536682:  97%|█████████▋| 6257/6470 [1:21:33<02:44,  1.30it/s]
1|6258|Loss: 0.8010616898536682:  97%|█████████▋| 6258/6470 [1:21:34<02:42,  1.30it/s]
1|6259|Loss: 1.076310396194458:  97%|█████████▋| 6258/6470 [1:21:34<02:42,  1.30it/s] 
1|6259|Loss: 1.076310396194458:  97%|█████████▋| 6259/6470 [1:21:34<02:38,  1.33it/s]
1|6260|Loss: 0.7819383144378662:  97%|█████████▋| 6259/6470 [1:21:35<02:38,  1.33it/s]
1|6260|Loss: 0.7819383144378662:  97%|█████████▋| 6260/6470 [1:21:35<02:39,  1.31it/s]
1|6261|Loss: 0.8162890672683716:  97%|█████████▋| 6260/6470 [1:21:36<02:39,  1.31it/s]
1|6261|Loss: 0.8162890672683716:  97%|█████████▋| 6261/6470 [1:21:36<02:40,  1.31it/s]
1|6262|Loss: 0.8074537515640259:  97%|█████████▋| 6261/6470 [1:21:36<02:40,  1.31it/s]
1|6262|Loss: 0.8074537515640259:  97%|█████████▋| 6262/6470 [1:21:36<02:29,  1.39it/s]
1|6263|Loss: 0.8859419822692871:  97%|█████████▋| 6262/6470 [1:21:37<02:29,  1.39it/s]
1|6263|Loss: 0.8859419822692871:  97%|█████████▋| 6263/6470 [1:21:37<02:36,  1.32it/s]
1|6264|Loss: 0.8427043557167053:  97%|█████████▋| 6263/6470 [1:21:38<02:36,  1.32it/s]
1|6264|Loss: 0.8427043557167053:  97%|█████████▋| 6264/6470 [1:21:38<02:39,  1.29it/s]
1|6265|Loss: 0.8020325899124146:  97%|█████████▋| 6264/6470 [1:21:39<02:39,  1.29it/s]
1|6265|Loss: 0.8020325899124146:  97%|█████████▋| 6265/6470 [1:21:39<02:43,  1.25it/s]
1|6266|Loss: 0.9772529602050781:  97%|█████████▋| 6265/6470 [1:21:39<02:43,  1.25it/s]
1|6266|Loss: 0.9772529602050781:  97%|█████████▋| 6266/6470 [1:21:40<02:41,  1.26it/s]
1|6267|Loss: 0.8700456619262695:  97%|█████████▋| 6266/6470 [1:21:40<02:41,  1.26it/s]
1|6267|Loss: 0.8700456619262695:  97%|█████████▋| 6267/6470 [1:21:41<02:46,  1.22it/s]
1|6268|Loss: 0.9047528505325317:  97%|█████████▋| 6267/6470 [1:21:41<02:46,  1.22it/s]
1|6268|Loss: 0.9047528505325317:  97%|█████████▋| 6268/6470 [1:21:42<02:48,  1.20it/s]
1|6269|Loss: 0.7559506297111511:  97%|█████████▋| 6268/6470 [1:21:42<02:48,  1.20it/s]
1|6269|Loss: 0.7559506297111511:  97%|█████████▋| 6269/6470 [1:21:42<02:54,  1.16it/s]
1|6270|Loss: 0.9946438074111938:  97%|█████████▋| 6269/6470 [1:21:43<02:54,  1.16it/s]
1|6270|Loss: 0.9946438074111938:  97%|█████████▋| 6270/6470 [1:21:43<02:46,  1.20it/s]
1|6271|Loss: 0.8414450883865356:  97%|█████████▋| 6270/6470 [1:21:44<02:46,  1.20it/s]
1|6271|Loss: 0.8414450883865356:  97%|█████████▋| 6271/6470 [1:21:44<02:41,  1.23it/s]
1|6272|Loss: 0.9874270558357239:  97%|█████████▋| 6271/6470 [1:21:44<02:41,  1.23it/s]
1|6272|Loss: 0.9874270558357239:  97%|█████████▋| 6272/6470 [1:21:45<02:39,  1.24it/s]
1|6273|Loss: 0.8850004076957703:  97%|█████████▋| 6272/6470 [1:21:45<02:39,  1.24it/s]
1|6273|Loss: 0.8850004076957703:  97%|█████████▋| 6273/6470 [1:21:45<02:27,  1.33it/s]
1|6274|Loss: 0.8374622464179993:  97%|█████████▋| 6273/6470 [1:21:46<02:27,  1.33it/s]
1|6274|Loss: 0.8374622464179993:  97%|█████████▋| 6274/6470 [1:21:46<02:22,  1.38it/s]
1|6275|Loss: 0.9754652976989746:  97%|█████████▋| 6274/6470 [1:21:46<02:22,  1.38it/s]
1|6275|Loss: 0.9754652976989746:  97%|█████████▋| 6275/6470 [1:21:47<02:17,  1.42it/s]
1|6276|Loss: 0.8649686574935913:  97%|█████████▋| 6275/6470 [1:21:47<02:17,  1.42it/s]
1|6276|Loss: 0.8649686574935913:  97%|█████████▋| 6276/6470 [1:21:47<02:18,  1.40it/s]
1|6277|Loss: 0.8381898999214172:  97%|█████████▋| 6276/6470 [1:21:48<02:18,  1.40it/s]
1|6277|Loss: 0.8381898999214172:  97%|█████████▋| 6277/6470 [1:21:48<02:22,  1.36it/s]
1|6278|Loss: 0.836056649684906:  97%|█████████▋| 6277/6470 [1:21:49<02:22,  1.36it/s] 
1|6278|Loss: 0.836056649684906:  97%|█████████▋| 6278/6470 [1:21:49<02:20,  1.37it/s]
1|6279|Loss: 0.8917171955108643:  97%|█████████▋| 6278/6470 [1:21:49<02:20,  1.37it/s]
1|6279|Loss: 0.8917171955108643:  97%|█████████▋| 6279/6470 [1:21:50<02:23,  1.33it/s]
1|6280|Loss: 0.8317588567733765:  97%|█████████▋| 6279/6470 [1:21:50<02:23,  1.33it/s]
1|6280|Loss: 0.8317588567733765:  97%|█████████▋| 6280/6470 [1:21:51<02:24,  1.32it/s]
1|6281|Loss: 0.9274888634681702:  97%|█████████▋| 6280/6470 [1:21:51<02:24,  1.32it/s]
1|6281|Loss: 0.9274888634681702:  97%|█████████▋| 6281/6470 [1:21:51<02:25,  1.29it/s]
1|6282|Loss: 0.796438992023468:  97%|█████████▋| 6281/6470 [1:21:52<02:25,  1.29it/s] 
1|6282|Loss: 0.796438992023468:  97%|█████████▋| 6282/6470 [1:21:52<02:22,  1.32it/s]
1|6283|Loss: 0.8885326981544495:  97%|█████████▋| 6282/6470 [1:21:52<02:22,  1.32it/s]
1|6283|Loss: 0.8885326981544495:  97%|█████████▋| 6283/6470 [1:21:53<02:17,  1.36it/s]
1|6284|Loss: 0.8902689814567566:  97%|█████████▋| 6283/6470 [1:21:53<02:17,  1.36it/s]
1|6284|Loss: 0.8902689814567566:  97%|█████████▋| 6284/6470 [1:21:54<02:18,  1.34it/s]
1|6285|Loss: 0.886053204536438:  97%|█████████▋| 6284/6470 [1:21:54<02:18,  1.34it/s] 
1|6285|Loss: 0.886053204536438:  97%|█████████▋| 6285/6470 [1:21:54<02:27,  1.26it/s]
1|6286|Loss: 0.8645914793014526:  97%|█████████▋| 6285/6470 [1:21:55<02:27,  1.26it/s]
1|6286|Loss: 0.8645914793014526:  97%|█████████▋| 6286/6470 [1:21:55<02:32,  1.20it/s]
1|6287|Loss: 0.8895246386528015:  97%|█████████▋| 6286/6470 [1:21:56<02:32,  1.20it/s]
1|6287|Loss: 0.8895246386528015:  97%|█████████▋| 6287/6470 [1:21:56<02:29,  1.22it/s]
1|6288|Loss: 0.8036554455757141:  97%|█████████▋| 6287/6470 [1:21:57<02:29,  1.22it/s]
1|6288|Loss: 0.8036554455757141:  97%|█████████▋| 6288/6470 [1:21:57<02:23,  1.27it/s]
1|6289|Loss: 0.622826337814331:  97%|█████████▋| 6288/6470 [1:21:57<02:23,  1.27it/s] 
1|6289|Loss: 0.622826337814331:  97%|█████████▋| 6289/6470 [1:21:58<02:16,  1.32it/s]
1|6290|Loss: 0.9237774014472961:  97%|█████████▋| 6289/6470 [1:21:58<02:16,  1.32it/s]
1|6290|Loss: 0.9237774014472961:  97%|█████████▋| 6290/6470 [1:21:58<02:15,  1.32it/s]
1|6291|Loss: 0.8270179629325867:  97%|█████████▋| 6290/6470 [1:21:59<02:15,  1.32it/s]
1|6291|Loss: 0.8270179629325867:  97%|█████████▋| 6291/6470 [1:21:59<02:10,  1.37it/s]
1|6292|Loss: 0.9728547930717468:  97%|█████████▋| 6291/6470 [1:21:59<02:10,  1.37it/s]
1|6292|Loss: 0.9728547930717468:  97%|█████████▋| 6292/6470 [1:22:00<02:12,  1.35it/s]
1|6293|Loss: 0.845303475856781:  97%|█████████▋| 6292/6470 [1:22:00<02:12,  1.35it/s] 
1|6293|Loss: 0.845303475856781:  97%|█████████▋| 6293/6470 [1:22:01<02:14,  1.31it/s]
1|6294|Loss: 0.9714971780776978:  97%|█████████▋| 6293/6470 [1:22:01<02:14,  1.31it/s]
1|6294|Loss: 0.9714971780776978:  97%|█████████▋| 6294/6470 [1:22:01<02:05,  1.40it/s]
1|6295|Loss: 0.7970297336578369:  97%|█████████▋| 6294/6470 [1:22:02<02:05,  1.40it/s]
1|6295|Loss: 0.7970297336578369:  97%|█████████▋| 6295/6470 [1:22:02<02:04,  1.40it/s]
1|6296|Loss: 0.817176103591919:  97%|█████████▋| 6295/6470 [1:22:02<02:04,  1.40it/s] 
1|6296|Loss: 0.817176103591919:  97%|█████████▋| 6296/6470 [1:22:02<01:57,  1.48it/s]
1|6297|Loss: 0.8532562851905823:  97%|█████████▋| 6296/6470 [1:22:03<01:57,  1.48it/s]
1|6297|Loss: 0.8532562851905823:  97%|█████████▋| 6297/6470 [1:22:03<01:58,  1.46it/s]
1|6298|Loss: 0.9740266799926758:  97%|█████████▋| 6297/6470 [1:22:04<01:58,  1.46it/s]
1|6298|Loss: 0.9740266799926758:  97%|█████████▋| 6298/6470 [1:22:04<02:09,  1.32it/s]
1|6299|Loss: 0.9407021403312683:  97%|█████████▋| 6298/6470 [1:22:05<02:09,  1.32it/s]
1|6299|Loss: 0.9407021403312683:  97%|█████████▋| 6299/6470 [1:22:05<02:14,  1.27it/s]
1|6300|Loss: 0.8499532341957092:  97%|█████████▋| 6299/6470 [1:22:05<02:14,  1.27it/s]
1|6300|Loss: 0.8499532341957092:  97%|█████████▋| 6300/6470 [1:22:06<02:04,  1.37it/s]
1|6301|Loss: 0.9406693577766418:  97%|█████████▋| 6300/6470 [1:22:06<02:04,  1.37it/s]
1|6301|Loss: 0.9406693577766418:  97%|█████████▋| 6301/6470 [1:22:06<01:58,  1.43it/s]
1|6302|Loss: 0.8932721614837646:  97%|█████████▋| 6301/6470 [1:22:07<01:58,  1.43it/s]
1|6302|Loss: 0.8932721614837646:  97%|█████████▋| 6302/6470 [1:22:07<01:56,  1.44it/s]
1|6303|Loss: 0.8484019041061401:  97%|█████████▋| 6302/6470 [1:22:07<01:56,  1.44it/s]
1|6303|Loss: 0.8484019041061401:  97%|█████████▋| 6303/6470 [1:22:08<01:56,  1.44it/s]
1|6304|Loss: 0.9460430145263672:  97%|█████████▋| 6303/6470 [1:22:08<01:56,  1.44it/s]
1|6304|Loss: 0.9460430145263672:  97%|█████████▋| 6304/6470 [1:22:08<01:54,  1.45it/s]
1|6305|Loss: 0.8912299275398254:  97%|█████████▋| 6304/6470 [1:22:09<01:54,  1.45it/s]
1|6305|Loss: 0.8912299275398254:  97%|█████████▋| 6305/6470 [1:22:09<02:00,  1.37it/s]
1|6306|Loss: 0.9265320301055908:  97%|█████████▋| 6305/6470 [1:22:09<02:00,  1.37it/s]
1|6306|Loss: 0.9265320301055908:  97%|█████████▋| 6306/6470 [1:22:10<01:54,  1.44it/s]
1|6307|Loss: 0.8516229391098022:  97%|█████████▋| 6306/6470 [1:22:10<01:54,  1.44it/s]
1|6307|Loss: 0.8516229391098022:  97%|█████████▋| 6307/6470 [1:22:10<01:53,  1.43it/s]
1|6308|Loss: 0.7892206311225891:  97%|█████████▋| 6307/6470 [1:22:11<01:53,  1.43it/s]
1|6308|Loss: 0.7892206311225891:  97%|█████████▋| 6308/6470 [1:22:11<01:57,  1.38it/s]
1|6309|Loss: 0.8187545537948608:  97%|█████████▋| 6308/6470 [1:22:12<01:57,  1.38it/s]
1|6309|Loss: 0.8187545537948608:  98%|█████████▊| 6309/6470 [1:22:12<01:55,  1.39it/s]
1|6310|Loss: 0.9474856853485107:  98%|█████████▊| 6309/6470 [1:22:12<01:55,  1.39it/s]
1|6310|Loss: 0.9474856853485107:  98%|█████████▊| 6310/6470 [1:22:13<01:55,  1.38it/s]
1|6311|Loss: 0.8526572585105896:  98%|█████████▊| 6310/6470 [1:22:13<01:55,  1.38it/s]
1|6311|Loss: 0.8526572585105896:  98%|█████████▊| 6311/6470 [1:22:13<01:53,  1.40it/s]
1|6312|Loss: 0.8611273765563965:  98%|█████████▊| 6311/6470 [1:22:14<01:53,  1.40it/s]
1|6312|Loss: 0.8611273765563965:  98%|█████████▊| 6312/6470 [1:22:14<01:52,  1.41it/s]
1|6313|Loss: 0.6110985279083252:  98%|█████████▊| 6312/6470 [1:22:14<01:52,  1.41it/s]
1|6313|Loss: 0.6110985279083252:  98%|█████████▊| 6313/6470 [1:22:15<01:45,  1.48it/s]
1|6314|Loss: 0.852268397808075:  98%|█████████▊| 6313/6470 [1:22:15<01:45,  1.48it/s] 
1|6314|Loss: 0.852268397808075:  98%|█████████▊| 6314/6470 [1:22:15<01:53,  1.37it/s]
1|6315|Loss: 0.8565444946289062:  98%|█████████▊| 6314/6470 [1:22:16<01:53,  1.37it/s]
1|6315|Loss: 0.8565444946289062:  98%|█████████▊| 6315/6470 [1:22:16<01:56,  1.33it/s]
1|6316|Loss: 0.7490400075912476:  98%|█████████▊| 6315/6470 [1:22:17<01:56,  1.33it/s]
1|6316|Loss: 0.7490400075912476:  98%|█████████▊| 6316/6470 [1:22:17<01:58,  1.30it/s]
1|6317|Loss: 0.8998432159423828:  98%|█████████▊| 6316/6470 [1:22:17<01:58,  1.30it/s]
1|6317|Loss: 0.8998432159423828:  98%|█████████▊| 6317/6470 [1:22:18<01:51,  1.37it/s]
1|6318|Loss: 0.7920019626617432:  98%|█████████▊| 6317/6470 [1:22:18<01:51,  1.37it/s]
1|6318|Loss: 0.7920019626617432:  98%|█████████▊| 6318/6470 [1:22:18<01:48,  1.40it/s]
1|6319|Loss: 0.7996819615364075:  98%|█████████▊| 6318/6470 [1:22:19<01:48,  1.40it/s]
1|6319|Loss: 0.7996819615364075:  98%|█████████▊| 6319/6470 [1:22:19<01:55,  1.30it/s]
1|6320|Loss: 0.8270958065986633:  98%|█████████▊| 6319/6470 [1:22:20<01:55,  1.30it/s]
1|6320|Loss: 0.8270958065986633:  98%|█████████▊| 6320/6470 [1:22:20<01:53,  1.33it/s]
1|6321|Loss: 0.7039879560470581:  98%|█████████▊| 6320/6470 [1:22:20<01:53,  1.33it/s]
1|6321|Loss: 0.7039879560470581:  98%|█████████▊| 6321/6470 [1:22:21<01:46,  1.40it/s]
1|6322|Loss: 0.9221852421760559:  98%|█████████▊| 6321/6470 [1:22:21<01:46,  1.40it/s]
1|6322|Loss: 0.9221852421760559:  98%|█████████▊| 6322/6470 [1:22:21<01:48,  1.37it/s]
1|6323|Loss: 0.8473677635192871:  98%|█████████▊| 6322/6470 [1:22:22<01:48,  1.37it/s]
1|6323|Loss: 0.8473677635192871:  98%|█████████▊| 6323/6470 [1:22:22<01:46,  1.38it/s]
1|6324|Loss: 0.9026070833206177:  98%|█████████▊| 6323/6470 [1:22:22<01:46,  1.38it/s]
1|6324|Loss: 0.9026070833206177:  98%|█████████▊| 6324/6470 [1:22:23<01:47,  1.35it/s]
1|6325|Loss: 0.8496434092521667:  98%|█████████▊| 6324/6470 [1:22:23<01:47,  1.35it/s]
1|6325|Loss: 0.8496434092521667:  98%|█████████▊| 6325/6470 [1:22:24<01:49,  1.32it/s]
1|6326|Loss: 0.7597365379333496:  98%|█████████▊| 6325/6470 [1:22:24<01:49,  1.32it/s]
1|6326|Loss: 0.7597365379333496:  98%|█████████▊| 6326/6470 [1:22:24<01:50,  1.31it/s]
1|6327|Loss: 0.8140220642089844:  98%|█████████▊| 6326/6470 [1:22:25<01:50,  1.31it/s]
1|6327|Loss: 0.8140220642089844:  98%|█████████▊| 6327/6470 [1:22:25<01:50,  1.29it/s]
1|6328|Loss: 0.7691590189933777:  98%|█████████▊| 6327/6470 [1:22:26<01:50,  1.29it/s]
1|6328|Loss: 0.7691590189933777:  98%|█████████▊| 6328/6470 [1:22:26<01:51,  1.28it/s]
1|6329|Loss: 0.9564577341079712:  98%|█████████▊| 6328/6470 [1:22:26<01:51,  1.28it/s]
1|6329|Loss: 0.9564577341079712:  98%|█████████▊| 6329/6470 [1:22:27<01:46,  1.32it/s]
1|6330|Loss: 0.9428356885910034:  98%|█████████▊| 6329/6470 [1:22:27<01:46,  1.32it/s]
1|6330|Loss: 0.9428356885910034:  98%|█████████▊| 6330/6470 [1:22:27<01:43,  1.35it/s]
1|6331|Loss: 0.9733328819274902:  98%|█████████▊| 6330/6470 [1:22:28<01:43,  1.35it/s]
1|6331|Loss: 0.9733328819274902:  98%|█████████▊| 6331/6470 [1:22:28<01:44,  1.33it/s]
1|6332|Loss: 0.7836126685142517:  98%|█████████▊| 6331/6470 [1:22:29<01:44,  1.33it/s]
1|6332|Loss: 0.7836126685142517:  98%|█████████▊| 6332/6470 [1:22:29<01:42,  1.34it/s]
1|6333|Loss: 0.6517758965492249:  98%|█████████▊| 6332/6470 [1:22:29<01:42,  1.34it/s]
1|6333|Loss: 0.6517758965492249:  98%|█████████▊| 6333/6470 [1:22:30<01:40,  1.37it/s]
1|6334|Loss: 0.7604163885116577:  98%|█████████▊| 6333/6470 [1:22:30<01:40,  1.37it/s]
1|6334|Loss: 0.7604163885116577:  98%|█████████▊| 6334/6470 [1:22:30<01:32,  1.47it/s]
1|6335|Loss: 0.9005928635597229:  98%|█████████▊| 6334/6470 [1:22:31<01:32,  1.47it/s]
1|6335|Loss: 0.9005928635597229:  98%|█████████▊| 6335/6470 [1:22:31<01:35,  1.42it/s]
1|6336|Loss: 0.7115620374679565:  98%|█████████▊| 6335/6470 [1:22:31<01:35,  1.42it/s]
1|6336|Loss: 0.7115620374679565:  98%|█████████▊| 6336/6470 [1:22:32<01:39,  1.35it/s]
1|6337|Loss: 0.7408416271209717:  98%|█████████▊| 6336/6470 [1:22:32<01:39,  1.35it/s]
1|6337|Loss: 0.7408416271209717:  98%|█████████▊| 6337/6470 [1:22:32<01:37,  1.37it/s]
1|6338|Loss: 0.862628161907196:  98%|█████████▊| 6337/6470 [1:22:33<01:37,  1.37it/s] 
1|6338|Loss: 0.862628161907196:  98%|█████████▊| 6338/6470 [1:22:33<01:38,  1.34it/s]
1|6339|Loss: 0.8985539078712463:  98%|█████████▊| 6338/6470 [1:22:34<01:38,  1.34it/s]
1|6339|Loss: 0.8985539078712463:  98%|█████████▊| 6339/6470 [1:22:34<01:34,  1.38it/s]
1|6340|Loss: 0.9844424724578857:  98%|█████████▊| 6339/6470 [1:22:34<01:34,  1.38it/s]
1|6340|Loss: 0.9844424724578857:  98%|█████████▊| 6340/6470 [1:22:35<01:37,  1.33it/s]
1|6341|Loss: 0.8349851965904236:  98%|█████████▊| 6340/6470 [1:22:35<01:37,  1.33it/s]
1|6341|Loss: 0.8349851965904236:  98%|█████████▊| 6341/6470 [1:22:36<01:41,  1.27it/s]
1|6342|Loss: 0.8757278323173523:  98%|█████████▊| 6341/6470 [1:22:36<01:41,  1.27it/s]
1|6342|Loss: 0.8757278323173523:  98%|█████████▊| 6342/6470 [1:22:36<01:32,  1.38it/s]
1|6343|Loss: 0.6548296213150024:  98%|█████████▊| 6342/6470 [1:22:37<01:32,  1.38it/s]
1|6343|Loss: 0.6548296213150024:  98%|█████████▊| 6343/6470 [1:22:37<01:43,  1.23it/s]
1|6344|Loss: 0.6820057034492493:  98%|█████████▊| 6343/6470 [1:22:38<01:43,  1.23it/s]
1|6344|Loss: 0.6820057034492493:  98%|█████████▊| 6344/6470 [1:22:38<01:48,  1.16it/s]
1|6345|Loss: 0.8921554088592529:  98%|█████████▊| 6344/6470 [1:22:39<01:48,  1.16it/s]
1|6345|Loss: 0.8921554088592529:  98%|█████████▊| 6345/6470 [1:22:39<01:50,  1.13it/s]
1|6346|Loss: 1.017282247543335:  98%|█████████▊| 6345/6470 [1:22:40<01:50,  1.13it/s] 
1|6346|Loss: 1.017282247543335:  98%|█████████▊| 6346/6470 [1:22:40<01:53,  1.09it/s]
1|6347|Loss: 0.7720613479614258:  98%|█████████▊| 6346/6470 [1:22:41<01:53,  1.09it/s]
1|6347|Loss: 0.7720613479614258:  98%|█████████▊| 6347/6470 [1:22:41<01:50,  1.11it/s]
1|6348|Loss: 0.83969646692276:  98%|█████████▊| 6347/6470 [1:22:41<01:50,  1.11it/s]  
1|6348|Loss: 0.83969646692276:  98%|█████████▊| 6348/6470 [1:22:42<01:48,  1.12it/s]
1|6349|Loss: 0.8308721780776978:  98%|█████████▊| 6348/6470 [1:22:42<01:48,  1.12it/s]
1|6349|Loss: 0.8308721780776978:  98%|█████████▊| 6349/6470 [1:22:43<01:41,  1.19it/s]
1|6350|Loss: 0.8089070916175842:  98%|█████████▊| 6349/6470 [1:22:43<01:41,  1.19it/s]
1|6350|Loss: 0.8089070916175842:  98%|█████████▊| 6350/6470 [1:22:43<01:40,  1.19it/s]
1|6351|Loss: 1.0271086692810059:  98%|█████████▊| 6350/6470 [1:22:44<01:40,  1.19it/s]
1|6351|Loss: 1.0271086692810059:  98%|█████████▊| 6351/6470 [1:22:44<01:35,  1.25it/s]
1|6352|Loss: 0.8699766993522644:  98%|█████████▊| 6351/6470 [1:22:44<01:35,  1.25it/s]
1|6352|Loss: 0.8699766993522644:  98%|█████████▊| 6352/6470 [1:22:45<01:29,  1.32it/s]
1|6353|Loss: 1.0297603607177734:  98%|█████████▊| 6352/6470 [1:22:45<01:29,  1.32it/s]
1|6353|Loss: 1.0297603607177734:  98%|█████████▊| 6353/6470 [1:22:46<01:33,  1.24it/s]
1|6354|Loss: 0.9560773372650146:  98%|█████████▊| 6353/6470 [1:22:46<01:33,  1.24it/s]
1|6354|Loss: 0.9560773372650146:  98%|█████████▊| 6354/6470 [1:22:47<01:34,  1.23it/s]
1|6355|Loss: 0.956399142742157:  98%|█████████▊| 6354/6470 [1:22:47<01:34,  1.23it/s] 
1|6355|Loss: 0.956399142742157:  98%|█████████▊| 6355/6470 [1:22:47<01:32,  1.24it/s]
1|6356|Loss: 0.6802344918251038:  98%|█████████▊| 6355/6470 [1:22:48<01:32,  1.24it/s]
1|6356|Loss: 0.6802344918251038:  98%|█████████▊| 6356/6470 [1:22:48<01:26,  1.31it/s]
1|6357|Loss: 0.7659482955932617:  98%|█████████▊| 6356/6470 [1:22:48<01:26,  1.31it/s]
1|6357|Loss: 0.7659482955932617:  98%|█████████▊| 6357/6470 [1:22:49<01:25,  1.32it/s]
1|6358|Loss: 0.8521436452865601:  98%|█████████▊| 6357/6470 [1:22:49<01:25,  1.32it/s]
1|6358|Loss: 0.8521436452865601:  98%|█████████▊| 6358/6470 [1:22:49<01:23,  1.34it/s]
1|6359|Loss: 0.923977792263031:  98%|█████████▊| 6358/6470 [1:22:50<01:23,  1.34it/s] 
1|6359|Loss: 0.923977792263031:  98%|█████████▊| 6359/6470 [1:22:50<01:23,  1.33it/s]
1|6360|Loss: 0.8912377953529358:  98%|█████████▊| 6359/6470 [1:22:51<01:23,  1.33it/s]
1|6360|Loss: 0.8912377953529358:  98%|█████████▊| 6360/6470 [1:22:51<01:16,  1.44it/s]
1|6361|Loss: 0.7354192137718201:  98%|█████████▊| 6360/6470 [1:22:51<01:16,  1.44it/s]
1|6361|Loss: 0.7354192137718201:  98%|█████████▊| 6361/6470 [1:22:51<01:12,  1.51it/s]
1|6362|Loss: 0.7978514432907104:  98%|█████████▊| 6361/6470 [1:22:52<01:12,  1.51it/s]
1|6362|Loss: 0.7978514432907104:  98%|█████████▊| 6362/6470 [1:22:52<01:17,  1.39it/s]
1|6363|Loss: 0.7719378471374512:  98%|█████████▊| 6362/6470 [1:22:53<01:17,  1.39it/s]
1|6363|Loss: 0.7719378471374512:  98%|█████████▊| 6363/6470 [1:22:53<01:20,  1.33it/s]
1|6364|Loss: 0.8314441442489624:  98%|█████████▊| 6363/6470 [1:22:53<01:20,  1.33it/s]
1|6364|Loss: 0.8314441442489624:  98%|█████████▊| 6364/6470 [1:22:54<01:18,  1.36it/s]
1|6365|Loss: 0.8696937561035156:  98%|█████████▊| 6364/6470 [1:22:54<01:18,  1.36it/s]
1|6365|Loss: 0.8696937561035156:  98%|█████████▊| 6365/6470 [1:22:55<01:18,  1.34it/s]
1|6366|Loss: 0.862964391708374:  98%|█████████▊| 6365/6470 [1:22:55<01:18,  1.34it/s] 
1|6366|Loss: 0.862964391708374:  98%|█████████▊| 6366/6470 [1:22:55<01:17,  1.35it/s]
1|6367|Loss: 0.8113174438476562:  98%|█████████▊| 6366/6470 [1:22:56<01:17,  1.35it/s]
1|6367|Loss: 0.8113174438476562:  98%|█████████▊| 6367/6470 [1:22:56<01:13,  1.40it/s]
1|6368|Loss: 0.9630663990974426:  98%|█████████▊| 6367/6470 [1:22:56<01:13,  1.40it/s]
1|6368|Loss: 0.9630663990974426:  98%|█████████▊| 6368/6470 [1:22:56<01:08,  1.49it/s]
1|6369|Loss: 0.6954124569892883:  98%|█████████▊| 6368/6470 [1:22:57<01:08,  1.49it/s]
1|6369|Loss: 0.6954124569892883:  98%|█████████▊| 6369/6470 [1:22:57<01:07,  1.49it/s]
1|6370|Loss: 0.9835910201072693:  98%|█████████▊| 6369/6470 [1:22:57<01:07,  1.49it/s]
1|6370|Loss: 0.9835910201072693:  98%|█████████▊| 6370/6470 [1:22:58<01:04,  1.55it/s]
1|6371|Loss: 0.9624916911125183:  98%|█████████▊| 6370/6470 [1:22:58<01:04,  1.55it/s]
1|6371|Loss: 0.9624916911125183:  98%|█████████▊| 6371/6470 [1:22:58<01:07,  1.47it/s]
1|6372|Loss: 0.8222779631614685:  98%|█████████▊| 6371/6470 [1:22:59<01:07,  1.47it/s]
1|6372|Loss: 0.8222779631614685:  98%|█████████▊| 6372/6470 [1:22:59<01:10,  1.39it/s]
1|6373|Loss: 0.8245574235916138:  98%|█████████▊| 6372/6470 [1:23:00<01:10,  1.39it/s]
1|6373|Loss: 0.8245574235916138:  99%|█████████▊| 6373/6470 [1:23:00<01:11,  1.36it/s]
1|6374|Loss: 0.8987447619438171:  99%|█████████▊| 6373/6470 [1:23:01<01:11,  1.36it/s]
1|6374|Loss: 0.8987447619438171:  99%|█████████▊| 6374/6470 [1:23:01<01:14,  1.29it/s]
1|6375|Loss: 0.7053219079971313:  99%|█████████▊| 6374/6470 [1:23:01<01:14,  1.29it/s]
1|6375|Loss: 0.7053219079971313:  99%|█████████▊| 6375/6470 [1:23:02<01:09,  1.36it/s]
1|6376|Loss: 0.8805253505706787:  99%|█████████▊| 6375/6470 [1:23:02<01:09,  1.36it/s]
1|6376|Loss: 0.8805253505706787:  99%|█████████▊| 6376/6470 [1:23:02<01:06,  1.41it/s]
1|6377|Loss: 0.771619439125061:  99%|█████████▊| 6376/6470 [1:23:03<01:06,  1.41it/s] 
1|6377|Loss: 0.771619439125061:  99%|█████████▊| 6377/6470 [1:23:03<01:05,  1.42it/s]
1|6378|Loss: 1.0131428241729736:  99%|█████████▊| 6377/6470 [1:23:03<01:05,  1.42it/s]
1|6378|Loss: 1.0131428241729736:  99%|█████████▊| 6378/6470 [1:23:04<01:06,  1.38it/s]
1|6379|Loss: 0.6925416588783264:  99%|█████████▊| 6378/6470 [1:23:04<01:06,  1.38it/s]
1|6379|Loss: 0.6925416588783264:  99%|█████████▊| 6379/6470 [1:23:04<01:06,  1.36it/s]
1|6380|Loss: 0.9120784401893616:  99%|█████████▊| 6379/6470 [1:23:05<01:06,  1.36it/s]
1|6380|Loss: 0.9120784401893616:  99%|█████████▊| 6380/6470 [1:23:05<01:07,  1.34it/s]
1|6381|Loss: 0.8139145374298096:  99%|█████████▊| 6380/6470 [1:23:06<01:07,  1.34it/s]
1|6381|Loss: 0.8139145374298096:  99%|█████████▊| 6381/6470 [1:23:06<01:07,  1.32it/s]
1|6382|Loss: 0.6827878355979919:  99%|█████████▊| 6381/6470 [1:23:06<01:07,  1.32it/s]
1|6382|Loss: 0.6827878355979919:  99%|█████████▊| 6382/6470 [1:23:07<01:01,  1.44it/s]
1|6383|Loss: 0.9661029577255249:  99%|█████████▊| 6382/6470 [1:23:07<01:01,  1.44it/s]
1|6383|Loss: 0.9661029577255249:  99%|█████████▊| 6383/6470 [1:23:07<01:03,  1.38it/s]
1|6384|Loss: 1.0319141149520874:  99%|█████████▊| 6383/6470 [1:23:08<01:03,  1.38it/s]
1|6384|Loss: 1.0319141149520874:  99%|█████████▊| 6384/6470 [1:23:08<01:07,  1.28it/s]
1|6385|Loss: 0.8802630305290222:  99%|█████████▊| 6384/6470 [1:23:09<01:07,  1.28it/s]
1|6385|Loss: 0.8802630305290222:  99%|█████████▊| 6385/6470 [1:23:09<01:09,  1.22it/s]
1|6386|Loss: 0.7836554646492004:  99%|█████████▊| 6385/6470 [1:23:10<01:09,  1.22it/s]
1|6386|Loss: 0.7836554646492004:  99%|█████████▊| 6386/6470 [1:23:10<01:06,  1.25it/s]
1|6387|Loss: 1.0692055225372314:  99%|█████████▊| 6386/6470 [1:23:10<01:06,  1.25it/s]
1|6387|Loss: 1.0692055225372314:  99%|█████████▊| 6387/6470 [1:23:11<01:03,  1.30it/s]
1|6388|Loss: 0.890794038772583:  99%|█████████▊| 6387/6470 [1:23:11<01:03,  1.30it/s] 
1|6388|Loss: 0.890794038772583:  99%|█████████▊| 6388/6470 [1:23:11<01:02,  1.31it/s]
1|6389|Loss: 0.8172942399978638:  99%|█████████▊| 6388/6470 [1:23:12<01:02,  1.31it/s]
1|6389|Loss: 0.8172942399978638:  99%|█████████▊| 6389/6470 [1:23:12<01:00,  1.33it/s]
1|6390|Loss: 1.069525957107544:  99%|█████████▊| 6389/6470 [1:23:12<01:00,  1.33it/s] 
1|6390|Loss: 1.069525957107544:  99%|█████████▉| 6390/6470 [1:23:13<00:57,  1.39it/s]
1|6391|Loss: 0.9419370889663696:  99%|█████████▉| 6390/6470 [1:23:13<00:57,  1.39it/s]
1|6391|Loss: 0.9419370889663696:  99%|█████████▉| 6391/6470 [1:23:13<00:55,  1.42it/s]
1|6392|Loss: 0.8934022784233093:  99%|█████████▉| 6391/6470 [1:23:14<00:55,  1.42it/s]
1|6392|Loss: 0.8934022784233093:  99%|█████████▉| 6392/6470 [1:23:14<00:59,  1.32it/s]
1|6393|Loss: 1.0383812189102173:  99%|█████████▉| 6392/6470 [1:23:15<00:59,  1.32it/s]
1|6393|Loss: 1.0383812189102173:  99%|█████████▉| 6393/6470 [1:23:15<01:00,  1.26it/s]
1|6394|Loss: 0.9622870683670044:  99%|█████████▉| 6393/6470 [1:23:16<01:00,  1.26it/s]
1|6394|Loss: 0.9622870683670044:  99%|█████████▉| 6394/6470 [1:23:16<01:02,  1.21it/s]
1|6395|Loss: 0.8930312395095825:  99%|█████████▉| 6394/6470 [1:23:16<01:02,  1.21it/s]
1|6395|Loss: 0.8930312395095825:  99%|█████████▉| 6395/6470 [1:23:17<01:01,  1.23it/s]
1|6396|Loss: 0.7852662801742554:  99%|█████████▉| 6395/6470 [1:23:17<01:01,  1.23it/s]
1|6396|Loss: 0.7852662801742554:  99%|█████████▉| 6396/6470 [1:23:18<00:57,  1.29it/s]
1|6397|Loss: 0.8118379712104797:  99%|█████████▉| 6396/6470 [1:23:18<00:57,  1.29it/s]
1|6397|Loss: 0.8118379712104797:  99%|█████████▉| 6397/6470 [1:23:18<00:53,  1.35it/s]
1|6398|Loss: 1.0906404256820679:  99%|█████████▉| 6397/6470 [1:23:19<00:53,  1.35it/s]
1|6398|Loss: 1.0906404256820679:  99%|█████████▉| 6398/6470 [1:23:19<00:52,  1.36it/s]
1|6399|Loss: 1.014812707901001:  99%|█████████▉| 6398/6470 [1:23:19<00:52,  1.36it/s] 
1|6399|Loss: 1.014812707901001:  99%|█████████▉| 6399/6470 [1:23:20<00:53,  1.33it/s]
1|6400|Loss: 1.002825379371643:  99%|█████████▉| 6399/6470 [1:23:20<00:53,  1.33it/s]
1|6400|Loss: 1.002825379371643:  99%|█████████▉| 6400/6470 [1:23:20<00:51,  1.36it/s]
1|6401|Loss: 0.7910144329071045:  99%|█████████▉| 6400/6470 [1:23:21<00:51,  1.36it/s]
1|6401|Loss: 0.7910144329071045:  99%|█████████▉| 6401/6470 [1:23:21<00:49,  1.40it/s]
1|6402|Loss: 0.8517197966575623:  99%|█████████▉| 6401/6470 [1:23:22<00:49,  1.40it/s]
1|6402|Loss: 0.8517197966575623:  99%|█████████▉| 6402/6470 [1:23:22<00:51,  1.32it/s]
1|6403|Loss: 0.9172127842903137:  99%|█████████▉| 6402/6470 [1:23:22<00:51,  1.32it/s]
1|6403|Loss: 0.9172127842903137:  99%|█████████▉| 6403/6470 [1:23:23<00:53,  1.25it/s]
1|6404|Loss: 0.7324901223182678:  99%|█████████▉| 6403/6470 [1:23:23<00:53,  1.25it/s]
1|6404|Loss: 0.7324901223182678:  99%|█████████▉| 6404/6470 [1:23:24<00:54,  1.20it/s]
1|6405|Loss: 0.842185378074646:  99%|█████████▉| 6404/6470 [1:23:24<00:54,  1.20it/s] 
1|6405|Loss: 0.842185378074646:  99%|█████████▉| 6405/6470 [1:23:25<00:54,  1.19it/s]
1|6406|Loss: 0.9444968700408936:  99%|█████████▉| 6405/6470 [1:23:25<00:54,  1.19it/s]
1|6406|Loss: 0.9444968700408936:  99%|█████████▉| 6406/6470 [1:23:25<00:50,  1.26it/s]
1|6407|Loss: 0.8784969449043274:  99%|█████████▉| 6406/6470 [1:23:26<00:50,  1.26it/s]
1|6407|Loss: 0.8784969449043274:  99%|█████████▉| 6407/6470 [1:23:26<00:50,  1.24it/s]
1|6408|Loss: 0.8396616578102112:  99%|█████████▉| 6407/6470 [1:23:27<00:50,  1.24it/s]
1|6408|Loss: 0.8396616578102112:  99%|█████████▉| 6408/6470 [1:23:27<00:48,  1.29it/s]
1|6409|Loss: 0.840336263179779:  99%|█████████▉| 6408/6470 [1:23:27<00:48,  1.29it/s] 
1|6409|Loss: 0.840336263179779:  99%|█████████▉| 6409/6470 [1:23:28<00:47,  1.29it/s]
1|6410|Loss: 0.8421429395675659:  99%|█████████▉| 6409/6470 [1:23:28<00:47,  1.29it/s]
1|6410|Loss: 0.8421429395675659:  99%|█████████▉| 6410/6470 [1:23:29<00:49,  1.22it/s]
1|6411|Loss: 0.8267058730125427:  99%|█████████▉| 6410/6470 [1:23:29<00:49,  1.22it/s]
1|6411|Loss: 0.8267058730125427:  99%|█████████▉| 6411/6470 [1:23:29<00:46,  1.26it/s]
1|6412|Loss: 0.8028870224952698:  99%|█████████▉| 6411/6470 [1:23:30<00:46,  1.26it/s]
1|6412|Loss: 0.8028870224952698:  99%|█████████▉| 6412/6470 [1:23:30<00:44,  1.31it/s]
1|6413|Loss: 0.9148483276367188:  99%|█████████▉| 6412/6470 [1:23:30<00:44,  1.31it/s]
1|6413|Loss: 0.9148483276367188:  99%|█████████▉| 6413/6470 [1:23:31<00:42,  1.34it/s]
1|6414|Loss: 0.8982489109039307:  99%|█████████▉| 6413/6470 [1:23:31<00:42,  1.34it/s]
1|6414|Loss: 0.8982489109039307:  99%|█████████▉| 6414/6470 [1:23:31<00:41,  1.36it/s]
1|6415|Loss: 0.7215455174446106:  99%|█████████▉| 6414/6470 [1:23:32<00:41,  1.36it/s]
1|6415|Loss: 0.7215455174446106:  99%|█████████▉| 6415/6470 [1:23:32<00:37,  1.46it/s]
1|6416|Loss: 0.8455218076705933:  99%|█████████▉| 6415/6470 [1:23:32<00:37,  1.46it/s]
1|6416|Loss: 0.8455218076705933:  99%|█████████▉| 6416/6470 [1:23:33<00:37,  1.46it/s]
1|6417|Loss: 0.9356620907783508:  99%|█████████▉| 6416/6470 [1:23:33<00:37,  1.46it/s]
1|6417|Loss: 0.9356620907783508:  99%|█████████▉| 6417/6470 [1:23:33<00:36,  1.46it/s]
1|6418|Loss: 0.9118683338165283:  99%|█████████▉| 6417/6470 [1:23:34<00:36,  1.46it/s]
1|6418|Loss: 0.9118683338165283:  99%|█████████▉| 6418/6470 [1:23:34<00:37,  1.39it/s]
1|6419|Loss: 0.8605366945266724:  99%|█████████▉| 6418/6470 [1:23:35<00:37,  1.39it/s]
1|6419|Loss: 0.8605366945266724:  99%|█████████▉| 6419/6470 [1:23:35<00:37,  1.34it/s]
1|6420|Loss: 0.8232404589653015:  99%|█████████▉| 6419/6470 [1:23:35<00:37,  1.34it/s]
1|6420|Loss: 0.8232404589653015:  99%|█████████▉| 6420/6470 [1:23:36<00:40,  1.25it/s]
1|6421|Loss: 0.7223426699638367:  99%|█████████▉| 6420/6470 [1:23:36<00:40,  1.25it/s]
1|6421|Loss: 0.7223426699638367:  99%|█████████▉| 6421/6470 [1:23:37<00:41,  1.17it/s]
1|6422|Loss: 0.9149422645568848:  99%|█████████▉| 6421/6470 [1:23:37<00:41,  1.17it/s]
1|6422|Loss: 0.9149422645568848:  99%|█████████▉| 6422/6470 [1:23:38<00:41,  1.17it/s]
1|6423|Loss: 0.9076100587844849:  99%|█████████▉| 6422/6470 [1:23:38<00:41,  1.17it/s]
1|6423|Loss: 0.9076100587844849:  99%|█████████▉| 6423/6470 [1:23:38<00:39,  1.18it/s]
1|6424|Loss: 0.7259137630462646:  99%|█████████▉| 6423/6470 [1:23:39<00:39,  1.18it/s]
1|6424|Loss: 0.7259137630462646:  99%|█████████▉| 6424/6470 [1:23:39<00:36,  1.25it/s]
1|6425|Loss: 0.9917751550674438:  99%|█████████▉| 6424/6470 [1:23:40<00:36,  1.25it/s]
1|6425|Loss: 0.9917751550674438:  99%|█████████▉| 6425/6470 [1:23:40<00:35,  1.26it/s]
1|6426|Loss: 0.6370582580566406:  99%|█████████▉| 6425/6470 [1:23:40<00:35,  1.26it/s]
1|6426|Loss: 0.6370582580566406:  99%|█████████▉| 6426/6470 [1:23:41<00:34,  1.28it/s]
1|6427|Loss: 1.0047106742858887:  99%|█████████▉| 6426/6470 [1:23:41<00:34,  1.28it/s]
1|6427|Loss: 1.0047106742858887:  99%|█████████▉| 6427/6470 [1:23:42<00:34,  1.24it/s]
1|6428|Loss: 0.8263046145439148:  99%|█████████▉| 6427/6470 [1:23:42<00:34,  1.24it/s]
1|6428|Loss: 0.8263046145439148:  99%|█████████▉| 6428/6470 [1:23:43<00:35,  1.18it/s]
1|6429|Loss: 0.9846327304840088:  99%|█████████▉| 6428/6470 [1:23:43<00:35,  1.18it/s]
1|6429|Loss: 0.9846327304840088:  99%|█████████▉| 6429/6470 [1:23:44<00:36,  1.13it/s]
1|6430|Loss: 1.0185664892196655:  99%|█████████▉| 6429/6470 [1:23:44<00:36,  1.13it/s]
1|6430|Loss: 1.0185664892196655:  99%|█████████▉| 6430/6470 [1:23:44<00:33,  1.18it/s]
1|6431|Loss: 0.8730307221412659:  99%|█████████▉| 6430/6470 [1:23:45<00:33,  1.18it/s]
1|6431|Loss: 0.8730307221412659:  99%|█████████▉| 6431/6470 [1:23:45<00:32,  1.22it/s]
1|6432|Loss: 0.8801915049552917:  99%|█████████▉| 6431/6470 [1:23:45<00:32,  1.22it/s]
1|6432|Loss: 0.8801915049552917:  99%|█████████▉| 6432/6470 [1:23:46<00:27,  1.36it/s]
1|6433|Loss: 0.9316046237945557:  99%|█████████▉| 6432/6470 [1:23:46<00:27,  1.36it/s]
1|6433|Loss: 0.9316046237945557:  99%|█████████▉| 6433/6470 [1:23:46<00:27,  1.36it/s]
1|6434|Loss: 0.8645622134208679:  99%|█████████▉| 6433/6470 [1:23:47<00:27,  1.36it/s]
1|6434|Loss: 0.8645622134208679:  99%|█████████▉| 6434/6470 [1:23:47<00:26,  1.36it/s]
1|6435|Loss: 0.9133846759796143:  99%|█████████▉| 6434/6470 [1:23:47<00:26,  1.36it/s]
1|6435|Loss: 0.9133846759796143:  99%|█████████▉| 6435/6470 [1:23:48<00:26,  1.33it/s]
1|6436|Loss: 0.9244232177734375:  99%|█████████▉| 6435/6470 [1:23:48<00:26,  1.33it/s]
1|6436|Loss: 0.9244232177734375:  99%|█████████▉| 6436/6470 [1:23:49<00:25,  1.35it/s]
1|6437|Loss: 0.766483724117279:  99%|█████████▉| 6436/6470 [1:23:49<00:25,  1.35it/s] 
1|6437|Loss: 0.766483724117279:  99%|█████████▉| 6437/6470 [1:23:49<00:23,  1.38it/s]
1|6438|Loss: 0.8855649828910828:  99%|█████████▉| 6437/6470 [1:23:50<00:23,  1.38it/s]
1|6438|Loss: 0.8855649828910828: 100%|█████████▉| 6438/6470 [1:23:50<00:24,  1.29it/s]
1|6439|Loss: 0.8548092842102051: 100%|█████████▉| 6438/6470 [1:23:51<00:24,  1.29it/s]
1|6439|Loss: 0.8548092842102051: 100%|█████████▉| 6439/6470 [1:23:51<00:25,  1.22it/s]
1|6440|Loss: 0.7222249507904053: 100%|█████████▉| 6439/6470 [1:23:52<00:25,  1.22it/s]
1|6440|Loss: 0.7222249507904053: 100%|█████████▉| 6440/6470 [1:23:52<00:24,  1.22it/s]
1|6441|Loss: 0.7882655262947083: 100%|█████████▉| 6440/6470 [1:23:52<00:24,  1.22it/s]
1|6441|Loss: 0.7882655262947083: 100%|█████████▉| 6441/6470 [1:23:53<00:23,  1.24it/s]
1|6442|Loss: 0.7823682427406311: 100%|█████████▉| 6441/6470 [1:23:53<00:23,  1.24it/s]
1|6442|Loss: 0.7823682427406311: 100%|█████████▉| 6442/6470 [1:23:53<00:22,  1.25it/s]
1|6443|Loss: 0.9554167985916138: 100%|█████████▉| 6442/6470 [1:23:54<00:22,  1.25it/s]
1|6443|Loss: 0.9554167985916138: 100%|█████████▉| 6443/6470 [1:23:54<00:21,  1.27it/s]
1|6444|Loss: 0.8726814389228821: 100%|█████████▉| 6443/6470 [1:23:55<00:21,  1.27it/s]
1|6444|Loss: 0.8726814389228821: 100%|█████████▉| 6444/6470 [1:23:55<00:20,  1.26it/s]
1|6445|Loss: 0.8234642744064331: 100%|█████████▉| 6444/6470 [1:23:55<00:20,  1.26it/s]
1|6445|Loss: 0.8234642744064331: 100%|█████████▉| 6445/6470 [1:23:56<00:20,  1.24it/s]
1|6446|Loss: 0.8447853326797485: 100%|█████████▉| 6445/6470 [1:23:56<00:20,  1.24it/s]
1|6446|Loss: 0.8447853326797485: 100%|█████████▉| 6446/6470 [1:23:57<00:18,  1.28it/s]
1|6447|Loss: 0.8039735555648804: 100%|█████████▉| 6446/6470 [1:23:57<00:18,  1.28it/s]
1|6447|Loss: 0.8039735555648804: 100%|█████████▉| 6447/6470 [1:23:57<00:17,  1.33it/s]
1|6448|Loss: 0.8744729161262512: 100%|█████████▉| 6447/6470 [1:23:58<00:17,  1.33it/s]
1|6448|Loss: 0.8744729161262512: 100%|█████████▉| 6448/6470 [1:23:58<00:15,  1.38it/s]
1|6449|Loss: 0.8138036727905273: 100%|█████████▉| 6448/6470 [1:23:58<00:15,  1.38it/s]
1|6449|Loss: 0.8138036727905273: 100%|█████████▉| 6449/6470 [1:23:59<00:15,  1.37it/s]
1|6450|Loss: 0.8335291743278503: 100%|█████████▉| 6449/6470 [1:23:59<00:15,  1.37it/s]
1|6450|Loss: 0.8335291743278503: 100%|█████████▉| 6450/6470 [1:23:59<00:14,  1.40it/s]
1|6451|Loss: 0.7784044742584229: 100%|█████████▉| 6450/6470 [1:24:00<00:14,  1.40it/s]
1|6451|Loss: 0.7784044742584229: 100%|█████████▉| 6451/6470 [1:24:00<00:13,  1.42it/s]
1|6452|Loss: 0.6067465543746948: 100%|█████████▉| 6451/6470 [1:24:00<00:13,  1.42it/s]
1|6452|Loss: 0.6067465543746948: 100%|█████████▉| 6452/6470 [1:24:01<00:12,  1.40it/s]
1|6453|Loss: 0.7792543768882751: 100%|█████████▉| 6452/6470 [1:24:01<00:12,  1.40it/s]
1|6453|Loss: 0.7792543768882751: 100%|█████████▉| 6453/6470 [1:24:02<00:12,  1.33it/s]
1|6454|Loss: 0.6770053505897522: 100%|█████████▉| 6453/6470 [1:24:02<00:12,  1.33it/s]
1|6454|Loss: 0.6770053505897522: 100%|█████████▉| 6454/6470 [1:24:02<00:11,  1.34it/s]
1|6455|Loss: 0.9585354924201965: 100%|█████████▉| 6454/6470 [1:24:03<00:11,  1.34it/s]
1|6455|Loss: 0.9585354924201965: 100%|█████████▉| 6455/6470 [1:24:03<00:11,  1.33it/s]
1|6456|Loss: 0.8813311457633972: 100%|█████████▉| 6455/6470 [1:24:04<00:11,  1.33it/s]
1|6456|Loss: 0.8813311457633972: 100%|█████████▉| 6456/6470 [1:24:04<00:11,  1.27it/s]
1|6457|Loss: 0.9165953993797302: 100%|█████████▉| 6456/6470 [1:24:04<00:11,  1.27it/s]
1|6457|Loss: 0.9165953993797302: 100%|█████████▉| 6457/6470 [1:24:05<00:09,  1.32it/s]
1|6458|Loss: 0.7929320931434631: 100%|█████████▉| 6457/6470 [1:24:05<00:09,  1.32it/s]
1|6458|Loss: 0.7929320931434631: 100%|█████████▉| 6458/6470 [1:24:05<00:09,  1.32it/s]
1|6459|Loss: 0.8229908347129822: 100%|█████████▉| 6458/6470 [1:24:06<00:09,  1.32it/s]
1|6459|Loss: 0.8229908347129822: 100%|█████████▉| 6459/6470 [1:24:06<00:08,  1.32it/s]
1|6460|Loss: 0.969156801700592: 100%|█████████▉| 6459/6470 [1:24:07<00:08,  1.32it/s] 
1|6460|Loss: 0.969156801700592: 100%|█████████▉| 6460/6470 [1:24:07<00:07,  1.31it/s]
1|6461|Loss: 0.7801384925842285: 100%|█████████▉| 6460/6470 [1:24:07<00:07,  1.31it/s]
1|6461|Loss: 0.7801384925842285: 100%|█████████▉| 6461/6470 [1:24:08<00:06,  1.41it/s]
1|6462|Loss: 0.8237682580947876: 100%|█████████▉| 6461/6470 [1:24:08<00:06,  1.41it/s]
1|6462|Loss: 0.8237682580947876: 100%|█████████▉| 6462/6470 [1:24:08<00:05,  1.39it/s]
1|6463|Loss: 0.7651375532150269: 100%|█████████▉| 6462/6470 [1:24:09<00:05,  1.39it/s]
1|6463|Loss: 0.7651375532150269: 100%|█████████▉| 6463/6470 [1:24:09<00:05,  1.34it/s]
1|6464|Loss: 0.8364762663841248: 100%|█████████▉| 6463/6470 [1:24:10<00:05,  1.34it/s]
1|6464|Loss: 0.8364762663841248: 100%|█████████▉| 6464/6470 [1:24:10<00:04,  1.31it/s]
1|6465|Loss: 0.9944437146186829: 100%|█████████▉| 6464/6470 [1:24:10<00:04,  1.31it/s]
1|6465|Loss: 0.9944437146186829: 100%|█████████▉| 6465/6470 [1:24:11<00:03,  1.26it/s]
1|6466|Loss: 0.9235989451408386: 100%|█████████▉| 6465/6470 [1:24:11<00:03,  1.26it/s]
1|6466|Loss: 0.9235989451408386: 100%|█████████▉| 6466/6470 [1:24:12<00:03,  1.22it/s]
1|6467|Loss: 0.8272281885147095: 100%|█████████▉| 6466/6470 [1:24:12<00:03,  1.22it/s]
1|6467|Loss: 0.8272281885147095: 100%|█████████▉| 6467/6470 [1:24:12<00:02,  1.20it/s]
1|6468|Loss: 0.9076261520385742: 100%|█████████▉| 6467/6470 [1:24:13<00:02,  1.20it/s]
1|6468|Loss: 0.9076261520385742: 100%|█████████▉| 6468/6470 [1:24:13<00:01,  1.26it/s]
1|6469|Loss: 0.7618200182914734: 100%|█████████▉| 6468/6470 [1:24:14<00:01,  1.26it/s]
1|6469|Loss: 0.7618200182914734: 100%|█████████▉| 6469/6470 [1:24:14<00:00,  1.32it/s]
1|6470|Loss: 0.8044407367706299: 100%|█████████▉| 6469/6470 [1:24:14<00:00,  1.32it/s]
1|6470|Loss: 0.8044407367706299: 100%|██████████| 6470/6470 [1:24:15<00:00,  1.36it/s]
1|6470|Loss: 0.8044407367706299: 100%|██████████| 6470/6470 [1:24:15<00:00,  1.28it/s]
Model checkpoint of size 25961 MB saved to output/alpaca-colorful-llama2-finetune/model_0_6470.ckpt
Skipping uploading to HuggingFace Hub (no repo id specified)

  0%|          | 0/6470 [00:00<?, ?it/s]
2|1|Loss: 1.0914278030395508:   0%|          | 0/6470 [00:00<?, ?it/s]
2|1|Loss: 1.0914278030395508:   0%|          | 1/6470 [00:00<56:13,  1.92it/s]
2|2|Loss: 0.808953583240509:   0%|          | 1/6470 [00:00<56:13,  1.92it/s] 
2|2|Loss: 0.808953583240509:   0%|          | 2/6470 [00:01<1:09:30,  1.55it/s]
2|3|Loss: 0.8922575116157532:   0%|          | 2/6470 [00:01<1:09:30,  1.55it/s]
2|3|Loss: 0.8922575116157532:   0%|          | 3/6470 [00:02<1:18:45,  1.37it/s]
2|4|Loss: 0.7488001585006714:   0%|          | 3/6470 [00:02<1:18:45,  1.37it/s]
2|4|Loss: 0.7488001585006714:   0%|          | 4/6470 [00:02<1:11:58,  1.50it/s]
2|5|Loss: 0.8377427458763123:   0%|          | 4/6470 [00:02<1:11:58,  1.50it/s]
2|5|Loss: 0.8377427458763123:   0%|          | 5/6470 [00:03<1:09:45,  1.54it/s]
2|6|Loss: 0.8568614721298218:   0%|          | 5/6470 [00:03<1:09:45,  1.54it/s]
2|6|Loss: 0.8568614721298218:   0%|          | 6/6470 [00:04<1:13:20,  1.47it/s]
2|7|Loss: 0.8895756602287292:   0%|          | 6/6470 [00:04<1:13:20,  1.47it/s]
2|7|Loss: 0.8895756602287292:   0%|          | 7/6470 [00:04<1:16:45,  1.40it/s]
2|8|Loss: 0.8381625413894653:   0%|          | 7/6470 [00:05<1:16:45,  1.40it/s]
2|8|Loss: 0.8381625413894653:   0%|          | 8/6470 [00:05<1:20:25,  1.34it/s]
2|9|Loss: 0.7197388410568237:   0%|          | 8/6470 [00:06<1:20:25,  1.34it/s]
2|9|Loss: 0.7197388410568237:   0%|          | 9/6470 [00:06<1:21:26,  1.32it/s]
2|10|Loss: 0.9951055645942688:   0%|          | 9/6470 [00:06<1:21:26,  1.32it/s]
2|10|Loss: 0.9951055645942688:   0%|          | 10/6470 [00:07<1:24:45,  1.27it/s]
2|11|Loss: 0.9283627271652222:   0%|          | 10/6470 [00:07<1:24:45,  1.27it/s]
2|11|Loss: 0.9283627271652222:   0%|          | 11/6470 [00:07<1:21:14,  1.33it/s]
2|12|Loss: 0.9003990292549133:   0%|          | 11/6470 [00:08<1:21:14,  1.33it/s]
2|12|Loss: 0.9003990292549133:   0%|          | 12/6470 [00:08<1:20:40,  1.33it/s]
2|13|Loss: 0.9008589386940002:   0%|          | 12/6470 [00:09<1:20:40,  1.33it/s]
2|13|Loss: 0.9008589386940002:   0%|          | 13/6470 [00:09<1:25:00,  1.27it/s]
2|14|Loss: 0.9581912755966187:   0%|          | 13/6470 [00:10<1:25:00,  1.27it/s]
2|14|Loss: 0.9581912755966187:   0%|          | 14/6470 [00:10<1:25:18,  1.26it/s]
2|15|Loss: 0.7369280457496643:   0%|          | 14/6470 [00:10<1:25:18,  1.26it/s]
2|15|Loss: 0.7369280457496643:   0%|          | 15/6470 [00:11<1:23:33,  1.29it/s]
2|16|Loss: 0.839326798915863:   0%|          | 15/6470 [00:11<1:23:33,  1.29it/s] 
2|16|Loss: 0.839326798915863:   0%|          | 16/6470 [00:11<1:20:33,  1.34it/s]
2|17|Loss: 0.8648098707199097:   0%|          | 16/6470 [00:12<1:20:33,  1.34it/s]
2|17|Loss: 0.8648098707199097:   0%|          | 17/6470 [00:12<1:15:31,  1.42it/s]
2|18|Loss: 0.7767547965049744:   0%|          | 17/6470 [00:12<1:15:31,  1.42it/s]
2|18|Loss: 0.7767547965049744:   0%|          | 18/6470 [00:13<1:16:54,  1.40it/s]
2|19|Loss: 0.9182208180427551:   0%|          | 18/6470 [00:13<1:16:54,  1.40it/s]
2|19|Loss: 0.9182208180427551:   0%|          | 19/6470 [00:13<1:21:41,  1.32it/s]
2|20|Loss: 0.9774668216705322:   0%|          | 19/6470 [00:14<1:21:41,  1.32it/s]
2|20|Loss: 0.9774668216705322:   0%|          | 20/6470 [00:14<1:24:07,  1.28it/s]
2|21|Loss: 0.7937353849411011:   0%|          | 20/6470 [00:15<1:24:07,  1.28it/s]
2|21|Loss: 0.7937353849411011:   0%|          | 21/6470 [00:15<1:21:27,  1.32it/s]
2|22|Loss: 0.8020921349525452:   0%|          | 21/6470 [00:15<1:21:27,  1.32it/s]
2|22|Loss: 0.8020921349525452:   0%|          | 22/6470 [00:16<1:18:42,  1.37it/s]
2|23|Loss: 0.8488864302635193:   0%|          | 22/6470 [00:16<1:18:42,  1.37it/s]
2|23|Loss: 0.8488864302635193:   0%|          | 23/6470 [00:16<1:20:34,  1.33it/s]
2|24|Loss: 0.7533907890319824:   0%|          | 23/6470 [00:17<1:20:34,  1.33it/s]
2|24|Loss: 0.7533907890319824:   0%|          | 24/6470 [00:17<1:13:51,  1.45it/s]
2|25|Loss: 0.745887279510498:   0%|          | 24/6470 [00:17<1:13:51,  1.45it/s] 
2|25|Loss: 0.745887279510498:   0%|          | 25/6470 [00:18<1:13:23,  1.46it/s]
2|26|Loss: 0.9293866157531738:   0%|          | 25/6470 [00:18<1:13:23,  1.46it/s]
2|26|Loss: 0.9293866157531738:   0%|          | 26/6470 [00:18<1:12:44,  1.48it/s]
2|27|Loss: 0.9551060199737549:   0%|          | 26/6470 [00:19<1:12:44,  1.48it/s]
2|27|Loss: 0.9551060199737549:   0%|          | 27/6470 [00:19<1:04:51,  1.66it/s]
2|28|Loss: 0.8619967699050903:   0%|          | 27/6470 [00:19<1:04:51,  1.66it/s]
2|28|Loss: 0.8619967699050903:   0%|          | 28/6470 [00:19<1:07:38,  1.59it/s]
2|29|Loss: 0.8951432704925537:   0%|          | 28/6470 [00:20<1:07:38,  1.59it/s]
2|29|Loss: 0.8951432704925537:   0%|          | 29/6470 [00:20<1:15:35,  1.42it/s]
2|30|Loss: 0.7116103768348694:   0%|          | 29/6470 [00:21<1:15:35,  1.42it/s]
2|30|Loss: 0.7116103768348694:   0%|          | 30/6470 [00:21<1:21:42,  1.31it/s]
2|31|Loss: 0.763126015663147:   0%|          | 30/6470 [00:22<1:21:42,  1.31it/s] 
2|31|Loss: 0.763126015663147:   0%|          | 31/6470 [00:22<1:20:41,  1.33it/s]
2|32|Loss: 0.7940241098403931:   0%|          | 31/6470 [00:22<1:20:41,  1.33it/s]
2|32|Loss: 0.7940241098403931:   0%|          | 32/6470 [00:23<1:18:13,  1.37it/s]
2|33|Loss: 0.9189254641532898:   0%|          | 32/6470 [00:23<1:18:13,  1.37it/s]
2|33|Loss: 0.9189254641532898:   1%|          | 33/6470 [00:23<1:16:21,  1.41it/s]
2|34|Loss: 0.7222902178764343:   1%|          | 33/6470 [00:24<1:16:21,  1.41it/s]
2|34|Loss: 0.7222902178764343:   1%|          | 34/6470 [00:24<1:17:51,  1.38it/s]
2|35|Loss: 0.8095594048500061:   1%|          | 34/6470 [00:24<1:17:51,  1.38it/s]
2|35|Loss: 0.8095594048500061:   1%|          | 35/6470 [00:25<1:15:37,  1.42it/s]
2|36|Loss: 0.844319224357605:   1%|          | 35/6470 [00:25<1:15:37,  1.42it/s] 
2|36|Loss: 0.844319224357605:   1%|          | 36/6470 [00:25<1:15:39,  1.42it/s]
2|37|Loss: 0.7209942936897278:   1%|          | 36/6470 [00:26<1:15:39,  1.42it/s]
2|37|Loss: 0.7209942936897278:   1%|          | 37/6470 [00:26<1:11:17,  1.50it/s]
2|38|Loss: 0.6577540636062622:   1%|          | 37/6470 [00:26<1:11:17,  1.50it/s]
2|38|Loss: 0.6577540636062622:   1%|          | 38/6470 [00:27<1:09:44,  1.54it/s]
2|39|Loss: 0.7565412521362305:   1%|          | 38/6470 [00:27<1:09:44,  1.54it/s]
2|39|Loss: 0.7565412521362305:   1%|          | 39/6470 [00:27<1:08:15,  1.57it/s]
2|40|Loss: 0.9126712083816528:   1%|          | 39/6470 [00:28<1:08:15,  1.57it/s]
2|40|Loss: 0.9126712083816528:   1%|          | 40/6470 [00:28<1:13:17,  1.46it/s]
2|41|Loss: 0.9155727624893188:   1%|          | 40/6470 [00:28<1:13:17,  1.46it/s]
2|41|Loss: 0.9155727624893188:   1%|          | 41/6470 [00:29<1:15:45,  1.41it/s]
2|42|Loss: 0.7118359208106995:   1%|          | 41/6470 [00:29<1:15:45,  1.41it/s]
2|42|Loss: 0.7118359208106995:   1%|          | 42/6470 [00:30<1:17:53,  1.38it/s]
2|43|Loss: 0.6593111753463745:   1%|          | 42/6470 [00:30<1:17:53,  1.38it/s]
2|43|Loss: 0.6593111753463745:   1%|          | 43/6470 [00:30<1:21:54,  1.31it/s]
2|44|Loss: 0.9193435907363892:   1%|          | 43/6470 [00:31<1:21:54,  1.31it/s]
2|44|Loss: 0.9193435907363892:   1%|          | 44/6470 [00:31<1:19:26,  1.35it/s]
2|45|Loss: 0.7529253959655762:   1%|          | 44/6470 [00:31<1:19:26,  1.35it/s]
2|45|Loss: 0.7529253959655762:   1%|          | 45/6470 [00:32<1:16:11,  1.41it/s]
2|46|Loss: 0.9055619835853577:   1%|          | 45/6470 [00:32<1:16:11,  1.41it/s]
2|46|Loss: 0.9055619835853577:   1%|          | 46/6470 [00:33<1:19:02,  1.35it/s]
2|47|Loss: 0.7754009366035461:   1%|          | 46/6470 [00:33<1:19:02,  1.35it/s]
2|47|Loss: 0.7754009366035461:   1%|          | 47/6470 [00:33<1:25:33,  1.25it/s]
2|48|Loss: 0.7380227446556091:   1%|          | 47/6470 [00:34<1:25:33,  1.25it/s]
2|48|Loss: 0.7380227446556091:   1%|          | 48/6470 [00:34<1:25:42,  1.25it/s]
2|49|Loss: 0.8512960076332092:   1%|          | 48/6470 [00:35<1:25:42,  1.25it/s]
2|49|Loss: 0.8512960076332092:   1%|          | 49/6470 [00:35<1:24:37,  1.26it/s]
2|50|Loss: 0.9237502813339233:   1%|          | 49/6470 [00:35<1:24:37,  1.26it/s]
2|50|Loss: 0.9237502813339233:   1%|          | 50/6470 [00:36<1:20:48,  1.32it/s]
2|51|Loss: 1.008745551109314:   1%|          | 50/6470 [00:36<1:20:48,  1.32it/s] 
2|51|Loss: 1.008745551109314:   1%|          | 51/6470 [00:37<1:24:46,  1.26it/s]
2|52|Loss: 0.9303588271141052:   1%|          | 51/6470 [00:37<1:24:46,  1.26it/s]
2|52|Loss: 0.9303588271141052:   1%|          | 52/6470 [00:37<1:22:53,  1.29it/s]
2|53|Loss: 0.9576234221458435:   1%|          | 52/6470 [00:38<1:22:53,  1.29it/s]
2|53|Loss: 0.9576234221458435:   1%|          | 53/6470 [00:38<1:17:55,  1.37it/s]
2|54|Loss: 1.0747123956680298:   1%|          | 53/6470 [00:38<1:17:55,  1.37it/s]
2|54|Loss: 1.0747123956680298:   1%|          | 54/6470 [00:39<1:21:04,  1.32it/s]
2|55|Loss: 0.853064775466919:   1%|          | 54/6470 [00:39<1:21:04,  1.32it/s] 
2|55|Loss: 0.853064775466919:   1%|          | 55/6470 [00:39<1:17:19,  1.38it/s]
2|56|Loss: 1.004630446434021:   1%|          | 55/6470 [00:40<1:17:19,  1.38it/s]
2|56|Loss: 1.004630446434021:   1%|          | 56/6470 [00:40<1:13:43,  1.45it/s]
2|57|Loss: 1.0470257997512817:   1%|          | 56/6470 [00:40<1:13:43,  1.45it/s]
2|57|Loss: 1.0470257997512817:   1%|          | 57/6470 [00:41<1:09:45,  1.53it/s]
2|58|Loss: 0.8857016563415527:   1%|          | 57/6470 [00:41<1:09:45,  1.53it/s]
2|58|Loss: 0.8857016563415527:   1%|          | 58/6470 [00:41<1:11:36,  1.49it/s]
2|59|Loss: 0.8667882680892944:   1%|          | 58/6470 [00:42<1:11:36,  1.49it/s]
2|59|Loss: 0.8667882680892944:   1%|          | 59/6470 [00:42<1:17:51,  1.37it/s]
2|60|Loss: 0.8606767058372498:   1%|          | 59/6470 [00:43<1:17:51,  1.37it/s]
2|60|Loss: 0.8606767058372498:   1%|          | 60/6470 [00:43<1:17:40,  1.38it/s]
2|61|Loss: 0.9009014368057251:   1%|          | 60/6470 [00:43<1:17:40,  1.38it/s]
2|61|Loss: 0.9009014368057251:   1%|          | 61/6470 [00:44<1:21:41,  1.31it/s]
2|62|Loss: 0.8370543122291565:   1%|          | 61/6470 [00:44<1:21:41,  1.31it/s]
2|62|Loss: 0.8370543122291565:   1%|          | 62/6470 [00:44<1:19:28,  1.34it/s]
2|63|Loss: 0.8350756764411926:   1%|          | 62/6470 [00:45<1:19:28,  1.34it/s]
2|63|Loss: 0.8350756764411926:   1%|          | 63/6470 [00:45<1:18:49,  1.35it/s]
2|64|Loss: 0.8632751703262329:   1%|          | 63/6470 [00:46<1:18:49,  1.35it/s]
2|64|Loss: 0.8632751703262329:   1%|          | 64/6470 [00:46<1:16:06,  1.40it/s]
2|65|Loss: 1.002428650856018:   1%|          | 64/6470 [00:46<1:16:06,  1.40it/s] 
2|65|Loss: 1.002428650856018:   1%|          | 65/6470 [00:46<1:12:34,  1.47it/s]
2|66|Loss: 1.0751380920410156:   1%|          | 65/6470 [00:47<1:12:34,  1.47it/s]
2|66|Loss: 1.0751380920410156:   1%|          | 66/6470 [00:47<1:18:33,  1.36it/s]
2|67|Loss: 0.9269602298736572:   1%|          | 66/6470 [00:48<1:18:33,  1.36it/s]
2|67|Loss: 0.9269602298736572:   1%|          | 67/6470 [00:48<1:14:43,  1.43it/s]
2|68|Loss: 0.7707357406616211:   1%|          | 67/6470 [00:48<1:14:43,  1.43it/s]
2|68|Loss: 0.7707357406616211:   1%|          | 68/6470 [00:48<1:09:30,  1.54it/s]
2|69|Loss: 0.7660759687423706:   1%|          | 68/6470 [00:49<1:09:30,  1.54it/s]
2|69|Loss: 0.7660759687423706:   1%|          | 69/6470 [00:49<1:12:13,  1.48it/s]
2|70|Loss: 0.772892415523529:   1%|          | 69/6470 [00:50<1:12:13,  1.48it/s] 
2|70|Loss: 0.772892415523529:   1%|          | 70/6470 [00:50<1:15:10,  1.42it/s]
2|71|Loss: 0.8608977794647217:   1%|          | 70/6470 [00:50<1:15:10,  1.42it/s]
2|71|Loss: 0.8608977794647217:   1%|          | 71/6470 [00:51<1:16:38,  1.39it/s]
2|72|Loss: 0.9029865264892578:   1%|          | 71/6470 [00:51<1:16:38,  1.39it/s]
2|72|Loss: 0.9029865264892578:   1%|          | 72/6470 [00:51<1:14:05,  1.44it/s]
2|73|Loss: 0.7954429984092712:   1%|          | 72/6470 [00:52<1:14:05,  1.44it/s]
2|73|Loss: 0.7954429984092712:   1%|          | 73/6470 [00:52<1:14:48,  1.43it/s]
2|74|Loss: 0.918239414691925:   1%|          | 73/6470 [00:52<1:14:48,  1.43it/s] 
2|74|Loss: 0.918239414691925:   1%|          | 74/6470 [00:53<1:14:09,  1.44it/s]
2|75|Loss: 1.195026159286499:   1%|          | 74/6470 [00:53<1:14:09,  1.44it/s]
2|75|Loss: 1.195026159286499:   1%|          | 75/6470 [00:54<1:19:21,  1.34it/s]
2|76|Loss: 0.8805401921272278:   1%|          | 75/6470 [00:54<1:19:21,  1.34it/s]
2|76|Loss: 0.8805401921272278:   1%|          | 76/6470 [00:54<1:18:34,  1.36it/s]
2|77|Loss: 0.9248477816581726:   1%|          | 76/6470 [00:55<1:18:34,  1.36it/s]
2|77|Loss: 0.9248477816581726:   1%|          | 77/6470 [00:55<1:15:15,  1.42it/s]
2|78|Loss: 0.821323573589325:   1%|          | 77/6470 [00:55<1:15:15,  1.42it/s] 
2|78|Loss: 0.821323573589325:   1%|          | 78/6470 [00:56<1:16:32,  1.39it/s]
2|79|Loss: 0.9369165301322937:   1%|          | 78/6470 [00:56<1:16:32,  1.39it/s]
2|79|Loss: 0.9369165301322937:   1%|          | 79/6470 [00:56<1:17:29,  1.37it/s]
2|80|Loss: 0.7562646269798279:   1%|          | 79/6470 [00:57<1:17:29,  1.37it/s]
2|80|Loss: 0.7562646269798279:   1%|          | 80/6470 [00:57<1:16:25,  1.39it/s]
2|81|Loss: 0.7703831791877747:   1%|          | 80/6470 [00:58<1:16:25,  1.39it/s]
2|81|Loss: 0.7703831791877747:   1%|▏         | 81/6470 [00:58<1:18:34,  1.36it/s]
2|82|Loss: 0.8386315703392029:   1%|▏         | 81/6470 [00:58<1:18:34,  1.36it/s]
2|82|Loss: 0.8386315703392029:   1%|▏         | 82/6470 [00:59<1:19:56,  1.33it/s]
2|83|Loss: 0.9248376488685608:   1%|▏         | 82/6470 [00:59<1:19:56,  1.33it/s]
2|83|Loss: 0.9248376488685608:   1%|▏         | 83/6470 [00:59<1:12:47,  1.46it/s]
2|84|Loss: 0.7799661755561829:   1%|▏         | 83/6470 [01:00<1:12:47,  1.46it/s]
2|84|Loss: 0.7799661755561829:   1%|▏         | 84/6470 [01:00<1:08:43,  1.55it/s]
2|85|Loss: 1.1530271768569946:   1%|▏         | 84/6470 [01:00<1:08:43,  1.55it/s]
2|85|Loss: 1.1530271768569946:   1%|▏         | 85/6470 [01:01<1:10:42,  1.50it/s]
2|86|Loss: 0.7670463919639587:   1%|▏         | 85/6470 [01:01<1:10:42,  1.50it/s]
2|86|Loss: 0.7670463919639587:   1%|▏         | 86/6470 [01:01<1:15:33,  1.41it/s]
2|87|Loss: 0.8755122423171997:   1%|▏         | 86/6470 [01:02<1:15:33,  1.41it/s]
2|87|Loss: 0.8755122423171997:   1%|▏         | 87/6470 [01:02<1:15:46,  1.40it/s]
2|88|Loss: 0.8763253688812256:   1%|▏         | 87/6470 [01:02<1:15:46,  1.40it/s]
2|88|Loss: 0.8763253688812256:   1%|▏         | 88/6470 [01:03<1:17:57,  1.36it/s]
2|89|Loss: 0.8635507225990295:   1%|▏         | 88/6470 [01:03<1:17:57,  1.36it/s]
2|89|Loss: 0.8635507225990295:   1%|▏         | 89/6470 [01:03<1:11:57,  1.48it/s]
2|90|Loss: 0.923606276512146:   1%|▏         | 89/6470 [01:04<1:11:57,  1.48it/s] 
2|90|Loss: 0.923606276512146:   1%|▏         | 90/6470 [01:04<1:16:12,  1.40it/s]
2|91|Loss: 0.9379302859306335:   1%|▏         | 90/6470 [01:05<1:16:12,  1.40it/s]
2|91|Loss: 0.9379302859306335:   1%|▏         | 91/6470 [01:05<1:18:30,  1.35it/s]
2|92|Loss: 0.9285001754760742:   1%|▏         | 91/6470 [01:05<1:18:30,  1.35it/s]
2|92|Loss: 0.9285001754760742:   1%|▏         | 92/6470 [01:06<1:21:42,  1.30it/s]
2|93|Loss: 0.7232888340950012:   1%|▏         | 92/6470 [01:06<1:21:42,  1.30it/s]
2|93|Loss: 0.7232888340950012:   1%|▏         | 93/6470 [01:07<1:26:20,  1.23it/s]
2|94|Loss: 0.7968599796295166:   1%|▏         | 93/6470 [01:07<1:26:20,  1.23it/s]
2|94|Loss: 0.7968599796295166:   1%|▏         | 94/6470 [01:07<1:16:52,  1.38it/s]
2|95|Loss: 0.8257997035980225:   1%|▏         | 94/6470 [01:08<1:16:52,  1.38it/s]
2|95|Loss: 0.8257997035980225:   1%|▏         | 95/6470 [01:08<1:12:13,  1.47it/s]
2|96|Loss: 0.8572731614112854:   1%|▏         | 95/6470 [01:08<1:12:13,  1.47it/s]
2|96|Loss: 0.8572731614112854:   1%|▏         | 96/6470 [01:09<1:13:14,  1.45it/s]
2|97|Loss: 0.6941709518432617:   1%|▏         | 96/6470 [01:09<1:13:14,  1.45it/s]
2|97|Loss: 0.6941709518432617:   1%|▏         | 97/6470 [01:09<1:18:47,  1.35it/s]
2|98|Loss: 0.9141939878463745:   1%|▏         | 97/6470 [01:10<1:18:47,  1.35it/s]
2|98|Loss: 0.9141939878463745:   2%|▏         | 98/6470 [01:10<1:18:14,  1.36it/s]
2|99|Loss: 0.8382993936538696:   2%|▏         | 98/6470 [01:11<1:18:14,  1.36it/s]
2|99|Loss: 0.8382993936538696:   2%|▏         | 99/6470 [01:11<1:16:09,  1.39it/s]
2|100|Loss: 0.9974405765533447:   2%|▏         | 99/6470 [01:11<1:16:09,  1.39it/s]
2|100|Loss: 0.9974405765533447:   2%|▏         | 100/6470 [01:12<1:20:02,  1.33it/s]
2|101|Loss: 0.8508592247962952:   2%|▏         | 100/6470 [01:12<1:20:02,  1.33it/s]
2|101|Loss: 0.8508592247962952:   2%|▏         | 101/6470 [01:12<1:23:02,  1.28it/s]
2|102|Loss: 0.9975312352180481:   2%|▏         | 101/6470 [01:13<1:23:02,  1.28it/s]
2|102|Loss: 0.9975312352180481:   2%|▏         | 102/6470 [01:13<1:24:17,  1.26it/s]
2|103|Loss: 0.8258328437805176:   2%|▏         | 102/6470 [01:14<1:24:17,  1.26it/s]
2|103|Loss: 0.8258328437805176:   2%|▏         | 103/6470 [01:14<1:13:58,  1.43it/s]
2|104|Loss: 0.8842313289642334:   2%|▏         | 103/6470 [01:14<1:13:58,  1.43it/s]
2|104|Loss: 0.8842313289642334:   2%|▏         | 104/6470 [01:15<1:17:14,  1.37it/s]
2|105|Loss: 0.8130571246147156:   2%|▏         | 104/6470 [01:15<1:17:14,  1.37it/s]
2|105|Loss: 0.8130571246147156:   2%|▏         | 105/6470 [01:15<1:15:29,  1.41it/s]
2|106|Loss: 1.0120128393173218:   2%|▏         | 105/6470 [01:16<1:15:29,  1.41it/s]
2|106|Loss: 1.0120128393173218:   2%|▏         | 106/6470 [01:16<1:13:25,  1.44it/s]
2|107|Loss: 0.783905565738678:   2%|▏         | 106/6470 [01:16<1:13:25,  1.44it/s] 
2|107|Loss: 0.783905565738678:   2%|▏         | 107/6470 [01:17<1:15:22,  1.41it/s]
2|108|Loss: 0.9587352275848389:   2%|▏         | 107/6470 [01:17<1:15:22,  1.41it/s]
2|108|Loss: 0.9587352275848389:   2%|▏         | 108/6470 [01:17<1:15:15,  1.41it/s]
2|109|Loss: 0.7855825424194336:   2%|▏         | 108/6470 [01:18<1:15:15,  1.41it/s]
2|109|Loss: 0.7855825424194336:   2%|▏         | 109/6470 [01:18<1:10:02,  1.51it/s]
2|110|Loss: 0.9945772886276245:   2%|▏         | 109/6470 [01:18<1:10:02,  1.51it/s]
2|110|Loss: 0.9945772886276245:   2%|▏         | 110/6470 [01:19<1:12:03,  1.47it/s]
2|111|Loss: 0.8989734649658203:   2%|▏         | 110/6470 [01:19<1:12:03,  1.47it/s]
2|111|Loss: 0.8989734649658203:   2%|▏         | 111/6470 [01:20<1:18:30,  1.35it/s]
2|112|Loss: 0.8696088194847107:   2%|▏         | 111/6470 [01:20<1:18:30,  1.35it/s]
2|112|Loss: 0.8696088194847107:   2%|▏         | 112/6470 [01:20<1:20:29,  1.32it/s]
2|113|Loss: 0.870871901512146:   2%|▏         | 112/6470 [01:21<1:20:29,  1.32it/s] 
2|113|Loss: 0.870871901512146:   2%|▏         | 113/6470 [01:21<1:20:30,  1.32it/s]
2|114|Loss: 0.6719153523445129:   2%|▏         | 113/6470 [01:21<1:20:30,  1.32it/s]
2|114|Loss: 0.6719153523445129:   2%|▏         | 114/6470 [01:22<1:17:50,  1.36it/s]
2|115|Loss: 0.9543333053588867:   2%|▏         | 114/6470 [01:22<1:17:50,  1.36it/s]
2|115|Loss: 0.9543333053588867:   2%|▏         | 115/6470 [01:22<1:14:16,  1.43it/s]
2|116|Loss: 0.8957409858703613:   2%|▏         | 115/6470 [01:23<1:14:16,  1.43it/s]
2|116|Loss: 0.8957409858703613:   2%|▏         | 116/6470 [01:23<1:13:22,  1.44it/s]
2|117|Loss: 0.8383695483207703:   2%|▏         | 116/6470 [01:23<1:13:22,  1.44it/s]
2|117|Loss: 0.8383695483207703:   2%|▏         | 117/6470 [01:24<1:15:59,  1.39it/s]
2|118|Loss: 1.0090670585632324:   2%|▏         | 117/6470 [01:24<1:15:59,  1.39it/s]
2|118|Loss: 1.0090670585632324:   2%|▏         | 118/6470 [01:25<1:21:15,  1.30it/s]
2|119|Loss: 0.9312034845352173:   2%|▏         | 118/6470 [01:25<1:21:15,  1.30it/s]
2|119|Loss: 0.9312034845352173:   2%|▏         | 119/6470 [01:26<1:25:34,  1.24it/s]
2|120|Loss: 0.8122230768203735:   2%|▏         | 119/6470 [01:26<1:25:34,  1.24it/s]
2|120|Loss: 0.8122230768203735:   2%|▏         | 120/6470 [01:26<1:27:20,  1.21it/s]
2|121|Loss: 0.8345277905464172:   2%|▏         | 120/6470 [01:27<1:27:20,  1.21it/s]
2|121|Loss: 0.8345277905464172:   2%|▏         | 121/6470 [01:27<1:21:48,  1.29it/s]
2|122|Loss: 0.9922012090682983:   2%|▏         | 121/6470 [01:27<1:21:48,  1.29it/s]
2|122|Loss: 0.9922012090682983:   2%|▏         | 122/6470 [01:28<1:17:18,  1.37it/s]
2|123|Loss: 0.8892475366592407:   2%|▏         | 122/6470 [01:28<1:17:18,  1.37it/s]
2|123|Loss: 0.8892475366592407:   2%|▏         | 123/6470 [01:28<1:12:00,  1.47it/s]
2|124|Loss: 0.9709327220916748:   2%|▏         | 123/6470 [01:29<1:12:00,  1.47it/s]
2|124|Loss: 0.9709327220916748:   2%|▏         | 124/6470 [01:29<1:13:48,  1.43it/s]
2|125|Loss: 0.9742552042007446:   2%|▏         | 124/6470 [01:30<1:13:48,  1.43it/s]
2|125|Loss: 0.9742552042007446:   2%|▏         | 125/6470 [01:30<1:17:14,  1.37it/s]
2|126|Loss: 0.7612498998641968:   2%|▏         | 125/6470 [01:30<1:17:14,  1.37it/s]
2|126|Loss: 0.7612498998641968:   2%|▏         | 126/6470 [01:30<1:13:25,  1.44it/s]
2|127|Loss: 0.9458726644515991:   2%|▏         | 126/6470 [01:31<1:13:25,  1.44it/s]
2|127|Loss: 0.9458726644515991:   2%|▏         | 127/6470 [01:31<1:14:26,  1.42it/s]
2|128|Loss: 0.9225215911865234:   2%|▏         | 127/6470 [01:32<1:14:26,  1.42it/s]
2|128|Loss: 0.9225215911865234:   2%|▏         | 128/6470 [01:32<1:14:38,  1.42it/s]
2|129|Loss: 0.8681542277336121:   2%|▏         | 128/6470 [01:32<1:14:38,  1.42it/s]
2|129|Loss: 0.8681542277336121:   2%|▏         | 129/6470 [01:33<1:15:33,  1.40it/s]
2|130|Loss: 0.846611499786377:   2%|▏         | 129/6470 [01:33<1:15:33,  1.40it/s] 
2|130|Loss: 0.846611499786377:   2%|▏         | 130/6470 [01:34<1:21:20,  1.30it/s]
2|131|Loss: 0.9323669075965881:   2%|▏         | 130/6470 [01:34<1:21:20,  1.30it/s]
2|131|Loss: 0.9323669075965881:   2%|▏         | 131/6470 [01:34<1:23:27,  1.27it/s]
2|132|Loss: 0.8308795690536499:   2%|▏         | 131/6470 [01:35<1:23:27,  1.27it/s]
2|132|Loss: 0.8308795690536499:   2%|▏         | 132/6470 [01:35<1:15:53,  1.39it/s]
2|133|Loss: 0.8904551863670349:   2%|▏         | 132/6470 [01:35<1:15:53,  1.39it/s]
2|133|Loss: 0.8904551863670349:   2%|▏         | 133/6470 [01:36<1:13:36,  1.43it/s]
2|134|Loss: 0.7786272168159485:   2%|▏         | 133/6470 [01:36<1:13:36,  1.43it/s]
2|134|Loss: 0.7786272168159485:   2%|▏         | 134/6470 [01:36<1:15:56,  1.39it/s]
2|135|Loss: 0.9899224042892456:   2%|▏         | 134/6470 [01:37<1:15:56,  1.39it/s]
2|135|Loss: 0.9899224042892456:   2%|▏         | 135/6470 [01:37<1:20:47,  1.31it/s]
2|136|Loss: 0.9800200462341309:   2%|▏         | 135/6470 [01:38<1:20:47,  1.31it/s]
2|136|Loss: 0.9800200462341309:   2%|▏         | 136/6470 [01:38<1:14:57,  1.41it/s]
2|137|Loss: 0.8069770932197571:   2%|▏         | 136/6470 [01:38<1:14:57,  1.41it/s]
2|137|Loss: 0.8069770932197571:   2%|▏         | 137/6470 [01:39<1:20:35,  1.31it/s]
2|138|Loss: 0.8420538902282715:   2%|▏         | 137/6470 [01:39<1:20:35,  1.31it/s]
2|138|Loss: 0.8420538902282715:   2%|▏         | 138/6470 [01:40<1:22:09,  1.28it/s]
2|139|Loss: 0.8943473696708679:   2%|▏         | 138/6470 [01:40<1:22:09,  1.28it/s]
2|139|Loss: 0.8943473696708679:   2%|▏         | 139/6470 [01:40<1:17:39,  1.36it/s]
2|140|Loss: 0.9151288866996765:   2%|▏         | 139/6470 [01:41<1:17:39,  1.36it/s]
2|140|Loss: 0.9151288866996765:   2%|▏         | 140/6470 [01:41<1:16:41,  1.38it/s]
2|141|Loss: 0.9810117483139038:   2%|▏         | 140/6470 [01:41<1:16:41,  1.38it/s]
2|141|Loss: 0.9810117483139038:   2%|▏         | 141/6470 [01:42<1:14:17,  1.42it/s]
2|142|Loss: 0.8534151315689087:   2%|▏         | 141/6470 [01:42<1:14:17,  1.42it/s]
2|142|Loss: 0.8534151315689087:   2%|▏         | 142/6470 [01:42<1:07:40,  1.56it/s]
2|143|Loss: 0.7525904774665833:   2%|▏         | 142/6470 [01:42<1:07:40,  1.56it/s]
2|143|Loss: 0.7525904774665833:   2%|▏         | 143/6470 [01:43<1:03:57,  1.65it/s]
2|144|Loss: 0.8795425295829773:   2%|▏         | 143/6470 [01:43<1:03:57,  1.65it/s]
2|144|Loss: 0.8795425295829773:   2%|▏         | 144/6470 [01:43<1:00:06,  1.75it/s]
2|145|Loss: 0.9671394228935242:   2%|▏         | 144/6470 [01:43<1:00:06,  1.75it/s]
2|145|Loss: 0.9671394228935242:   2%|▏         | 145/6470 [01:44<1:03:44,  1.65it/s]
2|146|Loss: 0.8939447999000549:   2%|▏         | 145/6470 [01:44<1:03:44,  1.65it/s]
2|146|Loss: 0.8939447999000549:   2%|▏         | 146/6470 [01:44<1:09:32,  1.52it/s]
2|147|Loss: 0.8710523247718811:   2%|▏         | 146/6470 [01:45<1:09:32,  1.52it/s]
2|147|Loss: 0.8710523247718811:   2%|▏         | 147/6470 [01:45<1:08:26,  1.54it/s]
2|148|Loss: 0.9138202667236328:   2%|▏         | 147/6470 [01:45<1:08:26,  1.54it/s]
2|148|Loss: 0.9138202667236328:   2%|▏         | 148/6470 [01:46<1:01:09,  1.72it/s]
2|149|Loss: 0.9360518455505371:   2%|▏         | 148/6470 [01:46<1:01:09,  1.72it/s]
2|149|Loss: 0.9360518455505371:   2%|▏         | 149/6470 [01:46<1:07:07,  1.57it/s]
2|150|Loss: 0.9762651324272156:   2%|▏         | 149/6470 [01:47<1:07:07,  1.57it/s]
2|150|Loss: 0.9762651324272156:   2%|▏         | 150/6470 [01:47<1:08:30,  1.54it/s]
2|151|Loss: 0.7931217551231384:   2%|▏         | 150/6470 [01:47<1:08:30,  1.54it/s]
2|151|Loss: 0.7931217551231384:   2%|▏         | 151/6470 [01:48<1:10:47,  1.49it/s]
2|152|Loss: 1.0096648931503296:   2%|▏         | 151/6470 [01:48<1:10:47,  1.49it/s]
2|152|Loss: 1.0096648931503296:   2%|▏         | 152/6470 [01:48<1:14:19,  1.42it/s]
2|153|Loss: 0.9335631132125854:   2%|▏         | 152/6470 [01:49<1:14:19,  1.42it/s]
2|153|Loss: 0.9335631132125854:   2%|▏         | 153/6470 [01:49<1:11:19,  1.48it/s]
2|154|Loss: 0.7673543095588684:   2%|▏         | 153/6470 [01:49<1:11:19,  1.48it/s]
2|154|Loss: 0.7673543095588684:   2%|▏         | 154/6470 [01:50<1:08:51,  1.53it/s]
2|155|Loss: 0.8279432058334351:   2%|▏         | 154/6470 [01:50<1:08:51,  1.53it/s]
2|155|Loss: 0.8279432058334351:   2%|▏         | 155/6470 [01:50<1:08:47,  1.53it/s]
2|156|Loss: 0.7735862135887146:   2%|▏         | 155/6470 [01:51<1:08:47,  1.53it/s]
2|156|Loss: 0.7735862135887146:   2%|▏         | 156/6470 [01:51<1:09:11,  1.52it/s]
2|157|Loss: 0.7536028027534485:   2%|▏         | 156/6470 [01:51<1:09:11,  1.52it/s]
2|157|Loss: 0.7536028027534485:   2%|▏         | 157/6470 [01:52<1:08:31,  1.54it/s]
2|158|Loss: 0.8781258463859558:   2%|▏         | 157/6470 [01:52<1:08:31,  1.54it/s]
2|158|Loss: 0.8781258463859558:   2%|▏         | 158/6470 [01:52<1:14:27,  1.41it/s]
2|159|Loss: 0.8257603645324707:   2%|▏         | 158/6470 [01:53<1:14:27,  1.41it/s]
2|159|Loss: 0.8257603645324707:   2%|▏         | 159/6470 [01:53<1:19:24,  1.32it/s]
2|160|Loss: 0.9932994246482849:   2%|▏         | 159/6470 [01:54<1:19:24,  1.32it/s]
2|160|Loss: 0.9932994246482849:   2%|▏         | 160/6470 [01:54<1:19:28,  1.32it/s]
2|161|Loss: 0.8750854730606079:   2%|▏         | 160/6470 [01:55<1:19:28,  1.32it/s]
2|161|Loss: 0.8750854730606079:   2%|▏         | 161/6470 [01:55<1:20:37,  1.30it/s]
2|162|Loss: 0.8804144859313965:   2%|▏         | 161/6470 [01:55<1:20:37,  1.30it/s]
2|162|Loss: 0.8804144859313965:   3%|▎         | 162/6470 [01:56<1:19:23,  1.32it/s]
2|163|Loss: 0.9831052422523499:   3%|▎         | 162/6470 [01:56<1:19:23,  1.32it/s]
2|163|Loss: 0.9831052422523499:   3%|▎         | 163/6470 [01:56<1:19:44,  1.32it/s]
2|164|Loss: 0.8747805953025818:   3%|▎         | 163/6470 [01:57<1:19:44,  1.32it/s]
2|164|Loss: 0.8747805953025818:   3%|▎         | 164/6470 [01:57<1:18:26,  1.34it/s]
2|165|Loss: 0.8201043009757996:   3%|▎         | 164/6470 [01:57<1:18:26,  1.34it/s]
2|165|Loss: 0.8201043009757996:   3%|▎         | 165/6470 [01:58<1:15:07,  1.40it/s]
2|166|Loss: 0.8227402567863464:   3%|▎         | 165/6470 [01:58<1:15:07,  1.40it/s]
2|166|Loss: 0.8227402567863464:   3%|▎         | 166/6470 [01:58<1:09:45,  1.51it/s]
2|167|Loss: 0.9352574348449707:   3%|▎         | 166/6470 [01:59<1:09:45,  1.51it/s]
2|167|Loss: 0.9352574348449707:   3%|▎         | 167/6470 [01:59<1:01:58,  1.69it/s]
2|168|Loss: 0.9272667765617371:   3%|▎         | 167/6470 [01:59<1:01:58,  1.69it/s]
2|168|Loss: 0.9272667765617371:   3%|▎         | 168/6470 [01:59<1:03:24,  1.66it/s]
2|169|Loss: 0.9388609528541565:   3%|▎         | 168/6470 [02:00<1:03:24,  1.66it/s]
2|169|Loss: 0.9388609528541565:   3%|▎         | 169/6470 [02:00<1:10:15,  1.49it/s]
2|170|Loss: 0.870965301990509:   3%|▎         | 169/6470 [02:01<1:10:15,  1.49it/s] 
2|170|Loss: 0.870965301990509:   3%|▎         | 170/6470 [02:01<1:12:22,  1.45it/s]
2|171|Loss: 0.954710841178894:   3%|▎         | 170/6470 [02:01<1:12:22,  1.45it/s]
2|171|Loss: 0.954710841178894:   3%|▎         | 171/6470 [02:02<1:17:12,  1.36it/s]
2|172|Loss: 0.8333350419998169:   3%|▎         | 171/6470 [02:02<1:17:12,  1.36it/s]
2|172|Loss: 0.8333350419998169:   3%|▎         | 172/6470 [02:03<1:23:12,  1.26it/s]
2|173|Loss: 0.9002498984336853:   3%|▎         | 172/6470 [02:03<1:23:12,  1.26it/s]
2|173|Loss: 0.9002498984336853:   3%|▎         | 173/6470 [02:03<1:17:24,  1.36it/s]
2|174|Loss: 0.7243698239326477:   3%|▎         | 173/6470 [02:04<1:17:24,  1.36it/s]
2|174|Loss: 0.7243698239326477:   3%|▎         | 174/6470 [02:04<1:16:01,  1.38it/s]
2|175|Loss: 0.936565637588501:   3%|▎         | 174/6470 [02:04<1:16:01,  1.38it/s] 
2|175|Loss: 0.936565637588501:   3%|▎         | 175/6470 [02:05<1:16:55,  1.36it/s]
2|176|Loss: 1.0310888290405273:   3%|▎         | 175/6470 [02:05<1:16:55,  1.36it/s]
2|176|Loss: 1.0310888290405273:   3%|▎         | 176/6470 [02:06<1:20:19,  1.31it/s]
2|177|Loss: 0.7955536246299744:   3%|▎         | 176/6470 [02:06<1:20:19,  1.31it/s]
2|177|Loss: 0.7955536246299744:   3%|▎         | 177/6470 [02:06<1:20:45,  1.30it/s]
2|178|Loss: 1.0729783773422241:   3%|▎         | 177/6470 [02:07<1:20:45,  1.30it/s]
2|178|Loss: 1.0729783773422241:   3%|▎         | 178/6470 [02:07<1:16:57,  1.36it/s]
2|179|Loss: 0.9377379417419434:   3%|▎         | 178/6470 [02:07<1:16:57,  1.36it/s]
2|179|Loss: 0.9377379417419434:   3%|▎         | 179/6470 [02:08<1:14:16,  1.41it/s]
2|180|Loss: 0.7539752721786499:   3%|▎         | 179/6470 [02:08<1:14:16,  1.41it/s]
2|180|Loss: 0.7539752721786499:   3%|▎         | 180/6470 [02:08<1:09:51,  1.50it/s]
2|181|Loss: 0.922163188457489:   3%|▎         | 180/6470 [02:09<1:09:51,  1.50it/s] 
2|181|Loss: 0.922163188457489:   3%|▎         | 181/6470 [02:09<1:09:15,  1.51it/s]
2|182|Loss: 0.8145303130149841:   3%|▎         | 181/6470 [02:09<1:09:15,  1.51it/s]
2|182|Loss: 0.8145303130149841:   3%|▎         | 182/6470 [02:10<1:13:31,  1.43it/s]
2|183|Loss: 0.9037642478942871:   3%|▎         | 182/6470 [02:10<1:13:31,  1.43it/s]
2|183|Loss: 0.9037642478942871:   3%|▎         | 183/6470 [02:11<1:19:38,  1.32it/s]
2|184|Loss: 0.76871657371521:   3%|▎         | 183/6470 [02:11<1:19:38,  1.32it/s]  
2|184|Loss: 0.76871657371521:   3%|▎         | 184/6470 [02:11<1:22:03,  1.28it/s]
2|185|Loss: 0.9674645066261292:   3%|▎         | 184/6470 [02:12<1:22:03,  1.28it/s]
2|185|Loss: 0.9674645066261292:   3%|▎         | 185/6470 [02:12<1:19:27,  1.32it/s]
2|186|Loss: 0.8723605275154114:   3%|▎         | 185/6470 [02:13<1:19:27,  1.32it/s]
2|186|Loss: 0.8723605275154114:   3%|▎         | 186/6470 [02:13<1:21:34,  1.28it/s]
2|187|Loss: 0.8888959884643555:   3%|▎         | 186/6470 [02:13<1:21:34,  1.28it/s]
2|187|Loss: 0.8888959884643555:   3%|▎         | 187/6470 [02:14<1:21:29,  1.29it/s]
2|188|Loss: 0.8450784683227539:   3%|▎         | 187/6470 [02:14<1:21:29,  1.29it/s]
2|188|Loss: 0.8450784683227539:   3%|▎         | 188/6470 [02:15<1:22:13,  1.27it/s]
2|189|Loss: 1.0561543703079224:   3%|▎         | 188/6470 [02:15<1:22:13,  1.27it/s]
2|189|Loss: 1.0561543703079224:   3%|▎         | 189/6470 [02:15<1:20:41,  1.30it/s]
2|190|Loss: 0.7460789680480957:   3%|▎         | 189/6470 [02:16<1:20:41,  1.30it/s]
2|190|Loss: 0.7460789680480957:   3%|▎         | 190/6470 [02:16<1:22:43,  1.27it/s]
2|191|Loss: 0.7700487971305847:   3%|▎         | 190/6470 [02:17<1:22:43,  1.27it/s]
2|191|Loss: 0.7700487971305847:   3%|▎         | 191/6470 [02:17<1:19:55,  1.31it/s]
2|192|Loss: 0.6901913285255432:   3%|▎         | 191/6470 [02:17<1:19:55,  1.31it/s]
2|192|Loss: 0.6901913285255432:   3%|▎         | 192/6470 [02:17<1:09:24,  1.51it/s]
2|193|Loss: 0.7851212024688721:   3%|▎         | 192/6470 [02:18<1:09:24,  1.51it/s]
2|193|Loss: 0.7851212024688721:   3%|▎         | 193/6470 [02:18<1:04:43,  1.62it/s]
2|194|Loss: 0.8116815686225891:   3%|▎         | 193/6470 [02:18<1:04:43,  1.62it/s]
2|194|Loss: 0.8116815686225891:   3%|▎         | 194/6470 [02:19<1:12:29,  1.44it/s]
2|195|Loss: 0.7718523740768433:   3%|▎         | 194/6470 [02:19<1:12:29,  1.44it/s]
2|195|Loss: 0.7718523740768433:   3%|▎         | 195/6470 [02:19<1:16:26,  1.37it/s]
2|196|Loss: 0.6161233186721802:   3%|▎         | 195/6470 [02:20<1:16:26,  1.37it/s]
2|196|Loss: 0.6161233186721802:   3%|▎         | 196/6470 [02:20<1:14:52,  1.40it/s]
2|197|Loss: 0.8313637971878052:   3%|▎         | 196/6470 [02:20<1:14:52,  1.40it/s]
2|197|Loss: 0.8313637971878052:   3%|▎         | 197/6470 [02:21<1:11:42,  1.46it/s]
2|198|Loss: 0.7432753443717957:   3%|▎         | 197/6470 [02:21<1:11:42,  1.46it/s]
2|198|Loss: 0.7432753443717957:   3%|▎         | 198/6470 [02:21<1:13:57,  1.41it/s]
2|199|Loss: 0.8695489168167114:   3%|▎         | 198/6470 [02:22<1:13:57,  1.41it/s]
2|199|Loss: 0.8695489168167114:   3%|▎         | 199/6470 [02:22<1:16:50,  1.36it/s]
2|200|Loss: 0.8608313798904419:   3%|▎         | 199/6470 [02:23<1:16:50,  1.36it/s]
2|200|Loss: 0.8608313798904419:   3%|▎         | 200/6470 [02:23<1:15:38,  1.38it/s]
2|201|Loss: 0.7960037589073181:   3%|▎         | 200/6470 [02:23<1:15:38,  1.38it/s]
2|201|Loss: 0.7960037589073181:   3%|▎         | 201/6470 [02:24<1:20:16,  1.30it/s]
2|202|Loss: 0.953932523727417:   3%|▎         | 201/6470 [02:24<1:20:16,  1.30it/s] 
2|202|Loss: 0.953932523727417:   3%|▎         | 202/6470 [02:25<1:23:56,  1.24it/s]
2|203|Loss: 0.8926041722297668:   3%|▎         | 202/6470 [02:25<1:23:56,  1.24it/s]
2|203|Loss: 0.8926041722297668:   3%|▎         | 203/6470 [02:25<1:17:43,  1.34it/s]
2|204|Loss: 0.6228679418563843:   3%|▎         | 203/6470 [02:26<1:17:43,  1.34it/s]
2|204|Loss: 0.6228679418563843:   3%|▎         | 204/6470 [02:26<1:11:39,  1.46it/s]
2|205|Loss: 0.8177396655082703:   3%|▎         | 204/6470 [02:26<1:11:39,  1.46it/s]
2|205|Loss: 0.8177396655082703:   3%|▎         | 205/6470 [02:27<1:12:19,  1.44it/s]
2|206|Loss: 0.8815363049507141:   3%|▎         | 205/6470 [02:27<1:12:19,  1.44it/s]
2|206|Loss: 0.8815363049507141:   3%|▎         | 206/6470 [02:27<1:11:59,  1.45it/s]
2|207|Loss: 0.9780519008636475:   3%|▎         | 206/6470 [02:28<1:11:59,  1.45it/s]
2|207|Loss: 0.9780519008636475:   3%|▎         | 207/6470 [02:28<1:14:38,  1.40it/s]
2|208|Loss: 0.7307475209236145:   3%|▎         | 207/6470 [02:29<1:14:38,  1.40it/s]
2|208|Loss: 0.7307475209236145:   3%|▎         | 208/6470 [02:29<1:22:07,  1.27it/s]
2|209|Loss: 0.88315349817276:   3%|▎         | 208/6470 [02:29<1:22:07,  1.27it/s]  
2|209|Loss: 0.88315349817276:   3%|▎         | 209/6470 [02:30<1:17:51,  1.34it/s]
2|210|Loss: 0.6860472559928894:   3%|▎         | 209/6470 [02:30<1:17:51,  1.34it/s]
2|210|Loss: 0.6860472559928894:   3%|▎         | 210/6470 [02:31<1:21:45,  1.28it/s]
2|211|Loss: 0.8691408038139343:   3%|▎         | 210/6470 [02:31<1:21:45,  1.28it/s]
2|211|Loss: 0.8691408038139343:   3%|▎         | 211/6470 [02:31<1:22:08,  1.27it/s]
2|212|Loss: 0.8114675283432007:   3%|▎         | 211/6470 [02:32<1:22:08,  1.27it/s]
2|212|Loss: 0.8114675283432007:   3%|▎         | 212/6470 [02:32<1:26:05,  1.21it/s]
2|213|Loss: 0.8168999552726746:   3%|▎         | 212/6470 [02:33<1:26:05,  1.21it/s]
2|213|Loss: 0.8168999552726746:   3%|▎         | 213/6470 [02:33<1:23:15,  1.25it/s]
2|214|Loss: 0.9279773235321045:   3%|▎         | 213/6470 [02:33<1:23:15,  1.25it/s]
2|214|Loss: 0.9279773235321045:   3%|▎         | 214/6470 [02:34<1:22:09,  1.27it/s]
2|215|Loss: 1.4251658916473389:   3%|▎         | 214/6470 [02:34<1:22:09,  1.27it/s]
2|215|Loss: 1.4251658916473389:   3%|▎         | 215/6470 [02:35<1:24:00,  1.24it/s]
2|216|Loss: 0.8346552848815918:   3%|▎         | 215/6470 [02:35<1:24:00,  1.24it/s]
2|216|Loss: 0.8346552848815918:   3%|▎         | 216/6470 [02:36<1:27:47,  1.19it/s]
2|217|Loss: 0.871345043182373:   3%|▎         | 216/6470 [02:36<1:27:47,  1.19it/s] 
2|217|Loss: 0.871345043182373:   3%|▎         | 217/6470 [02:36<1:27:42,  1.19it/s]
2|218|Loss: 0.7906252145767212:   3%|▎         | 217/6470 [02:37<1:27:42,  1.19it/s]
2|218|Loss: 0.7906252145767212:   3%|▎         | 218/6470 [02:37<1:28:26,  1.18it/s]
2|219|Loss: 0.9833583831787109:   3%|▎         | 218/6470 [02:38<1:28:26,  1.18it/s]
2|219|Loss: 0.9833583831787109:   3%|▎         | 219/6470 [02:38<1:23:29,  1.25it/s]
2|220|Loss: 0.8534232378005981:   3%|▎         | 219/6470 [02:38<1:23:29,  1.25it/s]
2|220|Loss: 0.8534232378005981:   3%|▎         | 220/6470 [02:39<1:18:58,  1.32it/s]
2|221|Loss: 0.7797607779502869:   3%|▎         | 220/6470 [02:39<1:18:58,  1.32it/s]
2|221|Loss: 0.7797607779502869:   3%|▎         | 221/6470 [02:39<1:18:08,  1.33it/s]
2|222|Loss: 0.8002520799636841:   3%|▎         | 221/6470 [02:40<1:18:08,  1.33it/s]
2|222|Loss: 0.8002520799636841:   3%|▎         | 222/6470 [02:40<1:14:51,  1.39it/s]
2|223|Loss: 0.9736561179161072:   3%|▎         | 222/6470 [02:40<1:14:51,  1.39it/s]
2|223|Loss: 0.9736561179161072:   3%|▎         | 223/6470 [02:41<1:12:54,  1.43it/s]
2|224|Loss: 0.7022221684455872:   3%|▎         | 223/6470 [02:41<1:12:54,  1.43it/s]
2|224|Loss: 0.7022221684455872:   3%|▎         | 224/6470 [02:42<1:20:18,  1.30it/s]
2|225|Loss: 0.9247711300849915:   3%|▎         | 224/6470 [02:42<1:20:18,  1.30it/s]
2|225|Loss: 0.9247711300849915:   3%|▎         | 225/6470 [02:42<1:20:10,  1.30it/s]
2|226|Loss: 0.7516542673110962:   3%|▎         | 225/6470 [02:43<1:20:10,  1.30it/s]
2|226|Loss: 0.7516542673110962:   3%|▎         | 226/6470 [02:43<1:22:01,  1.27it/s]
2|227|Loss: 0.845697820186615:   3%|▎         | 226/6470 [02:44<1:22:01,  1.27it/s] 
2|227|Loss: 0.845697820186615:   4%|▎         | 227/6470 [02:44<1:16:29,  1.36it/s]
2|228|Loss: 0.9595582485198975:   4%|▎         | 227/6470 [02:44<1:16:29,  1.36it/s]
2|228|Loss: 0.9595582485198975:   4%|▎         | 228/6470 [02:44<1:15:55,  1.37it/s]
2|229|Loss: 0.8340141773223877:   4%|▎         | 228/6470 [02:45<1:15:55,  1.37it/s]
2|229|Loss: 0.8340141773223877:   4%|▎         | 229/6470 [02:45<1:17:55,  1.33it/s]
2|230|Loss: 0.8847905993461609:   4%|▎         | 229/6470 [02:46<1:17:55,  1.33it/s]
2|230|Loss: 0.8847905993461609:   4%|▎         | 230/6470 [02:46<1:19:45,  1.30it/s]
2|231|Loss: 1.0497149229049683:   4%|▎         | 230/6470 [02:46<1:19:45,  1.30it/s]
2|231|Loss: 1.0497149229049683:   4%|▎         | 231/6470 [02:47<1:18:21,  1.33it/s]
2|232|Loss: 0.8168336153030396:   4%|▎         | 231/6470 [02:47<1:18:21,  1.33it/s]
2|232|Loss: 0.8168336153030396:   4%|▎         | 232/6470 [02:48<1:17:51,  1.34it/s]
2|233|Loss: 0.7496654987335205:   4%|▎         | 232/6470 [02:48<1:17:51,  1.34it/s]
2|233|Loss: 0.7496654987335205:   4%|▎         | 233/6470 [02:48<1:11:10,  1.46it/s]
2|234|Loss: 0.980384349822998:   4%|▎         | 233/6470 [02:48<1:11:10,  1.46it/s] 
2|234|Loss: 0.980384349822998:   4%|▎         | 234/6470 [02:49<1:12:12,  1.44it/s]
2|235|Loss: 0.9156076312065125:   4%|▎         | 234/6470 [02:49<1:12:12,  1.44it/s]
2|235|Loss: 0.9156076312065125:   4%|▎         | 235/6470 [02:50<1:13:15,  1.42it/s]
2|236|Loss: 0.7775726914405823:   4%|▎         | 235/6470 [02:50<1:13:15,  1.42it/s]
2|236|Loss: 0.7775726914405823:   4%|▎         | 236/6470 [02:50<1:09:50,  1.49it/s]
2|237|Loss: 0.8204666376113892:   4%|▎         | 236/6470 [02:51<1:09:50,  1.49it/s]
2|237|Loss: 0.8204666376113892:   4%|▎         | 237/6470 [02:51<1:12:38,  1.43it/s]
2|238|Loss: 0.9023943543434143:   4%|▎         | 237/6470 [02:51<1:12:38,  1.43it/s]
2|238|Loss: 0.9023943543434143:   4%|▎         | 238/6470 [02:52<1:15:49,  1.37it/s]
2|239|Loss: 0.7172465920448303:   4%|▎         | 238/6470 [02:52<1:15:49,  1.37it/s]
2|239|Loss: 0.7172465920448303:   4%|▎         | 239/6470 [02:53<1:19:57,  1.30it/s]
2|240|Loss: 0.9282411336898804:   4%|▎         | 239/6470 [02:53<1:19:57,  1.30it/s]
2|240|Loss: 0.9282411336898804:   4%|▎         | 240/6470 [02:53<1:21:09,  1.28it/s]
2|241|Loss: 0.8044249415397644:   4%|▎         | 240/6470 [02:54<1:21:09,  1.28it/s]
2|241|Loss: 0.8044249415397644:   4%|▎         | 241/6470 [02:54<1:19:11,  1.31it/s]
2|242|Loss: 0.8549536466598511:   4%|▎         | 241/6470 [02:54<1:19:11,  1.31it/s]
2|242|Loss: 0.8549536466598511:   4%|▎         | 242/6470 [02:55<1:19:49,  1.30it/s]
2|243|Loss: 0.8349253535270691:   4%|▎         | 242/6470 [02:55<1:19:49,  1.30it/s]
2|243|Loss: 0.8349253535270691:   4%|▍         | 243/6470 [02:56<1:18:39,  1.32it/s]
2|244|Loss: 0.7957119345664978:   4%|▍         | 243/6470 [02:56<1:18:39,  1.32it/s]
2|244|Loss: 0.7957119345664978:   4%|▍         | 244/6470 [02:56<1:14:13,  1.40it/s]
2|245|Loss: 0.8980386853218079:   4%|▍         | 244/6470 [02:57<1:14:13,  1.40it/s]
2|245|Loss: 0.8980386853218079:   4%|▍         | 245/6470 [02:57<1:12:04,  1.44it/s]
2|246|Loss: 0.8708791732788086:   4%|▍         | 245/6470 [02:57<1:12:04,  1.44it/s]
2|246|Loss: 0.8708791732788086:   4%|▍         | 246/6470 [02:58<1:13:28,  1.41it/s]
2|247|Loss: 0.803928017616272:   4%|▍         | 246/6470 [02:58<1:13:28,  1.41it/s] 
2|247|Loss: 0.803928017616272:   4%|▍         | 247/6470 [02:59<1:19:58,  1.30it/s]
2|248|Loss: 0.8334434032440186:   4%|▍         | 247/6470 [02:59<1:19:58,  1.30it/s]
2|248|Loss: 0.8334434032440186:   4%|▍         | 248/6470 [02:59<1:21:14,  1.28it/s]
2|249|Loss: 0.7836207151412964:   4%|▍         | 248/6470 [03:00<1:21:14,  1.28it/s]
2|249|Loss: 0.7836207151412964:   4%|▍         | 249/6470 [03:00<1:25:09,  1.22it/s]
2|250|Loss: 0.943544328212738:   4%|▍         | 249/6470 [03:01<1:25:09,  1.22it/s] 
2|250|Loss: 0.943544328212738:   4%|▍         | 250/6470 [03:01<1:22:01,  1.26it/s]
2|251|Loss: 0.7911151051521301:   4%|▍         | 250/6470 [03:01<1:22:01,  1.26it/s]
2|251|Loss: 0.7911151051521301:   4%|▍         | 251/6470 [03:02<1:23:54,  1.24it/s]
2|252|Loss: 0.8180892467498779:   4%|▍         | 251/6470 [03:02<1:23:54,  1.24it/s]
2|252|Loss: 0.8180892467498779:   4%|▍         | 252/6470 [03:03<1:26:38,  1.20it/s]
2|253|Loss: 1.049073576927185:   4%|▍         | 252/6470 [03:03<1:26:38,  1.20it/s] 
2|253|Loss: 1.049073576927185:   4%|▍         | 253/6470 [03:04<1:30:00,  1.15it/s]
2|254|Loss: 0.8438993096351624:   4%|▍         | 253/6470 [03:04<1:30:00,  1.15it/s]
2|254|Loss: 0.8438993096351624:   4%|▍         | 254/6470 [03:04<1:28:10,  1.17it/s]
2|255|Loss: 0.872452437877655:   4%|▍         | 254/6470 [03:05<1:28:10,  1.17it/s] 
2|255|Loss: 0.872452437877655:   4%|▍         | 255/6470 [03:05<1:23:40,  1.24it/s]
2|256|Loss: 0.8494833111763:   4%|▍         | 255/6470 [03:06<1:23:40,  1.24it/s]  
2|256|Loss: 0.8494833111763:   4%|▍         | 256/6470 [03:06<1:20:44,  1.28it/s]
2|257|Loss: 0.8763558864593506:   4%|▍         | 256/6470 [03:06<1:20:44,  1.28it/s]
2|257|Loss: 0.8763558864593506:   4%|▍         | 257/6470 [03:07<1:25:21,  1.21it/s]
2|258|Loss: 0.8608731627464294:   4%|▍         | 257/6470 [03:07<1:25:21,  1.21it/s]
2|258|Loss: 0.8608731627464294:   4%|▍         | 258/6470 [03:07<1:18:26,  1.32it/s]
2|259|Loss: 0.839116096496582:   4%|▍         | 258/6470 [03:08<1:18:26,  1.32it/s] 
2|259|Loss: 0.839116096496582:   4%|▍         | 259/6470 [03:08<1:14:07,  1.40it/s]
2|260|Loss: 0.8775779008865356:   4%|▍         | 259/6470 [03:08<1:14:07,  1.40it/s]
2|260|Loss: 0.8775779008865356:   4%|▍         | 260/6470 [03:09<1:14:56,  1.38it/s]
2|261|Loss: 0.7299087047576904:   4%|▍         | 260/6470 [03:09<1:14:56,  1.38it/s]
2|261|Loss: 0.7299087047576904:   4%|▍         | 261/6470 [03:10<1:17:13,  1.34it/s]
2|262|Loss: 0.7597379684448242:   4%|▍         | 261/6470 [03:10<1:17:13,  1.34it/s]
2|262|Loss: 0.7597379684448242:   4%|▍         | 262/6470 [03:10<1:14:35,  1.39it/s]
2|263|Loss: 0.9243589639663696:   4%|▍         | 262/6470 [03:11<1:14:35,  1.39it/s]
2|263|Loss: 0.9243589639663696:   4%|▍         | 263/6470 [03:11<1:21:27,  1.27it/s]
2|264|Loss: 0.9226365089416504:   4%|▍         | 263/6470 [03:12<1:21:27,  1.27it/s]
2|264|Loss: 0.9226365089416504:   4%|▍         | 264/6470 [03:12<1:18:57,  1.31it/s]
2|265|Loss: 0.9546836614608765:   4%|▍         | 264/6470 [03:12<1:18:57,  1.31it/s]
2|265|Loss: 0.9546836614608765:   4%|▍         | 265/6470 [03:13<1:20:57,  1.28it/s]
2|266|Loss: 0.9187583923339844:   4%|▍         | 265/6470 [03:13<1:20:57,  1.28it/s]
2|266|Loss: 0.9187583923339844:   4%|▍         | 266/6470 [03:14<1:23:40,  1.24it/s]
2|267|Loss: 0.9069213271141052:   4%|▍         | 266/6470 [03:14<1:23:40,  1.24it/s]
2|267|Loss: 0.9069213271141052:   4%|▍         | 267/6470 [03:14<1:23:45,  1.23it/s]
2|268|Loss: 0.9769405722618103:   4%|▍         | 267/6470 [03:15<1:23:45,  1.23it/s]
2|268|Loss: 0.9769405722618103:   4%|▍         | 268/6470 [03:15<1:21:13,  1.27it/s]
2|269|Loss: 0.7782060503959656:   4%|▍         | 268/6470 [03:15<1:21:13,  1.27it/s]
2|269|Loss: 0.7782060503959656:   4%|▍         | 269/6470 [03:16<1:18:07,  1.32it/s]
2|270|Loss: 0.9372918009757996:   4%|▍         | 269/6470 [03:16<1:18:07,  1.32it/s]
2|270|Loss: 0.9372918009757996:   4%|▍         | 270/6470 [03:17<1:16:26,  1.35it/s]
2|271|Loss: 0.800843358039856:   4%|▍         | 270/6470 [03:17<1:16:26,  1.35it/s] 
2|271|Loss: 0.800843358039856:   4%|▍         | 271/6470 [03:17<1:19:18,  1.30it/s]
2|272|Loss: 0.79546058177948:   4%|▍         | 271/6470 [03:18<1:19:18,  1.30it/s] 
2|272|Loss: 0.79546058177948:   4%|▍         | 272/6470 [03:18<1:24:39,  1.22it/s]
2|273|Loss: 0.8505662083625793:   4%|▍         | 272/6470 [03:19<1:24:39,  1.22it/s]
2|273|Loss: 0.8505662083625793:   4%|▍         | 273/6470 [03:19<1:19:41,  1.30it/s]
2|274|Loss: 0.9572278261184692:   4%|▍         | 273/6470 [03:19<1:19:41,  1.30it/s]
2|274|Loss: 0.9572278261184692:   4%|▍         | 274/6470 [03:20<1:16:18,  1.35it/s]
2|275|Loss: 0.8434631824493408:   4%|▍         | 274/6470 [03:20<1:16:18,  1.35it/s]
2|275|Loss: 0.8434631824493408:   4%|▍         | 275/6470 [03:20<1:19:40,  1.30it/s]
2|276|Loss: 0.9269572496414185:   4%|▍         | 275/6470 [03:21<1:19:40,  1.30it/s]
2|276|Loss: 0.9269572496414185:   4%|▍         | 276/6470 [03:21<1:15:17,  1.37it/s]
2|277|Loss: 0.9158272743225098:   4%|▍         | 276/6470 [03:22<1:15:17,  1.37it/s]
2|277|Loss: 0.9158272743225098:   4%|▍         | 277/6470 [03:22<1:19:54,  1.29it/s]
2|278|Loss: 0.8191898465156555:   4%|▍         | 277/6470 [03:22<1:19:54,  1.29it/s]
2|278|Loss: 0.8191898465156555:   4%|▍         | 278/6470 [03:23<1:18:53,  1.31it/s]
2|279|Loss: 0.847711980342865:   4%|▍         | 278/6470 [03:23<1:18:53,  1.31it/s] 
2|279|Loss: 0.847711980342865:   4%|▍         | 279/6470 [03:23<1:16:44,  1.34it/s]
2|280|Loss: 0.8971378207206726:   4%|▍         | 279/6470 [03:24<1:16:44,  1.34it/s]
2|280|Loss: 0.8971378207206726:   4%|▍         | 280/6470 [03:24<1:17:32,  1.33it/s]
2|281|Loss: 0.9186866879463196:   4%|▍         | 280/6470 [03:25<1:17:32,  1.33it/s]
2|281|Loss: 0.9186866879463196:   4%|▍         | 281/6470 [03:25<1:21:23,  1.27it/s]
2|282|Loss: 0.9269861578941345:   4%|▍         | 281/6470 [03:25<1:21:23,  1.27it/s]
2|282|Loss: 0.9269861578941345:   4%|▍         | 282/6470 [03:26<1:21:51,  1.26it/s]
2|283|Loss: 0.8435335755348206:   4%|▍         | 282/6470 [03:26<1:21:51,  1.26it/s]
2|283|Loss: 0.8435335755348206:   4%|▍         | 283/6470 [03:26<1:15:36,  1.36it/s]
2|284|Loss: 0.8284870386123657:   4%|▍         | 283/6470 [03:27<1:15:36,  1.36it/s]
2|284|Loss: 0.8284870386123657:   4%|▍         | 284/6470 [03:27<1:13:09,  1.41it/s]
2|285|Loss: 0.927946150302887:   4%|▍         | 284/6470 [03:28<1:13:09,  1.41it/s] 
2|285|Loss: 0.927946150302887:   4%|▍         | 285/6470 [03:28<1:15:27,  1.37it/s]
2|286|Loss: 0.8453920483589172:   4%|▍         | 285/6470 [03:28<1:15:27,  1.37it/s]
2|286|Loss: 0.8453920483589172:   4%|▍         | 286/6470 [03:29<1:13:12,  1.41it/s]
2|287|Loss: 0.9017038345336914:   4%|▍         | 286/6470 [03:29<1:13:12,  1.41it/s]
2|287|Loss: 0.9017038345336914:   4%|▍         | 287/6470 [03:29<1:13:46,  1.40it/s]
2|288|Loss: 0.942855179309845:   4%|▍         | 287/6470 [03:30<1:13:46,  1.40it/s] 
2|288|Loss: 0.942855179309845:   4%|▍         | 288/6470 [03:30<1:14:56,  1.37it/s]
2|289|Loss: 0.9326892495155334:   4%|▍         | 288/6470 [03:30<1:14:56,  1.37it/s]
2|289|Loss: 0.9326892495155334:   4%|▍         | 289/6470 [03:31<1:13:56,  1.39it/s]
2|290|Loss: 0.8479083776473999:   4%|▍         | 289/6470 [03:31<1:13:56,  1.39it/s]
2|290|Loss: 0.8479083776473999:   4%|▍         | 290/6470 [03:31<1:11:17,  1.44it/s]
2|291|Loss: 0.7538544535636902:   4%|▍         | 290/6470 [03:32<1:11:17,  1.44it/s]
2|291|Loss: 0.7538544535636902:   4%|▍         | 291/6470 [03:32<1:08:57,  1.49it/s]
2|292|Loss: 0.8891676068305969:   4%|▍         | 291/6470 [03:32<1:08:57,  1.49it/s]
2|292|Loss: 0.8891676068305969:   5%|▍         | 292/6470 [03:33<1:09:20,  1.48it/s]
2|293|Loss: 0.8872455954551697:   5%|▍         | 292/6470 [03:33<1:09:20,  1.48it/s]
2|293|Loss: 0.8872455954551697:   5%|▍         | 293/6470 [03:33<1:12:03,  1.43it/s]
2|294|Loss: 1.025630235671997:   5%|▍         | 293/6470 [03:34<1:12:03,  1.43it/s] 
2|294|Loss: 1.025630235671997:   5%|▍         | 294/6470 [03:34<1:17:41,  1.32it/s]
2|295|Loss: 0.9442448616027832:   5%|▍         | 294/6470 [03:35<1:17:41,  1.32it/s]
2|295|Loss: 0.9442448616027832:   5%|▍         | 295/6470 [03:35<1:18:45,  1.31it/s]
2|296|Loss: 0.8153459429740906:   5%|▍         | 295/6470 [03:35<1:18:45,  1.31it/s]
2|296|Loss: 0.8153459429740906:   5%|▍         | 296/6470 [03:36<1:16:16,  1.35it/s]
2|297|Loss: 0.8708704710006714:   5%|▍         | 296/6470 [03:36<1:16:16,  1.35it/s]
2|297|Loss: 0.8708704710006714:   5%|▍         | 297/6470 [03:37<1:16:11,  1.35it/s]
2|298|Loss: 0.7720577120780945:   5%|▍         | 297/6470 [03:37<1:16:11,  1.35it/s]
2|298|Loss: 0.7720577120780945:   5%|▍         | 298/6470 [03:37<1:14:06,  1.39it/s]
2|299|Loss: 0.7759312391281128:   5%|▍         | 298/6470 [03:38<1:14:06,  1.39it/s]
2|299|Loss: 0.7759312391281128:   5%|▍         | 299/6470 [03:38<1:13:18,  1.40it/s]
2|300|Loss: 0.9889050126075745:   5%|▍         | 299/6470 [03:38<1:13:18,  1.40it/s]
2|300|Loss: 0.9889050126075745:   5%|▍         | 300/6470 [03:39<1:15:21,  1.36it/s]
2|301|Loss: 0.9175450205802917:   5%|▍         | 300/6470 [03:39<1:15:21,  1.36it/s]
2|301|Loss: 0.9175450205802917:   5%|▍         | 301/6470 [03:39<1:17:25,  1.33it/s]
2|302|Loss: 0.7351785898208618:   5%|▍         | 301/6470 [03:40<1:17:25,  1.33it/s]
2|302|Loss: 0.7351785898208618:   5%|▍         | 302/6470 [03:40<1:21:08,  1.27it/s]
2|303|Loss: 0.844772219657898:   5%|▍         | 302/6470 [03:41<1:21:08,  1.27it/s] 
2|303|Loss: 0.844772219657898:   5%|▍         | 303/6470 [03:41<1:18:54,  1.30it/s]
2|304|Loss: 0.9675771594047546:   5%|▍         | 303/6470 [03:41<1:18:54,  1.30it/s]
2|304|Loss: 0.9675771594047546:   5%|▍         | 304/6470 [03:42<1:18:18,  1.31it/s]
2|305|Loss: 1.046526312828064:   5%|▍         | 304/6470 [03:42<1:18:18,  1.31it/s] 
2|305|Loss: 1.046526312828064:   5%|▍         | 305/6470 [03:43<1:21:36,  1.26it/s]
2|306|Loss: 0.8196993470191956:   5%|▍         | 305/6470 [03:43<1:21:36,  1.26it/s]
2|306|Loss: 0.8196993470191956:   5%|▍         | 306/6470 [03:43<1:19:50,  1.29it/s]
2|307|Loss: 0.8712809681892395:   5%|▍         | 306/6470 [03:44<1:19:50,  1.29it/s]
2|307|Loss: 0.8712809681892395:   5%|▍         | 307/6470 [03:44<1:16:21,  1.35it/s]
2|308|Loss: 0.8582481741905212:   5%|▍         | 307/6470 [03:44<1:16:21,  1.35it/s]
2|308|Loss: 0.8582481741905212:   5%|▍         | 308/6470 [03:44<1:06:47,  1.54it/s]
2|309|Loss: 0.7944261431694031:   5%|▍         | 308/6470 [03:45<1:06:47,  1.54it/s]
2|309|Loss: 0.7944261431694031:   5%|▍         | 309/6470 [03:45<1:12:08,  1.42it/s]
2|310|Loss: 0.9208038449287415:   5%|▍         | 309/6470 [03:46<1:12:08,  1.42it/s]
2|310|Loss: 0.9208038449287415:   5%|▍         | 310/6470 [03:46<1:16:38,  1.34it/s]
2|311|Loss: 0.9091403484344482:   5%|▍         | 310/6470 [03:47<1:16:38,  1.34it/s]
2|311|Loss: 0.9091403484344482:   5%|▍         | 311/6470 [03:47<1:16:11,  1.35it/s]
2|312|Loss: 0.7672385573387146:   5%|▍         | 311/6470 [03:47<1:16:11,  1.35it/s]
2|312|Loss: 0.7672385573387146:   5%|▍         | 312/6470 [03:48<1:20:07,  1.28it/s]
2|313|Loss: 0.9108700156211853:   5%|▍         | 312/6470 [03:48<1:20:07,  1.28it/s]
2|313|Loss: 0.9108700156211853:   5%|▍         | 313/6470 [03:49<1:21:02,  1.27it/s]
2|314|Loss: 0.8038317561149597:   5%|▍         | 313/6470 [03:49<1:21:02,  1.27it/s]
2|314|Loss: 0.8038317561149597:   5%|▍         | 314/6470 [03:49<1:24:36,  1.21it/s]
2|315|Loss: 0.8363363742828369:   5%|▍         | 314/6470 [03:50<1:24:36,  1.21it/s]
2|315|Loss: 0.8363363742828369:   5%|▍         | 315/6470 [03:50<1:29:05,  1.15it/s]
2|316|Loss: 0.8085547685623169:   5%|▍         | 315/6470 [03:51<1:29:05,  1.15it/s]
2|316|Loss: 0.8085547685623169:   5%|▍         | 316/6470 [03:51<1:28:40,  1.16it/s]
2|317|Loss: 0.8155708312988281:   5%|▍         | 316/6470 [03:52<1:28:40,  1.16it/s]
2|317|Loss: 0.8155708312988281:   5%|▍         | 317/6470 [03:52<1:22:47,  1.24it/s]
2|318|Loss: 0.7930759191513062:   5%|▍         | 317/6470 [03:52<1:22:47,  1.24it/s]
2|318|Loss: 0.7930759191513062:   5%|▍         | 318/6470 [03:53<1:21:19,  1.26it/s]
2|319|Loss: 0.8467912673950195:   5%|▍         | 318/6470 [03:53<1:21:19,  1.26it/s]
2|319|Loss: 0.8467912673950195:   5%|▍         | 319/6470 [03:54<1:21:14,  1.26it/s]
2|320|Loss: 0.9181691408157349:   5%|▍         | 319/6470 [03:54<1:21:14,  1.26it/s]
2|320|Loss: 0.9181691408157349:   5%|▍         | 320/6470 [03:54<1:15:49,  1.35it/s]
2|321|Loss: 0.7948079705238342:   5%|▍         | 320/6470 [03:55<1:15:49,  1.35it/s]
2|321|Loss: 0.7948079705238342:   5%|▍         | 321/6470 [03:55<1:17:19,  1.33it/s]
2|322|Loss: 0.6631450057029724:   5%|▍         | 321/6470 [03:55<1:17:19,  1.33it/s]
2|322|Loss: 0.6631450057029724:   5%|▍         | 322/6470 [03:56<1:20:16,  1.28it/s]
2|323|Loss: 0.8937600255012512:   5%|▍         | 322/6470 [03:56<1:20:16,  1.28it/s]
2|323|Loss: 0.8937600255012512:   5%|▍         | 323/6470 [03:57<1:20:38,  1.27it/s]
2|324|Loss: 0.8269271850585938:   5%|▍         | 323/6470 [03:57<1:20:38,  1.27it/s]
2|324|Loss: 0.8269271850585938:   5%|▌         | 324/6470 [03:57<1:22:54,  1.24it/s]
2|325|Loss: 0.6663625836372375:   5%|▌         | 324/6470 [03:58<1:22:54,  1.24it/s]
2|325|Loss: 0.6663625836372375:   5%|▌         | 325/6470 [03:58<1:22:54,  1.24it/s]
2|326|Loss: 0.7205515503883362:   5%|▌         | 325/6470 [03:59<1:22:54,  1.24it/s]
2|326|Loss: 0.7205515503883362:   5%|▌         | 326/6470 [03:59<1:22:35,  1.24it/s]
2|327|Loss: 0.822596549987793:   5%|▌         | 326/6470 [03:59<1:22:35,  1.24it/s] 
2|327|Loss: 0.822596549987793:   5%|▌         | 327/6470 [04:00<1:21:30,  1.26it/s]
2|328|Loss: 0.8053295016288757:   5%|▌         | 327/6470 [04:00<1:21:30,  1.26it/s]
2|328|Loss: 0.8053295016288757:   5%|▌         | 328/6470 [04:01<1:18:06,  1.31it/s]
2|329|Loss: 0.8182828426361084:   5%|▌         | 328/6470 [04:01<1:18:06,  1.31it/s]
2|329|Loss: 0.8182828426361084:   5%|▌         | 329/6470 [04:01<1:18:24,  1.31it/s]
2|330|Loss: 0.8240997791290283:   5%|▌         | 329/6470 [04:02<1:18:24,  1.31it/s]
2|330|Loss: 0.8240997791290283:   5%|▌         | 330/6470 [04:02<1:15:21,  1.36it/s]
2|331|Loss: 1.023073673248291:   5%|▌         | 330/6470 [04:02<1:15:21,  1.36it/s] 
2|331|Loss: 1.023073673248291:   5%|▌         | 331/6470 [04:03<1:13:20,  1.40it/s]
2|332|Loss: 0.8454577922821045:   5%|▌         | 331/6470 [04:03<1:13:20,  1.40it/s]
2|332|Loss: 0.8454577922821045:   5%|▌         | 332/6470 [04:03<1:15:09,  1.36it/s]
2|333|Loss: 0.8855029344558716:   5%|▌         | 332/6470 [04:04<1:15:09,  1.36it/s]
2|333|Loss: 0.8855029344558716:   5%|▌         | 333/6470 [04:04<1:17:02,  1.33it/s]
2|334|Loss: 0.9033262133598328:   5%|▌         | 333/6470 [04:05<1:17:02,  1.33it/s]
2|334|Loss: 0.9033262133598328:   5%|▌         | 334/6470 [04:05<1:13:07,  1.40it/s]
2|335|Loss: 0.9240754842758179:   5%|▌         | 334/6470 [04:05<1:13:07,  1.40it/s]
2|335|Loss: 0.9240754842758179:   5%|▌         | 335/6470 [04:06<1:13:28,  1.39it/s]
2|336|Loss: 0.7729331851005554:   5%|▌         | 335/6470 [04:06<1:13:28,  1.39it/s]
2|336|Loss: 0.7729331851005554:   5%|▌         | 336/6470 [04:06<1:09:21,  1.47it/s]
2|337|Loss: 0.8247726559638977:   5%|▌         | 336/6470 [04:06<1:09:21,  1.47it/s]
2|337|Loss: 0.8247726559638977:   5%|▌         | 337/6470 [04:07<1:03:34,  1.61it/s]
2|338|Loss: 0.8926002979278564:   5%|▌         | 337/6470 [04:07<1:03:34,  1.61it/s]
2|338|Loss: 0.8926002979278564:   5%|▌         | 338/6470 [04:07<1:05:48,  1.55it/s]
2|339|Loss: 0.8207629323005676:   5%|▌         | 338/6470 [04:08<1:05:48,  1.55it/s]
2|339|Loss: 0.8207629323005676:   5%|▌         | 339/6470 [04:08<1:09:35,  1.47it/s]
2|340|Loss: 0.8713569641113281:   5%|▌         | 339/6470 [04:08<1:09:35,  1.47it/s]
2|340|Loss: 0.8713569641113281:   5%|▌         | 340/6470 [04:09<1:03:07,  1.62it/s]
2|341|Loss: 0.7466345429420471:   5%|▌         | 340/6470 [04:09<1:03:07,  1.62it/s]
2|341|Loss: 0.7466345429420471:   5%|▌         | 341/6470 [04:09<1:05:24,  1.56it/s]
2|342|Loss: 0.8028485774993896:   5%|▌         | 341/6470 [04:10<1:05:24,  1.56it/s]
2|342|Loss: 0.8028485774993896:   5%|▌         | 342/6470 [04:10<1:08:59,  1.48it/s]
2|343|Loss: 0.9988619089126587:   5%|▌         | 342/6470 [04:10<1:08:59,  1.48it/s]
2|343|Loss: 0.9988619089126587:   5%|▌         | 343/6470 [04:11<1:15:39,  1.35it/s]
2|344|Loss: 0.821745753288269:   5%|▌         | 343/6470 [04:11<1:15:39,  1.35it/s] 
2|344|Loss: 0.821745753288269:   5%|▌         | 344/6470 [04:12<1:20:25,  1.27it/s]
2|345|Loss: 0.7626350522041321:   5%|▌         | 344/6470 [04:12<1:20:25,  1.27it/s]
2|345|Loss: 0.7626350522041321:   5%|▌         | 345/6470 [04:13<1:18:40,  1.30it/s]
2|346|Loss: 0.7580099701881409:   5%|▌         | 345/6470 [04:13<1:18:40,  1.30it/s]
2|346|Loss: 0.7580099701881409:   5%|▌         | 346/6470 [04:13<1:18:08,  1.31it/s]
2|347|Loss: 0.9355263710021973:   5%|▌         | 346/6470 [04:14<1:18:08,  1.31it/s]
2|347|Loss: 0.9355263710021973:   5%|▌         | 347/6470 [04:14<1:15:31,  1.35it/s]
2|348|Loss: 0.720088541507721:   5%|▌         | 347/6470 [04:14<1:15:31,  1.35it/s] 
2|348|Loss: 0.720088541507721:   5%|▌         | 348/6470 [04:15<1:13:23,  1.39it/s]
2|349|Loss: 1.036729335784912:   5%|▌         | 348/6470 [04:15<1:13:23,  1.39it/s]
2|349|Loss: 1.036729335784912:   5%|▌         | 349/6470 [04:15<1:10:47,  1.44it/s]
2|350|Loss: 0.7752320170402527:   5%|▌         | 349/6470 [04:16<1:10:47,  1.44it/s]
2|350|Loss: 0.7752320170402527:   5%|▌         | 350/6470 [04:16<1:16:01,  1.34it/s]
2|351|Loss: 1.0363905429840088:   5%|▌         | 350/6470 [04:17<1:16:01,  1.34it/s]
2|351|Loss: 1.0363905429840088:   5%|▌         | 351/6470 [04:17<1:18:55,  1.29it/s]
2|352|Loss: 0.8095101714134216:   5%|▌         | 351/6470 [04:17<1:18:55,  1.29it/s]
2|352|Loss: 0.8095101714134216:   5%|▌         | 352/6470 [04:18<1:19:09,  1.29it/s]
2|353|Loss: 0.7810435891151428:   5%|▌         | 352/6470 [04:18<1:19:09,  1.29it/s]
2|353|Loss: 0.7810435891151428:   5%|▌         | 353/6470 [04:19<1:22:42,  1.23it/s]
2|354|Loss: 0.9158432483673096:   5%|▌         | 353/6470 [04:19<1:22:42,  1.23it/s]
2|354|Loss: 0.9158432483673096:   5%|▌         | 354/6470 [04:19<1:22:20,  1.24it/s]
2|355|Loss: 0.8980720043182373:   5%|▌         | 354/6470 [04:20<1:22:20,  1.24it/s]
2|355|Loss: 0.8980720043182373:   5%|▌         | 355/6470 [04:20<1:20:53,  1.26it/s]
2|356|Loss: 0.7116811275482178:   5%|▌         | 355/6470 [04:21<1:20:53,  1.26it/s]
2|356|Loss: 0.7116811275482178:   6%|▌         | 356/6470 [04:21<1:14:57,  1.36it/s]
2|357|Loss: 0.8213507533073425:   6%|▌         | 356/6470 [04:21<1:14:57,  1.36it/s]
2|357|Loss: 0.8213507533073425:   6%|▌         | 357/6470 [04:22<1:16:02,  1.34it/s]
2|358|Loss: 0.8694670796394348:   6%|▌         | 357/6470 [04:22<1:16:02,  1.34it/s]
2|358|Loss: 0.8694670796394348:   6%|▌         | 358/6470 [04:22<1:20:56,  1.26it/s]
2|359|Loss: 0.8248348832130432:   6%|▌         | 358/6470 [04:23<1:20:56,  1.26it/s]
2|359|Loss: 0.8248348832130432:   6%|▌         | 359/6470 [04:23<1:24:37,  1.20it/s]
2|360|Loss: 0.8844664692878723:   6%|▌         | 359/6470 [04:24<1:24:37,  1.20it/s]
2|360|Loss: 0.8844664692878723:   6%|▌         | 360/6470 [04:24<1:24:27,  1.21it/s]
2|361|Loss: 0.9283483028411865:   6%|▌         | 360/6470 [04:25<1:24:27,  1.21it/s]
2|361|Loss: 0.9283483028411865:   6%|▌         | 361/6470 [04:25<1:26:16,  1.18it/s]
2|362|Loss: 0.8744041919708252:   6%|▌         | 361/6470 [04:26<1:26:16,  1.18it/s]
2|362|Loss: 0.8744041919708252:   6%|▌         | 362/6470 [04:26<1:22:21,  1.24it/s]
2|363|Loss: 0.8652094602584839:   6%|▌         | 362/6470 [04:26<1:22:21,  1.24it/s]
2|363|Loss: 0.8652094602584839:   6%|▌         | 363/6470 [04:27<1:27:35,  1.16it/s]
2|364|Loss: 0.8097918629646301:   6%|▌         | 363/6470 [04:27<1:27:35,  1.16it/s]
2|364|Loss: 0.8097918629646301:   6%|▌         | 364/6470 [04:28<1:28:51,  1.15it/s]
2|365|Loss: 0.8096482157707214:   6%|▌         | 364/6470 [04:28<1:28:51,  1.15it/s]
2|365|Loss: 0.8096482157707214:   6%|▌         | 365/6470 [04:29<1:26:25,  1.18it/s]
2|366|Loss: 0.7853010892868042:   6%|▌         | 365/6470 [04:29<1:26:25,  1.18it/s]
2|366|Loss: 0.7853010892868042:   6%|▌         | 366/6470 [04:29<1:22:02,  1.24it/s]
2|367|Loss: 0.7798118591308594:   6%|▌         | 366/6470 [04:30<1:22:02,  1.24it/s]
2|367|Loss: 0.7798118591308594:   6%|▌         | 367/6470 [04:30<1:21:08,  1.25it/s]
2|368|Loss: 0.8637479543685913:   6%|▌         | 367/6470 [04:30<1:21:08,  1.25it/s]
2|368|Loss: 0.8637479543685913:   6%|▌         | 368/6470 [04:31<1:20:30,  1.26it/s]
2|369|Loss: 0.7755391001701355:   6%|▌         | 368/6470 [04:31<1:20:30,  1.26it/s]
2|369|Loss: 0.7755391001701355:   6%|▌         | 369/6470 [04:32<1:20:58,  1.26it/s]
2|370|Loss: 0.9739825129508972:   6%|▌         | 369/6470 [04:32<1:20:58,  1.26it/s]
2|370|Loss: 0.9739825129508972:   6%|▌         | 370/6470 [04:32<1:18:35,  1.29it/s]
2|371|Loss: 0.6901883482933044:   6%|▌         | 370/6470 [04:33<1:18:35,  1.29it/s]
2|371|Loss: 0.6901883482933044:   6%|▌         | 371/6470 [04:33<1:14:42,  1.36it/s]
2|372|Loss: 0.9728467464447021:   6%|▌         | 371/6470 [04:33<1:14:42,  1.36it/s]
2|372|Loss: 0.9728467464447021:   6%|▌         | 372/6470 [04:34<1:19:06,  1.28it/s]
2|373|Loss: 0.7684304118156433:   6%|▌         | 372/6470 [04:34<1:19:06,  1.28it/s]
2|373|Loss: 0.7684304118156433:   6%|▌         | 373/6470 [04:34<1:15:10,  1.35it/s]
2|374|Loss: 1.0482103824615479:   6%|▌         | 373/6470 [04:35<1:15:10,  1.35it/s]
2|374|Loss: 1.0482103824615479:   6%|▌         | 374/6470 [04:35<1:10:55,  1.43it/s]
2|375|Loss: 0.8537620306015015:   6%|▌         | 374/6470 [04:35<1:10:55,  1.43it/s]
2|375|Loss: 0.8537620306015015:   6%|▌         | 375/6470 [04:36<1:06:42,  1.52it/s]
2|376|Loss: 0.8480246663093567:   6%|▌         | 375/6470 [04:36<1:06:42,  1.52it/s]
2|376|Loss: 0.8480246663093567:   6%|▌         | 376/6470 [04:36<1:09:20,  1.46it/s]
2|377|Loss: 0.827190637588501:   6%|▌         | 376/6470 [04:37<1:09:20,  1.46it/s] 
2|377|Loss: 0.827190637588501:   6%|▌         | 377/6470 [04:37<1:16:43,  1.32it/s]
2|378|Loss: 0.8374374508857727:   6%|▌         | 377/6470 [04:38<1:16:43,  1.32it/s]
2|378|Loss: 0.8374374508857727:   6%|▌         | 378/6470 [04:38<1:15:11,  1.35it/s]
2|379|Loss: 0.7085944414138794:   6%|▌         | 378/6470 [04:38<1:15:11,  1.35it/s]
2|379|Loss: 0.7085944414138794:   6%|▌         | 379/6470 [04:39<1:16:14,  1.33it/s]
2|380|Loss: 0.8444507122039795:   6%|▌         | 379/6470 [04:39<1:16:14,  1.33it/s]
2|380|Loss: 0.8444507122039795:   6%|▌         | 380/6470 [04:40<1:17:44,  1.31it/s]
2|381|Loss: 0.8820899724960327:   6%|▌         | 380/6470 [04:40<1:17:44,  1.31it/s]
2|381|Loss: 0.8820899724960327:   6%|▌         | 381/6470 [04:40<1:18:34,  1.29it/s]
2|382|Loss: 0.7095723748207092:   6%|▌         | 381/6470 [04:41<1:18:34,  1.29it/s]
2|382|Loss: 0.7095723748207092:   6%|▌         | 382/6470 [04:41<1:15:52,  1.34it/s]
2|383|Loss: 0.8592497110366821:   6%|▌         | 382/6470 [04:41<1:15:52,  1.34it/s]
2|383|Loss: 0.8592497110366821:   6%|▌         | 383/6470 [04:42<1:15:31,  1.34it/s]
2|384|Loss: 0.8049996495246887:   6%|▌         | 383/6470 [04:42<1:15:31,  1.34it/s]
2|384|Loss: 0.8049996495246887:   6%|▌         | 384/6470 [04:43<1:16:36,  1.32it/s]
2|385|Loss: 0.8184033632278442:   6%|▌         | 384/6470 [04:43<1:16:36,  1.32it/s]
2|385|Loss: 0.8184033632278442:   6%|▌         | 385/6470 [04:43<1:21:01,  1.25it/s]
2|386|Loss: 0.7640979886054993:   6%|▌         | 385/6470 [04:44<1:21:01,  1.25it/s]
2|386|Loss: 0.7640979886054993:   6%|▌         | 386/6470 [04:44<1:24:43,  1.20it/s]
2|387|Loss: 0.9802271723747253:   6%|▌         | 386/6470 [04:45<1:24:43,  1.20it/s]
2|387|Loss: 0.9802271723747253:   6%|▌         | 387/6470 [04:45<1:30:39,  1.12it/s]
2|388|Loss: 0.7794259786605835:   6%|▌         | 387/6470 [04:46<1:30:39,  1.12it/s]
2|388|Loss: 0.7794259786605835:   6%|▌         | 388/6470 [04:46<1:27:21,  1.16it/s]
2|389|Loss: 0.8729349970817566:   6%|▌         | 388/6470 [04:47<1:27:21,  1.16it/s]
2|389|Loss: 0.8729349970817566:   6%|▌         | 389/6470 [04:47<1:22:25,  1.23it/s]
2|390|Loss: 0.7482235431671143:   6%|▌         | 389/6470 [04:47<1:22:25,  1.23it/s]
2|390|Loss: 0.7482235431671143:   6%|▌         | 390/6470 [04:48<1:21:13,  1.25it/s]
2|391|Loss: 0.7943955659866333:   6%|▌         | 390/6470 [04:48<1:21:13,  1.25it/s]
2|391|Loss: 0.7943955659866333:   6%|▌         | 391/6470 [04:48<1:18:40,  1.29it/s]
2|392|Loss: 0.73499596118927:   6%|▌         | 391/6470 [04:49<1:18:40,  1.29it/s]  
2|392|Loss: 0.73499596118927:   6%|▌         | 392/6470 [04:49<1:11:37,  1.41it/s]
2|393|Loss: 0.8314815759658813:   6%|▌         | 392/6470 [04:49<1:11:37,  1.41it/s]
2|393|Loss: 0.8314815759658813:   6%|▌         | 393/6470 [04:50<1:06:22,  1.53it/s]
2|394|Loss: 0.907808244228363:   6%|▌         | 393/6470 [04:50<1:06:22,  1.53it/s] 
2|394|Loss: 0.907808244228363:   6%|▌         | 394/6470 [04:50<1:09:03,  1.47it/s]
2|395|Loss: 0.8905548453330994:   6%|▌         | 394/6470 [04:51<1:09:03,  1.47it/s]
2|395|Loss: 0.8905548453330994:   6%|▌         | 395/6470 [04:51<1:07:57,  1.49it/s]
2|396|Loss: 0.9573745131492615:   6%|▌         | 395/6470 [04:51<1:07:57,  1.49it/s]
2|396|Loss: 0.9573745131492615:   6%|▌         | 396/6470 [04:51<1:02:51,  1.61it/s]
2|397|Loss: 0.8424157500267029:   6%|▌         | 396/6470 [04:52<1:02:51,  1.61it/s]
2|397|Loss: 0.8424157500267029:   6%|▌         | 397/6470 [04:52<1:03:49,  1.59it/s]
2|398|Loss: 0.8250817060470581:   6%|▌         | 397/6470 [04:52<1:03:49,  1.59it/s]
2|398|Loss: 0.8250817060470581:   6%|▌         | 398/6470 [04:53<1:05:14,  1.55it/s]
2|399|Loss: 0.9060521125793457:   6%|▌         | 398/6470 [04:53<1:05:14,  1.55it/s]
2|399|Loss: 0.9060521125793457:   6%|▌         | 399/6470 [04:53<1:08:33,  1.48it/s]
2|400|Loss: 0.8435655832290649:   6%|▌         | 399/6470 [04:54<1:08:33,  1.48it/s]
2|400|Loss: 0.8435655832290649:   6%|▌         | 400/6470 [04:54<1:13:38,  1.37it/s]
2|401|Loss: 0.909598171710968:   6%|▌         | 400/6470 [04:55<1:13:38,  1.37it/s] 
2|401|Loss: 0.909598171710968:   6%|▌         | 401/6470 [04:55<1:19:38,  1.27it/s]
2|402|Loss: 0.6688686013221741:   6%|▌         | 401/6470 [04:56<1:19:38,  1.27it/s]
2|402|Loss: 0.6688686013221741:   6%|▌         | 402/6470 [04:56<1:18:47,  1.28it/s]
2|403|Loss: 0.814602255821228:   6%|▌         | 402/6470 [04:56<1:18:47,  1.28it/s] 
2|403|Loss: 0.814602255821228:   6%|▌         | 403/6470 [04:57<1:17:47,  1.30it/s]
2|404|Loss: 0.9571389555931091:   6%|▌         | 403/6470 [04:57<1:17:47,  1.30it/s]
2|404|Loss: 0.9571389555931091:   6%|▌         | 404/6470 [04:57<1:13:14,  1.38it/s]
2|405|Loss: 0.7976601719856262:   6%|▌         | 404/6470 [04:58<1:13:14,  1.38it/s]
2|405|Loss: 0.7976601719856262:   6%|▋         | 405/6470 [04:58<1:09:56,  1.45it/s]
2|406|Loss: 0.6759331822395325:   6%|▋         | 405/6470 [04:58<1:09:56,  1.45it/s]
2|406|Loss: 0.6759331822395325:   6%|▋         | 406/6470 [04:59<1:18:23,  1.29it/s]
2|407|Loss: 0.8913782835006714:   6%|▋         | 406/6470 [04:59<1:18:23,  1.29it/s]
2|407|Loss: 0.8913782835006714:   6%|▋         | 407/6470 [05:00<1:20:18,  1.26it/s]
2|408|Loss: 0.9392579793930054:   6%|▋         | 407/6470 [05:00<1:20:18,  1.26it/s]
2|408|Loss: 0.9392579793930054:   6%|▋         | 408/6470 [05:00<1:16:01,  1.33it/s]
2|409|Loss: 0.9533504247665405:   6%|▋         | 408/6470 [05:01<1:16:01,  1.33it/s]
2|409|Loss: 0.9533504247665405:   6%|▋         | 409/6470 [05:01<1:16:17,  1.32it/s]
2|410|Loss: 0.7898666858673096:   6%|▋         | 409/6470 [05:02<1:16:17,  1.32it/s]
2|410|Loss: 0.7898666858673096:   6%|▋         | 410/6470 [05:02<1:15:15,  1.34it/s]
2|411|Loss: 0.7035272717475891:   6%|▋         | 410/6470 [05:02<1:15:15,  1.34it/s]
2|411|Loss: 0.7035272717475891:   6%|▋         | 411/6470 [05:03<1:12:09,  1.40it/s]
2|412|Loss: 0.9448913335800171:   6%|▋         | 411/6470 [05:03<1:12:09,  1.40it/s]
2|412|Loss: 0.9448913335800171:   6%|▋         | 412/6470 [05:03<1:09:55,  1.44it/s]
2|413|Loss: 0.7996057868003845:   6%|▋         | 412/6470 [05:04<1:09:55,  1.44it/s]
2|413|Loss: 0.7996057868003845:   6%|▋         | 413/6470 [05:04<1:09:51,  1.44it/s]
2|414|Loss: 0.8840843439102173:   6%|▋         | 413/6470 [05:04<1:09:51,  1.44it/s]
2|414|Loss: 0.8840843439102173:   6%|▋         | 414/6470 [05:05<1:12:54,  1.38it/s]
2|415|Loss: 0.8232811093330383:   6%|▋         | 414/6470 [05:05<1:12:54,  1.38it/s]
2|415|Loss: 0.8232811093330383:   6%|▋         | 415/6470 [05:05<1:13:13,  1.38it/s]
2|416|Loss: 0.734003484249115:   6%|▋         | 415/6470 [05:06<1:13:13,  1.38it/s] 
2|416|Loss: 0.734003484249115:   6%|▋         | 416/6470 [05:06<1:10:45,  1.43it/s]
2|417|Loss: 0.8969544172286987:   6%|▋         | 416/6470 [05:07<1:10:45,  1.43it/s]
2|417|Loss: 0.8969544172286987:   6%|▋         | 417/6470 [05:07<1:13:12,  1.38it/s]
2|418|Loss: 0.7514613270759583:   6%|▋         | 417/6470 [05:07<1:13:12,  1.38it/s]
2|418|Loss: 0.7514613270759583:   6%|▋         | 418/6470 [05:07<1:07:53,  1.49it/s]
2|419|Loss: 0.9611724019050598:   6%|▋         | 418/6470 [05:08<1:07:53,  1.49it/s]
2|419|Loss: 0.9611724019050598:   6%|▋         | 419/6470 [05:08<1:06:17,  1.52it/s]
2|420|Loss: 0.9162818789482117:   6%|▋         | 419/6470 [05:08<1:06:17,  1.52it/s]
2|420|Loss: 0.9162818789482117:   6%|▋         | 420/6470 [05:09<1:07:55,  1.48it/s]
2|421|Loss: 0.8362702131271362:   6%|▋         | 420/6470 [05:09<1:07:55,  1.48it/s]
2|421|Loss: 0.8362702131271362:   7%|▋         | 421/6470 [05:10<1:10:14,  1.44it/s]
2|422|Loss: 0.9714763164520264:   7%|▋         | 421/6470 [05:10<1:10:14,  1.44it/s]
2|422|Loss: 0.9714763164520264:   7%|▋         | 422/6470 [05:10<1:07:24,  1.50it/s]
2|423|Loss: 0.8164536356925964:   7%|▋         | 422/6470 [05:10<1:07:24,  1.50it/s]
2|423|Loss: 0.8164536356925964:   7%|▋         | 423/6470 [05:11<1:09:49,  1.44it/s]
2|424|Loss: 0.8623970150947571:   7%|▋         | 423/6470 [05:11<1:09:49,  1.44it/s]
2|424|Loss: 0.8623970150947571:   7%|▋         | 424/6470 [05:12<1:15:03,  1.34it/s]
2|425|Loss: 0.8615316152572632:   7%|▋         | 424/6470 [05:12<1:15:03,  1.34it/s]
2|425|Loss: 0.8615316152572632:   7%|▋         | 425/6470 [05:12<1:14:46,  1.35it/s]
2|426|Loss: 0.7979563474655151:   7%|▋         | 425/6470 [05:13<1:14:46,  1.35it/s]
2|426|Loss: 0.7979563474655151:   7%|▋         | 426/6470 [05:13<1:14:52,  1.35it/s]
2|427|Loss: 0.9289580583572388:   7%|▋         | 426/6470 [05:14<1:14:52,  1.35it/s]
2|427|Loss: 0.9289580583572388:   7%|▋         | 427/6470 [05:14<1:17:24,  1.30it/s]
2|428|Loss: 0.7924616932868958:   7%|▋         | 427/6470 [05:15<1:17:24,  1.30it/s]
2|428|Loss: 0.7924616932868958:   7%|▋         | 428/6470 [05:15<1:20:26,  1.25it/s]
2|429|Loss: 0.8490580916404724:   7%|▋         | 428/6470 [05:15<1:20:26,  1.25it/s]
2|429|Loss: 0.8490580916404724:   7%|▋         | 429/6470 [05:16<1:24:11,  1.20it/s]
2|430|Loss: 0.9396153092384338:   7%|▋         | 429/6470 [05:16<1:24:11,  1.20it/s]
2|430|Loss: 0.9396153092384338:   7%|▋         | 430/6470 [05:17<1:23:49,  1.20it/s]
2|431|Loss: 1.0299111604690552:   7%|▋         | 430/6470 [05:17<1:23:49,  1.20it/s]
2|431|Loss: 1.0299111604690552:   7%|▋         | 431/6470 [05:18<1:25:24,  1.18it/s]
2|432|Loss: 0.8783791661262512:   7%|▋         | 431/6470 [05:18<1:25:24,  1.18it/s]
2|432|Loss: 0.8783791661262512:   7%|▋         | 432/6470 [05:18<1:25:58,  1.17it/s]
2|433|Loss: 0.978173553943634:   7%|▋         | 432/6470 [05:19<1:25:58,  1.17it/s] 
2|433|Loss: 0.978173553943634:   7%|▋         | 433/6470 [05:19<1:24:55,  1.18it/s]
2|434|Loss: 0.7514738440513611:   7%|▋         | 433/6470 [05:20<1:24:55,  1.18it/s]
2|434|Loss: 0.7514738440513611:   7%|▋         | 434/6470 [05:20<1:19:33,  1.26it/s]
2|435|Loss: 0.9101388454437256:   7%|▋         | 434/6470 [05:20<1:19:33,  1.26it/s]
2|435|Loss: 0.9101388454437256:   7%|▋         | 435/6470 [05:21<1:16:28,  1.32it/s]
2|436|Loss: 0.7626568675041199:   7%|▋         | 435/6470 [05:21<1:16:28,  1.32it/s]
2|436|Loss: 0.7626568675041199:   7%|▋         | 436/6470 [05:21<1:14:34,  1.35it/s]
2|437|Loss: 0.8563455939292908:   7%|▋         | 436/6470 [05:22<1:14:34,  1.35it/s]
2|437|Loss: 0.8563455939292908:   7%|▋         | 437/6470 [05:22<1:14:43,  1.35it/s]
2|438|Loss: 0.9058101773262024:   7%|▋         | 437/6470 [05:22<1:14:43,  1.35it/s]
2|438|Loss: 0.9058101773262024:   7%|▋         | 438/6470 [05:23<1:13:46,  1.36it/s]
2|439|Loss: 0.8839215040206909:   7%|▋         | 438/6470 [05:23<1:13:46,  1.36it/s]
2|439|Loss: 0.8839215040206909:   7%|▋         | 439/6470 [05:23<1:14:30,  1.35it/s]
2|440|Loss: 0.9906385540962219:   7%|▋         | 439/6470 [05:24<1:14:30,  1.35it/s]
2|440|Loss: 0.9906385540962219:   7%|▋         | 440/6470 [05:24<1:15:56,  1.32it/s]
2|441|Loss: 0.7678383588790894:   7%|▋         | 440/6470 [05:25<1:15:56,  1.32it/s]
2|441|Loss: 0.7678383588790894:   7%|▋         | 441/6470 [05:25<1:15:04,  1.34it/s]
2|442|Loss: 0.8457224369049072:   7%|▋         | 441/6470 [05:25<1:15:04,  1.34it/s]
2|442|Loss: 0.8457224369049072:   7%|▋         | 442/6470 [05:26<1:12:59,  1.38it/s]
2|443|Loss: 0.8545525670051575:   7%|▋         | 442/6470 [05:26<1:12:59,  1.38it/s]
2|443|Loss: 0.8545525670051575:   7%|▋         | 443/6470 [05:26<1:12:01,  1.39it/s]
2|444|Loss: 0.9463760256767273:   7%|▋         | 443/6470 [05:27<1:12:01,  1.39it/s]
2|444|Loss: 0.9463760256767273:   7%|▋         | 444/6470 [05:27<1:16:31,  1.31it/s]
2|445|Loss: 0.835648775100708:   7%|▋         | 444/6470 [05:28<1:16:31,  1.31it/s] 
2|445|Loss: 0.835648775100708:   7%|▋         | 445/6470 [05:28<1:19:28,  1.26it/s]
2|446|Loss: 0.7842841148376465:   7%|▋         | 445/6470 [05:29<1:19:28,  1.26it/s]
2|446|Loss: 0.7842841148376465:   7%|▋         | 446/6470 [05:29<1:20:07,  1.25it/s]
2|447|Loss: 0.7622565031051636:   7%|▋         | 446/6470 [05:29<1:20:07,  1.25it/s]
2|447|Loss: 0.7622565031051636:   7%|▋         | 447/6470 [05:30<1:20:00,  1.25it/s]
2|448|Loss: 0.8132761120796204:   7%|▋         | 447/6470 [05:30<1:20:00,  1.25it/s]
2|448|Loss: 0.8132761120796204:   7%|▋         | 448/6470 [05:31<1:23:48,  1.20it/s]
2|449|Loss: 0.8286646008491516:   7%|▋         | 448/6470 [05:31<1:23:48,  1.20it/s]
2|449|Loss: 0.8286646008491516:   7%|▋         | 449/6470 [05:31<1:23:56,  1.20it/s]
2|450|Loss: 0.7357479929924011:   7%|▋         | 449/6470 [05:32<1:23:56,  1.20it/s]
2|450|Loss: 0.7357479929924011:   7%|▋         | 450/6470 [05:32<1:18:43,  1.27it/s]
2|451|Loss: 0.9330026507377625:   7%|▋         | 450/6470 [05:33<1:18:43,  1.27it/s]
2|451|Loss: 0.9330026507377625:   7%|▋         | 451/6470 [05:33<1:18:05,  1.28it/s]
2|452|Loss: 0.9493913650512695:   7%|▋         | 451/6470 [05:33<1:18:05,  1.28it/s]
2|452|Loss: 0.9493913650512695:   7%|▋         | 452/6470 [05:34<1:13:22,  1.37it/s]
2|453|Loss: 0.8027662038803101:   7%|▋         | 452/6470 [05:34<1:13:22,  1.37it/s]
2|453|Loss: 0.8027662038803101:   7%|▋         | 453/6470 [05:34<1:13:55,  1.36it/s]
2|454|Loss: 0.9356964230537415:   7%|▋         | 453/6470 [05:35<1:13:55,  1.36it/s]
2|454|Loss: 0.9356964230537415:   7%|▋         | 454/6470 [05:35<1:17:49,  1.29it/s]
2|455|Loss: 0.8396832942962646:   7%|▋         | 454/6470 [05:36<1:17:49,  1.29it/s]
2|455|Loss: 0.8396832942962646:   7%|▋         | 455/6470 [05:36<1:21:44,  1.23it/s]
2|456|Loss: 0.8881262540817261:   7%|▋         | 455/6470 [05:36<1:21:44,  1.23it/s]
2|456|Loss: 0.8881262540817261:   7%|▋         | 456/6470 [05:37<1:17:46,  1.29it/s]
2|457|Loss: 0.8710392713546753:   7%|▋         | 456/6470 [05:37<1:17:46,  1.29it/s]
2|457|Loss: 0.8710392713546753:   7%|▋         | 457/6470 [05:37<1:11:24,  1.40it/s]
2|458|Loss: 0.9859424233436584:   7%|▋         | 457/6470 [05:38<1:11:24,  1.40it/s]
2|458|Loss: 0.9859424233436584:   7%|▋         | 458/6470 [05:38<1:11:38,  1.40it/s]
2|459|Loss: 0.8991819024085999:   7%|▋         | 458/6470 [05:38<1:11:38,  1.40it/s]
2|459|Loss: 0.8991819024085999:   7%|▋         | 459/6470 [05:39<1:06:28,  1.51it/s]
2|460|Loss: 0.887897253036499:   7%|▋         | 459/6470 [05:39<1:06:28,  1.51it/s] 
2|460|Loss: 0.887897253036499:   7%|▋         | 460/6470 [05:39<1:04:14,  1.56it/s]
2|461|Loss: 0.8570340871810913:   7%|▋         | 460/6470 [05:40<1:04:14,  1.56it/s]
2|461|Loss: 0.8570340871810913:   7%|▋         | 461/6470 [05:40<1:08:57,  1.45it/s]
2|462|Loss: 0.7443957924842834:   7%|▋         | 461/6470 [05:40<1:08:57,  1.45it/s]
2|462|Loss: 0.7443957924842834:   7%|▋         | 462/6470 [05:41<1:11:32,  1.40it/s]
2|463|Loss: 0.7643794417381287:   7%|▋         | 462/6470 [05:41<1:11:32,  1.40it/s]
2|463|Loss: 0.7643794417381287:   7%|▋         | 463/6470 [05:42<1:12:59,  1.37it/s]
2|464|Loss: 0.8964233994483948:   7%|▋         | 463/6470 [05:42<1:12:59,  1.37it/s]
2|464|Loss: 0.8964233994483948:   7%|▋         | 464/6470 [05:42<1:12:46,  1.38it/s]
2|465|Loss: 0.8002004027366638:   7%|▋         | 464/6470 [05:43<1:12:46,  1.38it/s]
2|465|Loss: 0.8002004027366638:   7%|▋         | 465/6470 [05:43<1:14:14,  1.35it/s]
2|466|Loss: 0.7621258497238159:   7%|▋         | 465/6470 [05:43<1:14:14,  1.35it/s]
2|466|Loss: 0.7621258497238159:   7%|▋         | 466/6470 [05:44<1:12:34,  1.38it/s]
2|467|Loss: 0.8546074628829956:   7%|▋         | 466/6470 [05:44<1:12:34,  1.38it/s]
2|467|Loss: 0.8546074628829956:   7%|▋         | 467/6470 [05:44<1:14:24,  1.34it/s]
2|468|Loss: 0.7003669142723083:   7%|▋         | 467/6470 [05:45<1:14:24,  1.34it/s]
2|468|Loss: 0.7003669142723083:   7%|▋         | 468/6470 [05:45<1:16:36,  1.31it/s]
2|469|Loss: 0.9874308705329895:   7%|▋         | 468/6470 [05:46<1:16:36,  1.31it/s]
2|469|Loss: 0.9874308705329895:   7%|▋         | 469/6470 [05:46<1:15:17,  1.33it/s]
2|470|Loss: 0.9349232316017151:   7%|▋         | 469/6470 [05:46<1:15:17,  1.33it/s]
2|470|Loss: 0.9349232316017151:   7%|▋         | 470/6470 [05:47<1:18:56,  1.27it/s]
2|471|Loss: 0.7286585569381714:   7%|▋         | 470/6470 [05:47<1:18:56,  1.27it/s]
2|471|Loss: 0.7286585569381714:   7%|▋         | 471/6470 [05:48<1:17:35,  1.29it/s]
2|472|Loss: 0.8228263258934021:   7%|▋         | 471/6470 [05:48<1:17:35,  1.29it/s]
2|472|Loss: 0.8228263258934021:   7%|▋         | 472/6470 [05:48<1:14:55,  1.33it/s]
2|473|Loss: 0.9396460056304932:   7%|▋         | 472/6470 [05:49<1:14:55,  1.33it/s]
2|473|Loss: 0.9396460056304932:   7%|▋         | 473/6470 [05:49<1:09:46,  1.43it/s]
2|474|Loss: 0.9604623913764954:   7%|▋         | 473/6470 [05:49<1:09:46,  1.43it/s]
2|474|Loss: 0.9604623913764954:   7%|▋         | 474/6470 [05:50<1:13:27,  1.36it/s]
2|475|Loss: 0.8817922472953796:   7%|▋         | 474/6470 [05:50<1:13:27,  1.36it/s]
2|475|Loss: 0.8817922472953796:   7%|▋         | 475/6470 [05:50<1:11:22,  1.40it/s]
2|476|Loss: 0.8826338052749634:   7%|▋         | 475/6470 [05:51<1:11:22,  1.40it/s]
2|476|Loss: 0.8826338052749634:   7%|▋         | 476/6470 [05:51<1:14:16,  1.35it/s]
2|477|Loss: 0.8969029188156128:   7%|▋         | 476/6470 [05:52<1:14:16,  1.35it/s]
2|477|Loss: 0.8969029188156128:   7%|▋         | 477/6470 [05:52<1:18:22,  1.27it/s]
2|478|Loss: 0.9800127744674683:   7%|▋         | 477/6470 [05:52<1:18:22,  1.27it/s]
2|478|Loss: 0.9800127744674683:   7%|▋         | 478/6470 [05:53<1:16:50,  1.30it/s]
2|479|Loss: 0.8412730097770691:   7%|▋         | 478/6470 [05:53<1:16:50,  1.30it/s]
2|479|Loss: 0.8412730097770691:   7%|▋         | 479/6470 [05:53<1:14:12,  1.35it/s]
2|480|Loss: 0.8885846138000488:   7%|▋         | 479/6470 [05:54<1:14:12,  1.35it/s]
2|480|Loss: 0.8885846138000488:   7%|▋         | 480/6470 [05:54<1:15:36,  1.32it/s]
2|481|Loss: 0.7442673444747925:   7%|▋         | 480/6470 [05:55<1:15:36,  1.32it/s]
2|481|Loss: 0.7442673444747925:   7%|▋         | 481/6470 [05:55<1:14:38,  1.34it/s]
2|482|Loss: 0.8797756433486938:   7%|▋         | 481/6470 [05:55<1:14:38,  1.34it/s]
2|482|Loss: 0.8797756433486938:   7%|▋         | 482/6470 [05:56<1:14:44,  1.34it/s]
2|483|Loss: 0.9530715942382812:   7%|▋         | 482/6470 [05:56<1:14:44,  1.34it/s]
2|483|Loss: 0.9530715942382812:   7%|▋         | 483/6470 [05:57<1:15:59,  1.31it/s]
2|484|Loss: 0.8612781167030334:   7%|▋         | 483/6470 [05:57<1:15:59,  1.31it/s]
2|484|Loss: 0.8612781167030334:   7%|▋         | 484/6470 [05:57<1:14:21,  1.34it/s]
2|485|Loss: 0.8517708778381348:   7%|▋         | 484/6470 [05:58<1:14:21,  1.34it/s]
2|485|Loss: 0.8517708778381348:   7%|▋         | 485/6470 [05:58<1:17:20,  1.29it/s]
2|486|Loss: 0.737316906452179:   7%|▋         | 485/6470 [05:59<1:17:20,  1.29it/s] 
2|486|Loss: 0.737316906452179:   8%|▊         | 486/6470 [05:59<1:16:08,  1.31it/s]
2|487|Loss: 0.8334347605705261:   8%|▊         | 486/6470 [05:59<1:16:08,  1.31it/s]
2|487|Loss: 0.8334347605705261:   8%|▊         | 487/6470 [06:00<1:16:02,  1.31it/s]
2|488|Loss: 0.9083377122879028:   8%|▊         | 487/6470 [06:00<1:16:02,  1.31it/s]
2|488|Loss: 0.9083377122879028:   8%|▊         | 488/6470 [06:00<1:16:48,  1.30it/s]
2|489|Loss: 0.837365984916687:   8%|▊         | 488/6470 [06:01<1:16:48,  1.30it/s] 
2|489|Loss: 0.837365984916687:   8%|▊         | 489/6470 [06:01<1:14:37,  1.34it/s]
2|490|Loss: 1.110986351966858:   8%|▊         | 489/6470 [06:01<1:14:37,  1.34it/s]
2|490|Loss: 1.110986351966858:   8%|▊         | 490/6470 [06:02<1:14:38,  1.34it/s]
2|491|Loss: 0.7183769345283508:   8%|▊         | 490/6470 [06:02<1:14:38,  1.34it/s]
2|491|Loss: 0.7183769345283508:   8%|▊         | 491/6470 [06:02<1:09:22,  1.44it/s]
2|492|Loss: 0.9593111276626587:   8%|▊         | 491/6470 [06:03<1:09:22,  1.44it/s]
2|492|Loss: 0.9593111276626587:   8%|▊         | 492/6470 [06:03<1:11:20,  1.40it/s]
2|493|Loss: 0.9278073906898499:   8%|▊         | 492/6470 [06:04<1:11:20,  1.40it/s]
2|493|Loss: 0.9278073906898499:   8%|▊         | 493/6470 [06:04<1:13:03,  1.36it/s]
2|494|Loss: 1.0512088537216187:   8%|▊         | 493/6470 [06:04<1:13:03,  1.36it/s]
2|494|Loss: 1.0512088537216187:   8%|▊         | 494/6470 [06:05<1:11:56,  1.38it/s]
2|495|Loss: 0.8982692360877991:   8%|▊         | 494/6470 [06:05<1:11:56,  1.38it/s]
2|495|Loss: 0.8982692360877991:   8%|▊         | 495/6470 [06:05<1:09:06,  1.44it/s]
2|496|Loss: 0.7586892247200012:   8%|▊         | 495/6470 [06:06<1:09:06,  1.44it/s]
2|496|Loss: 0.7586892247200012:   8%|▊         | 496/6470 [06:06<1:06:59,  1.49it/s]
2|497|Loss: 0.8686601519584656:   8%|▊         | 496/6470 [06:06<1:06:59,  1.49it/s]
2|497|Loss: 0.8686601519584656:   8%|▊         | 497/6470 [06:07<1:07:30,  1.47it/s]
2|498|Loss: 0.838325023651123:   8%|▊         | 497/6470 [06:07<1:07:30,  1.47it/s] 
2|498|Loss: 0.838325023651123:   8%|▊         | 498/6470 [06:07<1:07:31,  1.47it/s]
2|499|Loss: 0.8016669154167175:   8%|▊         | 498/6470 [06:08<1:07:31,  1.47it/s]
2|499|Loss: 0.8016669154167175:   8%|▊         | 499/6470 [06:08<1:10:01,  1.42it/s]
2|500|Loss: 0.9720777273178101:   8%|▊         | 499/6470 [06:08<1:10:01,  1.42it/s]
2|500|Loss: 0.9720777273178101:   8%|▊         | 500/6470 [06:09<1:07:51,  1.47it/s]
2|501|Loss: 1.0329898595809937:   8%|▊         | 500/6470 [06:09<1:07:51,  1.47it/s]
2|501|Loss: 1.0329898595809937:   8%|▊         | 501/6470 [06:10<1:12:46,  1.37it/s]
2|502|Loss: 0.807252049446106:   8%|▊         | 501/6470 [06:10<1:12:46,  1.37it/s] 
2|502|Loss: 0.807252049446106:   8%|▊         | 502/6470 [06:10<1:14:11,  1.34it/s]
2|503|Loss: 0.9094198346138:   8%|▊         | 502/6470 [06:11<1:14:11,  1.34it/s]  
2|503|Loss: 0.9094198346138:   8%|▊         | 503/6470 [06:11<1:12:35,  1.37it/s]
2|504|Loss: 0.8453670740127563:   8%|▊         | 503/6470 [06:11<1:12:35,  1.37it/s]
2|504|Loss: 0.8453670740127563:   8%|▊         | 504/6470 [06:12<1:10:41,  1.41it/s]
2|505|Loss: 0.9068771004676819:   8%|▊         | 504/6470 [06:12<1:10:41,  1.41it/s]
2|505|Loss: 0.9068771004676819:   8%|▊         | 505/6470 [06:12<1:08:21,  1.45it/s]
2|506|Loss: 0.8810960054397583:   8%|▊         | 505/6470 [06:13<1:08:21,  1.45it/s]
2|506|Loss: 0.8810960054397583:   8%|▊         | 506/6470 [06:13<1:06:30,  1.49it/s]
2|507|Loss: 0.7897691130638123:   8%|▊         | 506/6470 [06:13<1:06:30,  1.49it/s]
2|507|Loss: 0.7897691130638123:   8%|▊         | 507/6470 [06:14<1:12:31,  1.37it/s]
2|508|Loss: 0.9266112446784973:   8%|▊         | 507/6470 [06:14<1:12:31,  1.37it/s]
2|508|Loss: 0.9266112446784973:   8%|▊         | 508/6470 [06:15<1:17:20,  1.28it/s]
2|509|Loss: 0.780622661113739:   8%|▊         | 508/6470 [06:15<1:17:20,  1.28it/s] 
2|509|Loss: 0.780622661113739:   8%|▊         | 509/6470 [06:16<1:21:48,  1.21it/s]
2|510|Loss: 0.8925508260726929:   8%|▊         | 509/6470 [06:16<1:21:48,  1.21it/s]
2|510|Loss: 0.8925508260726929:   8%|▊         | 510/6470 [06:16<1:16:19,  1.30it/s]
2|511|Loss: 0.7944982051849365:   8%|▊         | 510/6470 [06:17<1:16:19,  1.30it/s]
2|511|Loss: 0.7944982051849365:   8%|▊         | 511/6470 [06:17<1:08:52,  1.44it/s]
2|512|Loss: 0.9696449041366577:   8%|▊         | 511/6470 [06:17<1:08:52,  1.44it/s]
2|512|Loss: 0.9696449041366577:   8%|▊         | 512/6470 [06:17<1:10:14,  1.41it/s]
2|513|Loss: 0.8437302112579346:   8%|▊         | 512/6470 [06:18<1:10:14,  1.41it/s]
2|513|Loss: 0.8437302112579346:   8%|▊         | 513/6470 [06:18<1:10:54,  1.40it/s]
2|514|Loss: 0.9904151558876038:   8%|▊         | 513/6470 [06:19<1:10:54,  1.40it/s]
2|514|Loss: 0.9904151558876038:   8%|▊         | 514/6470 [06:19<1:09:43,  1.42it/s]
2|515|Loss: 0.8805995583534241:   8%|▊         | 514/6470 [06:19<1:09:43,  1.42it/s]
2|515|Loss: 0.8805995583534241:   8%|▊         | 515/6470 [06:20<1:09:48,  1.42it/s]
2|516|Loss: 0.8430579900741577:   8%|▊         | 515/6470 [06:20<1:09:48,  1.42it/s]
2|516|Loss: 0.8430579900741577:   8%|▊         | 516/6470 [06:20<1:14:00,  1.34it/s]
2|517|Loss: 0.8287951350212097:   8%|▊         | 516/6470 [06:21<1:14:00,  1.34it/s]
2|517|Loss: 0.8287951350212097:   8%|▊         | 517/6470 [06:21<1:15:23,  1.32it/s]
2|518|Loss: 0.7873397469520569:   8%|▊         | 517/6470 [06:22<1:15:23,  1.32it/s]
2|518|Loss: 0.7873397469520569:   8%|▊         | 518/6470 [06:22<1:12:23,  1.37it/s]
2|519|Loss: 0.8235333561897278:   8%|▊         | 518/6470 [06:22<1:12:23,  1.37it/s]
2|519|Loss: 0.8235333561897278:   8%|▊         | 519/6470 [06:22<1:05:37,  1.51it/s]
2|520|Loss: 0.8330566883087158:   8%|▊         | 519/6470 [06:23<1:05:37,  1.51it/s]
2|520|Loss: 0.8330566883087158:   8%|▊         | 520/6470 [06:23<1:10:38,  1.40it/s]
2|521|Loss: 0.7869638204574585:   8%|▊         | 520/6470 [06:24<1:10:38,  1.40it/s]
2|521|Loss: 0.7869638204574585:   8%|▊         | 521/6470 [06:24<1:10:06,  1.41it/s]
2|522|Loss: 0.7808122038841248:   8%|▊         | 521/6470 [06:24<1:10:06,  1.41it/s]
2|522|Loss: 0.7808122038841248:   8%|▊         | 522/6470 [06:25<1:09:05,  1.43it/s]
2|523|Loss: 0.8746005296707153:   8%|▊         | 522/6470 [06:25<1:09:05,  1.43it/s]
2|523|Loss: 0.8746005296707153:   8%|▊         | 523/6470 [06:25<1:13:42,  1.34it/s]
2|524|Loss: 0.8084253072738647:   8%|▊         | 523/6470 [06:26<1:13:42,  1.34it/s]
2|524|Loss: 0.8084253072738647:   8%|▊         | 524/6470 [06:26<1:12:46,  1.36it/s]
2|525|Loss: 0.9116613864898682:   8%|▊         | 524/6470 [06:27<1:12:46,  1.36it/s]
2|525|Loss: 0.9116613864898682:   8%|▊         | 525/6470 [06:27<1:13:03,  1.36it/s]
2|526|Loss: 0.9556046724319458:   8%|▊         | 525/6470 [06:27<1:13:03,  1.36it/s]
2|526|Loss: 0.9556046724319458:   8%|▊         | 526/6470 [06:28<1:08:53,  1.44it/s]
2|527|Loss: 0.7557279467582703:   8%|▊         | 526/6470 [06:28<1:08:53,  1.44it/s]
2|527|Loss: 0.7557279467582703:   8%|▊         | 527/6470 [06:28<1:10:21,  1.41it/s]
2|528|Loss: 0.9456050992012024:   8%|▊         | 527/6470 [06:29<1:10:21,  1.41it/s]
2|528|Loss: 0.9456050992012024:   8%|▊         | 528/6470 [06:29<1:10:20,  1.41it/s]
2|529|Loss: 0.9244284629821777:   8%|▊         | 528/6470 [06:29<1:10:20,  1.41it/s]
2|529|Loss: 0.9244284629821777:   8%|▊         | 529/6470 [06:30<1:09:57,  1.42it/s]
2|530|Loss: 0.9023905992507935:   8%|▊         | 529/6470 [06:30<1:09:57,  1.42it/s]
2|530|Loss: 0.9023905992507935:   8%|▊         | 530/6470 [06:31<1:14:09,  1.34it/s]
2|531|Loss: 0.8122464418411255:   8%|▊         | 530/6470 [06:31<1:14:09,  1.34it/s]
2|531|Loss: 0.8122464418411255:   8%|▊         | 531/6470 [06:31<1:14:15,  1.33it/s]
2|532|Loss: 0.9986257553100586:   8%|▊         | 531/6470 [06:32<1:14:15,  1.33it/s]
2|532|Loss: 0.9986257553100586:   8%|▊         | 532/6470 [06:32<1:12:57,  1.36it/s]
2|533|Loss: 0.7794238328933716:   8%|▊         | 532/6470 [06:32<1:12:57,  1.36it/s]
2|533|Loss: 0.7794238328933716:   8%|▊         | 533/6470 [06:33<1:16:15,  1.30it/s]
2|534|Loss: 0.7442508935928345:   8%|▊         | 533/6470 [06:33<1:16:15,  1.30it/s]
2|534|Loss: 0.7442508935928345:   8%|▊         | 534/6470 [06:34<1:22:42,  1.20it/s]
2|535|Loss: 0.9301052093505859:   8%|▊         | 534/6470 [06:34<1:22:42,  1.20it/s]
2|535|Loss: 0.9301052093505859:   8%|▊         | 535/6470 [06:35<1:26:06,  1.15it/s]
2|536|Loss: 0.7195996642112732:   8%|▊         | 535/6470 [06:35<1:26:06,  1.15it/s]
2|536|Loss: 0.7195996642112732:   8%|▊         | 536/6470 [06:36<1:24:01,  1.18it/s]
2|537|Loss: 0.774011492729187:   8%|▊         | 536/6470 [06:36<1:24:01,  1.18it/s] 
2|537|Loss: 0.774011492729187:   8%|▊         | 537/6470 [06:36<1:22:35,  1.20it/s]
2|538|Loss: 0.907137393951416:   8%|▊         | 537/6470 [06:37<1:22:35,  1.20it/s]
2|538|Loss: 0.907137393951416:   8%|▊         | 538/6470 [06:37<1:19:19,  1.25it/s]
2|539|Loss: 0.9659484624862671:   8%|▊         | 538/6470 [06:37<1:19:19,  1.25it/s]
2|539|Loss: 0.9659484624862671:   8%|▊         | 539/6470 [06:38<1:14:47,  1.32it/s]
2|540|Loss: 0.9258354306221008:   8%|▊         | 539/6470 [06:38<1:14:47,  1.32it/s]
2|540|Loss: 0.9258354306221008:   8%|▊         | 540/6470 [06:39<1:18:28,  1.26it/s]
2|541|Loss: 0.8484992384910583:   8%|▊         | 540/6470 [06:39<1:18:28,  1.26it/s]
2|541|Loss: 0.8484992384910583:   8%|▊         | 541/6470 [06:39<1:16:33,  1.29it/s]
2|542|Loss: 1.032193899154663:   8%|▊         | 541/6470 [06:40<1:16:33,  1.29it/s] 
2|542|Loss: 1.032193899154663:   8%|▊         | 542/6470 [06:40<1:10:13,  1.41it/s]
2|543|Loss: 0.81707364320755:   8%|▊         | 542/6470 [06:40<1:10:13,  1.41it/s] 
2|543|Loss: 0.81707364320755:   8%|▊         | 543/6470 [06:41<1:14:56,  1.32it/s]
2|544|Loss: 0.7783350944519043:   8%|▊         | 543/6470 [06:41<1:14:56,  1.32it/s]
2|544|Loss: 0.7783350944519043:   8%|▊         | 544/6470 [06:42<1:16:56,  1.28it/s]
2|545|Loss: 0.8139802813529968:   8%|▊         | 544/6470 [06:42<1:16:56,  1.28it/s]
2|545|Loss: 0.8139802813529968:   8%|▊         | 545/6470 [06:42<1:15:13,  1.31it/s]
2|546|Loss: 0.8451023101806641:   8%|▊         | 545/6470 [06:43<1:15:13,  1.31it/s]
2|546|Loss: 0.8451023101806641:   8%|▊         | 546/6470 [06:43<1:15:30,  1.31it/s]
2|547|Loss: 0.7904356718063354:   8%|▊         | 546/6470 [06:44<1:15:30,  1.31it/s]
2|547|Loss: 0.7904356718063354:   8%|▊         | 547/6470 [06:44<1:18:20,  1.26it/s]
2|548|Loss: 0.7194725275039673:   8%|▊         | 547/6470 [06:44<1:18:20,  1.26it/s]
2|548|Loss: 0.7194725275039673:   8%|▊         | 548/6470 [06:45<1:16:34,  1.29it/s]
2|549|Loss: 0.8179233074188232:   8%|▊         | 548/6470 [06:45<1:16:34,  1.29it/s]
2|549|Loss: 0.8179233074188232:   8%|▊         | 549/6470 [06:45<1:11:53,  1.37it/s]
2|550|Loss: 0.7642188668251038:   8%|▊         | 549/6470 [06:46<1:11:53,  1.37it/s]
2|550|Loss: 0.7642188668251038:   9%|▊         | 550/6470 [06:46<1:17:57,  1.27it/s]
2|551|Loss: 0.8926903009414673:   9%|▊         | 550/6470 [06:47<1:17:57,  1.27it/s]
2|551|Loss: 0.8926903009414673:   9%|▊         | 551/6470 [06:47<1:19:59,  1.23it/s]
2|552|Loss: 1.05137300491333:   9%|▊         | 551/6470 [06:48<1:19:59,  1.23it/s]  
2|552|Loss: 1.05137300491333:   9%|▊         | 552/6470 [06:48<1:22:37,  1.19it/s]
2|553|Loss: 0.8581444621086121:   9%|▊         | 552/6470 [06:48<1:22:37,  1.19it/s]
2|553|Loss: 0.8581444621086121:   9%|▊         | 553/6470 [06:49<1:19:56,  1.23it/s]
2|554|Loss: 0.7962249517440796:   9%|▊         | 553/6470 [06:49<1:19:56,  1.23it/s]
2|554|Loss: 0.7962249517440796:   9%|▊         | 554/6470 [06:49<1:14:13,  1.33it/s]
2|555|Loss: 0.810257077217102:   9%|▊         | 554/6470 [06:50<1:14:13,  1.33it/s] 
2|555|Loss: 0.810257077217102:   9%|▊         | 555/6470 [06:50<1:09:20,  1.42it/s]
2|556|Loss: 0.8091166615486145:   9%|▊         | 555/6470 [06:50<1:09:20,  1.42it/s]
2|556|Loss: 0.8091166615486145:   9%|▊         | 556/6470 [06:51<1:07:54,  1.45it/s]
2|557|Loss: 0.9792590141296387:   9%|▊         | 556/6470 [06:51<1:07:54,  1.45it/s]
2|557|Loss: 0.9792590141296387:   9%|▊         | 557/6470 [06:51<1:11:02,  1.39it/s]
2|558|Loss: 0.8622469902038574:   9%|▊         | 557/6470 [06:52<1:11:02,  1.39it/s]
2|558|Loss: 0.8622469902038574:   9%|▊         | 558/6470 [06:52<1:15:40,  1.30it/s]
2|559|Loss: 0.8554012775421143:   9%|▊         | 558/6470 [06:53<1:15:40,  1.30it/s]
2|559|Loss: 0.8554012775421143:   9%|▊         | 559/6470 [06:53<1:17:01,  1.28it/s]
2|560|Loss: 0.8194321393966675:   9%|▊         | 559/6470 [06:54<1:17:01,  1.28it/s]
2|560|Loss: 0.8194321393966675:   9%|▊         | 560/6470 [06:54<1:15:09,  1.31it/s]
2|561|Loss: 0.943784236907959:   9%|▊         | 560/6470 [06:54<1:15:09,  1.31it/s] 
2|561|Loss: 0.943784236907959:   9%|▊         | 561/6470 [06:55<1:14:59,  1.31it/s]
2|562|Loss: 0.8527563214302063:   9%|▊         | 561/6470 [06:55<1:14:59,  1.31it/s]
2|562|Loss: 0.8527563214302063:   9%|▊         | 562/6470 [06:55<1:11:08,  1.38it/s]
2|563|Loss: 0.9256826639175415:   9%|▊         | 562/6470 [06:56<1:11:08,  1.38it/s]
2|563|Loss: 0.9256826639175415:   9%|▊         | 563/6470 [06:56<1:12:12,  1.36it/s]
2|564|Loss: 0.7383975982666016:   9%|▊         | 563/6470 [06:56<1:12:12,  1.36it/s]
2|564|Loss: 0.7383975982666016:   9%|▊         | 564/6470 [06:57<1:11:50,  1.37it/s]
2|565|Loss: 0.927310049533844:   9%|▊         | 564/6470 [06:57<1:11:50,  1.37it/s] 
2|565|Loss: 0.927310049533844:   9%|▊         | 565/6470 [06:57<1:09:18,  1.42it/s]
2|566|Loss: 0.9924559593200684:   9%|▊         | 565/6470 [06:58<1:09:18,  1.42it/s]
2|566|Loss: 0.9924559593200684:   9%|▊         | 566/6470 [06:58<1:13:25,  1.34it/s]
2|567|Loss: 0.8886861801147461:   9%|▊         | 566/6470 [06:59<1:13:25,  1.34it/s]
2|567|Loss: 0.8886861801147461:   9%|▉         | 567/6470 [06:59<1:12:31,  1.36it/s]
2|568|Loss: 0.7820935845375061:   9%|▉         | 567/6470 [06:59<1:12:31,  1.36it/s]
2|568|Loss: 0.7820935845375061:   9%|▉         | 568/6470 [07:00<1:11:25,  1.38it/s]
2|569|Loss: 0.864154577255249:   9%|▉         | 568/6470 [07:00<1:11:25,  1.38it/s] 
2|569|Loss: 0.864154577255249:   9%|▉         | 569/6470 [07:00<1:10:43,  1.39it/s]
2|570|Loss: 0.8558048009872437:   9%|▉         | 569/6470 [07:01<1:10:43,  1.39it/s]
2|570|Loss: 0.8558048009872437:   9%|▉         | 570/6470 [07:01<1:11:47,  1.37it/s]
2|571|Loss: 0.9590646624565125:   9%|▉         | 570/6470 [07:01<1:11:47,  1.37it/s]
2|571|Loss: 0.9590646624565125:   9%|▉         | 571/6470 [07:02<1:11:24,  1.38it/s]
2|572|Loss: 0.7989709377288818:   9%|▉         | 571/6470 [07:02<1:11:24,  1.38it/s]
2|572|Loss: 0.7989709377288818:   9%|▉         | 572/6470 [07:03<1:12:48,  1.35it/s]
2|573|Loss: 0.9243397116661072:   9%|▉         | 572/6470 [07:03<1:12:48,  1.35it/s]
2|573|Loss: 0.9243397116661072:   9%|▉         | 573/6470 [07:03<1:13:26,  1.34it/s]
2|574|Loss: 0.729441225528717:   9%|▉         | 573/6470 [07:04<1:13:26,  1.34it/s] 
2|574|Loss: 0.729441225528717:   9%|▉         | 574/6470 [07:04<1:10:37,  1.39it/s]
2|575|Loss: 0.7895681262016296:   9%|▉         | 574/6470 [07:04<1:10:37,  1.39it/s]
2|575|Loss: 0.7895681262016296:   9%|▉         | 575/6470 [07:05<1:09:46,  1.41it/s]
2|576|Loss: 0.7930455803871155:   9%|▉         | 575/6470 [07:05<1:09:46,  1.41it/s]
2|576|Loss: 0.7930455803871155:   9%|▉         | 576/6470 [07:05<1:09:34,  1.41it/s]
2|577|Loss: 0.8803486227989197:   9%|▉         | 576/6470 [07:06<1:09:34,  1.41it/s]
2|577|Loss: 0.8803486227989197:   9%|▉         | 577/6470 [07:06<1:12:05,  1.36it/s]
2|578|Loss: 0.7026377320289612:   9%|▉         | 577/6470 [07:07<1:12:05,  1.36it/s]
2|578|Loss: 0.7026377320289612:   9%|▉         | 578/6470 [07:07<1:08:22,  1.44it/s]
2|579|Loss: 0.9459335207939148:   9%|▉         | 578/6470 [07:07<1:08:22,  1.44it/s]
2|579|Loss: 0.9459335207939148:   9%|▉         | 579/6470 [07:07<1:04:54,  1.51it/s]
2|580|Loss: 0.7350860834121704:   9%|▉         | 579/6470 [07:08<1:04:54,  1.51it/s]
2|580|Loss: 0.7350860834121704:   9%|▉         | 580/6470 [07:08<1:11:30,  1.37it/s]
2|581|Loss: 1.0007014274597168:   9%|▉         | 580/6470 [07:09<1:11:30,  1.37it/s]
2|581|Loss: 1.0007014274597168:   9%|▉         | 581/6470 [07:09<1:14:33,  1.32it/s]
2|582|Loss: 0.8463200330734253:   9%|▉         | 581/6470 [07:09<1:14:33,  1.32it/s]
2|582|Loss: 0.8463200330734253:   9%|▉         | 582/6470 [07:10<1:12:32,  1.35it/s]
2|583|Loss: 0.8702081441879272:   9%|▉         | 582/6470 [07:10<1:12:32,  1.35it/s]
2|583|Loss: 0.8702081441879272:   9%|▉         | 583/6470 [07:11<1:17:16,  1.27it/s]
2|584|Loss: 0.9291659593582153:   9%|▉         | 583/6470 [07:11<1:17:16,  1.27it/s]
2|584|Loss: 0.9291659593582153:   9%|▉         | 584/6470 [07:11<1:17:58,  1.26it/s]
2|585|Loss: 0.8517274856567383:   9%|▉         | 584/6470 [07:12<1:17:58,  1.26it/s]
2|585|Loss: 0.8517274856567383:   9%|▉         | 585/6470 [07:12<1:14:22,  1.32it/s]
2|586|Loss: 1.1203792095184326:   9%|▉         | 585/6470 [07:13<1:14:22,  1.32it/s]
2|586|Loss: 1.1203792095184326:   9%|▉         | 586/6470 [07:13<1:13:40,  1.33it/s]
2|587|Loss: 0.9816490411758423:   9%|▉         | 586/6470 [07:13<1:13:40,  1.33it/s]
2|587|Loss: 0.9816490411758423:   9%|▉         | 587/6470 [07:14<1:14:06,  1.32it/s]
2|588|Loss: 0.8762308955192566:   9%|▉         | 587/6470 [07:14<1:14:06,  1.32it/s]
2|588|Loss: 0.8762308955192566:   9%|▉         | 588/6470 [07:15<1:21:15,  1.21it/s]
2|589|Loss: 0.8853234052658081:   9%|▉         | 588/6470 [07:15<1:21:15,  1.21it/s]
2|589|Loss: 0.8853234052658081:   9%|▉         | 589/6470 [07:15<1:14:32,  1.31it/s]
2|590|Loss: 0.7415649890899658:   9%|▉         | 589/6470 [07:16<1:14:32,  1.31it/s]
2|590|Loss: 0.7415649890899658:   9%|▉         | 590/6470 [07:16<1:12:26,  1.35it/s]
2|591|Loss: 0.861858606338501:   9%|▉         | 590/6470 [07:16<1:12:26,  1.35it/s] 
2|591|Loss: 0.861858606338501:   9%|▉         | 591/6470 [07:17<1:12:01,  1.36it/s]
2|592|Loss: 0.8226338624954224:   9%|▉         | 591/6470 [07:17<1:12:01,  1.36it/s]
2|592|Loss: 0.8226338624954224:   9%|▉         | 592/6470 [07:17<1:05:32,  1.49it/s]
2|593|Loss: 0.8852962255477905:   9%|▉         | 592/6470 [07:18<1:05:32,  1.49it/s]
2|593|Loss: 0.8852962255477905:   9%|▉         | 593/6470 [07:18<1:05:01,  1.51it/s]
2|594|Loss: 1.0102863311767578:   9%|▉         | 593/6470 [07:18<1:05:01,  1.51it/s]
2|594|Loss: 1.0102863311767578:   9%|▉         | 594/6470 [07:18<1:05:35,  1.49it/s]
2|595|Loss: 0.7196548581123352:   9%|▉         | 594/6470 [07:19<1:05:35,  1.49it/s]
2|595|Loss: 0.7196548581123352:   9%|▉         | 595/6470 [07:19<1:06:58,  1.46it/s]
2|596|Loss: 0.9531217217445374:   9%|▉         | 595/6470 [07:20<1:06:58,  1.46it/s]
2|596|Loss: 0.9531217217445374:   9%|▉         | 596/6470 [07:20<1:06:58,  1.46it/s]
2|597|Loss: 0.9634859561920166:   9%|▉         | 596/6470 [07:20<1:06:58,  1.46it/s]
2|597|Loss: 0.9634859561920166:   9%|▉         | 597/6470 [07:21<1:09:10,  1.41it/s]
2|598|Loss: 0.7184785604476929:   9%|▉         | 597/6470 [07:21<1:09:10,  1.41it/s]
2|598|Loss: 0.7184785604476929:   9%|▉         | 598/6470 [07:21<1:09:37,  1.41it/s]
2|599|Loss: 1.0054079294204712:   9%|▉         | 598/6470 [07:22<1:09:37,  1.41it/s]
2|599|Loss: 1.0054079294204712:   9%|▉         | 599/6470 [07:22<1:09:37,  1.41it/s]
2|600|Loss: 0.9847540259361267:   9%|▉         | 599/6470 [07:23<1:09:37,  1.41it/s]
2|600|Loss: 0.9847540259361267:   9%|▉         | 600/6470 [07:23<1:11:32,  1.37it/s]
2|601|Loss: 0.867322564125061:   9%|▉         | 600/6470 [07:23<1:11:32,  1.37it/s] 
2|601|Loss: 0.867322564125061:   9%|▉         | 601/6470 [07:24<1:10:32,  1.39it/s]
2|602|Loss: 0.7786412835121155:   9%|▉         | 601/6470 [07:24<1:10:32,  1.39it/s]
2|602|Loss: 0.7786412835121155:   9%|▉         | 602/6470 [07:24<1:11:44,  1.36it/s]
2|603|Loss: 0.911590039730072:   9%|▉         | 602/6470 [07:25<1:11:44,  1.36it/s] 
2|603|Loss: 0.911590039730072:   9%|▉         | 603/6470 [07:25<1:11:00,  1.38it/s]
2|604|Loss: 0.8546310663223267:   9%|▉         | 603/6470 [07:26<1:11:00,  1.38it/s]
2|604|Loss: 0.8546310663223267:   9%|▉         | 604/6470 [07:26<1:16:30,  1.28it/s]
2|605|Loss: 0.8172691464424133:   9%|▉         | 604/6470 [07:26<1:16:30,  1.28it/s]
2|605|Loss: 0.8172691464424133:   9%|▉         | 605/6470 [07:27<1:18:27,  1.25it/s]
2|606|Loss: 0.832181453704834:   9%|▉         | 605/6470 [07:27<1:18:27,  1.25it/s] 
2|606|Loss: 0.832181453704834:   9%|▉         | 606/6470 [07:28<1:22:48,  1.18it/s]
2|607|Loss: 0.9573519825935364:   9%|▉         | 606/6470 [07:28<1:22:48,  1.18it/s]
2|607|Loss: 0.9573519825935364:   9%|▉         | 607/6470 [07:29<1:22:56,  1.18it/s]
2|608|Loss: 0.857796311378479:   9%|▉         | 607/6470 [07:29<1:22:56,  1.18it/s] 
2|608|Loss: 0.857796311378479:   9%|▉         | 608/6470 [07:29<1:21:47,  1.19it/s]
2|609|Loss: 0.8854753375053406:   9%|▉         | 608/6470 [07:30<1:21:47,  1.19it/s]
2|609|Loss: 0.8854753375053406:   9%|▉         | 609/6470 [07:30<1:17:57,  1.25it/s]
2|610|Loss: 0.7925751209259033:   9%|▉         | 609/6470 [07:30<1:17:57,  1.25it/s]
2|610|Loss: 0.7925751209259033:   9%|▉         | 610/6470 [07:31<1:13:30,  1.33it/s]
2|611|Loss: 0.9280475378036499:   9%|▉         | 610/6470 [07:31<1:13:30,  1.33it/s]
2|611|Loss: 0.9280475378036499:   9%|▉         | 611/6470 [07:32<1:13:17,  1.33it/s]
2|612|Loss: 0.7294613718986511:   9%|▉         | 611/6470 [07:32<1:13:17,  1.33it/s]
2|612|Loss: 0.7294613718986511:   9%|▉         | 612/6470 [07:32<1:13:20,  1.33it/s]
2|613|Loss: 0.8419046998023987:   9%|▉         | 612/6470 [07:33<1:13:20,  1.33it/s]
2|613|Loss: 0.8419046998023987:   9%|▉         | 613/6470 [07:33<1:11:58,  1.36it/s]
2|614|Loss: 0.8766751289367676:   9%|▉         | 613/6470 [07:33<1:11:58,  1.36it/s]
2|614|Loss: 0.8766751289367676:   9%|▉         | 614/6470 [07:34<1:12:09,  1.35it/s]
2|615|Loss: 0.8366710543632507:   9%|▉         | 614/6470 [07:34<1:12:09,  1.35it/s]
2|615|Loss: 0.8366710543632507:  10%|▉         | 615/6470 [07:35<1:18:47,  1.24it/s]
2|616|Loss: 0.9368520975112915:  10%|▉         | 615/6470 [07:35<1:18:47,  1.24it/s]
2|616|Loss: 0.9368520975112915:  10%|▉         | 616/6470 [07:36<1:22:45,  1.18it/s]
2|617|Loss: 0.8187204599380493:  10%|▉         | 616/6470 [07:36<1:22:45,  1.18it/s]
2|617|Loss: 0.8187204599380493:  10%|▉         | 617/6470 [07:36<1:22:23,  1.18it/s]
2|618|Loss: 1.0019147396087646:  10%|▉         | 617/6470 [07:37<1:22:23,  1.18it/s]
2|618|Loss: 1.0019147396087646:  10%|▉         | 618/6470 [07:37<1:21:41,  1.19it/s]
2|619|Loss: 1.0951790809631348:  10%|▉         | 618/6470 [07:38<1:21:41,  1.19it/s]
2|619|Loss: 1.0951790809631348:  10%|▉         | 619/6470 [07:38<1:19:58,  1.22it/s]
2|620|Loss: 0.8183587193489075:  10%|▉         | 619/6470 [07:38<1:19:58,  1.22it/s]
2|620|Loss: 0.8183587193489075:  10%|▉         | 620/6470 [07:39<1:13:32,  1.33it/s]
2|621|Loss: 0.7607437372207642:  10%|▉         | 620/6470 [07:39<1:13:32,  1.33it/s]
2|621|Loss: 0.7607437372207642:  10%|▉         | 621/6470 [07:39<1:13:47,  1.32it/s]
2|622|Loss: 0.8690650463104248:  10%|▉         | 621/6470 [07:40<1:13:47,  1.32it/s]
2|622|Loss: 0.8690650463104248:  10%|▉         | 622/6470 [07:40<1:13:57,  1.32it/s]
2|623|Loss: 0.902887225151062:  10%|▉         | 622/6470 [07:41<1:13:57,  1.32it/s] 
2|623|Loss: 0.902887225151062:  10%|▉         | 623/6470 [07:41<1:15:50,  1.28it/s]
2|624|Loss: 0.7838997840881348:  10%|▉         | 623/6470 [07:41<1:15:50,  1.28it/s]
2|624|Loss: 0.7838997840881348:  10%|▉         | 624/6470 [07:42<1:18:48,  1.24it/s]
2|625|Loss: 0.712748110294342:  10%|▉         | 624/6470 [07:42<1:18:48,  1.24it/s] 
2|625|Loss: 0.712748110294342:  10%|▉         | 625/6470 [07:43<1:15:54,  1.28it/s]
2|626|Loss: 0.8063111901283264:  10%|▉         | 625/6470 [07:43<1:15:54,  1.28it/s]
2|626|Loss: 0.8063111901283264:  10%|▉         | 626/6470 [07:43<1:15:46,  1.29it/s]
2|627|Loss: 0.7788690328598022:  10%|▉         | 626/6470 [07:44<1:15:46,  1.29it/s]
2|627|Loss: 0.7788690328598022:  10%|▉         | 627/6470 [07:44<1:11:48,  1.36it/s]
2|628|Loss: 0.8352067470550537:  10%|▉         | 627/6470 [07:44<1:11:48,  1.36it/s]
2|628|Loss: 0.8352067470550537:  10%|▉         | 628/6470 [07:45<1:12:08,  1.35it/s]
2|629|Loss: 0.8312576413154602:  10%|▉         | 628/6470 [07:45<1:12:08,  1.35it/s]
2|629|Loss: 0.8312576413154602:  10%|▉         | 629/6470 [07:46<1:12:41,  1.34it/s]
2|630|Loss: 0.8384511470794678:  10%|▉         | 629/6470 [07:46<1:12:41,  1.34it/s]
2|630|Loss: 0.8384511470794678:  10%|▉         | 630/6470 [07:46<1:11:58,  1.35it/s]
2|631|Loss: 0.884779155254364:  10%|▉         | 630/6470 [07:47<1:11:58,  1.35it/s] 
2|631|Loss: 0.884779155254364:  10%|▉         | 631/6470 [07:47<1:11:00,  1.37it/s]
2|632|Loss: 0.7541758418083191:  10%|▉         | 631/6470 [07:47<1:11:00,  1.37it/s]
2|632|Loss: 0.7541758418083191:  10%|▉         | 632/6470 [07:47<1:02:15,  1.56it/s]
2|633|Loss: 0.8538320660591125:  10%|▉         | 632/6470 [07:48<1:02:15,  1.56it/s]
2|633|Loss: 0.8538320660591125:  10%|▉         | 633/6470 [07:48<58:41,  1.66it/s]  
2|634|Loss: 0.8435706496238708:  10%|▉         | 633/6470 [07:48<58:41,  1.66it/s]
2|634|Loss: 0.8435706496238708:  10%|▉         | 634/6470 [07:49<58:23,  1.67it/s]
2|635|Loss: 1.085006594657898:  10%|▉         | 634/6470 [07:49<58:23,  1.67it/s] 
2|635|Loss: 1.085006594657898:  10%|▉         | 635/6470 [07:49<1:06:14,  1.47it/s]
2|636|Loss: 1.0299937725067139:  10%|▉         | 635/6470 [07:50<1:06:14,  1.47it/s]
2|636|Loss: 1.0299937725067139:  10%|▉         | 636/6470 [07:50<1:07:36,  1.44it/s]
2|637|Loss: 1.0278122425079346:  10%|▉         | 636/6470 [07:51<1:07:36,  1.44it/s]
2|637|Loss: 1.0278122425079346:  10%|▉         | 637/6470 [07:51<1:13:28,  1.32it/s]
2|638|Loss: 0.9178184866905212:  10%|▉         | 637/6470 [07:51<1:13:28,  1.32it/s]
2|638|Loss: 0.9178184866905212:  10%|▉         | 638/6470 [07:52<1:12:03,  1.35it/s]
2|639|Loss: 0.7029849290847778:  10%|▉         | 638/6470 [07:52<1:12:03,  1.35it/s]
2|639|Loss: 0.7029849290847778:  10%|▉         | 639/6470 [07:52<1:09:25,  1.40it/s]
2|640|Loss: 0.8144571781158447:  10%|▉         | 639/6470 [07:53<1:09:25,  1.40it/s]
2|640|Loss: 0.8144571781158447:  10%|▉         | 640/6470 [07:53<1:08:51,  1.41it/s]
2|641|Loss: 0.959168553352356:  10%|▉         | 640/6470 [07:53<1:08:51,  1.41it/s] 
2|641|Loss: 0.959168553352356:  10%|▉         | 641/6470 [07:54<1:10:50,  1.37it/s]
2|642|Loss: 0.9186177253723145:  10%|▉         | 641/6470 [07:54<1:10:50,  1.37it/s]
2|642|Loss: 0.9186177253723145:  10%|▉         | 642/6470 [07:55<1:09:45,  1.39it/s]
2|643|Loss: 0.846968412399292:  10%|▉         | 642/6470 [07:55<1:09:45,  1.39it/s] 
2|643|Loss: 0.846968412399292:  10%|▉         | 643/6470 [07:55<1:09:41,  1.39it/s]
2|644|Loss: 0.7418035268783569:  10%|▉         | 643/6470 [07:56<1:09:41,  1.39it/s]
2|644|Loss: 0.7418035268783569:  10%|▉         | 644/6470 [07:56<1:08:41,  1.41it/s]
2|645|Loss: 0.856529951095581:  10%|▉         | 644/6470 [07:56<1:08:41,  1.41it/s] 
2|645|Loss: 0.856529951095581:  10%|▉         | 645/6470 [07:57<1:09:39,  1.39it/s]
2|646|Loss: 0.9493677020072937:  10%|▉         | 645/6470 [07:57<1:09:39,  1.39it/s]
2|646|Loss: 0.9493677020072937:  10%|▉         | 646/6470 [07:57<1:08:53,  1.41it/s]
2|647|Loss: 0.8334080576896667:  10%|▉         | 646/6470 [07:58<1:08:53,  1.41it/s]
2|647|Loss: 0.8334080576896667:  10%|█         | 647/6470 [07:58<1:14:15,  1.31it/s]
2|648|Loss: 0.7171937227249146:  10%|█         | 647/6470 [07:59<1:14:15,  1.31it/s]
2|648|Loss: 0.7171937227249146:  10%|█         | 648/6470 [07:59<1:17:22,  1.25it/s]
2|649|Loss: 0.9841166734695435:  10%|█         | 648/6470 [08:00<1:17:22,  1.25it/s]
2|649|Loss: 0.9841166734695435:  10%|█         | 649/6470 [08:00<1:20:28,  1.21it/s]
2|650|Loss: 0.749814510345459:  10%|█         | 649/6470 [08:00<1:20:28,  1.21it/s] 
2|650|Loss: 0.749814510345459:  10%|█         | 650/6470 [08:01<1:20:31,  1.20it/s]
2|651|Loss: 0.8469210863113403:  10%|█         | 650/6470 [08:01<1:20:31,  1.20it/s]
2|651|Loss: 0.8469210863113403:  10%|█         | 651/6470 [08:02<1:17:20,  1.25it/s]
2|652|Loss: 0.869459331035614:  10%|█         | 651/6470 [08:02<1:17:20,  1.25it/s] 
2|652|Loss: 0.869459331035614:  10%|█         | 652/6470 [08:02<1:20:18,  1.21it/s]
2|653|Loss: 0.9065396189689636:  10%|█         | 652/6470 [08:03<1:20:18,  1.21it/s]
2|653|Loss: 0.9065396189689636:  10%|█         | 653/6470 [08:03<1:23:42,  1.16it/s]
2|654|Loss: 0.844433069229126:  10%|█         | 653/6470 [08:04<1:23:42,  1.16it/s] 
2|654|Loss: 0.844433069229126:  10%|█         | 654/6470 [08:04<1:22:37,  1.17it/s]
2|655|Loss: 0.8477227687835693:  10%|█         | 654/6470 [08:05<1:22:37,  1.17it/s]
2|655|Loss: 0.8477227687835693:  10%|█         | 655/6470 [08:05<1:17:38,  1.25it/s]
2|656|Loss: 1.0132867097854614:  10%|█         | 655/6470 [08:05<1:17:38,  1.25it/s]
2|656|Loss: 1.0132867097854614:  10%|█         | 656/6470 [08:06<1:19:36,  1.22it/s]
2|657|Loss: 0.8221426606178284:  10%|█         | 656/6470 [08:06<1:19:36,  1.22it/s]
2|657|Loss: 0.8221426606178284:  10%|█         | 657/6470 [08:07<1:21:58,  1.18it/s]
2|658|Loss: 0.8222464323043823:  10%|█         | 657/6470 [08:07<1:21:58,  1.18it/s]
2|658|Loss: 0.8222464323043823:  10%|█         | 658/6470 [08:08<1:24:05,  1.15it/s]
2|659|Loss: 0.8354482054710388:  10%|█         | 658/6470 [08:08<1:24:05,  1.15it/s]
2|659|Loss: 0.8354482054710388:  10%|█         | 659/6470 [08:08<1:20:20,  1.21it/s]
2|660|Loss: 1.0783882141113281:  10%|█         | 659/6470 [08:09<1:20:20,  1.21it/s]
2|660|Loss: 1.0783882141113281:  10%|█         | 660/6470 [08:09<1:17:54,  1.24it/s]
2|661|Loss: 0.9078022241592407:  10%|█         | 660/6470 [08:10<1:17:54,  1.24it/s]
2|661|Loss: 0.9078022241592407:  10%|█         | 661/6470 [08:10<1:16:31,  1.27it/s]
2|662|Loss: 0.902504563331604:  10%|█         | 661/6470 [08:10<1:16:31,  1.27it/s] 
2|662|Loss: 0.902504563331604:  10%|█         | 662/6470 [08:11<1:19:15,  1.22it/s]
2|663|Loss: 0.7527225613594055:  10%|█         | 662/6470 [08:11<1:19:15,  1.22it/s]
2|663|Loss: 0.7527225613594055:  10%|█         | 663/6470 [08:12<1:19:17,  1.22it/s]
2|664|Loss: 0.8587988018989563:  10%|█         | 663/6470 [08:12<1:19:17,  1.22it/s]
2|664|Loss: 0.8587988018989563:  10%|█         | 664/6470 [08:12<1:18:12,  1.24it/s]
2|665|Loss: 0.9100998640060425:  10%|█         | 664/6470 [08:13<1:18:12,  1.24it/s]
2|665|Loss: 0.9100998640060425:  10%|█         | 665/6470 [08:13<1:10:10,  1.38it/s]
2|666|Loss: 0.8318197131156921:  10%|█         | 665/6470 [08:13<1:10:10,  1.38it/s]
2|666|Loss: 0.8318197131156921:  10%|█         | 666/6470 [08:14<1:07:27,  1.43it/s]
2|667|Loss: 0.8663323521614075:  10%|█         | 666/6470 [08:14<1:07:27,  1.43it/s]
2|667|Loss: 0.8663323521614075:  10%|█         | 667/6470 [08:14<1:04:19,  1.50it/s]
2|668|Loss: 0.8008677959442139:  10%|█         | 667/6470 [08:14<1:04:19,  1.50it/s]
2|668|Loss: 0.8008677959442139:  10%|█         | 668/6470 [08:15<1:00:44,  1.59it/s]
2|669|Loss: 0.8259944319725037:  10%|█         | 668/6470 [08:15<1:00:44,  1.59it/s]
2|669|Loss: 0.8259944319725037:  10%|█         | 669/6470 [08:15<1:06:27,  1.45it/s]
2|670|Loss: 0.9195736646652222:  10%|█         | 669/6470 [08:16<1:06:27,  1.45it/s]
2|670|Loss: 0.9195736646652222:  10%|█         | 670/6470 [08:16<1:13:52,  1.31it/s]
2|671|Loss: 0.9791898131370544:  10%|█         | 670/6470 [08:17<1:13:52,  1.31it/s]
2|671|Loss: 0.9791898131370544:  10%|█         | 671/6470 [08:17<1:16:20,  1.27it/s]
2|672|Loss: 0.8620325922966003:  10%|█         | 671/6470 [08:18<1:16:20,  1.27it/s]
2|672|Loss: 0.8620325922966003:  10%|█         | 672/6470 [08:18<1:18:57,  1.22it/s]
2|673|Loss: 0.7642788887023926:  10%|█         | 672/6470 [08:19<1:18:57,  1.22it/s]
2|673|Loss: 0.7642788887023926:  10%|█         | 673/6470 [08:19<1:15:52,  1.27it/s]
2|674|Loss: 0.9913354516029358:  10%|█         | 673/6470 [08:19<1:15:52,  1.27it/s]
2|674|Loss: 0.9913354516029358:  10%|█         | 674/6470 [08:20<1:14:32,  1.30it/s]
2|675|Loss: 0.8537355065345764:  10%|█         | 674/6470 [08:20<1:14:32,  1.30it/s]
2|675|Loss: 0.8537355065345764:  10%|█         | 675/6470 [08:20<1:14:48,  1.29it/s]
2|676|Loss: 0.7377959489822388:  10%|█         | 675/6470 [08:21<1:14:48,  1.29it/s]
2|676|Loss: 0.7377959489822388:  10%|█         | 676/6470 [08:21<1:08:10,  1.42it/s]
2|677|Loss: 0.8965109586715698:  10%|█         | 676/6470 [08:21<1:08:10,  1.42it/s]
2|677|Loss: 0.8965109586715698:  10%|█         | 677/6470 [08:22<1:06:38,  1.45it/s]
2|678|Loss: 0.8102027177810669:  10%|█         | 677/6470 [08:22<1:06:38,  1.45it/s]
2|678|Loss: 0.8102027177810669:  10%|█         | 678/6470 [08:22<1:11:17,  1.35it/s]
2|679|Loss: 0.8523802757263184:  10%|█         | 678/6470 [08:23<1:11:17,  1.35it/s]
2|679|Loss: 0.8523802757263184:  10%|█         | 679/6470 [08:23<1:16:03,  1.27it/s]
2|680|Loss: 0.9090221524238586:  10%|█         | 679/6470 [08:24<1:16:03,  1.27it/s]
2|680|Loss: 0.9090221524238586:  11%|█         | 680/6470 [08:24<1:16:55,  1.25it/s]
2|681|Loss: 0.8640387058258057:  11%|█         | 680/6470 [08:25<1:16:55,  1.25it/s]
2|681|Loss: 0.8640387058258057:  11%|█         | 681/6470 [08:25<1:18:46,  1.22it/s]
2|682|Loss: 0.8732392191886902:  11%|█         | 681/6470 [08:25<1:18:46,  1.22it/s]
2|682|Loss: 0.8732392191886902:  11%|█         | 682/6470 [08:26<1:18:35,  1.23it/s]
2|683|Loss: 0.8121412396430969:  11%|█         | 682/6470 [08:26<1:18:35,  1.23it/s]
2|683|Loss: 0.8121412396430969:  11%|█         | 683/6470 [08:27<1:17:39,  1.24it/s]
2|684|Loss: 0.8596543073654175:  11%|█         | 683/6470 [08:27<1:17:39,  1.24it/s]
2|684|Loss: 0.8596543073654175:  11%|█         | 684/6470 [08:27<1:14:59,  1.29it/s]
2|685|Loss: 1.1796960830688477:  11%|█         | 684/6470 [08:28<1:14:59,  1.29it/s]
2|685|Loss: 1.1796960830688477:  11%|█         | 685/6470 [08:28<1:14:18,  1.30it/s]
2|686|Loss: 0.8901950120925903:  11%|█         | 685/6470 [08:28<1:14:18,  1.30it/s]
2|686|Loss: 0.8901950120925903:  11%|█         | 686/6470 [08:29<1:12:46,  1.32it/s]
2|687|Loss: 0.9465667009353638:  11%|█         | 686/6470 [08:29<1:12:46,  1.32it/s]
2|687|Loss: 0.9465667009353638:  11%|█         | 687/6470 [08:30<1:13:18,  1.31it/s]
2|688|Loss: 0.8763099312782288:  11%|█         | 687/6470 [08:30<1:13:18,  1.31it/s]
2|688|Loss: 0.8763099312782288:  11%|█         | 688/6470 [08:30<1:17:10,  1.25it/s]
2|689|Loss: 0.7745534181594849:  11%|█         | 688/6470 [08:31<1:17:10,  1.25it/s]
2|689|Loss: 0.7745534181594849:  11%|█         | 689/6470 [08:31<1:15:52,  1.27it/s]
2|690|Loss: 0.8730108737945557:  11%|█         | 689/6470 [08:32<1:15:52,  1.27it/s]
2|690|Loss: 0.8730108737945557:  11%|█         | 690/6470 [08:32<1:15:05,  1.28it/s]
2|691|Loss: 0.7620375156402588:  11%|█         | 690/6470 [08:32<1:15:05,  1.28it/s]
2|691|Loss: 0.7620375156402588:  11%|█         | 691/6470 [08:33<1:20:14,  1.20it/s]
2|692|Loss: 0.6956171989440918:  11%|█         | 691/6470 [08:33<1:20:14,  1.20it/s]
2|692|Loss: 0.6956171989440918:  11%|█         | 692/6470 [08:34<1:18:21,  1.23it/s]
2|693|Loss: 1.127698540687561:  11%|█         | 692/6470 [08:34<1:18:21,  1.23it/s] 
2|693|Loss: 1.127698540687561:  11%|█         | 693/6470 [08:34<1:15:35,  1.27it/s]
2|694|Loss: 0.8370537757873535:  11%|█         | 693/6470 [08:35<1:15:35,  1.27it/s]
2|694|Loss: 0.8370537757873535:  11%|█         | 694/6470 [08:35<1:12:46,  1.32it/s]
2|695|Loss: 0.8080236315727234:  11%|█         | 694/6470 [08:36<1:12:46,  1.32it/s]
2|695|Loss: 0.8080236315727234:  11%|█         | 695/6470 [08:36<1:12:16,  1.33it/s]
2|696|Loss: 0.843700647354126:  11%|█         | 695/6470 [08:36<1:12:16,  1.33it/s] 
2|696|Loss: 0.843700647354126:  11%|█         | 696/6470 [08:37<1:11:05,  1.35it/s]
2|697|Loss: 0.7822044491767883:  11%|█         | 696/6470 [08:37<1:11:05,  1.35it/s]
2|697|Loss: 0.7822044491767883:  11%|█         | 697/6470 [08:37<1:05:36,  1.47it/s]
2|698|Loss: 0.7148582935333252:  11%|█         | 697/6470 [08:37<1:05:36,  1.47it/s]
2|698|Loss: 0.7148582935333252:  11%|█         | 698/6470 [08:38<1:04:58,  1.48it/s]
2|699|Loss: 0.9740157723426819:  11%|█         | 698/6470 [08:38<1:04:58,  1.48it/s]
2|699|Loss: 0.9740157723426819:  11%|█         | 699/6470 [08:38<1:06:11,  1.45it/s]
2|700|Loss: 0.8003721833229065:  11%|█         | 699/6470 [08:39<1:06:11,  1.45it/s]
2|700|Loss: 0.8003721833229065:  11%|█         | 700/6470 [08:39<1:11:42,  1.34it/s]
2|701|Loss: 0.8834957480430603:  11%|█         | 700/6470 [08:40<1:11:42,  1.34it/s]
2|701|Loss: 0.8834957480430603:  11%|█         | 701/6470 [08:40<1:11:29,  1.35it/s]
2|702|Loss: 0.8941531777381897:  11%|█         | 701/6470 [08:41<1:11:29,  1.35it/s]
2|702|Loss: 0.8941531777381897:  11%|█         | 702/6470 [08:41<1:18:45,  1.22it/s]
2|703|Loss: 0.767538845539093:  11%|█         | 702/6470 [08:42<1:18:45,  1.22it/s] 
2|703|Loss: 0.767538845539093:  11%|█         | 703/6470 [08:42<1:17:49,  1.24it/s]
2|704|Loss: 0.9376156330108643:  11%|█         | 703/6470 [08:42<1:17:49,  1.24it/s]
2|704|Loss: 0.9376156330108643:  11%|█         | 704/6470 [08:43<1:16:26,  1.26it/s]
2|705|Loss: 0.9801754355430603:  11%|█         | 704/6470 [08:43<1:16:26,  1.26it/s]
2|705|Loss: 0.9801754355430603:  11%|█         | 705/6470 [08:43<1:08:51,  1.40it/s]
2|706|Loss: 0.8674159049987793:  11%|█         | 705/6470 [08:44<1:08:51,  1.40it/s]
2|706|Loss: 0.8674159049987793:  11%|█         | 706/6470 [08:44<1:06:53,  1.44it/s]
2|707|Loss: 0.9426243901252747:  11%|█         | 706/6470 [08:44<1:06:53,  1.44it/s]
2|707|Loss: 0.9426243901252747:  11%|█         | 707/6470 [08:45<1:07:44,  1.42it/s]
2|708|Loss: 0.7730911374092102:  11%|█         | 707/6470 [08:45<1:07:44,  1.42it/s]
2|708|Loss: 0.7730911374092102:  11%|█         | 708/6470 [08:45<1:13:14,  1.31it/s]
2|709|Loss: 0.956080973148346:  11%|█         | 708/6470 [08:46<1:13:14,  1.31it/s] 
2|709|Loss: 0.956080973148346:  11%|█         | 709/6470 [08:46<1:17:24,  1.24it/s]
2|710|Loss: 0.8119491934776306:  11%|█         | 709/6470 [08:47<1:17:24,  1.24it/s]
2|710|Loss: 0.8119491934776306:  11%|█         | 710/6470 [08:47<1:16:47,  1.25it/s]
2|711|Loss: 0.8024746179580688:  11%|█         | 710/6470 [08:48<1:16:47,  1.25it/s]
2|711|Loss: 0.8024746179580688:  11%|█         | 711/6470 [08:48<1:19:08,  1.21it/s]
2|712|Loss: 0.8724154829978943:  11%|█         | 711/6470 [08:48<1:19:08,  1.21it/s]
2|712|Loss: 0.8724154829978943:  11%|█         | 712/6470 [08:49<1:18:59,  1.21it/s]
2|713|Loss: 0.9692932963371277:  11%|█         | 712/6470 [08:49<1:18:59,  1.21it/s]
2|713|Loss: 0.9692932963371277:  11%|█         | 713/6470 [08:50<1:19:54,  1.20it/s]
2|714|Loss: 0.8424625992774963:  11%|█         | 713/6470 [08:50<1:19:54,  1.20it/s]
2|714|Loss: 0.8424625992774963:  11%|█         | 714/6470 [08:50<1:18:33,  1.22it/s]
2|715|Loss: 0.8192040920257568:  11%|█         | 714/6470 [08:51<1:18:33,  1.22it/s]
2|715|Loss: 0.8192040920257568:  11%|█         | 715/6470 [08:52<1:24:01,  1.14it/s]
2|716|Loss: 0.877640962600708:  11%|█         | 715/6470 [08:52<1:24:01,  1.14it/s] 
2|716|Loss: 0.877640962600708:  11%|█         | 716/6470 [08:52<1:16:12,  1.26it/s]
2|717|Loss: 0.8710021376609802:  11%|█         | 716/6470 [08:52<1:16:12,  1.26it/s]
2|717|Loss: 0.8710021376609802:  11%|█         | 717/6470 [08:53<1:12:17,  1.33it/s]
2|718|Loss: 0.6372862458229065:  11%|█         | 717/6470 [08:53<1:12:17,  1.33it/s]
2|718|Loss: 0.6372862458229065:  11%|█         | 718/6470 [08:54<1:19:24,  1.21it/s]
2|719|Loss: 0.7940396070480347:  11%|█         | 718/6470 [08:54<1:19:24,  1.21it/s]
2|719|Loss: 0.7940396070480347:  11%|█         | 719/6470 [08:55<1:20:18,  1.19it/s]
2|720|Loss: 0.7527729272842407:  11%|█         | 719/6470 [08:55<1:20:18,  1.19it/s]
2|720|Loss: 0.7527729272842407:  11%|█         | 720/6470 [08:55<1:19:20,  1.21it/s]
2|721|Loss: 0.8107907772064209:  11%|█         | 720/6470 [08:56<1:19:20,  1.21it/s]
2|721|Loss: 0.8107907772064209:  11%|█         | 721/6470 [08:56<1:14:48,  1.28it/s]
2|722|Loss: 0.8459481596946716:  11%|█         | 721/6470 [08:57<1:14:48,  1.28it/s]
2|722|Loss: 0.8459481596946716:  11%|█         | 722/6470 [08:57<1:18:09,  1.23it/s]
2|723|Loss: 0.8071697950363159:  11%|█         | 722/6470 [08:57<1:18:09,  1.23it/s]
2|723|Loss: 0.8071697950363159:  11%|█         | 723/6470 [08:58<1:18:56,  1.21it/s]
2|724|Loss: 0.7897305488586426:  11%|█         | 723/6470 [08:58<1:18:56,  1.21it/s]
2|724|Loss: 0.7897305488586426:  11%|█         | 724/6470 [08:59<1:17:25,  1.24it/s]
2|725|Loss: 0.696234405040741:  11%|█         | 724/6470 [08:59<1:17:25,  1.24it/s] 
2|725|Loss: 0.696234405040741:  11%|█         | 725/6470 [09:00<1:22:37,  1.16it/s]
2|726|Loss: 0.8055397868156433:  11%|█         | 725/6470 [09:00<1:22:37,  1.16it/s]
2|726|Loss: 0.8055397868156433:  11%|█         | 726/6470 [09:01<1:26:11,  1.11it/s]
2|727|Loss: 0.802237868309021:  11%|█         | 726/6470 [09:01<1:26:11,  1.11it/s] 
2|727|Loss: 0.802237868309021:  11%|█         | 727/6470 [09:01<1:23:56,  1.14it/s]
2|728|Loss: 0.6955173015594482:  11%|█         | 727/6470 [09:02<1:23:56,  1.14it/s]
2|728|Loss: 0.6955173015594482:  11%|█▏        | 728/6470 [09:02<1:21:31,  1.17it/s]
2|729|Loss: 0.8636620044708252:  11%|█▏        | 728/6470 [09:03<1:21:31,  1.17it/s]
2|729|Loss: 0.8636620044708252:  11%|█▏        | 729/6470 [09:03<1:16:51,  1.24it/s]
2|730|Loss: 0.9233415722846985:  11%|█▏        | 729/6470 [09:03<1:16:51,  1.24it/s]
2|730|Loss: 0.9233415722846985:  11%|█▏        | 730/6470 [09:04<1:19:29,  1.20it/s]
2|731|Loss: 0.9186766743659973:  11%|█▏        | 730/6470 [09:04<1:19:29,  1.20it/s]
2|731|Loss: 0.9186766743659973:  11%|█▏        | 731/6470 [09:05<1:19:29,  1.20it/s]
2|732|Loss: 0.8058193922042847:  11%|█▏        | 731/6470 [09:05<1:19:29,  1.20it/s]
2|732|Loss: 0.8058193922042847:  11%|█▏        | 732/6470 [09:05<1:15:51,  1.26it/s]
2|733|Loss: 0.7833632826805115:  11%|█▏        | 732/6470 [09:06<1:15:51,  1.26it/s]
2|733|Loss: 0.7833632826805115:  11%|█▏        | 733/6470 [09:06<1:12:13,  1.32it/s]
2|734|Loss: 0.8216680884361267:  11%|█▏        | 733/6470 [09:06<1:12:13,  1.32it/s]
2|734|Loss: 0.8216680884361267:  11%|█▏        | 734/6470 [09:07<1:11:44,  1.33it/s]
2|735|Loss: 0.7962138652801514:  11%|█▏        | 734/6470 [09:07<1:11:44,  1.33it/s]
2|735|Loss: 0.7962138652801514:  11%|█▏        | 735/6470 [09:07<1:10:48,  1.35it/s]
2|736|Loss: 0.8878017663955688:  11%|█▏        | 735/6470 [09:08<1:10:48,  1.35it/s]
2|736|Loss: 0.8878017663955688:  11%|█▏        | 736/6470 [09:08<1:05:54,  1.45it/s]
2|737|Loss: 0.9465023279190063:  11%|█▏        | 736/6470 [09:08<1:05:54,  1.45it/s]
2|737|Loss: 0.9465023279190063:  11%|█▏        | 737/6470 [09:09<1:07:49,  1.41it/s]
2|738|Loss: 0.8448207974433899:  11%|█▏        | 737/6470 [09:09<1:07:49,  1.41it/s]
2|738|Loss: 0.8448207974433899:  11%|█▏        | 738/6470 [09:09<1:07:33,  1.41it/s]
2|739|Loss: 0.9150151610374451:  11%|█▏        | 738/6470 [09:10<1:07:33,  1.41it/s]
2|739|Loss: 0.9150151610374451:  11%|█▏        | 739/6470 [09:10<1:11:59,  1.33it/s]
2|740|Loss: 0.7989131212234497:  11%|█▏        | 739/6470 [09:11<1:11:59,  1.33it/s]
2|740|Loss: 0.7989131212234497:  11%|█▏        | 740/6470 [09:11<1:09:35,  1.37it/s]
2|741|Loss: 0.8886140584945679:  11%|█▏        | 740/6470 [09:11<1:09:35,  1.37it/s]
2|741|Loss: 0.8886140584945679:  11%|█▏        | 741/6470 [09:12<1:12:18,  1.32it/s]
2|742|Loss: 0.8323084115982056:  11%|█▏        | 741/6470 [09:12<1:12:18,  1.32it/s]
2|742|Loss: 0.8323084115982056:  11%|█▏        | 742/6470 [09:13<1:11:02,  1.34it/s]
2|743|Loss: 0.8995601534843445:  11%|█▏        | 742/6470 [09:13<1:11:02,  1.34it/s]
2|743|Loss: 0.8995601534843445:  11%|█▏        | 743/6470 [09:13<1:11:25,  1.34it/s]
2|744|Loss: 0.9731184840202332:  11%|█▏        | 743/6470 [09:14<1:11:25,  1.34it/s]
2|744|Loss: 0.9731184840202332:  11%|█▏        | 744/6470 [09:14<1:10:22,  1.36it/s]
2|745|Loss: 0.83318692445755:  11%|█▏        | 744/6470 [09:14<1:10:22,  1.36it/s]  
2|745|Loss: 0.83318692445755:  12%|█▏        | 745/6470 [09:15<1:13:26,  1.30it/s]
2|746|Loss: 0.8463438153266907:  12%|█▏        | 745/6470 [09:15<1:13:26,  1.30it/s]
2|746|Loss: 0.8463438153266907:  12%|█▏        | 746/6470 [09:15<1:07:19,  1.42it/s]
2|747|Loss: 0.9397853016853333:  12%|█▏        | 746/6470 [09:16<1:07:19,  1.42it/s]
2|747|Loss: 0.9397853016853333:  12%|█▏        | 747/6470 [09:16<1:05:15,  1.46it/s]
2|748|Loss: 0.8398342728614807:  12%|█▏        | 747/6470 [09:16<1:05:15,  1.46it/s]
2|748|Loss: 0.8398342728614807:  12%|█▏        | 748/6470 [09:17<1:08:05,  1.40it/s]
2|749|Loss: 0.6727302074432373:  12%|█▏        | 748/6470 [09:17<1:08:05,  1.40it/s]
2|749|Loss: 0.6727302074432373:  12%|█▏        | 749/6470 [09:17<1:05:04,  1.47it/s]
2|750|Loss: 0.8116344809532166:  12%|█▏        | 749/6470 [09:18<1:05:04,  1.47it/s]
2|750|Loss: 0.8116344809532166:  12%|█▏        | 750/6470 [09:18<1:02:48,  1.52it/s]
2|751|Loss: 0.8247528672218323:  12%|█▏        | 750/6470 [09:18<1:02:48,  1.52it/s]
2|751|Loss: 0.8247528672218323:  12%|█▏        | 751/6470 [09:19<1:03:57,  1.49it/s]
2|752|Loss: 0.7227783799171448:  12%|█▏        | 751/6470 [09:19<1:03:57,  1.49it/s]
2|752|Loss: 0.7227783799171448:  12%|█▏        | 752/6470 [09:19<1:04:51,  1.47it/s]
2|753|Loss: 0.8928709030151367:  12%|█▏        | 752/6470 [09:20<1:04:51,  1.47it/s]
2|753|Loss: 0.8928709030151367:  12%|█▏        | 753/6470 [09:20<1:10:02,  1.36it/s]
2|754|Loss: 0.7906118035316467:  12%|█▏        | 753/6470 [09:21<1:10:02,  1.36it/s]
2|754|Loss: 0.7906118035316467:  12%|█▏        | 754/6470 [09:21<1:13:02,  1.30it/s]
2|755|Loss: 0.8937163949012756:  12%|█▏        | 754/6470 [09:22<1:13:02,  1.30it/s]
2|755|Loss: 0.8937163949012756:  12%|█▏        | 755/6470 [09:22<1:17:16,  1.23it/s]
2|756|Loss: 1.0387225151062012:  12%|█▏        | 755/6470 [09:23<1:17:16,  1.23it/s]
2|756|Loss: 1.0387225151062012:  12%|█▏        | 756/6470 [09:23<1:15:41,  1.26it/s]
2|757|Loss: 0.8837016224861145:  12%|█▏        | 756/6470 [09:23<1:15:41,  1.26it/s]
2|757|Loss: 0.8837016224861145:  12%|█▏        | 757/6470 [09:24<1:12:38,  1.31it/s]
2|758|Loss: 0.8841999769210815:  12%|█▏        | 757/6470 [09:24<1:12:38,  1.31it/s]
2|758|Loss: 0.8841999769210815:  12%|█▏        | 758/6470 [09:24<1:13:12,  1.30it/s]
2|759|Loss: 0.7967047691345215:  12%|█▏        | 758/6470 [09:25<1:13:12,  1.30it/s]
2|759|Loss: 0.7967047691345215:  12%|█▏        | 759/6470 [09:25<1:18:33,  1.21it/s]
2|760|Loss: 0.9330202341079712:  12%|█▏        | 759/6470 [09:26<1:18:33,  1.21it/s]
2|760|Loss: 0.9330202341079712:  12%|█▏        | 760/6470 [09:26<1:19:07,  1.20it/s]
2|761|Loss: 0.9464709162712097:  12%|█▏        | 760/6470 [09:26<1:19:07,  1.20it/s]
2|761|Loss: 0.9464709162712097:  12%|█▏        | 761/6470 [09:27<1:12:27,  1.31it/s]
2|762|Loss: 0.9101340770721436:  12%|█▏        | 761/6470 [09:27<1:12:27,  1.31it/s]
2|762|Loss: 0.9101340770721436:  12%|█▏        | 762/6470 [09:28<1:16:43,  1.24it/s]
2|763|Loss: 0.7864876985549927:  12%|█▏        | 762/6470 [09:28<1:16:43,  1.24it/s]
2|763|Loss: 0.7864876985549927:  12%|█▏        | 763/6470 [09:29<1:19:24,  1.20it/s]
2|764|Loss: 0.8265490531921387:  12%|█▏        | 763/6470 [09:29<1:19:24,  1.20it/s]
2|764|Loss: 0.8265490531921387:  12%|█▏        | 764/6470 [09:30<1:24:58,  1.12it/s]
2|765|Loss: 1.0434272289276123:  12%|█▏        | 764/6470 [09:30<1:24:58,  1.12it/s]
2|765|Loss: 1.0434272289276123:  12%|█▏        | 765/6470 [09:30<1:20:38,  1.18it/s]
2|766|Loss: 0.718345046043396:  12%|█▏        | 765/6470 [09:31<1:20:38,  1.18it/s] 
2|766|Loss: 0.718345046043396:  12%|█▏        | 766/6470 [09:31<1:24:26,  1.13it/s]
2|767|Loss: 0.7960894107818604:  12%|█▏        | 766/6470 [09:32<1:24:26,  1.13it/s]
2|767|Loss: 0.7960894107818604:  12%|█▏        | 767/6470 [09:32<1:24:35,  1.12it/s]
2|768|Loss: 0.8151129484176636:  12%|█▏        | 767/6470 [09:33<1:24:35,  1.12it/s]
2|768|Loss: 0.8151129484176636:  12%|█▏        | 768/6470 [09:33<1:25:04,  1.12it/s]
2|769|Loss: 0.7038633823394775:  12%|█▏        | 768/6470 [09:33<1:25:04,  1.12it/s]
2|769|Loss: 0.7038633823394775:  12%|█▏        | 769/6470 [09:34<1:15:39,  1.26it/s]
2|770|Loss: 1.005259394645691:  12%|█▏        | 769/6470 [09:34<1:15:39,  1.26it/s] 
2|770|Loss: 1.005259394645691:  12%|█▏        | 770/6470 [09:34<1:13:55,  1.29it/s]
2|771|Loss: 0.904507040977478:  12%|█▏        | 770/6470 [09:35<1:13:55,  1.29it/s]
2|771|Loss: 0.904507040977478:  12%|█▏        | 771/6470 [09:35<1:13:40,  1.29it/s]
2|772|Loss: 0.7757850885391235:  12%|█▏        | 771/6470 [09:36<1:13:40,  1.29it/s]
2|772|Loss: 0.7757850885391235:  12%|█▏        | 772/6470 [09:36<1:09:36,  1.36it/s]
2|773|Loss: 0.9203529953956604:  12%|█▏        | 772/6470 [09:36<1:09:36,  1.36it/s]
2|773|Loss: 0.9203529953956604:  12%|█▏        | 773/6470 [09:37<1:10:09,  1.35it/s]
2|774|Loss: 0.8338379859924316:  12%|█▏        | 773/6470 [09:37<1:10:09,  1.35it/s]
2|774|Loss: 0.8338379859924316:  12%|█▏        | 774/6470 [09:37<1:09:40,  1.36it/s]
2|775|Loss: 0.9171616435050964:  12%|█▏        | 774/6470 [09:38<1:09:40,  1.36it/s]
2|775|Loss: 0.9171616435050964:  12%|█▏        | 775/6470 [09:38<1:14:20,  1.28it/s]
2|776|Loss: 0.9417882561683655:  12%|█▏        | 775/6470 [09:39<1:14:20,  1.28it/s]
2|776|Loss: 0.9417882561683655:  12%|█▏        | 776/6470 [09:39<1:15:52,  1.25it/s]
2|777|Loss: 0.8004929423332214:  12%|█▏        | 776/6470 [09:39<1:15:52,  1.25it/s]
2|777|Loss: 0.8004929423332214:  12%|█▏        | 777/6470 [09:40<1:17:51,  1.22it/s]
2|778|Loss: 0.7695116400718689:  12%|█▏        | 777/6470 [09:40<1:17:51,  1.22it/s]
2|778|Loss: 0.7695116400718689:  12%|█▏        | 778/6470 [09:41<1:15:17,  1.26it/s]
2|779|Loss: 0.8808003067970276:  12%|█▏        | 778/6470 [09:41<1:15:17,  1.26it/s]
2|779|Loss: 0.8808003067970276:  12%|█▏        | 779/6470 [09:41<1:14:31,  1.27it/s]
2|780|Loss: 0.9517338871955872:  12%|█▏        | 779/6470 [09:42<1:14:31,  1.27it/s]
2|780|Loss: 0.9517338871955872:  12%|█▏        | 780/6470 [09:42<1:10:44,  1.34it/s]
2|781|Loss: 0.8652254343032837:  12%|█▏        | 780/6470 [09:42<1:10:44,  1.34it/s]
2|781|Loss: 0.8652254343032837:  12%|█▏        | 781/6470 [09:43<1:11:30,  1.33it/s]
2|782|Loss: 0.9251594543457031:  12%|█▏        | 781/6470 [09:43<1:11:30,  1.33it/s]
2|782|Loss: 0.9251594543457031:  12%|█▏        | 782/6470 [09:44<1:12:47,  1.30it/s]
2|783|Loss: 0.9921729564666748:  12%|█▏        | 782/6470 [09:44<1:12:47,  1.30it/s]
2|783|Loss: 0.9921729564666748:  12%|█▏        | 783/6470 [09:44<1:13:51,  1.28it/s]
2|784|Loss: 0.8526086211204529:  12%|█▏        | 783/6470 [09:45<1:13:51,  1.28it/s]
2|784|Loss: 0.8526086211204529:  12%|█▏        | 784/6470 [09:45<1:14:56,  1.26it/s]
2|785|Loss: 1.0642929077148438:  12%|█▏        | 784/6470 [09:46<1:14:56,  1.26it/s]
2|785|Loss: 1.0642929077148438:  12%|█▏        | 785/6470 [09:46<1:15:55,  1.25it/s]
2|786|Loss: 0.8729838728904724:  12%|█▏        | 785/6470 [09:46<1:15:55,  1.25it/s]
2|786|Loss: 0.8729838728904724:  12%|█▏        | 786/6470 [09:47<1:15:44,  1.25it/s]
2|787|Loss: 0.8184583187103271:  12%|█▏        | 786/6470 [09:47<1:15:44,  1.25it/s]
2|787|Loss: 0.8184583187103271:  12%|█▏        | 787/6470 [09:48<1:15:44,  1.25it/s]
2|788|Loss: 0.8400371074676514:  12%|█▏        | 787/6470 [09:48<1:15:44,  1.25it/s]
2|788|Loss: 0.8400371074676514:  12%|█▏        | 788/6470 [09:49<1:18:04,  1.21it/s]
2|789|Loss: 0.9906778335571289:  12%|█▏        | 788/6470 [09:49<1:18:04,  1.21it/s]
2|789|Loss: 0.9906778335571289:  12%|█▏        | 789/6470 [09:49<1:17:21,  1.22it/s]
2|790|Loss: 0.8680434823036194:  12%|█▏        | 789/6470 [09:50<1:17:21,  1.22it/s]
2|790|Loss: 0.8680434823036194:  12%|█▏        | 790/6470 [09:50<1:13:39,  1.29it/s]
2|791|Loss: 0.935401976108551:  12%|█▏        | 790/6470 [09:50<1:13:39,  1.29it/s] 
2|791|Loss: 0.935401976108551:  12%|█▏        | 791/6470 [09:51<1:11:15,  1.33it/s]
2|792|Loss: 0.7107942700386047:  12%|█▏        | 791/6470 [09:51<1:11:15,  1.33it/s]
2|792|Loss: 0.7107942700386047:  12%|█▏        | 792/6470 [09:52<1:14:22,  1.27it/s]
2|793|Loss: 0.9270992875099182:  12%|█▏        | 792/6470 [09:52<1:14:22,  1.27it/s]
2|793|Loss: 0.9270992875099182:  12%|█▏        | 793/6470 [09:52<1:15:22,  1.26it/s]
2|794|Loss: 0.9193735718727112:  12%|█▏        | 793/6470 [09:53<1:15:22,  1.26it/s]
2|794|Loss: 0.9193735718727112:  12%|█▏        | 794/6470 [09:53<1:13:18,  1.29it/s]
2|795|Loss: 0.9503260850906372:  12%|█▏        | 794/6470 [09:54<1:13:18,  1.29it/s]
2|795|Loss: 0.9503260850906372:  12%|█▏        | 795/6470 [09:54<1:12:37,  1.30it/s]
2|796|Loss: 0.8377951979637146:  12%|█▏        | 795/6470 [09:54<1:12:37,  1.30it/s]
2|796|Loss: 0.8377951979637146:  12%|█▏        | 796/6470 [09:55<1:13:21,  1.29it/s]
2|797|Loss: 0.8793575167655945:  12%|█▏        | 796/6470 [09:55<1:13:21,  1.29it/s]
2|797|Loss: 0.8793575167655945:  12%|█▏        | 797/6470 [09:55<1:11:37,  1.32it/s]
2|798|Loss: 0.9157570004463196:  12%|█▏        | 797/6470 [09:56<1:11:37,  1.32it/s]
2|798|Loss: 0.9157570004463196:  12%|█▏        | 798/6470 [09:56<1:09:23,  1.36it/s]
2|799|Loss: 0.8861597776412964:  12%|█▏        | 798/6470 [09:56<1:09:23,  1.36it/s]
2|799|Loss: 0.8861597776412964:  12%|█▏        | 799/6470 [09:57<1:11:10,  1.33it/s]
2|800|Loss: 0.7438315153121948:  12%|█▏        | 799/6470 [09:57<1:11:10,  1.33it/s]
2|800|Loss: 0.7438315153121948:  12%|█▏        | 800/6470 [09:57<1:08:08,  1.39it/s]
2|801|Loss: 0.769889771938324:  12%|█▏        | 800/6470 [09:58<1:08:08,  1.39it/s] 
2|801|Loss: 0.769889771938324:  12%|█▏        | 801/6470 [09:58<1:09:40,  1.36it/s]
2|802|Loss: 0.871908962726593:  12%|█▏        | 801/6470 [09:59<1:09:40,  1.36it/s]
2|802|Loss: 0.871908962726593:  12%|█▏        | 802/6470 [09:59<1:11:34,  1.32it/s]
2|803|Loss: 0.8599017858505249:  12%|█▏        | 802/6470 [09:59<1:11:34,  1.32it/s]
2|803|Loss: 0.8599017858505249:  12%|█▏        | 803/6470 [10:00<1:07:27,  1.40it/s]
2|804|Loss: 0.7929036617279053:  12%|█▏        | 803/6470 [10:00<1:07:27,  1.40it/s]
2|804|Loss: 0.7929036617279053:  12%|█▏        | 804/6470 [10:00<1:06:11,  1.43it/s]
2|805|Loss: 0.8224503397941589:  12%|█▏        | 804/6470 [10:01<1:06:11,  1.43it/s]
2|805|Loss: 0.8224503397941589:  12%|█▏        | 805/6470 [10:01<1:10:56,  1.33it/s]
2|806|Loss: 0.9357049465179443:  12%|█▏        | 805/6470 [10:02<1:10:56,  1.33it/s]
2|806|Loss: 0.9357049465179443:  12%|█▏        | 806/6470 [10:02<1:14:26,  1.27it/s]
2|807|Loss: 0.6378715634346008:  12%|█▏        | 806/6470 [10:03<1:14:26,  1.27it/s]
2|807|Loss: 0.6378715634346008:  12%|█▏        | 807/6470 [10:03<1:13:06,  1.29it/s]
2|808|Loss: 0.7663320899009705:  12%|█▏        | 807/6470 [10:03<1:13:06,  1.29it/s]
2|808|Loss: 0.7663320899009705:  12%|█▏        | 808/6470 [10:04<1:16:00,  1.24it/s]
2|809|Loss: 0.8490046858787537:  12%|█▏        | 808/6470 [10:04<1:16:00,  1.24it/s]
2|809|Loss: 0.8490046858787537:  13%|█▎        | 809/6470 [10:04<1:10:40,  1.33it/s]
2|810|Loss: 0.7113962173461914:  13%|█▎        | 809/6470 [10:05<1:10:40,  1.33it/s]
2|810|Loss: 0.7113962173461914:  13%|█▎        | 810/6470 [10:05<1:10:58,  1.33it/s]
2|811|Loss: 0.7972657084465027:  13%|█▎        | 810/6470 [10:06<1:10:58,  1.33it/s]
2|811|Loss: 0.7972657084465027:  13%|█▎        | 811/6470 [10:06<1:12:47,  1.30it/s]
2|812|Loss: 0.8934042453765869:  13%|█▎        | 811/6470 [10:06<1:12:47,  1.30it/s]
2|812|Loss: 0.8934042453765869:  13%|█▎        | 812/6470 [10:07<1:16:08,  1.24it/s]
2|813|Loss: 0.9209563136100769:  13%|█▎        | 812/6470 [10:07<1:16:08,  1.24it/s]
2|813|Loss: 0.9209563136100769:  13%|█▎        | 813/6470 [10:07<1:12:38,  1.30it/s]
2|814|Loss: 0.9035558104515076:  13%|█▎        | 813/6470 [10:08<1:12:38,  1.30it/s]
2|814|Loss: 0.9035558104515076:  13%|█▎        | 814/6470 [10:08<1:11:44,  1.31it/s]
2|815|Loss: 0.8720519542694092:  13%|█▎        | 814/6470 [10:09<1:11:44,  1.31it/s]
2|815|Loss: 0.8720519542694092:  13%|█▎        | 815/6470 [10:09<1:12:05,  1.31it/s]
2|816|Loss: 0.8226686716079712:  13%|█▎        | 815/6470 [10:09<1:12:05,  1.31it/s]
2|816|Loss: 0.8226686716079712:  13%|█▎        | 816/6470 [10:10<1:08:28,  1.38it/s]
2|817|Loss: 0.8399897217750549:  13%|█▎        | 816/6470 [10:10<1:08:28,  1.38it/s]
2|817|Loss: 0.8399897217750549:  13%|█▎        | 817/6470 [10:10<1:11:59,  1.31it/s]
2|818|Loss: 0.9416171908378601:  13%|█▎        | 817/6470 [10:11<1:11:59,  1.31it/s]
2|818|Loss: 0.9416171908378601:  13%|█▎        | 818/6470 [10:11<1:12:52,  1.29it/s]
2|819|Loss: 0.7492145299911499:  13%|█▎        | 818/6470 [10:12<1:12:52,  1.29it/s]
2|819|Loss: 0.7492145299911499:  13%|█▎        | 819/6470 [10:12<1:15:35,  1.25it/s]
2|820|Loss: 0.7519813776016235:  13%|█▎        | 819/6470 [10:13<1:15:35,  1.25it/s]
2|820|Loss: 0.7519813776016235:  13%|█▎        | 820/6470 [10:13<1:10:17,  1.34it/s]
2|821|Loss: 0.8033989667892456:  13%|█▎        | 820/6470 [10:13<1:10:17,  1.34it/s]
2|821|Loss: 0.8033989667892456:  13%|█▎        | 821/6470 [10:13<1:06:12,  1.42it/s]
2|822|Loss: 0.7531992793083191:  13%|█▎        | 821/6470 [10:14<1:06:12,  1.42it/s]
2|822|Loss: 0.7531992793083191:  13%|█▎        | 822/6470 [10:14<1:05:49,  1.43it/s]
2|823|Loss: 0.804767906665802:  13%|█▎        | 822/6470 [10:14<1:05:49,  1.43it/s] 
2|823|Loss: 0.804767906665802:  13%|█▎        | 823/6470 [10:15<1:08:34,  1.37it/s]
2|824|Loss: 0.975075900554657:  13%|█▎        | 823/6470 [10:15<1:08:34,  1.37it/s]
2|824|Loss: 0.975075900554657:  13%|█▎        | 824/6470 [10:16<1:08:39,  1.37it/s]
2|825|Loss: 0.8411017060279846:  13%|█▎        | 824/6470 [10:16<1:08:39,  1.37it/s]
2|825|Loss: 0.8411017060279846:  13%|█▎        | 825/6470 [10:16<1:05:24,  1.44it/s]
2|826|Loss: 0.8717726469039917:  13%|█▎        | 825/6470 [10:17<1:05:24,  1.44it/s]
2|826|Loss: 0.8717726469039917:  13%|█▎        | 826/6470 [10:17<1:05:02,  1.45it/s]
2|827|Loss: 0.9742997288703918:  13%|█▎        | 826/6470 [10:17<1:05:02,  1.45it/s]
2|827|Loss: 0.9742997288703918:  13%|█▎        | 827/6470 [10:18<1:05:24,  1.44it/s]
2|828|Loss: 0.86274653673172:  13%|█▎        | 827/6470 [10:18<1:05:24,  1.44it/s]  
2|828|Loss: 0.86274653673172:  13%|█▎        | 828/6470 [10:18<1:03:47,  1.47it/s]
2|829|Loss: 0.798168420791626:  13%|█▎        | 828/6470 [10:19<1:03:47,  1.47it/s]
2|829|Loss: 0.798168420791626:  13%|█▎        | 829/6470 [10:19<1:05:12,  1.44it/s]
2|830|Loss: 0.8339502215385437:  13%|█▎        | 829/6470 [10:19<1:05:12,  1.44it/s]
2|830|Loss: 0.8339502215385437:  13%|█▎        | 830/6470 [10:20<1:05:49,  1.43it/s]
2|831|Loss: 0.8088239431381226:  13%|█▎        | 830/6470 [10:20<1:05:49,  1.43it/s]
2|831|Loss: 0.8088239431381226:  13%|█▎        | 831/6470 [10:20<1:07:09,  1.40it/s]
2|832|Loss: 0.8110731244087219:  13%|█▎        | 831/6470 [10:21<1:07:09,  1.40it/s]
2|832|Loss: 0.8110731244087219:  13%|█▎        | 832/6470 [10:21<1:09:15,  1.36it/s]
2|833|Loss: 0.7998161315917969:  13%|█▎        | 832/6470 [10:22<1:09:15,  1.36it/s]
2|833|Loss: 0.7998161315917969:  13%|█▎        | 833/6470 [10:22<1:11:00,  1.32it/s]
2|834|Loss: 0.8731154203414917:  13%|█▎        | 833/6470 [10:22<1:11:00,  1.32it/s]
2|834|Loss: 0.8731154203414917:  13%|█▎        | 834/6470 [10:23<1:10:03,  1.34it/s]
2|835|Loss: 0.8830259442329407:  13%|█▎        | 834/6470 [10:23<1:10:03,  1.34it/s]
2|835|Loss: 0.8830259442329407:  13%|█▎        | 835/6470 [10:23<1:08:10,  1.38it/s]
2|836|Loss: 0.7532538771629333:  13%|█▎        | 835/6470 [10:24<1:08:10,  1.38it/s]
2|836|Loss: 0.7532538771629333:  13%|█▎        | 836/6470 [10:24<1:07:02,  1.40it/s]
2|837|Loss: 0.7244934439659119:  13%|█▎        | 836/6470 [10:24<1:07:02,  1.40it/s]
2|837|Loss: 0.7244934439659119:  13%|█▎        | 837/6470 [10:25<1:06:27,  1.41it/s]
2|838|Loss: 0.828140377998352:  13%|█▎        | 837/6470 [10:25<1:06:27,  1.41it/s] 
2|838|Loss: 0.828140377998352:  13%|█▎        | 838/6470 [10:26<1:07:25,  1.39it/s]
2|839|Loss: 0.9091503024101257:  13%|█▎        | 838/6470 [10:26<1:07:25,  1.39it/s]
2|839|Loss: 0.9091503024101257:  13%|█▎        | 839/6470 [10:26<1:08:41,  1.37it/s]
2|840|Loss: 0.8113047480583191:  13%|█▎        | 839/6470 [10:27<1:08:41,  1.37it/s]
2|840|Loss: 0.8113047480583191:  13%|█▎        | 840/6470 [10:27<1:08:13,  1.38it/s]
2|841|Loss: 0.9278340339660645:  13%|█▎        | 840/6470 [10:27<1:08:13,  1.38it/s]
2|841|Loss: 0.9278340339660645:  13%|█▎        | 841/6470 [10:28<1:02:28,  1.50it/s]
2|842|Loss: 0.9518119692802429:  13%|█▎        | 841/6470 [10:28<1:02:28,  1.50it/s]
2|842|Loss: 0.9518119692802429:  13%|█▎        | 842/6470 [10:28<1:04:11,  1.46it/s]
2|843|Loss: 0.919633686542511:  13%|█▎        | 842/6470 [10:29<1:04:11,  1.46it/s] 
2|843|Loss: 0.919633686542511:  13%|█▎        | 843/6470 [10:29<1:07:41,  1.39it/s]
2|844|Loss: 0.798687219619751:  13%|█▎        | 843/6470 [10:30<1:07:41,  1.39it/s]
2|844|Loss: 0.798687219619751:  13%|█▎        | 844/6470 [10:30<1:10:29,  1.33it/s]
2|845|Loss: 0.8254308104515076:  13%|█▎        | 844/6470 [10:30<1:10:29,  1.33it/s]
2|845|Loss: 0.8254308104515076:  13%|█▎        | 845/6470 [10:31<1:09:44,  1.34it/s]
2|846|Loss: 0.8819612264633179:  13%|█▎        | 845/6470 [10:31<1:09:44,  1.34it/s]
2|846|Loss: 0.8819612264633179:  13%|█▎        | 846/6470 [10:31<1:11:06,  1.32it/s]
2|847|Loss: 0.798933207988739:  13%|█▎        | 846/6470 [10:32<1:11:06,  1.32it/s] 
2|847|Loss: 0.798933207988739:  13%|█▎        | 847/6470 [10:32<1:14:49,  1.25it/s]
2|848|Loss: 0.7181891202926636:  13%|█▎        | 847/6470 [10:33<1:14:49,  1.25it/s]
2|848|Loss: 0.7181891202926636:  13%|█▎        | 848/6470 [10:33<1:14:30,  1.26it/s]
2|849|Loss: 0.8772415518760681:  13%|█▎        | 848/6470 [10:33<1:14:30,  1.26it/s]
2|849|Loss: 0.8772415518760681:  13%|█▎        | 849/6470 [10:34<1:10:31,  1.33it/s]
2|850|Loss: 0.700825035572052:  13%|█▎        | 849/6470 [10:34<1:10:31,  1.33it/s] 
2|850|Loss: 0.700825035572052:  13%|█▎        | 850/6470 [10:35<1:10:40,  1.33it/s]
2|851|Loss: 1.0258491039276123:  13%|█▎        | 850/6470 [10:35<1:10:40,  1.33it/s]
2|851|Loss: 1.0258491039276123:  13%|█▎        | 851/6470 [10:35<1:10:53,  1.32it/s]
2|852|Loss: 0.9396629333496094:  13%|█▎        | 851/6470 [10:36<1:10:53,  1.32it/s]
2|852|Loss: 0.9396629333496094:  13%|█▎        | 852/6470 [10:36<1:08:58,  1.36it/s]
2|853|Loss: 0.8615926504135132:  13%|█▎        | 852/6470 [10:36<1:08:58,  1.36it/s]
2|853|Loss: 0.8615926504135132:  13%|█▎        | 853/6470 [10:37<1:04:37,  1.45it/s]
2|854|Loss: 0.8077091574668884:  13%|█▎        | 853/6470 [10:37<1:04:37,  1.45it/s]
2|854|Loss: 0.8077091574668884:  13%|█▎        | 854/6470 [10:37<1:06:10,  1.41it/s]
2|855|Loss: 0.8661792278289795:  13%|█▎        | 854/6470 [10:38<1:06:10,  1.41it/s]
2|855|Loss: 0.8661792278289795:  13%|█▎        | 855/6470 [10:38<1:07:51,  1.38it/s]
2|856|Loss: 0.8186184167861938:  13%|█▎        | 855/6470 [10:38<1:07:51,  1.38it/s]
2|856|Loss: 0.8186184167861938:  13%|█▎        | 856/6470 [10:39<1:09:35,  1.34it/s]
2|857|Loss: 0.8616242408752441:  13%|█▎        | 856/6470 [10:39<1:09:35,  1.34it/s]
2|857|Loss: 0.8616242408752441:  13%|█▎        | 857/6470 [10:40<1:12:41,  1.29it/s]
2|858|Loss: 0.8952930569648743:  13%|█▎        | 857/6470 [10:40<1:12:41,  1.29it/s]
2|858|Loss: 0.8952930569648743:  13%|█▎        | 858/6470 [10:40<1:11:38,  1.31it/s]
2|859|Loss: 0.879156768321991:  13%|█▎        | 858/6470 [10:41<1:11:38,  1.31it/s] 
2|859|Loss: 0.879156768321991:  13%|█▎        | 859/6470 [10:41<1:15:09,  1.24it/s]
2|860|Loss: 0.9568052887916565:  13%|█▎        | 859/6470 [10:42<1:15:09,  1.24it/s]
2|860|Loss: 0.9568052887916565:  13%|█▎        | 860/6470 [10:42<1:18:46,  1.19it/s]
2|861|Loss: 0.8841750621795654:  13%|█▎        | 860/6470 [10:43<1:18:46,  1.19it/s]
2|861|Loss: 0.8841750621795654:  13%|█▎        | 861/6470 [10:43<1:14:49,  1.25it/s]
2|862|Loss: 0.8537145853042603:  13%|█▎        | 861/6470 [10:43<1:14:49,  1.25it/s]
2|862|Loss: 0.8537145853042603:  13%|█▎        | 862/6470 [10:44<1:15:57,  1.23it/s]
2|863|Loss: 0.8371063470840454:  13%|█▎        | 862/6470 [10:44<1:15:57,  1.23it/s]
2|863|Loss: 0.8371063470840454:  13%|█▎        | 863/6470 [10:45<1:15:45,  1.23it/s]
2|864|Loss: 0.9455414414405823:  13%|█▎        | 863/6470 [10:45<1:15:45,  1.23it/s]
2|864|Loss: 0.9455414414405823:  13%|█▎        | 864/6470 [10:45<1:17:20,  1.21it/s]
2|865|Loss: 0.7735261917114258:  13%|█▎        | 864/6470 [10:46<1:17:20,  1.21it/s]
2|865|Loss: 0.7735261917114258:  13%|█▎        | 865/6470 [10:46<1:14:06,  1.26it/s]
2|866|Loss: 0.82264244556427:  13%|█▎        | 865/6470 [10:47<1:14:06,  1.26it/s]  
2|866|Loss: 0.82264244556427:  13%|█▎        | 866/6470 [10:47<1:11:06,  1.31it/s]
2|867|Loss: 0.8152707815170288:  13%|█▎        | 866/6470 [10:47<1:11:06,  1.31it/s]
2|867|Loss: 0.8152707815170288:  13%|█▎        | 867/6470 [10:48<1:11:20,  1.31it/s]
2|868|Loss: 0.8603649735450745:  13%|█▎        | 867/6470 [10:48<1:11:20,  1.31it/s]
2|868|Loss: 0.8603649735450745:  13%|█▎        | 868/6470 [10:48<1:12:06,  1.29it/s]
2|869|Loss: 0.6769700646400452:  13%|█▎        | 868/6470 [10:49<1:12:06,  1.29it/s]
2|869|Loss: 0.6769700646400452:  13%|█▎        | 869/6470 [10:49<1:07:19,  1.39it/s]
2|870|Loss: 0.9112164974212646:  13%|█▎        | 869/6470 [10:49<1:07:19,  1.39it/s]
2|870|Loss: 0.9112164974212646:  13%|█▎        | 870/6470 [10:50<1:03:24,  1.47it/s]
2|871|Loss: 0.8077487349510193:  13%|█▎        | 870/6470 [10:50<1:03:24,  1.47it/s]
2|871|Loss: 0.8077487349510193:  13%|█▎        | 871/6470 [10:50<1:03:22,  1.47it/s]
2|872|Loss: 0.7793174386024475:  13%|█▎        | 871/6470 [10:51<1:03:22,  1.47it/s]
2|872|Loss: 0.7793174386024475:  13%|█▎        | 872/6470 [10:51<1:03:29,  1.47it/s]
2|873|Loss: 0.8735868334770203:  13%|█▎        | 872/6470 [10:51<1:03:29,  1.47it/s]
2|873|Loss: 0.8735868334770203:  13%|█▎        | 873/6470 [10:52<1:03:04,  1.48it/s]
2|874|Loss: 0.8018091917037964:  13%|█▎        | 873/6470 [10:52<1:03:04,  1.48it/s]
2|874|Loss: 0.8018091917037964:  14%|█▎        | 874/6470 [10:52<1:06:23,  1.40it/s]
2|875|Loss: 0.8448060750961304:  14%|█▎        | 874/6470 [10:53<1:06:23,  1.40it/s]
2|875|Loss: 0.8448060750961304:  14%|█▎        | 875/6470 [10:53<1:03:29,  1.47it/s]
2|876|Loss: 0.9449660181999207:  14%|█▎        | 875/6470 [10:53<1:03:29,  1.47it/s]
2|876|Loss: 0.9449660181999207:  14%|█▎        | 876/6470 [10:54<1:00:32,  1.54it/s]
2|877|Loss: 0.5938726663589478:  14%|█▎        | 876/6470 [10:54<1:00:32,  1.54it/s]
2|877|Loss: 0.5938726663589478:  14%|█▎        | 877/6470 [10:54<58:18,  1.60it/s]  
2|878|Loss: 0.9388038516044617:  14%|█▎        | 877/6470 [10:55<58:18,  1.60it/s]
2|878|Loss: 0.9388038516044617:  14%|█▎        | 878/6470 [10:55<1:03:40,  1.46it/s]
2|879|Loss: 0.8730335831642151:  14%|█▎        | 878/6470 [10:55<1:03:40,  1.46it/s]
2|879|Loss: 0.8730335831642151:  14%|█▎        | 879/6470 [10:56<1:07:07,  1.39it/s]
2|880|Loss: 0.7752233743667603:  14%|█▎        | 879/6470 [10:56<1:07:07,  1.39it/s]
2|880|Loss: 0.7752233743667603:  14%|█▎        | 880/6470 [10:56<1:00:32,  1.54it/s]
2|881|Loss: 0.824674129486084:  14%|█▎        | 880/6470 [10:57<1:00:32,  1.54it/s] 
2|881|Loss: 0.824674129486084:  14%|█▎        | 881/6470 [10:57<1:02:22,  1.49it/s]
2|882|Loss: 0.8437188863754272:  14%|█▎        | 881/6470 [10:57<1:02:22,  1.49it/s]
2|882|Loss: 0.8437188863754272:  14%|█▎        | 882/6470 [10:58<1:06:36,  1.40it/s]
2|883|Loss: 0.7291528582572937:  14%|█▎        | 882/6470 [10:58<1:06:36,  1.40it/s]
2|883|Loss: 0.7291528582572937:  14%|█▎        | 883/6470 [10:59<1:07:09,  1.39it/s]
2|884|Loss: 0.6809251308441162:  14%|█▎        | 883/6470 [10:59<1:07:09,  1.39it/s]
2|884|Loss: 0.6809251308441162:  14%|█▎        | 884/6470 [10:59<1:05:59,  1.41it/s]
2|885|Loss: 0.7135713696479797:  14%|█▎        | 884/6470 [11:00<1:05:59,  1.41it/s]
2|885|Loss: 0.7135713696479797:  14%|█▎        | 885/6470 [11:00<1:07:52,  1.37it/s]
2|886|Loss: 0.8268390893936157:  14%|█▎        | 885/6470 [11:01<1:07:52,  1.37it/s]
2|886|Loss: 0.8268390893936157:  14%|█▎        | 886/6470 [11:01<1:11:31,  1.30it/s]
2|887|Loss: 0.7721601724624634:  14%|█▎        | 886/6470 [11:01<1:11:31,  1.30it/s]
2|887|Loss: 0.7721601724624634:  14%|█▎        | 887/6470 [11:02<1:10:43,  1.32it/s]
2|888|Loss: 0.8995344638824463:  14%|█▎        | 887/6470 [11:02<1:10:43,  1.32it/s]
2|888|Loss: 0.8995344638824463:  14%|█▎        | 888/6470 [11:02<1:11:38,  1.30it/s]
2|889|Loss: 0.8693814277648926:  14%|█▎        | 888/6470 [11:03<1:11:38,  1.30it/s]
2|889|Loss: 0.8693814277648926:  14%|█▎        | 889/6470 [11:03<1:06:43,  1.39it/s]
2|890|Loss: 0.8262548446655273:  14%|█▎        | 889/6470 [11:03<1:06:43,  1.39it/s]
2|890|Loss: 0.8262548446655273:  14%|█▍        | 890/6470 [11:04<1:04:24,  1.44it/s]
2|891|Loss: 0.8565192222595215:  14%|█▍        | 890/6470 [11:04<1:04:24,  1.44it/s]
2|891|Loss: 0.8565192222595215:  14%|█▍        | 891/6470 [11:04<1:06:55,  1.39it/s]
2|892|Loss: 0.7971656918525696:  14%|█▍        | 891/6470 [11:05<1:06:55,  1.39it/s]
2|892|Loss: 0.7971656918525696:  14%|█▍        | 892/6470 [11:05<1:05:26,  1.42it/s]
2|893|Loss: 1.0731184482574463:  14%|█▍        | 892/6470 [11:05<1:05:26,  1.42it/s]
2|893|Loss: 1.0731184482574463:  14%|█▍        | 893/6470 [11:06<1:01:50,  1.50it/s]
2|894|Loss: 0.9072356820106506:  14%|█▍        | 893/6470 [11:06<1:01:50,  1.50it/s]
2|894|Loss: 0.9072356820106506:  14%|█▍        | 894/6470 [11:07<1:06:49,  1.39it/s]
2|895|Loss: 0.9375296831130981:  14%|█▍        | 894/6470 [11:07<1:06:49,  1.39it/s]
2|895|Loss: 0.9375296831130981:  14%|█▍        | 895/6470 [11:07<1:07:13,  1.38it/s]
2|896|Loss: 0.840959370136261:  14%|█▍        | 895/6470 [11:08<1:07:13,  1.38it/s] 
2|896|Loss: 0.840959370136261:  14%|█▍        | 896/6470 [11:08<1:08:54,  1.35it/s]
2|897|Loss: 0.812186598777771:  14%|█▍        | 896/6470 [11:08<1:08:54,  1.35it/s]
2|897|Loss: 0.812186598777771:  14%|█▍        | 897/6470 [11:09<1:07:43,  1.37it/s]
2|898|Loss: 0.8634291291236877:  14%|█▍        | 897/6470 [11:09<1:07:43,  1.37it/s]
2|898|Loss: 0.8634291291236877:  14%|█▍        | 898/6470 [11:10<1:15:51,  1.22it/s]
2|899|Loss: 0.8907781839370728:  14%|█▍        | 898/6470 [11:10<1:15:51,  1.22it/s]
2|899|Loss: 0.8907781839370728:  14%|█▍        | 899/6470 [11:11<1:17:18,  1.20it/s]
2|900|Loss: 0.9903517961502075:  14%|█▍        | 899/6470 [11:11<1:17:18,  1.20it/s]
2|900|Loss: 0.9903517961502075:  14%|█▍        | 900/6470 [11:11<1:13:44,  1.26it/s]
2|901|Loss: 0.9425605535507202:  14%|█▍        | 900/6470 [11:12<1:13:44,  1.26it/s]
2|901|Loss: 0.9425605535507202:  14%|█▍        | 901/6470 [11:12<1:15:17,  1.23it/s]
2|902|Loss: 0.7488133311271667:  14%|█▍        | 901/6470 [11:13<1:15:17,  1.23it/s]
2|902|Loss: 0.7488133311271667:  14%|█▍        | 902/6470 [11:13<1:18:52,  1.18it/s]
2|903|Loss: 0.8890870213508606:  14%|█▍        | 902/6470 [11:14<1:18:52,  1.18it/s]
2|903|Loss: 0.8890870213508606:  14%|█▍        | 903/6470 [11:14<1:16:22,  1.21it/s]
2|904|Loss: 0.7775253653526306:  14%|█▍        | 903/6470 [11:14<1:16:22,  1.21it/s]
2|904|Loss: 0.7775253653526306:  14%|█▍        | 904/6470 [11:14<1:09:26,  1.34it/s]
2|905|Loss: 0.9636258482933044:  14%|█▍        | 904/6470 [11:15<1:09:26,  1.34it/s]
2|905|Loss: 0.9636258482933044:  14%|█▍        | 905/6470 [11:15<1:05:01,  1.43it/s]
2|906|Loss: 0.903297483921051:  14%|█▍        | 905/6470 [11:15<1:05:01,  1.43it/s] 
2|906|Loss: 0.903297483921051:  14%|█▍        | 906/6470 [11:16<1:03:37,  1.46it/s]
2|907|Loss: 0.7789860963821411:  14%|█▍        | 906/6470 [11:16<1:03:37,  1.46it/s]
2|907|Loss: 0.7789860963821411:  14%|█▍        | 907/6470 [11:16<1:04:00,  1.45it/s]
2|908|Loss: 0.8687873482704163:  14%|█▍        | 907/6470 [11:17<1:04:00,  1.45it/s]
2|908|Loss: 0.8687873482704163:  14%|█▍        | 908/6470 [11:17<1:04:43,  1.43it/s]
2|909|Loss: 0.9101058840751648:  14%|█▍        | 908/6470 [11:18<1:04:43,  1.43it/s]
2|909|Loss: 0.9101058840751648:  14%|█▍        | 909/6470 [11:18<1:06:11,  1.40it/s]
2|910|Loss: 0.8634434938430786:  14%|█▍        | 909/6470 [11:18<1:06:11,  1.40it/s]
2|910|Loss: 0.8634434938430786:  14%|█▍        | 910/6470 [11:19<1:08:50,  1.35it/s]
2|911|Loss: 0.7289263010025024:  14%|█▍        | 910/6470 [11:19<1:08:50,  1.35it/s]
2|911|Loss: 0.7289263010025024:  14%|█▍        | 911/6470 [11:19<1:08:39,  1.35it/s]
2|912|Loss: 0.6778115630149841:  14%|█▍        | 911/6470 [11:20<1:08:39,  1.35it/s]
2|912|Loss: 0.6778115630149841:  14%|█▍        | 912/6470 [11:20<1:10:05,  1.32it/s]
2|913|Loss: 0.9798537492752075:  14%|█▍        | 912/6470 [11:21<1:10:05,  1.32it/s]
2|913|Loss: 0.9798537492752075:  14%|█▍        | 913/6470 [11:21<1:08:36,  1.35it/s]
2|914|Loss: 0.8236609697341919:  14%|█▍        | 913/6470 [11:21<1:08:36,  1.35it/s]
2|914|Loss: 0.8236609697341919:  14%|█▍        | 914/6470 [11:22<1:04:10,  1.44it/s]
2|915|Loss: 0.8191514015197754:  14%|█▍        | 914/6470 [11:22<1:04:10,  1.44it/s]
2|915|Loss: 0.8191514015197754:  14%|█▍        | 915/6470 [11:22<1:02:20,  1.49it/s]
2|916|Loss: 0.8557440638542175:  14%|█▍        | 915/6470 [11:23<1:02:20,  1.49it/s]
2|916|Loss: 0.8557440638542175:  14%|█▍        | 916/6470 [11:23<1:04:10,  1.44it/s]
2|917|Loss: 0.9360376000404358:  14%|█▍        | 916/6470 [11:23<1:04:10,  1.44it/s]
2|917|Loss: 0.9360376000404358:  14%|█▍        | 917/6470 [11:24<1:03:59,  1.45it/s]
2|918|Loss: 0.8208523392677307:  14%|█▍        | 917/6470 [11:24<1:03:59,  1.45it/s]
2|918|Loss: 0.8208523392677307:  14%|█▍        | 918/6470 [11:24<1:05:22,  1.42it/s]
2|919|Loss: 0.8556462526321411:  14%|█▍        | 918/6470 [11:25<1:05:22,  1.42it/s]
2|919|Loss: 0.8556462526321411:  14%|█▍        | 919/6470 [11:25<1:10:44,  1.31it/s]
2|920|Loss: 0.791511058807373:  14%|█▍        | 919/6470 [11:26<1:10:44,  1.31it/s] 
2|920|Loss: 0.791511058807373:  14%|█▍        | 920/6470 [11:26<1:12:52,  1.27it/s]
2|921|Loss: 0.8096602559089661:  14%|█▍        | 920/6470 [11:26<1:12:52,  1.27it/s]
2|921|Loss: 0.8096602559089661:  14%|█▍        | 921/6470 [11:27<1:12:47,  1.27it/s]
2|922|Loss: 0.774454653263092:  14%|█▍        | 921/6470 [11:27<1:12:47,  1.27it/s] 
2|922|Loss: 0.774454653263092:  14%|█▍        | 922/6470 [11:27<1:07:32,  1.37it/s]
2|923|Loss: 0.8353831768035889:  14%|█▍        | 922/6470 [11:28<1:07:32,  1.37it/s]
2|923|Loss: 0.8353831768035889:  14%|█▍        | 923/6470 [11:28<1:06:03,  1.40it/s]
2|924|Loss: 0.7777559161186218:  14%|█▍        | 923/6470 [11:29<1:06:03,  1.40it/s]
2|924|Loss: 0.7777559161186218:  14%|█▍        | 924/6470 [11:29<1:06:03,  1.40it/s]
2|925|Loss: 0.8137621879577637:  14%|█▍        | 924/6470 [11:29<1:06:03,  1.40it/s]
2|925|Loss: 0.8137621879577637:  14%|█▍        | 925/6470 [11:30<1:05:16,  1.42it/s]
2|926|Loss: 0.9579537510871887:  14%|█▍        | 925/6470 [11:30<1:05:16,  1.42it/s]
2|926|Loss: 0.9579537510871887:  14%|█▍        | 926/6470 [11:30<1:09:15,  1.33it/s]
2|927|Loss: 0.9045202732086182:  14%|█▍        | 926/6470 [11:31<1:09:15,  1.33it/s]
2|927|Loss: 0.9045202732086182:  14%|█▍        | 927/6470 [11:31<1:11:10,  1.30it/s]
2|928|Loss: 0.9334860444068909:  14%|█▍        | 927/6470 [11:32<1:11:10,  1.30it/s]
2|928|Loss: 0.9334860444068909:  14%|█▍        | 928/6470 [11:32<1:14:31,  1.24it/s]
2|929|Loss: 0.7990339994430542:  14%|█▍        | 928/6470 [11:33<1:14:31,  1.24it/s]
2|929|Loss: 0.7990339994430542:  14%|█▍        | 929/6470 [11:33<1:12:47,  1.27it/s]
2|930|Loss: 0.8202813267707825:  14%|█▍        | 929/6470 [11:33<1:12:47,  1.27it/s]
2|930|Loss: 0.8202813267707825:  14%|█▍        | 930/6470 [11:34<1:12:10,  1.28it/s]
2|931|Loss: 1.016743540763855:  14%|█▍        | 930/6470 [11:34<1:12:10,  1.28it/s] 
2|931|Loss: 1.016743540763855:  14%|█▍        | 931/6470 [11:34<1:07:10,  1.37it/s]
2|932|Loss: 0.7611097693443298:  14%|█▍        | 931/6470 [11:35<1:07:10,  1.37it/s]
2|932|Loss: 0.7611097693443298:  14%|█▍        | 932/6470 [11:35<1:05:28,  1.41it/s]
2|933|Loss: 0.9007441997528076:  14%|█▍        | 932/6470 [11:35<1:05:28,  1.41it/s]
2|933|Loss: 0.9007441997528076:  14%|█▍        | 933/6470 [11:36<1:04:51,  1.42it/s]
2|934|Loss: 0.8363077044487:  14%|█▍        | 933/6470 [11:36<1:04:51,  1.42it/s]   
2|934|Loss: 0.8363077044487:  14%|█▍        | 934/6470 [11:36<1:04:09,  1.44it/s]
2|935|Loss: 0.9482160806655884:  14%|█▍        | 934/6470 [11:37<1:04:09,  1.44it/s]
2|935|Loss: 0.9482160806655884:  14%|█▍        | 935/6470 [11:37<1:01:41,  1.50it/s]
2|936|Loss: 0.7950685024261475:  14%|█▍        | 935/6470 [11:37<1:01:41,  1.50it/s]
2|936|Loss: 0.7950685024261475:  14%|█▍        | 936/6470 [11:37<59:27,  1.55it/s]  
2|937|Loss: 0.9697489142417908:  14%|█▍        | 936/6470 [11:38<59:27,  1.55it/s]
2|937|Loss: 0.9697489142417908:  14%|█▍        | 937/6470 [11:38<1:02:56,  1.47it/s]
2|938|Loss: 0.7814448475837708:  14%|█▍        | 937/6470 [11:39<1:02:56,  1.47it/s]
2|938|Loss: 0.7814448475837708:  14%|█▍        | 938/6470 [11:39<1:08:20,  1.35it/s]
2|939|Loss: 0.9497565031051636:  14%|█▍        | 938/6470 [11:40<1:08:20,  1.35it/s]
2|939|Loss: 0.9497565031051636:  15%|█▍        | 939/6470 [11:40<1:12:20,  1.27it/s]
2|940|Loss: 0.8372634053230286:  15%|█▍        | 939/6470 [11:40<1:12:20,  1.27it/s]
2|940|Loss: 0.8372634053230286:  15%|█▍        | 940/6470 [11:41<1:11:06,  1.30it/s]
2|941|Loss: 0.7755232453346252:  15%|█▍        | 940/6470 [11:41<1:11:06,  1.30it/s]
2|941|Loss: 0.7755232453346252:  15%|█▍        | 941/6470 [11:41<1:11:47,  1.28it/s]
2|942|Loss: 0.8168445229530334:  15%|█▍        | 941/6470 [11:42<1:11:47,  1.28it/s]
2|942|Loss: 0.8168445229530334:  15%|█▍        | 942/6470 [11:42<1:12:24,  1.27it/s]
2|943|Loss: 0.9378647804260254:  15%|█▍        | 942/6470 [11:43<1:12:24,  1.27it/s]
2|943|Loss: 0.9378647804260254:  15%|█▍        | 943/6470 [11:43<1:10:00,  1.32it/s]
2|944|Loss: 0.8995636105537415:  15%|█▍        | 943/6470 [11:43<1:10:00,  1.32it/s]
2|944|Loss: 0.8995636105537415:  15%|█▍        | 944/6470 [11:44<1:08:31,  1.34it/s]
2|945|Loss: 1.0383124351501465:  15%|█▍        | 944/6470 [11:44<1:08:31,  1.34it/s]
2|945|Loss: 1.0383124351501465:  15%|█▍        | 945/6470 [11:45<1:11:18,  1.29it/s]
2|946|Loss: 0.870475172996521:  15%|█▍        | 945/6470 [11:45<1:11:18,  1.29it/s] 
2|946|Loss: 0.870475172996521:  15%|█▍        | 946/6470 [11:45<1:11:43,  1.28it/s]
2|947|Loss: 0.8209118247032166:  15%|█▍        | 946/6470 [11:46<1:11:43,  1.28it/s]
2|947|Loss: 0.8209118247032166:  15%|█▍        | 947/6470 [11:46<1:06:24,  1.39it/s]
2|948|Loss: 0.8758495450019836:  15%|█▍        | 947/6470 [11:46<1:06:24,  1.39it/s]
2|948|Loss: 0.8758495450019836:  15%|█▍        | 948/6470 [11:47<1:06:15,  1.39it/s]
2|949|Loss: 0.8655853867530823:  15%|█▍        | 948/6470 [11:47<1:06:15,  1.39it/s]
2|949|Loss: 0.8655853867530823:  15%|█▍        | 949/6470 [11:47<1:04:33,  1.43it/s]
2|950|Loss: 0.9628210067749023:  15%|█▍        | 949/6470 [11:48<1:04:33,  1.43it/s]
2|950|Loss: 0.9628210067749023:  15%|█▍        | 950/6470 [11:48<1:09:15,  1.33it/s]
2|951|Loss: 0.9360179901123047:  15%|█▍        | 950/6470 [11:49<1:09:15,  1.33it/s]
2|951|Loss: 0.9360179901123047:  15%|█▍        | 951/6470 [11:49<1:07:39,  1.36it/s]
2|952|Loss: 1.0516828298568726:  15%|█▍        | 951/6470 [11:49<1:07:39,  1.36it/s]
2|952|Loss: 1.0516828298568726:  15%|█▍        | 952/6470 [11:50<1:08:31,  1.34it/s]
2|953|Loss: 1.1502948999404907:  15%|█▍        | 952/6470 [11:50<1:08:31,  1.34it/s]
2|953|Loss: 1.1502948999404907:  15%|█▍        | 953/6470 [11:50<1:10:06,  1.31it/s]
2|954|Loss: 0.7910721898078918:  15%|█▍        | 953/6470 [11:51<1:10:06,  1.31it/s]
2|954|Loss: 0.7910721898078918:  15%|█▍        | 954/6470 [11:51<1:11:17,  1.29it/s]
2|955|Loss: 0.8450947403907776:  15%|█▍        | 954/6470 [11:52<1:11:17,  1.29it/s]
2|955|Loss: 0.8450947403907776:  15%|█▍        | 955/6470 [11:52<1:06:57,  1.37it/s]
2|956|Loss: 0.877547025680542:  15%|█▍        | 955/6470 [11:52<1:06:57,  1.37it/s] 
2|956|Loss: 0.877547025680542:  15%|█▍        | 956/6470 [11:53<1:07:35,  1.36it/s]
2|957|Loss: 0.8578193187713623:  15%|█▍        | 956/6470 [11:53<1:07:35,  1.36it/s]
2|957|Loss: 0.8578193187713623:  15%|█▍        | 957/6470 [11:53<1:07:13,  1.37it/s]
2|958|Loss: 0.8164935111999512:  15%|█▍        | 957/6470 [11:54<1:07:13,  1.37it/s]
2|958|Loss: 0.8164935111999512:  15%|█▍        | 958/6470 [11:54<59:20,  1.55it/s]  
2|959|Loss: 0.8059973120689392:  15%|█▍        | 958/6470 [11:54<59:20,  1.55it/s]
2|959|Loss: 0.8059973120689392:  15%|█▍        | 959/6470 [11:54<57:11,  1.61it/s]
2|960|Loss: 0.8615933060646057:  15%|█▍        | 959/6470 [11:55<57:11,  1.61it/s]
2|960|Loss: 0.8615933060646057:  15%|█▍        | 960/6470 [11:55<53:49,  1.71it/s]
2|961|Loss: 0.8642461895942688:  15%|█▍        | 960/6470 [11:55<53:49,  1.71it/s]
2|961|Loss: 0.8642461895942688:  15%|█▍        | 961/6470 [11:56<59:25,  1.55it/s]
2|962|Loss: 0.7513605952262878:  15%|█▍        | 961/6470 [11:56<59:25,  1.55it/s]
2|962|Loss: 0.7513605952262878:  15%|█▍        | 962/6470 [11:56<1:04:11,  1.43it/s]
2|963|Loss: 0.8451878428459167:  15%|█▍        | 962/6470 [11:57<1:04:11,  1.43it/s]
2|963|Loss: 0.8451878428459167:  15%|█▍        | 963/6470 [11:57<1:04:38,  1.42it/s]
2|964|Loss: 0.8296629190444946:  15%|█▍        | 963/6470 [11:58<1:04:38,  1.42it/s]
2|964|Loss: 0.8296629190444946:  15%|█▍        | 964/6470 [11:58<1:00:56,  1.51it/s]
2|965|Loss: 0.7408601641654968:  15%|█▍        | 964/6470 [11:58<1:00:56,  1.51it/s]
2|965|Loss: 0.7408601641654968:  15%|█▍        | 965/6470 [11:59<1:05:16,  1.41it/s]
2|966|Loss: 0.7241127490997314:  15%|█▍        | 965/6470 [11:59<1:05:16,  1.41it/s]
2|966|Loss: 0.7241127490997314:  15%|█▍        | 966/6470 [11:59<58:52,  1.56it/s]  
2|967|Loss: 0.6670979857444763:  15%|█▍        | 966/6470 [11:59<58:52,  1.56it/s]
2|967|Loss: 0.6670979857444763:  15%|█▍        | 967/6470 [12:00<1:00:24,  1.52it/s]
2|968|Loss: 0.7284415364265442:  15%|█▍        | 967/6470 [12:00<1:00:24,  1.52it/s]
2|968|Loss: 0.7284415364265442:  15%|█▍        | 968/6470 [12:01<1:04:22,  1.42it/s]
2|969|Loss: 0.7901896834373474:  15%|█▍        | 968/6470 [12:01<1:04:22,  1.42it/s]
2|969|Loss: 0.7901896834373474:  15%|█▍        | 969/6470 [12:01<1:03:25,  1.45it/s]
2|970|Loss: 0.8460683822631836:  15%|█▍        | 969/6470 [12:02<1:03:25,  1.45it/s]
2|970|Loss: 0.8460683822631836:  15%|█▍        | 970/6470 [12:02<1:03:16,  1.45it/s]
2|971|Loss: 0.8754967451095581:  15%|█▍        | 970/6470 [12:02<1:03:16,  1.45it/s]
2|971|Loss: 0.8754967451095581:  15%|█▌        | 971/6470 [12:02<1:00:49,  1.51it/s]
2|972|Loss: 1.0291367769241333:  15%|█▌        | 971/6470 [12:03<1:00:49,  1.51it/s]
2|972|Loss: 1.0291367769241333:  15%|█▌        | 972/6470 [12:03<1:02:29,  1.47it/s]
2|973|Loss: 0.8366851210594177:  15%|█▌        | 972/6470 [12:04<1:02:29,  1.47it/s]
2|973|Loss: 0.8366851210594177:  15%|█▌        | 973/6470 [12:04<1:01:45,  1.48it/s]
2|974|Loss: 0.8111271858215332:  15%|█▌        | 973/6470 [12:04<1:01:45,  1.48it/s]
2|974|Loss: 0.8111271858215332:  15%|█▌        | 974/6470 [12:04<58:38,  1.56it/s]  
2|975|Loss: 0.707148551940918:  15%|█▌        | 974/6470 [12:05<58:38,  1.56it/s] 
2|975|Loss: 0.707148551940918:  15%|█▌        | 975/6470 [12:05<54:47,  1.67it/s]
2|976|Loss: 0.7913375496864319:  15%|█▌        | 975/6470 [12:05<54:47,  1.67it/s]
2|976|Loss: 0.7913375496864319:  15%|█▌        | 976/6470 [12:06<1:00:57,  1.50it/s]
2|977|Loss: 0.9116883873939514:  15%|█▌        | 976/6470 [12:06<1:00:57,  1.50it/s]
2|977|Loss: 0.9116883873939514:  15%|█▌        | 977/6470 [12:07<1:09:51,  1.31it/s]
2|978|Loss: 0.8986810445785522:  15%|█▌        | 977/6470 [12:07<1:09:51,  1.31it/s]
2|978|Loss: 0.8986810445785522:  15%|█▌        | 978/6470 [12:08<1:14:41,  1.23it/s]
2|979|Loss: 0.8916842937469482:  15%|█▌        | 978/6470 [12:08<1:14:41,  1.23it/s]
2|979|Loss: 0.8916842937469482:  15%|█▌        | 979/6470 [12:08<1:12:46,  1.26it/s]
2|980|Loss: 0.9112993478775024:  15%|█▌        | 979/6470 [12:09<1:12:46,  1.26it/s]
2|980|Loss: 0.9112993478775024:  15%|█▌        | 980/6470 [12:09<1:12:36,  1.26it/s]
2|981|Loss: 0.8765940070152283:  15%|█▌        | 980/6470 [12:10<1:12:36,  1.26it/s]
2|981|Loss: 0.8765940070152283:  15%|█▌        | 981/6470 [12:10<1:14:27,  1.23it/s]
2|982|Loss: 0.8284348845481873:  15%|█▌        | 981/6470 [12:11<1:14:27,  1.23it/s]
2|982|Loss: 0.8284348845481873:  15%|█▌        | 982/6470 [12:11<1:18:14,  1.17it/s]
2|983|Loss: 0.8713841438293457:  15%|█▌        | 982/6470 [12:12<1:18:14,  1.17it/s]
2|983|Loss: 0.8713841438293457:  15%|█▌        | 983/6470 [12:12<1:17:06,  1.19it/s]
2|984|Loss: 0.7744583487510681:  15%|█▌        | 983/6470 [12:12<1:17:06,  1.19it/s]
2|984|Loss: 0.7744583487510681:  15%|█▌        | 984/6470 [12:13<1:13:26,  1.24it/s]
2|985|Loss: 0.7701553702354431:  15%|█▌        | 984/6470 [12:13<1:13:26,  1.24it/s]
2|985|Loss: 0.7701553702354431:  15%|█▌        | 985/6470 [12:13<1:07:03,  1.36it/s]
2|986|Loss: 0.8262264728546143:  15%|█▌        | 985/6470 [12:14<1:07:03,  1.36it/s]
2|986|Loss: 0.8262264728546143:  15%|█▌        | 986/6470 [12:14<1:08:54,  1.33it/s]
2|987|Loss: 0.8721092343330383:  15%|█▌        | 986/6470 [12:14<1:08:54,  1.33it/s]
2|987|Loss: 0.8721092343330383:  15%|█▌        | 987/6470 [12:15<1:10:26,  1.30it/s]
2|988|Loss: 0.8512917160987854:  15%|█▌        | 987/6470 [12:15<1:10:26,  1.30it/s]
2|988|Loss: 0.8512917160987854:  15%|█▌        | 988/6470 [12:16<1:11:00,  1.29it/s]
2|989|Loss: 0.8877724409103394:  15%|█▌        | 988/6470 [12:16<1:11:00,  1.29it/s]
2|989|Loss: 0.8877724409103394:  15%|█▌        | 989/6470 [12:16<1:14:22,  1.23it/s]
2|990|Loss: 0.9695133566856384:  15%|█▌        | 989/6470 [12:17<1:14:22,  1.23it/s]
2|990|Loss: 0.9695133566856384:  15%|█▌        | 990/6470 [12:17<1:12:52,  1.25it/s]
2|991|Loss: 0.7965843081474304:  15%|█▌        | 990/6470 [12:18<1:12:52,  1.25it/s]
2|991|Loss: 0.7965843081474304:  15%|█▌        | 991/6470 [12:18<1:14:51,  1.22it/s]
2|992|Loss: 1.0396087169647217:  15%|█▌        | 991/6470 [12:19<1:14:51,  1.22it/s]
2|992|Loss: 1.0396087169647217:  15%|█▌        | 992/6470 [12:19<1:16:02,  1.20it/s]
2|993|Loss: 0.7954558730125427:  15%|█▌        | 992/6470 [12:19<1:16:02,  1.20it/s]
2|993|Loss: 0.7954558730125427:  15%|█▌        | 993/6470 [12:20<1:14:31,  1.22it/s]
2|994|Loss: 0.9432176351547241:  15%|█▌        | 993/6470 [12:20<1:14:31,  1.22it/s]
2|994|Loss: 0.9432176351547241:  15%|█▌        | 994/6470 [12:20<1:13:13,  1.25it/s]
2|995|Loss: 0.7808330059051514:  15%|█▌        | 994/6470 [12:21<1:13:13,  1.25it/s]
2|995|Loss: 0.7808330059051514:  15%|█▌        | 995/6470 [12:21<1:10:55,  1.29it/s]
2|996|Loss: 0.7857765555381775:  15%|█▌        | 995/6470 [12:22<1:10:55,  1.29it/s]
2|996|Loss: 0.7857765555381775:  15%|█▌        | 996/6470 [12:22<1:14:04,  1.23it/s]
2|997|Loss: 0.8658574819564819:  15%|█▌        | 996/6470 [12:23<1:14:04,  1.23it/s]
2|997|Loss: 0.8658574819564819:  15%|█▌        | 997/6470 [12:23<1:14:51,  1.22it/s]
2|998|Loss: 0.869482159614563:  15%|█▌        | 997/6470 [12:23<1:14:51,  1.22it/s] 
2|998|Loss: 0.869482159614563:  15%|█▌        | 998/6470 [12:24<1:12:29,  1.26it/s]
2|999|Loss: 0.9900736212730408:  15%|█▌        | 998/6470 [12:24<1:12:29,  1.26it/s]
2|999|Loss: 0.9900736212730408:  15%|█▌        | 999/6470 [12:24<1:07:54,  1.34it/s]
2|1000|Loss: 0.917302131652832:  15%|█▌        | 999/6470 [12:25<1:07:54,  1.34it/s]
2|1000|Loss: 0.917302131652832:  15%|█▌        | 1000/6470 [12:25<1:06:32,  1.37it/s]
2|1001|Loss: 0.6789922714233398:  15%|█▌        | 1000/6470 [12:25<1:06:32,  1.37it/s]
2|1001|Loss: 0.6789922714233398:  15%|█▌        | 1001/6470 [12:26<1:02:18,  1.46it/s]
2|1002|Loss: 0.8889457583427429:  15%|█▌        | 1001/6470 [12:26<1:02:18,  1.46it/s]
2|1002|Loss: 0.8889457583427429:  15%|█▌        | 1002/6470 [12:26<1:05:59,  1.38it/s]
2|1003|Loss: 0.7671635150909424:  15%|█▌        | 1002/6470 [12:27<1:05:59,  1.38it/s]
2|1003|Loss: 0.7671635150909424:  16%|█▌        | 1003/6470 [12:27<1:11:41,  1.27it/s]
2|1004|Loss: 0.7946855425834656:  16%|█▌        | 1003/6470 [12:28<1:11:41,  1.27it/s]
2|1004|Loss: 0.7946855425834656:  16%|█▌        | 1004/6470 [12:28<1:11:01,  1.28it/s]
2|1005|Loss: 0.8234887719154358:  16%|█▌        | 1004/6470 [12:28<1:11:01,  1.28it/s]
2|1005|Loss: 0.8234887719154358:  16%|█▌        | 1005/6470 [12:29<1:07:26,  1.35it/s]
2|1006|Loss: 0.8294726014137268:  16%|█▌        | 1005/6470 [12:29<1:07:26,  1.35it/s]
2|1006|Loss: 0.8294726014137268:  16%|█▌        | 1006/6470 [12:30<1:08:23,  1.33it/s]
2|1007|Loss: 1.000478744506836:  16%|█▌        | 1006/6470 [12:30<1:08:23,  1.33it/s] 
2|1007|Loss: 1.000478744506836:  16%|█▌        | 1007/6470 [12:30<1:09:14,  1.31it/s]
2|1008|Loss: 0.7160468697547913:  16%|█▌        | 1007/6470 [12:31<1:09:14,  1.31it/s]
2|1008|Loss: 0.7160468697547913:  16%|█▌        | 1008/6470 [12:31<1:15:50,  1.20it/s]
2|1009|Loss: 0.8093326091766357:  16%|█▌        | 1008/6470 [12:32<1:15:50,  1.20it/s]
2|1009|Loss: 0.8093326091766357:  16%|█▌        | 1009/6470 [12:32<1:15:33,  1.20it/s]
2|1010|Loss: 0.9171653389930725:  16%|█▌        | 1009/6470 [12:33<1:15:33,  1.20it/s]
2|1010|Loss: 0.9171653389930725:  16%|█▌        | 1010/6470 [12:33<1:13:30,  1.24it/s]
2|1011|Loss: 0.8099378347396851:  16%|█▌        | 1010/6470 [12:33<1:13:30,  1.24it/s]
2|1011|Loss: 0.8099378347396851:  16%|█▌        | 1011/6470 [12:34<1:15:23,  1.21it/s]
2|1012|Loss: 0.7383486032485962:  16%|█▌        | 1011/6470 [12:34<1:15:23,  1.21it/s]
2|1012|Loss: 0.7383486032485962:  16%|█▌        | 1012/6470 [12:35<1:15:17,  1.21it/s]
2|1013|Loss: 0.8820802569389343:  16%|█▌        | 1012/6470 [12:35<1:15:17,  1.21it/s]
2|1013|Loss: 0.8820802569389343:  16%|█▌        | 1013/6470 [12:35<1:14:05,  1.23it/s]
2|1014|Loss: 0.869408130645752:  16%|█▌        | 1013/6470 [12:36<1:14:05,  1.23it/s] 
2|1014|Loss: 0.869408130645752:  16%|█▌        | 1014/6470 [12:36<1:11:21,  1.27it/s]
2|1015|Loss: 0.8361794948577881:  16%|█▌        | 1014/6470 [12:37<1:11:21,  1.27it/s]
2|1015|Loss: 0.8361794948577881:  16%|█▌        | 1015/6470 [12:37<1:13:36,  1.24it/s]
2|1016|Loss: 0.6918116211891174:  16%|█▌        | 1015/6470 [12:37<1:13:36,  1.24it/s]
2|1016|Loss: 0.6918116211891174:  16%|█▌        | 1016/6470 [12:38<1:11:37,  1.27it/s]
2|1017|Loss: 0.8271970152854919:  16%|█▌        | 1016/6470 [12:38<1:11:37,  1.27it/s]
2|1017|Loss: 0.8271970152854919:  16%|█▌        | 1017/6470 [12:38<1:09:44,  1.30it/s]
2|1018|Loss: 0.8737055659294128:  16%|█▌        | 1017/6470 [12:39<1:09:44,  1.30it/s]
2|1018|Loss: 0.8737055659294128:  16%|█▌        | 1018/6470 [12:39<1:12:59,  1.24it/s]
2|1019|Loss: 0.7700421810150146:  16%|█▌        | 1018/6470 [12:40<1:12:59,  1.24it/s]
2|1019|Loss: 0.7700421810150146:  16%|█▌        | 1019/6470 [12:40<1:11:02,  1.28it/s]
2|1020|Loss: 0.7162148952484131:  16%|█▌        | 1019/6470 [12:41<1:11:02,  1.28it/s]
2|1020|Loss: 0.7162148952484131:  16%|█▌        | 1020/6470 [12:41<1:20:59,  1.12it/s]
2|1021|Loss: 0.8275715112686157:  16%|█▌        | 1020/6470 [12:42<1:20:59,  1.12it/s]
2|1021|Loss: 0.8275715112686157:  16%|█▌        | 1021/6470 [12:42<1:22:05,  1.11it/s]
2|1022|Loss: 0.7929794788360596:  16%|█▌        | 1021/6470 [12:42<1:22:05,  1.11it/s]
2|1022|Loss: 0.7929794788360596:  16%|█▌        | 1022/6470 [12:43<1:14:32,  1.22it/s]
2|1023|Loss: 0.8203675746917725:  16%|█▌        | 1022/6470 [12:43<1:14:32,  1.22it/s]
2|1023|Loss: 0.8203675746917725:  16%|█▌        | 1023/6470 [12:43<1:08:57,  1.32it/s]
2|1024|Loss: 1.0036004781723022:  16%|█▌        | 1023/6470 [12:44<1:08:57,  1.32it/s]
2|1024|Loss: 1.0036004781723022:  16%|█▌        | 1024/6470 [12:44<1:11:11,  1.28it/s]
2|1025|Loss: 1.1021244525909424:  16%|█▌        | 1024/6470 [12:45<1:11:11,  1.28it/s]
2|1025|Loss: 1.1021244525909424:  16%|█▌        | 1025/6470 [12:45<1:11:50,  1.26it/s]
2|1026|Loss: 0.8631958365440369:  16%|█▌        | 1025/6470 [12:45<1:11:50,  1.26it/s]
2|1026|Loss: 0.8631958365440369:  16%|█▌        | 1026/6470 [12:46<1:11:05,  1.28it/s]
2|1027|Loss: 0.6994048953056335:  16%|█▌        | 1026/6470 [12:46<1:11:05,  1.28it/s]
2|1027|Loss: 0.6994048953056335:  16%|█▌        | 1027/6470 [12:47<1:12:58,  1.24it/s]
2|1028|Loss: 0.8187738656997681:  16%|█▌        | 1027/6470 [12:47<1:12:58,  1.24it/s]
2|1028|Loss: 0.8187738656997681:  16%|█▌        | 1028/6470 [12:47<1:12:21,  1.25it/s]
2|1029|Loss: 0.7141618728637695:  16%|█▌        | 1028/6470 [12:48<1:12:21,  1.25it/s]
2|1029|Loss: 0.7141618728637695:  16%|█▌        | 1029/6470 [12:48<1:09:25,  1.31it/s]
2|1030|Loss: 0.8374761343002319:  16%|█▌        | 1029/6470 [12:49<1:09:25,  1.31it/s]
2|1030|Loss: 0.8374761343002319:  16%|█▌        | 1030/6470 [12:49<1:11:56,  1.26it/s]
2|1031|Loss: 0.7675797343254089:  16%|█▌        | 1030/6470 [12:49<1:11:56,  1.26it/s]
2|1031|Loss: 0.7675797343254089:  16%|█▌        | 1031/6470 [12:50<1:12:58,  1.24it/s]
2|1032|Loss: 0.8055933117866516:  16%|█▌        | 1031/6470 [12:50<1:12:58,  1.24it/s]
2|1032|Loss: 0.8055933117866516:  16%|█▌        | 1032/6470 [12:50<1:10:31,  1.29it/s]
2|1033|Loss: 0.7693501114845276:  16%|█▌        | 1032/6470 [12:51<1:10:31,  1.29it/s]
2|1033|Loss: 0.7693501114845276:  16%|█▌        | 1033/6470 [12:51<1:09:43,  1.30it/s]
2|1034|Loss: 0.9738829731941223:  16%|█▌        | 1033/6470 [12:52<1:09:43,  1.30it/s]
2|1034|Loss: 0.9738829731941223:  16%|█▌        | 1034/6470 [12:52<1:09:36,  1.30it/s]
2|1035|Loss: 0.8489235043525696:  16%|█▌        | 1034/6470 [12:52<1:09:36,  1.30it/s]
2|1035|Loss: 0.8489235043525696:  16%|█▌        | 1035/6470 [12:53<1:13:30,  1.23it/s]
2|1036|Loss: 0.934055745601654:  16%|█▌        | 1035/6470 [12:53<1:13:30,  1.23it/s] 
2|1036|Loss: 0.934055745601654:  16%|█▌        | 1036/6470 [12:54<1:11:25,  1.27it/s]
2|1037|Loss: 0.9931924939155579:  16%|█▌        | 1036/6470 [12:54<1:11:25,  1.27it/s]
2|1037|Loss: 0.9931924939155579:  16%|█▌        | 1037/6470 [12:54<1:11:21,  1.27it/s]
2|1038|Loss: 0.9146428108215332:  16%|█▌        | 1037/6470 [12:55<1:11:21,  1.27it/s]
2|1038|Loss: 0.9146428108215332:  16%|█▌        | 1038/6470 [12:55<1:11:22,  1.27it/s]
2|1039|Loss: 0.7054204940795898:  16%|█▌        | 1038/6470 [12:56<1:11:22,  1.27it/s]
2|1039|Loss: 0.7054204940795898:  16%|█▌        | 1039/6470 [12:56<1:09:29,  1.30it/s]
2|1040|Loss: 0.9923322796821594:  16%|█▌        | 1039/6470 [12:56<1:09:29,  1.30it/s]
2|1040|Loss: 0.9923322796821594:  16%|█▌        | 1040/6470 [12:57<1:12:56,  1.24it/s]
2|1041|Loss: 0.8211470246315002:  16%|█▌        | 1040/6470 [12:57<1:12:56,  1.24it/s]
2|1041|Loss: 0.8211470246315002:  16%|█▌        | 1041/6470 [12:58<1:13:49,  1.23it/s]
2|1042|Loss: 0.69678795337677:  16%|█▌        | 1041/6470 [12:58<1:13:49,  1.23it/s]  
2|1042|Loss: 0.69678795337677:  16%|█▌        | 1042/6470 [12:59<1:17:54,  1.16it/s]
2|1043|Loss: 0.786945641040802:  16%|█▌        | 1042/6470 [12:59<1:17:54,  1.16it/s]
2|1043|Loss: 0.786945641040802:  16%|█▌        | 1043/6470 [13:00<1:19:49,  1.13it/s]
2|1044|Loss: 0.7516914010047913:  16%|█▌        | 1043/6470 [13:00<1:19:49,  1.13it/s]
2|1044|Loss: 0.7516914010047913:  16%|█▌        | 1044/6470 [13:01<1:21:51,  1.10it/s]
2|1045|Loss: 0.8802958130836487:  16%|█▌        | 1044/6470 [13:01<1:21:51,  1.10it/s]
2|1045|Loss: 0.8802958130836487:  16%|█▌        | 1045/6470 [13:01<1:22:43,  1.09it/s]
2|1046|Loss: 0.9337815046310425:  16%|█▌        | 1045/6470 [13:02<1:22:43,  1.09it/s]
2|1046|Loss: 0.9337815046310425:  16%|█▌        | 1046/6470 [13:02<1:20:28,  1.12it/s]
2|1047|Loss: 0.8792851567268372:  16%|█▌        | 1046/6470 [13:03<1:20:28,  1.12it/s]
2|1047|Loss: 0.8792851567268372:  16%|█▌        | 1047/6470 [13:03<1:18:27,  1.15it/s]
2|1048|Loss: 0.8923171758651733:  16%|█▌        | 1047/6470 [13:04<1:18:27,  1.15it/s]
2|1048|Loss: 0.8923171758651733:  16%|█▌        | 1048/6470 [13:04<1:13:36,  1.23it/s]
2|1049|Loss: 0.8619389533996582:  16%|█▌        | 1048/6470 [13:04<1:13:36,  1.23it/s]
2|1049|Loss: 0.8619389533996582:  16%|█▌        | 1049/6470 [13:05<1:10:18,  1.29it/s]
2|1050|Loss: 0.7617507576942444:  16%|█▌        | 1049/6470 [13:05<1:10:18,  1.29it/s]
2|1050|Loss: 0.7617507576942444:  16%|█▌        | 1050/6470 [13:05<1:03:27,  1.42it/s]
2|1051|Loss: 0.9283698201179504:  16%|█▌        | 1050/6470 [13:05<1:03:27,  1.42it/s]
2|1051|Loss: 0.9283698201179504:  16%|█▌        | 1051/6470 [13:06<1:00:09,  1.50it/s]
2|1052|Loss: 0.777668297290802:  16%|█▌        | 1051/6470 [13:06<1:00:09,  1.50it/s] 
2|1052|Loss: 0.777668297290802:  16%|█▋        | 1052/6470 [13:06<57:13,  1.58it/s]  
2|1053|Loss: 0.8838137984275818:  16%|█▋        | 1052/6470 [13:07<57:13,  1.58it/s]
2|1053|Loss: 0.8838137984275818:  16%|█▋        | 1053/6470 [13:07<1:03:39,  1.42it/s]
2|1054|Loss: 1.0450682640075684:  16%|█▋        | 1053/6470 [13:08<1:03:39,  1.42it/s]
2|1054|Loss: 1.0450682640075684:  16%|█▋        | 1054/6470 [13:08<1:07:06,  1.35it/s]
2|1055|Loss: 0.7565932273864746:  16%|█▋        | 1054/6470 [13:08<1:07:06,  1.35it/s]
2|1055|Loss: 0.7565932273864746:  16%|█▋        | 1055/6470 [13:09<1:11:28,  1.26it/s]
2|1056|Loss: 0.8510019779205322:  16%|█▋        | 1055/6470 [13:09<1:11:28,  1.26it/s]
2|1056|Loss: 0.8510019779205322:  16%|█▋        | 1056/6470 [13:10<1:10:29,  1.28it/s]
2|1057|Loss: 0.8720637559890747:  16%|█▋        | 1056/6470 [13:10<1:10:29,  1.28it/s]
2|1057|Loss: 0.8720637559890747:  16%|█▋        | 1057/6470 [13:10<1:10:39,  1.28it/s]
2|1058|Loss: 0.7154431939125061:  16%|█▋        | 1057/6470 [13:11<1:10:39,  1.28it/s]
2|1058|Loss: 0.7154431939125061:  16%|█▋        | 1058/6470 [13:11<1:14:40,  1.21it/s]
2|1059|Loss: 0.7570862770080566:  16%|█▋        | 1058/6470 [13:12<1:14:40,  1.21it/s]
2|1059|Loss: 0.7570862770080566:  16%|█▋        | 1059/6470 [13:12<1:14:13,  1.22it/s]
2|1060|Loss: 1.08548903465271:  16%|█▋        | 1059/6470 [13:12<1:14:13,  1.22it/s]  
2|1060|Loss: 1.08548903465271:  16%|█▋        | 1060/6470 [13:13<1:07:45,  1.33it/s]
2|1061|Loss: 1.0367623567581177:  16%|█▋        | 1060/6470 [13:13<1:07:45,  1.33it/s]
2|1061|Loss: 1.0367623567581177:  16%|█▋        | 1061/6470 [13:13<1:03:10,  1.43it/s]
2|1062|Loss: 0.7502572536468506:  16%|█▋        | 1061/6470 [13:14<1:03:10,  1.43it/s]
2|1062|Loss: 0.7502572536468506:  16%|█▋        | 1062/6470 [13:14<1:03:51,  1.41it/s]
2|1063|Loss: 0.7670552730560303:  16%|█▋        | 1062/6470 [13:14<1:03:51,  1.41it/s]
2|1063|Loss: 0.7670552730560303:  16%|█▋        | 1063/6470 [13:15<1:06:02,  1.36it/s]
2|1064|Loss: 0.8754469156265259:  16%|█▋        | 1063/6470 [13:15<1:06:02,  1.36it/s]
2|1064|Loss: 0.8754469156265259:  16%|█▋        | 1064/6470 [13:15<1:04:58,  1.39it/s]
2|1065|Loss: 0.8851621150970459:  16%|█▋        | 1064/6470 [13:16<1:04:58,  1.39it/s]
2|1065|Loss: 0.8851621150970459:  16%|█▋        | 1065/6470 [13:16<1:11:26,  1.26it/s]
2|1066|Loss: 0.8502475619316101:  16%|█▋        | 1065/6470 [13:17<1:11:26,  1.26it/s]
2|1066|Loss: 0.8502475619316101:  16%|█▋        | 1066/6470 [13:17<1:18:32,  1.15it/s]
2|1067|Loss: 0.8026571273803711:  16%|█▋        | 1066/6470 [13:18<1:18:32,  1.15it/s]
2|1067|Loss: 0.8026571273803711:  16%|█▋        | 1067/6470 [13:18<1:21:59,  1.10it/s]
2|1068|Loss: 0.8409653306007385:  16%|█▋        | 1067/6470 [13:19<1:21:59,  1.10it/s]
2|1068|Loss: 0.8409653306007385:  17%|█▋        | 1068/6470 [13:19<1:17:25,  1.16it/s]
2|1069|Loss: 0.8658128380775452:  17%|█▋        | 1068/6470 [13:20<1:17:25,  1.16it/s]
2|1069|Loss: 0.8658128380775452:  17%|█▋        | 1069/6470 [13:20<1:16:32,  1.18it/s]
2|1070|Loss: 0.9977556467056274:  17%|█▋        | 1069/6470 [13:21<1:16:32,  1.18it/s]
2|1070|Loss: 0.9977556467056274:  17%|█▋        | 1070/6470 [13:21<1:17:39,  1.16it/s]
2|1071|Loss: 0.8443129658699036:  17%|█▋        | 1070/6470 [13:21<1:17:39,  1.16it/s]
2|1071|Loss: 0.8443129658699036:  17%|█▋        | 1071/6470 [13:22<1:18:07,  1.15it/s]
2|1072|Loss: 0.861990749835968:  17%|█▋        | 1071/6470 [13:22<1:18:07,  1.15it/s] 
2|1072|Loss: 0.861990749835968:  17%|█▋        | 1072/6470 [13:23<1:17:15,  1.16it/s]
2|1073|Loss: 0.8275086879730225:  17%|█▋        | 1072/6470 [13:23<1:17:15,  1.16it/s]
2|1073|Loss: 0.8275086879730225:  17%|█▋        | 1073/6470 [13:24<1:18:48,  1.14it/s]
2|1074|Loss: 0.9525718092918396:  17%|█▋        | 1073/6470 [13:24<1:18:48,  1.14it/s]
2|1074|Loss: 0.9525718092918396:  17%|█▋        | 1074/6470 [13:24<1:14:54,  1.20it/s]
2|1075|Loss: 0.895272970199585:  17%|█▋        | 1074/6470 [13:25<1:14:54,  1.20it/s] 
2|1075|Loss: 0.895272970199585:  17%|█▋        | 1075/6470 [13:25<1:12:30,  1.24it/s]
2|1076|Loss: 0.9230908751487732:  17%|█▋        | 1075/6470 [13:25<1:12:30,  1.24it/s]
2|1076|Loss: 0.9230908751487732:  17%|█▋        | 1076/6470 [13:26<1:09:19,  1.30it/s]
2|1077|Loss: 0.8669145107269287:  17%|█▋        | 1076/6470 [13:26<1:09:19,  1.30it/s]
2|1077|Loss: 0.8669145107269287:  17%|█▋        | 1077/6470 [13:27<1:11:28,  1.26it/s]
2|1078|Loss: 0.7753843069076538:  17%|█▋        | 1077/6470 [13:27<1:11:28,  1.26it/s]
2|1078|Loss: 0.7753843069076538:  17%|█▋        | 1078/6470 [13:27<1:12:13,  1.24it/s]
2|1079|Loss: 0.6410968899726868:  17%|█▋        | 1078/6470 [13:28<1:12:13,  1.24it/s]
2|1079|Loss: 0.6410968899726868:  17%|█▋        | 1079/6470 [13:28<1:14:01,  1.21it/s]
2|1080|Loss: 0.8986842036247253:  17%|█▋        | 1079/6470 [13:29<1:14:01,  1.21it/s]
2|1080|Loss: 0.8986842036247253:  17%|█▋        | 1080/6470 [13:29<1:13:33,  1.22it/s]
2|1081|Loss: 0.9301964044570923:  17%|█▋        | 1080/6470 [13:29<1:13:33,  1.22it/s]
2|1081|Loss: 0.9301964044570923:  17%|█▋        | 1081/6470 [13:30<1:07:00,  1.34it/s]
2|1082|Loss: 0.8894789218902588:  17%|█▋        | 1081/6470 [13:30<1:07:00,  1.34it/s]
2|1082|Loss: 0.8894789218902588:  17%|█▋        | 1082/6470 [13:30<1:05:42,  1.37it/s]
2|1083|Loss: 0.9410340785980225:  17%|█▋        | 1082/6470 [13:31<1:05:42,  1.37it/s]
2|1083|Loss: 0.9410340785980225:  17%|█▋        | 1083/6470 [13:31<1:07:01,  1.34it/s]
2|1084|Loss: 0.8184508681297302:  17%|█▋        | 1083/6470 [13:32<1:07:01,  1.34it/s]
2|1084|Loss: 0.8184508681297302:  17%|█▋        | 1084/6470 [13:32<1:08:39,  1.31it/s]
2|1085|Loss: 0.7884405851364136:  17%|█▋        | 1084/6470 [13:32<1:08:39,  1.31it/s]
2|1085|Loss: 0.7884405851364136:  17%|█▋        | 1085/6470 [13:33<1:07:26,  1.33it/s]
2|1086|Loss: 0.9444194436073303:  17%|█▋        | 1085/6470 [13:33<1:07:26,  1.33it/s]
2|1086|Loss: 0.9444194436073303:  17%|█▋        | 1086/6470 [13:33<1:08:05,  1.32it/s]
2|1087|Loss: 0.8628279566764832:  17%|█▋        | 1086/6470 [13:34<1:08:05,  1.32it/s]
2|1087|Loss: 0.8628279566764832:  17%|█▋        | 1087/6470 [13:34<1:11:58,  1.25it/s]
2|1088|Loss: 0.7574032545089722:  17%|█▋        | 1087/6470 [13:35<1:11:58,  1.25it/s]
2|1088|Loss: 0.7574032545089722:  17%|█▋        | 1088/6470 [13:35<1:12:43,  1.23it/s]
2|1089|Loss: 0.8375535011291504:  17%|█▋        | 1088/6470 [13:36<1:12:43,  1.23it/s]
2|1089|Loss: 0.8375535011291504:  17%|█▋        | 1089/6470 [13:36<1:07:35,  1.33it/s]
2|1090|Loss: 0.890892744064331:  17%|█▋        | 1089/6470 [13:36<1:07:35,  1.33it/s] 
2|1090|Loss: 0.890892744064331:  17%|█▋        | 1090/6470 [13:37<1:07:41,  1.32it/s]
2|1091|Loss: 0.7717563509941101:  17%|█▋        | 1090/6470 [13:37<1:07:41,  1.32it/s]
2|1091|Loss: 0.7717563509941101:  17%|█▋        | 1091/6470 [13:37<1:09:08,  1.30it/s]
2|1092|Loss: 0.908280611038208:  17%|█▋        | 1091/6470 [13:38<1:09:08,  1.30it/s] 
2|1092|Loss: 0.908280611038208:  17%|█▋        | 1092/6470 [13:38<1:10:18,  1.27it/s]
2|1093|Loss: 0.8196284174919128:  17%|█▋        | 1092/6470 [13:39<1:10:18,  1.27it/s]
2|1093|Loss: 0.8196284174919128:  17%|█▋        | 1093/6470 [13:39<1:12:11,  1.24it/s]
2|1094|Loss: 0.7715964913368225:  17%|█▋        | 1093/6470 [13:39<1:12:11,  1.24it/s]
2|1094|Loss: 0.7715964913368225:  17%|█▋        | 1094/6470 [13:40<1:09:37,  1.29it/s]
2|1095|Loss: 0.9326733946800232:  17%|█▋        | 1094/6470 [13:40<1:09:37,  1.29it/s]
2|1095|Loss: 0.9326733946800232:  17%|█▋        | 1095/6470 [13:40<1:07:32,  1.33it/s]
2|1096|Loss: 0.8538867235183716:  17%|█▋        | 1095/6470 [13:41<1:07:32,  1.33it/s]
2|1096|Loss: 0.8538867235183716:  17%|█▋        | 1096/6470 [13:41<1:06:08,  1.35it/s]
2|1097|Loss: 0.6812034845352173:  17%|█▋        | 1096/6470 [13:42<1:06:08,  1.35it/s]
2|1097|Loss: 0.6812034845352173:  17%|█▋        | 1097/6470 [13:42<1:12:53,  1.23it/s]
2|1098|Loss: 0.8373730778694153:  17%|█▋        | 1097/6470 [13:43<1:12:53,  1.23it/s]
2|1098|Loss: 0.8373730778694153:  17%|█▋        | 1098/6470 [13:43<1:13:52,  1.21it/s]
2|1099|Loss: 0.7537659406661987:  17%|█▋        | 1098/6470 [13:43<1:13:52,  1.21it/s]
2|1099|Loss: 0.7537659406661987:  17%|█▋        | 1099/6470 [13:44<1:07:59,  1.32it/s]
2|1100|Loss: 0.854770302772522:  17%|█▋        | 1099/6470 [13:44<1:07:59,  1.32it/s] 
2|1100|Loss: 0.854770302772522:  17%|█▋        | 1100/6470 [13:44<1:09:14,  1.29it/s]
2|1101|Loss: 0.8316794037818909:  17%|█▋        | 1100/6470 [13:45<1:09:14,  1.29it/s]
2|1101|Loss: 0.8316794037818909:  17%|█▋        | 1101/6470 [13:45<1:04:29,  1.39it/s]
2|1102|Loss: 0.8106322884559631:  17%|█▋        | 1101/6470 [13:45<1:04:29,  1.39it/s]
2|1102|Loss: 0.8106322884559631:  17%|█▋        | 1102/6470 [13:46<58:47,  1.52it/s]  
2|1103|Loss: 0.9849064350128174:  17%|█▋        | 1102/6470 [13:46<58:47,  1.52it/s]
2|1103|Loss: 0.9849064350128174:  17%|█▋        | 1103/6470 [13:46<1:01:05,  1.46it/s]
2|1104|Loss: 0.8404841423034668:  17%|█▋        | 1103/6470 [13:47<1:01:05,  1.46it/s]
2|1104|Loss: 0.8404841423034668:  17%|█▋        | 1104/6470 [13:47<1:03:23,  1.41it/s]
2|1105|Loss: 0.9155951738357544:  17%|█▋        | 1104/6470 [13:47<1:03:23,  1.41it/s]
2|1105|Loss: 0.9155951738357544:  17%|█▋        | 1105/6470 [13:48<1:03:31,  1.41it/s]
2|1106|Loss: 0.8497973680496216:  17%|█▋        | 1105/6470 [13:48<1:03:31,  1.41it/s]
2|1106|Loss: 0.8497973680496216:  17%|█▋        | 1106/6470 [13:48<1:01:55,  1.44it/s]
2|1107|Loss: 0.9528322815895081:  17%|█▋        | 1106/6470 [13:49<1:01:55,  1.44it/s]
2|1107|Loss: 0.9528322815895081:  17%|█▋        | 1107/6470 [13:49<1:01:39,  1.45it/s]
2|1108|Loss: 0.8440417051315308:  17%|█▋        | 1107/6470 [13:49<1:01:39,  1.45it/s]
2|1108|Loss: 0.8440417051315308:  17%|█▋        | 1108/6470 [13:50<1:04:03,  1.39it/s]
2|1109|Loss: 0.9071846008300781:  17%|█▋        | 1108/6470 [13:50<1:04:03,  1.39it/s]
2|1109|Loss: 0.9071846008300781:  17%|█▋        | 1109/6470 [13:51<1:03:38,  1.40it/s]
2|1110|Loss: 0.9242112040519714:  17%|█▋        | 1109/6470 [13:51<1:03:38,  1.40it/s]
2|1110|Loss: 0.9242112040519714:  17%|█▋        | 1110/6470 [13:51<1:01:36,  1.45it/s]
2|1111|Loss: 0.8734333515167236:  17%|█▋        | 1110/6470 [13:52<1:01:36,  1.45it/s]
2|1111|Loss: 0.8734333515167236:  17%|█▋        | 1111/6470 [13:52<1:05:52,  1.36it/s]
2|1112|Loss: 0.8811388611793518:  17%|█▋        | 1111/6470 [13:53<1:05:52,  1.36it/s]
2|1112|Loss: 0.8811388611793518:  17%|█▋        | 1112/6470 [13:53<1:11:28,  1.25it/s]
2|1113|Loss: 0.898015558719635:  17%|█▋        | 1112/6470 [13:54<1:11:28,  1.25it/s] 
2|1113|Loss: 0.898015558719635:  17%|█▋        | 1113/6470 [13:54<1:14:58,  1.19it/s]
2|1114|Loss: 0.9498879909515381:  17%|█▋        | 1113/6470 [13:54<1:14:58,  1.19it/s]
2|1114|Loss: 0.9498879909515381:  17%|█▋        | 1114/6470 [13:55<1:12:34,  1.23it/s]
2|1115|Loss: 0.8708299994468689:  17%|█▋        | 1114/6470 [13:55<1:12:34,  1.23it/s]
2|1115|Loss: 0.8708299994468689:  17%|█▋        | 1115/6470 [13:55<1:09:16,  1.29it/s]
2|1116|Loss: 0.9905363917350769:  17%|█▋        | 1115/6470 [13:56<1:09:16,  1.29it/s]
2|1116|Loss: 0.9905363917350769:  17%|█▋        | 1116/6470 [13:56<1:07:32,  1.32it/s]
2|1117|Loss: 0.6676405072212219:  17%|█▋        | 1116/6470 [13:56<1:07:32,  1.32it/s]
2|1117|Loss: 0.6676405072212219:  17%|█▋        | 1117/6470 [13:57<1:01:15,  1.46it/s]
2|1118|Loss: 0.8273320198059082:  17%|█▋        | 1117/6470 [13:57<1:01:15,  1.46it/s]
2|1118|Loss: 0.8273320198059082:  17%|█▋        | 1118/6470 [13:57<1:01:24,  1.45it/s]
2|1119|Loss: 0.8316084146499634:  17%|█▋        | 1118/6470 [13:58<1:01:24,  1.45it/s]
2|1119|Loss: 0.8316084146499634:  17%|█▋        | 1119/6470 [13:58<57:32,  1.55it/s]  
2|1120|Loss: 0.9489352107048035:  17%|█▋        | 1119/6470 [13:58<57:32,  1.55it/s]
2|1120|Loss: 0.9489352107048035:  17%|█▋        | 1120/6470 [13:59<58:47,  1.52it/s]
2|1121|Loss: 0.8778534531593323:  17%|█▋        | 1120/6470 [13:59<58:47,  1.52it/s]
2|1121|Loss: 0.8778534531593323:  17%|█▋        | 1121/6470 [13:59<1:02:35,  1.42it/s]
2|1122|Loss: 0.9431896805763245:  17%|█▋        | 1121/6470 [14:00<1:02:35,  1.42it/s]
2|1122|Loss: 0.9431896805763245:  17%|█▋        | 1122/6470 [14:00<1:04:48,  1.38it/s]
2|1123|Loss: 0.8149492144584656:  17%|█▋        | 1122/6470 [14:01<1:04:48,  1.38it/s]
2|1123|Loss: 0.8149492144584656:  17%|█▋        | 1123/6470 [14:01<1:05:32,  1.36it/s]
2|1124|Loss: 0.7437798380851746:  17%|█▋        | 1123/6470 [14:01<1:05:32,  1.36it/s]
2|1124|Loss: 0.7437798380851746:  17%|█▋        | 1124/6470 [14:02<1:05:00,  1.37it/s]
2|1125|Loss: 0.9708637595176697:  17%|█▋        | 1124/6470 [14:02<1:05:00,  1.37it/s]
2|1125|Loss: 0.9708637595176697:  17%|█▋        | 1125/6470 [14:02<1:10:07,  1.27it/s]
2|1126|Loss: 0.791753888130188:  17%|█▋        | 1125/6470 [14:03<1:10:07,  1.27it/s] 
2|1126|Loss: 0.791753888130188:  17%|█▋        | 1126/6470 [14:03<1:09:00,  1.29it/s]
2|1127|Loss: 0.9472989439964294:  17%|█▋        | 1126/6470 [14:04<1:09:00,  1.29it/s]
2|1127|Loss: 0.9472989439964294:  17%|█▋        | 1127/6470 [14:04<1:02:08,  1.43it/s]
2|1128|Loss: 0.7658466100692749:  17%|█▋        | 1127/6470 [14:04<1:02:08,  1.43it/s]
2|1128|Loss: 0.7658466100692749:  17%|█▋        | 1128/6470 [14:04<1:01:57,  1.44it/s]
2|1129|Loss: 1.0313867330551147:  17%|█▋        | 1128/6470 [14:05<1:01:57,  1.44it/s]
2|1129|Loss: 1.0313867330551147:  17%|█▋        | 1129/6470 [14:05<1:02:56,  1.41it/s]
2|1130|Loss: 0.9411032795906067:  17%|█▋        | 1129/6470 [14:06<1:02:56,  1.41it/s]
2|1130|Loss: 0.9411032795906067:  17%|█▋        | 1130/6470 [14:06<1:05:38,  1.36it/s]
2|1131|Loss: 0.7701697945594788:  17%|█▋        | 1130/6470 [14:06<1:05:38,  1.36it/s]
2|1131|Loss: 0.7701697945594788:  17%|█▋        | 1131/6470 [14:07<1:07:43,  1.31it/s]
2|1132|Loss: 0.9160448312759399:  17%|█▋        | 1131/6470 [14:07<1:07:43,  1.31it/s]
2|1132|Loss: 0.9160448312759399:  17%|█▋        | 1132/6470 [14:07<59:58,  1.48it/s]  
2|1133|Loss: 0.7493525147438049:  17%|█▋        | 1132/6470 [14:08<59:58,  1.48it/s]
2|1133|Loss: 0.7493525147438049:  18%|█▊        | 1133/6470 [14:08<1:01:38,  1.44it/s]
2|1134|Loss: 0.8501014113426208:  18%|█▊        | 1133/6470 [14:08<1:01:38,  1.44it/s]
2|1134|Loss: 0.8501014113426208:  18%|█▊        | 1134/6470 [14:09<1:02:40,  1.42it/s]
2|1135|Loss: 0.8239234685897827:  18%|█▊        | 1134/6470 [14:09<1:02:40,  1.42it/s]
2|1135|Loss: 0.8239234685897827:  18%|█▊        | 1135/6470 [14:10<1:03:56,  1.39it/s]
2|1136|Loss: 0.8450610041618347:  18%|█▊        | 1135/6470 [14:10<1:03:56,  1.39it/s]
2|1136|Loss: 0.8450610041618347:  18%|█▊        | 1136/6470 [14:10<1:09:48,  1.27it/s]
2|1137|Loss: 0.9305863380432129:  18%|█▊        | 1136/6470 [14:11<1:09:48,  1.27it/s]
2|1137|Loss: 0.9305863380432129:  18%|█▊        | 1137/6470 [14:11<1:11:34,  1.24it/s]
2|1138|Loss: 0.7808117866516113:  18%|█▊        | 1137/6470 [14:12<1:11:34,  1.24it/s]
2|1138|Loss: 0.7808117866516113:  18%|█▊        | 1138/6470 [14:12<1:04:29,  1.38it/s]
2|1139|Loss: 0.8575469255447388:  18%|█▊        | 1138/6470 [14:12<1:04:29,  1.38it/s]
2|1139|Loss: 0.8575469255447388:  18%|█▊        | 1139/6470 [14:12<1:01:15,  1.45it/s]
2|1140|Loss: 0.7512948513031006:  18%|█▊        | 1139/6470 [14:13<1:01:15,  1.45it/s]
2|1140|Loss: 0.7512948513031006:  18%|█▊        | 1140/6470 [14:13<57:41,  1.54it/s]  
2|1141|Loss: 0.9267442226409912:  18%|█▊        | 1140/6470 [14:13<57:41,  1.54it/s]
2|1141|Loss: 0.9267442226409912:  18%|█▊        | 1141/6470 [14:14<56:04,  1.58it/s]
2|1142|Loss: 0.9472683668136597:  18%|█▊        | 1141/6470 [14:14<56:04,  1.58it/s]
2|1142|Loss: 0.9472683668136597:  18%|█▊        | 1142/6470 [14:14<54:29,  1.63it/s]
2|1143|Loss: 1.0454999208450317:  18%|█▊        | 1142/6470 [14:15<54:29,  1.63it/s]
2|1143|Loss: 1.0454999208450317:  18%|█▊        | 1143/6470 [14:15<55:17,  1.61it/s]
2|1144|Loss: 0.8928106427192688:  18%|█▊        | 1143/6470 [14:15<55:17,  1.61it/s]
2|1144|Loss: 0.8928106427192688:  18%|█▊        | 1144/6470 [14:16<59:30,  1.49it/s]
2|1145|Loss: 0.8172087073326111:  18%|█▊        | 1144/6470 [14:16<59:30,  1.49it/s]
2|1145|Loss: 0.8172087073326111:  18%|█▊        | 1145/6470 [14:16<1:05:17,  1.36it/s]
2|1146|Loss: 0.9482934474945068:  18%|█▊        | 1145/6470 [14:17<1:05:17,  1.36it/s]
2|1146|Loss: 0.9482934474945068:  18%|█▊        | 1146/6470 [14:17<1:03:53,  1.39it/s]
2|1147|Loss: 0.8951835632324219:  18%|█▊        | 1146/6470 [14:18<1:03:53,  1.39it/s]
2|1147|Loss: 0.8951835632324219:  18%|█▊        | 1147/6470 [14:18<1:01:39,  1.44it/s]
2|1148|Loss: 0.7906761169433594:  18%|█▊        | 1147/6470 [14:18<1:01:39,  1.44it/s]
2|1148|Loss: 0.7906761169433594:  18%|█▊        | 1148/6470 [14:19<1:02:51,  1.41it/s]
2|1149|Loss: 0.7855668663978577:  18%|█▊        | 1148/6470 [14:19<1:02:51,  1.41it/s]
2|1149|Loss: 0.7855668663978577:  18%|█▊        | 1149/6470 [14:19<1:01:48,  1.43it/s]
2|1150|Loss: 0.8262540698051453:  18%|█▊        | 1149/6470 [14:20<1:01:48,  1.43it/s]
2|1150|Loss: 0.8262540698051453:  18%|█▊        | 1150/6470 [14:20<1:00:38,  1.46it/s]
2|1151|Loss: 0.7836527824401855:  18%|█▊        | 1150/6470 [14:20<1:00:38,  1.46it/s]
2|1151|Loss: 0.7836527824401855:  18%|█▊        | 1151/6470 [14:21<1:06:04,  1.34it/s]
2|1152|Loss: 0.8144919276237488:  18%|█▊        | 1151/6470 [14:21<1:06:04,  1.34it/s]
2|1152|Loss: 0.8144919276237488:  18%|█▊        | 1152/6470 [14:21<1:00:00,  1.48it/s]
2|1153|Loss: 0.9370648264884949:  18%|█▊        | 1152/6470 [14:22<1:00:00,  1.48it/s]
2|1153|Loss: 0.9370648264884949:  18%|█▊        | 1153/6470 [14:22<1:01:06,  1.45it/s]
2|1154|Loss: 0.8575558066368103:  18%|█▊        | 1153/6470 [14:22<1:01:06,  1.45it/s]
2|1154|Loss: 0.8575558066368103:  18%|█▊        | 1154/6470 [14:23<1:05:45,  1.35it/s]
2|1155|Loss: 0.9036788940429688:  18%|█▊        | 1154/6470 [14:23<1:05:45,  1.35it/s]
2|1155|Loss: 0.9036788940429688:  18%|█▊        | 1155/6470 [14:23<1:01:03,  1.45it/s]
2|1156|Loss: 0.7948083281517029:  18%|█▊        | 1155/6470 [14:24<1:01:03,  1.45it/s]
2|1156|Loss: 0.7948083281517029:  18%|█▊        | 1156/6470 [14:24<55:55,  1.58it/s]  
2|1157|Loss: 0.9658740162849426:  18%|█▊        | 1156/6470 [14:24<55:55,  1.58it/s]
2|1157|Loss: 0.9658740162849426:  18%|█▊        | 1157/6470 [14:25<1:00:08,  1.47it/s]
2|1158|Loss: 0.9151725172996521:  18%|█▊        | 1157/6470 [14:25<1:00:08,  1.47it/s]
2|1158|Loss: 0.9151725172996521:  18%|█▊        | 1158/6470 [14:26<1:06:38,  1.33it/s]
2|1159|Loss: 0.8788254857063293:  18%|█▊        | 1158/6470 [14:26<1:06:38,  1.33it/s]
2|1159|Loss: 0.8788254857063293:  18%|█▊        | 1159/6470 [14:27<1:10:48,  1.25it/s]
2|1160|Loss: 0.7453091144561768:  18%|█▊        | 1159/6470 [14:27<1:10:48,  1.25it/s]
2|1160|Loss: 0.7453091144561768:  18%|█▊        | 1160/6470 [14:27<1:13:20,  1.21it/s]
2|1161|Loss: 0.787948489189148:  18%|█▊        | 1160/6470 [14:28<1:13:20,  1.21it/s] 
2|1161|Loss: 0.787948489189148:  18%|█▊        | 1161/6470 [14:28<1:12:38,  1.22it/s]
2|1162|Loss: 0.849260151386261:  18%|█▊        | 1161/6470 [14:29<1:12:38,  1.22it/s]
2|1162|Loss: 0.849260151386261:  18%|█▊        | 1162/6470 [14:29<1:14:00,  1.20it/s]
2|1163|Loss: 0.8276900053024292:  18%|█▊        | 1162/6470 [14:30<1:14:00,  1.20it/s]
2|1163|Loss: 0.8276900053024292:  18%|█▊        | 1163/6470 [14:30<1:10:40,  1.25it/s]
2|1164|Loss: 0.775862991809845:  18%|█▊        | 1163/6470 [14:30<1:10:40,  1.25it/s] 
2|1164|Loss: 0.775862991809845:  18%|█▊        | 1164/6470 [14:30<1:05:39,  1.35it/s]
2|1165|Loss: 0.7792466282844543:  18%|█▊        | 1164/6470 [14:31<1:05:39,  1.35it/s]
2|1165|Loss: 0.7792466282844543:  18%|█▊        | 1165/6470 [14:31<1:00:54,  1.45it/s]
2|1166|Loss: 1.004165768623352:  18%|█▊        | 1165/6470 [14:31<1:00:54,  1.45it/s] 
2|1166|Loss: 1.004165768623352:  18%|█▊        | 1166/6470 [14:32<1:05:49,  1.34it/s]
2|1167|Loss: 0.7651890516281128:  18%|█▊        | 1166/6470 [14:32<1:05:49,  1.34it/s]
2|1167|Loss: 0.7651890516281128:  18%|█▊        | 1167/6470 [14:33<1:08:22,  1.29it/s]
2|1168|Loss: 1.1120171546936035:  18%|█▊        | 1167/6470 [14:33<1:08:22,  1.29it/s]
2|1168|Loss: 1.1120171546936035:  18%|█▊        | 1168/6470 [14:34<1:13:53,  1.20it/s]
2|1169|Loss: 0.7571290731430054:  18%|█▊        | 1168/6470 [14:34<1:13:53,  1.20it/s]
2|1169|Loss: 0.7571290731430054:  18%|█▊        | 1169/6470 [14:35<1:13:20,  1.20it/s]
2|1170|Loss: 0.7620992064476013:  18%|█▊        | 1169/6470 [14:35<1:13:20,  1.20it/s]
2|1170|Loss: 0.7620992064476013:  18%|█▊        | 1170/6470 [14:35<1:12:27,  1.22it/s]
2|1171|Loss: 0.7281076312065125:  18%|█▊        | 1170/6470 [14:36<1:12:27,  1.22it/s]
2|1171|Loss: 0.7281076312065125:  18%|█▊        | 1171/6470 [14:36<1:12:10,  1.22it/s]
2|1172|Loss: 0.826567530632019:  18%|█▊        | 1171/6470 [14:37<1:12:10,  1.22it/s] 
2|1172|Loss: 0.826567530632019:  18%|█▊        | 1172/6470 [14:37<1:11:04,  1.24it/s]
2|1173|Loss: 1.0049668550491333:  18%|█▊        | 1172/6470 [14:37<1:11:04,  1.24it/s]
2|1173|Loss: 1.0049668550491333:  18%|█▊        | 1173/6470 [14:38<1:07:55,  1.30it/s]
2|1174|Loss: 0.6039686799049377:  18%|█▊        | 1173/6470 [14:38<1:07:55,  1.30it/s]
2|1174|Loss: 0.6039686799049377:  18%|█▊        | 1174/6470 [14:38<1:07:08,  1.31it/s]
2|1175|Loss: 0.827772319316864:  18%|█▊        | 1174/6470 [14:39<1:07:08,  1.31it/s] 
2|1175|Loss: 0.827772319316864:  18%|█▊        | 1175/6470 [14:39<1:04:34,  1.37it/s]
2|1176|Loss: 0.6312311887741089:  18%|█▊        | 1175/6470 [14:39<1:04:34,  1.37it/s]
2|1176|Loss: 0.6312311887741089:  18%|█▊        | 1176/6470 [14:40<1:02:36,  1.41it/s]
2|1177|Loss: 0.7966518998146057:  18%|█▊        | 1176/6470 [14:40<1:02:36,  1.41it/s]
2|1177|Loss: 0.7966518998146057:  18%|█▊        | 1177/6470 [14:40<1:05:52,  1.34it/s]
2|1178|Loss: 0.8020147681236267:  18%|█▊        | 1177/6470 [14:41<1:05:52,  1.34it/s]
2|1178|Loss: 0.8020147681236267:  18%|█▊        | 1178/6470 [14:41<1:07:13,  1.31it/s]
2|1179|Loss: 0.97157883644104:  18%|█▊        | 1178/6470 [14:42<1:07:13,  1.31it/s]  
2|1179|Loss: 0.97157883644104:  18%|█▊        | 1179/6470 [14:42<1:07:16,  1.31it/s]
2|1180|Loss: 0.8923637270927429:  18%|█▊        | 1179/6470 [14:43<1:07:16,  1.31it/s]
2|1180|Loss: 0.8923637270927429:  18%|█▊        | 1180/6470 [14:43<1:11:09,  1.24it/s]
2|1181|Loss: 0.7559774518013:  18%|█▊        | 1180/6470 [14:43<1:11:09,  1.24it/s]   
2|1181|Loss: 0.7559774518013:  18%|█▊        | 1181/6470 [14:44<1:06:35,  1.32it/s]
2|1182|Loss: 0.8595765233039856:  18%|█▊        | 1181/6470 [14:44<1:06:35,  1.32it/s]
2|1182|Loss: 0.8595765233039856:  18%|█▊        | 1182/6470 [14:44<1:10:35,  1.25it/s]
2|1183|Loss: 0.6775656342506409:  18%|█▊        | 1182/6470 [14:45<1:10:35,  1.25it/s]
2|1183|Loss: 0.6775656342506409:  18%|█▊        | 1183/6470 [14:45<1:07:39,  1.30it/s]
2|1184|Loss: 0.8676398992538452:  18%|█▊        | 1183/6470 [14:45<1:07:39,  1.30it/s]
2|1184|Loss: 0.8676398992538452:  18%|█▊        | 1184/6470 [14:46<1:01:50,  1.42it/s]
2|1185|Loss: 1.034894347190857:  18%|█▊        | 1184/6470 [14:46<1:01:50,  1.42it/s] 
2|1185|Loss: 1.034894347190857:  18%|█▊        | 1185/6470 [14:46<1:02:23,  1.41it/s]
2|1186|Loss: 0.7958686351776123:  18%|█▊        | 1185/6470 [14:47<1:02:23,  1.41it/s]
2|1186|Loss: 0.7958686351776123:  18%|█▊        | 1186/6470 [14:47<1:08:46,  1.28it/s]
2|1187|Loss: 0.9476773738861084:  18%|█▊        | 1186/6470 [14:48<1:08:46,  1.28it/s]
2|1187|Loss: 0.9476773738861084:  18%|█▊        | 1187/6470 [14:48<1:08:31,  1.28it/s]
2|1188|Loss: 0.7956403493881226:  18%|█▊        | 1187/6470 [14:49<1:08:31,  1.28it/s]
2|1188|Loss: 0.7956403493881226:  18%|█▊        | 1188/6470 [14:49<1:03:03,  1.40it/s]
2|1189|Loss: 0.8984357714653015:  18%|█▊        | 1188/6470 [14:49<1:03:03,  1.40it/s]
2|1189|Loss: 0.8984357714653015:  18%|█▊        | 1189/6470 [14:50<1:06:00,  1.33it/s]
2|1190|Loss: 0.9350810647010803:  18%|█▊        | 1189/6470 [14:50<1:06:00,  1.33it/s]
2|1190|Loss: 0.9350810647010803:  18%|█▊        | 1190/6470 [14:50<1:10:34,  1.25it/s]
2|1191|Loss: 0.8070928454399109:  18%|█▊        | 1190/6470 [14:51<1:10:34,  1.25it/s]
2|1191|Loss: 0.8070928454399109:  18%|█▊        | 1191/6470 [14:51<1:13:09,  1.20it/s]
2|1192|Loss: 0.8351465463638306:  18%|█▊        | 1191/6470 [14:52<1:13:09,  1.20it/s]
2|1192|Loss: 0.8351465463638306:  18%|█▊        | 1192/6470 [14:52<1:12:18,  1.22it/s]
2|1193|Loss: 0.7436133623123169:  18%|█▊        | 1192/6470 [14:53<1:12:18,  1.22it/s]
2|1193|Loss: 0.7436133623123169:  18%|█▊        | 1193/6470 [14:53<1:07:51,  1.30it/s]
2|1194|Loss: 0.7830080389976501:  18%|█▊        | 1193/6470 [14:53<1:07:51,  1.30it/s]
2|1194|Loss: 0.7830080389976501:  18%|█▊        | 1194/6470 [14:54<1:07:12,  1.31it/s]
2|1195|Loss: 0.7168246507644653:  18%|█▊        | 1194/6470 [14:54<1:07:12,  1.31it/s]
2|1195|Loss: 0.7168246507644653:  18%|█▊        | 1195/6470 [14:54<1:04:03,  1.37it/s]
2|1196|Loss: 0.7728655934333801:  18%|█▊        | 1195/6470 [14:55<1:04:03,  1.37it/s]
2|1196|Loss: 0.7728655934333801:  18%|█▊        | 1196/6470 [14:55<1:07:50,  1.30it/s]
2|1197|Loss: 0.886299192905426:  18%|█▊        | 1196/6470 [14:56<1:07:50,  1.30it/s] 
2|1197|Loss: 0.886299192905426:  19%|█▊        | 1197/6470 [14:56<1:07:11,  1.31it/s]
2|1198|Loss: 0.9882409572601318:  19%|█▊        | 1197/6470 [14:56<1:07:11,  1.31it/s]
2|1198|Loss: 0.9882409572601318:  19%|█▊        | 1198/6470 [14:57<1:05:31,  1.34it/s]
2|1199|Loss: 0.8390337824821472:  19%|█▊        | 1198/6470 [14:57<1:05:31,  1.34it/s]
2|1199|Loss: 0.8390337824821472:  19%|█▊        | 1199/6470 [14:57<1:04:05,  1.37it/s]
2|1200|Loss: 0.9167635440826416:  19%|█▊        | 1199/6470 [14:58<1:04:05,  1.37it/s]
2|1200|Loss: 0.9167635440826416:  19%|█▊        | 1200/6470 [14:58<1:00:06,  1.46it/s]
2|1201|Loss: 0.8495675921440125:  19%|█▊        | 1200/6470 [14:58<1:00:06,  1.46it/s]
2|1201|Loss: 0.8495675921440125:  19%|█▊        | 1201/6470 [14:59<1:01:26,  1.43it/s]
2|1202|Loss: 0.7886974811553955:  19%|█▊        | 1201/6470 [14:59<1:01:26,  1.43it/s]
2|1202|Loss: 0.7886974811553955:  19%|█▊        | 1202/6470 [14:59<1:05:50,  1.33it/s]
2|1203|Loss: 0.7752262353897095:  19%|█▊        | 1202/6470 [15:00<1:05:50,  1.33it/s]
2|1203|Loss: 0.7752262353897095:  19%|█▊        | 1203/6470 [15:00<1:05:31,  1.34it/s]
2|1204|Loss: 0.7601885795593262:  19%|█▊        | 1203/6470 [15:01<1:05:31,  1.34it/s]
2|1204|Loss: 0.7601885795593262:  19%|█▊        | 1204/6470 [15:01<1:03:55,  1.37it/s]
2|1205|Loss: 0.8072723746299744:  19%|█▊        | 1204/6470 [15:01<1:03:55,  1.37it/s]
2|1205|Loss: 0.8072723746299744:  19%|█▊        | 1205/6470 [15:01<59:37,  1.47it/s]  
2|1206|Loss: 0.9623413681983948:  19%|█▊        | 1205/6470 [15:02<59:37,  1.47it/s]
2|1206|Loss: 0.9623413681983948:  19%|█▊        | 1206/6470 [15:02<1:03:05,  1.39it/s]
2|1207|Loss: 0.6912615895271301:  19%|█▊        | 1206/6470 [15:03<1:03:05,  1.39it/s]
2|1207|Loss: 0.6912615895271301:  19%|█▊        | 1207/6470 [15:03<1:02:47,  1.40it/s]
2|1208|Loss: 0.8743450045585632:  19%|█▊        | 1207/6470 [15:03<1:02:47,  1.40it/s]
2|1208|Loss: 0.8743450045585632:  19%|█▊        | 1208/6470 [15:03<58:16,  1.51it/s]  
2|1209|Loss: 0.9881574511528015:  19%|█▊        | 1208/6470 [15:04<58:16,  1.51it/s]
2|1209|Loss: 0.9881574511528015:  19%|█▊        | 1209/6470 [15:04<58:48,  1.49it/s]
2|1210|Loss: 0.5846751928329468:  19%|█▊        | 1209/6470 [15:05<58:48,  1.49it/s]
2|1210|Loss: 0.5846751928329468:  19%|█▊        | 1210/6470 [15:05<58:51,  1.49it/s]
2|1211|Loss: 0.8293430209159851:  19%|█▊        | 1210/6470 [15:05<58:51,  1.49it/s]
2|1211|Loss: 0.8293430209159851:  19%|█▊        | 1211/6470 [15:05<58:08,  1.51it/s]
2|1212|Loss: 0.8969926238059998:  19%|█▊        | 1211/6470 [15:06<58:08,  1.51it/s]
2|1212|Loss: 0.8969926238059998:  19%|█▊        | 1212/6470 [15:06<1:00:36,  1.45it/s]
2|1213|Loss: 0.9085889458656311:  19%|█▊        | 1212/6470 [15:07<1:00:36,  1.45it/s]
2|1213|Loss: 0.9085889458656311:  19%|█▊        | 1213/6470 [15:07<1:06:53,  1.31it/s]
2|1214|Loss: 0.7809191346168518:  19%|█▊        | 1213/6470 [15:08<1:06:53,  1.31it/s]
2|1214|Loss: 0.7809191346168518:  19%|█▉        | 1214/6470 [15:08<1:06:04,  1.33it/s]
2|1215|Loss: 0.7887699007987976:  19%|█▉        | 1214/6470 [15:08<1:06:04,  1.33it/s]
2|1215|Loss: 0.7887699007987976:  19%|█▉        | 1215/6470 [15:09<1:07:08,  1.30it/s]
2|1216|Loss: 0.7799124121665955:  19%|█▉        | 1215/6470 [15:09<1:07:08,  1.30it/s]
2|1216|Loss: 0.7799124121665955:  19%|█▉        | 1216/6470 [15:09<59:01,  1.48it/s]  
2|1217|Loss: 0.8325768113136292:  19%|█▉        | 1216/6470 [15:09<59:01,  1.48it/s]
2|1217|Loss: 0.8325768113136292:  19%|█▉        | 1217/6470 [15:10<57:55,  1.51it/s]
2|1218|Loss: 0.868877112865448:  19%|█▉        | 1217/6470 [15:10<57:55,  1.51it/s] 
2|1218|Loss: 0.868877112865448:  19%|█▉        | 1218/6470 [15:11<1:05:50,  1.33it/s]
2|1219|Loss: 1.025580883026123:  19%|█▉        | 1218/6470 [15:11<1:05:50,  1.33it/s]
2|1219|Loss: 1.025580883026123:  19%|█▉        | 1219/6470 [15:12<1:11:13,  1.23it/s]
2|1220|Loss: 0.8124978542327881:  19%|█▉        | 1219/6470 [15:12<1:11:13,  1.23it/s]
2|1220|Loss: 0.8124978542327881:  19%|█▉        | 1220/6470 [15:12<1:09:10,  1.26it/s]
2|1221|Loss: 0.7657660245895386:  19%|█▉        | 1220/6470 [15:13<1:09:10,  1.26it/s]
2|1221|Loss: 0.7657660245895386:  19%|█▉        | 1221/6470 [15:13<1:11:12,  1.23it/s]
2|1222|Loss: 1.011003851890564:  19%|█▉        | 1221/6470 [15:14<1:11:12,  1.23it/s] 
2|1222|Loss: 1.011003851890564:  19%|█▉        | 1222/6470 [15:14<1:11:46,  1.22it/s]
2|1223|Loss: 0.6283760070800781:  19%|█▉        | 1222/6470 [15:15<1:11:46,  1.22it/s]
2|1223|Loss: 0.6283760070800781:  19%|█▉        | 1223/6470 [15:15<1:11:11,  1.23it/s]
2|1224|Loss: 0.888769268989563:  19%|█▉        | 1223/6470 [15:15<1:11:11,  1.23it/s] 
2|1224|Loss: 0.888769268989563:  19%|█▉        | 1224/6470 [15:16<1:10:49,  1.23it/s]
2|1225|Loss: 0.7227101922035217:  19%|█▉        | 1224/6470 [15:16<1:10:49,  1.23it/s]
2|1225|Loss: 0.7227101922035217:  19%|█▉        | 1225/6470 [15:16<1:06:02,  1.32it/s]
2|1226|Loss: 0.9147650003433228:  19%|█▉        | 1225/6470 [15:17<1:06:02,  1.32it/s]
2|1226|Loss: 0.9147650003433228:  19%|█▉        | 1226/6470 [15:17<1:05:07,  1.34it/s]
2|1227|Loss: 0.7965899109840393:  19%|█▉        | 1226/6470 [15:18<1:05:07,  1.34it/s]
2|1227|Loss: 0.7965899109840393:  19%|█▉        | 1227/6470 [15:18<1:04:29,  1.35it/s]
2|1228|Loss: 0.9199144244194031:  19%|█▉        | 1227/6470 [15:18<1:04:29,  1.35it/s]
2|1228|Loss: 0.9199144244194031:  19%|█▉        | 1228/6470 [15:18<1:00:30,  1.44it/s]
2|1229|Loss: 0.7864610552787781:  19%|█▉        | 1228/6470 [15:19<1:00:30,  1.44it/s]
2|1229|Loss: 0.7864610552787781:  19%|█▉        | 1229/6470 [15:19<58:09,  1.50it/s]  
2|1230|Loss: 0.8426518440246582:  19%|█▉        | 1229/6470 [15:19<58:09,  1.50it/s]
2|1230|Loss: 0.8426518440246582:  19%|█▉        | 1230/6470 [15:20<1:03:19,  1.38it/s]
2|1231|Loss: 0.8551040291786194:  19%|█▉        | 1230/6470 [15:20<1:03:19,  1.38it/s]
2|1231|Loss: 0.8551040291786194:  19%|█▉        | 1231/6470 [15:21<1:08:50,  1.27it/s]
2|1232|Loss: 0.8565770387649536:  19%|█▉        | 1231/6470 [15:21<1:08:50,  1.27it/s]
2|1232|Loss: 0.8565770387649536:  19%|█▉        | 1232/6470 [15:22<1:12:32,  1.20it/s]
2|1233|Loss: 0.7713385820388794:  19%|█▉        | 1232/6470 [15:22<1:12:32,  1.20it/s]
2|1233|Loss: 0.7713385820388794:  19%|█▉        | 1233/6470 [15:22<1:08:12,  1.28it/s]
2|1234|Loss: 0.7909492254257202:  19%|█▉        | 1233/6470 [15:23<1:08:12,  1.28it/s]
2|1234|Loss: 0.7909492254257202:  19%|█▉        | 1234/6470 [15:23<1:07:10,  1.30it/s]
2|1235|Loss: 0.8777530789375305:  19%|█▉        | 1234/6470 [15:24<1:07:10,  1.30it/s]
2|1235|Loss: 0.8777530789375305:  19%|█▉        | 1235/6470 [15:24<1:05:57,  1.32it/s]
2|1236|Loss: 1.0036866664886475:  19%|█▉        | 1235/6470 [15:24<1:05:57,  1.32it/s]
2|1236|Loss: 1.0036866664886475:  19%|█▉        | 1236/6470 [15:25<1:09:37,  1.25it/s]
2|1237|Loss: 0.8838802576065063:  19%|█▉        | 1236/6470 [15:25<1:09:37,  1.25it/s]
2|1237|Loss: 0.8838802576065063:  19%|█▉        | 1237/6470 [15:26<1:13:09,  1.19it/s]
2|1238|Loss: 0.9240529537200928:  19%|█▉        | 1237/6470 [15:26<1:13:09,  1.19it/s]
2|1238|Loss: 0.9240529537200928:  19%|█▉        | 1238/6470 [15:27<1:13:16,  1.19it/s]
2|1239|Loss: 1.0038435459136963:  19%|█▉        | 1238/6470 [15:27<1:13:16,  1.19it/s]
2|1239|Loss: 1.0038435459136963:  19%|█▉        | 1239/6470 [15:27<1:15:21,  1.16it/s]
2|1240|Loss: 0.8786022067070007:  19%|█▉        | 1239/6470 [15:28<1:15:21,  1.16it/s]
2|1240|Loss: 0.8786022067070007:  19%|█▉        | 1240/6470 [15:28<1:14:53,  1.16it/s]
2|1241|Loss: 0.7679422497749329:  19%|█▉        | 1240/6470 [15:29<1:14:53,  1.16it/s]
2|1241|Loss: 0.7679422497749329:  19%|█▉        | 1241/6470 [15:29<1:18:08,  1.12it/s]
2|1242|Loss: 0.7436797618865967:  19%|█▉        | 1241/6470 [15:30<1:18:08,  1.12it/s]
2|1242|Loss: 0.7436797618865967:  19%|█▉        | 1242/6470 [15:30<1:14:15,  1.17it/s]
2|1243|Loss: 0.9290856122970581:  19%|█▉        | 1242/6470 [15:30<1:14:15,  1.17it/s]
2|1243|Loss: 0.9290856122970581:  19%|█▉        | 1243/6470 [15:31<1:09:52,  1.25it/s]
2|1244|Loss: 0.7423453330993652:  19%|█▉        | 1243/6470 [15:31<1:09:52,  1.25it/s]
2|1244|Loss: 0.7423453330993652:  19%|█▉        | 1244/6470 [15:31<1:05:34,  1.33it/s]
2|1245|Loss: 0.8597445487976074:  19%|█▉        | 1244/6470 [15:32<1:05:34,  1.33it/s]
2|1245|Loss: 0.8597445487976074:  19%|█▉        | 1245/6470 [15:32<1:03:56,  1.36it/s]
2|1246|Loss: 1.0376849174499512:  19%|█▉        | 1245/6470 [15:32<1:03:56,  1.36it/s]
2|1246|Loss: 1.0376849174499512:  19%|█▉        | 1246/6470 [15:33<1:00:26,  1.44it/s]
2|1247|Loss: 0.8504185080528259:  19%|█▉        | 1246/6470 [15:33<1:00:26,  1.44it/s]
2|1247|Loss: 0.8504185080528259:  19%|█▉        | 1247/6470 [15:33<58:04,  1.50it/s]  
2|1248|Loss: 0.8433386087417603:  19%|█▉        | 1247/6470 [15:34<58:04,  1.50it/s]
2|1248|Loss: 0.8433386087417603:  19%|█▉        | 1248/6470 [15:34<57:55,  1.50it/s]
2|1249|Loss: 0.8416632413864136:  19%|█▉        | 1248/6470 [15:34<57:55,  1.50it/s]
2|1249|Loss: 0.8416632413864136:  19%|█▉        | 1249/6470 [15:35<1:00:07,  1.45it/s]
2|1250|Loss: 0.8074445724487305:  19%|█▉        | 1249/6470 [15:35<1:00:07,  1.45it/s]
2|1250|Loss: 0.8074445724487305:  19%|█▉        | 1250/6470 [15:36<1:04:25,  1.35it/s]
2|1251|Loss: 0.9812264442443848:  19%|█▉        | 1250/6470 [15:36<1:04:25,  1.35it/s]
2|1251|Loss: 0.9812264442443848:  19%|█▉        | 1251/6470 [15:37<1:10:50,  1.23it/s]
2|1252|Loss: 0.8963991403579712:  19%|█▉        | 1251/6470 [15:37<1:10:50,  1.23it/s]
2|1252|Loss: 0.8963991403579712:  19%|█▉        | 1252/6470 [15:37<1:11:56,  1.21it/s]
2|1253|Loss: 0.8667882680892944:  19%|█▉        | 1252/6470 [15:38<1:11:56,  1.21it/s]
2|1253|Loss: 0.8667882680892944:  19%|█▉        | 1253/6470 [15:38<1:10:29,  1.23it/s]
2|1254|Loss: 0.8678537607192993:  19%|█▉        | 1253/6470 [15:39<1:10:29,  1.23it/s]
2|1254|Loss: 0.8678537607192993:  19%|█▉        | 1254/6470 [15:39<1:07:10,  1.29it/s]
2|1255|Loss: 0.7919096350669861:  19%|█▉        | 1254/6470 [15:39<1:07:10,  1.29it/s]
2|1255|Loss: 0.7919096350669861:  19%|█▉        | 1255/6470 [15:40<1:05:26,  1.33it/s]
2|1256|Loss: 0.8431901931762695:  19%|█▉        | 1255/6470 [15:40<1:05:26,  1.33it/s]
2|1256|Loss: 0.8431901931762695:  19%|█▉        | 1256/6470 [15:40<1:08:48,  1.26it/s]
2|1257|Loss: 0.9350321888923645:  19%|█▉        | 1256/6470 [15:41<1:08:48,  1.26it/s]
2|1257|Loss: 0.9350321888923645:  19%|█▉        | 1257/6470 [15:41<1:08:54,  1.26it/s]
2|1258|Loss: 0.8376456499099731:  19%|█▉        | 1257/6470 [15:42<1:08:54,  1.26it/s]
2|1258|Loss: 0.8376456499099731:  19%|█▉        | 1258/6470 [15:42<1:10:48,  1.23it/s]
2|1259|Loss: 0.7304596900939941:  19%|█▉        | 1258/6470 [15:42<1:10:48,  1.23it/s]
2|1259|Loss: 0.7304596900939941:  19%|█▉        | 1259/6470 [15:43<1:06:54,  1.30it/s]
2|1260|Loss: 0.913368284702301:  19%|█▉        | 1259/6470 [15:43<1:06:54,  1.30it/s] 
2|1260|Loss: 0.913368284702301:  19%|█▉        | 1260/6470 [15:43<1:03:08,  1.38it/s]
2|1261|Loss: 0.7899126410484314:  19%|█▉        | 1260/6470 [15:44<1:03:08,  1.38it/s]
2|1261|Loss: 0.7899126410484314:  19%|█▉        | 1261/6470 [15:44<1:00:24,  1.44it/s]
2|1262|Loss: 0.8858592510223389:  19%|█▉        | 1261/6470 [15:44<1:00:24,  1.44it/s]
2|1262|Loss: 0.8858592510223389:  20%|█▉        | 1262/6470 [15:45<1:01:35,  1.41it/s]
2|1263|Loss: 0.795115053653717:  20%|█▉        | 1262/6470 [15:45<1:01:35,  1.41it/s] 
2|1263|Loss: 0.795115053653717:  20%|█▉        | 1263/6470 [15:46<1:02:55,  1.38it/s]
2|1264|Loss: 0.7637646198272705:  20%|█▉        | 1263/6470 [15:46<1:02:55,  1.38it/s]
2|1264|Loss: 0.7637646198272705:  20%|█▉        | 1264/6470 [15:46<55:44,  1.56it/s]  
2|1265|Loss: 0.9941120743751526:  20%|█▉        | 1264/6470 [15:46<55:44,  1.56it/s]
2|1265|Loss: 0.9941120743751526:  20%|█▉        | 1265/6470 [15:47<55:38,  1.56it/s]
2|1266|Loss: 0.7947840094566345:  20%|█▉        | 1265/6470 [15:47<55:38,  1.56it/s]
2|1266|Loss: 0.7947840094566345:  20%|█▉        | 1266/6470 [15:47<57:38,  1.50it/s]
2|1267|Loss: 0.8385457992553711:  20%|█▉        | 1266/6470 [15:48<57:38,  1.50it/s]
2|1267|Loss: 0.8385457992553711:  20%|█▉        | 1267/6470 [15:48<58:14,  1.49it/s]
2|1268|Loss: 0.9022416472434998:  20%|█▉        | 1267/6470 [15:48<58:14,  1.49it/s]
2|1268|Loss: 0.9022416472434998:  20%|█▉        | 1268/6470 [15:49<1:00:51,  1.42it/s]
2|1269|Loss: 0.946692705154419:  20%|█▉        | 1268/6470 [15:49<1:00:51,  1.42it/s] 
2|1269|Loss: 0.946692705154419:  20%|█▉        | 1269/6470 [15:50<1:04:46,  1.34it/s]
2|1270|Loss: 0.8202508091926575:  20%|█▉        | 1269/6470 [15:50<1:04:46,  1.34it/s]
2|1270|Loss: 0.8202508091926575:  20%|█▉        | 1270/6470 [15:50<1:03:59,  1.35it/s]
2|1271|Loss: 0.8164228200912476:  20%|█▉        | 1270/6470 [15:51<1:03:59,  1.35it/s]
2|1271|Loss: 0.8164228200912476:  20%|█▉        | 1271/6470 [15:51<1:03:10,  1.37it/s]
2|1272|Loss: 0.881037175655365:  20%|█▉        | 1271/6470 [15:51<1:03:10,  1.37it/s] 
2|1272|Loss: 0.881037175655365:  20%|█▉        | 1272/6470 [15:52<1:04:31,  1.34it/s]
2|1273|Loss: 0.9902468323707581:  20%|█▉        | 1272/6470 [15:52<1:04:31,  1.34it/s]
2|1273|Loss: 0.9902468323707581:  20%|█▉        | 1273/6470 [15:53<1:02:27,  1.39it/s]
2|1274|Loss: 0.7645283341407776:  20%|█▉        | 1273/6470 [15:53<1:02:27,  1.39it/s]
2|1274|Loss: 0.7645283341407776:  20%|█▉        | 1274/6470 [15:53<1:02:34,  1.38it/s]
2|1275|Loss: 0.7197458148002625:  20%|█▉        | 1274/6470 [15:54<1:02:34,  1.38it/s]
2|1275|Loss: 0.7197458148002625:  20%|█▉        | 1275/6470 [15:54<1:02:38,  1.38it/s]
2|1276|Loss: 0.7942373156547546:  20%|█▉        | 1275/6470 [15:55<1:02:38,  1.38it/s]
2|1276|Loss: 0.7942373156547546:  20%|█▉        | 1276/6470 [15:55<1:11:53,  1.20it/s]
2|1277|Loss: 0.8690962195396423:  20%|█▉        | 1276/6470 [15:56<1:11:53,  1.20it/s]
2|1277|Loss: 0.8690962195396423:  20%|█▉        | 1277/6470 [15:56<1:12:10,  1.20it/s]
2|1278|Loss: 0.7941208481788635:  20%|█▉        | 1277/6470 [15:57<1:12:10,  1.20it/s]
2|1278|Loss: 0.7941208481788635:  20%|█▉        | 1278/6470 [15:57<1:24:37,  1.02it/s]
2|1279|Loss: 0.9613382816314697:  20%|█▉        | 1278/6470 [15:58<1:24:37,  1.02it/s]
2|1279|Loss: 0.9613382816314697:  20%|█▉        | 1279/6470 [15:58<1:18:07,  1.11it/s]
2|1280|Loss: 0.8530741930007935:  20%|█▉        | 1279/6470 [15:58<1:18:07,  1.11it/s]
2|1280|Loss: 0.8530741930007935:  20%|█▉        | 1280/6470 [15:59<1:15:01,  1.15it/s]
2|1281|Loss: 0.7640377283096313:  20%|█▉        | 1280/6470 [15:59<1:15:01,  1.15it/s]
2|1281|Loss: 0.7640377283096313:  20%|█▉        | 1281/6470 [15:59<1:10:49,  1.22it/s]
2|1282|Loss: 0.918695867061615:  20%|█▉        | 1281/6470 [16:00<1:10:49,  1.22it/s] 
2|1282|Loss: 0.918695867061615:  20%|█▉        | 1282/6470 [16:00<1:10:08,  1.23it/s]
2|1283|Loss: 0.9689244031906128:  20%|█▉        | 1282/6470 [16:01<1:10:08,  1.23it/s]
2|1283|Loss: 0.9689244031906128:  20%|█▉        | 1283/6470 [16:01<1:07:59,  1.27it/s]
2|1284|Loss: 0.8578304648399353:  20%|█▉        | 1283/6470 [16:01<1:07:59,  1.27it/s]
2|1284|Loss: 0.8578304648399353:  20%|█▉        | 1284/6470 [16:02<1:05:17,  1.32it/s]
2|1285|Loss: 0.8720733523368835:  20%|█▉        | 1284/6470 [16:02<1:05:17,  1.32it/s]
2|1285|Loss: 0.8720733523368835:  20%|█▉        | 1285/6470 [16:02<1:05:31,  1.32it/s]
2|1286|Loss: 0.9088747501373291:  20%|█▉        | 1285/6470 [16:03<1:05:31,  1.32it/s]
2|1286|Loss: 0.9088747501373291:  20%|█▉        | 1286/6470 [16:03<1:05:54,  1.31it/s]
2|1287|Loss: 0.7185407280921936:  20%|█▉        | 1286/6470 [16:04<1:05:54,  1.31it/s]
2|1287|Loss: 0.7185407280921936:  20%|█▉        | 1287/6470 [16:04<1:01:15,  1.41it/s]
2|1288|Loss: 0.8949992656707764:  20%|█▉        | 1287/6470 [16:04<1:01:15,  1.41it/s]
2|1288|Loss: 0.8949992656707764:  20%|█▉        | 1288/6470 [16:04<1:02:15,  1.39it/s]
2|1289|Loss: 0.7058069705963135:  20%|█▉        | 1288/6470 [16:05<1:02:15,  1.39it/s]
2|1289|Loss: 0.7058069705963135:  20%|█▉        | 1289/6470 [16:05<1:06:49,  1.29it/s]
2|1290|Loss: 0.9059694409370422:  20%|█▉        | 1289/6470 [16:06<1:06:49,  1.29it/s]
2|1290|Loss: 0.9059694409370422:  20%|█▉        | 1290/6470 [16:06<1:03:45,  1.35it/s]
2|1291|Loss: 0.8279961347579956:  20%|█▉        | 1290/6470 [16:06<1:03:45,  1.35it/s]
2|1291|Loss: 0.8279961347579956:  20%|█▉        | 1291/6470 [16:07<1:01:20,  1.41it/s]
2|1292|Loss: 0.8454310297966003:  20%|█▉        | 1291/6470 [16:07<1:01:20,  1.41it/s]
2|1292|Loss: 0.8454310297966003:  20%|█▉        | 1292/6470 [16:08<1:08:35,  1.26it/s]
2|1293|Loss: 0.7710696458816528:  20%|█▉        | 1292/6470 [16:08<1:08:35,  1.26it/s]
2|1293|Loss: 0.7710696458816528:  20%|█▉        | 1293/6470 [16:09<1:14:01,  1.17it/s]
2|1294|Loss: 0.9696229696273804:  20%|█▉        | 1293/6470 [16:09<1:14:01,  1.17it/s]
2|1294|Loss: 0.9696229696273804:  20%|██        | 1294/6470 [16:10<1:14:16,  1.16it/s]
2|1295|Loss: 0.9001619815826416:  20%|██        | 1294/6470 [16:10<1:14:16,  1.16it/s]
2|1295|Loss: 0.9001619815826416:  20%|██        | 1295/6470 [16:10<1:14:38,  1.16it/s]
2|1296|Loss: 0.8050509095191956:  20%|██        | 1295/6470 [16:11<1:14:38,  1.16it/s]
2|1296|Loss: 0.8050509095191956:  20%|██        | 1296/6470 [16:11<1:13:14,  1.18it/s]
2|1297|Loss: 0.9127741456031799:  20%|██        | 1296/6470 [16:12<1:13:14,  1.18it/s]
2|1297|Loss: 0.9127741456031799:  20%|██        | 1297/6470 [16:12<1:13:53,  1.17it/s]
2|1298|Loss: 0.7667354941368103:  20%|██        | 1297/6470 [16:13<1:13:53,  1.17it/s]
2|1298|Loss: 0.7667354941368103:  20%|██        | 1298/6470 [16:13<1:11:58,  1.20it/s]
2|1299|Loss: 0.9426871538162231:  20%|██        | 1298/6470 [16:13<1:11:58,  1.20it/s]
2|1299|Loss: 0.9426871538162231:  20%|██        | 1299/6470 [16:14<1:12:23,  1.19it/s]
2|1300|Loss: 0.9113069176673889:  20%|██        | 1299/6470 [16:14<1:12:23,  1.19it/s]
2|1300|Loss: 0.9113069176673889:  20%|██        | 1300/6470 [16:14<1:09:08,  1.25it/s]
2|1301|Loss: 0.7256545424461365:  20%|██        | 1300/6470 [16:15<1:09:08,  1.25it/s]
2|1301|Loss: 0.7256545424461365:  20%|██        | 1301/6470 [16:15<1:04:54,  1.33it/s]
2|1302|Loss: 0.9516746401786804:  20%|██        | 1301/6470 [16:15<1:04:54,  1.33it/s]
2|1302|Loss: 0.9516746401786804:  20%|██        | 1302/6470 [16:16<1:04:21,  1.34it/s]
2|1303|Loss: 0.7860041260719299:  20%|██        | 1302/6470 [16:16<1:04:21,  1.34it/s]
2|1303|Loss: 0.7860041260719299:  20%|██        | 1303/6470 [16:17<1:04:52,  1.33it/s]
2|1304|Loss: 0.9546735286712646:  20%|██        | 1303/6470 [16:17<1:04:52,  1.33it/s]
2|1304|Loss: 0.9546735286712646:  20%|██        | 1304/6470 [16:17<1:02:49,  1.37it/s]
2|1305|Loss: 1.0073015689849854:  20%|██        | 1304/6470 [16:18<1:02:49,  1.37it/s]
2|1305|Loss: 1.0073015689849854:  20%|██        | 1305/6470 [16:18<1:03:55,  1.35it/s]
2|1306|Loss: 0.7326235771179199:  20%|██        | 1305/6470 [16:18<1:03:55,  1.35it/s]
2|1306|Loss: 0.7326235771179199:  20%|██        | 1306/6470 [16:19<1:03:08,  1.36it/s]
2|1307|Loss: 0.8073152899742126:  20%|██        | 1306/6470 [16:19<1:03:08,  1.36it/s]
2|1307|Loss: 0.8073152899742126:  20%|██        | 1307/6470 [16:20<1:04:09,  1.34it/s]
2|1308|Loss: 0.898773729801178:  20%|██        | 1307/6470 [16:20<1:04:09,  1.34it/s] 
2|1308|Loss: 0.898773729801178:  20%|██        | 1308/6470 [16:20<1:08:06,  1.26it/s]
2|1309|Loss: 0.7631503343582153:  20%|██        | 1308/6470 [16:21<1:08:06,  1.26it/s]
2|1309|Loss: 0.7631503343582153:  20%|██        | 1309/6470 [16:21<1:07:16,  1.28it/s]
2|1310|Loss: 0.8329789042472839:  20%|██        | 1309/6470 [16:22<1:07:16,  1.28it/s]
2|1310|Loss: 0.8329789042472839:  20%|██        | 1310/6470 [16:22<1:07:11,  1.28it/s]
2|1311|Loss: 0.8108057379722595:  20%|██        | 1310/6470 [16:22<1:07:11,  1.28it/s]
2|1311|Loss: 0.8108057379722595:  20%|██        | 1311/6470 [16:23<1:07:04,  1.28it/s]
2|1312|Loss: 0.8483737111091614:  20%|██        | 1311/6470 [16:23<1:07:04,  1.28it/s]
2|1312|Loss: 0.8483737111091614:  20%|██        | 1312/6470 [16:24<1:06:35,  1.29it/s]
2|1313|Loss: 0.6360659599304199:  20%|██        | 1312/6470 [16:24<1:06:35,  1.29it/s]
2|1313|Loss: 0.6360659599304199:  20%|██        | 1313/6470 [16:24<1:06:40,  1.29it/s]
2|1314|Loss: 0.8817129731178284:  20%|██        | 1313/6470 [16:25<1:06:40,  1.29it/s]
2|1314|Loss: 0.8817129731178284:  20%|██        | 1314/6470 [16:25<1:00:33,  1.42it/s]
2|1315|Loss: 0.8752879500389099:  20%|██        | 1314/6470 [16:25<1:00:33,  1.42it/s]
2|1315|Loss: 0.8752879500389099:  20%|██        | 1315/6470 [16:25<58:33,  1.47it/s]  
2|1316|Loss: 0.9308913946151733:  20%|██        | 1315/6470 [16:26<58:33,  1.47it/s]
2|1316|Loss: 0.9308913946151733:  20%|██        | 1316/6470 [16:26<1:00:29,  1.42it/s]
2|1317|Loss: 0.8259553909301758:  20%|██        | 1316/6470 [16:27<1:00:29,  1.42it/s]
2|1317|Loss: 0.8259553909301758:  20%|██        | 1317/6470 [16:27<1:03:16,  1.36it/s]
2|1318|Loss: 0.9894343018531799:  20%|██        | 1317/6470 [16:27<1:03:16,  1.36it/s]
2|1318|Loss: 0.9894343018531799:  20%|██        | 1318/6470 [16:28<1:05:20,  1.31it/s]
2|1319|Loss: 0.7447514533996582:  20%|██        | 1318/6470 [16:28<1:05:20,  1.31it/s]
2|1319|Loss: 0.7447514533996582:  20%|██        | 1319/6470 [16:29<1:09:20,  1.24it/s]
2|1320|Loss: 0.734539270401001:  20%|██        | 1319/6470 [16:29<1:09:20,  1.24it/s] 
2|1320|Loss: 0.734539270401001:  20%|██        | 1320/6470 [16:30<1:13:29,  1.17it/s]
2|1321|Loss: 0.7823998332023621:  20%|██        | 1320/6470 [16:30<1:13:29,  1.17it/s]
2|1321|Loss: 0.7823998332023621:  20%|██        | 1321/6470 [16:31<1:11:22,  1.20it/s]
2|1322|Loss: 0.8260807394981384:  20%|██        | 1321/6470 [16:31<1:11:22,  1.20it/s]
2|1322|Loss: 0.8260807394981384:  20%|██        | 1322/6470 [16:31<1:07:52,  1.26it/s]
2|1323|Loss: 0.7823944687843323:  20%|██        | 1322/6470 [16:32<1:07:52,  1.26it/s]
2|1323|Loss: 0.7823944687843323:  20%|██        | 1323/6470 [16:32<1:10:39,  1.21it/s]
2|1324|Loss: 0.9312046766281128:  20%|██        | 1323/6470 [16:33<1:10:39,  1.21it/s]
2|1324|Loss: 0.9312046766281128:  20%|██        | 1324/6470 [16:33<1:10:17,  1.22it/s]
2|1325|Loss: 0.8344246745109558:  20%|██        | 1324/6470 [16:33<1:10:17,  1.22it/s]
2|1325|Loss: 0.8344246745109558:  20%|██        | 1325/6470 [16:34<1:09:27,  1.23it/s]
2|1326|Loss: 0.8453497886657715:  20%|██        | 1325/6470 [16:34<1:09:27,  1.23it/s]
2|1326|Loss: 0.8453497886657715:  20%|██        | 1326/6470 [16:34<1:08:09,  1.26it/s]
2|1327|Loss: 0.903390109539032:  20%|██        | 1326/6470 [16:35<1:08:09,  1.26it/s] 
2|1327|Loss: 0.903390109539032:  21%|██        | 1327/6470 [16:35<1:08:07,  1.26it/s]
2|1328|Loss: 0.8267471790313721:  21%|██        | 1327/6470 [16:36<1:08:07,  1.26it/s]
2|1328|Loss: 0.8267471790313721:  21%|██        | 1328/6470 [16:36<1:05:29,  1.31it/s]
2|1329|Loss: 0.9052622318267822:  21%|██        | 1328/6470 [16:36<1:05:29,  1.31it/s]
2|1329|Loss: 0.9052622318267822:  21%|██        | 1329/6470 [16:37<1:04:48,  1.32it/s]
2|1330|Loss: 0.8429386019706726:  21%|██        | 1329/6470 [16:37<1:04:48,  1.32it/s]
2|1330|Loss: 0.8429386019706726:  21%|██        | 1330/6470 [16:37<58:53,  1.45it/s]  
2|1331|Loss: 1.130988359451294:  21%|██        | 1330/6470 [16:38<58:53,  1.45it/s] 
2|1331|Loss: 1.130988359451294:  21%|██        | 1331/6470 [16:38<1:06:28,  1.29it/s]
2|1332|Loss: 0.9443455934524536:  21%|██        | 1331/6470 [16:39<1:06:28,  1.29it/s]
2|1332|Loss: 0.9443455934524536:  21%|██        | 1332/6470 [16:39<1:11:24,  1.20it/s]
2|1333|Loss: 0.9124971628189087:  21%|██        | 1332/6470 [16:40<1:11:24,  1.20it/s]
2|1333|Loss: 0.9124971628189087:  21%|██        | 1333/6470 [16:40<1:07:50,  1.26it/s]
2|1334|Loss: 0.7586278319358826:  21%|██        | 1333/6470 [16:40<1:07:50,  1.26it/s]
2|1334|Loss: 0.7586278319358826:  21%|██        | 1334/6470 [16:40<1:03:14,  1.35it/s]
2|1335|Loss: 0.7281852960586548:  21%|██        | 1334/6470 [16:41<1:03:14,  1.35it/s]
2|1335|Loss: 0.7281852960586548:  21%|██        | 1335/6470 [16:41<1:04:01,  1.34it/s]
2|1336|Loss: 0.8486013412475586:  21%|██        | 1335/6470 [16:42<1:04:01,  1.34it/s]
2|1336|Loss: 0.8486013412475586:  21%|██        | 1336/6470 [16:42<1:03:23,  1.35it/s]
2|1337|Loss: 0.7215206623077393:  21%|██        | 1336/6470 [16:42<1:03:23,  1.35it/s]
2|1337|Loss: 0.7215206623077393:  21%|██        | 1337/6470 [16:43<1:02:25,  1.37it/s]
2|1338|Loss: 0.8256285786628723:  21%|██        | 1337/6470 [16:43<1:02:25,  1.37it/s]
2|1338|Loss: 0.8256285786628723:  21%|██        | 1338/6470 [16:43<1:01:25,  1.39it/s]
2|1339|Loss: 0.7963517904281616:  21%|██        | 1338/6470 [16:44<1:01:25,  1.39it/s]
2|1339|Loss: 0.7963517904281616:  21%|██        | 1339/6470 [16:44<1:04:17,  1.33it/s]
2|1340|Loss: 0.7059723734855652:  21%|██        | 1339/6470 [16:45<1:04:17,  1.33it/s]
2|1340|Loss: 0.7059723734855652:  21%|██        | 1340/6470 [16:45<1:05:00,  1.32it/s]
2|1341|Loss: 0.8588936924934387:  21%|██        | 1340/6470 [16:45<1:05:00,  1.32it/s]
2|1341|Loss: 0.8588936924934387:  21%|██        | 1341/6470 [16:46<1:05:12,  1.31it/s]
2|1342|Loss: 0.9611501097679138:  21%|██        | 1341/6470 [16:46<1:05:12,  1.31it/s]
2|1342|Loss: 0.9611501097679138:  21%|██        | 1342/6470 [16:46<1:03:22,  1.35it/s]
2|1343|Loss: 0.7977114319801331:  21%|██        | 1342/6470 [16:47<1:03:22,  1.35it/s]
2|1343|Loss: 0.7977114319801331:  21%|██        | 1343/6470 [16:47<57:32,  1.49it/s]  
2|1344|Loss: 0.8469878435134888:  21%|██        | 1343/6470 [16:47<57:32,  1.49it/s]
2|1344|Loss: 0.8469878435134888:  21%|██        | 1344/6470 [16:48<58:44,  1.45it/s]
2|1345|Loss: 0.8557068705558777:  21%|██        | 1344/6470 [16:48<58:44,  1.45it/s]
2|1345|Loss: 0.8557068705558777:  21%|██        | 1345/6470 [16:48<59:37,  1.43it/s]
2|1346|Loss: 0.8142699003219604:  21%|██        | 1345/6470 [16:49<59:37,  1.43it/s]
2|1346|Loss: 0.8142699003219604:  21%|██        | 1346/6470 [16:49<55:59,  1.53it/s]
2|1347|Loss: 0.9478479027748108:  21%|██        | 1346/6470 [16:49<55:59,  1.53it/s]
2|1347|Loss: 0.9478479027748108:  21%|██        | 1347/6470 [16:50<55:30,  1.54it/s]
2|1348|Loss: 0.8773199915885925:  21%|██        | 1347/6470 [16:50<55:30,  1.54it/s]
2|1348|Loss: 0.8773199915885925:  21%|██        | 1348/6470 [16:50<1:02:13,  1.37it/s]
2|1349|Loss: 0.829182505607605:  21%|██        | 1348/6470 [16:51<1:02:13,  1.37it/s] 
2|1349|Loss: 0.829182505607605:  21%|██        | 1349/6470 [16:51<1:02:28,  1.37it/s]
2|1350|Loss: 1.0084794759750366:  21%|██        | 1349/6470 [16:52<1:02:28,  1.37it/s]
2|1350|Loss: 1.0084794759750366:  21%|██        | 1350/6470 [16:52<1:06:16,  1.29it/s]
2|1351|Loss: 0.8630890250205994:  21%|██        | 1350/6470 [16:53<1:06:16,  1.29it/s]
2|1351|Loss: 0.8630890250205994:  21%|██        | 1351/6470 [16:53<1:06:47,  1.28it/s]
2|1352|Loss: 0.8822846412658691:  21%|██        | 1351/6470 [16:53<1:06:47,  1.28it/s]
2|1352|Loss: 0.8822846412658691:  21%|██        | 1352/6470 [16:54<1:07:10,  1.27it/s]
2|1353|Loss: 1.0381819009780884:  21%|██        | 1352/6470 [16:54<1:07:10,  1.27it/s]
2|1353|Loss: 1.0381819009780884:  21%|██        | 1353/6470 [16:54<1:01:55,  1.38it/s]
2|1354|Loss: 0.7584782242774963:  21%|██        | 1353/6470 [16:55<1:01:55,  1.38it/s]
2|1354|Loss: 0.7584782242774963:  21%|██        | 1354/6470 [16:55<58:04,  1.47it/s]  
2|1355|Loss: 1.0309669971466064:  21%|██        | 1354/6470 [16:55<58:04,  1.47it/s]
2|1355|Loss: 1.0309669971466064:  21%|██        | 1355/6470 [16:55<55:43,  1.53it/s]
2|1356|Loss: 0.6720570921897888:  21%|██        | 1355/6470 [16:56<55:43,  1.53it/s]
2|1356|Loss: 0.6720570921897888:  21%|██        | 1356/6470 [16:56<58:55,  1.45it/s]
2|1357|Loss: 0.7496329545974731:  21%|██        | 1356/6470 [16:57<58:55,  1.45it/s]
2|1357|Loss: 0.7496329545974731:  21%|██        | 1357/6470 [16:57<1:01:26,  1.39it/s]
2|1358|Loss: 0.8133561611175537:  21%|██        | 1357/6470 [16:57<1:01:26,  1.39it/s]
2|1358|Loss: 0.8133561611175537:  21%|██        | 1358/6470 [16:58<1:02:42,  1.36it/s]
2|1359|Loss: 0.9262710213661194:  21%|██        | 1358/6470 [16:58<1:02:42,  1.36it/s]
2|1359|Loss: 0.9262710213661194:  21%|██        | 1359/6470 [16:58<59:57,  1.42it/s]  
2|1360|Loss: 0.874140202999115:  21%|██        | 1359/6470 [16:59<59:57,  1.42it/s] 
2|1360|Loss: 0.874140202999115:  21%|██        | 1360/6470 [16:59<59:15,  1.44it/s]
2|1361|Loss: 0.849017858505249:  21%|██        | 1360/6470 [17:00<59:15,  1.44it/s]
2|1361|Loss: 0.849017858505249:  21%|██        | 1361/6470 [17:00<1:04:23,  1.32it/s]
2|1362|Loss: 0.7510434985160828:  21%|██        | 1361/6470 [17:00<1:04:23,  1.32it/s]
2|1362|Loss: 0.7510434985160828:  21%|██        | 1362/6470 [17:01<1:01:32,  1.38it/s]
2|1363|Loss: 0.9103193879127502:  21%|██        | 1362/6470 [17:01<1:01:32,  1.38it/s]
2|1363|Loss: 0.9103193879127502:  21%|██        | 1363/6470 [17:01<58:25,  1.46it/s]  
2|1364|Loss: 0.8338977098464966:  21%|██        | 1363/6470 [17:02<58:25,  1.46it/s]
2|1364|Loss: 0.8338977098464966:  21%|██        | 1364/6470 [17:02<1:02:13,  1.37it/s]
2|1365|Loss: 0.7948682904243469:  21%|██        | 1364/6470 [17:03<1:02:13,  1.37it/s]
2|1365|Loss: 0.7948682904243469:  21%|██        | 1365/6470 [17:03<1:02:40,  1.36it/s]
2|1366|Loss: 0.7069256901741028:  21%|██        | 1365/6470 [17:03<1:02:40,  1.36it/s]
2|1366|Loss: 0.7069256901741028:  21%|██        | 1366/6470 [17:04<1:04:19,  1.32it/s]
2|1367|Loss: 0.7879437208175659:  21%|██        | 1366/6470 [17:04<1:04:19,  1.32it/s]
2|1367|Loss: 0.7879437208175659:  21%|██        | 1367/6470 [17:04<59:02,  1.44it/s]  
2|1368|Loss: 0.9394537806510925:  21%|██        | 1367/6470 [17:04<59:02,  1.44it/s]
2|1368|Loss: 0.9394537806510925:  21%|██        | 1368/6470 [17:05<53:32,  1.59it/s]
2|1369|Loss: 0.7673925757408142:  21%|██        | 1368/6470 [17:05<53:32,  1.59it/s]
2|1369|Loss: 0.7673925757408142:  21%|██        | 1369/6470 [17:05<57:51,  1.47it/s]
2|1370|Loss: 1.0344469547271729:  21%|██        | 1369/6470 [17:06<57:51,  1.47it/s]
2|1370|Loss: 1.0344469547271729:  21%|██        | 1370/6470 [17:06<1:03:09,  1.35it/s]
2|1371|Loss: 0.9071784615516663:  21%|██        | 1370/6470 [17:07<1:03:09,  1.35it/s]
2|1371|Loss: 0.9071784615516663:  21%|██        | 1371/6470 [17:07<1:04:26,  1.32it/s]
2|1372|Loss: 0.7665404677391052:  21%|██        | 1371/6470 [17:08<1:04:26,  1.32it/s]
2|1372|Loss: 0.7665404677391052:  21%|██        | 1372/6470 [17:08<1:04:35,  1.32it/s]
2|1373|Loss: 0.7214312553405762:  21%|██        | 1372/6470 [17:08<1:04:35,  1.32it/s]
2|1373|Loss: 0.7214312553405762:  21%|██        | 1373/6470 [17:09<1:01:03,  1.39it/s]
2|1374|Loss: 0.9264448285102844:  21%|██        | 1373/6470 [17:09<1:01:03,  1.39it/s]
2|1374|Loss: 0.9264448285102844:  21%|██        | 1374/6470 [17:09<56:22,  1.51it/s]  
2|1375|Loss: 0.6725292205810547:  21%|██        | 1374/6470 [17:09<56:22,  1.51it/s]
2|1375|Loss: 0.6725292205810547:  21%|██▏       | 1375/6470 [17:10<59:11,  1.43it/s]
2|1376|Loss: 0.8522325754165649:  21%|██▏       | 1375/6470 [17:10<59:11,  1.43it/s]
2|1376|Loss: 0.8522325754165649:  21%|██▏       | 1376/6470 [17:10<56:57,  1.49it/s]
2|1377|Loss: 0.7018590569496155:  21%|██▏       | 1376/6470 [17:11<56:57,  1.49it/s]
2|1377|Loss: 0.7018590569496155:  21%|██▏       | 1377/6470 [17:11<1:01:38,  1.38it/s]
2|1378|Loss: 0.8316795229911804:  21%|██▏       | 1377/6470 [17:12<1:01:38,  1.38it/s]
2|1378|Loss: 0.8316795229911804:  21%|██▏       | 1378/6470 [17:12<1:02:24,  1.36it/s]
2|1379|Loss: 0.8469566106796265:  21%|██▏       | 1378/6470 [17:12<1:02:24,  1.36it/s]
2|1379|Loss: 0.8469566106796265:  21%|██▏       | 1379/6470 [17:13<1:03:39,  1.33it/s]
2|1380|Loss: 0.7773391008377075:  21%|██▏       | 1379/6470 [17:13<1:03:39,  1.33it/s]
2|1380|Loss: 0.7773391008377075:  21%|██▏       | 1380/6470 [17:14<1:02:13,  1.36it/s]
2|1381|Loss: 0.8394041657447815:  21%|██▏       | 1380/6470 [17:14<1:02:13,  1.36it/s]
2|1381|Loss: 0.8394041657447815:  21%|██▏       | 1381/6470 [17:14<56:46,  1.49it/s]  
2|1382|Loss: 0.9051459431648254:  21%|██▏       | 1381/6470 [17:14<56:46,  1.49it/s]
2|1382|Loss: 0.9051459431648254:  21%|██▏       | 1382/6470 [17:15<56:02,  1.51it/s]
2|1383|Loss: 0.9263405799865723:  21%|██▏       | 1382/6470 [17:15<56:02,  1.51it/s]
2|1383|Loss: 0.9263405799865723:  21%|██▏       | 1383/6470 [17:16<1:00:41,  1.40it/s]
2|1384|Loss: 0.9390875101089478:  21%|██▏       | 1383/6470 [17:16<1:00:41,  1.40it/s]
2|1384|Loss: 0.9390875101089478:  21%|██▏       | 1384/6470 [17:16<1:05:28,  1.29it/s]
2|1385|Loss: 0.9841683506965637:  21%|██▏       | 1384/6470 [17:17<1:05:28,  1.29it/s]
2|1385|Loss: 0.9841683506965637:  21%|██▏       | 1385/6470 [17:18<1:14:19,  1.14it/s]
2|1386|Loss: 0.7866470813751221:  21%|██▏       | 1385/6470 [17:18<1:14:19,  1.14it/s]
2|1386|Loss: 0.7866470813751221:  21%|██▏       | 1386/6470 [17:18<1:10:53,  1.20it/s]
2|1387|Loss: 0.8823322653770447:  21%|██▏       | 1386/6470 [17:19<1:10:53,  1.20it/s]
2|1387|Loss: 0.8823322653770447:  21%|██▏       | 1387/6470 [17:19<1:11:47,  1.18it/s]
2|1388|Loss: 0.963306725025177:  21%|██▏       | 1387/6470 [17:20<1:11:47,  1.18it/s] 
2|1388|Loss: 0.963306725025177:  21%|██▏       | 1388/6470 [17:20<1:11:40,  1.18it/s]
2|1389|Loss: 0.9272447228431702:  21%|██▏       | 1388/6470 [17:20<1:11:40,  1.18it/s]
2|1389|Loss: 0.9272447228431702:  21%|██▏       | 1389/6470 [17:21<1:10:07,  1.21it/s]
2|1390|Loss: 0.7714884877204895:  21%|██▏       | 1389/6470 [17:21<1:10:07,  1.21it/s]
2|1390|Loss: 0.7714884877204895:  21%|██▏       | 1390/6470 [17:22<1:07:03,  1.26it/s]
2|1391|Loss: 0.9138607382774353:  21%|██▏       | 1390/6470 [17:22<1:07:03,  1.26it/s]
2|1391|Loss: 0.9138607382774353:  21%|██▏       | 1391/6470 [17:22<1:03:14,  1.34it/s]
2|1392|Loss: 0.9246527552604675:  21%|██▏       | 1391/6470 [17:23<1:03:14,  1.34it/s]
2|1392|Loss: 0.9246527552604675:  22%|██▏       | 1392/6470 [17:23<1:01:49,  1.37it/s]
2|1393|Loss: 0.7778476476669312:  22%|██▏       | 1392/6470 [17:23<1:01:49,  1.37it/s]
2|1393|Loss: 0.7778476476669312:  22%|██▏       | 1393/6470 [17:23<57:47,  1.46it/s]  
2|1394|Loss: 0.9157540202140808:  22%|██▏       | 1393/6470 [17:24<57:47,  1.46it/s]
2|1394|Loss: 0.9157540202140808:  22%|██▏       | 1394/6470 [17:24<1:00:33,  1.40it/s]
2|1395|Loss: 0.9501054883003235:  22%|██▏       | 1394/6470 [17:25<1:00:33,  1.40it/s]
2|1395|Loss: 0.9501054883003235:  22%|██▏       | 1395/6470 [17:25<1:03:47,  1.33it/s]
2|1396|Loss: 0.7546610236167908:  22%|██▏       | 1395/6470 [17:25<1:03:47,  1.33it/s]
2|1396|Loss: 0.7546610236167908:  22%|██▏       | 1396/6470 [17:26<1:02:27,  1.35it/s]
2|1397|Loss: 0.7700590491294861:  22%|██▏       | 1396/6470 [17:26<1:02:27,  1.35it/s]
2|1397|Loss: 0.7700590491294861:  22%|██▏       | 1397/6470 [17:26<58:24,  1.45it/s]  
2|1398|Loss: 0.7801844477653503:  22%|██▏       | 1397/6470 [17:27<58:24,  1.45it/s]
2|1398|Loss: 0.7801844477653503:  22%|██▏       | 1398/6470 [17:27<1:01:09,  1.38it/s]
2|1399|Loss: 0.718719482421875:  22%|██▏       | 1398/6470 [17:28<1:01:09,  1.38it/s] 
2|1399|Loss: 0.718719482421875:  22%|██▏       | 1399/6470 [17:28<58:22,  1.45it/s]  
2|1400|Loss: 0.7888211011886597:  22%|██▏       | 1399/6470 [17:28<58:22,  1.45it/s]
2|1400|Loss: 0.7888211011886597:  22%|██▏       | 1400/6470 [17:28<53:20,  1.58it/s]
2|1401|Loss: 0.8765637278556824:  22%|██▏       | 1400/6470 [17:29<53:20,  1.58it/s]
2|1401|Loss: 0.8765637278556824:  22%|██▏       | 1401/6470 [17:29<51:47,  1.63it/s]
2|1402|Loss: 0.8191318511962891:  22%|██▏       | 1401/6470 [17:29<51:47,  1.63it/s]
2|1402|Loss: 0.8191318511962891:  22%|██▏       | 1402/6470 [17:30<53:23,  1.58it/s]
2|1403|Loss: 0.9757763743400574:  22%|██▏       | 1402/6470 [17:30<53:23,  1.58it/s]
2|1403|Loss: 0.9757763743400574:  22%|██▏       | 1403/6470 [17:30<56:47,  1.49it/s]
2|1404|Loss: 0.9452961087226868:  22%|██▏       | 1403/6470 [17:31<56:47,  1.49it/s]
2|1404|Loss: 0.9452961087226868:  22%|██▏       | 1404/6470 [17:31<1:00:05,  1.41it/s]
2|1405|Loss: 0.8832005262374878:  22%|██▏       | 1404/6470 [17:31<1:00:05,  1.41it/s]
2|1405|Loss: 0.8832005262374878:  22%|██▏       | 1405/6470 [17:32<57:15,  1.47it/s]  
2|1406|Loss: 1.0037685632705688:  22%|██▏       | 1405/6470 [17:32<57:15,  1.47it/s]
2|1406|Loss: 1.0037685632705688:  22%|██▏       | 1406/6470 [17:32<52:33,  1.61it/s]
2|1407|Loss: 0.9153013229370117:  22%|██▏       | 1406/6470 [17:33<52:33,  1.61it/s]
2|1407|Loss: 0.9153013229370117:  22%|██▏       | 1407/6470 [17:33<54:22,  1.55it/s]
2|1408|Loss: 0.8294599056243896:  22%|██▏       | 1407/6470 [17:33<54:22,  1.55it/s]
2|1408|Loss: 0.8294599056243896:  22%|██▏       | 1408/6470 [17:34<54:43,  1.54it/s]
2|1409|Loss: 0.787049412727356:  22%|██▏       | 1408/6470 [17:34<54:43,  1.54it/s] 
2|1409|Loss: 0.787049412727356:  22%|██▏       | 1409/6470 [17:34<55:42,  1.51it/s]
2|1410|Loss: 0.948657214641571:  22%|██▏       | 1409/6470 [17:35<55:42,  1.51it/s]
2|1410|Loss: 0.948657214641571:  22%|██▏       | 1410/6470 [17:35<53:28,  1.58it/s]
2|1411|Loss: 0.8701255917549133:  22%|██▏       | 1410/6470 [17:35<53:28,  1.58it/s]
2|1411|Loss: 0.8701255917549133:  22%|██▏       | 1411/6470 [17:36<59:06,  1.43it/s]
2|1412|Loss: 0.993815541267395:  22%|██▏       | 1411/6470 [17:36<59:06,  1.43it/s] 
2|1412|Loss: 0.993815541267395:  22%|██▏       | 1412/6470 [17:36<57:14,  1.47it/s]
2|1413|Loss: 0.8753982186317444:  22%|██▏       | 1412/6470 [17:37<57:14,  1.47it/s]
2|1413|Loss: 0.8753982186317444:  22%|██▏       | 1413/6470 [17:37<55:11,  1.53it/s]
2|1414|Loss: 0.6452131271362305:  22%|██▏       | 1413/6470 [17:37<55:11,  1.53it/s]
2|1414|Loss: 0.6452131271362305:  22%|██▏       | 1414/6470 [17:38<56:02,  1.50it/s]
2|1415|Loss: 0.9700431227684021:  22%|██▏       | 1414/6470 [17:38<56:02,  1.50it/s]
2|1415|Loss: 0.9700431227684021:  22%|██▏       | 1415/6470 [17:38<59:18,  1.42it/s]
2|1416|Loss: 0.7714856863021851:  22%|██▏       | 1415/6470 [17:39<59:18,  1.42it/s]
2|1416|Loss: 0.7714856863021851:  22%|██▏       | 1416/6470 [17:39<56:05,  1.50it/s]
2|1417|Loss: 0.9148501753807068:  22%|██▏       | 1416/6470 [17:39<56:05,  1.50it/s]
2|1417|Loss: 0.9148501753807068:  22%|██▏       | 1417/6470 [17:40<57:23,  1.47it/s]
2|1418|Loss: 0.8459137678146362:  22%|██▏       | 1417/6470 [17:40<57:23,  1.47it/s]
2|1418|Loss: 0.8459137678146362:  22%|██▏       | 1418/6470 [17:41<1:01:41,  1.36it/s]
2|1419|Loss: 0.8578425049781799:  22%|██▏       | 1418/6470 [17:41<1:01:41,  1.36it/s]
2|1419|Loss: 0.8578425049781799:  22%|██▏       | 1419/6470 [17:41<1:03:08,  1.33it/s]
2|1420|Loss: 0.8016389012336731:  22%|██▏       | 1419/6470 [17:42<1:03:08,  1.33it/s]
2|1420|Loss: 0.8016389012336731:  22%|██▏       | 1420/6470 [17:42<1:04:00,  1.31it/s]
2|1421|Loss: 0.7089427709579468:  22%|██▏       | 1420/6470 [17:42<1:04:00,  1.31it/s]
2|1421|Loss: 0.7089427709579468:  22%|██▏       | 1421/6470 [17:43<1:02:24,  1.35it/s]
2|1422|Loss: 0.8963171243667603:  22%|██▏       | 1421/6470 [17:43<1:02:24,  1.35it/s]
2|1422|Loss: 0.8963171243667603:  22%|██▏       | 1422/6470 [17:44<1:05:33,  1.28it/s]
2|1423|Loss: 0.8962706327438354:  22%|██▏       | 1422/6470 [17:44<1:05:33,  1.28it/s]
2|1423|Loss: 0.8962706327438354:  22%|██▏       | 1423/6470 [17:44<1:06:08,  1.27it/s]
2|1424|Loss: 0.860907256603241:  22%|██▏       | 1423/6470 [17:45<1:06:08,  1.27it/s] 
2|1424|Loss: 0.860907256603241:  22%|██▏       | 1424/6470 [17:45<1:02:53,  1.34it/s]
2|1425|Loss: 0.8629747033119202:  22%|██▏       | 1424/6470 [17:45<1:02:53,  1.34it/s]
2|1425|Loss: 0.8629747033119202:  22%|██▏       | 1425/6470 [17:46<1:01:26,  1.37it/s]
2|1426|Loss: 0.8274722695350647:  22%|██▏       | 1425/6470 [17:46<1:01:26,  1.37it/s]
2|1426|Loss: 0.8274722695350647:  22%|██▏       | 1426/6470 [17:47<1:05:29,  1.28it/s]
2|1427|Loss: 0.8285015821456909:  22%|██▏       | 1426/6470 [17:47<1:05:29,  1.28it/s]
2|1427|Loss: 0.8285015821456909:  22%|██▏       | 1427/6470 [17:47<1:05:06,  1.29it/s]
2|1428|Loss: 0.9234139919281006:  22%|██▏       | 1427/6470 [17:48<1:05:06,  1.29it/s]
2|1428|Loss: 0.9234139919281006:  22%|██▏       | 1428/6470 [17:48<1:02:58,  1.33it/s]
2|1429|Loss: 0.9745066165924072:  22%|██▏       | 1428/6470 [17:48<1:02:58,  1.33it/s]
2|1429|Loss: 0.9745066165924072:  22%|██▏       | 1429/6470 [17:49<57:21,  1.46it/s]  
2|1430|Loss: 0.9536949992179871:  22%|██▏       | 1429/6470 [17:49<57:21,  1.46it/s]
2|1430|Loss: 0.9536949992179871:  22%|██▏       | 1430/6470 [17:49<58:45,  1.43it/s]
2|1431|Loss: 0.8649014234542847:  22%|██▏       | 1430/6470 [17:50<58:45,  1.43it/s]
2|1431|Loss: 0.8649014234542847:  22%|██▏       | 1431/6470 [17:50<1:04:26,  1.30it/s]
2|1432|Loss: 0.9152421951293945:  22%|██▏       | 1431/6470 [17:51<1:04:26,  1.30it/s]
2|1432|Loss: 0.9152421951293945:  22%|██▏       | 1432/6470 [17:51<1:05:22,  1.28it/s]
2|1433|Loss: 0.8739734292030334:  22%|██▏       | 1432/6470 [17:52<1:05:22,  1.28it/s]
2|1433|Loss: 0.8739734292030334:  22%|██▏       | 1433/6470 [17:52<1:03:14,  1.33it/s]
2|1434|Loss: 0.9052654504776001:  22%|██▏       | 1433/6470 [17:52<1:03:14,  1.33it/s]
2|1434|Loss: 0.9052654504776001:  22%|██▏       | 1434/6470 [17:52<1:00:58,  1.38it/s]
2|1435|Loss: 0.8556287288665771:  22%|██▏       | 1434/6470 [17:53<1:00:58,  1.38it/s]
2|1435|Loss: 0.8556287288665771:  22%|██▏       | 1435/6470 [17:53<1:03:26,  1.32it/s]
2|1436|Loss: 0.8492733240127563:  22%|██▏       | 1435/6470 [17:54<1:03:26,  1.32it/s]
2|1436|Loss: 0.8492733240127563:  22%|██▏       | 1436/6470 [17:54<1:04:59,  1.29it/s]
2|1437|Loss: 0.8575409650802612:  22%|██▏       | 1436/6470 [17:55<1:04:59,  1.29it/s]
2|1437|Loss: 0.8575409650802612:  22%|██▏       | 1437/6470 [17:55<1:10:08,  1.20it/s]
2|1438|Loss: 0.9061586856842041:  22%|██▏       | 1437/6470 [17:56<1:10:08,  1.20it/s]
2|1438|Loss: 0.9061586856842041:  22%|██▏       | 1438/6470 [17:56<1:09:41,  1.20it/s]
2|1439|Loss: 0.7781716585159302:  22%|██▏       | 1438/6470 [17:56<1:09:41,  1.20it/s]
2|1439|Loss: 0.7781716585159302:  22%|██▏       | 1439/6470 [17:57<1:07:30,  1.24it/s]
2|1440|Loss: 0.8981308341026306:  22%|██▏       | 1439/6470 [17:57<1:07:30,  1.24it/s]
2|1440|Loss: 0.8981308341026306:  22%|██▏       | 1440/6470 [17:57<1:06:15,  1.27it/s]
2|1441|Loss: 0.7600708603858948:  22%|██▏       | 1440/6470 [17:58<1:06:15,  1.27it/s]
2|1441|Loss: 0.7600708603858948:  22%|██▏       | 1441/6470 [17:58<1:05:29,  1.28it/s]
2|1442|Loss: 0.7767812609672546:  22%|██▏       | 1441/6470 [17:59<1:05:29,  1.28it/s]
2|1442|Loss: 0.7767812609672546:  22%|██▏       | 1442/6470 [17:59<1:07:39,  1.24it/s]
2|1443|Loss: 0.9412304759025574:  22%|██▏       | 1442/6470 [18:00<1:07:39,  1.24it/s]
2|1443|Loss: 0.9412304759025574:  22%|██▏       | 1443/6470 [18:00<1:09:13,  1.21it/s]
2|1444|Loss: 0.81334388256073:  22%|██▏       | 1443/6470 [18:00<1:09:13,  1.21it/s]  
2|1444|Loss: 0.81334388256073:  22%|██▏       | 1444/6470 [18:01<1:07:05,  1.25it/s]
2|1445|Loss: 0.8523330688476562:  22%|██▏       | 1444/6470 [18:01<1:07:05,  1.25it/s]
2|1445|Loss: 0.8523330688476562:  22%|██▏       | 1445/6470 [18:01<1:00:03,  1.39it/s]
2|1446|Loss: 0.8130257725715637:  22%|██▏       | 1445/6470 [18:02<1:00:03,  1.39it/s]
2|1446|Loss: 0.8130257725715637:  22%|██▏       | 1446/6470 [18:02<58:09,  1.44it/s]  
2|1447|Loss: 0.8460142016410828:  22%|██▏       | 1446/6470 [18:02<58:09,  1.44it/s]
2|1447|Loss: 0.8460142016410828:  22%|██▏       | 1447/6470 [18:03<59:47,  1.40it/s]
2|1448|Loss: 0.8354173302650452:  22%|██▏       | 1447/6470 [18:03<59:47,  1.40it/s]
2|1448|Loss: 0.8354173302650452:  22%|██▏       | 1448/6470 [18:04<1:04:31,  1.30it/s]
2|1449|Loss: 0.6860134601593018:  22%|██▏       | 1448/6470 [18:04<1:04:31,  1.30it/s]
2|1449|Loss: 0.6860134601593018:  22%|██▏       | 1449/6470 [18:04<1:03:40,  1.31it/s]
2|1450|Loss: 0.9335473775863647:  22%|██▏       | 1449/6470 [18:05<1:03:40,  1.31it/s]
2|1450|Loss: 0.9335473775863647:  22%|██▏       | 1450/6470 [18:05<58:53,  1.42it/s]  
2|1451|Loss: 0.7706798315048218:  22%|██▏       | 1450/6470 [18:05<58:53,  1.42it/s]
2|1451|Loss: 0.7706798315048218:  22%|██▏       | 1451/6470 [18:06<59:57,  1.40it/s]
2|1452|Loss: 0.9372418522834778:  22%|██▏       | 1451/6470 [18:06<59:57,  1.40it/s]
2|1452|Loss: 0.9372418522834778:  22%|██▏       | 1452/6470 [18:06<59:43,  1.40it/s]
2|1453|Loss: 0.9184006452560425:  22%|██▏       | 1452/6470 [18:07<59:43,  1.40it/s]
2|1453|Loss: 0.9184006452560425:  22%|██▏       | 1453/6470 [18:07<1:04:38,  1.29it/s]
2|1454|Loss: 0.8794853091239929:  22%|██▏       | 1453/6470 [18:08<1:04:38,  1.29it/s]
2|1454|Loss: 0.8794853091239929:  22%|██▏       | 1454/6470 [18:08<1:06:41,  1.25it/s]
2|1455|Loss: 0.8319731950759888:  22%|██▏       | 1454/6470 [18:08<1:06:41,  1.25it/s]
2|1455|Loss: 0.8319731950759888:  22%|██▏       | 1455/6470 [18:09<1:04:19,  1.30it/s]
2|1456|Loss: 0.8531666994094849:  22%|██▏       | 1455/6470 [18:09<1:04:19,  1.30it/s]
2|1456|Loss: 0.8531666994094849:  23%|██▎       | 1456/6470 [18:09<1:03:24,  1.32it/s]
2|1457|Loss: 0.9864467978477478:  23%|██▎       | 1456/6470 [18:10<1:03:24,  1.32it/s]
2|1457|Loss: 0.9864467978477478:  23%|██▎       | 1457/6470 [18:10<1:04:20,  1.30it/s]
2|1458|Loss: 0.9237334728240967:  23%|██▎       | 1457/6470 [18:11<1:04:20,  1.30it/s]
2|1458|Loss: 0.9237334728240967:  23%|██▎       | 1458/6470 [18:11<1:05:46,  1.27it/s]
2|1459|Loss: 0.9201263785362244:  23%|██▎       | 1458/6470 [18:11<1:05:46,  1.27it/s]
2|1459|Loss: 0.9201263785362244:  23%|██▎       | 1459/6470 [18:12<1:00:38,  1.38it/s]
2|1460|Loss: 0.7489009499549866:  23%|██▎       | 1459/6470 [18:12<1:00:38,  1.38it/s]
2|1460|Loss: 0.7489009499549866:  23%|██▎       | 1460/6470 [18:13<1:05:32,  1.27it/s]
2|1461|Loss: 0.7770138382911682:  23%|██▎       | 1460/6470 [18:13<1:05:32,  1.27it/s]
2|1461|Loss: 0.7770138382911682:  23%|██▎       | 1461/6470 [18:13<1:01:22,  1.36it/s]
2|1462|Loss: 0.8964386582374573:  23%|██▎       | 1461/6470 [18:14<1:01:22,  1.36it/s]
2|1462|Loss: 0.8964386582374573:  23%|██▎       | 1462/6470 [18:14<1:00:41,  1.38it/s]
2|1463|Loss: 0.8213236331939697:  23%|██▎       | 1462/6470 [18:14<1:00:41,  1.38it/s]
2|1463|Loss: 0.8213236331939697:  23%|██▎       | 1463/6470 [18:15<1:04:36,  1.29it/s]
2|1464|Loss: 0.9606539011001587:  23%|██▎       | 1463/6470 [18:15<1:04:36,  1.29it/s]
2|1464|Loss: 0.9606539011001587:  23%|██▎       | 1464/6470 [18:16<1:05:26,  1.27it/s]
2|1465|Loss: 0.9484105110168457:  23%|██▎       | 1464/6470 [18:16<1:05:26,  1.27it/s]
2|1465|Loss: 0.9484105110168457:  23%|██▎       | 1465/6470 [18:16<1:05:47,  1.27it/s]
2|1466|Loss: 0.9091688990592957:  23%|██▎       | 1465/6470 [18:17<1:05:47,  1.27it/s]
2|1466|Loss: 0.9091688990592957:  23%|██▎       | 1466/6470 [18:17<1:06:03,  1.26it/s]
2|1467|Loss: 0.9378183484077454:  23%|██▎       | 1466/6470 [18:18<1:06:03,  1.26it/s]
2|1467|Loss: 0.9378183484077454:  23%|██▎       | 1467/6470 [18:18<1:04:27,  1.29it/s]
2|1468|Loss: 0.7951026558876038:  23%|██▎       | 1467/6470 [18:18<1:04:27,  1.29it/s]
2|1468|Loss: 0.7951026558876038:  23%|██▎       | 1468/6470 [18:19<1:01:17,  1.36it/s]
2|1469|Loss: 0.7394147515296936:  23%|██▎       | 1468/6470 [18:19<1:01:17,  1.36it/s]
2|1469|Loss: 0.7394147515296936:  23%|██▎       | 1469/6470 [18:19<59:35,  1.40it/s]  
2|1470|Loss: 0.7700057625770569:  23%|██▎       | 1469/6470 [18:20<59:35,  1.40it/s]
2|1470|Loss: 0.7700057625770569:  23%|██▎       | 1470/6470 [18:20<59:23,  1.40it/s]
2|1471|Loss: 0.9674561619758606:  23%|██▎       | 1470/6470 [18:20<59:23,  1.40it/s]
2|1471|Loss: 0.9674561619758606:  23%|██▎       | 1471/6470 [18:21<1:01:18,  1.36it/s]
2|1472|Loss: 0.8797878623008728:  23%|██▎       | 1471/6470 [18:21<1:01:18,  1.36it/s]
2|1472|Loss: 0.8797878623008728:  23%|██▎       | 1472/6470 [18:22<1:05:36,  1.27it/s]
2|1473|Loss: 0.7331293821334839:  23%|██▎       | 1472/6470 [18:22<1:05:36,  1.27it/s]
2|1473|Loss: 0.7331293821334839:  23%|██▎       | 1473/6470 [18:22<1:05:43,  1.27it/s]
2|1474|Loss: 0.8817699551582336:  23%|██▎       | 1473/6470 [18:23<1:05:43,  1.27it/s]
2|1474|Loss: 0.8817699551582336:  23%|██▎       | 1474/6470 [18:23<59:25,  1.40it/s]  
2|1475|Loss: 0.9357232451438904:  23%|██▎       | 1474/6470 [18:23<59:25,  1.40it/s]
2|1475|Loss: 0.9357232451438904:  23%|██▎       | 1475/6470 [18:24<55:13,  1.51it/s]
2|1476|Loss: 0.9059740304946899:  23%|██▎       | 1475/6470 [18:24<55:13,  1.51it/s]
2|1476|Loss: 0.9059740304946899:  23%|██▎       | 1476/6470 [18:24<55:04,  1.51it/s]
2|1477|Loss: 0.9337977766990662:  23%|██▎       | 1476/6470 [18:25<55:04,  1.51it/s]
2|1477|Loss: 0.9337977766990662:  23%|██▎       | 1477/6470 [18:25<1:00:23,  1.38it/s]
2|1478|Loss: 0.8678421974182129:  23%|██▎       | 1477/6470 [18:26<1:00:23,  1.38it/s]
2|1478|Loss: 0.8678421974182129:  23%|██▎       | 1478/6470 [18:26<1:02:13,  1.34it/s]
2|1479|Loss: 0.8619547486305237:  23%|██▎       | 1478/6470 [18:26<1:02:13,  1.34it/s]
2|1479|Loss: 0.8619547486305237:  23%|██▎       | 1479/6470 [18:27<1:08:33,  1.21it/s]
2|1480|Loss: 0.9286231398582458:  23%|██▎       | 1479/6470 [18:27<1:08:33,  1.21it/s]
2|1480|Loss: 0.9286231398582458:  23%|██▎       | 1480/6470 [18:27<1:02:50,  1.32it/s]
2|1481|Loss: 0.8296844363212585:  23%|██▎       | 1480/6470 [18:28<1:02:50,  1.32it/s]
2|1481|Loss: 0.8296844363212585:  23%|██▎       | 1481/6470 [18:28<1:04:30,  1.29it/s]
2|1482|Loss: 0.8191073536872864:  23%|██▎       | 1481/6470 [18:29<1:04:30,  1.29it/s]
2|1482|Loss: 0.8191073536872864:  23%|██▎       | 1482/6470 [18:29<1:04:41,  1.29it/s]
2|1483|Loss: 0.909748911857605:  23%|██▎       | 1482/6470 [18:30<1:04:41,  1.29it/s] 
2|1483|Loss: 0.909748911857605:  23%|██▎       | 1483/6470 [18:30<1:05:23,  1.27it/s]
2|1484|Loss: 0.8260058164596558:  23%|██▎       | 1483/6470 [18:30<1:05:23,  1.27it/s]
2|1484|Loss: 0.8260058164596558:  23%|██▎       | 1484/6470 [18:31<1:05:08,  1.28it/s]
2|1485|Loss: 0.7359150052070618:  23%|██▎       | 1484/6470 [18:31<1:05:08,  1.28it/s]
2|1485|Loss: 0.7359150052070618:  23%|██▎       | 1485/6470 [18:31<1:03:21,  1.31it/s]
2|1486|Loss: 0.8182217478752136:  23%|██▎       | 1485/6470 [18:32<1:03:21,  1.31it/s]
2|1486|Loss: 0.8182217478752136:  23%|██▎       | 1486/6470 [18:32<1:05:21,  1.27it/s]
2|1487|Loss: 0.8243701457977295:  23%|██▎       | 1486/6470 [18:33<1:05:21,  1.27it/s]
2|1487|Loss: 0.8243701457977295:  23%|██▎       | 1487/6470 [18:33<1:01:37,  1.35it/s]
2|1488|Loss: 1.0048366785049438:  23%|██▎       | 1487/6470 [18:33<1:01:37,  1.35it/s]
2|1488|Loss: 1.0048366785049438:  23%|██▎       | 1488/6470 [18:34<59:43,  1.39it/s]  
2|1489|Loss: 0.8936070203781128:  23%|██▎       | 1488/6470 [18:34<59:43,  1.39it/s]
2|1489|Loss: 0.8936070203781128:  23%|██▎       | 1489/6470 [18:34<1:03:23,  1.31it/s]
2|1490|Loss: 0.7470216155052185:  23%|██▎       | 1489/6470 [18:35<1:03:23,  1.31it/s]
2|1490|Loss: 0.7470216155052185:  23%|██▎       | 1490/6470 [18:35<1:02:47,  1.32it/s]
2|1491|Loss: 0.8878320455551147:  23%|██▎       | 1490/6470 [18:36<1:02:47,  1.32it/s]
2|1491|Loss: 0.8878320455551147:  23%|██▎       | 1491/6470 [18:36<1:06:09,  1.25it/s]
2|1492|Loss: 0.8721609115600586:  23%|██▎       | 1491/6470 [18:37<1:06:09,  1.25it/s]
2|1492|Loss: 0.8721609115600586:  23%|██▎       | 1492/6470 [18:37<1:07:12,  1.23it/s]
2|1493|Loss: 0.7544860243797302:  23%|██▎       | 1492/6470 [18:37<1:07:12,  1.23it/s]
2|1493|Loss: 0.7544860243797302:  23%|██▎       | 1493/6470 [18:38<1:08:59,  1.20it/s]
2|1494|Loss: 0.7569452524185181:  23%|██▎       | 1493/6470 [18:38<1:08:59,  1.20it/s]
2|1494|Loss: 0.7569452524185181:  23%|██▎       | 1494/6470 [18:38<1:06:48,  1.24it/s]
2|1495|Loss: 1.3169429302215576:  23%|██▎       | 1494/6470 [18:39<1:06:48,  1.24it/s]
2|1495|Loss: 1.3169429302215576:  23%|██▎       | 1495/6470 [18:39<1:06:22,  1.25it/s]
2|1496|Loss: 0.8491658568382263:  23%|██▎       | 1495/6470 [18:40<1:06:22,  1.25it/s]
2|1496|Loss: 0.8491658568382263:  23%|██▎       | 1496/6470 [18:40<1:01:50,  1.34it/s]
2|1497|Loss: 0.6369519233703613:  23%|██▎       | 1496/6470 [18:41<1:01:50,  1.34it/s]
2|1497|Loss: 0.6369519233703613:  23%|██▎       | 1497/6470 [18:41<1:20:38,  1.03it/s]
2|1498|Loss: 0.7721690535545349:  23%|██▎       | 1497/6470 [18:42<1:20:38,  1.03it/s]
2|1498|Loss: 0.7721690535545349:  23%|██▎       | 1498/6470 [18:42<1:19:30,  1.04it/s]
2|1499|Loss: 0.8522841930389404:  23%|██▎       | 1498/6470 [18:43<1:19:30,  1.04it/s]
2|1499|Loss: 0.8522841930389404:  23%|██▎       | 1499/6470 [18:43<1:18:01,  1.06it/s]
2|1500|Loss: 0.8498137593269348:  23%|██▎       | 1499/6470 [18:44<1:18:01,  1.06it/s]
2|1500|Loss: 0.8498137593269348:  23%|██▎       | 1500/6470 [18:44<1:15:37,  1.10it/s]
2|1501|Loss: 0.8585647344589233:  23%|██▎       | 1500/6470 [18:44<1:15:37,  1.10it/s]
2|1501|Loss: 0.8585647344589233:  23%|██▎       | 1501/6470 [18:45<1:10:19,  1.18it/s]
2|1502|Loss: 0.8107236623764038:  23%|██▎       | 1501/6470 [18:45<1:10:19,  1.18it/s]
2|1502|Loss: 0.8107236623764038:  23%|██▎       | 1502/6470 [18:46<1:12:57,  1.13it/s]
2|1503|Loss: 0.9508084058761597:  23%|██▎       | 1502/6470 [18:46<1:12:57,  1.13it/s]
2|1503|Loss: 0.9508084058761597:  23%|██▎       | 1503/6470 [18:47<1:14:46,  1.11it/s]
2|1504|Loss: 0.8578361868858337:  23%|██▎       | 1503/6470 [18:47<1:14:46,  1.11it/s]
2|1504|Loss: 0.8578361868858337:  23%|██▎       | 1504/6470 [18:48<1:15:12,  1.10it/s]
2|1505|Loss: 0.7123288512229919:  23%|██▎       | 1504/6470 [18:48<1:15:12,  1.10it/s]
2|1505|Loss: 0.7123288512229919:  23%|██▎       | 1505/6470 [18:49<1:15:22,  1.10it/s]
2|1506|Loss: 0.8526583909988403:  23%|██▎       | 1505/6470 [18:49<1:15:22,  1.10it/s]
2|1506|Loss: 0.8526583909988403:  23%|██▎       | 1506/6470 [18:49<1:12:10,  1.15it/s]
2|1507|Loss: 0.8039266467094421:  23%|██▎       | 1506/6470 [18:50<1:12:10,  1.15it/s]
2|1507|Loss: 0.8039266467094421:  23%|██▎       | 1507/6470 [18:50<1:07:01,  1.23it/s]
2|1508|Loss: 0.8500306606292725:  23%|██▎       | 1507/6470 [18:50<1:07:01,  1.23it/s]
2|1508|Loss: 0.8500306606292725:  23%|██▎       | 1508/6470 [18:51<1:09:25,  1.19it/s]
2|1509|Loss: 0.8088482022285461:  23%|██▎       | 1508/6470 [18:51<1:09:25,  1.19it/s]
2|1509|Loss: 0.8088482022285461:  23%|██▎       | 1509/6470 [18:52<1:11:59,  1.15it/s]
2|1510|Loss: 0.9360398054122925:  23%|██▎       | 1509/6470 [18:52<1:11:59,  1.15it/s]
2|1510|Loss: 0.9360398054122925:  23%|██▎       | 1510/6470 [18:53<1:10:21,  1.17it/s]
2|1511|Loss: 0.8217160105705261:  23%|██▎       | 1510/6470 [18:53<1:10:21,  1.17it/s]
2|1511|Loss: 0.8217160105705261:  23%|██▎       | 1511/6470 [18:53<1:08:56,  1.20it/s]
2|1512|Loss: 0.9357243776321411:  23%|██▎       | 1511/6470 [18:54<1:08:56,  1.20it/s]
2|1512|Loss: 0.9357243776321411:  23%|██▎       | 1512/6470 [18:54<1:10:21,  1.17it/s]
2|1513|Loss: 0.8570821285247803:  23%|██▎       | 1512/6470 [18:55<1:10:21,  1.17it/s]
2|1513|Loss: 0.8570821285247803:  23%|██▎       | 1513/6470 [18:55<1:14:40,  1.11it/s]
2|1514|Loss: 0.8409914970397949:  23%|██▎       | 1513/6470 [18:56<1:14:40,  1.11it/s]
2|1514|Loss: 0.8409914970397949:  23%|██▎       | 1514/6470 [18:56<1:16:55,  1.07it/s]
2|1515|Loss: 0.9203270077705383:  23%|██▎       | 1514/6470 [18:57<1:16:55,  1.07it/s]
2|1515|Loss: 0.9203270077705383:  23%|██▎       | 1515/6470 [18:57<1:16:29,  1.08it/s]
2|1516|Loss: 0.9001350402832031:  23%|██▎       | 1515/6470 [18:58<1:16:29,  1.08it/s]
2|1516|Loss: 0.9001350402832031:  23%|██▎       | 1516/6470 [18:58<1:13:11,  1.13it/s]
2|1517|Loss: 0.8056660890579224:  23%|██▎       | 1516/6470 [18:58<1:13:11,  1.13it/s]
2|1517|Loss: 0.8056660890579224:  23%|██▎       | 1517/6470 [18:59<1:06:48,  1.24it/s]
2|1518|Loss: 0.9271814227104187:  23%|██▎       | 1517/6470 [18:59<1:06:48,  1.24it/s]
2|1518|Loss: 0.9271814227104187:  23%|██▎       | 1518/6470 [18:59<1:04:47,  1.27it/s]
2|1519|Loss: 0.8948512077331543:  23%|██▎       | 1518/6470 [19:00<1:04:47,  1.27it/s]
2|1519|Loss: 0.8948512077331543:  23%|██▎       | 1519/6470 [19:00<1:05:21,  1.26it/s]
2|1520|Loss: 0.9370521903038025:  23%|██▎       | 1519/6470 [19:01<1:05:21,  1.26it/s]
2|1520|Loss: 0.9370521903038025:  23%|██▎       | 1520/6470 [19:01<1:00:41,  1.36it/s]
2|1521|Loss: 1.0611720085144043:  23%|██▎       | 1520/6470 [19:01<1:00:41,  1.36it/s]
2|1521|Loss: 1.0611720085144043:  24%|██▎       | 1521/6470 [19:02<1:00:18,  1.37it/s]
2|1522|Loss: 0.9837962985038757:  24%|██▎       | 1521/6470 [19:02<1:00:18,  1.37it/s]
2|1522|Loss: 0.9837962985038757:  24%|██▎       | 1522/6470 [19:02<1:02:17,  1.32it/s]
2|1523|Loss: 0.8739511370658875:  24%|██▎       | 1522/6470 [19:03<1:02:17,  1.32it/s]
2|1523|Loss: 0.8739511370658875:  24%|██▎       | 1523/6470 [19:03<58:29,  1.41it/s]  
2|1524|Loss: 0.7180181741714478:  24%|██▎       | 1523/6470 [19:03<58:29,  1.41it/s]
2|1524|Loss: 0.7180181741714478:  24%|██▎       | 1524/6470 [19:04<1:01:53,  1.33it/s]
2|1525|Loss: 0.905157744884491:  24%|██▎       | 1524/6470 [19:04<1:01:53,  1.33it/s] 
2|1525|Loss: 0.905157744884491:  24%|██▎       | 1525/6470 [19:05<1:03:48,  1.29it/s]
2|1526|Loss: 0.9046790599822998:  24%|██▎       | 1525/6470 [19:05<1:03:48,  1.29it/s]
2|1526|Loss: 0.9046790599822998:  24%|██▎       | 1526/6470 [19:05<1:01:27,  1.34it/s]
2|1527|Loss: 0.886988639831543:  24%|██▎       | 1526/6470 [19:06<1:01:27,  1.34it/s] 
2|1527|Loss: 0.886988639831543:  24%|██▎       | 1527/6470 [19:06<1:02:02,  1.33it/s]
2|1528|Loss: 0.7212547659873962:  24%|██▎       | 1527/6470 [19:06<1:02:02,  1.33it/s]
2|1528|Loss: 0.7212547659873962:  24%|██▎       | 1528/6470 [19:07<1:01:30,  1.34it/s]
2|1529|Loss: 0.7996441125869751:  24%|██▎       | 1528/6470 [19:07<1:01:30,  1.34it/s]
2|1529|Loss: 0.7996441125869751:  24%|██▎       | 1529/6470 [19:08<1:02:40,  1.31it/s]
2|1530|Loss: 0.7972049117088318:  24%|██▎       | 1529/6470 [19:08<1:02:40,  1.31it/s]
2|1530|Loss: 0.7972049117088318:  24%|██▎       | 1530/6470 [19:08<58:47,  1.40it/s]  
2|1531|Loss: 0.8567561507225037:  24%|██▎       | 1530/6470 [19:09<58:47,  1.40it/s]
2|1531|Loss: 0.8567561507225037:  24%|██▎       | 1531/6470 [19:09<56:04,  1.47it/s]
2|1532|Loss: 0.8973305225372314:  24%|██▎       | 1531/6470 [19:09<56:04,  1.47it/s]
2|1532|Loss: 0.8973305225372314:  24%|██▎       | 1532/6470 [19:10<56:28,  1.46it/s]
2|1533|Loss: 0.8810299634933472:  24%|██▎       | 1532/6470 [19:10<56:28,  1.46it/s]
2|1533|Loss: 0.8810299634933472:  24%|██▎       | 1533/6470 [19:10<1:01:34,  1.34it/s]
2|1534|Loss: 0.9255974292755127:  24%|██▎       | 1533/6470 [19:11<1:01:34,  1.34it/s]
2|1534|Loss: 0.9255974292755127:  24%|██▎       | 1534/6470 [19:11<1:06:00,  1.25it/s]
2|1535|Loss: 1.0099897384643555:  24%|██▎       | 1534/6470 [19:12<1:06:00,  1.25it/s]
2|1535|Loss: 1.0099897384643555:  24%|██▎       | 1535/6470 [19:12<1:05:49,  1.25it/s]
2|1536|Loss: 0.98395174741745:  24%|██▎       | 1535/6470 [19:13<1:05:49,  1.25it/s]  
2|1536|Loss: 0.98395174741745:  24%|██▎       | 1536/6470 [19:13<1:04:02,  1.28it/s]
2|1537|Loss: 0.7776334881782532:  24%|██▎       | 1536/6470 [19:13<1:04:02,  1.28it/s]
2|1537|Loss: 0.7776334881782532:  24%|██▍       | 1537/6470 [19:14<1:03:29,  1.30it/s]
2|1538|Loss: 0.915207028388977:  24%|██▍       | 1537/6470 [19:14<1:03:29,  1.30it/s] 
2|1538|Loss: 0.915207028388977:  24%|██▍       | 1538/6470 [19:14<1:03:10,  1.30it/s]
2|1539|Loss: 0.7323045134544373:  24%|██▍       | 1538/6470 [19:15<1:03:10,  1.30it/s]
2|1539|Loss: 0.7323045134544373:  24%|██▍       | 1539/6470 [19:15<1:02:00,  1.33it/s]
2|1540|Loss: 0.697644054889679:  24%|██▍       | 1539/6470 [19:16<1:02:00,  1.33it/s] 
2|1540|Loss: 0.697644054889679:  24%|██▍       | 1540/6470 [19:16<1:08:10,  1.21it/s]
2|1541|Loss: 0.7863267064094543:  24%|██▍       | 1540/6470 [19:17<1:08:10,  1.21it/s]
2|1541|Loss: 0.7863267064094543:  24%|██▍       | 1541/6470 [19:17<1:10:59,  1.16it/s]
2|1542|Loss: 0.8246229290962219:  24%|██▍       | 1541/6470 [19:17<1:10:59,  1.16it/s]
2|1542|Loss: 0.8246229290962219:  24%|██▍       | 1542/6470 [19:18<1:07:57,  1.21it/s]
2|1543|Loss: 0.7476122379302979:  24%|██▍       | 1542/6470 [19:18<1:07:57,  1.21it/s]
2|1543|Loss: 0.7476122379302979:  24%|██▍       | 1543/6470 [19:19<1:06:45,  1.23it/s]
2|1544|Loss: 0.8384225368499756:  24%|██▍       | 1543/6470 [19:19<1:06:45,  1.23it/s]
2|1544|Loss: 0.8384225368499756:  24%|██▍       | 1544/6470 [19:19<1:06:44,  1.23it/s]
2|1545|Loss: 0.7803295850753784:  24%|██▍       | 1544/6470 [19:20<1:06:44,  1.23it/s]
2|1545|Loss: 0.7803295850753784:  24%|██▍       | 1545/6470 [19:20<1:03:45,  1.29it/s]
2|1546|Loss: 0.9624164700508118:  24%|██▍       | 1545/6470 [19:20<1:03:45,  1.29it/s]
2|1546|Loss: 0.9624164700508118:  24%|██▍       | 1546/6470 [19:21<1:02:52,  1.31it/s]
2|1547|Loss: 0.6510897874832153:  24%|██▍       | 1546/6470 [19:21<1:02:52,  1.31it/s]
2|1547|Loss: 0.6510897874832153:  24%|██▍       | 1547/6470 [19:21<1:00:02,  1.37it/s]
2|1548|Loss: 0.7621353268623352:  24%|██▍       | 1547/6470 [19:22<1:00:02,  1.37it/s]
2|1548|Loss: 0.7621353268623352:  24%|██▍       | 1548/6470 [19:22<58:07,  1.41it/s]  
2|1549|Loss: 0.8513980507850647:  24%|██▍       | 1548/6470 [19:22<58:07,  1.41it/s]
2|1549|Loss: 0.8513980507850647:  24%|██▍       | 1549/6470 [19:23<54:09,  1.51it/s]
2|1550|Loss: 0.8581206202507019:  24%|██▍       | 1549/6470 [19:23<54:09,  1.51it/s]
2|1550|Loss: 0.8581206202507019:  24%|██▍       | 1550/6470 [19:23<51:51,  1.58it/s]
2|1551|Loss: 0.8736265301704407:  24%|██▍       | 1550/6470 [19:24<51:51,  1.58it/s]
2|1551|Loss: 0.8736265301704407:  24%|██▍       | 1551/6470 [19:24<50:42,  1.62it/s]
2|1552|Loss: 0.7112932801246643:  24%|██▍       | 1551/6470 [19:24<50:42,  1.62it/s]
2|1552|Loss: 0.7112932801246643:  24%|██▍       | 1552/6470 [19:24<51:19,  1.60it/s]
2|1553|Loss: 0.8218810558319092:  24%|██▍       | 1552/6470 [19:25<51:19,  1.60it/s]
2|1553|Loss: 0.8218810558319092:  24%|██▍       | 1553/6470 [19:25<51:48,  1.58it/s]
2|1554|Loss: 0.9230855107307434:  24%|██▍       | 1553/6470 [19:26<51:48,  1.58it/s]
2|1554|Loss: 0.9230855107307434:  24%|██▍       | 1554/6470 [19:26<57:26,  1.43it/s]
2|1555|Loss: 0.9240995049476624:  24%|██▍       | 1554/6470 [19:26<57:26,  1.43it/s]
2|1555|Loss: 0.9240995049476624:  24%|██▍       | 1555/6470 [19:27<58:33,  1.40it/s]
2|1556|Loss: 0.9454497694969177:  24%|██▍       | 1555/6470 [19:27<58:33,  1.40it/s]
2|1556|Loss: 0.9454497694969177:  24%|██▍       | 1556/6470 [19:27<59:13,  1.38it/s]
2|1557|Loss: 0.8730390071868896:  24%|██▍       | 1556/6470 [19:28<59:13,  1.38it/s]
2|1557|Loss: 0.8730390071868896:  24%|██▍       | 1557/6470 [19:28<56:18,  1.45it/s]
2|1558|Loss: 0.8976779580116272:  24%|██▍       | 1557/6470 [19:28<56:18,  1.45it/s]
2|1558|Loss: 0.8976779580116272:  24%|██▍       | 1558/6470 [19:29<52:14,  1.57it/s]
2|1559|Loss: 0.8912177085876465:  24%|██▍       | 1558/6470 [19:29<52:14,  1.57it/s]
2|1559|Loss: 0.8912177085876465:  24%|██▍       | 1559/6470 [19:29<50:38,  1.62it/s]
2|1560|Loss: 0.7910908460617065:  24%|██▍       | 1559/6470 [19:30<50:38,  1.62it/s]
2|1560|Loss: 0.7910908460617065:  24%|██▍       | 1560/6470 [19:30<50:37,  1.62it/s]
2|1561|Loss: 0.6340093612670898:  24%|██▍       | 1560/6470 [19:30<50:37,  1.62it/s]
2|1561|Loss: 0.6340093612670898:  24%|██▍       | 1561/6470 [19:31<59:33,  1.37it/s]
2|1562|Loss: 0.7195863127708435:  24%|██▍       | 1561/6470 [19:31<59:33,  1.37it/s]
2|1562|Loss: 0.7195863127708435:  24%|██▍       | 1562/6470 [19:31<59:05,  1.38it/s]
2|1563|Loss: 0.8160800337791443:  24%|██▍       | 1562/6470 [19:32<59:05,  1.38it/s]
2|1563|Loss: 0.8160800337791443:  24%|██▍       | 1563/6470 [19:32<55:51,  1.46it/s]
2|1564|Loss: 0.9811987280845642:  24%|██▍       | 1563/6470 [19:32<55:51,  1.46it/s]
2|1564|Loss: 0.9811987280845642:  24%|██▍       | 1564/6470 [19:33<57:58,  1.41it/s]
2|1565|Loss: 0.8485323190689087:  24%|██▍       | 1564/6470 [19:33<57:58,  1.41it/s]
2|1565|Loss: 0.8485323190689087:  24%|██▍       | 1565/6470 [19:34<59:56,  1.36it/s]
2|1566|Loss: 0.751870334148407:  24%|██▍       | 1565/6470 [19:34<59:56,  1.36it/s] 
2|1566|Loss: 0.751870334148407:  24%|██▍       | 1566/6470 [19:34<53:51,  1.52it/s]
2|1567|Loss: 0.8138720989227295:  24%|██▍       | 1566/6470 [19:34<53:51,  1.52it/s]
2|1567|Loss: 0.8138720989227295:  24%|██▍       | 1567/6470 [19:35<54:32,  1.50it/s]
2|1568|Loss: 1.0850110054016113:  24%|██▍       | 1567/6470 [19:35<54:32,  1.50it/s]
2|1568|Loss: 1.0850110054016113:  24%|██▍       | 1568/6470 [19:36<56:56,  1.43it/s]
2|1569|Loss: 0.886743426322937:  24%|██▍       | 1568/6470 [19:36<56:56,  1.43it/s] 
2|1569|Loss: 0.886743426322937:  24%|██▍       | 1569/6470 [19:36<55:25,  1.47it/s]
2|1570|Loss: 0.848261833190918:  24%|██▍       | 1569/6470 [19:37<55:25,  1.47it/s]
2|1570|Loss: 0.848261833190918:  24%|██▍       | 1570/6470 [19:37<55:31,  1.47it/s]
2|1571|Loss: 0.7400758862495422:  24%|██▍       | 1570/6470 [19:37<55:31,  1.47it/s]
2|1571|Loss: 0.7400758862495422:  24%|██▍       | 1571/6470 [19:38<57:12,  1.43it/s]
2|1572|Loss: 0.8955448269844055:  24%|██▍       | 1571/6470 [19:38<57:12,  1.43it/s]
2|1572|Loss: 0.8955448269844055:  24%|██▍       | 1572/6470 [19:38<59:13,  1.38it/s]
2|1573|Loss: 0.9170830845832825:  24%|██▍       | 1572/6470 [19:39<59:13,  1.38it/s]
2|1573|Loss: 0.9170830845832825:  24%|██▍       | 1573/6470 [19:39<1:03:45,  1.28it/s]
2|1574|Loss: 0.8879538774490356:  24%|██▍       | 1573/6470 [19:40<1:03:45,  1.28it/s]
2|1574|Loss: 0.8879538774490356:  24%|██▍       | 1574/6470 [19:40<1:05:29,  1.25it/s]
2|1575|Loss: 0.7928912043571472:  24%|██▍       | 1574/6470 [19:41<1:05:29,  1.25it/s]
2|1575|Loss: 0.7928912043571472:  24%|██▍       | 1575/6470 [19:41<1:05:34,  1.24it/s]
2|1576|Loss: 0.7862203121185303:  24%|██▍       | 1575/6470 [19:41<1:05:34,  1.24it/s]
2|1576|Loss: 0.7862203121185303:  24%|██▍       | 1576/6470 [19:42<1:04:33,  1.26it/s]
2|1577|Loss: 0.9843176007270813:  24%|██▍       | 1576/6470 [19:42<1:04:33,  1.26it/s]
2|1577|Loss: 0.9843176007270813:  24%|██▍       | 1577/6470 [19:42<1:02:06,  1.31it/s]
2|1578|Loss: 0.8655152320861816:  24%|██▍       | 1577/6470 [19:43<1:02:06,  1.31it/s]
2|1578|Loss: 0.8655152320861816:  24%|██▍       | 1578/6470 [19:43<59:50,  1.36it/s]  
2|1579|Loss: 1.0000267028808594:  24%|██▍       | 1578/6470 [19:43<59:50,  1.36it/s]
2|1579|Loss: 1.0000267028808594:  24%|██▍       | 1579/6470 [19:44<56:55,  1.43it/s]
2|1580|Loss: 0.9206796288490295:  24%|██▍       | 1579/6470 [19:44<56:55,  1.43it/s]
2|1580|Loss: 0.9206796288490295:  24%|██▍       | 1580/6470 [19:44<55:34,  1.47it/s]
2|1581|Loss: 0.851477324962616:  24%|██▍       | 1580/6470 [19:45<55:34,  1.47it/s] 
2|1581|Loss: 0.851477324962616:  24%|██▍       | 1581/6470 [19:45<57:50,  1.41it/s]
2|1582|Loss: 0.8949135541915894:  24%|██▍       | 1581/6470 [19:46<57:50,  1.41it/s]
2|1582|Loss: 0.8949135541915894:  24%|██▍       | 1582/6470 [19:46<1:00:00,  1.36it/s]
2|1583|Loss: 0.8323962688446045:  24%|██▍       | 1582/6470 [19:46<1:00:00,  1.36it/s]
2|1583|Loss: 0.8323962688446045:  24%|██▍       | 1583/6470 [19:47<1:06:19,  1.23it/s]
2|1584|Loss: 0.7750513553619385:  24%|██▍       | 1583/6470 [19:47<1:06:19,  1.23it/s]
2|1584|Loss: 0.7750513553619385:  24%|██▍       | 1584/6470 [19:48<1:07:14,  1.21it/s]
2|1585|Loss: 0.9280882477760315:  24%|██▍       | 1584/6470 [19:48<1:07:14,  1.21it/s]
2|1585|Loss: 0.9280882477760315:  24%|██▍       | 1585/6470 [19:48<1:00:20,  1.35it/s]
2|1586|Loss: 1.0250762701034546:  24%|██▍       | 1585/6470 [19:49<1:00:20,  1.35it/s]
2|1586|Loss: 1.0250762701034546:  25%|██▍       | 1586/6470 [19:49<56:30,  1.44it/s]  
2|1587|Loss: 0.9567476511001587:  25%|██▍       | 1586/6470 [19:49<56:30,  1.44it/s]
2|1587|Loss: 0.9567476511001587:  25%|██▍       | 1587/6470 [19:50<55:48,  1.46it/s]
2|1588|Loss: 0.9105234742164612:  25%|██▍       | 1587/6470 [19:50<55:48,  1.46it/s]
2|1588|Loss: 0.9105234742164612:  25%|██▍       | 1588/6470 [19:50<58:05,  1.40it/s]
2|1589|Loss: 0.9672247767448425:  25%|██▍       | 1588/6470 [19:51<58:05,  1.40it/s]
2|1589|Loss: 0.9672247767448425:  25%|██▍       | 1589/6470 [19:51<1:02:43,  1.30it/s]
2|1590|Loss: 0.9849934577941895:  25%|██▍       | 1589/6470 [19:52<1:02:43,  1.30it/s]
2|1590|Loss: 0.9849934577941895:  25%|██▍       | 1590/6470 [19:52<1:01:40,  1.32it/s]
2|1591|Loss: 0.8000513911247253:  25%|██▍       | 1590/6470 [19:52<1:01:40,  1.32it/s]
2|1591|Loss: 0.8000513911247253:  25%|██▍       | 1591/6470 [19:53<1:06:57,  1.21it/s]
2|1592|Loss: 0.7801823616027832:  25%|██▍       | 1591/6470 [19:53<1:06:57,  1.21it/s]
2|1592|Loss: 0.7801823616027832:  25%|██▍       | 1592/6470 [19:54<1:07:23,  1.21it/s]
2|1593|Loss: 0.828682005405426:  25%|██▍       | 1592/6470 [19:54<1:07:23,  1.21it/s] 
2|1593|Loss: 0.828682005405426:  25%|██▍       | 1593/6470 [19:55<1:06:49,  1.22it/s]
2|1594|Loss: 0.9783573746681213:  25%|██▍       | 1593/6470 [19:55<1:06:49,  1.22it/s]
2|1594|Loss: 0.9783573746681213:  25%|██▍       | 1594/6470 [19:55<1:04:27,  1.26it/s]
2|1595|Loss: 0.9223198890686035:  25%|██▍       | 1594/6470 [19:56<1:04:27,  1.26it/s]
2|1595|Loss: 0.9223198890686035:  25%|██▍       | 1595/6470 [19:56<1:05:44,  1.24it/s]
2|1596|Loss: 0.9541283845901489:  25%|██▍       | 1595/6470 [19:57<1:05:44,  1.24it/s]
2|1596|Loss: 0.9541283845901489:  25%|██▍       | 1596/6470 [19:57<1:01:57,  1.31it/s]
2|1597|Loss: 0.8554390668869019:  25%|██▍       | 1596/6470 [19:57<1:01:57,  1.31it/s]
2|1597|Loss: 0.8554390668869019:  25%|██▍       | 1597/6470 [19:58<1:01:03,  1.33it/s]
2|1598|Loss: 0.889438271522522:  25%|██▍       | 1597/6470 [19:58<1:01:03,  1.33it/s] 
2|1598|Loss: 0.889438271522522:  25%|██▍       | 1598/6470 [19:58<59:56,  1.35it/s]  
2|1599|Loss: 0.8290325999259949:  25%|██▍       | 1598/6470 [19:59<59:56,  1.35it/s]
2|1599|Loss: 0.8290325999259949:  25%|██▍       | 1599/6470 [19:59<1:05:53,  1.23it/s]
2|1600|Loss: 0.9811206459999084:  25%|██▍       | 1599/6470 [20:00<1:05:53,  1.23it/s]
2|1600|Loss: 0.9811206459999084:  25%|██▍       | 1600/6470 [20:00<1:04:37,  1.26it/s]
2|1601|Loss: 0.7953523397445679:  25%|██▍       | 1600/6470 [20:00<1:04:37,  1.26it/s]
2|1601|Loss: 0.7953523397445679:  25%|██▍       | 1601/6470 [20:01<1:03:59,  1.27it/s]
2|1602|Loss: 0.9145857691764832:  25%|██▍       | 1601/6470 [20:01<1:03:59,  1.27it/s]
2|1602|Loss: 0.9145857691764832:  25%|██▍       | 1602/6470 [20:02<1:12:12,  1.12it/s]
2|1603|Loss: 0.8577239513397217:  25%|██▍       | 1602/6470 [20:02<1:12:12,  1.12it/s]
2|1603|Loss: 0.8577239513397217:  25%|██▍       | 1603/6470 [20:03<1:14:50,  1.08it/s]
2|1604|Loss: 0.7470902800559998:  25%|██▍       | 1603/6470 [20:03<1:14:50,  1.08it/s]
2|1604|Loss: 0.7470902800559998:  25%|██▍       | 1604/6470 [20:04<1:12:21,  1.12it/s]
2|1605|Loss: 0.8519757390022278:  25%|██▍       | 1604/6470 [20:04<1:12:21,  1.12it/s]
2|1605|Loss: 0.8519757390022278:  25%|██▍       | 1605/6470 [20:05<1:11:52,  1.13it/s]
2|1606|Loss: 0.9256975054740906:  25%|██▍       | 1605/6470 [20:05<1:11:52,  1.13it/s]
2|1606|Loss: 0.9256975054740906:  25%|██▍       | 1606/6470 [20:05<1:08:09,  1.19it/s]
2|1607|Loss: 1.0123200416564941:  25%|██▍       | 1606/6470 [20:06<1:08:09,  1.19it/s]
2|1607|Loss: 1.0123200416564941:  25%|██▍       | 1607/6470 [20:06<1:04:54,  1.25it/s]
2|1608|Loss: 0.778154730796814:  25%|██▍       | 1607/6470 [20:07<1:04:54,  1.25it/s] 
2|1608|Loss: 0.778154730796814:  25%|██▍       | 1608/6470 [20:07<1:07:07,  1.21it/s]
2|1609|Loss: 0.8606698513031006:  25%|██▍       | 1608/6470 [20:08<1:07:07,  1.21it/s]
2|1609|Loss: 0.8606698513031006:  25%|██▍       | 1609/6470 [20:08<1:08:00,  1.19it/s]
2|1610|Loss: 0.7793949246406555:  25%|██▍       | 1609/6470 [20:08<1:08:00,  1.19it/s]
2|1610|Loss: 0.7793949246406555:  25%|██▍       | 1610/6470 [20:08<1:03:58,  1.27it/s]
2|1611|Loss: 0.7497479319572449:  25%|██▍       | 1610/6470 [20:09<1:03:58,  1.27it/s]
2|1611|Loss: 0.7497479319572449:  25%|██▍       | 1611/6470 [20:09<1:01:05,  1.33it/s]
2|1612|Loss: 0.8179854154586792:  25%|██▍       | 1611/6470 [20:10<1:01:05,  1.33it/s]
2|1612|Loss: 0.8179854154586792:  25%|██▍       | 1612/6470 [20:10<57:26,  1.41it/s]  
2|1613|Loss: 0.7843848466873169:  25%|██▍       | 1612/6470 [20:10<57:26,  1.41it/s]
2|1613|Loss: 0.7843848466873169:  25%|██▍       | 1613/6470 [20:10<52:21,  1.55it/s]
2|1614|Loss: 0.9802666306495667:  25%|██▍       | 1613/6470 [20:11<52:21,  1.55it/s]
2|1614|Loss: 0.9802666306495667:  25%|██▍       | 1614/6470 [20:11<52:11,  1.55it/s]
2|1615|Loss: 0.8557002544403076:  25%|██▍       | 1614/6470 [20:11<52:11,  1.55it/s]
2|1615|Loss: 0.8557002544403076:  25%|██▍       | 1615/6470 [20:12<54:37,  1.48it/s]
2|1616|Loss: 0.7386408448219299:  25%|██▍       | 1615/6470 [20:12<54:37,  1.48it/s]
2|1616|Loss: 0.7386408448219299:  25%|██▍       | 1616/6470 [20:12<52:58,  1.53it/s]
2|1617|Loss: 0.6699267029762268:  25%|██▍       | 1616/6470 [20:13<52:58,  1.53it/s]
2|1617|Loss: 0.6699267029762268:  25%|██▍       | 1617/6470 [20:13<57:40,  1.40it/s]
2|1618|Loss: 0.8290948271751404:  25%|██▍       | 1617/6470 [20:14<57:40,  1.40it/s]
2|1618|Loss: 0.8290948271751404:  25%|██▌       | 1618/6470 [20:14<1:01:09,  1.32it/s]
2|1619|Loss: 0.7852947115898132:  25%|██▌       | 1618/6470 [20:14<1:01:09,  1.32it/s]
2|1619|Loss: 0.7852947115898132:  25%|██▌       | 1619/6470 [20:14<54:30,  1.48it/s]  
2|1620|Loss: 0.9470089077949524:  25%|██▌       | 1619/6470 [20:15<54:30,  1.48it/s]
2|1620|Loss: 0.9470089077949524:  25%|██▌       | 1620/6470 [20:15<54:42,  1.48it/s]
2|1621|Loss: 0.8769587278366089:  25%|██▌       | 1620/6470 [20:16<54:42,  1.48it/s]
2|1621|Loss: 0.8769587278366089:  25%|██▌       | 1621/6470 [20:16<57:51,  1.40it/s]
2|1622|Loss: 0.7872502207756042:  25%|██▌       | 1621/6470 [20:16<57:51,  1.40it/s]
2|1622|Loss: 0.7872502207756042:  25%|██▌       | 1622/6470 [20:17<56:35,  1.43it/s]
2|1623|Loss: 0.7581456303596497:  25%|██▌       | 1622/6470 [20:17<56:35,  1.43it/s]
2|1623|Loss: 0.7581456303596497:  25%|██▌       | 1623/6470 [20:17<53:09,  1.52it/s]
2|1624|Loss: 0.7320873141288757:  25%|██▌       | 1623/6470 [20:18<53:09,  1.52it/s]
2|1624|Loss: 0.7320873141288757:  25%|██▌       | 1624/6470 [20:18<59:30,  1.36it/s]
2|1625|Loss: 0.9103005528450012:  25%|██▌       | 1624/6470 [20:19<59:30,  1.36it/s]
2|1625|Loss: 0.9103005528450012:  25%|██▌       | 1625/6470 [20:19<1:03:41,  1.27it/s]
2|1626|Loss: 0.7494873404502869:  25%|██▌       | 1625/6470 [20:19<1:03:41,  1.27it/s]
2|1626|Loss: 0.7494873404502869:  25%|██▌       | 1626/6470 [20:20<1:04:24,  1.25it/s]
2|1627|Loss: 0.9413048624992371:  25%|██▌       | 1626/6470 [20:20<1:04:24,  1.25it/s]
2|1627|Loss: 0.9413048624992371:  25%|██▌       | 1627/6470 [20:21<1:06:14,  1.22it/s]
2|1628|Loss: 0.9461610317230225:  25%|██▌       | 1627/6470 [20:21<1:06:14,  1.22it/s]
2|1628|Loss: 0.9461610317230225:  25%|██▌       | 1628/6470 [20:22<1:06:40,  1.21it/s]
2|1629|Loss: 0.8615311980247498:  25%|██▌       | 1628/6470 [20:22<1:06:40,  1.21it/s]
2|1629|Loss: 0.8615311980247498:  25%|██▌       | 1629/6470 [20:23<1:11:25,  1.13it/s]
2|1630|Loss: 0.8846672773361206:  25%|██▌       | 1629/6470 [20:23<1:11:25,  1.13it/s]
2|1630|Loss: 0.8846672773361206:  25%|██▌       | 1630/6470 [20:23<1:10:54,  1.14it/s]
2|1631|Loss: 0.9330040216445923:  25%|██▌       | 1630/6470 [20:24<1:10:54,  1.14it/s]
2|1631|Loss: 0.9330040216445923:  25%|██▌       | 1631/6470 [20:24<1:08:44,  1.17it/s]
2|1632|Loss: 0.8138643503189087:  25%|██▌       | 1631/6470 [20:25<1:08:44,  1.17it/s]
2|1632|Loss: 0.8138643503189087:  25%|██▌       | 1632/6470 [20:25<1:09:22,  1.16it/s]
2|1633|Loss: 0.915363073348999:  25%|██▌       | 1632/6470 [20:26<1:09:22,  1.16it/s] 
2|1633|Loss: 0.915363073348999:  25%|██▌       | 1633/6470 [20:26<1:12:15,  1.12it/s]
2|1634|Loss: 0.8500258326530457:  25%|██▌       | 1633/6470 [20:26<1:12:15,  1.12it/s]
2|1634|Loss: 0.8500258326530457:  25%|██▌       | 1634/6470 [20:27<1:07:49,  1.19it/s]
2|1635|Loss: 0.8489340543746948:  25%|██▌       | 1634/6470 [20:27<1:07:49,  1.19it/s]
2|1635|Loss: 0.8489340543746948:  25%|██▌       | 1635/6470 [20:27<1:03:13,  1.27it/s]
2|1636|Loss: 0.7711135149002075:  25%|██▌       | 1635/6470 [20:28<1:03:13,  1.27it/s]
2|1636|Loss: 0.7711135149002075:  25%|██▌       | 1636/6470 [20:28<58:26,  1.38it/s]  
2|1637|Loss: 1.4167449474334717:  25%|██▌       | 1636/6470 [20:28<58:26,  1.38it/s]
2|1637|Loss: 1.4167449474334717:  25%|██▌       | 1637/6470 [20:29<58:48,  1.37it/s]
2|1638|Loss: 0.9904571771621704:  25%|██▌       | 1637/6470 [20:29<58:48,  1.37it/s]
2|1638|Loss: 0.9904571771621704:  25%|██▌       | 1638/6470 [20:29<58:37,  1.37it/s]
2|1639|Loss: 0.9938706159591675:  25%|██▌       | 1638/6470 [20:30<58:37,  1.37it/s]
2|1639|Loss: 0.9938706159591675:  25%|██▌       | 1639/6470 [20:30<57:12,  1.41it/s]
2|1640|Loss: 0.8847473859786987:  25%|██▌       | 1639/6470 [20:31<57:12,  1.41it/s]
2|1640|Loss: 0.8847473859786987:  25%|██▌       | 1640/6470 [20:31<57:49,  1.39it/s]
2|1641|Loss: 1.0159006118774414:  25%|██▌       | 1640/6470 [20:31<57:49,  1.39it/s]
2|1641|Loss: 1.0159006118774414:  25%|██▌       | 1641/6470 [20:32<1:00:01,  1.34it/s]
2|1642|Loss: 0.9568203091621399:  25%|██▌       | 1641/6470 [20:32<1:00:01,  1.34it/s]
2|1642|Loss: 0.9568203091621399:  25%|██▌       | 1642/6470 [20:32<59:35,  1.35it/s]  
2|1643|Loss: 0.9836530685424805:  25%|██▌       | 1642/6470 [20:33<59:35,  1.35it/s]
2|1643|Loss: 0.9836530685424805:  25%|██▌       | 1643/6470 [20:33<1:00:02,  1.34it/s]
2|1644|Loss: 1.0100176334381104:  25%|██▌       | 1643/6470 [20:34<1:00:02,  1.34it/s]
2|1644|Loss: 1.0100176334381104:  25%|██▌       | 1644/6470 [20:34<58:24,  1.38it/s]  
2|1645|Loss: 1.057183027267456:  25%|██▌       | 1644/6470 [20:34<58:24,  1.38it/s] 
2|1645|Loss: 1.057183027267456:  25%|██▌       | 1645/6470 [20:35<1:01:35,  1.31it/s]
2|1646|Loss: 0.8000792264938354:  25%|██▌       | 1645/6470 [20:35<1:01:35,  1.31it/s]
2|1646|Loss: 0.8000792264938354:  25%|██▌       | 1646/6470 [20:35<59:23,  1.35it/s]  
2|1647|Loss: 0.8469094634056091:  25%|██▌       | 1646/6470 [20:36<59:23,  1.35it/s]
2|1647|Loss: 0.8469094634056091:  25%|██▌       | 1647/6470 [20:36<57:54,  1.39it/s]
2|1648|Loss: 0.9091899991035461:  25%|██▌       | 1647/6470 [20:36<57:54,  1.39it/s]
2|1648|Loss: 0.9091899991035461:  25%|██▌       | 1648/6470 [20:37<51:23,  1.56it/s]
2|1649|Loss: 0.8213289976119995:  25%|██▌       | 1648/6470 [20:37<51:23,  1.56it/s]
2|1649|Loss: 0.8213289976119995:  25%|██▌       | 1649/6470 [20:37<55:14,  1.45it/s]
2|1650|Loss: 0.8859682679176331:  25%|██▌       | 1649/6470 [20:38<55:14,  1.45it/s]
2|1650|Loss: 0.8859682679176331:  26%|██▌       | 1650/6470 [20:38<53:57,  1.49it/s]
2|1651|Loss: 0.9079467058181763:  26%|██▌       | 1650/6470 [20:38<53:57,  1.49it/s]
2|1651|Loss: 0.9079467058181763:  26%|██▌       | 1651/6470 [20:39<55:47,  1.44it/s]
2|1652|Loss: 0.6719638109207153:  26%|██▌       | 1651/6470 [20:39<55:47,  1.44it/s]
2|1652|Loss: 0.6719638109207153:  26%|██▌       | 1652/6470 [20:40<1:00:38,  1.32it/s]
2|1653|Loss: 0.8634976744651794:  26%|██▌       | 1652/6470 [20:40<1:00:38,  1.32it/s]
2|1653|Loss: 0.8634976744651794:  26%|██▌       | 1653/6470 [20:40<57:51,  1.39it/s]  
2|1654|Loss: 0.9323806166648865:  26%|██▌       | 1653/6470 [20:41<57:51,  1.39it/s]
2|1654|Loss: 0.9323806166648865:  26%|██▌       | 1654/6470 [20:41<58:39,  1.37it/s]
2|1655|Loss: 1.1190340518951416:  26%|██▌       | 1654/6470 [20:41<58:39,  1.37it/s]
2|1655|Loss: 1.1190340518951416:  26%|██▌       | 1655/6470 [20:42<59:30,  1.35it/s]
2|1656|Loss: 0.9431222677230835:  26%|██▌       | 1655/6470 [20:42<59:30,  1.35it/s]
2|1656|Loss: 0.9431222677230835:  26%|██▌       | 1656/6470 [20:43<1:00:59,  1.32it/s]
2|1657|Loss: 0.9436503648757935:  26%|██▌       | 1656/6470 [20:43<1:00:59,  1.32it/s]
2|1657|Loss: 0.9436503648757935:  26%|██▌       | 1657/6470 [20:43<58:47,  1.36it/s]  
2|1658|Loss: 0.7739003896713257:  26%|██▌       | 1657/6470 [20:44<58:47,  1.36it/s]
2|1658|Loss: 0.7739003896713257:  26%|██▌       | 1658/6470 [20:44<59:04,  1.36it/s]
2|1659|Loss: 0.6866376996040344:  26%|██▌       | 1658/6470 [20:44<59:04,  1.36it/s]
2|1659|Loss: 0.6866376996040344:  26%|██▌       | 1659/6470 [20:45<59:36,  1.35it/s]
2|1660|Loss: 0.9159773588180542:  26%|██▌       | 1659/6470 [20:45<59:36,  1.35it/s]
2|1660|Loss: 0.9159773588180542:  26%|██▌       | 1660/6470 [20:45<57:37,  1.39it/s]
2|1661|Loss: 0.7865543961524963:  26%|██▌       | 1660/6470 [20:46<57:37,  1.39it/s]
2|1661|Loss: 0.7865543961524963:  26%|██▌       | 1661/6470 [20:46<56:52,  1.41it/s]
2|1662|Loss: 0.7490784525871277:  26%|██▌       | 1661/6470 [20:46<56:52,  1.41it/s]
2|1662|Loss: 0.7490784525871277:  26%|██▌       | 1662/6470 [20:47<56:36,  1.42it/s]
2|1663|Loss: 0.8138535618782043:  26%|██▌       | 1662/6470 [20:47<56:36,  1.42it/s]
2|1663|Loss: 0.8138535618782043:  26%|██▌       | 1663/6470 [20:47<55:51,  1.43it/s]
2|1664|Loss: 0.6663095951080322:  26%|██▌       | 1663/6470 [20:48<55:51,  1.43it/s]
2|1664|Loss: 0.6663095951080322:  26%|██▌       | 1664/6470 [20:48<1:01:48,  1.30it/s]
2|1665|Loss: 0.784669041633606:  26%|██▌       | 1664/6470 [20:49<1:01:48,  1.30it/s] 
2|1665|Loss: 0.784669041633606:  26%|██▌       | 1665/6470 [20:49<1:01:47,  1.30it/s]
2|1666|Loss: 0.701528012752533:  26%|██▌       | 1665/6470 [20:49<1:01:47,  1.30it/s]
2|1666|Loss: 0.701528012752533:  26%|██▌       | 1666/6470 [20:50<53:13,  1.50it/s]  
2|1667|Loss: 0.9282811880111694:  26%|██▌       | 1666/6470 [20:50<53:13,  1.50it/s]
2|1667|Loss: 0.9282811880111694:  26%|██▌       | 1667/6470 [20:50<56:55,  1.41it/s]
2|1668|Loss: 0.8254263997077942:  26%|██▌       | 1667/6470 [20:51<56:55,  1.41it/s]
2|1668|Loss: 0.8254263997077942:  26%|██▌       | 1668/6470 [20:51<58:25,  1.37it/s]
2|1669|Loss: 0.7991708517074585:  26%|██▌       | 1668/6470 [20:52<58:25,  1.37it/s]
2|1669|Loss: 0.7991708517074585:  26%|██▌       | 1669/6470 [20:52<57:21,  1.39it/s]
2|1670|Loss: 0.8095589876174927:  26%|██▌       | 1669/6470 [20:52<57:21,  1.39it/s]
2|1670|Loss: 0.8095589876174927:  26%|██▌       | 1670/6470 [20:53<1:00:41,  1.32it/s]
2|1671|Loss: 0.755742073059082:  26%|██▌       | 1670/6470 [20:53<1:00:41,  1.32it/s] 
2|1671|Loss: 0.755742073059082:  26%|██▌       | 1671/6470 [20:54<1:10:49,  1.13it/s]
2|1672|Loss: 0.8419294357299805:  26%|██▌       | 1671/6470 [20:54<1:10:49,  1.13it/s]
2|1672|Loss: 0.8419294357299805:  26%|██▌       | 1672/6470 [20:55<1:10:52,  1.13it/s]
2|1673|Loss: 0.8306794762611389:  26%|██▌       | 1672/6470 [20:55<1:10:52,  1.13it/s]
2|1673|Loss: 0.8306794762611389:  26%|██▌       | 1673/6470 [20:56<1:07:40,  1.18it/s]
2|1674|Loss: 0.8900761604309082:  26%|██▌       | 1673/6470 [20:56<1:07:40,  1.18it/s]
2|1674|Loss: 0.8900761604309082:  26%|██▌       | 1674/6470 [20:56<1:02:34,  1.28it/s]
2|1675|Loss: 0.8221807479858398:  26%|██▌       | 1674/6470 [20:57<1:02:34,  1.28it/s]
2|1675|Loss: 0.8221807479858398:  26%|██▌       | 1675/6470 [20:57<1:03:07,  1.27it/s]
2|1676|Loss: 1.0651271343231201:  26%|██▌       | 1675/6470 [20:57<1:03:07,  1.27it/s]
2|1676|Loss: 1.0651271343231201:  26%|██▌       | 1676/6470 [20:58<1:06:18,  1.20it/s]
2|1677|Loss: 0.9346333146095276:  26%|██▌       | 1676/6470 [20:58<1:06:18,  1.20it/s]
2|1677|Loss: 0.9346333146095276:  26%|██▌       | 1677/6470 [20:59<1:04:06,  1.25it/s]
2|1678|Loss: 1.0512380599975586:  26%|██▌       | 1677/6470 [20:59<1:04:06,  1.25it/s]
2|1678|Loss: 1.0512380599975586:  26%|██▌       | 1678/6470 [20:59<1:03:53,  1.25it/s]
2|1679|Loss: 0.7409957051277161:  26%|██▌       | 1678/6470 [21:00<1:03:53,  1.25it/s]
2|1679|Loss: 0.7409957051277161:  26%|██▌       | 1679/6470 [21:00<1:02:01,  1.29it/s]
2|1680|Loss: 1.0172138214111328:  26%|██▌       | 1679/6470 [21:01<1:02:01,  1.29it/s]
2|1680|Loss: 1.0172138214111328:  26%|██▌       | 1680/6470 [21:01<1:03:57,  1.25it/s]
2|1681|Loss: 0.8006595373153687:  26%|██▌       | 1680/6470 [21:01<1:03:57,  1.25it/s]
2|1681|Loss: 0.8006595373153687:  26%|██▌       | 1681/6470 [21:02<1:01:06,  1.31it/s]
2|1682|Loss: 0.733620822429657:  26%|██▌       | 1681/6470 [21:02<1:01:06,  1.31it/s] 
2|1682|Loss: 0.733620822429657:  26%|██▌       | 1682/6470 [21:02<57:39,  1.38it/s]  
2|1683|Loss: 0.9448314309120178:  26%|██▌       | 1682/6470 [21:03<57:39,  1.38it/s]
2|1683|Loss: 0.9448314309120178:  26%|██▌       | 1683/6470 [21:03<1:00:16,  1.32it/s]
2|1684|Loss: 0.8988287448883057:  26%|██▌       | 1683/6470 [21:04<1:00:16,  1.32it/s]
2|1684|Loss: 0.8988287448883057:  26%|██▌       | 1684/6470 [21:04<1:01:46,  1.29it/s]
2|1685|Loss: 0.8926798701286316:  26%|██▌       | 1684/6470 [21:04<1:01:46,  1.29it/s]
2|1685|Loss: 0.8926798701286316:  26%|██▌       | 1685/6470 [21:05<1:02:13,  1.28it/s]
2|1686|Loss: 0.8425158262252808:  26%|██▌       | 1685/6470 [21:05<1:02:13,  1.28it/s]
2|1686|Loss: 0.8425158262252808:  26%|██▌       | 1686/6470 [21:06<1:01:57,  1.29it/s]
2|1687|Loss: 0.7788028717041016:  26%|██▌       | 1686/6470 [21:06<1:01:57,  1.29it/s]
2|1687|Loss: 0.7788028717041016:  26%|██▌       | 1687/6470 [21:07<1:07:16,  1.18it/s]
2|1688|Loss: 0.8103646039962769:  26%|██▌       | 1687/6470 [21:07<1:07:16,  1.18it/s]
2|1688|Loss: 0.8103646039962769:  26%|██▌       | 1688/6470 [21:08<1:10:24,  1.13it/s]
2|1689|Loss: 0.874352753162384:  26%|██▌       | 1688/6470 [21:08<1:10:24,  1.13it/s] 
2|1689|Loss: 0.874352753162384:  26%|██▌       | 1689/6470 [21:08<1:07:10,  1.19it/s]
2|1690|Loss: 0.857832133769989:  26%|██▌       | 1689/6470 [21:09<1:07:10,  1.19it/s]
2|1690|Loss: 0.857832133769989:  26%|██▌       | 1690/6470 [21:09<1:05:50,  1.21it/s]
2|1691|Loss: 0.864016592502594:  26%|██▌       | 1690/6470 [21:09<1:05:50,  1.21it/s]
2|1691|Loss: 0.864016592502594:  26%|██▌       | 1691/6470 [21:10<1:05:12,  1.22it/s]
2|1692|Loss: 0.7057377696037292:  26%|██▌       | 1691/6470 [21:10<1:05:12,  1.22it/s]
2|1692|Loss: 0.7057377696037292:  26%|██▌       | 1692/6470 [21:11<1:09:52,  1.14it/s]
2|1693|Loss: 0.8882470726966858:  26%|██▌       | 1692/6470 [21:11<1:09:52,  1.14it/s]
2|1693|Loss: 0.8882470726966858:  26%|██▌       | 1693/6470 [21:12<1:08:45,  1.16it/s]
2|1694|Loss: 0.9332806468009949:  26%|██▌       | 1693/6470 [21:12<1:08:45,  1.16it/s]
2|1694|Loss: 0.9332806468009949:  26%|██▌       | 1694/6470 [21:13<1:10:05,  1.14it/s]
2|1695|Loss: 0.709287703037262:  26%|██▌       | 1694/6470 [21:13<1:10:05,  1.14it/s] 
2|1695|Loss: 0.709287703037262:  26%|██▌       | 1695/6470 [21:14<1:10:22,  1.13it/s]
2|1696|Loss: 0.8523926734924316:  26%|██▌       | 1695/6470 [21:14<1:10:22,  1.13it/s]
2|1696|Loss: 0.8523926734924316:  26%|██▌       | 1696/6470 [21:14<1:06:43,  1.19it/s]
2|1697|Loss: 0.8035377264022827:  26%|██▌       | 1696/6470 [21:15<1:06:43,  1.19it/s]
2|1697|Loss: 0.8035377264022827:  26%|██▌       | 1697/6470 [21:15<1:10:21,  1.13it/s]
2|1698|Loss: 0.9548191428184509:  26%|██▌       | 1697/6470 [21:16<1:10:21,  1.13it/s]
2|1698|Loss: 0.9548191428184509:  26%|██▌       | 1698/6470 [21:16<1:09:11,  1.15it/s]
2|1699|Loss: 0.8384072780609131:  26%|██▌       | 1698/6470 [21:16<1:09:11,  1.15it/s]
2|1699|Loss: 0.8384072780609131:  26%|██▋       | 1699/6470 [21:17<1:05:47,  1.21it/s]
2|1700|Loss: 0.8086836934089661:  26%|██▋       | 1699/6470 [21:17<1:05:47,  1.21it/s]
2|1700|Loss: 0.8086836934089661:  26%|██▋       | 1700/6470 [21:17<1:01:20,  1.30it/s]
2|1701|Loss: 0.9275822639465332:  26%|██▋       | 1700/6470 [21:18<1:01:20,  1.30it/s]
2|1701|Loss: 0.9275822639465332:  26%|██▋       | 1701/6470 [21:18<57:41,  1.38it/s]  
2|1702|Loss: 0.8912699818611145:  26%|██▋       | 1701/6470 [21:19<57:41,  1.38it/s]
2|1702|Loss: 0.8912699818611145:  26%|██▋       | 1702/6470 [21:19<1:00:08,  1.32it/s]
2|1703|Loss: 0.8207289576530457:  26%|██▋       | 1702/6470 [21:19<1:00:08,  1.32it/s]
2|1703|Loss: 0.8207289576530457:  26%|██▋       | 1703/6470 [21:20<58:36,  1.36it/s]  
2|1704|Loss: 0.9231086373329163:  26%|██▋       | 1703/6470 [21:20<58:36,  1.36it/s]
2|1704|Loss: 0.9231086373329163:  26%|██▋       | 1704/6470 [21:20<59:11,  1.34it/s]
2|1705|Loss: 1.2582132816314697:  26%|██▋       | 1704/6470 [21:21<59:11,  1.34it/s]
2|1705|Loss: 1.2582132816314697:  26%|██▋       | 1705/6470 [21:21<56:37,  1.40it/s]
2|1706|Loss: 0.7657010555267334:  26%|██▋       | 1705/6470 [21:21<56:37,  1.40it/s]
2|1706|Loss: 0.7657010555267334:  26%|██▋       | 1706/6470 [21:22<53:05,  1.50it/s]
2|1707|Loss: 1.0294880867004395:  26%|██▋       | 1706/6470 [21:22<53:05,  1.50it/s]
2|1707|Loss: 1.0294880867004395:  26%|██▋       | 1707/6470 [21:22<53:08,  1.49it/s]
2|1708|Loss: 0.8279125094413757:  26%|██▋       | 1707/6470 [21:23<53:08,  1.49it/s]
2|1708|Loss: 0.8279125094413757:  26%|██▋       | 1708/6470 [21:23<55:52,  1.42it/s]
2|1709|Loss: 0.9010750651359558:  26%|██▋       | 1708/6470 [21:23<55:52,  1.42it/s]
2|1709|Loss: 0.9010750651359558:  26%|██▋       | 1709/6470 [21:24<59:15,  1.34it/s]
2|1710|Loss: 0.8906043767929077:  26%|██▋       | 1709/6470 [21:24<59:15,  1.34it/s]
2|1710|Loss: 0.8906043767929077:  26%|██▋       | 1710/6470 [21:25<1:03:23,  1.25it/s]
2|1711|Loss: 0.761152982711792:  26%|██▋       | 1710/6470 [21:25<1:03:23,  1.25it/s] 
2|1711|Loss: 0.761152982711792:  26%|██▋       | 1711/6470 [21:26<1:04:20,  1.23it/s]
2|1712|Loss: 0.8360697627067566:  26%|██▋       | 1711/6470 [21:26<1:04:20,  1.23it/s]
2|1712|Loss: 0.8360697627067566:  26%|██▋       | 1712/6470 [21:26<1:03:33,  1.25it/s]
2|1713|Loss: 0.9661039113998413:  26%|██▋       | 1712/6470 [21:27<1:03:33,  1.25it/s]
2|1713|Loss: 0.9661039113998413:  26%|██▋       | 1713/6470 [21:27<1:00:18,  1.31it/s]
2|1714|Loss: 0.9332415461540222:  26%|██▋       | 1713/6470 [21:27<1:00:18,  1.31it/s]
2|1714|Loss: 0.9332415461540222:  26%|██▋       | 1714/6470 [21:28<58:27,  1.36it/s]  
2|1715|Loss: 0.8268924951553345:  26%|██▋       | 1714/6470 [21:28<58:27,  1.36it/s]
2|1715|Loss: 0.8268924951553345:  27%|██▋       | 1715/6470 [21:28<56:04,  1.41it/s]
2|1716|Loss: 0.7280818819999695:  27%|██▋       | 1715/6470 [21:29<56:04,  1.41it/s]
2|1716|Loss: 0.7280818819999695:  27%|██▋       | 1716/6470 [21:29<54:53,  1.44it/s]
2|1717|Loss: 0.7522388696670532:  27%|██▋       | 1716/6470 [21:29<54:53,  1.44it/s]
2|1717|Loss: 0.7522388696670532:  27%|██▋       | 1717/6470 [21:30<52:00,  1.52it/s]
2|1718|Loss: 0.8044305443763733:  27%|██▋       | 1717/6470 [21:30<52:00,  1.52it/s]
2|1718|Loss: 0.8044305443763733:  27%|██▋       | 1718/6470 [21:30<56:39,  1.40it/s]
2|1719|Loss: 0.6790036559104919:  27%|██▋       | 1718/6470 [21:31<56:39,  1.40it/s]
2|1719|Loss: 0.6790036559104919:  27%|██▋       | 1719/6470 [21:32<1:04:38,  1.23it/s]
2|1720|Loss: 0.929031491279602:  27%|██▋       | 1719/6470 [21:32<1:04:38,  1.23it/s] 
2|1720|Loss: 0.929031491279602:  27%|██▋       | 1720/6470 [21:33<1:13:37,  1.08it/s]
2|1721|Loss: 0.8766720294952393:  27%|██▋       | 1720/6470 [21:33<1:13:37,  1.08it/s]
2|1721|Loss: 0.8766720294952393:  27%|██▋       | 1721/6470 [21:34<1:11:21,  1.11it/s]
2|1722|Loss: 1.0034719705581665:  27%|██▋       | 1721/6470 [21:34<1:11:21,  1.11it/s]
2|1722|Loss: 1.0034719705581665:  27%|██▋       | 1722/6470 [21:34<1:06:39,  1.19it/s]
2|1723|Loss: 0.8490436673164368:  27%|██▋       | 1722/6470 [21:35<1:06:39,  1.19it/s]
2|1723|Loss: 0.8490436673164368:  27%|██▋       | 1723/6470 [21:35<1:06:58,  1.18it/s]
2|1724|Loss: 0.7568861246109009:  27%|██▋       | 1723/6470 [21:36<1:06:58,  1.18it/s]
2|1724|Loss: 0.7568861246109009:  27%|██▋       | 1724/6470 [21:36<1:03:39,  1.24it/s]
2|1725|Loss: 0.9376339912414551:  27%|██▋       | 1724/6470 [21:36<1:03:39,  1.24it/s]
2|1725|Loss: 0.9376339912414551:  27%|██▋       | 1725/6470 [21:37<1:02:55,  1.26it/s]
2|1726|Loss: 0.8502804636955261:  27%|██▋       | 1725/6470 [21:37<1:02:55,  1.26it/s]
2|1726|Loss: 0.8502804636955261:  27%|██▋       | 1726/6470 [21:37<1:05:05,  1.21it/s]
2|1727|Loss: 0.6778104901313782:  27%|██▋       | 1726/6470 [21:38<1:05:05,  1.21it/s]
2|1727|Loss: 0.6778104901313782:  27%|██▋       | 1727/6470 [21:38<57:36,  1.37it/s]  
2|1728|Loss: 0.8874480128288269:  27%|██▋       | 1727/6470 [21:38<57:36,  1.37it/s]
2|1728|Loss: 0.8874480128288269:  27%|██▋       | 1728/6470 [21:39<59:02,  1.34it/s]
2|1729|Loss: 0.8651962280273438:  27%|██▋       | 1728/6470 [21:39<59:02,  1.34it/s]
2|1729|Loss: 0.8651962280273438:  27%|██▋       | 1729/6470 [21:40<1:02:04,  1.27it/s]
2|1730|Loss: 0.7882466912269592:  27%|██▋       | 1729/6470 [21:40<1:02:04,  1.27it/s]
2|1730|Loss: 0.7882466912269592:  27%|██▋       | 1730/6470 [21:40<1:01:11,  1.29it/s]
2|1731|Loss: 1.1200143098831177:  27%|██▋       | 1730/6470 [21:41<1:01:11,  1.29it/s]
2|1731|Loss: 1.1200143098831177:  27%|██▋       | 1731/6470 [21:41<1:04:50,  1.22it/s]
2|1732|Loss: 0.7691739201545715:  27%|██▋       | 1731/6470 [21:42<1:04:50,  1.22it/s]
2|1732|Loss: 0.7691739201545715:  27%|██▋       | 1732/6470 [21:42<1:03:45,  1.24it/s]
2|1733|Loss: 0.9732335805892944:  27%|██▋       | 1732/6470 [21:42<1:03:45,  1.24it/s]
2|1733|Loss: 0.9732335805892944:  27%|██▋       | 1733/6470 [21:43<58:22,  1.35it/s]  
2|1734|Loss: 0.7706325054168701:  27%|██▋       | 1733/6470 [21:43<58:22,  1.35it/s]
2|1734|Loss: 0.7706325054168701:  27%|██▋       | 1734/6470 [21:43<58:35,  1.35it/s]
2|1735|Loss: 0.7083985805511475:  27%|██▋       | 1734/6470 [21:44<58:35,  1.35it/s]
2|1735|Loss: 0.7083985805511475:  27%|██▋       | 1735/6470 [21:44<58:11,  1.36it/s]
2|1736|Loss: 0.9101694226264954:  27%|██▋       | 1735/6470 [21:45<58:11,  1.36it/s]
2|1736|Loss: 0.9101694226264954:  27%|██▋       | 1736/6470 [21:45<57:20,  1.38it/s]
2|1737|Loss: 0.9621908664703369:  27%|██▋       | 1736/6470 [21:45<57:20,  1.38it/s]
2|1737|Loss: 0.9621908664703369:  27%|██▋       | 1737/6470 [21:46<1:00:01,  1.31it/s]
2|1738|Loss: 0.7759416103363037:  27%|██▋       | 1737/6470 [21:46<1:00:01,  1.31it/s]
2|1738|Loss: 0.7759416103363037:  27%|██▋       | 1738/6470 [21:46<1:00:43,  1.30it/s]
2|1739|Loss: 0.9513648152351379:  27%|██▋       | 1738/6470 [21:47<1:00:43,  1.30it/s]
2|1739|Loss: 0.9513648152351379:  27%|██▋       | 1739/6470 [21:47<56:31,  1.39it/s]  
2|1740|Loss: 0.8975093364715576:  27%|██▋       | 1739/6470 [21:47<56:31,  1.39it/s]
2|1740|Loss: 0.8975093364715576:  27%|██▋       | 1740/6470 [21:48<56:56,  1.38it/s]
2|1741|Loss: 0.8209465742111206:  27%|██▋       | 1740/6470 [21:48<56:56,  1.38it/s]
2|1741|Loss: 0.8209465742111206:  27%|██▋       | 1741/6470 [21:49<58:40,  1.34it/s]
2|1742|Loss: 0.8475950360298157:  27%|██▋       | 1741/6470 [21:49<58:40,  1.34it/s]
2|1742|Loss: 0.8475950360298157:  27%|██▋       | 1742/6470 [21:49<56:28,  1.40it/s]
2|1743|Loss: 0.8027059435844421:  27%|██▋       | 1742/6470 [21:50<56:28,  1.40it/s]
2|1743|Loss: 0.8027059435844421:  27%|██▋       | 1743/6470 [21:50<57:15,  1.38it/s]
2|1744|Loss: 0.9093446731567383:  27%|██▋       | 1743/6470 [21:50<57:15,  1.38it/s]
2|1744|Loss: 0.9093446731567383:  27%|██▋       | 1744/6470 [21:51<58:55,  1.34it/s]
2|1745|Loss: 0.8784443140029907:  27%|██▋       | 1744/6470 [21:51<58:55,  1.34it/s]
2|1745|Loss: 0.8784443140029907:  27%|██▋       | 1745/6470 [21:51<57:03,  1.38it/s]
2|1746|Loss: 0.8703067898750305:  27%|██▋       | 1745/6470 [21:52<57:03,  1.38it/s]
2|1746|Loss: 0.8703067898750305:  27%|██▋       | 1746/6470 [21:52<57:59,  1.36it/s]
2|1747|Loss: 0.8819677233695984:  27%|██▋       | 1746/6470 [21:53<57:59,  1.36it/s]
2|1747|Loss: 0.8819677233695984:  27%|██▋       | 1747/6470 [21:53<58:43,  1.34it/s]
2|1748|Loss: 0.946306049823761:  27%|██▋       | 1747/6470 [21:54<58:43,  1.34it/s] 
2|1748|Loss: 0.946306049823761:  27%|██▋       | 1748/6470 [21:54<1:02:23,  1.26it/s]
2|1749|Loss: 0.7257958650588989:  27%|██▋       | 1748/6470 [21:54<1:02:23,  1.26it/s]
2|1749|Loss: 0.7257958650588989:  27%|██▋       | 1749/6470 [21:55<1:03:33,  1.24it/s]
2|1750|Loss: 0.6277608871459961:  27%|██▋       | 1749/6470 [21:55<1:03:33,  1.24it/s]
2|1750|Loss: 0.6277608871459961:  27%|██▋       | 1750/6470 [21:56<1:04:50,  1.21it/s]
2|1751|Loss: 0.7712663412094116:  27%|██▋       | 1750/6470 [21:56<1:04:50,  1.21it/s]
2|1751|Loss: 0.7712663412094116:  27%|██▋       | 1751/6470 [21:57<1:06:31,  1.18it/s]
2|1752|Loss: 0.9297690391540527:  27%|██▋       | 1751/6470 [21:57<1:06:31,  1.18it/s]
2|1752|Loss: 0.9297690391540527:  27%|██▋       | 1752/6470 [21:57<1:05:23,  1.20it/s]
2|1753|Loss: 0.9878442883491516:  27%|██▋       | 1752/6470 [21:58<1:05:23,  1.20it/s]
2|1753|Loss: 0.9878442883491516:  27%|██▋       | 1753/6470 [21:58<1:07:30,  1.16it/s]
2|1754|Loss: 0.8433976769447327:  27%|██▋       | 1753/6470 [21:59<1:07:30,  1.16it/s]
2|1754|Loss: 0.8433976769447327:  27%|██▋       | 1754/6470 [21:59<1:06:23,  1.18it/s]
2|1755|Loss: 0.9759637117385864:  27%|██▋       | 1754/6470 [21:59<1:06:23,  1.18it/s]
2|1755|Loss: 0.9759637117385864:  27%|██▋       | 1755/6470 [22:00<1:03:37,  1.24it/s]
2|1756|Loss: 0.8735799193382263:  27%|██▋       | 1755/6470 [22:00<1:03:37,  1.24it/s]
2|1756|Loss: 0.8735799193382263:  27%|██▋       | 1756/6470 [22:01<1:04:23,  1.22it/s]
2|1757|Loss: 0.8445277214050293:  27%|██▋       | 1756/6470 [22:01<1:04:23,  1.22it/s]
2|1757|Loss: 0.8445277214050293:  27%|██▋       | 1757/6470 [22:02<1:07:51,  1.16it/s]
2|1758|Loss: 0.7739797234535217:  27%|██▋       | 1757/6470 [22:02<1:07:51,  1.16it/s]
2|1758|Loss: 0.7739797234535217:  27%|██▋       | 1758/6470 [22:02<1:03:08,  1.24it/s]
2|1759|Loss: 0.987795889377594:  27%|██▋       | 1758/6470 [22:03<1:03:08,  1.24it/s] 
2|1759|Loss: 0.987795889377594:  27%|██▋       | 1759/6470 [22:03<1:01:56,  1.27it/s]
2|1760|Loss: 0.8297194242477417:  27%|██▋       | 1759/6470 [22:03<1:01:56,  1.27it/s]
2|1760|Loss: 0.8297194242477417:  27%|██▋       | 1760/6470 [22:04<1:02:19,  1.26it/s]
2|1761|Loss: 0.8194149136543274:  27%|██▋       | 1760/6470 [22:04<1:02:19,  1.26it/s]
2|1761|Loss: 0.8194149136543274:  27%|██▋       | 1761/6470 [22:05<1:01:40,  1.27it/s]
2|1762|Loss: 0.8991082906723022:  27%|██▋       | 1761/6470 [22:05<1:01:40,  1.27it/s]
2|1762|Loss: 0.8991082906723022:  27%|██▋       | 1762/6470 [22:05<57:15,  1.37it/s]  
2|1763|Loss: 0.7178271412849426:  27%|██▋       | 1762/6470 [22:06<57:15,  1.37it/s]
2|1763|Loss: 0.7178271412849426:  27%|██▋       | 1763/6470 [22:06<59:39,  1.31it/s]
2|1764|Loss: 0.9296135306358337:  27%|██▋       | 1763/6470 [22:07<59:39,  1.31it/s]
2|1764|Loss: 0.9296135306358337:  27%|██▋       | 1764/6470 [22:07<1:02:40,  1.25it/s]
2|1765|Loss: 0.8441494107246399:  27%|██▋       | 1764/6470 [22:07<1:02:40,  1.25it/s]
2|1765|Loss: 0.8441494107246399:  27%|██▋       | 1765/6470 [22:08<1:03:14,  1.24it/s]
2|1766|Loss: 0.8687784075737:  27%|██▋       | 1765/6470 [22:08<1:03:14,  1.24it/s]   
2|1766|Loss: 0.8687784075737:  27%|██▋       | 1766/6470 [22:08<1:01:03,  1.28it/s]
2|1767|Loss: 0.7993040084838867:  27%|██▋       | 1766/6470 [22:09<1:01:03,  1.28it/s]
2|1767|Loss: 0.7993040084838867:  27%|██▋       | 1767/6470 [22:09<1:01:18,  1.28it/s]
2|1768|Loss: 0.8855281472206116:  27%|██▋       | 1767/6470 [22:10<1:01:18,  1.28it/s]
2|1768|Loss: 0.8855281472206116:  27%|██▋       | 1768/6470 [22:10<58:21,  1.34it/s]  
2|1769|Loss: 0.9576939344406128:  27%|██▋       | 1768/6470 [22:10<58:21,  1.34it/s]
2|1769|Loss: 0.9576939344406128:  27%|██▋       | 1769/6470 [22:11<55:36,  1.41it/s]
2|1770|Loss: 0.7421464920043945:  27%|██▋       | 1769/6470 [22:11<55:36,  1.41it/s]
2|1770|Loss: 0.7421464920043945:  27%|██▋       | 1770/6470 [22:12<1:03:41,  1.23it/s]
2|1771|Loss: 0.928774893283844:  27%|██▋       | 1770/6470 [22:12<1:03:41,  1.23it/s] 
2|1771|Loss: 0.928774893283844:  27%|██▋       | 1771/6470 [22:12<1:00:13,  1.30it/s]
2|1772|Loss: 0.9845931529998779:  27%|██▋       | 1771/6470 [22:13<1:00:13,  1.30it/s]
2|1772|Loss: 0.9845931529998779:  27%|██▋       | 1772/6470 [22:13<57:15,  1.37it/s]  
2|1773|Loss: 0.7508429288864136:  27%|██▋       | 1772/6470 [22:13<57:15,  1.37it/s]
2|1773|Loss: 0.7508429288864136:  27%|██▋       | 1773/6470 [22:13<54:04,  1.45it/s]
2|1774|Loss: 0.8569589853286743:  27%|██▋       | 1773/6470 [22:14<54:04,  1.45it/s]
2|1774|Loss: 0.8569589853286743:  27%|██▋       | 1774/6470 [22:14<53:39,  1.46it/s]
2|1775|Loss: 0.8334897756576538:  27%|██▋       | 1774/6470 [22:15<53:39,  1.46it/s]
2|1775|Loss: 0.8334897756576538:  27%|██▋       | 1775/6470 [22:15<57:31,  1.36it/s]
2|1776|Loss: 0.757775068283081:  27%|██▋       | 1775/6470 [22:15<57:31,  1.36it/s] 
2|1776|Loss: 0.757775068283081:  27%|██▋       | 1776/6470 [22:16<54:19,  1.44it/s]
2|1777|Loss: 0.8517591953277588:  27%|██▋       | 1776/6470 [22:16<54:19,  1.44it/s]
2|1777|Loss: 0.8517591953277588:  27%|██▋       | 1777/6470 [22:16<56:51,  1.38it/s]
2|1778|Loss: 0.9699357151985168:  27%|██▋       | 1777/6470 [22:17<56:51,  1.38it/s]
2|1778|Loss: 0.9699357151985168:  27%|██▋       | 1778/6470 [22:17<58:45,  1.33it/s]
2|1779|Loss: 0.9272037148475647:  27%|██▋       | 1778/6470 [22:18<58:45,  1.33it/s]
2|1779|Loss: 0.9272037148475647:  27%|██▋       | 1779/6470 [22:18<57:03,  1.37it/s]
2|1780|Loss: 0.8905158638954163:  27%|██▋       | 1779/6470 [22:18<57:03,  1.37it/s]
2|1780|Loss: 0.8905158638954163:  28%|██▊       | 1780/6470 [22:19<57:30,  1.36it/s]
2|1781|Loss: 0.8188556432723999:  28%|██▊       | 1780/6470 [22:19<57:30,  1.36it/s]
2|1781|Loss: 0.8188556432723999:  28%|██▊       | 1781/6470 [22:19<1:00:00,  1.30it/s]
2|1782|Loss: 0.8685476779937744:  28%|██▊       | 1781/6470 [22:20<1:00:00,  1.30it/s]
2|1782|Loss: 0.8685476779937744:  28%|██▊       | 1782/6470 [22:20<58:58,  1.32it/s]  
2|1783|Loss: 0.8623952865600586:  28%|██▊       | 1782/6470 [22:21<58:58,  1.32it/s]
2|1783|Loss: 0.8623952865600586:  28%|██▊       | 1783/6470 [22:21<57:40,  1.35it/s]
2|1784|Loss: 1.0005600452423096:  28%|██▊       | 1783/6470 [22:21<57:40,  1.35it/s]
2|1784|Loss: 1.0005600452423096:  28%|██▊       | 1784/6470 [22:22<56:38,  1.38it/s]
2|1785|Loss: 0.805133044719696:  28%|██▊       | 1784/6470 [22:22<56:38,  1.38it/s] 
2|1785|Loss: 0.805133044719696:  28%|██▊       | 1785/6470 [22:22<56:11,  1.39it/s]
2|1786|Loss: 0.8118290901184082:  28%|██▊       | 1785/6470 [22:23<56:11,  1.39it/s]
2|1786|Loss: 0.8118290901184082:  28%|██▊       | 1786/6470 [22:23<54:42,  1.43it/s]
2|1787|Loss: 0.9273747205734253:  28%|██▊       | 1786/6470 [22:23<54:42,  1.43it/s]
2|1787|Loss: 0.9273747205734253:  28%|██▊       | 1787/6470 [22:24<51:32,  1.51it/s]
2|1788|Loss: 0.7325836420059204:  28%|██▊       | 1787/6470 [22:24<51:32,  1.51it/s]
2|1788|Loss: 0.7325836420059204:  28%|██▊       | 1788/6470 [22:24<46:03,  1.69it/s]
2|1789|Loss: 0.9443538784980774:  28%|██▊       | 1788/6470 [22:24<46:03,  1.69it/s]
2|1789|Loss: 0.9443538784980774:  28%|██▊       | 1789/6470 [22:25<48:02,  1.62it/s]
2|1790|Loss: 0.974945068359375:  28%|██▊       | 1789/6470 [22:25<48:02,  1.62it/s] 
2|1790|Loss: 0.974945068359375:  28%|██▊       | 1790/6470 [22:25<53:23,  1.46it/s]
2|1791|Loss: 0.8934566974639893:  28%|██▊       | 1790/6470 [22:26<53:23,  1.46it/s]
2|1791|Loss: 0.8934566974639893:  28%|██▊       | 1791/6470 [22:26<54:21,  1.43it/s]
2|1792|Loss: 0.7755483388900757:  28%|██▊       | 1791/6470 [22:27<54:21,  1.43it/s]
2|1792|Loss: 0.7755483388900757:  28%|██▊       | 1792/6470 [22:27<56:05,  1.39it/s]
2|1793|Loss: 0.8407220840454102:  28%|██▊       | 1792/6470 [22:27<56:05,  1.39it/s]
2|1793|Loss: 0.8407220840454102:  28%|██▊       | 1793/6470 [22:28<57:43,  1.35it/s]
2|1794|Loss: 0.8405438661575317:  28%|██▊       | 1793/6470 [22:28<57:43,  1.35it/s]
2|1794|Loss: 0.8405438661575317:  28%|██▊       | 1794/6470 [22:28<57:06,  1.36it/s]
2|1795|Loss: 0.8188190460205078:  28%|██▊       | 1794/6470 [22:29<57:06,  1.36it/s]
2|1795|Loss: 0.8188190460205078:  28%|██▊       | 1795/6470 [22:29<56:12,  1.39it/s]
2|1796|Loss: 0.7824895977973938:  28%|██▊       | 1795/6470 [22:30<56:12,  1.39it/s]
2|1796|Loss: 0.7824895977973938:  28%|██▊       | 1796/6470 [22:30<55:12,  1.41it/s]
2|1797|Loss: 0.7620832324028015:  28%|██▊       | 1796/6470 [22:30<55:12,  1.41it/s]
2|1797|Loss: 0.7620832324028015:  28%|██▊       | 1797/6470 [22:31<56:12,  1.39it/s]
2|1798|Loss: 0.8579888939857483:  28%|██▊       | 1797/6470 [22:31<56:12,  1.39it/s]
2|1798|Loss: 0.8579888939857483:  28%|██▊       | 1798/6470 [22:32<1:00:18,  1.29it/s]
2|1799|Loss: 1.0174227952957153:  28%|██▊       | 1798/6470 [22:32<1:00:18,  1.29it/s]
2|1799|Loss: 1.0174227952957153:  28%|██▊       | 1799/6470 [22:32<58:41,  1.33it/s]  
2|1800|Loss: 0.9086606502532959:  28%|██▊       | 1799/6470 [22:33<58:41,  1.33it/s]
2|1800|Loss: 0.9086606502532959:  28%|██▊       | 1800/6470 [22:33<56:59,  1.37it/s]
2|1801|Loss: 0.8357005715370178:  28%|██▊       | 1800/6470 [22:33<56:59,  1.37it/s]
2|1801|Loss: 0.8357005715370178:  28%|██▊       | 1801/6470 [22:34<58:15,  1.34it/s]
2|1802|Loss: 0.8095031380653381:  28%|██▊       | 1801/6470 [22:34<58:15,  1.34it/s]
2|1802|Loss: 0.8095031380653381:  28%|██▊       | 1802/6470 [22:34<55:44,  1.40it/s]
2|1803|Loss: 0.7785753011703491:  28%|██▊       | 1802/6470 [22:35<55:44,  1.40it/s]
2|1803|Loss: 0.7785753011703491:  28%|██▊       | 1803/6470 [22:35<54:39,  1.42it/s]
2|1804|Loss: 0.6829484105110168:  28%|██▊       | 1803/6470 [22:35<54:39,  1.42it/s]
2|1804|Loss: 0.6829484105110168:  28%|██▊       | 1804/6470 [22:36<56:38,  1.37it/s]
2|1805|Loss: 0.923694908618927:  28%|██▊       | 1804/6470 [22:36<56:38,  1.37it/s] 
2|1805|Loss: 0.923694908618927:  28%|██▊       | 1805/6470 [22:37<58:26,  1.33it/s]
2|1806|Loss: 0.8245393633842468:  28%|██▊       | 1805/6470 [22:37<58:26,  1.33it/s]
2|1806|Loss: 0.8245393633842468:  28%|██▊       | 1806/6470 [22:37<57:17,  1.36it/s]
2|1807|Loss: 0.8339512348175049:  28%|██▊       | 1806/6470 [22:38<57:17,  1.36it/s]
2|1807|Loss: 0.8339512348175049:  28%|██▊       | 1807/6470 [22:38<1:01:24,  1.27it/s]
2|1808|Loss: 0.9344425797462463:  28%|██▊       | 1807/6470 [22:39<1:01:24,  1.27it/s]
2|1808|Loss: 0.9344425797462463:  28%|██▊       | 1808/6470 [22:39<58:44,  1.32it/s]  
2|1809|Loss: 0.7186079025268555:  28%|██▊       | 1808/6470 [22:39<58:44,  1.32it/s]
2|1809|Loss: 0.7186079025268555:  28%|██▊       | 1809/6470 [22:40<57:11,  1.36it/s]
2|1810|Loss: 0.8667930960655212:  28%|██▊       | 1809/6470 [22:40<57:11,  1.36it/s]
2|1810|Loss: 0.8667930960655212:  28%|██▊       | 1810/6470 [22:40<1:00:14,  1.29it/s]
2|1811|Loss: 0.8432470560073853:  28%|██▊       | 1810/6470 [22:41<1:00:14,  1.29it/s]
2|1811|Loss: 0.8432470560073853:  28%|██▊       | 1811/6470 [22:41<59:27,  1.31it/s]  
2|1812|Loss: 0.8577637672424316:  28%|██▊       | 1811/6470 [22:42<59:27,  1.31it/s]
2|1812|Loss: 0.8577637672424316:  28%|██▊       | 1812/6470 [22:42<58:16,  1.33it/s]
2|1813|Loss: 0.8315290808677673:  28%|██▊       | 1812/6470 [22:42<58:16,  1.33it/s]
2|1813|Loss: 0.8315290808677673:  28%|██▊       | 1813/6470 [22:43<55:15,  1.40it/s]
2|1814|Loss: 0.9574151039123535:  28%|██▊       | 1813/6470 [22:43<55:15,  1.40it/s]
2|1814|Loss: 0.9574151039123535:  28%|██▊       | 1814/6470 [22:43<55:57,  1.39it/s]
2|1815|Loss: 0.9920103549957275:  28%|██▊       | 1814/6470 [22:44<55:57,  1.39it/s]
2|1815|Loss: 0.9920103549957275:  28%|██▊       | 1815/6470 [22:44<1:00:13,  1.29it/s]
2|1816|Loss: 0.8370279669761658:  28%|██▊       | 1815/6470 [22:45<1:00:13,  1.29it/s]
2|1816|Loss: 0.8370279669761658:  28%|██▊       | 1816/6470 [22:45<57:31,  1.35it/s]  
2|1817|Loss: 0.7055222392082214:  28%|██▊       | 1816/6470 [22:45<57:31,  1.35it/s]
2|1817|Loss: 0.7055222392082214:  28%|██▊       | 1817/6470 [22:46<56:57,  1.36it/s]
2|1818|Loss: 0.875454306602478:  28%|██▊       | 1817/6470 [22:46<56:57,  1.36it/s] 
2|1818|Loss: 0.875454306602478:  28%|██▊       | 1818/6470 [22:46<56:24,  1.37it/s]
2|1819|Loss: 0.9351484179496765:  28%|██▊       | 1818/6470 [22:47<56:24,  1.37it/s]
2|1819|Loss: 0.9351484179496765:  28%|██▊       | 1819/6470 [22:47<1:00:17,  1.29it/s]
2|1820|Loss: 0.8699158430099487:  28%|██▊       | 1819/6470 [22:48<1:00:17,  1.29it/s]
2|1820|Loss: 0.8699158430099487:  28%|██▊       | 1820/6470 [22:48<1:00:41,  1.28it/s]
2|1821|Loss: 0.8937978148460388:  28%|██▊       | 1820/6470 [22:48<1:00:41,  1.28it/s]
2|1821|Loss: 0.8937978148460388:  28%|██▊       | 1821/6470 [22:49<1:02:52,  1.23it/s]
2|1822|Loss: 0.7769608497619629:  28%|██▊       | 1821/6470 [22:49<1:02:52,  1.23it/s]
2|1822|Loss: 0.7769608497619629:  28%|██▊       | 1822/6470 [22:50<1:02:56,  1.23it/s]
2|1823|Loss: 0.8750578165054321:  28%|██▊       | 1822/6470 [22:50<1:02:56,  1.23it/s]
2|1823|Loss: 0.8750578165054321:  28%|██▊       | 1823/6470 [22:50<1:00:55,  1.27it/s]
2|1824|Loss: 0.8641142845153809:  28%|██▊       | 1823/6470 [22:51<1:00:55,  1.27it/s]
2|1824|Loss: 0.8641142845153809:  28%|██▊       | 1824/6470 [22:51<58:35,  1.32it/s]  
2|1825|Loss: 0.9759765863418579:  28%|██▊       | 1824/6470 [22:51<58:35,  1.32it/s]
2|1825|Loss: 0.9759765863418579:  28%|██▊       | 1825/6470 [22:52<1:01:04,  1.27it/s]
2|1826|Loss: 0.8373494148254395:  28%|██▊       | 1825/6470 [22:52<1:01:04,  1.27it/s]
2|1826|Loss: 0.8373494148254395:  28%|██▊       | 1826/6470 [22:53<1:03:46,  1.21it/s]
2|1827|Loss: 0.9133023023605347:  28%|██▊       | 1826/6470 [22:53<1:03:46,  1.21it/s]
2|1827|Loss: 0.9133023023605347:  28%|██▊       | 1827/6470 [22:54<1:01:09,  1.27it/s]
2|1828|Loss: 0.8995821475982666:  28%|██▊       | 1827/6470 [22:54<1:01:09,  1.27it/s]
2|1828|Loss: 0.8995821475982666:  28%|██▊       | 1828/6470 [22:54<1:00:48,  1.27it/s]
2|1829|Loss: 0.8561719059944153:  28%|██▊       | 1828/6470 [22:55<1:00:48,  1.27it/s]
2|1829|Loss: 0.8561719059944153:  28%|██▊       | 1829/6470 [22:55<1:02:35,  1.24it/s]
2|1830|Loss: 0.7663825154304504:  28%|██▊       | 1829/6470 [22:56<1:02:35,  1.24it/s]
2|1830|Loss: 0.7663825154304504:  28%|██▊       | 1830/6470 [22:56<1:02:32,  1.24it/s]
2|1831|Loss: 0.8346080780029297:  28%|██▊       | 1830/6470 [22:56<1:02:32,  1.24it/s]
2|1831|Loss: 0.8346080780029297:  28%|██▊       | 1831/6470 [22:57<1:00:41,  1.27it/s]
2|1832|Loss: 0.7517041563987732:  28%|██▊       | 1831/6470 [22:57<1:00:41,  1.27it/s]
2|1832|Loss: 0.7517041563987732:  28%|██▊       | 1832/6470 [22:57<58:45,  1.32it/s]  
2|1833|Loss: 0.8257195353507996:  28%|██▊       | 1832/6470 [22:58<58:45,  1.32it/s]
2|1833|Loss: 0.8257195353507996:  28%|██▊       | 1833/6470 [22:58<57:19,  1.35it/s]
2|1834|Loss: 0.7871207594871521:  28%|██▊       | 1833/6470 [22:59<57:19,  1.35it/s]
2|1834|Loss: 0.7871207594871521:  28%|██▊       | 1834/6470 [22:59<58:22,  1.32it/s]
2|1835|Loss: 0.9383779168128967:  28%|██▊       | 1834/6470 [22:59<58:22,  1.32it/s]
2|1835|Loss: 0.9383779168128967:  28%|██▊       | 1835/6470 [23:00<57:02,  1.35it/s]
2|1836|Loss: 0.8409872651100159:  28%|██▊       | 1835/6470 [23:00<57:02,  1.35it/s]
2|1836|Loss: 0.8409872651100159:  28%|██▊       | 1836/6470 [23:00<57:49,  1.34it/s]
2|1837|Loss: 0.49929723143577576:  28%|██▊       | 1836/6470 [23:01<57:49,  1.34it/s]
2|1837|Loss: 0.49929723143577576:  28%|██▊       | 1837/6470 [23:01<1:05:32,  1.18it/s]
2|1838|Loss: 0.8795368671417236:  28%|██▊       | 1837/6470 [23:02<1:05:32,  1.18it/s] 
2|1838|Loss: 0.8795368671417236:  28%|██▊       | 1838/6470 [23:02<1:03:51,  1.21it/s]
2|1839|Loss: 0.8228660821914673:  28%|██▊       | 1838/6470 [23:03<1:03:51,  1.21it/s]
2|1839|Loss: 0.8228660821914673:  28%|██▊       | 1839/6470 [23:03<59:11,  1.30it/s]  
2|1840|Loss: 0.9071308374404907:  28%|██▊       | 1839/6470 [23:03<59:11,  1.30it/s]
2|1840|Loss: 0.9071308374404907:  28%|██▊       | 1840/6470 [23:04<56:45,  1.36it/s]
2|1841|Loss: 0.8336036205291748:  28%|██▊       | 1840/6470 [23:04<56:45,  1.36it/s]
2|1841|Loss: 0.8336036205291748:  28%|██▊       | 1841/6470 [23:04<53:17,  1.45it/s]
2|1842|Loss: 0.9430359601974487:  28%|██▊       | 1841/6470 [23:05<53:17,  1.45it/s]
2|1842|Loss: 0.9430359601974487:  28%|██▊       | 1842/6470 [23:05<56:39,  1.36it/s]
2|1843|Loss: 0.9541813731193542:  28%|██▊       | 1842/6470 [23:05<56:39,  1.36it/s]
2|1843|Loss: 0.9541813731193542:  28%|██▊       | 1843/6470 [23:06<1:00:14,  1.28it/s]
2|1844|Loss: 0.6923427581787109:  28%|██▊       | 1843/6470 [23:06<1:00:14,  1.28it/s]
2|1844|Loss: 0.6923427581787109:  29%|██▊       | 1844/6470 [23:07<1:00:55,  1.27it/s]
2|1845|Loss: 0.8092588782310486:  29%|██▊       | 1844/6470 [23:07<1:00:55,  1.27it/s]
2|1845|Loss: 0.8092588782310486:  29%|██▊       | 1845/6470 [23:07<55:19,  1.39it/s]  
2|1846|Loss: 0.8710426092147827:  29%|██▊       | 1845/6470 [23:08<55:19,  1.39it/s]
2|1846|Loss: 0.8710426092147827:  29%|██▊       | 1846/6470 [23:08<54:40,  1.41it/s]
2|1847|Loss: 0.8125911951065063:  29%|██▊       | 1846/6470 [23:08<54:40,  1.41it/s]
2|1847|Loss: 0.8125911951065063:  29%|██▊       | 1847/6470 [23:09<56:41,  1.36it/s]
2|1848|Loss: 0.942242443561554:  29%|██▊       | 1847/6470 [23:09<56:41,  1.36it/s] 
2|1848|Loss: 0.942242443561554:  29%|██▊       | 1848/6470 [23:09<55:24,  1.39it/s]
2|1849|Loss: 0.8976708650588989:  29%|██▊       | 1848/6470 [23:10<55:24,  1.39it/s]
2|1849|Loss: 0.8976708650588989:  29%|██▊       | 1849/6470 [23:10<51:35,  1.49it/s]
2|1850|Loss: 0.8160108327865601:  29%|██▊       | 1849/6470 [23:10<51:35,  1.49it/s]
2|1850|Loss: 0.8160108327865601:  29%|██▊       | 1850/6470 [23:11<53:28,  1.44it/s]
2|1851|Loss: 0.8401763439178467:  29%|██▊       | 1850/6470 [23:11<53:28,  1.44it/s]
2|1851|Loss: 0.8401763439178467:  29%|██▊       | 1851/6470 [23:11<53:04,  1.45it/s]
2|1852|Loss: 1.00883150100708:  29%|██▊       | 1851/6470 [23:12<53:04,  1.45it/s]  
2|1852|Loss: 1.00883150100708:  29%|██▊       | 1852/6470 [23:12<56:25,  1.36it/s]
2|1853|Loss: 0.7886359095573425:  29%|██▊       | 1852/6470 [23:13<56:25,  1.36it/s]
2|1853|Loss: 0.7886359095573425:  29%|██▊       | 1853/6470 [23:13<57:20,  1.34it/s]
2|1854|Loss: 0.8264204263687134:  29%|██▊       | 1853/6470 [23:13<57:20,  1.34it/s]
2|1854|Loss: 0.8264204263687134:  29%|██▊       | 1854/6470 [23:14<58:17,  1.32it/s]
2|1855|Loss: 0.6740220785140991:  29%|██▊       | 1854/6470 [23:14<58:17,  1.32it/s]
2|1855|Loss: 0.6740220785140991:  29%|██▊       | 1855/6470 [23:15<1:02:18,  1.23it/s]
2|1856|Loss: 0.8892918229103088:  29%|██▊       | 1855/6470 [23:15<1:02:18,  1.23it/s]
2|1856|Loss: 0.8892918229103088:  29%|██▊       | 1856/6470 [23:15<59:04,  1.30it/s]  
2|1857|Loss: 0.7447224259376526:  29%|██▊       | 1856/6470 [23:16<59:04,  1.30it/s]
2|1857|Loss: 0.7447224259376526:  29%|██▊       | 1857/6470 [23:16<56:52,  1.35it/s]
2|1858|Loss: 0.7941441535949707:  29%|██▊       | 1857/6470 [23:16<56:52,  1.35it/s]
2|1858|Loss: 0.7941441535949707:  29%|██▊       | 1858/6470 [23:17<57:50,  1.33it/s]
2|1859|Loss: 0.9684950709342957:  29%|██▊       | 1858/6470 [23:17<57:50,  1.33it/s]
2|1859|Loss: 0.9684950709342957:  29%|██▊       | 1859/6470 [23:18<1:00:43,  1.27it/s]
2|1860|Loss: 0.7659687995910645:  29%|██▊       | 1859/6470 [23:18<1:00:43,  1.27it/s]
2|1860|Loss: 0.7659687995910645:  29%|██▊       | 1860/6470 [23:18<56:52,  1.35it/s]  
2|1861|Loss: 0.860023021697998:  29%|██▊       | 1860/6470 [23:19<56:52,  1.35it/s] 
2|1861|Loss: 0.860023021697998:  29%|██▉       | 1861/6470 [23:19<55:20,  1.39it/s]
2|1862|Loss: 0.7732763886451721:  29%|██▉       | 1861/6470 [23:19<55:20,  1.39it/s]
2|1862|Loss: 0.7732763886451721:  29%|██▉       | 1862/6470 [23:20<55:22,  1.39it/s]
2|1863|Loss: 0.9427696466445923:  29%|██▉       | 1862/6470 [23:20<55:22,  1.39it/s]
2|1863|Loss: 0.9427696466445923:  29%|██▉       | 1863/6470 [23:20<57:04,  1.35it/s]
2|1864|Loss: 0.886377215385437:  29%|██▉       | 1863/6470 [23:21<57:04,  1.35it/s] 
2|1864|Loss: 0.886377215385437:  29%|██▉       | 1864/6470 [23:21<55:51,  1.37it/s]
2|1865|Loss: 0.866511344909668:  29%|██▉       | 1864/6470 [23:21<55:51,  1.37it/s]
2|1865|Loss: 0.866511344909668:  29%|██▉       | 1865/6470 [23:22<51:05,  1.50it/s]
2|1866|Loss: 1.010766625404358:  29%|██▉       | 1865/6470 [23:22<51:05,  1.50it/s]
2|1866|Loss: 1.010766625404358:  29%|██▉       | 1866/6470 [23:23<55:15,  1.39it/s]
2|1867|Loss: 0.8705427050590515:  29%|██▉       | 1866/6470 [23:23<55:15,  1.39it/s]
2|1867|Loss: 0.8705427050590515:  29%|██▉       | 1867/6470 [23:23<53:37,  1.43it/s]
2|1868|Loss: 0.6816107034683228:  29%|██▉       | 1867/6470 [23:23<53:37,  1.43it/s]
2|1868|Loss: 0.6816107034683228:  29%|██▉       | 1868/6470 [23:24<49:27,  1.55it/s]
2|1869|Loss: 0.7764018774032593:  29%|██▉       | 1868/6470 [23:24<49:27,  1.55it/s]
2|1869|Loss: 0.7764018774032593:  29%|██▉       | 1869/6470 [23:24<47:31,  1.61it/s]
2|1870|Loss: 1.0527721643447876:  29%|██▉       | 1869/6470 [23:25<47:31,  1.61it/s]
2|1870|Loss: 1.0527721643447876:  29%|██▉       | 1870/6470 [23:25<51:07,  1.50it/s]
2|1871|Loss: 0.8112109899520874:  29%|██▉       | 1870/6470 [23:26<51:07,  1.50it/s]
2|1871|Loss: 0.8112109899520874:  29%|██▉       | 1871/6470 [23:26<55:40,  1.38it/s]
2|1872|Loss: 0.7905320525169373:  29%|██▉       | 1871/6470 [23:26<55:40,  1.38it/s]
2|1872|Loss: 0.7905320525169373:  29%|██▉       | 1872/6470 [23:27<58:18,  1.31it/s]
2|1873|Loss: 1.003224492073059:  29%|██▉       | 1872/6470 [23:27<58:18,  1.31it/s] 
2|1873|Loss: 1.003224492073059:  29%|██▉       | 1873/6470 [23:28<59:51,  1.28it/s]
2|1874|Loss: 0.9457831382751465:  29%|██▉       | 1873/6470 [23:28<59:51,  1.28it/s]
2|1874|Loss: 0.9457831382751465:  29%|██▉       | 1874/6470 [23:28<58:34,  1.31it/s]
2|1875|Loss: 0.8076491951942444:  29%|██▉       | 1874/6470 [23:29<58:34,  1.31it/s]
2|1875|Loss: 0.8076491951942444:  29%|██▉       | 1875/6470 [23:29<56:55,  1.35it/s]
2|1876|Loss: 0.9655890464782715:  29%|██▉       | 1875/6470 [23:29<56:55,  1.35it/s]
2|1876|Loss: 0.9655890464782715:  29%|██▉       | 1876/6470 [23:30<56:11,  1.36it/s]
2|1877|Loss: 0.7176648378372192:  29%|██▉       | 1876/6470 [23:30<56:11,  1.36it/s]
2|1877|Loss: 0.7176648378372192:  29%|██▉       | 1877/6470 [23:31<59:27,  1.29it/s]
2|1878|Loss: 0.9835130572319031:  29%|██▉       | 1877/6470 [23:31<59:27,  1.29it/s]
2|1878|Loss: 0.9835130572319031:  29%|██▉       | 1878/6470 [23:31<57:18,  1.34it/s]
2|1879|Loss: 0.8095314502716064:  29%|██▉       | 1878/6470 [23:32<57:18,  1.34it/s]
2|1879|Loss: 0.8095314502716064:  29%|██▉       | 1879/6470 [23:32<54:47,  1.40it/s]
2|1880|Loss: 0.8601827621459961:  29%|██▉       | 1879/6470 [23:32<54:47,  1.40it/s]
2|1880|Loss: 0.8601827621459961:  29%|██▉       | 1880/6470 [23:33<53:55,  1.42it/s]
2|1881|Loss: 0.982170045375824:  29%|██▉       | 1880/6470 [23:33<53:55,  1.42it/s] 
2|1881|Loss: 0.982170045375824:  29%|██▉       | 1881/6470 [23:33<55:45,  1.37it/s]
2|1882|Loss: 0.8914009928703308:  29%|██▉       | 1881/6470 [23:34<55:45,  1.37it/s]
2|1882|Loss: 0.8914009928703308:  29%|██▉       | 1882/6470 [23:34<53:34,  1.43it/s]
2|1883|Loss: 0.8008170127868652:  29%|██▉       | 1882/6470 [23:34<53:34,  1.43it/s]
2|1883|Loss: 0.8008170127868652:  29%|██▉       | 1883/6470 [23:35<51:17,  1.49it/s]
2|1884|Loss: 0.7492426633834839:  29%|██▉       | 1883/6470 [23:35<51:17,  1.49it/s]
2|1884|Loss: 0.7492426633834839:  29%|██▉       | 1884/6470 [23:35<49:11,  1.55it/s]
2|1885|Loss: 0.7360697984695435:  29%|██▉       | 1884/6470 [23:36<49:11,  1.55it/s]
2|1885|Loss: 0.7360697984695435:  29%|██▉       | 1885/6470 [23:36<49:13,  1.55it/s]
2|1886|Loss: 0.8868959546089172:  29%|██▉       | 1885/6470 [23:36<49:13,  1.55it/s]
2|1886|Loss: 0.8868959546089172:  29%|██▉       | 1886/6470 [23:37<52:18,  1.46it/s]
2|1887|Loss: 0.6104779243469238:  29%|██▉       | 1886/6470 [23:37<52:18,  1.46it/s]
2|1887|Loss: 0.6104779243469238:  29%|██▉       | 1887/6470 [23:38<59:44,  1.28it/s]
2|1888|Loss: 0.8962111473083496:  29%|██▉       | 1887/6470 [23:38<59:44,  1.28it/s]
2|1888|Loss: 0.8962111473083496:  29%|██▉       | 1888/6470 [23:38<1:00:53,  1.25it/s]
2|1889|Loss: 0.7493029832839966:  29%|██▉       | 1888/6470 [23:39<1:00:53,  1.25it/s]
2|1889|Loss: 0.7493029832839966:  29%|██▉       | 1889/6470 [23:39<1:00:39,  1.26it/s]
2|1890|Loss: 0.8701826333999634:  29%|██▉       | 1889/6470 [23:40<1:00:39,  1.26it/s]
2|1890|Loss: 0.8701826333999634:  29%|██▉       | 1890/6470 [23:40<57:25,  1.33it/s]  
2|1891|Loss: 0.8339775204658508:  29%|██▉       | 1890/6470 [23:40<57:25,  1.33it/s]
2|1891|Loss: 0.8339775204658508:  29%|██▉       | 1891/6470 [23:41<56:46,  1.34it/s]
2|1892|Loss: 0.8094688057899475:  29%|██▉       | 1891/6470 [23:41<56:46,  1.34it/s]
2|1892|Loss: 0.8094688057899475:  29%|██▉       | 1892/6470 [23:41<55:59,  1.36it/s]
2|1893|Loss: 0.9175020456314087:  29%|██▉       | 1892/6470 [23:42<55:59,  1.36it/s]
2|1893|Loss: 0.9175020456314087:  29%|██▉       | 1893/6470 [23:42<53:45,  1.42it/s]
2|1894|Loss: 0.9144920706748962:  29%|██▉       | 1893/6470 [23:42<53:45,  1.42it/s]
2|1894|Loss: 0.9144920706748962:  29%|██▉       | 1894/6470 [23:43<53:10,  1.43it/s]
2|1895|Loss: 0.8300154805183411:  29%|██▉       | 1894/6470 [23:43<53:10,  1.43it/s]
2|1895|Loss: 0.8300154805183411:  29%|██▉       | 1895/6470 [23:43<56:18,  1.35it/s]
2|1896|Loss: 0.8346553444862366:  29%|██▉       | 1895/6470 [23:44<56:18,  1.35it/s]
2|1896|Loss: 0.8346553444862366:  29%|██▉       | 1896/6470 [23:44<56:16,  1.35it/s]
2|1897|Loss: 0.7861322164535522:  29%|██▉       | 1896/6470 [23:45<56:16,  1.35it/s]
2|1897|Loss: 0.7861322164535522:  29%|██▉       | 1897/6470 [23:45<51:15,  1.49it/s]
2|1898|Loss: 0.892234206199646:  29%|██▉       | 1897/6470 [23:45<51:15,  1.49it/s] 
2|1898|Loss: 0.892234206199646:  29%|██▉       | 1898/6470 [23:45<51:39,  1.48it/s]
2|1899|Loss: 0.8537485003471375:  29%|██▉       | 1898/6470 [23:46<51:39,  1.48it/s]
2|1899|Loss: 0.8537485003471375:  29%|██▉       | 1899/6470 [23:46<54:32,  1.40it/s]
2|1900|Loss: 0.9474111795425415:  29%|██▉       | 1899/6470 [23:47<54:32,  1.40it/s]
2|1900|Loss: 0.9474111795425415:  29%|██▉       | 1900/6470 [23:47<54:31,  1.40it/s]
2|1901|Loss: 0.7010135650634766:  29%|██▉       | 1900/6470 [23:47<54:31,  1.40it/s]
2|1901|Loss: 0.7010135650634766:  29%|██▉       | 1901/6470 [23:48<56:27,  1.35it/s]
2|1902|Loss: 0.8389204144477844:  29%|██▉       | 1901/6470 [23:48<56:27,  1.35it/s]
2|1902|Loss: 0.8389204144477844:  29%|██▉       | 1902/6470 [23:48<56:15,  1.35it/s]
2|1903|Loss: 0.7302733659744263:  29%|██▉       | 1902/6470 [23:49<56:15,  1.35it/s]
2|1903|Loss: 0.7302733659744263:  29%|██▉       | 1903/6470 [23:49<57:09,  1.33it/s]
2|1904|Loss: 0.8424032330513:  29%|██▉       | 1903/6470 [23:50<57:09,  1.33it/s]   
2|1904|Loss: 0.8424032330513:  29%|██▉       | 1904/6470 [23:50<55:01,  1.38it/s]
2|1905|Loss: 0.7529892921447754:  29%|██▉       | 1904/6470 [23:50<55:01,  1.38it/s]
2|1905|Loss: 0.7529892921447754:  29%|██▉       | 1905/6470 [23:51<55:53,  1.36it/s]
2|1906|Loss: 0.8821452260017395:  29%|██▉       | 1905/6470 [23:51<55:53,  1.36it/s]
2|1906|Loss: 0.8821452260017395:  29%|██▉       | 1906/6470 [23:51<52:53,  1.44it/s]
2|1907|Loss: 0.8940261602401733:  29%|██▉       | 1906/6470 [23:52<52:53,  1.44it/s]
2|1907|Loss: 0.8940261602401733:  29%|██▉       | 1907/6470 [23:52<50:41,  1.50it/s]
2|1908|Loss: 0.7136843800544739:  29%|██▉       | 1907/6470 [23:52<50:41,  1.50it/s]
2|1908|Loss: 0.7136843800544739:  29%|██▉       | 1908/6470 [23:52<48:29,  1.57it/s]
2|1909|Loss: 0.9531540274620056:  29%|██▉       | 1908/6470 [23:53<48:29,  1.57it/s]
2|1909|Loss: 0.9531540274620056:  30%|██▉       | 1909/6470 [23:53<43:36,  1.74it/s]
2|1910|Loss: 0.9701319336891174:  30%|██▉       | 1909/6470 [23:53<43:36,  1.74it/s]
2|1910|Loss: 0.9701319336891174:  30%|██▉       | 1910/6470 [23:54<46:49,  1.62it/s]
2|1911|Loss: 0.9877276420593262:  30%|██▉       | 1910/6470 [23:54<46:49,  1.62it/s]
2|1911|Loss: 0.9877276420593262:  30%|██▉       | 1911/6470 [23:54<51:03,  1.49it/s]
2|1912|Loss: 0.8071150779724121:  30%|██▉       | 1911/6470 [23:55<51:03,  1.49it/s]
2|1912|Loss: 0.8071150779724121:  30%|██▉       | 1912/6470 [23:56<1:03:08,  1.20it/s]
2|1913|Loss: 0.802497148513794:  30%|██▉       | 1912/6470 [23:56<1:03:08,  1.20it/s] 
2|1913|Loss: 0.802497148513794:  30%|██▉       | 1913/6470 [23:56<1:04:12,  1.18it/s]
2|1914|Loss: 0.8523873090744019:  30%|██▉       | 1913/6470 [23:57<1:04:12,  1.18it/s]
2|1914|Loss: 0.8523873090744019:  30%|██▉       | 1914/6470 [23:57<1:02:22,  1.22it/s]
2|1915|Loss: 0.7435300350189209:  30%|██▉       | 1914/6470 [23:58<1:02:22,  1.22it/s]
2|1915|Loss: 0.7435300350189209:  30%|██▉       | 1915/6470 [23:58<1:01:18,  1.24it/s]
2|1916|Loss: 0.824651837348938:  30%|██▉       | 1915/6470 [23:58<1:01:18,  1.24it/s] 
2|1916|Loss: 0.824651837348938:  30%|██▉       | 1916/6470 [23:59<1:00:32,  1.25it/s]
2|1917|Loss: 0.7810195684432983:  30%|██▉       | 1916/6470 [23:59<1:00:32,  1.25it/s]
2|1917|Loss: 0.7810195684432983:  30%|██▉       | 1917/6470 [24:00<1:02:14,  1.22it/s]
2|1918|Loss: 1.059513807296753:  30%|██▉       | 1917/6470 [24:00<1:02:14,  1.22it/s] 
2|1918|Loss: 1.059513807296753:  30%|██▉       | 1918/6470 [24:00<59:28,  1.28it/s]  
2|1919|Loss: 0.7485268115997314:  30%|██▉       | 1918/6470 [24:01<59:28,  1.28it/s]
2|1919|Loss: 0.7485268115997314:  30%|██▉       | 1919/6470 [24:01<1:01:26,  1.23it/s]
2|1920|Loss: 1.0416946411132812:  30%|██▉       | 1919/6470 [24:02<1:01:26,  1.23it/s]
2|1920|Loss: 1.0416946411132812:  30%|██▉       | 1920/6470 [24:02<1:03:54,  1.19it/s]
2|1921|Loss: 0.7217988967895508:  30%|██▉       | 1920/6470 [24:03<1:03:54,  1.19it/s]
2|1921|Loss: 0.7217988967895508:  30%|██▉       | 1921/6470 [24:03<1:01:05,  1.24it/s]
2|1922|Loss: 0.7221232056617737:  30%|██▉       | 1921/6470 [24:03<1:01:05,  1.24it/s]
2|1922|Loss: 0.7221232056617737:  30%|██▉       | 1922/6470 [24:04<1:02:29,  1.21it/s]
2|1923|Loss: 1.171413540840149:  30%|██▉       | 1922/6470 [24:04<1:02:29,  1.21it/s] 
2|1923|Loss: 1.171413540840149:  30%|██▉       | 1923/6470 [24:05<1:02:07,  1.22it/s]
2|1924|Loss: 0.9634689092636108:  30%|██▉       | 1923/6470 [24:05<1:02:07,  1.22it/s]
2|1924|Loss: 0.9634689092636108:  30%|██▉       | 1924/6470 [24:05<1:01:20,  1.24it/s]
2|1925|Loss: 0.7545945048332214:  30%|██▉       | 1924/6470 [24:06<1:01:20,  1.24it/s]
2|1925|Loss: 0.7545945048332214:  30%|██▉       | 1925/6470 [24:06<1:03:22,  1.20it/s]
2|1926|Loss: 0.9273982644081116:  30%|██▉       | 1925/6470 [24:07<1:03:22,  1.20it/s]
2|1926|Loss: 0.9273982644081116:  30%|██▉       | 1926/6470 [24:07<1:03:21,  1.20it/s]
2|1927|Loss: 0.8175957798957825:  30%|██▉       | 1926/6470 [24:08<1:03:21,  1.20it/s]
2|1927|Loss: 0.8175957798957825:  30%|██▉       | 1927/6470 [24:08<1:02:16,  1.22it/s]
2|1928|Loss: 0.850620448589325:  30%|██▉       | 1927/6470 [24:08<1:02:16,  1.22it/s] 
2|1928|Loss: 0.850620448589325:  30%|██▉       | 1928/6470 [24:09<1:03:03,  1.20it/s]
2|1929|Loss: 0.7056642770767212:  30%|██▉       | 1928/6470 [24:09<1:03:03,  1.20it/s]
2|1929|Loss: 0.7056642770767212:  30%|██▉       | 1929/6470 [24:10<1:04:37,  1.17it/s]
2|1930|Loss: 1.0058667659759521:  30%|██▉       | 1929/6470 [24:10<1:04:37,  1.17it/s]
2|1930|Loss: 1.0058667659759521:  30%|██▉       | 1930/6470 [24:10<1:01:31,  1.23it/s]
2|1931|Loss: 0.8621594309806824:  30%|██▉       | 1930/6470 [24:11<1:01:31,  1.23it/s]
2|1931|Loss: 0.8621594309806824:  30%|██▉       | 1931/6470 [24:11<1:00:48,  1.24it/s]
2|1932|Loss: 0.7441390752792358:  30%|██▉       | 1931/6470 [24:12<1:00:48,  1.24it/s]
2|1932|Loss: 0.7441390752792358:  30%|██▉       | 1932/6470 [24:12<57:32,  1.31it/s]  
2|1933|Loss: 0.7382537126541138:  30%|██▉       | 1932/6470 [24:12<57:32,  1.31it/s]
2|1933|Loss: 0.7382537126541138:  30%|██▉       | 1933/6470 [24:12<52:19,  1.45it/s]
2|1934|Loss: 0.8014489412307739:  30%|██▉       | 1933/6470 [24:13<52:19,  1.45it/s]
2|1934|Loss: 0.8014489412307739:  30%|██▉       | 1934/6470 [24:13<56:18,  1.34it/s]
2|1935|Loss: 0.8640118837356567:  30%|██▉       | 1934/6470 [24:14<56:18,  1.34it/s]
2|1935|Loss: 0.8640118837356567:  30%|██▉       | 1935/6470 [24:14<55:20,  1.37it/s]
2|1936|Loss: 0.9018976092338562:  30%|██▉       | 1935/6470 [24:14<55:20,  1.37it/s]
2|1936|Loss: 0.9018976092338562:  30%|██▉       | 1936/6470 [24:15<55:13,  1.37it/s]
2|1937|Loss: 0.7222020030021667:  30%|██▉       | 1936/6470 [24:15<55:13,  1.37it/s]
2|1937|Loss: 0.7222020030021667:  30%|██▉       | 1937/6470 [24:15<51:23,  1.47it/s]
2|1938|Loss: 0.9130188226699829:  30%|██▉       | 1937/6470 [24:16<51:23,  1.47it/s]
2|1938|Loss: 0.9130188226699829:  30%|██▉       | 1938/6470 [24:16<49:33,  1.52it/s]
2|1939|Loss: 0.8920025825500488:  30%|██▉       | 1938/6470 [24:16<49:33,  1.52it/s]
2|1939|Loss: 0.8920025825500488:  30%|██▉       | 1939/6470 [24:16<47:35,  1.59it/s]
2|1940|Loss: 0.898614227771759:  30%|██▉       | 1939/6470 [24:17<47:35,  1.59it/s] 
2|1940|Loss: 0.898614227771759:  30%|██▉       | 1940/6470 [24:17<49:18,  1.53it/s]
2|1941|Loss: 0.8954006433486938:  30%|██▉       | 1940/6470 [24:18<49:18,  1.53it/s]
2|1941|Loss: 0.8954006433486938:  30%|███       | 1941/6470 [24:18<55:49,  1.35it/s]
2|1942|Loss: 1.039415955543518:  30%|███       | 1941/6470 [24:18<55:49,  1.35it/s] 
2|1942|Loss: 1.039415955543518:  30%|███       | 1942/6470 [24:19<52:39,  1.43it/s]
2|1943|Loss: 0.8904176950454712:  30%|███       | 1942/6470 [24:19<52:39,  1.43it/s]
2|1943|Loss: 0.8904176950454712:  30%|███       | 1943/6470 [24:19<53:26,  1.41it/s]
2|1944|Loss: 0.8939991593360901:  30%|███       | 1943/6470 [24:20<53:26,  1.41it/s]
2|1944|Loss: 0.8939991593360901:  30%|███       | 1944/6470 [24:20<54:46,  1.38it/s]
2|1945|Loss: 0.7925894260406494:  30%|███       | 1944/6470 [24:20<54:46,  1.38it/s]
2|1945|Loss: 0.7925894260406494:  30%|███       | 1945/6470 [24:21<52:59,  1.42it/s]
2|1946|Loss: 0.9562848806381226:  30%|███       | 1945/6470 [24:21<52:59,  1.42it/s]
2|1946|Loss: 0.9562848806381226:  30%|███       | 1946/6470 [24:21<51:35,  1.46it/s]
2|1947|Loss: 0.9769785404205322:  30%|███       | 1946/6470 [24:22<51:35,  1.46it/s]
2|1947|Loss: 0.9769785404205322:  30%|███       | 1947/6470 [24:22<54:46,  1.38it/s]
2|1948|Loss: 0.9214929342269897:  30%|███       | 1947/6470 [24:23<54:46,  1.38it/s]
2|1948|Loss: 0.9214929342269897:  30%|███       | 1948/6470 [24:23<56:34,  1.33it/s]
2|1949|Loss: 0.7720972895622253:  30%|███       | 1948/6470 [24:23<56:34,  1.33it/s]
2|1949|Loss: 0.7720972895622253:  30%|███       | 1949/6470 [24:24<55:38,  1.35it/s]
2|1950|Loss: 0.9542512893676758:  30%|███       | 1949/6470 [24:24<55:38,  1.35it/s]
2|1950|Loss: 0.9542512893676758:  30%|███       | 1950/6470 [24:24<49:00,  1.54it/s]
2|1951|Loss: 0.7614671587944031:  30%|███       | 1950/6470 [24:25<49:00,  1.54it/s]
2|1951|Loss: 0.7614671587944031:  30%|███       | 1951/6470 [24:25<50:33,  1.49it/s]
2|1952|Loss: 0.8906162977218628:  30%|███       | 1951/6470 [24:25<50:33,  1.49it/s]
2|1952|Loss: 0.8906162977218628:  30%|███       | 1952/6470 [24:26<49:09,  1.53it/s]
2|1953|Loss: 0.8019182085990906:  30%|███       | 1952/6470 [24:26<49:09,  1.53it/s]
2|1953|Loss: 0.8019182085990906:  30%|███       | 1953/6470 [24:26<46:59,  1.60it/s]
2|1954|Loss: 0.8254460096359253:  30%|███       | 1953/6470 [24:26<46:59,  1.60it/s]
2|1954|Loss: 0.8254460096359253:  30%|███       | 1954/6470 [24:27<47:23,  1.59it/s]
2|1955|Loss: 1.0192967653274536:  30%|███       | 1954/6470 [24:27<47:23,  1.59it/s]
2|1955|Loss: 1.0192967653274536:  30%|███       | 1955/6470 [24:27<50:35,  1.49it/s]
2|1956|Loss: 0.9085587859153748:  30%|███       | 1955/6470 [24:28<50:35,  1.49it/s]
2|1956|Loss: 0.9085587859153748:  30%|███       | 1956/6470 [24:28<52:21,  1.44it/s]
2|1957|Loss: 0.8506121635437012:  30%|███       | 1956/6470 [24:29<52:21,  1.44it/s]
2|1957|Loss: 0.8506121635437012:  30%|███       | 1957/6470 [24:29<49:45,  1.51it/s]
2|1958|Loss: 0.856087327003479:  30%|███       | 1957/6470 [24:29<49:45,  1.51it/s] 
2|1958|Loss: 0.856087327003479:  30%|███       | 1958/6470 [24:29<49:50,  1.51it/s]
2|1959|Loss: 0.8918924927711487:  30%|███       | 1958/6470 [24:30<49:50,  1.51it/s]
2|1959|Loss: 0.8918924927711487:  30%|███       | 1959/6470 [24:30<48:22,  1.55it/s]
2|1960|Loss: 0.8373774290084839:  30%|███       | 1959/6470 [24:30<48:22,  1.55it/s]
2|1960|Loss: 0.8373774290084839:  30%|███       | 1960/6470 [24:31<51:01,  1.47it/s]
2|1961|Loss: 0.7270976901054382:  30%|███       | 1960/6470 [24:31<51:01,  1.47it/s]
2|1961|Loss: 0.7270976901054382:  30%|███       | 1961/6470 [24:32<56:30,  1.33it/s]
2|1962|Loss: 0.848318338394165:  30%|███       | 1961/6470 [24:32<56:30,  1.33it/s] 
2|1962|Loss: 0.848318338394165:  30%|███       | 1962/6470 [24:32<55:10,  1.36it/s]
2|1963|Loss: 0.7751370072364807:  30%|███       | 1962/6470 [24:33<55:10,  1.36it/s]
2|1963|Loss: 0.7751370072364807:  30%|███       | 1963/6470 [24:33<55:48,  1.35it/s]
2|1964|Loss: 0.9083497524261475:  30%|███       | 1963/6470 [24:34<55:48,  1.35it/s]
2|1964|Loss: 0.9083497524261475:  30%|███       | 1964/6470 [24:34<59:48,  1.26it/s]
2|1965|Loss: 0.9758859872817993:  30%|███       | 1964/6470 [24:35<59:48,  1.26it/s]
2|1965|Loss: 0.9758859872817993:  30%|███       | 1965/6470 [24:35<59:51,  1.25it/s]
2|1966|Loss: 0.9496964812278748:  30%|███       | 1965/6470 [24:35<59:51,  1.25it/s]
2|1966|Loss: 0.9496964812278748:  30%|███       | 1966/6470 [24:36<57:34,  1.30it/s]
2|1967|Loss: 0.8849035501480103:  30%|███       | 1966/6470 [24:36<57:34,  1.30it/s]
2|1967|Loss: 0.8849035501480103:  30%|███       | 1967/6470 [24:36<55:55,  1.34it/s]
2|1968|Loss: 0.841280996799469:  30%|███       | 1967/6470 [24:37<55:55,  1.34it/s] 
2|1968|Loss: 0.841280996799469:  30%|███       | 1968/6470 [24:37<54:49,  1.37it/s]
2|1969|Loss: 0.8897197246551514:  30%|███       | 1968/6470 [24:38<54:49,  1.37it/s]
2|1969|Loss: 0.8897197246551514:  30%|███       | 1969/6470 [24:38<58:24,  1.28it/s]
2|1970|Loss: 0.9469103813171387:  30%|███       | 1969/6470 [24:38<58:24,  1.28it/s]
2|1970|Loss: 0.9469103813171387:  30%|███       | 1970/6470 [24:39<58:00,  1.29it/s]
2|1971|Loss: 0.9317396879196167:  30%|███       | 1970/6470 [24:39<58:00,  1.29it/s]
2|1971|Loss: 0.9317396879196167:  30%|███       | 1971/6470 [24:39<55:12,  1.36it/s]
2|1972|Loss: 0.806674063205719:  30%|███       | 1971/6470 [24:40<55:12,  1.36it/s] 
2|1972|Loss: 0.806674063205719:  30%|███       | 1972/6470 [24:40<56:11,  1.33it/s]
2|1973|Loss: 0.795033872127533:  30%|███       | 1972/6470 [24:41<56:11,  1.33it/s]
2|1973|Loss: 0.795033872127533:  30%|███       | 1973/6470 [24:41<55:39,  1.35it/s]
2|1974|Loss: 0.8807880878448486:  30%|███       | 1973/6470 [24:41<55:39,  1.35it/s]
2|1974|Loss: 0.8807880878448486:  31%|███       | 1974/6470 [24:41<53:38,  1.40it/s]
2|1975|Loss: 0.9163153171539307:  31%|███       | 1974/6470 [24:42<53:38,  1.40it/s]
2|1975|Loss: 0.9163153171539307:  31%|███       | 1975/6470 [24:42<53:58,  1.39it/s]
2|1976|Loss: 0.7890375256538391:  31%|███       | 1975/6470 [24:43<53:58,  1.39it/s]
2|1976|Loss: 0.7890375256538391:  31%|███       | 1976/6470 [24:43<50:56,  1.47it/s]
2|1977|Loss: 0.8299862742424011:  31%|███       | 1976/6470 [24:43<50:56,  1.47it/s]
2|1977|Loss: 0.8299862742424011:  31%|███       | 1977/6470 [24:44<54:54,  1.36it/s]
2|1978|Loss: 0.7440468072891235:  31%|███       | 1977/6470 [24:44<54:54,  1.36it/s]
2|1978|Loss: 0.7440468072891235:  31%|███       | 1978/6470 [24:44<49:52,  1.50it/s]
2|1979|Loss: 0.8752747178077698:  31%|███       | 1978/6470 [24:44<49:52,  1.50it/s]
2|1979|Loss: 0.8752747178077698:  31%|███       | 1979/6470 [24:45<46:23,  1.61it/s]
2|1980|Loss: 0.955491840839386:  31%|███       | 1979/6470 [24:45<46:23,  1.61it/s] 
2|1980|Loss: 0.955491840839386:  31%|███       | 1980/6470 [24:45<48:40,  1.54it/s]
2|1981|Loss: 0.8956158757209778:  31%|███       | 1980/6470 [24:46<48:40,  1.54it/s]
2|1981|Loss: 0.8956158757209778:  31%|███       | 1981/6470 [24:46<53:38,  1.39it/s]
2|1982|Loss: 0.8486028909683228:  31%|███       | 1981/6470 [24:47<53:38,  1.39it/s]
2|1982|Loss: 0.8486028909683228:  31%|███       | 1982/6470 [24:47<51:44,  1.45it/s]
2|1983|Loss: 0.7280654311180115:  31%|███       | 1982/6470 [24:47<51:44,  1.45it/s]
2|1983|Loss: 0.7280654311180115:  31%|███       | 1983/6470 [24:48<50:35,  1.48it/s]
2|1984|Loss: 0.84591144323349:  31%|███       | 1983/6470 [24:48<50:35,  1.48it/s]  
2|1984|Loss: 0.84591144323349:  31%|███       | 1984/6470 [24:48<50:46,  1.47it/s]
2|1985|Loss: 0.8939897418022156:  31%|███       | 1984/6470 [24:49<50:46,  1.47it/s]
2|1985|Loss: 0.8939897418022156:  31%|███       | 1985/6470 [24:49<51:00,  1.47it/s]
2|1986|Loss: 0.809969425201416:  31%|███       | 1985/6470 [24:49<51:00,  1.47it/s] 
2|1986|Loss: 0.809969425201416:  31%|███       | 1986/6470 [24:49<47:25,  1.58it/s]
2|1987|Loss: 0.9638770818710327:  31%|███       | 1986/6470 [24:50<47:25,  1.58it/s]
2|1987|Loss: 0.9638770818710327:  31%|███       | 1987/6470 [24:50<51:41,  1.45it/s]
2|1988|Loss: 0.8723963499069214:  31%|███       | 1987/6470 [24:51<51:41,  1.45it/s]
2|1988|Loss: 0.8723963499069214:  31%|███       | 1988/6470 [24:51<53:50,  1.39it/s]
2|1989|Loss: 0.9479062557220459:  31%|███       | 1988/6470 [24:52<53:50,  1.39it/s]
2|1989|Loss: 0.9479062557220459:  31%|███       | 1989/6470 [24:52<56:46,  1.32it/s]
2|1990|Loss: 0.792782187461853:  31%|███       | 1989/6470 [24:52<56:46,  1.32it/s] 
2|1990|Loss: 0.792782187461853:  31%|███       | 1990/6470 [24:52<52:19,  1.43it/s]
2|1991|Loss: 0.8691588044166565:  31%|███       | 1990/6470 [24:53<52:19,  1.43it/s]
2|1991|Loss: 0.8691588044166565:  31%|███       | 1991/6470 [24:53<52:11,  1.43it/s]
2|1992|Loss: 0.9140506386756897:  31%|███       | 1991/6470 [24:54<52:11,  1.43it/s]
2|1992|Loss: 0.9140506386756897:  31%|███       | 1992/6470 [24:54<53:38,  1.39it/s]
2|1993|Loss: 0.8410040140151978:  31%|███       | 1992/6470 [24:54<53:38,  1.39it/s]
2|1993|Loss: 0.8410040140151978:  31%|███       | 1993/6470 [24:55<58:18,  1.28it/s]
2|1994|Loss: 0.874171257019043:  31%|███       | 1993/6470 [24:55<58:18,  1.28it/s] 
2|1994|Loss: 0.874171257019043:  31%|███       | 1994/6470 [24:56<58:53,  1.27it/s]
2|1995|Loss: 0.8694011569023132:  31%|███       | 1994/6470 [24:56<58:53,  1.27it/s]
2|1995|Loss: 0.8694011569023132:  31%|███       | 1995/6470 [24:56<57:14,  1.30it/s]
2|1996|Loss: 0.794035792350769:  31%|███       | 1995/6470 [24:57<57:14,  1.30it/s] 
2|1996|Loss: 0.794035792350769:  31%|███       | 1996/6470 [24:57<1:00:03,  1.24it/s]
2|1997|Loss: 0.9198905229568481:  31%|███       | 1996/6470 [24:58<1:00:03,  1.24it/s]
2|1997|Loss: 0.9198905229568481:  31%|███       | 1997/6470 [24:58<1:02:12,  1.20it/s]
2|1998|Loss: 0.866498589515686:  31%|███       | 1997/6470 [24:59<1:02:12,  1.20it/s] 
2|1998|Loss: 0.866498589515686:  31%|███       | 1998/6470 [24:59<1:03:50,  1.17it/s]
2|1999|Loss: 0.9081647396087646:  31%|███       | 1998/6470 [25:00<1:03:50,  1.17it/s]
2|1999|Loss: 0.9081647396087646:  31%|███       | 1999/6470 [25:00<1:00:47,  1.23it/s]
2|2000|Loss: 0.8298199772834778:  31%|███       | 1999/6470 [25:00<1:00:47,  1.23it/s]
2|2000|Loss: 0.8298199772834778:  31%|███       | 2000/6470 [25:00<57:39,  1.29it/s]  
2|2001|Loss: 0.6832205057144165:  31%|███       | 2000/6470 [25:01<57:39,  1.29it/s]
2|2001|Loss: 0.6832205057144165:  31%|███       | 2001/6470 [25:01<1:01:24,  1.21it/s]
2|2002|Loss: 0.782686710357666:  31%|███       | 2001/6470 [25:02<1:01:24,  1.21it/s] 
2|2002|Loss: 0.782686710357666:  31%|███       | 2002/6470 [25:02<1:00:12,  1.24it/s]
2|2003|Loss: 0.8920173645019531:  31%|███       | 2002/6470 [25:03<1:00:12,  1.24it/s]
2|2003|Loss: 0.8920173645019531:  31%|███       | 2003/6470 [25:03<59:48,  1.24it/s]  
2|2004|Loss: 0.9674027562141418:  31%|███       | 2003/6470 [25:03<59:48,  1.24it/s]
2|2004|Loss: 0.9674027562141418:  31%|███       | 2004/6470 [25:04<59:38,  1.25it/s]
2|2005|Loss: 0.8950017690658569:  31%|███       | 2004/6470 [25:04<59:38,  1.25it/s]
2|2005|Loss: 0.8950017690658569:  31%|███       | 2005/6470 [25:05<57:51,  1.29it/s]
2|2006|Loss: 0.7881001234054565:  31%|███       | 2005/6470 [25:05<57:51,  1.29it/s]
2|2006|Loss: 0.7881001234054565:  31%|███       | 2006/6470 [25:05<58:06,  1.28it/s]
2|2007|Loss: 0.7845512628555298:  31%|███       | 2006/6470 [25:06<58:06,  1.28it/s]
2|2007|Loss: 0.7845512628555298:  31%|███       | 2007/6470 [25:06<55:40,  1.34it/s]
2|2008|Loss: 0.7303336262702942:  31%|███       | 2007/6470 [25:06<55:40,  1.34it/s]
2|2008|Loss: 0.7303336262702942:  31%|███       | 2008/6470 [25:06<48:38,  1.53it/s]
2|2009|Loss: 0.8894984722137451:  31%|███       | 2008/6470 [25:07<48:38,  1.53it/s]
2|2009|Loss: 0.8894984722137451:  31%|███       | 2009/6470 [25:07<45:36,  1.63it/s]
2|2010|Loss: 0.7690721154212952:  31%|███       | 2009/6470 [25:07<45:36,  1.63it/s]
2|2010|Loss: 0.7690721154212952:  31%|███       | 2010/6470 [25:08<47:43,  1.56it/s]
2|2011|Loss: 0.841306746006012:  31%|███       | 2010/6470 [25:08<47:43,  1.56it/s] 
2|2011|Loss: 0.841306746006012:  31%|███       | 2011/6470 [25:08<45:19,  1.64it/s]
2|2012|Loss: 0.7974108457565308:  31%|███       | 2011/6470 [25:09<45:19,  1.64it/s]
2|2012|Loss: 0.7974108457565308:  31%|███       | 2012/6470 [25:09<45:43,  1.62it/s]
2|2013|Loss: 0.8571712374687195:  31%|███       | 2012/6470 [25:09<45:43,  1.62it/s]
2|2013|Loss: 0.8571712374687195:  31%|███       | 2013/6470 [25:09<46:20,  1.60it/s]
2|2014|Loss: 0.8151755332946777:  31%|███       | 2013/6470 [25:10<46:20,  1.60it/s]
2|2014|Loss: 0.8151755332946777:  31%|███       | 2014/6470 [25:10<46:17,  1.60it/s]
2|2015|Loss: 0.7149540781974792:  31%|███       | 2014/6470 [25:11<46:17,  1.60it/s]
2|2015|Loss: 0.7149540781974792:  31%|███       | 2015/6470 [25:11<52:24,  1.42it/s]
2|2016|Loss: 0.8376208543777466:  31%|███       | 2015/6470 [25:11<52:24,  1.42it/s]
2|2016|Loss: 0.8376208543777466:  31%|███       | 2016/6470 [25:12<55:38,  1.33it/s]
2|2017|Loss: 0.917455792427063:  31%|███       | 2016/6470 [25:12<55:38,  1.33it/s] 
2|2017|Loss: 0.917455792427063:  31%|███       | 2017/6470 [25:13<56:41,  1.31it/s]
2|2018|Loss: 0.8935052156448364:  31%|███       | 2017/6470 [25:13<56:41,  1.31it/s]
2|2018|Loss: 0.8935052156448364:  31%|███       | 2018/6470 [25:13<56:52,  1.30it/s]
2|2019|Loss: 0.7183212637901306:  31%|███       | 2018/6470 [25:14<56:52,  1.30it/s]
2|2019|Loss: 0.7183212637901306:  31%|███       | 2019/6470 [25:14<57:15,  1.30it/s]
2|2020|Loss: 0.9172791838645935:  31%|███       | 2019/6470 [25:15<57:15,  1.30it/s]
2|2020|Loss: 0.9172791838645935:  31%|███       | 2020/6470 [25:15<55:32,  1.34it/s]
2|2021|Loss: 0.7799006700515747:  31%|███       | 2020/6470 [25:15<55:32,  1.34it/s]
2|2021|Loss: 0.7799006700515747:  31%|███       | 2021/6470 [25:16<57:36,  1.29it/s]
2|2022|Loss: 0.8586345314979553:  31%|███       | 2021/6470 [25:16<57:36,  1.29it/s]
2|2022|Loss: 0.8586345314979553:  31%|███▏      | 2022/6470 [25:17<58:28,  1.27it/s]
2|2023|Loss: 0.81160569190979:  31%|███▏      | 2022/6470 [25:17<58:28,  1.27it/s]  
2|2023|Loss: 0.81160569190979:  31%|███▏      | 2023/6470 [25:17<56:54,  1.30it/s]
2|2024|Loss: 0.6481188535690308:  31%|███▏      | 2023/6470 [25:18<56:54,  1.30it/s]
2|2024|Loss: 0.6481188535690308:  31%|███▏      | 2024/6470 [25:18<52:34,  1.41it/s]
2|2025|Loss: 0.8036207556724548:  31%|███▏      | 2024/6470 [25:18<52:34,  1.41it/s]
2|2025|Loss: 0.8036207556724548:  31%|███▏      | 2025/6470 [25:19<52:41,  1.41it/s]
2|2026|Loss: 0.6898198127746582:  31%|███▏      | 2025/6470 [25:19<52:41,  1.41it/s]
2|2026|Loss: 0.6898198127746582:  31%|███▏      | 2026/6470 [25:19<49:28,  1.50it/s]
2|2027|Loss: 0.8349048495292664:  31%|███▏      | 2026/6470 [25:19<49:28,  1.50it/s]
2|2027|Loss: 0.8349048495292664:  31%|███▏      | 2027/6470 [25:20<50:18,  1.47it/s]
2|2028|Loss: 0.7595476508140564:  31%|███▏      | 2027/6470 [25:20<50:18,  1.47it/s]
2|2028|Loss: 0.7595476508140564:  31%|███▏      | 2028/6470 [25:21<54:47,  1.35it/s]
2|2029|Loss: 0.9029795527458191:  31%|███▏      | 2028/6470 [25:21<54:47,  1.35it/s]
2|2029|Loss: 0.9029795527458191:  31%|███▏      | 2029/6470 [25:21<49:21,  1.50it/s]
2|2030|Loss: 0.7056944966316223:  31%|███▏      | 2029/6470 [25:22<49:21,  1.50it/s]
2|2030|Loss: 0.7056944966316223:  31%|███▏      | 2030/6470 [25:22<50:14,  1.47it/s]
2|2031|Loss: 0.650847315788269:  31%|███▏      | 2030/6470 [25:22<50:14,  1.47it/s] 
2|2031|Loss: 0.650847315788269:  31%|███▏      | 2031/6470 [25:23<50:12,  1.47it/s]
2|2032|Loss: 1.043135643005371:  31%|███▏      | 2031/6470 [25:23<50:12,  1.47it/s]
2|2032|Loss: 1.043135643005371:  31%|███▏      | 2032/6470 [25:23<53:42,  1.38it/s]
2|2033|Loss: 0.8966876864433289:  31%|███▏      | 2032/6470 [25:24<53:42,  1.38it/s]
2|2033|Loss: 0.8966876864433289:  31%|███▏      | 2033/6470 [25:24<53:35,  1.38it/s]
2|2034|Loss: 0.8033353090286255:  31%|███▏      | 2033/6470 [25:25<53:35,  1.38it/s]
2|2034|Loss: 0.8033353090286255:  31%|███▏      | 2034/6470 [25:25<56:06,  1.32it/s]
2|2035|Loss: 0.759749710559845:  31%|███▏      | 2034/6470 [25:25<56:06,  1.32it/s] 
2|2035|Loss: 0.759749710559845:  31%|███▏      | 2035/6470 [25:25<49:54,  1.48it/s]
2|2036|Loss: 0.9499542713165283:  31%|███▏      | 2035/6470 [25:26<49:54,  1.48it/s]
2|2036|Loss: 0.9499542713165283:  31%|███▏      | 2036/6470 [25:26<52:07,  1.42it/s]
2|2037|Loss: 0.8478014469146729:  31%|███▏      | 2036/6470 [25:27<52:07,  1.42it/s]
2|2037|Loss: 0.8478014469146729:  31%|███▏      | 2037/6470 [25:27<57:13,  1.29it/s]
2|2038|Loss: 0.8416369557380676:  31%|███▏      | 2037/6470 [25:28<57:13,  1.29it/s]
2|2038|Loss: 0.8416369557380676:  31%|███▏      | 2038/6470 [25:28<57:56,  1.27it/s]
2|2039|Loss: 0.7429143786430359:  31%|███▏      | 2038/6470 [25:28<57:56,  1.27it/s]
2|2039|Loss: 0.7429143786430359:  32%|███▏      | 2039/6470 [25:29<59:43,  1.24it/s]
2|2040|Loss: 0.9151882529258728:  32%|███▏      | 2039/6470 [25:29<59:43,  1.24it/s]
2|2040|Loss: 0.9151882529258728:  32%|███▏      | 2040/6470 [25:29<55:34,  1.33it/s]
2|2041|Loss: 0.9082557559013367:  32%|███▏      | 2040/6470 [25:30<55:34,  1.33it/s]
2|2041|Loss: 0.9082557559013367:  32%|███▏      | 2041/6470 [25:30<52:58,  1.39it/s]
2|2042|Loss: 0.8956771492958069:  32%|███▏      | 2041/6470 [25:31<52:58,  1.39it/s]
2|2042|Loss: 0.8956771492958069:  32%|███▏      | 2042/6470 [25:31<55:56,  1.32it/s]
2|2043|Loss: 0.8330860137939453:  32%|███▏      | 2042/6470 [25:31<55:56,  1.32it/s]
2|2043|Loss: 0.8330860137939453:  32%|███▏      | 2043/6470 [25:31<49:46,  1.48it/s]
2|2044|Loss: 0.7898615002632141:  32%|███▏      | 2043/6470 [25:32<49:46,  1.48it/s]
2|2044|Loss: 0.7898615002632141:  32%|███▏      | 2044/6470 [25:32<52:04,  1.42it/s]
2|2045|Loss: 0.9546544551849365:  32%|███▏      | 2044/6470 [25:33<52:04,  1.42it/s]
2|2045|Loss: 0.9546544551849365:  32%|███▏      | 2045/6470 [25:33<53:59,  1.37it/s]
2|2046|Loss: 0.6972696781158447:  32%|███▏      | 2045/6470 [25:33<53:59,  1.37it/s]
2|2046|Loss: 0.6972696781158447:  32%|███▏      | 2046/6470 [25:34<55:25,  1.33it/s]
2|2047|Loss: 0.8473143577575684:  32%|███▏      | 2046/6470 [25:34<55:25,  1.33it/s]
2|2047|Loss: 0.8473143577575684:  32%|███▏      | 2047/6470 [25:34<53:27,  1.38it/s]
2|2048|Loss: 0.8662164211273193:  32%|███▏      | 2047/6470 [25:35<53:27,  1.38it/s]
2|2048|Loss: 0.8662164211273193:  32%|███▏      | 2048/6470 [25:35<54:20,  1.36it/s]
2|2049|Loss: 0.9112516045570374:  32%|███▏      | 2048/6470 [25:36<54:20,  1.36it/s]
2|2049|Loss: 0.9112516045570374:  32%|███▏      | 2049/6470 [25:36<51:38,  1.43it/s]
2|2050|Loss: 0.8342323899269104:  32%|███▏      | 2049/6470 [25:36<51:38,  1.43it/s]
2|2050|Loss: 0.8342323899269104:  32%|███▏      | 2050/6470 [25:36<50:29,  1.46it/s]
2|2051|Loss: 0.9038546681404114:  32%|███▏      | 2050/6470 [25:37<50:29,  1.46it/s]
2|2051|Loss: 0.9038546681404114:  32%|███▏      | 2051/6470 [25:37<52:09,  1.41it/s]
2|2052|Loss: 0.8680325150489807:  32%|███▏      | 2051/6470 [25:38<52:09,  1.41it/s]
2|2052|Loss: 0.8680325150489807:  32%|███▏      | 2052/6470 [25:38<52:32,  1.40it/s]
2|2053|Loss: 0.810634434223175:  32%|███▏      | 2052/6470 [25:38<52:32,  1.40it/s] 
2|2053|Loss: 0.810634434223175:  32%|███▏      | 2053/6470 [25:39<49:39,  1.48it/s]
2|2054|Loss: 0.8217649459838867:  32%|███▏      | 2053/6470 [25:39<49:39,  1.48it/s]
2|2054|Loss: 0.8217649459838867:  32%|███▏      | 2054/6470 [25:39<55:02,  1.34it/s]
2|2055|Loss: 0.7946664690971375:  32%|███▏      | 2054/6470 [25:40<55:02,  1.34it/s]
2|2055|Loss: 0.7946664690971375:  32%|███▏      | 2055/6470 [25:40<59:43,  1.23it/s]
2|2056|Loss: 0.9385371208190918:  32%|███▏      | 2055/6470 [25:41<59:43,  1.23it/s]
2|2056|Loss: 0.9385371208190918:  32%|███▏      | 2056/6470 [25:41<56:38,  1.30it/s]
2|2057|Loss: 0.8368498086929321:  32%|███▏      | 2056/6470 [25:42<56:38,  1.30it/s]
2|2057|Loss: 0.8368498086929321:  32%|███▏      | 2057/6470 [25:42<56:40,  1.30it/s]
2|2058|Loss: 0.705580472946167:  32%|███▏      | 2057/6470 [25:42<56:40,  1.30it/s] 
2|2058|Loss: 0.705580472946167:  32%|███▏      | 2058/6470 [25:43<57:21,  1.28it/s]
2|2059|Loss: 0.8487931489944458:  32%|███▏      | 2058/6470 [25:43<57:21,  1.28it/s]
2|2059|Loss: 0.8487931489944458:  32%|███▏      | 2059/6470 [25:43<54:58,  1.34it/s]
2|2060|Loss: 0.9194434881210327:  32%|███▏      | 2059/6470 [25:44<54:58,  1.34it/s]
2|2060|Loss: 0.9194434881210327:  32%|███▏      | 2060/6470 [25:44<54:31,  1.35it/s]
2|2061|Loss: 0.8011584877967834:  32%|███▏      | 2060/6470 [25:45<54:31,  1.35it/s]
2|2061|Loss: 0.8011584877967834:  32%|███▏      | 2061/6470 [25:45<54:41,  1.34it/s]
2|2062|Loss: 0.8148311376571655:  32%|███▏      | 2061/6470 [25:45<54:41,  1.34it/s]
2|2062|Loss: 0.8148311376571655:  32%|███▏      | 2062/6470 [25:46<54:04,  1.36it/s]
2|2063|Loss: 0.8836956024169922:  32%|███▏      | 2062/6470 [25:46<54:04,  1.36it/s]
2|2063|Loss: 0.8836956024169922:  32%|███▏      | 2063/6470 [25:46<56:33,  1.30it/s]
2|2064|Loss: 0.8804579973220825:  32%|███▏      | 2063/6470 [25:47<56:33,  1.30it/s]
2|2064|Loss: 0.8804579973220825:  32%|███▏      | 2064/6470 [25:47<1:00:51,  1.21it/s]
2|2065|Loss: 0.7246460914611816:  32%|███▏      | 2064/6470 [25:48<1:00:51,  1.21it/s]
2|2065|Loss: 0.7246460914611816:  32%|███▏      | 2065/6470 [25:48<1:01:21,  1.20it/s]
2|2066|Loss: 0.8690275549888611:  32%|███▏      | 2065/6470 [25:49<1:01:21,  1.20it/s]
2|2066|Loss: 0.8690275549888611:  32%|███▏      | 2066/6470 [25:49<56:08,  1.31it/s]  
2|2067|Loss: 1.0833791494369507:  32%|███▏      | 2066/6470 [25:49<56:08,  1.31it/s]
2|2067|Loss: 1.0833791494369507:  32%|███▏      | 2067/6470 [25:50<56:48,  1.29it/s]
2|2068|Loss: 0.8851765394210815:  32%|███▏      | 2067/6470 [25:50<56:48,  1.29it/s]
2|2068|Loss: 0.8851765394210815:  32%|███▏      | 2068/6470 [25:50<56:19,  1.30it/s]
2|2069|Loss: 0.8494336009025574:  32%|███▏      | 2068/6470 [25:51<56:19,  1.30it/s]
2|2069|Loss: 0.8494336009025574:  32%|███▏      | 2069/6470 [25:51<53:15,  1.38it/s]
2|2070|Loss: 0.862095296382904:  32%|███▏      | 2069/6470 [25:51<53:15,  1.38it/s] 
2|2070|Loss: 0.862095296382904:  32%|███▏      | 2070/6470 [25:52<56:12,  1.30it/s]
2|2071|Loss: 0.7800807952880859:  32%|███▏      | 2070/6470 [25:52<56:12,  1.30it/s]
2|2071|Loss: 0.7800807952880859:  32%|███▏      | 2071/6470 [25:53<1:00:00,  1.22it/s]
2|2072|Loss: 0.8458062410354614:  32%|███▏      | 2071/6470 [25:53<1:00:00,  1.22it/s]
2|2072|Loss: 0.8458062410354614:  32%|███▏      | 2072/6470 [25:54<58:47,  1.25it/s]  
2|2073|Loss: 0.9385781288146973:  32%|███▏      | 2072/6470 [25:54<58:47,  1.25it/s]
2|2073|Loss: 0.9385781288146973:  32%|███▏      | 2073/6470 [25:54<57:53,  1.27it/s]
2|2074|Loss: 0.7578058242797852:  32%|███▏      | 2073/6470 [25:55<57:53,  1.27it/s]
2|2074|Loss: 0.7578058242797852:  32%|███▏      | 2074/6470 [25:55<59:03,  1.24it/s]
2|2075|Loss: 0.8423101305961609:  32%|███▏      | 2074/6470 [25:56<59:03,  1.24it/s]
2|2075|Loss: 0.8423101305961609:  32%|███▏      | 2075/6470 [25:56<58:53,  1.24it/s]
2|2076|Loss: 0.8925795555114746:  32%|███▏      | 2075/6470 [25:56<58:53,  1.24it/s]
2|2076|Loss: 0.8925795555114746:  32%|███▏      | 2076/6470 [25:57<57:11,  1.28it/s]
2|2077|Loss: 0.8914102911949158:  32%|███▏      | 2076/6470 [25:57<57:11,  1.28it/s]
2|2077|Loss: 0.8914102911949158:  32%|███▏      | 2077/6470 [25:57<53:45,  1.36it/s]
2|2078|Loss: 0.7880429029464722:  32%|███▏      | 2077/6470 [25:58<53:45,  1.36it/s]
2|2078|Loss: 0.7880429029464722:  32%|███▏      | 2078/6470 [25:58<52:44,  1.39it/s]
2|2079|Loss: 0.7957598567008972:  32%|███▏      | 2078/6470 [25:58<52:44,  1.39it/s]
2|2079|Loss: 0.7957598567008972:  32%|███▏      | 2079/6470 [25:59<52:24,  1.40it/s]
2|2080|Loss: 0.7700478434562683:  32%|███▏      | 2079/6470 [25:59<52:24,  1.40it/s]
2|2080|Loss: 0.7700478434562683:  32%|███▏      | 2080/6470 [25:59<50:47,  1.44it/s]
2|2081|Loss: 0.8713559508323669:  32%|███▏      | 2080/6470 [26:00<50:47,  1.44it/s]
2|2081|Loss: 0.8713559508323669:  32%|███▏      | 2081/6470 [26:00<55:09,  1.33it/s]
2|2082|Loss: 0.9756879210472107:  32%|███▏      | 2081/6470 [26:01<55:09,  1.33it/s]
2|2082|Loss: 0.9756879210472107:  32%|███▏      | 2082/6470 [26:01<59:34,  1.23it/s]
2|2083|Loss: 0.9125702977180481:  32%|███▏      | 2082/6470 [26:02<59:34,  1.23it/s]
2|2083|Loss: 0.9125702977180481:  32%|███▏      | 2083/6470 [26:02<56:35,  1.29it/s]
2|2084|Loss: 0.9905452728271484:  32%|███▏      | 2083/6470 [26:02<56:35,  1.29it/s]
2|2084|Loss: 0.9905452728271484:  32%|███▏      | 2084/6470 [26:03<53:58,  1.35it/s]
2|2085|Loss: 0.9023265838623047:  32%|███▏      | 2084/6470 [26:03<53:58,  1.35it/s]
2|2085|Loss: 0.9023265838623047:  32%|███▏      | 2085/6470 [26:03<54:18,  1.35it/s]
2|2086|Loss: 0.8744919300079346:  32%|███▏      | 2085/6470 [26:04<54:18,  1.35it/s]
2|2086|Loss: 0.8744919300079346:  32%|███▏      | 2086/6470 [26:04<53:43,  1.36it/s]
2|2087|Loss: 0.8537803888320923:  32%|███▏      | 2086/6470 [26:04<53:43,  1.36it/s]
2|2087|Loss: 0.8537803888320923:  32%|███▏      | 2087/6470 [26:05<51:47,  1.41it/s]
2|2088|Loss: 0.8145411610603333:  32%|███▏      | 2087/6470 [26:05<51:47,  1.41it/s]
2|2088|Loss: 0.8145411610603333:  32%|███▏      | 2088/6470 [26:05<51:09,  1.43it/s]
2|2089|Loss: 0.7785364389419556:  32%|███▏      | 2088/6470 [26:06<51:09,  1.43it/s]
2|2089|Loss: 0.7785364389419556:  32%|███▏      | 2089/6470 [26:06<55:37,  1.31it/s]
2|2090|Loss: 0.7988072037696838:  32%|███▏      | 2089/6470 [26:07<55:37,  1.31it/s]
2|2090|Loss: 0.7988072037696838:  32%|███▏      | 2090/6470 [26:07<56:41,  1.29it/s]
2|2091|Loss: 0.889628529548645:  32%|███▏      | 2090/6470 [26:08<56:41,  1.29it/s] 
2|2091|Loss: 0.889628529548645:  32%|███▏      | 2091/6470 [26:08<59:12,  1.23it/s]
2|2092|Loss: 0.844808042049408:  32%|███▏      | 2091/6470 [26:08<59:12,  1.23it/s]
2|2092|Loss: 0.844808042049408:  32%|███▏      | 2092/6470 [26:09<59:30,  1.23it/s]
2|2093|Loss: 0.7596235871315002:  32%|███▏      | 2092/6470 [26:09<59:30,  1.23it/s]
2|2093|Loss: 0.7596235871315002:  32%|███▏      | 2093/6470 [26:09<52:17,  1.40it/s]
2|2094|Loss: 0.9139559864997864:  32%|███▏      | 2093/6470 [26:10<52:17,  1.40it/s]
2|2094|Loss: 0.9139559864997864:  32%|███▏      | 2094/6470 [26:10<52:52,  1.38it/s]
2|2095|Loss: 0.8050705790519714:  32%|███▏      | 2094/6470 [26:11<52:52,  1.38it/s]
2|2095|Loss: 0.8050705790519714:  32%|███▏      | 2095/6470 [26:11<57:19,  1.27it/s]
2|2096|Loss: 0.923470675945282:  32%|███▏      | 2095/6470 [26:11<57:19,  1.27it/s] 
2|2096|Loss: 0.923470675945282:  32%|███▏      | 2096/6470 [26:12<56:42,  1.29it/s]
2|2097|Loss: 0.845947802066803:  32%|███▏      | 2096/6470 [26:12<56:42,  1.29it/s]
2|2097|Loss: 0.845947802066803:  32%|███▏      | 2097/6470 [26:13<58:06,  1.25it/s]
2|2098|Loss: 0.7956305146217346:  32%|███▏      | 2097/6470 [26:13<58:06,  1.25it/s]
2|2098|Loss: 0.7956305146217346:  32%|███▏      | 2098/6470 [26:13<57:58,  1.26it/s]
2|2099|Loss: 0.9310197234153748:  32%|███▏      | 2098/6470 [26:14<57:58,  1.26it/s]
2|2099|Loss: 0.9310197234153748:  32%|███▏      | 2099/6470 [26:14<55:35,  1.31it/s]
2|2100|Loss: 0.8166632652282715:  32%|███▏      | 2099/6470 [26:14<55:35,  1.31it/s]
2|2100|Loss: 0.8166632652282715:  32%|███▏      | 2100/6470 [26:15<55:15,  1.32it/s]
2|2101|Loss: 0.8829692602157593:  32%|███▏      | 2100/6470 [26:15<55:15,  1.32it/s]
2|2101|Loss: 0.8829692602157593:  32%|███▏      | 2101/6470 [26:16<57:10,  1.27it/s]
2|2102|Loss: 0.8243839740753174:  32%|███▏      | 2101/6470 [26:16<57:10,  1.27it/s]
2|2102|Loss: 0.8243839740753174:  32%|███▏      | 2102/6470 [26:16<55:16,  1.32it/s]
2|2103|Loss: 0.8143165111541748:  32%|███▏      | 2102/6470 [26:17<55:16,  1.32it/s]
2|2103|Loss: 0.8143165111541748:  33%|███▎      | 2103/6470 [26:17<54:34,  1.33it/s]
2|2104|Loss: 0.7938129305839539:  33%|███▎      | 2103/6470 [26:17<54:34,  1.33it/s]
2|2104|Loss: 0.7938129305839539:  33%|███▎      | 2104/6470 [26:18<53:42,  1.36it/s]
2|2105|Loss: 0.7705861926078796:  33%|███▎      | 2104/6470 [26:18<53:42,  1.36it/s]
2|2105|Loss: 0.7705861926078796:  33%|███▎      | 2105/6470 [26:19<57:14,  1.27it/s]
2|2106|Loss: 0.8896613121032715:  33%|███▎      | 2105/6470 [26:19<57:14,  1.27it/s]
2|2106|Loss: 0.8896613121032715:  33%|███▎      | 2106/6470 [26:19<56:25,  1.29it/s]
2|2107|Loss: 0.8485810160636902:  33%|███▎      | 2106/6470 [26:20<56:25,  1.29it/s]
2|2107|Loss: 0.8485810160636902:  33%|███▎      | 2107/6470 [26:20<51:58,  1.40it/s]
2|2108|Loss: 0.8354033827781677:  33%|███▎      | 2107/6470 [26:20<51:58,  1.40it/s]
2|2108|Loss: 0.8354033827781677:  33%|███▎      | 2108/6470 [26:21<52:58,  1.37it/s]
2|2109|Loss: 0.9374338388442993:  33%|███▎      | 2108/6470 [26:21<52:58,  1.37it/s]
2|2109|Loss: 0.9374338388442993:  33%|███▎      | 2109/6470 [26:21<52:41,  1.38it/s]
2|2110|Loss: 0.8561316132545471:  33%|███▎      | 2109/6470 [26:22<52:41,  1.38it/s]
2|2110|Loss: 0.8561316132545471:  33%|███▎      | 2110/6470 [26:22<50:10,  1.45it/s]
2|2111|Loss: 0.9761509299278259:  33%|███▎      | 2110/6470 [26:22<50:10,  1.45it/s]
2|2111|Loss: 0.9761509299278259:  33%|███▎      | 2111/6470 [26:23<48:37,  1.49it/s]
2|2112|Loss: 0.7635695338249207:  33%|███▎      | 2111/6470 [26:23<48:37,  1.49it/s]
2|2112|Loss: 0.7635695338249207:  33%|███▎      | 2112/6470 [26:23<48:31,  1.50it/s]
2|2113|Loss: 0.6977131366729736:  33%|███▎      | 2112/6470 [26:24<48:31,  1.50it/s]
2|2113|Loss: 0.6977131366729736:  33%|███▎      | 2113/6470 [26:24<53:43,  1.35it/s]
2|2114|Loss: 0.8191112875938416:  33%|███▎      | 2113/6470 [26:25<53:43,  1.35it/s]
2|2114|Loss: 0.8191112875938416:  33%|███▎      | 2114/6470 [26:25<58:11,  1.25it/s]
2|2115|Loss: 0.8543049693107605:  33%|███▎      | 2114/6470 [26:26<58:11,  1.25it/s]
2|2115|Loss: 0.8543049693107605:  33%|███▎      | 2115/6470 [26:26<1:00:04,  1.21it/s]
2|2116|Loss: 1.0053156614303589:  33%|███▎      | 2115/6470 [26:27<1:00:04,  1.21it/s]
2|2116|Loss: 1.0053156614303589:  33%|███▎      | 2116/6470 [26:27<59:53,  1.21it/s]  
2|2117|Loss: 0.8932051062583923:  33%|███▎      | 2116/6470 [26:27<59:53,  1.21it/s]
2|2117|Loss: 0.8932051062583923:  33%|███▎      | 2117/6470 [26:28<55:51,  1.30it/s]
2|2118|Loss: 0.88656085729599:  33%|███▎      | 2117/6470 [26:28<55:51,  1.30it/s]  
2|2118|Loss: 0.88656085729599:  33%|███▎      | 2118/6470 [26:28<50:55,  1.42it/s]
2|2119|Loss: 0.9057372808456421:  33%|███▎      | 2118/6470 [26:28<50:55,  1.42it/s]
2|2119|Loss: 0.9057372808456421:  33%|███▎      | 2119/6470 [26:29<53:01,  1.37it/s]
2|2120|Loss: 0.830534815788269:  33%|███▎      | 2119/6470 [26:29<53:01,  1.37it/s] 
2|2120|Loss: 0.830534815788269:  33%|███▎      | 2120/6470 [26:30<50:55,  1.42it/s]
2|2121|Loss: 0.8206350803375244:  33%|███▎      | 2120/6470 [26:30<50:55,  1.42it/s]
2|2121|Loss: 0.8206350803375244:  33%|███▎      | 2121/6470 [26:30<55:01,  1.32it/s]
2|2122|Loss: 0.7975530028343201:  33%|███▎      | 2121/6470 [26:31<55:01,  1.32it/s]
2|2122|Loss: 0.7975530028343201:  33%|███▎      | 2122/6470 [26:31<57:17,  1.26it/s]
2|2123|Loss: 0.7083820700645447:  33%|███▎      | 2122/6470 [26:32<57:17,  1.26it/s]
2|2123|Loss: 0.7083820700645447:  33%|███▎      | 2123/6470 [26:32<56:53,  1.27it/s]
2|2124|Loss: 0.8073912262916565:  33%|███▎      | 2123/6470 [26:33<56:53,  1.27it/s]
2|2124|Loss: 0.8073912262916565:  33%|███▎      | 2124/6470 [26:33<59:31,  1.22it/s]
2|2125|Loss: 0.8679364919662476:  33%|███▎      | 2124/6470 [26:33<59:31,  1.22it/s]
2|2125|Loss: 0.8679364919662476:  33%|███▎      | 2125/6470 [26:34<57:34,  1.26it/s]
2|2126|Loss: 0.8226709365844727:  33%|███▎      | 2125/6470 [26:34<57:34,  1.26it/s]
2|2126|Loss: 0.8226709365844727:  33%|███▎      | 2126/6470 [26:34<55:34,  1.30it/s]
2|2127|Loss: 0.8146572709083557:  33%|███▎      | 2126/6470 [26:35<55:34,  1.30it/s]
2|2127|Loss: 0.8146572709083557:  33%|███▎      | 2127/6470 [26:35<58:11,  1.24it/s]
2|2128|Loss: 0.9426617622375488:  33%|███▎      | 2127/6470 [26:36<58:11,  1.24it/s]
2|2128|Loss: 0.9426617622375488:  33%|███▎      | 2128/6470 [26:36<58:16,  1.24it/s]
2|2129|Loss: 0.9628071188926697:  33%|███▎      | 2128/6470 [26:37<58:16,  1.24it/s]
2|2129|Loss: 0.9628071188926697:  33%|███▎      | 2129/6470 [26:37<59:10,  1.22it/s]
2|2130|Loss: 0.9743400812149048:  33%|███▎      | 2129/6470 [26:37<59:10,  1.22it/s]
2|2130|Loss: 0.9743400812149048:  33%|███▎      | 2130/6470 [26:38<1:00:26,  1.20it/s]
2|2131|Loss: 0.7270515561103821:  33%|███▎      | 2130/6470 [26:38<1:00:26,  1.20it/s]
2|2131|Loss: 0.7270515561103821:  33%|███▎      | 2131/6470 [26:38<56:04,  1.29it/s]  
2|2132|Loss: 0.9283934235572815:  33%|███▎      | 2131/6470 [26:39<56:04,  1.29it/s]
2|2132|Loss: 0.9283934235572815:  33%|███▎      | 2132/6470 [26:39<55:16,  1.31it/s]
2|2133|Loss: 0.8021802306175232:  33%|███▎      | 2132/6470 [26:40<55:16,  1.31it/s]
2|2133|Loss: 0.8021802306175232:  33%|███▎      | 2133/6470 [26:40<53:06,  1.36it/s]
2|2134|Loss: 0.6698322892189026:  33%|███▎      | 2133/6470 [26:40<53:06,  1.36it/s]
2|2134|Loss: 0.6698322892189026:  33%|███▎      | 2134/6470 [26:41<53:51,  1.34it/s]
2|2135|Loss: 0.8139029145240784:  33%|███▎      | 2134/6470 [26:41<53:51,  1.34it/s]
2|2135|Loss: 0.8139029145240784:  33%|███▎      | 2135/6470 [26:41<55:02,  1.31it/s]
2|2136|Loss: 0.8392752408981323:  33%|███▎      | 2135/6470 [26:42<55:02,  1.31it/s]
2|2136|Loss: 0.8392752408981323:  33%|███▎      | 2136/6470 [26:42<55:56,  1.29it/s]
2|2137|Loss: 0.8545345664024353:  33%|███▎      | 2136/6470 [26:43<55:56,  1.29it/s]
2|2137|Loss: 0.8545345664024353:  33%|███▎      | 2137/6470 [26:43<56:05,  1.29it/s]
2|2138|Loss: 0.753369927406311:  33%|███▎      | 2137/6470 [26:43<56:05,  1.29it/s] 
2|2138|Loss: 0.753369927406311:  33%|███▎      | 2138/6470 [26:44<56:27,  1.28it/s]
2|2139|Loss: 0.876488208770752:  33%|███▎      | 2138/6470 [26:44<56:27,  1.28it/s]
2|2139|Loss: 0.876488208770752:  33%|███▎      | 2139/6470 [26:45<57:16,  1.26it/s]
2|2140|Loss: 0.874512255191803:  33%|███▎      | 2139/6470 [26:45<57:16,  1.26it/s]
2|2140|Loss: 0.874512255191803:  33%|███▎      | 2140/6470 [26:45<57:30,  1.25it/s]
2|2141|Loss: 0.7993244528770447:  33%|███▎      | 2140/6470 [26:46<57:30,  1.25it/s]
2|2141|Loss: 0.7993244528770447:  33%|███▎      | 2141/6470 [26:46<54:25,  1.33it/s]
2|2142|Loss: 0.720231294631958:  33%|███▎      | 2141/6470 [26:47<54:25,  1.33it/s] 
2|2142|Loss: 0.720231294631958:  33%|███▎      | 2142/6470 [26:47<56:24,  1.28it/s]
2|2143|Loss: 0.9079374670982361:  33%|███▎      | 2142/6470 [26:47<56:24,  1.28it/s]
2|2143|Loss: 0.9079374670982361:  33%|███▎      | 2143/6470 [26:48<57:09,  1.26it/s]
2|2144|Loss: 0.9406803250312805:  33%|███▎      | 2143/6470 [26:48<57:09,  1.26it/s]
2|2144|Loss: 0.9406803250312805:  33%|███▎      | 2144/6470 [26:48<55:47,  1.29it/s]
2|2145|Loss: 0.9102081656455994:  33%|███▎      | 2144/6470 [26:49<55:47,  1.29it/s]
2|2145|Loss: 0.9102081656455994:  33%|███▎      | 2145/6470 [26:49<52:15,  1.38it/s]
2|2146|Loss: 0.8945091366767883:  33%|███▎      | 2145/6470 [26:49<52:15,  1.38it/s]
2|2146|Loss: 0.8945091366767883:  33%|███▎      | 2146/6470 [26:50<52:32,  1.37it/s]
2|2147|Loss: 0.9385995268821716:  33%|███▎      | 2146/6470 [26:50<52:32,  1.37it/s]
2|2147|Loss: 0.9385995268821716:  33%|███▎      | 2147/6470 [26:50<48:32,  1.48it/s]
2|2148|Loss: 0.9162362217903137:  33%|███▎      | 2147/6470 [26:51<48:32,  1.48it/s]
2|2148|Loss: 0.9162362217903137:  33%|███▎      | 2148/6470 [26:51<49:53,  1.44it/s]
2|2149|Loss: 0.9298343658447266:  33%|███▎      | 2148/6470 [26:51<49:53,  1.44it/s]
2|2149|Loss: 0.9298343658447266:  33%|███▎      | 2149/6470 [26:52<45:00,  1.60it/s]
2|2150|Loss: 0.7666442394256592:  33%|███▎      | 2149/6470 [26:52<45:00,  1.60it/s]
2|2150|Loss: 0.7666442394256592:  33%|███▎      | 2150/6470 [26:52<41:21,  1.74it/s]
2|2151|Loss: 0.9118243455886841:  33%|███▎      | 2150/6470 [26:52<41:21,  1.74it/s]
2|2151|Loss: 0.9118243455886841:  33%|███▎      | 2151/6470 [26:53<43:54,  1.64it/s]
2|2152|Loss: 0.6789563894271851:  33%|███▎      | 2151/6470 [26:53<43:54,  1.64it/s]
2|2152|Loss: 0.6789563894271851:  33%|███▎      | 2152/6470 [26:53<43:26,  1.66it/s]
2|2153|Loss: 0.9579125046730042:  33%|███▎      | 2152/6470 [26:54<43:26,  1.66it/s]
2|2153|Loss: 0.9579125046730042:  33%|███▎      | 2153/6470 [26:54<41:35,  1.73it/s]
2|2154|Loss: 0.8953157663345337:  33%|███▎      | 2153/6470 [26:54<41:35,  1.73it/s]
2|2154|Loss: 0.8953157663345337:  33%|███▎      | 2154/6470 [26:55<45:02,  1.60it/s]
2|2155|Loss: 0.8208551406860352:  33%|███▎      | 2154/6470 [26:55<45:02,  1.60it/s]
2|2155|Loss: 0.8208551406860352:  33%|███▎      | 2155/6470 [26:55<50:33,  1.42it/s]
2|2156|Loss: 0.8652326464653015:  33%|███▎      | 2155/6470 [26:56<50:33,  1.42it/s]
2|2156|Loss: 0.8652326464653015:  33%|███▎      | 2156/6470 [26:56<47:43,  1.51it/s]
2|2157|Loss: 0.8149109482765198:  33%|███▎      | 2156/6470 [26:56<47:43,  1.51it/s]
2|2157|Loss: 0.8149109482765198:  33%|███▎      | 2157/6470 [26:57<47:07,  1.53it/s]
2|2158|Loss: 0.8773301243782043:  33%|███▎      | 2157/6470 [26:57<47:07,  1.53it/s]
2|2158|Loss: 0.8773301243782043:  33%|███▎      | 2158/6470 [26:57<46:23,  1.55it/s]
2|2159|Loss: 0.9404965043067932:  33%|███▎      | 2158/6470 [26:58<46:23,  1.55it/s]
2|2159|Loss: 0.9404965043067932:  33%|███▎      | 2159/6470 [26:58<50:43,  1.42it/s]
2|2160|Loss: 0.8948597311973572:  33%|███▎      | 2159/6470 [26:59<50:43,  1.42it/s]
2|2160|Loss: 0.8948597311973572:  33%|███▎      | 2160/6470 [26:59<1:02:53,  1.14it/s]
2|2161|Loss: 0.7115008234977722:  33%|███▎      | 2160/6470 [27:00<1:02:53,  1.14it/s]
2|2161|Loss: 0.7115008234977722:  33%|███▎      | 2161/6470 [27:00<1:01:55,  1.16it/s]
2|2162|Loss: 0.7693626880645752:  33%|███▎      | 2161/6470 [27:01<1:01:55,  1.16it/s]
2|2162|Loss: 0.7693626880645752:  33%|███▎      | 2162/6470 [27:01<1:00:20,  1.19it/s]
2|2163|Loss: 0.7593193054199219:  33%|███▎      | 2162/6470 [27:01<1:00:20,  1.19it/s]
2|2163|Loss: 0.7593193054199219:  33%|███▎      | 2163/6470 [27:02<56:52,  1.26it/s]  
2|2164|Loss: 0.8937179446220398:  33%|███▎      | 2163/6470 [27:02<56:52,  1.26it/s]
2|2164|Loss: 0.8937179446220398:  33%|███▎      | 2164/6470 [27:02<53:19,  1.35it/s]
2|2165|Loss: 0.9256921410560608:  33%|███▎      | 2164/6470 [27:03<53:19,  1.35it/s]
2|2165|Loss: 0.9256921410560608:  33%|███▎      | 2165/6470 [27:03<51:44,  1.39it/s]
2|2166|Loss: 0.9715046286582947:  33%|███▎      | 2165/6470 [27:03<51:44,  1.39it/s]
2|2166|Loss: 0.9715046286582947:  33%|███▎      | 2166/6470 [27:04<51:14,  1.40it/s]
2|2167|Loss: 0.7969208359718323:  33%|███▎      | 2166/6470 [27:04<51:14,  1.40it/s]
2|2167|Loss: 0.7969208359718323:  33%|███▎      | 2167/6470 [27:05<55:40,  1.29it/s]
2|2168|Loss: 0.7559337019920349:  33%|███▎      | 2167/6470 [27:05<55:40,  1.29it/s]
2|2168|Loss: 0.7559337019920349:  34%|███▎      | 2168/6470 [27:05<51:27,  1.39it/s]
2|2169|Loss: 0.7752413153648376:  34%|███▎      | 2168/6470 [27:06<51:27,  1.39it/s]
2|2169|Loss: 0.7752413153648376:  34%|███▎      | 2169/6470 [27:06<51:39,  1.39it/s]
2|2170|Loss: 0.7953348159790039:  34%|███▎      | 2169/6470 [27:06<51:39,  1.39it/s]
2|2170|Loss: 0.7953348159790039:  34%|███▎      | 2170/6470 [27:06<48:00,  1.49it/s]
2|2171|Loss: 0.9954869747161865:  34%|███▎      | 2170/6470 [27:07<48:00,  1.49it/s]
2|2171|Loss: 0.9954869747161865:  34%|███▎      | 2171/6470 [27:07<45:51,  1.56it/s]
2|2172|Loss: 1.1242039203643799:  34%|███▎      | 2171/6470 [27:07<45:51,  1.56it/s]
2|2172|Loss: 1.1242039203643799:  34%|███▎      | 2172/6470 [27:08<44:15,  1.62it/s]
2|2173|Loss: 0.8117027282714844:  34%|███▎      | 2172/6470 [27:08<44:15,  1.62it/s]
2|2173|Loss: 0.8117027282714844:  34%|███▎      | 2173/6470 [27:08<45:16,  1.58it/s]
2|2174|Loss: 0.6798700094223022:  34%|███▎      | 2173/6470 [27:09<45:16,  1.58it/s]
2|2174|Loss: 0.6798700094223022:  34%|███▎      | 2174/6470 [27:09<47:59,  1.49it/s]
2|2175|Loss: 0.7916605472564697:  34%|███▎      | 2174/6470 [27:09<47:59,  1.49it/s]
2|2175|Loss: 0.7916605472564697:  34%|███▎      | 2175/6470 [27:10<50:04,  1.43it/s]
2|2176|Loss: 0.924001932144165:  34%|███▎      | 2175/6470 [27:10<50:04,  1.43it/s] 
2|2176|Loss: 0.924001932144165:  34%|███▎      | 2176/6470 [27:11<52:10,  1.37it/s]
2|2177|Loss: 1.0539411306381226:  34%|███▎      | 2176/6470 [27:11<52:10,  1.37it/s]
2|2177|Loss: 1.0539411306381226:  34%|███▎      | 2177/6470 [27:11<53:48,  1.33it/s]
2|2178|Loss: 0.918333113193512:  34%|███▎      | 2177/6470 [27:12<53:48,  1.33it/s] 
2|2178|Loss: 0.918333113193512:  34%|███▎      | 2178/6470 [27:12<54:29,  1.31it/s]
2|2179|Loss: 0.8439768552780151:  34%|███▎      | 2178/6470 [27:13<54:29,  1.31it/s]
2|2179|Loss: 0.8439768552780151:  34%|███▎      | 2179/6470 [27:13<52:58,  1.35it/s]
2|2180|Loss: 0.7493850588798523:  34%|███▎      | 2179/6470 [27:13<52:58,  1.35it/s]
2|2180|Loss: 0.7493850588798523:  34%|███▎      | 2180/6470 [27:13<49:56,  1.43it/s]
2|2181|Loss: 0.8181470632553101:  34%|███▎      | 2180/6470 [27:14<49:56,  1.43it/s]
2|2181|Loss: 0.8181470632553101:  34%|███▎      | 2181/6470 [27:14<54:20,  1.32it/s]
2|2182|Loss: 0.8790149688720703:  34%|███▎      | 2181/6470 [27:15<54:20,  1.32it/s]
2|2182|Loss: 0.8790149688720703:  34%|███▎      | 2182/6470 [27:15<58:34,  1.22it/s]
2|2183|Loss: 0.7969300746917725:  34%|███▎      | 2182/6470 [27:16<58:34,  1.22it/s]
2|2183|Loss: 0.7969300746917725:  34%|███▎      | 2183/6470 [27:16<57:06,  1.25it/s]
2|2184|Loss: 0.9231225252151489:  34%|███▎      | 2183/6470 [27:17<57:06,  1.25it/s]
2|2184|Loss: 0.9231225252151489:  34%|███▍      | 2184/6470 [27:17<56:44,  1.26it/s]
2|2185|Loss: 0.9187433123588562:  34%|███▍      | 2184/6470 [27:17<56:44,  1.26it/s]
2|2185|Loss: 0.9187433123588562:  34%|███▍      | 2185/6470 [27:18<53:18,  1.34it/s]
2|2186|Loss: 0.8305965662002563:  34%|███▍      | 2185/6470 [27:18<53:18,  1.34it/s]
2|2186|Loss: 0.8305965662002563:  34%|███▍      | 2186/6470 [27:18<51:55,  1.38it/s]
2|2187|Loss: 0.81207674741745:  34%|███▍      | 2186/6470 [27:19<51:55,  1.38it/s]  
2|2187|Loss: 0.81207674741745:  34%|███▍      | 2187/6470 [27:19<53:27,  1.34it/s]
2|2188|Loss: 0.8429556488990784:  34%|███▍      | 2187/6470 [27:19<53:27,  1.34it/s]
2|2188|Loss: 0.8429556488990784:  34%|███▍      | 2188/6470 [27:20<52:45,  1.35it/s]
2|2189|Loss: 0.8394421339035034:  34%|███▍      | 2188/6470 [27:20<52:45,  1.35it/s]
2|2189|Loss: 0.8394421339035034:  34%|███▍      | 2189/6470 [27:20<52:55,  1.35it/s]
2|2190|Loss: 0.7799534797668457:  34%|███▍      | 2189/6470 [27:21<52:55,  1.35it/s]
2|2190|Loss: 0.7799534797668457:  34%|███▍      | 2190/6470 [27:21<51:41,  1.38it/s]
2|2191|Loss: 0.7566130757331848:  34%|███▍      | 2190/6470 [27:22<51:41,  1.38it/s]
2|2191|Loss: 0.7566130757331848:  34%|███▍      | 2191/6470 [27:22<52:54,  1.35it/s]
2|2192|Loss: 0.7677210569381714:  34%|███▍      | 2191/6470 [27:22<52:54,  1.35it/s]
2|2192|Loss: 0.7677210569381714:  34%|███▍      | 2192/6470 [27:23<53:55,  1.32it/s]
2|2193|Loss: 0.8430113196372986:  34%|███▍      | 2192/6470 [27:23<53:55,  1.32it/s]
2|2193|Loss: 0.8430113196372986:  34%|███▍      | 2193/6470 [27:23<54:17,  1.31it/s]
2|2194|Loss: 0.9165980815887451:  34%|███▍      | 2193/6470 [27:24<54:17,  1.31it/s]
2|2194|Loss: 0.9165980815887451:  34%|███▍      | 2194/6470 [27:24<54:44,  1.30it/s]
2|2195|Loss: 0.835081934928894:  34%|███▍      | 2194/6470 [27:25<54:44,  1.30it/s] 
2|2195|Loss: 0.835081934928894:  34%|███▍      | 2195/6470 [27:25<52:58,  1.34it/s]
2|2196|Loss: 0.8179306983947754:  34%|███▍      | 2195/6470 [27:25<52:58,  1.34it/s]
2|2196|Loss: 0.8179306983947754:  34%|███▍      | 2196/6470 [27:26<53:10,  1.34it/s]
2|2197|Loss: 0.8358654975891113:  34%|███▍      | 2196/6470 [27:26<53:10,  1.34it/s]
2|2197|Loss: 0.8358654975891113:  34%|███▍      | 2197/6470 [27:26<50:12,  1.42it/s]
2|2198|Loss: 0.8429374098777771:  34%|███▍      | 2197/6470 [27:27<50:12,  1.42it/s]
2|2198|Loss: 0.8429374098777771:  34%|███▍      | 2198/6470 [27:27<50:32,  1.41it/s]
2|2199|Loss: 0.8544673323631287:  34%|███▍      | 2198/6470 [27:27<50:32,  1.41it/s]
2|2199|Loss: 0.8544673323631287:  34%|███▍      | 2199/6470 [27:28<50:30,  1.41it/s]
2|2200|Loss: 0.8222365975379944:  34%|███▍      | 2199/6470 [27:28<50:30,  1.41it/s]
2|2200|Loss: 0.8222365975379944:  34%|███▍      | 2200/6470 [27:29<54:13,  1.31it/s]
2|2201|Loss: 0.9561170339584351:  34%|███▍      | 2200/6470 [27:29<54:13,  1.31it/s]
2|2201|Loss: 0.9561170339584351:  34%|███▍      | 2201/6470 [27:30<57:49,  1.23it/s]
2|2202|Loss: 0.7000094652175903:  34%|███▍      | 2201/6470 [27:30<57:49,  1.23it/s]
2|2202|Loss: 0.7000094652175903:  34%|███▍      | 2202/6470 [27:30<55:18,  1.29it/s]
2|2203|Loss: 0.8175696730613708:  34%|███▍      | 2202/6470 [27:31<55:18,  1.29it/s]
2|2203|Loss: 0.8175696730613708:  34%|███▍      | 2203/6470 [27:32<1:07:14,  1.06it/s]
2|2204|Loss: 0.7493485808372498:  34%|███▍      | 2203/6470 [27:32<1:07:14,  1.06it/s]
2|2204|Loss: 0.7493485808372498:  34%|███▍      | 2204/6470 [27:32<1:06:00,  1.08it/s]
2|2205|Loss: 1.0854781866073608:  34%|███▍      | 2204/6470 [27:33<1:06:00,  1.08it/s]
2|2205|Loss: 1.0854781866073608:  34%|███▍      | 2205/6470 [27:33<1:01:49,  1.15it/s]
2|2206|Loss: 0.8128276467323303:  34%|███▍      | 2205/6470 [27:34<1:01:49,  1.15it/s]
2|2206|Loss: 0.8128276467323303:  34%|███▍      | 2206/6470 [27:34<53:09,  1.34it/s]  
2|2207|Loss: 0.8151208162307739:  34%|███▍      | 2206/6470 [27:34<53:09,  1.34it/s]
2|2207|Loss: 0.8151208162307739:  34%|███▍      | 2207/6470 [27:35<56:13,  1.26it/s]
2|2208|Loss: 0.7165541648864746:  34%|███▍      | 2207/6470 [27:35<56:13,  1.26it/s]
2|2208|Loss: 0.7165541648864746:  34%|███▍      | 2208/6470 [27:35<58:27,  1.22it/s]
2|2209|Loss: 0.8308934569358826:  34%|███▍      | 2208/6470 [27:36<58:27,  1.22it/s]
2|2209|Loss: 0.8308934569358826:  34%|███▍      | 2209/6470 [27:36<58:03,  1.22it/s]
2|2210|Loss: 0.8185903429985046:  34%|███▍      | 2209/6470 [27:37<58:03,  1.22it/s]
2|2210|Loss: 0.8185903429985046:  34%|███▍      | 2210/6470 [27:37<55:14,  1.29it/s]
2|2211|Loss: 0.7806458473205566:  34%|███▍      | 2210/6470 [27:37<55:14,  1.29it/s]
2|2211|Loss: 0.7806458473205566:  34%|███▍      | 2211/6470 [27:38<57:22,  1.24it/s]
2|2212|Loss: 0.9301853775978088:  34%|███▍      | 2211/6470 [27:38<57:22,  1.24it/s]
2|2212|Loss: 0.9301853775978088:  34%|███▍      | 2212/6470 [27:39<58:59,  1.20it/s]
2|2213|Loss: 1.0900635719299316:  34%|███▍      | 2212/6470 [27:39<58:59,  1.20it/s]
2|2213|Loss: 1.0900635719299316:  34%|███▍      | 2213/6470 [27:39<55:46,  1.27it/s]
2|2214|Loss: 0.7888429164886475:  34%|███▍      | 2213/6470 [27:40<55:46,  1.27it/s]
2|2214|Loss: 0.7888429164886475:  34%|███▍      | 2214/6470 [27:40<52:57,  1.34it/s]
2|2215|Loss: 0.8099446296691895:  34%|███▍      | 2214/6470 [27:40<52:57,  1.34it/s]
2|2215|Loss: 0.8099446296691895:  34%|███▍      | 2215/6470 [27:41<53:06,  1.34it/s]
2|2216|Loss: 0.7399410009384155:  34%|███▍      | 2215/6470 [27:41<53:06,  1.34it/s]
2|2216|Loss: 0.7399410009384155:  34%|███▍      | 2216/6470 [27:42<52:34,  1.35it/s]
2|2217|Loss: 0.8667347431182861:  34%|███▍      | 2216/6470 [27:42<52:34,  1.35it/s]
2|2217|Loss: 0.8667347431182861:  34%|███▍      | 2217/6470 [27:42<52:53,  1.34it/s]
2|2218|Loss: 0.866689145565033:  34%|███▍      | 2217/6470 [27:43<52:53,  1.34it/s] 
2|2218|Loss: 0.866689145565033:  34%|███▍      | 2218/6470 [27:43<52:27,  1.35it/s]
2|2219|Loss: 0.8659612536430359:  34%|███▍      | 2218/6470 [27:43<52:27,  1.35it/s]
2|2219|Loss: 0.8659612536430359:  34%|███▍      | 2219/6470 [27:44<48:07,  1.47it/s]
2|2220|Loss: 0.7616925835609436:  34%|███▍      | 2219/6470 [27:44<48:07,  1.47it/s]
2|2220|Loss: 0.7616925835609436:  34%|███▍      | 2220/6470 [27:44<47:17,  1.50it/s]
2|2221|Loss: 0.925785481929779:  34%|███▍      | 2220/6470 [27:45<47:17,  1.50it/s] 
2|2221|Loss: 0.925785481929779:  34%|███▍      | 2221/6470 [27:45<48:08,  1.47it/s]
2|2222|Loss: 0.8289796710014343:  34%|███▍      | 2221/6470 [27:45<48:08,  1.47it/s]
2|2222|Loss: 0.8289796710014343:  34%|███▍      | 2222/6470 [27:46<50:13,  1.41it/s]
2|2223|Loss: 0.8869516253471375:  34%|███▍      | 2222/6470 [27:46<50:13,  1.41it/s]
2|2223|Loss: 0.8869516253471375:  34%|███▍      | 2223/6470 [27:46<51:56,  1.36it/s]
2|2224|Loss: 0.808681845664978:  34%|███▍      | 2223/6470 [27:47<51:56,  1.36it/s] 
2|2224|Loss: 0.808681845664978:  34%|███▍      | 2224/6470 [27:47<55:29,  1.28it/s]
2|2225|Loss: 0.794684886932373:  34%|███▍      | 2224/6470 [27:48<55:29,  1.28it/s]
2|2225|Loss: 0.794684886932373:  34%|███▍      | 2225/6470 [27:48<52:35,  1.35it/s]
2|2226|Loss: 0.8429556488990784:  34%|███▍      | 2225/6470 [27:48<52:35,  1.35it/s]
2|2226|Loss: 0.8429556488990784:  34%|███▍      | 2226/6470 [27:49<53:56,  1.31it/s]
2|2227|Loss: 0.9814618229866028:  34%|███▍      | 2226/6470 [27:49<53:56,  1.31it/s]
2|2227|Loss: 0.9814618229866028:  34%|███▍      | 2227/6470 [27:50<57:06,  1.24it/s]
2|2228|Loss: 0.8772203326225281:  34%|███▍      | 2227/6470 [27:50<57:06,  1.24it/s]
2|2228|Loss: 0.8772203326225281:  34%|███▍      | 2228/6470 [27:51<59:20,  1.19it/s]
2|2229|Loss: 0.8608589172363281:  34%|███▍      | 2228/6470 [27:51<59:20,  1.19it/s]
2|2229|Loss: 0.8608589172363281:  34%|███▍      | 2229/6470 [27:51<58:15,  1.21it/s]
2|2230|Loss: 0.8943161368370056:  34%|███▍      | 2229/6470 [27:52<58:15,  1.21it/s]
2|2230|Loss: 0.8943161368370056:  34%|███▍      | 2230/6470 [27:52<54:56,  1.29it/s]
2|2231|Loss: 0.790764570236206:  34%|███▍      | 2230/6470 [27:52<54:56,  1.29it/s] 
2|2231|Loss: 0.790764570236206:  34%|███▍      | 2231/6470 [27:53<50:12,  1.41it/s]
2|2232|Loss: 0.8138203024864197:  34%|███▍      | 2231/6470 [27:53<50:12,  1.41it/s]
2|2232|Loss: 0.8138203024864197:  34%|███▍      | 2232/6470 [27:53<47:09,  1.50it/s]
2|2233|Loss: 0.8796415328979492:  34%|███▍      | 2232/6470 [27:54<47:09,  1.50it/s]
2|2233|Loss: 0.8796415328979492:  35%|███▍      | 2233/6470 [27:54<51:15,  1.38it/s]
2|2234|Loss: 0.7955775260925293:  35%|███▍      | 2233/6470 [27:55<51:15,  1.38it/s]
2|2234|Loss: 0.7955775260925293:  35%|███▍      | 2234/6470 [27:55<51:37,  1.37it/s]
2|2235|Loss: 0.8598285913467407:  35%|███▍      | 2234/6470 [27:55<51:37,  1.37it/s]
2|2235|Loss: 0.8598285913467407:  35%|███▍      | 2235/6470 [27:56<52:51,  1.34it/s]
2|2236|Loss: 1.0005288124084473:  35%|███▍      | 2235/6470 [27:56<52:51,  1.34it/s]
2|2236|Loss: 1.0005288124084473:  35%|███▍      | 2236/6470 [27:56<52:29,  1.34it/s]
2|2237|Loss: 0.9370487332344055:  35%|███▍      | 2236/6470 [27:57<52:29,  1.34it/s]
2|2237|Loss: 0.9370487332344055:  35%|███▍      | 2237/6470 [27:57<53:29,  1.32it/s]
2|2238|Loss: 0.7739484310150146:  35%|███▍      | 2237/6470 [27:58<53:29,  1.32it/s]
2|2238|Loss: 0.7739484310150146:  35%|███▍      | 2238/6470 [27:58<51:32,  1.37it/s]
2|2239|Loss: 0.8418068289756775:  35%|███▍      | 2238/6470 [27:58<51:32,  1.37it/s]
2|2239|Loss: 0.8418068289756775:  35%|███▍      | 2239/6470 [27:59<54:20,  1.30it/s]
2|2240|Loss: 0.8200176358222961:  35%|███▍      | 2239/6470 [27:59<54:20,  1.30it/s]
2|2240|Loss: 0.8200176358222961:  35%|███▍      | 2240/6470 [28:00<55:41,  1.27it/s]
2|2241|Loss: 0.7261862754821777:  35%|███▍      | 2240/6470 [28:00<55:41,  1.27it/s]
2|2241|Loss: 0.7261862754821777:  35%|███▍      | 2241/6470 [28:00<59:09,  1.19it/s]
2|2242|Loss: 0.6117238402366638:  35%|███▍      | 2241/6470 [28:01<59:09,  1.19it/s]
2|2242|Loss: 0.6117238402366638:  35%|███▍      | 2242/6470 [28:01<57:00,  1.24it/s]
2|2243|Loss: 0.9462727904319763:  35%|███▍      | 2242/6470 [28:02<57:00,  1.24it/s]
2|2243|Loss: 0.9462727904319763:  35%|███▍      | 2243/6470 [28:02<58:10,  1.21it/s]
2|2244|Loss: 0.8870522379875183:  35%|███▍      | 2243/6470 [28:03<58:10,  1.21it/s]
2|2244|Loss: 0.8870522379875183:  35%|███▍      | 2244/6470 [28:03<56:14,  1.25it/s]
2|2245|Loss: 0.8904820084571838:  35%|███▍      | 2244/6470 [28:03<56:14,  1.25it/s]
2|2245|Loss: 0.8904820084571838:  35%|███▍      | 2245/6470 [28:04<55:19,  1.27it/s]
2|2246|Loss: 0.8834139704704285:  35%|███▍      | 2245/6470 [28:04<55:19,  1.27it/s]
2|2246|Loss: 0.8834139704704285:  35%|███▍      | 2246/6470 [28:04<52:41,  1.34it/s]
2|2247|Loss: 0.7267123460769653:  35%|███▍      | 2246/6470 [28:05<52:41,  1.34it/s]
2|2247|Loss: 0.7267123460769653:  35%|███▍      | 2247/6470 [28:05<57:18,  1.23it/s]
2|2248|Loss: 0.9091503024101257:  35%|███▍      | 2247/6470 [28:06<57:18,  1.23it/s]
2|2248|Loss: 0.9091503024101257:  35%|███▍      | 2248/6470 [28:06<58:07,  1.21it/s]
2|2249|Loss: 0.6555032730102539:  35%|███▍      | 2248/6470 [28:07<58:07,  1.21it/s]
2|2249|Loss: 0.6555032730102539:  35%|███▍      | 2249/6470 [28:07<58:35,  1.20it/s]
2|2250|Loss: 0.7076241970062256:  35%|███▍      | 2249/6470 [28:08<58:35,  1.20it/s]
2|2250|Loss: 0.7076241970062256:  35%|███▍      | 2250/6470 [28:08<1:05:16,  1.08it/s]
2|2251|Loss: 0.8065997362136841:  35%|███▍      | 2250/6470 [28:09<1:05:16,  1.08it/s]
2|2251|Loss: 0.8065997362136841:  35%|███▍      | 2251/6470 [28:09<1:05:03,  1.08it/s]
2|2252|Loss: 0.8925280570983887:  35%|███▍      | 2251/6470 [28:09<1:05:03,  1.08it/s]
2|2252|Loss: 0.8925280570983887:  35%|███▍      | 2252/6470 [28:10<1:00:56,  1.15it/s]
2|2253|Loss: 0.891612708568573:  35%|███▍      | 2252/6470 [28:10<1:00:56,  1.15it/s] 
2|2253|Loss: 0.891612708568573:  35%|███▍      | 2253/6470 [28:11<1:00:01,  1.17it/s]
2|2254|Loss: 0.8066226840019226:  35%|███▍      | 2253/6470 [28:11<1:00:01,  1.17it/s]
2|2254|Loss: 0.8066226840019226:  35%|███▍      | 2254/6470 [28:11<59:07,  1.19it/s]  
2|2255|Loss: 0.7076717019081116:  35%|███▍      | 2254/6470 [28:12<59:07,  1.19it/s]
2|2255|Loss: 0.7076717019081116:  35%|███▍      | 2255/6470 [28:12<52:44,  1.33it/s]
2|2256|Loss: 0.9827839136123657:  35%|███▍      | 2255/6470 [28:12<52:44,  1.33it/s]
2|2256|Loss: 0.9827839136123657:  35%|███▍      | 2256/6470 [28:13<50:07,  1.40it/s]
2|2257|Loss: 0.8464173078536987:  35%|███▍      | 2256/6470 [28:13<50:07,  1.40it/s]
2|2257|Loss: 0.8464173078536987:  35%|███▍      | 2257/6470 [28:13<50:14,  1.40it/s]
2|2258|Loss: 0.8799436688423157:  35%|███▍      | 2257/6470 [28:14<50:14,  1.40it/s]
2|2258|Loss: 0.8799436688423157:  35%|███▍      | 2258/6470 [28:14<46:00,  1.53it/s]
2|2259|Loss: 0.952648937702179:  35%|███▍      | 2258/6470 [28:14<46:00,  1.53it/s] 
2|2259|Loss: 0.952648937702179:  35%|███▍      | 2259/6470 [28:14<43:35,  1.61it/s]
2|2260|Loss: 0.8795934319496155:  35%|███▍      | 2259/6470 [28:15<43:35,  1.61it/s]
2|2260|Loss: 0.8795934319496155:  35%|███▍      | 2260/6470 [28:15<43:17,  1.62it/s]
2|2261|Loss: 0.8794291615486145:  35%|███▍      | 2260/6470 [28:15<43:17,  1.62it/s]
2|2261|Loss: 0.8794291615486145:  35%|███▍      | 2261/6470 [28:15<42:08,  1.66it/s]
2|2262|Loss: 0.8736417293548584:  35%|███▍      | 2261/6470 [28:16<42:08,  1.66it/s]
2|2262|Loss: 0.8736417293548584:  35%|███▍      | 2262/6470 [28:16<45:29,  1.54it/s]
2|2263|Loss: 0.8236623406410217:  35%|███▍      | 2262/6470 [28:17<45:29,  1.54it/s]
2|2263|Loss: 0.8236623406410217:  35%|███▍      | 2263/6470 [28:17<48:11,  1.45it/s]
2|2264|Loss: 0.9871777296066284:  35%|███▍      | 2263/6470 [28:18<48:11,  1.45it/s]
2|2264|Loss: 0.9871777296066284:  35%|███▍      | 2264/6470 [28:18<54:47,  1.28it/s]
2|2265|Loss: 0.7695971131324768:  35%|███▍      | 2264/6470 [28:18<54:47,  1.28it/s]
2|2265|Loss: 0.7695971131324768:  35%|███▌      | 2265/6470 [28:19<51:58,  1.35it/s]
2|2266|Loss: 0.873958170413971:  35%|███▌      | 2265/6470 [28:19<51:58,  1.35it/s] 
2|2266|Loss: 0.873958170413971:  35%|███▌      | 2266/6470 [28:19<51:46,  1.35it/s]
2|2267|Loss: 0.8970519304275513:  35%|███▌      | 2266/6470 [28:20<51:46,  1.35it/s]
2|2267|Loss: 0.8970519304275513:  35%|███▌      | 2267/6470 [28:20<46:14,  1.52it/s]
2|2268|Loss: 0.9320887327194214:  35%|███▌      | 2267/6470 [28:20<46:14,  1.52it/s]
2|2268|Loss: 0.9320887327194214:  35%|███▌      | 2268/6470 [28:21<45:50,  1.53it/s]
2|2269|Loss: 0.7301383018493652:  35%|███▌      | 2268/6470 [28:21<45:50,  1.53it/s]
2|2269|Loss: 0.7301383018493652:  35%|███▌      | 2269/6470 [28:21<49:30,  1.41it/s]
2|2270|Loss: 0.8198047280311584:  35%|███▌      | 2269/6470 [28:22<49:30,  1.41it/s]
2|2270|Loss: 0.8198047280311584:  35%|███▌      | 2270/6470 [28:22<53:52,  1.30it/s]
2|2271|Loss: 1.0413233041763306:  35%|███▌      | 2270/6470 [28:23<53:52,  1.30it/s]
2|2271|Loss: 1.0413233041763306:  35%|███▌      | 2271/6470 [28:23<51:38,  1.36it/s]
2|2272|Loss: 0.7559664249420166:  35%|███▌      | 2271/6470 [28:23<51:38,  1.36it/s]
2|2272|Loss: 0.7559664249420166:  35%|███▌      | 2272/6470 [28:24<54:18,  1.29it/s]
2|2273|Loss: 0.9193031191825867:  35%|███▌      | 2272/6470 [28:24<54:18,  1.29it/s]
2|2273|Loss: 0.9193031191825867:  35%|███▌      | 2273/6470 [28:25<53:46,  1.30it/s]
2|2274|Loss: 0.7866963148117065:  35%|███▌      | 2273/6470 [28:25<53:46,  1.30it/s]
2|2274|Loss: 0.7866963148117065:  35%|███▌      | 2274/6470 [28:26<1:01:05,  1.14it/s]
2|2275|Loss: 0.8747424483299255:  35%|███▌      | 2274/6470 [28:26<1:01:05,  1.14it/s]
2|2275|Loss: 0.8747424483299255:  35%|███▌      | 2275/6470 [28:26<59:35,  1.17it/s]  
2|2276|Loss: 0.8819969892501831:  35%|███▌      | 2275/6470 [28:27<59:35,  1.17it/s]
2|2276|Loss: 0.8819969892501831:  35%|███▌      | 2276/6470 [28:27<59:57,  1.17it/s]
2|2277|Loss: 0.776994526386261:  35%|███▌      | 2276/6470 [28:28<59:57,  1.17it/s] 
2|2277|Loss: 0.776994526386261:  35%|███▌      | 2277/6470 [28:28<57:33,  1.21it/s]
2|2278|Loss: 0.7379146814346313:  35%|███▌      | 2277/6470 [28:28<57:33,  1.21it/s]
2|2278|Loss: 0.7379146814346313:  35%|███▌      | 2278/6470 [28:29<56:40,  1.23it/s]
2|2279|Loss: 0.9219123125076294:  35%|███▌      | 2278/6470 [28:29<56:40,  1.23it/s]
2|2279|Loss: 0.9219123125076294:  35%|███▌      | 2279/6470 [28:30<54:23,  1.28it/s]
2|2280|Loss: 0.7569603323936462:  35%|███▌      | 2279/6470 [28:30<54:23,  1.28it/s]
2|2280|Loss: 0.7569603323936462:  35%|███▌      | 2280/6470 [28:30<56:50,  1.23it/s]
2|2281|Loss: 0.9399149417877197:  35%|███▌      | 2280/6470 [28:31<56:50,  1.23it/s]
2|2281|Loss: 0.9399149417877197:  35%|███▌      | 2281/6470 [28:31<56:25,  1.24it/s]
2|2282|Loss: 0.8267503380775452:  35%|███▌      | 2281/6470 [28:32<56:25,  1.24it/s]
2|2282|Loss: 0.8267503380775452:  35%|███▌      | 2282/6470 [28:32<51:53,  1.34it/s]
2|2283|Loss: 0.9798643589019775:  35%|███▌      | 2282/6470 [28:32<51:53,  1.34it/s]
2|2283|Loss: 0.9798643589019775:  35%|███▌      | 2283/6470 [28:33<53:47,  1.30it/s]
2|2284|Loss: 0.8464266657829285:  35%|███▌      | 2283/6470 [28:33<53:47,  1.30it/s]
2|2284|Loss: 0.8464266657829285:  35%|███▌      | 2284/6470 [28:34<56:37,  1.23it/s]
2|2285|Loss: 0.803392767906189:  35%|███▌      | 2284/6470 [28:34<56:37,  1.23it/s] 
2|2285|Loss: 0.803392767906189:  35%|███▌      | 2285/6470 [28:34<52:09,  1.34it/s]
2|2286|Loss: 0.9323880672454834:  35%|███▌      | 2285/6470 [28:35<52:09,  1.34it/s]
2|2286|Loss: 0.9323880672454834:  35%|███▌      | 2286/6470 [28:35<52:57,  1.32it/s]
2|2287|Loss: 0.9006459712982178:  35%|███▌      | 2286/6470 [28:35<52:57,  1.32it/s]
2|2287|Loss: 0.9006459712982178:  35%|███▌      | 2287/6470 [28:36<48:50,  1.43it/s]
2|2288|Loss: 0.8819694519042969:  35%|███▌      | 2287/6470 [28:36<48:50,  1.43it/s]
2|2288|Loss: 0.8819694519042969:  35%|███▌      | 2288/6470 [28:36<51:30,  1.35it/s]
2|2289|Loss: 0.8862485885620117:  35%|███▌      | 2288/6470 [28:37<51:30,  1.35it/s]
2|2289|Loss: 0.8862485885620117:  35%|███▌      | 2289/6470 [28:37<55:39,  1.25it/s]
2|2290|Loss: 0.8579726815223694:  35%|███▌      | 2289/6470 [28:38<55:39,  1.25it/s]
2|2290|Loss: 0.8579726815223694:  35%|███▌      | 2290/6470 [28:38<54:18,  1.28it/s]
2|2291|Loss: 0.7568815350532532:  35%|███▌      | 2290/6470 [28:38<54:18,  1.28it/s]
2|2291|Loss: 0.7568815350532532:  35%|███▌      | 2291/6470 [28:39<54:13,  1.28it/s]
2|2292|Loss: 0.8002642393112183:  35%|███▌      | 2291/6470 [28:39<54:13,  1.28it/s]
2|2292|Loss: 0.8002642393112183:  35%|███▌      | 2292/6470 [28:40<54:52,  1.27it/s]
2|2293|Loss: 0.8028078675270081:  35%|███▌      | 2292/6470 [28:40<54:52,  1.27it/s]
2|2293|Loss: 0.8028078675270081:  35%|███▌      | 2293/6470 [28:41<57:50,  1.20it/s]
2|2294|Loss: 0.7274871468544006:  35%|███▌      | 2293/6470 [28:41<57:50,  1.20it/s]
2|2294|Loss: 0.7274871468544006:  35%|███▌      | 2294/6470 [28:41<57:48,  1.20it/s]
2|2295|Loss: 0.7968379855155945:  35%|███▌      | 2294/6470 [28:42<57:48,  1.20it/s]
2|2295|Loss: 0.7968379855155945:  35%|███▌      | 2295/6470 [28:42<51:51,  1.34it/s]
2|2296|Loss: 1.0369848012924194:  35%|███▌      | 2295/6470 [28:42<51:51,  1.34it/s]
2|2296|Loss: 1.0369848012924194:  35%|███▌      | 2296/6470 [28:43<51:26,  1.35it/s]
2|2297|Loss: 0.855208694934845:  35%|███▌      | 2296/6470 [28:43<51:26,  1.35it/s] 
2|2297|Loss: 0.855208694934845:  36%|███▌      | 2297/6470 [28:43<52:51,  1.32it/s]
2|2298|Loss: 0.9394001960754395:  36%|███▌      | 2297/6470 [28:44<52:51,  1.32it/s]
2|2298|Loss: 0.9394001960754395:  36%|███▌      | 2298/6470 [28:44<53:30,  1.30it/s]
2|2299|Loss: 0.790417492389679:  36%|███▌      | 2298/6470 [28:45<53:30,  1.30it/s] 
2|2299|Loss: 0.790417492389679:  36%|███▌      | 2299/6470 [28:45<53:43,  1.29it/s]
2|2300|Loss: 0.8378056883811951:  36%|███▌      | 2299/6470 [28:45<53:43,  1.29it/s]
2|2300|Loss: 0.8378056883811951:  36%|███▌      | 2300/6470 [28:46<52:21,  1.33it/s]
2|2301|Loss: 0.8144863247871399:  36%|███▌      | 2300/6470 [28:46<52:21,  1.33it/s]
2|2301|Loss: 0.8144863247871399:  36%|███▌      | 2301/6470 [28:46<51:25,  1.35it/s]
2|2302|Loss: 0.9660360217094421:  36%|███▌      | 2301/6470 [28:47<51:25,  1.35it/s]
2|2302|Loss: 0.9660360217094421:  36%|███▌      | 2302/6470 [28:47<54:57,  1.26it/s]
2|2303|Loss: 0.8441641330718994:  36%|███▌      | 2302/6470 [28:48<54:57,  1.26it/s]
2|2303|Loss: 0.8441641330718994:  36%|███▌      | 2303/6470 [28:48<52:20,  1.33it/s]
2|2304|Loss: 0.8608596324920654:  36%|███▌      | 2303/6470 [28:48<52:20,  1.33it/s]
2|2304|Loss: 0.8608596324920654:  36%|███▌      | 2304/6470 [28:49<53:33,  1.30it/s]
2|2305|Loss: 0.8933479189872742:  36%|███▌      | 2304/6470 [28:49<53:33,  1.30it/s]
2|2305|Loss: 0.8933479189872742:  36%|███▌      | 2305/6470 [28:50<57:26,  1.21it/s]
2|2306|Loss: 0.8859140872955322:  36%|███▌      | 2305/6470 [28:50<57:26,  1.21it/s]
2|2306|Loss: 0.8859140872955322:  36%|███▌      | 2306/6470 [28:51<59:23,  1.17it/s]
2|2307|Loss: 0.823125958442688:  36%|███▌      | 2306/6470 [28:51<59:23,  1.17it/s] 
2|2307|Loss: 0.823125958442688:  36%|███▌      | 2307/6470 [28:52<59:31,  1.17it/s]
2|2308|Loss: 0.8982879519462585:  36%|███▌      | 2307/6470 [28:52<59:31,  1.17it/s]
2|2308|Loss: 0.8982879519462585:  36%|███▌      | 2308/6470 [28:52<58:28,  1.19it/s]
2|2309|Loss: 0.9190490245819092:  36%|███▌      | 2308/6470 [28:53<58:28,  1.19it/s]
2|2309|Loss: 0.9190490245819092:  36%|███▌      | 2309/6470 [28:53<57:41,  1.20it/s]
2|2310|Loss: 0.7633472084999084:  36%|███▌      | 2309/6470 [28:54<57:41,  1.20it/s]
2|2310|Loss: 0.7633472084999084:  36%|███▌      | 2310/6470 [28:54<58:03,  1.19it/s]
2|2311|Loss: 0.7670403718948364:  36%|███▌      | 2310/6470 [28:54<58:03,  1.19it/s]
2|2311|Loss: 0.7670403718948364:  36%|███▌      | 2311/6470 [28:55<54:50,  1.26it/s]
2|2312|Loss: 0.9125442504882812:  36%|███▌      | 2311/6470 [28:55<54:50,  1.26it/s]
2|2312|Loss: 0.9125442504882812:  36%|███▌      | 2312/6470 [28:55<50:48,  1.36it/s]
2|2313|Loss: 0.7230390310287476:  36%|███▌      | 2312/6470 [28:56<50:48,  1.36it/s]
2|2313|Loss: 0.7230390310287476:  36%|███▌      | 2313/6470 [28:56<51:41,  1.34it/s]
2|2314|Loss: 0.7587222456932068:  36%|███▌      | 2313/6470 [28:56<51:41,  1.34it/s]
2|2314|Loss: 0.7587222456932068:  36%|███▌      | 2314/6470 [28:57<49:53,  1.39it/s]
2|2315|Loss: 0.8230569958686829:  36%|███▌      | 2314/6470 [28:57<49:53,  1.39it/s]
2|2315|Loss: 0.8230569958686829:  36%|███▌      | 2315/6470 [28:57<48:15,  1.44it/s]
2|2316|Loss: 0.8668626546859741:  36%|███▌      | 2315/6470 [28:58<48:15,  1.44it/s]
2|2316|Loss: 0.8668626546859741:  36%|███▌      | 2316/6470 [28:58<49:37,  1.40it/s]
2|2317|Loss: 0.8579398989677429:  36%|███▌      | 2316/6470 [28:59<49:37,  1.40it/s]
2|2317|Loss: 0.8579398989677429:  36%|███▌      | 2317/6470 [28:59<53:32,  1.29it/s]
2|2318|Loss: 0.9083525538444519:  36%|███▌      | 2317/6470 [29:00<53:32,  1.29it/s]
2|2318|Loss: 0.9083525538444519:  36%|███▌      | 2318/6470 [29:00<53:00,  1.31it/s]
2|2319|Loss: 0.9507508873939514:  36%|███▌      | 2318/6470 [29:00<53:00,  1.31it/s]
2|2319|Loss: 0.9507508873939514:  36%|███▌      | 2319/6470 [29:01<54:31,  1.27it/s]
2|2320|Loss: 0.9115793108940125:  36%|███▌      | 2319/6470 [29:01<54:31,  1.27it/s]
2|2320|Loss: 0.9115793108940125:  36%|███▌      | 2320/6470 [29:01<54:32,  1.27it/s]
2|2321|Loss: 1.1086218357086182:  36%|███▌      | 2320/6470 [29:02<54:32,  1.27it/s]
2|2321|Loss: 1.1086218357086182:  36%|███▌      | 2321/6470 [29:02<54:47,  1.26it/s]
2|2322|Loss: 0.9048593640327454:  36%|███▌      | 2321/6470 [29:03<54:47,  1.26it/s]
2|2322|Loss: 0.9048593640327454:  36%|███▌      | 2322/6470 [29:03<53:13,  1.30it/s]
2|2323|Loss: 0.8602883219718933:  36%|███▌      | 2322/6470 [29:03<53:13,  1.30it/s]
2|2323|Loss: 0.8602883219718933:  36%|███▌      | 2323/6470 [29:04<48:40,  1.42it/s]
2|2324|Loss: 0.8472115397453308:  36%|███▌      | 2323/6470 [29:04<48:40,  1.42it/s]
2|2324|Loss: 0.8472115397453308:  36%|███▌      | 2324/6470 [29:04<46:06,  1.50it/s]
2|2325|Loss: 1.1064469814300537:  36%|███▌      | 2324/6470 [29:04<46:06,  1.50it/s]
2|2325|Loss: 1.1064469814300537:  36%|███▌      | 2325/6470 [29:05<47:28,  1.45it/s]
2|2326|Loss: 0.8969603776931763:  36%|███▌      | 2325/6470 [29:05<47:28,  1.45it/s]
2|2326|Loss: 0.8969603776931763:  36%|███▌      | 2326/6470 [29:06<50:51,  1.36it/s]
2|2327|Loss: 0.8334259986877441:  36%|███▌      | 2326/6470 [29:06<50:51,  1.36it/s]
2|2327|Loss: 0.8334259986877441:  36%|███▌      | 2327/6470 [29:07<54:52,  1.26it/s]
2|2328|Loss: 0.7664405703544617:  36%|███▌      | 2327/6470 [29:07<54:52,  1.26it/s]
2|2328|Loss: 0.7664405703544617:  36%|███▌      | 2328/6470 [29:07<52:20,  1.32it/s]
2|2329|Loss: 1.0147467851638794:  36%|███▌      | 2328/6470 [29:08<52:20,  1.32it/s]
2|2329|Loss: 1.0147467851638794:  36%|███▌      | 2329/6470 [29:08<52:39,  1.31it/s]
2|2330|Loss: 0.8161064386367798:  36%|███▌      | 2329/6470 [29:09<52:39,  1.31it/s]
2|2330|Loss: 0.8161064386367798:  36%|███▌      | 2330/6470 [29:09<53:22,  1.29it/s]
2|2331|Loss: 0.8876882791519165:  36%|███▌      | 2330/6470 [29:09<53:22,  1.29it/s]
2|2331|Loss: 0.8876882791519165:  36%|███▌      | 2331/6470 [29:10<55:01,  1.25it/s]
2|2332|Loss: 0.841392457485199:  36%|███▌      | 2331/6470 [29:10<55:01,  1.25it/s] 
2|2332|Loss: 0.841392457485199:  36%|███▌      | 2332/6470 [29:10<53:26,  1.29it/s]
2|2333|Loss: 0.8258259296417236:  36%|███▌      | 2332/6470 [29:11<53:26,  1.29it/s]
2|2333|Loss: 0.8258259296417236:  36%|███▌      | 2333/6470 [29:11<55:47,  1.24it/s]
2|2334|Loss: 0.7613421678543091:  36%|███▌      | 2333/6470 [29:12<55:47,  1.24it/s]
2|2334|Loss: 0.7613421678543091:  36%|███▌      | 2334/6470 [29:12<59:43,  1.15it/s]
2|2335|Loss: 0.8517753481864929:  36%|███▌      | 2334/6470 [29:13<59:43,  1.15it/s]
2|2335|Loss: 0.8517753481864929:  36%|███▌      | 2335/6470 [29:13<58:54,  1.17it/s]
2|2336|Loss: 0.8776179552078247:  36%|███▌      | 2335/6470 [29:14<58:54,  1.17it/s]
2|2336|Loss: 0.8776179552078247:  36%|███▌      | 2336/6470 [29:14<53:36,  1.29it/s]
2|2337|Loss: 0.8797980546951294:  36%|███▌      | 2336/6470 [29:14<53:36,  1.29it/s]
2|2337|Loss: 0.8797980546951294:  36%|███▌      | 2337/6470 [29:14<51:23,  1.34it/s]
2|2338|Loss: 0.7583319544792175:  36%|███▌      | 2337/6470 [29:15<51:23,  1.34it/s]
2|2338|Loss: 0.7583319544792175:  36%|███▌      | 2338/6470 [29:15<46:52,  1.47it/s]
2|2339|Loss: 0.6459431648254395:  36%|███▌      | 2338/6470 [29:15<46:52,  1.47it/s]
2|2339|Loss: 0.6459431648254395:  36%|███▌      | 2339/6470 [29:16<45:33,  1.51it/s]
2|2340|Loss: 0.9056715965270996:  36%|███▌      | 2339/6470 [29:16<45:33,  1.51it/s]
2|2340|Loss: 0.9056715965270996:  36%|███▌      | 2340/6470 [29:16<47:24,  1.45it/s]
2|2341|Loss: 0.8821888566017151:  36%|███▌      | 2340/6470 [29:17<47:24,  1.45it/s]
2|2341|Loss: 0.8821888566017151:  36%|███▌      | 2341/6470 [29:17<52:25,  1.31it/s]
2|2342|Loss: 0.8137717843055725:  36%|███▌      | 2341/6470 [29:18<52:25,  1.31it/s]
2|2342|Loss: 0.8137717843055725:  36%|███▌      | 2342/6470 [29:18<56:47,  1.21it/s]
2|2343|Loss: 0.8675329089164734:  36%|███▌      | 2342/6470 [29:19<56:47,  1.21it/s]
2|2343|Loss: 0.8675329089164734:  36%|███▌      | 2343/6470 [29:19<57:16,  1.20it/s]
2|2344|Loss: 0.8280428647994995:  36%|███▌      | 2343/6470 [29:19<57:16,  1.20it/s]
2|2344|Loss: 0.8280428647994995:  36%|███▌      | 2344/6470 [29:20<55:54,  1.23it/s]
2|2345|Loss: 0.6746897101402283:  36%|███▌      | 2344/6470 [29:20<55:54,  1.23it/s]
2|2345|Loss: 0.6746897101402283:  36%|███▌      | 2345/6470 [29:20<51:54,  1.32it/s]
2|2346|Loss: 0.9130536913871765:  36%|███▌      | 2345/6470 [29:21<51:54,  1.32it/s]
2|2346|Loss: 0.9130536913871765:  36%|███▋      | 2346/6470 [29:21<53:54,  1.28it/s]
2|2347|Loss: 0.7398046851158142:  36%|███▋      | 2346/6470 [29:22<53:54,  1.28it/s]
2|2347|Loss: 0.7398046851158142:  36%|███▋      | 2347/6470 [29:22<53:49,  1.28it/s]
2|2348|Loss: 0.9333616495132446:  36%|███▋      | 2347/6470 [29:22<53:49,  1.28it/s]
2|2348|Loss: 0.9333616495132446:  36%|███▋      | 2348/6470 [29:23<49:48,  1.38it/s]
2|2349|Loss: 0.9219774007797241:  36%|███▋      | 2348/6470 [29:23<49:48,  1.38it/s]
2|2349|Loss: 0.9219774007797241:  36%|███▋      | 2349/6470 [29:23<49:47,  1.38it/s]
2|2350|Loss: 0.8785343766212463:  36%|███▋      | 2349/6470 [29:24<49:47,  1.38it/s]
2|2350|Loss: 0.8785343766212463:  36%|███▋      | 2350/6470 [29:24<50:33,  1.36it/s]
2|2351|Loss: 0.8131545186042786:  36%|███▋      | 2350/6470 [29:25<50:33,  1.36it/s]
2|2351|Loss: 0.8131545186042786:  36%|███▋      | 2351/6470 [29:25<48:23,  1.42it/s]
2|2352|Loss: 0.9546604156494141:  36%|███▋      | 2351/6470 [29:25<48:23,  1.42it/s]
2|2352|Loss: 0.9546604156494141:  36%|███▋      | 2352/6470 [29:25<47:27,  1.45it/s]
2|2353|Loss: 0.7457970380783081:  36%|███▋      | 2352/6470 [29:26<47:27,  1.45it/s]
2|2353|Loss: 0.7457970380783081:  36%|███▋      | 2353/6470 [29:26<47:08,  1.46it/s]
2|2354|Loss: 0.9219213724136353:  36%|███▋      | 2353/6470 [29:27<47:08,  1.46it/s]
2|2354|Loss: 0.9219213724136353:  36%|███▋      | 2354/6470 [29:27<46:53,  1.46it/s]
2|2355|Loss: 0.7576713562011719:  36%|███▋      | 2354/6470 [29:27<46:53,  1.46it/s]
2|2355|Loss: 0.7576713562011719:  36%|███▋      | 2355/6470 [29:27<46:29,  1.48it/s]
2|2356|Loss: 0.8539184927940369:  36%|███▋      | 2355/6470 [29:28<46:29,  1.48it/s]
2|2356|Loss: 0.8539184927940369:  36%|███▋      | 2356/6470 [29:28<48:35,  1.41it/s]
2|2357|Loss: 0.7718156576156616:  36%|███▋      | 2356/6470 [29:29<48:35,  1.41it/s]
2|2357|Loss: 0.7718156576156616:  36%|███▋      | 2357/6470 [29:29<48:18,  1.42it/s]
2|2358|Loss: 0.8220517635345459:  36%|███▋      | 2357/6470 [29:29<48:18,  1.42it/s]
2|2358|Loss: 0.8220517635345459:  36%|███▋      | 2358/6470 [29:30<48:09,  1.42it/s]
2|2359|Loss: 0.828309953212738:  36%|███▋      | 2358/6470 [29:30<48:09,  1.42it/s] 
2|2359|Loss: 0.828309953212738:  36%|███▋      | 2359/6470 [29:30<48:11,  1.42it/s]
2|2360|Loss: 0.7645840048789978:  36%|███▋      | 2359/6470 [29:31<48:11,  1.42it/s]
2|2360|Loss: 0.7645840048789978:  36%|███▋      | 2360/6470 [29:31<52:31,  1.30it/s]
2|2361|Loss: 0.864453911781311:  36%|███▋      | 2360/6470 [29:32<52:31,  1.30it/s] 
2|2361|Loss: 0.864453911781311:  36%|███▋      | 2361/6470 [29:32<58:47,  1.16it/s]
2|2362|Loss: 0.7670120596885681:  36%|███▋      | 2361/6470 [29:33<58:47,  1.16it/s]
2|2362|Loss: 0.7670120596885681:  37%|███▋      | 2362/6470 [29:33<56:45,  1.21it/s]
2|2363|Loss: 0.8850541710853577:  37%|███▋      | 2362/6470 [29:34<56:45,  1.21it/s]
2|2363|Loss: 0.8850541710853577:  37%|███▋      | 2363/6470 [29:34<56:45,  1.21it/s]
2|2364|Loss: 0.8913971781730652:  37%|███▋      | 2363/6470 [29:34<56:45,  1.21it/s]
2|2364|Loss: 0.8913971781730652:  37%|███▋      | 2364/6470 [29:35<58:31,  1.17it/s]
2|2365|Loss: 0.8158831596374512:  37%|███▋      | 2364/6470 [29:35<58:31,  1.17it/s]
2|2365|Loss: 0.8158831596374512:  37%|███▋      | 2365/6470 [29:36<55:36,  1.23it/s]
2|2366|Loss: 0.9034328460693359:  37%|███▋      | 2365/6470 [29:36<55:36,  1.23it/s]
2|2366|Loss: 0.9034328460693359:  37%|███▋      | 2366/6470 [29:36<56:59,  1.20it/s]
2|2367|Loss: 0.8392600417137146:  37%|███▋      | 2366/6470 [29:37<56:59,  1.20it/s]
2|2367|Loss: 0.8392600417137146:  37%|███▋      | 2367/6470 [29:37<59:34,  1.15it/s]
2|2368|Loss: 0.8748517036437988:  37%|███▋      | 2367/6470 [29:38<59:34,  1.15it/s]
2|2368|Loss: 0.8748517036437988:  37%|███▋      | 2368/6470 [29:38<59:06,  1.16it/s]
2|2369|Loss: 0.805809497833252:  37%|███▋      | 2368/6470 [29:39<59:06,  1.16it/s] 
2|2369|Loss: 0.805809497833252:  37%|███▋      | 2369/6470 [29:39<56:14,  1.22it/s]
2|2370|Loss: 0.6807458996772766:  37%|███▋      | 2369/6470 [29:39<56:14,  1.22it/s]
2|2370|Loss: 0.6807458996772766:  37%|███▋      | 2370/6470 [29:40<53:41,  1.27it/s]
2|2371|Loss: 0.8118178248405457:  37%|███▋      | 2370/6470 [29:40<53:41,  1.27it/s]
2|2371|Loss: 0.8118178248405457:  37%|███▋      | 2371/6470 [29:40<53:53,  1.27it/s]
2|2372|Loss: 0.9396653771400452:  37%|███▋      | 2371/6470 [29:41<53:53,  1.27it/s]
2|2372|Loss: 0.9396653771400452:  37%|███▋      | 2372/6470 [29:41<53:42,  1.27it/s]
2|2373|Loss: 0.9121877551078796:  37%|███▋      | 2372/6470 [29:42<53:42,  1.27it/s]
2|2373|Loss: 0.9121877551078796:  37%|███▋      | 2373/6470 [29:42<49:35,  1.38it/s]
2|2374|Loss: 0.7193996906280518:  37%|███▋      | 2373/6470 [29:42<49:35,  1.38it/s]
2|2374|Loss: 0.7193996906280518:  37%|███▋      | 2374/6470 [29:43<51:06,  1.34it/s]
2|2375|Loss: 0.8383798003196716:  37%|███▋      | 2374/6470 [29:43<51:06,  1.34it/s]
2|2375|Loss: 0.8383798003196716:  37%|███▋      | 2375/6470 [29:43<51:36,  1.32it/s]
2|2376|Loss: 0.8463002443313599:  37%|███▋      | 2375/6470 [29:44<51:36,  1.32it/s]
2|2376|Loss: 0.8463002443313599:  37%|███▋      | 2376/6470 [29:44<52:31,  1.30it/s]
2|2377|Loss: 0.9200183153152466:  37%|███▋      | 2376/6470 [29:45<52:31,  1.30it/s]
2|2377|Loss: 0.9200183153152466:  37%|███▋      | 2377/6470 [29:45<53:30,  1.27it/s]
2|2378|Loss: 0.7729408740997314:  37%|███▋      | 2377/6470 [29:45<53:30,  1.27it/s]
2|2378|Loss: 0.7729408740997314:  37%|███▋      | 2378/6470 [29:46<51:45,  1.32it/s]
2|2379|Loss: 0.9056616425514221:  37%|███▋      | 2378/6470 [29:46<51:45,  1.32it/s]
2|2379|Loss: 0.9056616425514221:  37%|███▋      | 2379/6470 [29:47<54:29,  1.25it/s]
2|2380|Loss: 0.9682433009147644:  37%|███▋      | 2379/6470 [29:47<54:29,  1.25it/s]
2|2380|Loss: 0.9682433009147644:  37%|███▋      | 2380/6470 [29:47<50:45,  1.34it/s]
2|2381|Loss: 0.9110161662101746:  37%|███▋      | 2380/6470 [29:48<50:45,  1.34it/s]
2|2381|Loss: 0.9110161662101746:  37%|███▋      | 2381/6470 [29:48<47:12,  1.44it/s]
2|2382|Loss: 0.7641472220420837:  37%|███▋      | 2381/6470 [29:48<47:12,  1.44it/s]
2|2382|Loss: 0.7641472220420837:  37%|███▋      | 2382/6470 [29:49<48:38,  1.40it/s]
2|2383|Loss: 0.8784309029579163:  37%|███▋      | 2382/6470 [29:49<48:38,  1.40it/s]
2|2383|Loss: 0.8784309029579163:  37%|███▋      | 2383/6470 [29:49<45:10,  1.51it/s]
2|2384|Loss: 0.8386756777763367:  37%|███▋      | 2383/6470 [29:49<45:10,  1.51it/s]
2|2384|Loss: 0.8386756777763367:  37%|███▋      | 2384/6470 [29:50<46:38,  1.46it/s]
2|2385|Loss: 0.8024327754974365:  37%|███▋      | 2384/6470 [29:50<46:38,  1.46it/s]
2|2385|Loss: 0.8024327754974365:  37%|███▋      | 2385/6470 [29:51<48:20,  1.41it/s]
2|2386|Loss: 0.6340791583061218:  37%|███▋      | 2385/6470 [29:51<48:20,  1.41it/s]
2|2386|Loss: 0.6340791583061218:  37%|███▋      | 2386/6470 [29:51<47:05,  1.45it/s]
2|2387|Loss: 0.7811479568481445:  37%|███▋      | 2386/6470 [29:52<47:05,  1.45it/s]
2|2387|Loss: 0.7811479568481445:  37%|███▋      | 2387/6470 [29:52<48:51,  1.39it/s]
2|2388|Loss: 0.8892322778701782:  37%|███▋      | 2387/6470 [29:53<48:51,  1.39it/s]
2|2388|Loss: 0.8892322778701782:  37%|███▋      | 2388/6470 [29:53<53:41,  1.27it/s]
2|2389|Loss: 1.001088261604309:  37%|███▋      | 2388/6470 [29:54<53:41,  1.27it/s] 
2|2389|Loss: 1.001088261604309:  37%|███▋      | 2389/6470 [29:54<54:16,  1.25it/s]
2|2390|Loss: 0.8895975947380066:  37%|███▋      | 2389/6470 [29:54<54:16,  1.25it/s]
2|2390|Loss: 0.8895975947380066:  37%|███▋      | 2390/6470 [29:55<54:03,  1.26it/s]
2|2391|Loss: 0.7341786026954651:  37%|███▋      | 2390/6470 [29:55<54:03,  1.26it/s]
2|2391|Loss: 0.7341786026954651:  37%|███▋      | 2391/6470 [29:55<55:24,  1.23it/s]
2|2392|Loss: 0.7645799517631531:  37%|███▋      | 2391/6470 [29:56<55:24,  1.23it/s]
2|2392|Loss: 0.7645799517631531:  37%|███▋      | 2392/6470 [29:56<53:54,  1.26it/s]
2|2393|Loss: 0.8410575985908508:  37%|███▋      | 2392/6470 [29:57<53:54,  1.26it/s]
2|2393|Loss: 0.8410575985908508:  37%|███▋      | 2393/6470 [29:57<56:08,  1.21it/s]
2|2394|Loss: 1.038695216178894:  37%|███▋      | 2393/6470 [29:58<56:08,  1.21it/s] 
2|2394|Loss: 1.038695216178894:  37%|███▋      | 2394/6470 [29:58<54:12,  1.25it/s]
2|2395|Loss: 0.9216718077659607:  37%|███▋      | 2394/6470 [29:58<54:12,  1.25it/s]
2|2395|Loss: 0.9216718077659607:  37%|███▋      | 2395/6470 [29:59<51:37,  1.32it/s]
2|2396|Loss: 0.695270299911499:  37%|███▋      | 2395/6470 [29:59<51:37,  1.32it/s] 
2|2396|Loss: 0.695270299911499:  37%|███▋      | 2396/6470 [29:59<48:57,  1.39it/s]
2|2397|Loss: 0.8044515252113342:  37%|███▋      | 2396/6470 [30:00<48:57,  1.39it/s]
2|2397|Loss: 0.8044515252113342:  37%|███▋      | 2397/6470 [30:00<50:55,  1.33it/s]
2|2398|Loss: 0.809202253818512:  37%|███▋      | 2397/6470 [30:00<50:55,  1.33it/s] 
2|2398|Loss: 0.809202253818512:  37%|███▋      | 2398/6470 [30:01<52:01,  1.30it/s]
2|2399|Loss: 0.8026641011238098:  37%|███▋      | 2398/6470 [30:01<52:01,  1.30it/s]
2|2399|Loss: 0.8026641011238098:  37%|███▋      | 2399/6470 [30:02<54:34,  1.24it/s]
2|2400|Loss: 0.8827283382415771:  37%|███▋      | 2399/6470 [30:02<54:34,  1.24it/s]
2|2400|Loss: 0.8827283382415771:  37%|███▋      | 2400/6470 [30:02<54:17,  1.25it/s]
2|2401|Loss: 0.8967946171760559:  37%|███▋      | 2400/6470 [30:03<54:17,  1.25it/s]
2|2401|Loss: 0.8967946171760559:  37%|███▋      | 2401/6470 [30:03<51:54,  1.31it/s]
2|2402|Loss: 0.9328229427337646:  37%|███▋      | 2401/6470 [30:04<51:54,  1.31it/s]
2|2402|Loss: 0.9328229427337646:  37%|███▋      | 2402/6470 [30:04<54:16,  1.25it/s]
2|2403|Loss: 0.8754791617393494:  37%|███▋      | 2402/6470 [30:04<54:16,  1.25it/s]
2|2403|Loss: 0.8754791617393494:  37%|███▋      | 2403/6470 [30:05<52:42,  1.29it/s]
2|2404|Loss: 0.9084405303001404:  37%|███▋      | 2403/6470 [30:05<52:42,  1.29it/s]
2|2404|Loss: 0.9084405303001404:  37%|███▋      | 2404/6470 [30:05<51:09,  1.32it/s]
2|2405|Loss: 0.7417764663696289:  37%|███▋      | 2404/6470 [30:06<51:09,  1.32it/s]
2|2405|Loss: 0.7417764663696289:  37%|███▋      | 2405/6470 [30:06<54:14,  1.25it/s]
2|2406|Loss: 0.9299914240837097:  37%|███▋      | 2405/6470 [30:07<54:14,  1.25it/s]
2|2406|Loss: 0.9299914240837097:  37%|███▋      | 2406/6470 [30:07<56:54,  1.19it/s]
2|2407|Loss: 0.8733795881271362:  37%|███▋      | 2406/6470 [30:08<56:54,  1.19it/s]
2|2407|Loss: 0.8733795881271362:  37%|███▋      | 2407/6470 [30:08<56:31,  1.20it/s]
2|2408|Loss: 0.9644884467124939:  37%|███▋      | 2407/6470 [30:09<56:31,  1.20it/s]
2|2408|Loss: 0.9644884467124939:  37%|███▋      | 2408/6470 [30:09<55:05,  1.23it/s]
2|2409|Loss: 0.8743199110031128:  37%|███▋      | 2408/6470 [30:09<55:05,  1.23it/s]
2|2409|Loss: 0.8743199110031128:  37%|███▋      | 2409/6470 [30:10<54:41,  1.24it/s]
2|2410|Loss: 0.792961835861206:  37%|███▋      | 2409/6470 [30:10<54:41,  1.24it/s] 
2|2410|Loss: 0.792961835861206:  37%|███▋      | 2410/6470 [30:10<54:50,  1.23it/s]
2|2411|Loss: 0.8398855328559875:  37%|███▋      | 2410/6470 [30:11<54:50,  1.23it/s]
2|2411|Loss: 0.8398855328559875:  37%|███▋      | 2411/6470 [30:11<53:15,  1.27it/s]
2|2412|Loss: 0.8664337396621704:  37%|███▋      | 2411/6470 [30:12<53:15,  1.27it/s]
2|2412|Loss: 0.8664337396621704:  37%|███▋      | 2412/6470 [30:12<54:18,  1.25it/s]
2|2413|Loss: 0.8152691721916199:  37%|███▋      | 2412/6470 [30:13<54:18,  1.25it/s]
2|2413|Loss: 0.8152691721916199:  37%|███▋      | 2413/6470 [30:13<58:36,  1.15it/s]
2|2414|Loss: 0.8270836472511292:  37%|███▋      | 2413/6470 [30:14<58:36,  1.15it/s]
2|2414|Loss: 0.8270836472511292:  37%|███▋      | 2414/6470 [30:14<54:11,  1.25it/s]
2|2415|Loss: 0.8409600853919983:  37%|███▋      | 2414/6470 [30:14<54:11,  1.25it/s]
2|2415|Loss: 0.8409600853919983:  37%|███▋      | 2415/6470 [30:15<55:24,  1.22it/s]
2|2416|Loss: 0.8965069055557251:  37%|███▋      | 2415/6470 [30:15<55:24,  1.22it/s]
2|2416|Loss: 0.8965069055557251:  37%|███▋      | 2416/6470 [30:15<55:28,  1.22it/s]
2|2417|Loss: 0.9344936609268188:  37%|███▋      | 2416/6470 [30:16<55:28,  1.22it/s]
2|2417|Loss: 0.9344936609268188:  37%|███▋      | 2417/6470 [30:16<50:03,  1.35it/s]
2|2418|Loss: 0.769313633441925:  37%|███▋      | 2417/6470 [30:16<50:03,  1.35it/s] 
2|2418|Loss: 0.769313633441925:  37%|███▋      | 2418/6470 [30:17<49:49,  1.36it/s]
2|2419|Loss: 0.9603461623191833:  37%|███▋      | 2418/6470 [30:17<49:49,  1.36it/s]
2|2419|Loss: 0.9603461623191833:  37%|███▋      | 2419/6470 [30:17<50:42,  1.33it/s]
2|2420|Loss: 0.7471541166305542:  37%|███▋      | 2419/6470 [30:18<50:42,  1.33it/s]
2|2420|Loss: 0.7471541166305542:  37%|███▋      | 2420/6470 [30:18<51:31,  1.31it/s]
2|2421|Loss: 0.8427425026893616:  37%|███▋      | 2420/6470 [30:19<51:31,  1.31it/s]
2|2421|Loss: 0.8427425026893616:  37%|███▋      | 2421/6470 [30:19<51:24,  1.31it/s]
2|2422|Loss: 0.7251732349395752:  37%|███▋      | 2421/6470 [30:19<51:24,  1.31it/s]
2|2422|Loss: 0.7251732349395752:  37%|███▋      | 2422/6470 [30:20<46:50,  1.44it/s]
2|2423|Loss: 0.7566617727279663:  37%|███▋      | 2422/6470 [30:20<46:50,  1.44it/s]
2|2423|Loss: 0.7566617727279663:  37%|███▋      | 2423/6470 [30:20<46:50,  1.44it/s]
2|2424|Loss: 0.835883378982544:  37%|███▋      | 2423/6470 [30:21<46:50,  1.44it/s] 
2|2424|Loss: 0.835883378982544:  37%|███▋      | 2424/6470 [30:21<50:40,  1.33it/s]
2|2425|Loss: 0.8715766668319702:  37%|███▋      | 2424/6470 [30:22<50:40,  1.33it/s]
2|2425|Loss: 0.8715766668319702:  37%|███▋      | 2425/6470 [30:22<50:08,  1.34it/s]
2|2426|Loss: 0.7673575282096863:  37%|███▋      | 2425/6470 [30:22<50:08,  1.34it/s]
2|2426|Loss: 0.7673575282096863:  37%|███▋      | 2426/6470 [30:23<50:20,  1.34it/s]
2|2427|Loss: 0.9169855713844299:  37%|███▋      | 2426/6470 [30:23<50:20,  1.34it/s]
2|2427|Loss: 0.9169855713844299:  38%|███▊      | 2427/6470 [30:23<47:28,  1.42it/s]
2|2428|Loss: 0.9517828822135925:  38%|███▊      | 2427/6470 [30:24<47:28,  1.42it/s]
2|2428|Loss: 0.9517828822135925:  38%|███▊      | 2428/6470 [30:24<46:48,  1.44it/s]
2|2429|Loss: 0.7972385287284851:  38%|███▊      | 2428/6470 [30:24<46:48,  1.44it/s]
2|2429|Loss: 0.7972385287284851:  38%|███▊      | 2429/6470 [30:25<48:01,  1.40it/s]
2|2430|Loss: 0.8425086736679077:  38%|███▊      | 2429/6470 [30:25<48:01,  1.40it/s]
2|2430|Loss: 0.8425086736679077:  38%|███▊      | 2430/6470 [30:25<49:50,  1.35it/s]
2|2431|Loss: 0.8120460510253906:  38%|███▊      | 2430/6470 [30:26<49:50,  1.35it/s]
2|2431|Loss: 0.8120460510253906:  38%|███▊      | 2431/6470 [30:26<53:25,  1.26it/s]
2|2432|Loss: 0.7251624464988708:  38%|███▊      | 2431/6470 [30:27<53:25,  1.26it/s]
2|2432|Loss: 0.7251624464988708:  38%|███▊      | 2432/6470 [30:27<52:20,  1.29it/s]
2|2433|Loss: 0.7723262906074524:  38%|███▊      | 2432/6470 [30:27<52:20,  1.29it/s]
2|2433|Loss: 0.7723262906074524:  38%|███▊      | 2433/6470 [30:28<49:27,  1.36it/s]
2|2434|Loss: 0.8498864769935608:  38%|███▊      | 2433/6470 [30:28<49:27,  1.36it/s]
2|2434|Loss: 0.8498864769935608:  38%|███▊      | 2434/6470 [30:29<50:15,  1.34it/s]
2|2435|Loss: 0.8512430191040039:  38%|███▊      | 2434/6470 [30:29<50:15,  1.34it/s]
2|2435|Loss: 0.8512430191040039:  38%|███▊      | 2435/6470 [30:29<49:37,  1.36it/s]
2|2436|Loss: 0.974922776222229:  38%|███▊      | 2435/6470 [30:30<49:37,  1.36it/s] 
2|2436|Loss: 0.974922776222229:  38%|███▊      | 2436/6470 [30:30<49:03,  1.37it/s]
2|2437|Loss: 0.8101608753204346:  38%|███▊      | 2436/6470 [30:30<49:03,  1.37it/s]
2|2437|Loss: 0.8101608753204346:  38%|███▊      | 2437/6470 [30:31<48:08,  1.40it/s]
2|2438|Loss: 0.9308981895446777:  38%|███▊      | 2437/6470 [30:31<48:08,  1.40it/s]
2|2438|Loss: 0.9308981895446777:  38%|███▊      | 2438/6470 [30:31<49:27,  1.36it/s]
2|2439|Loss: 0.9865974187850952:  38%|███▊      | 2438/6470 [30:32<49:27,  1.36it/s]
2|2439|Loss: 0.9865974187850952:  38%|███▊      | 2439/6470 [30:32<50:17,  1.34it/s]
2|2440|Loss: 0.8394575119018555:  38%|███▊      | 2439/6470 [30:33<50:17,  1.34it/s]
2|2440|Loss: 0.8394575119018555:  38%|███▊      | 2440/6470 [30:33<48:17,  1.39it/s]
2|2441|Loss: 0.879463255405426:  38%|███▊      | 2440/6470 [30:33<48:17,  1.39it/s] 
2|2441|Loss: 0.879463255405426:  38%|███▊      | 2441/6470 [30:34<51:22,  1.31it/s]
2|2442|Loss: 0.8724173903465271:  38%|███▊      | 2441/6470 [30:34<51:22,  1.31it/s]
2|2442|Loss: 0.8724173903465271:  38%|███▊      | 2442/6470 [30:34<50:43,  1.32it/s]
2|2443|Loss: 0.8124454617500305:  38%|███▊      | 2442/6470 [30:35<50:43,  1.32it/s]
2|2443|Loss: 0.8124454617500305:  38%|███▊      | 2443/6470 [30:35<53:31,  1.25it/s]
2|2444|Loss: 0.9158597588539124:  38%|███▊      | 2443/6470 [30:36<53:31,  1.25it/s]
2|2444|Loss: 0.9158597588539124:  38%|███▊      | 2444/6470 [30:36<53:43,  1.25it/s]
2|2445|Loss: 0.9678478240966797:  38%|███▊      | 2444/6470 [30:37<53:43,  1.25it/s]
2|2445|Loss: 0.9678478240966797:  38%|███▊      | 2445/6470 [30:37<51:40,  1.30it/s]
2|2446|Loss: 0.7794427275657654:  38%|███▊      | 2445/6470 [30:37<51:40,  1.30it/s]
2|2446|Loss: 0.7794427275657654:  38%|███▊      | 2446/6470 [30:38<49:05,  1.37it/s]
2|2447|Loss: 0.8584527969360352:  38%|███▊      | 2446/6470 [30:38<49:05,  1.37it/s]
2|2447|Loss: 0.8584527969360352:  38%|███▊      | 2447/6470 [30:38<51:33,  1.30it/s]
2|2448|Loss: 0.7700495719909668:  38%|███▊      | 2447/6470 [30:39<51:33,  1.30it/s]
2|2448|Loss: 0.7700495719909668:  38%|███▊      | 2448/6470 [30:39<51:59,  1.29it/s]
2|2449|Loss: 0.8268938660621643:  38%|███▊      | 2448/6470 [30:40<51:59,  1.29it/s]
2|2449|Loss: 0.8268938660621643:  38%|███▊      | 2449/6470 [30:40<52:10,  1.28it/s]
2|2450|Loss: 0.7814498543739319:  38%|███▊      | 2449/6470 [30:40<52:10,  1.28it/s]
2|2450|Loss: 0.7814498543739319:  38%|███▊      | 2450/6470 [30:41<48:41,  1.38it/s]
2|2451|Loss: 0.8936949968338013:  38%|███▊      | 2450/6470 [30:41<48:41,  1.38it/s]
2|2451|Loss: 0.8936949968338013:  38%|███▊      | 2451/6470 [30:41<47:40,  1.40it/s]
2|2452|Loss: 0.9339433312416077:  38%|███▊      | 2451/6470 [30:42<47:40,  1.40it/s]
2|2452|Loss: 0.9339433312416077:  38%|███▊      | 2452/6470 [30:42<43:51,  1.53it/s]
2|2453|Loss: 0.9352777004241943:  38%|███▊      | 2452/6470 [30:42<43:51,  1.53it/s]
2|2453|Loss: 0.9352777004241943:  38%|███▊      | 2453/6470 [30:42<40:46,  1.64it/s]
2|2454|Loss: 0.9231510162353516:  38%|███▊      | 2453/6470 [30:43<40:46,  1.64it/s]
2|2454|Loss: 0.9231510162353516:  38%|███▊      | 2454/6470 [30:43<42:58,  1.56it/s]
2|2455|Loss: 0.8359275460243225:  38%|███▊      | 2454/6470 [30:43<42:58,  1.56it/s]
2|2455|Loss: 0.8359275460243225:  38%|███▊      | 2455/6470 [30:44<46:10,  1.45it/s]
2|2456|Loss: 0.742361843585968:  38%|███▊      | 2455/6470 [30:44<46:10,  1.45it/s] 
2|2456|Loss: 0.742361843585968:  38%|███▊      | 2456/6470 [30:44<45:22,  1.47it/s]
2|2457|Loss: 0.8611869215965271:  38%|███▊      | 2456/6470 [30:45<45:22,  1.47it/s]
2|2457|Loss: 0.8611869215965271:  38%|███▊      | 2457/6470 [30:45<46:25,  1.44it/s]
2|2458|Loss: 0.7990529537200928:  38%|███▊      | 2457/6470 [30:46<46:25,  1.44it/s]
2|2458|Loss: 0.7990529537200928:  38%|███▊      | 2458/6470 [30:46<44:28,  1.50it/s]
2|2459|Loss: 0.9464289546012878:  38%|███▊      | 2458/6470 [30:46<44:28,  1.50it/s]
2|2459|Loss: 0.9464289546012878:  38%|███▊      | 2459/6470 [30:46<45:25,  1.47it/s]
2|2460|Loss: 0.8249133229255676:  38%|███▊      | 2459/6470 [30:47<45:25,  1.47it/s]
2|2460|Loss: 0.8249133229255676:  38%|███▊      | 2460/6470 [30:47<45:32,  1.47it/s]
2|2461|Loss: 0.8397284746170044:  38%|███▊      | 2460/6470 [30:48<45:32,  1.47it/s]
2|2461|Loss: 0.8397284746170044:  38%|███▊      | 2461/6470 [30:48<45:57,  1.45it/s]
2|2462|Loss: 0.8510398268699646:  38%|███▊      | 2461/6470 [30:48<45:57,  1.45it/s]
2|2462|Loss: 0.8510398268699646:  38%|███▊      | 2462/6470 [30:49<45:18,  1.47it/s]
2|2463|Loss: 0.8603039383888245:  38%|███▊      | 2462/6470 [30:49<45:18,  1.47it/s]
2|2463|Loss: 0.8603039383888245:  38%|███▊      | 2463/6470 [30:49<47:12,  1.41it/s]
2|2464|Loss: 1.0087870359420776:  38%|███▊      | 2463/6470 [30:50<47:12,  1.41it/s]
2|2464|Loss: 1.0087870359420776:  38%|███▊      | 2464/6470 [30:50<48:31,  1.38it/s]
2|2465|Loss: 0.7804926037788391:  38%|███▊      | 2464/6470 [30:50<48:31,  1.38it/s]
2|2465|Loss: 0.7804926037788391:  38%|███▊      | 2465/6470 [30:51<47:59,  1.39it/s]
2|2466|Loss: 0.7304123044013977:  38%|███▊      | 2465/6470 [30:51<47:59,  1.39it/s]
2|2466|Loss: 0.7304123044013977:  38%|███▊      | 2466/6470 [30:52<50:47,  1.31it/s]
2|2467|Loss: 0.91387939453125:  38%|███▊      | 2466/6470 [30:52<50:47,  1.31it/s]  
2|2467|Loss: 0.91387939453125:  38%|███▊      | 2467/6470 [30:53<53:36,  1.24it/s]
2|2468|Loss: 0.8163312673568726:  38%|███▊      | 2467/6470 [30:53<53:36,  1.24it/s]
2|2468|Loss: 0.8163312673568726:  38%|███▊      | 2468/6470 [30:53<55:08,  1.21it/s]
2|2469|Loss: 0.7861605286598206:  38%|███▊      | 2468/6470 [30:54<55:08,  1.21it/s]
2|2469|Loss: 0.7861605286598206:  38%|███▊      | 2469/6470 [30:54<57:01,  1.17it/s]
2|2470|Loss: 0.9055402874946594:  38%|███▊      | 2469/6470 [30:55<57:01,  1.17it/s]
2|2470|Loss: 0.9055402874946594:  38%|███▊      | 2470/6470 [30:55<53:04,  1.26it/s]
2|2471|Loss: 0.8186054229736328:  38%|███▊      | 2470/6470 [30:55<53:04,  1.26it/s]
2|2471|Loss: 0.8186054229736328:  38%|███▊      | 2471/6470 [30:56<51:37,  1.29it/s]
2|2472|Loss: 0.8259192705154419:  38%|███▊      | 2471/6470 [30:56<51:37,  1.29it/s]
2|2472|Loss: 0.8259192705154419:  38%|███▊      | 2472/6470 [30:57<52:14,  1.28it/s]
2|2473|Loss: 0.8017574548721313:  38%|███▊      | 2472/6470 [30:57<52:14,  1.28it/s]
2|2473|Loss: 0.8017574548721313:  38%|███▊      | 2473/6470 [30:57<55:08,  1.21it/s]
2|2474|Loss: 0.7630656361579895:  38%|███▊      | 2473/6470 [30:58<55:08,  1.21it/s]
2|2474|Loss: 0.7630656361579895:  38%|███▊      | 2474/6470 [30:58<52:31,  1.27it/s]
2|2475|Loss: 0.8801410794258118:  38%|███▊      | 2474/6470 [30:59<52:31,  1.27it/s]
2|2475|Loss: 0.8801410794258118:  38%|███▊      | 2475/6470 [30:59<53:29,  1.24it/s]
2|2476|Loss: 0.7984176874160767:  38%|███▊      | 2475/6470 [30:59<53:29,  1.24it/s]
2|2476|Loss: 0.7984176874160767:  38%|███▊      | 2476/6470 [31:00<53:36,  1.24it/s]
2|2477|Loss: 0.7303067445755005:  38%|███▊      | 2476/6470 [31:00<53:36,  1.24it/s]
2|2477|Loss: 0.7303067445755005:  38%|███▊      | 2477/6470 [31:01<52:39,  1.26it/s]
2|2478|Loss: 0.7585890293121338:  38%|███▊      | 2477/6470 [31:01<52:39,  1.26it/s]
2|2478|Loss: 0.7585890293121338:  38%|███▊      | 2478/6470 [31:01<52:32,  1.27it/s]
2|2479|Loss: 0.7728027105331421:  38%|███▊      | 2478/6470 [31:02<52:32,  1.27it/s]
2|2479|Loss: 0.7728027105331421:  38%|███▊      | 2479/6470 [31:02<49:51,  1.33it/s]
2|2480|Loss: 0.7020699381828308:  38%|███▊      | 2479/6470 [31:02<49:51,  1.33it/s]
2|2480|Loss: 0.7020699381828308:  38%|███▊      | 2480/6470 [31:03<50:03,  1.33it/s]
2|2481|Loss: 0.854892909526825:  38%|███▊      | 2480/6470 [31:03<50:03,  1.33it/s] 
2|2481|Loss: 0.854892909526825:  38%|███▊      | 2481/6470 [31:03<48:54,  1.36it/s]
2|2482|Loss: 0.9402509331703186:  38%|███▊      | 2481/6470 [31:04<48:54,  1.36it/s]
2|2482|Loss: 0.9402509331703186:  38%|███▊      | 2482/6470 [31:04<49:01,  1.36it/s]
2|2483|Loss: 0.7666955590248108:  38%|███▊      | 2482/6470 [31:05<49:01,  1.36it/s]
2|2483|Loss: 0.7666955590248108:  38%|███▊      | 2483/6470 [31:05<45:01,  1.48it/s]
2|2484|Loss: 0.8743616342544556:  38%|███▊      | 2483/6470 [31:05<45:01,  1.48it/s]
2|2484|Loss: 0.8743616342544556:  38%|███▊      | 2484/6470 [31:05<45:29,  1.46it/s]
2|2485|Loss: 0.925945520401001:  38%|███▊      | 2484/6470 [31:06<45:29,  1.46it/s] 
2|2485|Loss: 0.925945520401001:  38%|███▊      | 2485/6470 [31:06<48:02,  1.38it/s]
2|2486|Loss: 0.7618522644042969:  38%|███▊      | 2485/6470 [31:07<48:02,  1.38it/s]
2|2486|Loss: 0.7618522644042969:  38%|███▊      | 2486/6470 [31:07<49:07,  1.35it/s]
2|2487|Loss: 0.9360505938529968:  38%|███▊      | 2486/6470 [31:07<49:07,  1.35it/s]
2|2487|Loss: 0.9360505938529968:  38%|███▊      | 2487/6470 [31:08<51:18,  1.29it/s]
2|2488|Loss: 0.7921605706214905:  38%|███▊      | 2487/6470 [31:08<51:18,  1.29it/s]
2|2488|Loss: 0.7921605706214905:  38%|███▊      | 2488/6470 [31:09<52:03,  1.28it/s]
2|2489|Loss: 1.042988657951355:  38%|███▊      | 2488/6470 [31:09<52:03,  1.28it/s] 
2|2489|Loss: 1.042988657951355:  38%|███▊      | 2489/6470 [31:09<52:29,  1.26it/s]
2|2490|Loss: 0.7761364579200745:  38%|███▊      | 2489/6470 [31:10<52:29,  1.26it/s]
2|2490|Loss: 0.7761364579200745:  38%|███▊      | 2490/6470 [31:10<53:22,  1.24it/s]
2|2491|Loss: 0.801728367805481:  38%|███▊      | 2490/6470 [31:11<53:22,  1.24it/s] 
2|2491|Loss: 0.801728367805481:  39%|███▊      | 2491/6470 [31:11<51:29,  1.29it/s]
2|2492|Loss: 0.8425790071487427:  39%|███▊      | 2491/6470 [31:11<51:29,  1.29it/s]
2|2492|Loss: 0.8425790071487427:  39%|███▊      | 2492/6470 [31:12<49:12,  1.35it/s]
2|2493|Loss: 0.7917405962944031:  39%|███▊      | 2492/6470 [31:12<49:12,  1.35it/s]
2|2493|Loss: 0.7917405962944031:  39%|███▊      | 2493/6470 [31:12<48:28,  1.37it/s]
2|2494|Loss: 0.9060091376304626:  39%|███▊      | 2493/6470 [31:13<48:28,  1.37it/s]
2|2494|Loss: 0.9060091376304626:  39%|███▊      | 2494/6470 [31:13<48:55,  1.35it/s]
2|2495|Loss: 0.8246762752532959:  39%|███▊      | 2494/6470 [31:14<48:55,  1.35it/s]
2|2495|Loss: 0.8246762752532959:  39%|███▊      | 2495/6470 [31:14<46:24,  1.43it/s]
2|2496|Loss: 0.7052755355834961:  39%|███▊      | 2495/6470 [31:14<46:24,  1.43it/s]
2|2496|Loss: 0.7052755355834961:  39%|███▊      | 2496/6470 [31:14<46:52,  1.41it/s]
2|2497|Loss: 0.8275564908981323:  39%|███▊      | 2496/6470 [31:15<46:52,  1.41it/s]
2|2497|Loss: 0.8275564908981323:  39%|███▊      | 2497/6470 [31:15<51:02,  1.30it/s]
2|2498|Loss: 0.8649336099624634:  39%|███▊      | 2497/6470 [31:16<51:02,  1.30it/s]
2|2498|Loss: 0.8649336099624634:  39%|███▊      | 2498/6470 [31:16<54:44,  1.21it/s]
2|2499|Loss: 0.7838283777236938:  39%|███▊      | 2498/6470 [31:17<54:44,  1.21it/s]
2|2499|Loss: 0.7838283777236938:  39%|███▊      | 2499/6470 [31:17<54:38,  1.21it/s]
2|2500|Loss: 0.9656386971473694:  39%|███▊      | 2499/6470 [31:18<54:38,  1.21it/s]
2|2500|Loss: 0.9656386971473694:  39%|███▊      | 2500/6470 [31:18<55:49,  1.19it/s]
2|2501|Loss: 0.7340810298919678:  39%|███▊      | 2500/6470 [31:19<55:49,  1.19it/s]
2|2501|Loss: 0.7340810298919678:  39%|███▊      | 2501/6470 [31:19<54:31,  1.21it/s]
2|2502|Loss: 0.907489001750946:  39%|███▊      | 2501/6470 [31:19<54:31,  1.21it/s] 
2|2502|Loss: 0.907489001750946:  39%|███▊      | 2502/6470 [31:20<53:06,  1.25it/s]
2|2503|Loss: 0.8427577614784241:  39%|███▊      | 2502/6470 [31:20<53:06,  1.25it/s]
2|2503|Loss: 0.8427577614784241:  39%|███▊      | 2503/6470 [31:20<53:17,  1.24it/s]
2|2504|Loss: 0.7401672601699829:  39%|███▊      | 2503/6470 [31:21<53:17,  1.24it/s]
2|2504|Loss: 0.7401672601699829:  39%|███▊      | 2504/6470 [31:21<55:32,  1.19it/s]
2|2505|Loss: 0.8867718577384949:  39%|███▊      | 2504/6470 [31:22<55:32,  1.19it/s]
2|2505|Loss: 0.8867718577384949:  39%|███▊      | 2505/6470 [31:22<55:58,  1.18it/s]
2|2506|Loss: 0.7387531995773315:  39%|███▊      | 2505/6470 [31:23<55:58,  1.18it/s]
2|2506|Loss: 0.7387531995773315:  39%|███▊      | 2506/6470 [31:23<53:05,  1.24it/s]
2|2507|Loss: 0.7149912118911743:  39%|███▊      | 2506/6470 [31:23<53:05,  1.24it/s]
2|2507|Loss: 0.7149912118911743:  39%|███▊      | 2507/6470 [31:24<50:12,  1.32it/s]
2|2508|Loss: 0.8701953291893005:  39%|███▊      | 2507/6470 [31:24<50:12,  1.32it/s]
2|2508|Loss: 0.8701953291893005:  39%|███▉      | 2508/6470 [31:24<47:34,  1.39it/s]
2|2509|Loss: 0.8506177067756653:  39%|███▉      | 2508/6470 [31:25<47:34,  1.39it/s]
2|2509|Loss: 0.8506177067756653:  39%|███▉      | 2509/6470 [31:25<43:17,  1.53it/s]
2|2510|Loss: 0.9579868316650391:  39%|███▉      | 2509/6470 [31:25<43:17,  1.53it/s]
2|2510|Loss: 0.9579868316650391:  39%|███▉      | 2510/6470 [31:25<44:56,  1.47it/s]
2|2511|Loss: 0.7882624864578247:  39%|███▉      | 2510/6470 [31:26<44:56,  1.47it/s]
2|2511|Loss: 0.7882624864578247:  39%|███▉      | 2511/6470 [31:26<45:49,  1.44it/s]
2|2512|Loss: 0.9062435030937195:  39%|███▉      | 2511/6470 [31:27<45:49,  1.44it/s]
2|2512|Loss: 0.9062435030937195:  39%|███▉      | 2512/6470 [31:27<45:33,  1.45it/s]
2|2513|Loss: 0.8118067383766174:  39%|███▉      | 2512/6470 [31:27<45:33,  1.45it/s]
2|2513|Loss: 0.8118067383766174:  39%|███▉      | 2513/6470 [31:28<46:26,  1.42it/s]
2|2514|Loss: 0.911005973815918:  39%|███▉      | 2513/6470 [31:28<46:26,  1.42it/s] 
2|2514|Loss: 0.911005973815918:  39%|███▉      | 2514/6470 [31:28<50:05,  1.32it/s]
2|2515|Loss: 0.8343362212181091:  39%|███▉      | 2514/6470 [31:29<50:05,  1.32it/s]
2|2515|Loss: 0.8343362212181091:  39%|███▉      | 2515/6470 [31:29<52:51,  1.25it/s]
2|2516|Loss: 0.8230241537094116:  39%|███▉      | 2515/6470 [31:30<52:51,  1.25it/s]
2|2516|Loss: 0.8230241537094116:  39%|███▉      | 2516/6470 [31:30<48:43,  1.35it/s]
2|2517|Loss: 0.9044601321220398:  39%|███▉      | 2516/6470 [31:30<48:43,  1.35it/s]
2|2517|Loss: 0.9044601321220398:  39%|███▉      | 2517/6470 [31:31<45:17,  1.45it/s]
2|2518|Loss: 1.010716199874878:  39%|███▉      | 2517/6470 [31:31<45:17,  1.45it/s] 
2|2518|Loss: 1.010716199874878:  39%|███▉      | 2518/6470 [31:31<50:40,  1.30it/s]
2|2519|Loss: 0.9411000609397888:  39%|███▉      | 2518/6470 [31:32<50:40,  1.30it/s]
2|2519|Loss: 0.9411000609397888:  39%|███▉      | 2519/6470 [31:32<50:37,  1.30it/s]
2|2520|Loss: 0.9076384902000427:  39%|███▉      | 2519/6470 [31:33<50:37,  1.30it/s]
2|2520|Loss: 0.9076384902000427:  39%|███▉      | 2520/6470 [31:33<50:31,  1.30it/s]
2|2521|Loss: 0.8009023666381836:  39%|███▉      | 2520/6470 [31:33<50:31,  1.30it/s]
2|2521|Loss: 0.8009023666381836:  39%|███▉      | 2521/6470 [31:34<46:33,  1.41it/s]
2|2522|Loss: 0.7572782039642334:  39%|███▉      | 2521/6470 [31:34<46:33,  1.41it/s]
2|2522|Loss: 0.7572782039642334:  39%|███▉      | 2522/6470 [31:34<45:04,  1.46it/s]
2|2523|Loss: 0.7087595462799072:  39%|███▉      | 2522/6470 [31:35<45:04,  1.46it/s]
2|2523|Loss: 0.7087595462799072:  39%|███▉      | 2523/6470 [31:35<49:21,  1.33it/s]
2|2524|Loss: 0.7118059992790222:  39%|███▉      | 2523/6470 [31:36<49:21,  1.33it/s]
2|2524|Loss: 0.7118059992790222:  39%|███▉      | 2524/6470 [31:36<49:11,  1.34it/s]
2|2525|Loss: 0.9343445301055908:  39%|███▉      | 2524/6470 [31:36<49:11,  1.34it/s]
2|2525|Loss: 0.9343445301055908:  39%|███▉      | 2525/6470 [31:37<48:16,  1.36it/s]
2|2526|Loss: 0.8250202536582947:  39%|███▉      | 2525/6470 [31:37<48:16,  1.36it/s]
2|2526|Loss: 0.8250202536582947:  39%|███▉      | 2526/6470 [31:37<45:05,  1.46it/s]
2|2527|Loss: 1.0098730325698853:  39%|███▉      | 2526/6470 [31:38<45:05,  1.46it/s]
2|2527|Loss: 1.0098730325698853:  39%|███▉      | 2527/6470 [31:38<47:17,  1.39it/s]
2|2528|Loss: 0.845383882522583:  39%|███▉      | 2527/6470 [31:38<47:17,  1.39it/s] 
2|2528|Loss: 0.845383882522583:  39%|███▉      | 2528/6470 [31:39<47:35,  1.38it/s]
2|2529|Loss: 0.793017566204071:  39%|███▉      | 2528/6470 [31:39<47:35,  1.38it/s]
2|2529|Loss: 0.793017566204071:  39%|███▉      | 2529/6470 [31:39<47:19,  1.39it/s]
2|2530|Loss: 1.0088378190994263:  39%|███▉      | 2529/6470 [31:40<47:19,  1.39it/s]
2|2530|Loss: 1.0088378190994263:  39%|███▉      | 2530/6470 [31:40<49:24,  1.33it/s]
2|2531|Loss: 0.8211593627929688:  39%|███▉      | 2530/6470 [31:41<49:24,  1.33it/s]
2|2531|Loss: 0.8211593627929688:  39%|███▉      | 2531/6470 [31:41<52:21,  1.25it/s]
2|2532|Loss: 0.7387292981147766:  39%|███▉      | 2531/6470 [31:42<52:21,  1.25it/s]
2|2532|Loss: 0.7387292981147766:  39%|███▉      | 2532/6470 [31:42<53:02,  1.24it/s]
2|2533|Loss: 1.0451604127883911:  39%|███▉      | 2532/6470 [31:42<53:02,  1.24it/s]
2|2533|Loss: 1.0451604127883911:  39%|███▉      | 2533/6470 [31:43<51:24,  1.28it/s]
2|2534|Loss: 0.9910885691642761:  39%|███▉      | 2533/6470 [31:43<51:24,  1.28it/s]
2|2534|Loss: 0.9910885691642761:  39%|███▉      | 2534/6470 [31:43<51:41,  1.27it/s]
2|2535|Loss: 1.034262776374817:  39%|███▉      | 2534/6470 [31:44<51:41,  1.27it/s] 
2|2535|Loss: 1.034262776374817:  39%|███▉      | 2535/6470 [31:44<50:06,  1.31it/s]
2|2536|Loss: 0.9438287019729614:  39%|███▉      | 2535/6470 [31:45<50:06,  1.31it/s]
2|2536|Loss: 0.9438287019729614:  39%|███▉      | 2536/6470 [31:45<45:15,  1.45it/s]
2|2537|Loss: 0.7772810459136963:  39%|███▉      | 2536/6470 [31:45<45:15,  1.45it/s]
2|2537|Loss: 0.7772810459136963:  39%|███▉      | 2537/6470 [31:46<54:40,  1.20it/s]
2|2538|Loss: 0.7160312533378601:  39%|███▉      | 2537/6470 [31:46<54:40,  1.20it/s]
2|2538|Loss: 0.7160312533378601:  39%|███▉      | 2538/6470 [31:47<55:01,  1.19it/s]
2|2539|Loss: 0.8907765746116638:  39%|███▉      | 2538/6470 [31:47<55:01,  1.19it/s]
2|2539|Loss: 0.8907765746116638:  39%|███▉      | 2539/6470 [31:48<55:40,  1.18it/s]
2|2540|Loss: 0.7678984999656677:  39%|███▉      | 2539/6470 [31:48<55:40,  1.18it/s]
2|2540|Loss: 0.7678984999656677:  39%|███▉      | 2540/6470 [31:48<53:31,  1.22it/s]
2|2541|Loss: 0.663661539554596:  39%|███▉      | 2540/6470 [31:49<53:31,  1.22it/s] 
2|2541|Loss: 0.663661539554596:  39%|███▉      | 2541/6470 [31:49<54:27,  1.20it/s]
2|2542|Loss: 0.8736810684204102:  39%|███▉      | 2541/6470 [31:50<54:27,  1.20it/s]
2|2542|Loss: 0.8736810684204102:  39%|███▉      | 2542/6470 [31:50<52:40,  1.24it/s]
2|2543|Loss: 0.7616586685180664:  39%|███▉      | 2542/6470 [31:50<52:40,  1.24it/s]
2|2543|Loss: 0.7616586685180664:  39%|███▉      | 2543/6470 [31:51<49:21,  1.33it/s]
2|2544|Loss: 0.7555721402168274:  39%|███▉      | 2543/6470 [31:51<49:21,  1.33it/s]
2|2544|Loss: 0.7555721402168274:  39%|███▉      | 2544/6470 [31:51<47:57,  1.36it/s]
2|2545|Loss: 0.8029786348342896:  39%|███▉      | 2544/6470 [31:52<47:57,  1.36it/s]
2|2545|Loss: 0.8029786348342896:  39%|███▉      | 2545/6470 [31:52<47:31,  1.38it/s]
2|2546|Loss: 0.9405941963195801:  39%|███▉      | 2545/6470 [31:52<47:31,  1.38it/s]
2|2546|Loss: 0.9405941963195801:  39%|███▉      | 2546/6470 [31:53<48:39,  1.34it/s]
2|2547|Loss: 0.9233996868133545:  39%|███▉      | 2546/6470 [31:53<48:39,  1.34it/s]
2|2547|Loss: 0.9233996868133545:  39%|███▉      | 2547/6470 [31:53<46:55,  1.39it/s]
2|2548|Loss: 0.826254665851593:  39%|███▉      | 2547/6470 [31:54<46:55,  1.39it/s] 
2|2548|Loss: 0.826254665851593:  39%|███▉      | 2548/6470 [31:54<45:38,  1.43it/s]
2|2549|Loss: 0.8090389966964722:  39%|███▉      | 2548/6470 [31:54<45:38,  1.43it/s]
2|2549|Loss: 0.8090389966964722:  39%|███▉      | 2549/6470 [31:55<46:37,  1.40it/s]
2|2550|Loss: 0.8430535197257996:  39%|███▉      | 2549/6470 [31:55<46:37,  1.40it/s]
2|2550|Loss: 0.8430535197257996:  39%|███▉      | 2550/6470 [31:55<45:53,  1.42it/s]
2|2551|Loss: 0.8603782653808594:  39%|███▉      | 2550/6470 [31:56<45:53,  1.42it/s]
2|2551|Loss: 0.8603782653808594:  39%|███▉      | 2551/6470 [31:56<47:24,  1.38it/s]
2|2552|Loss: 0.675162136554718:  39%|███▉      | 2551/6470 [31:57<47:24,  1.38it/s] 
2|2552|Loss: 0.675162136554718:  39%|███▉      | 2552/6470 [31:57<45:41,  1.43it/s]
2|2553|Loss: 0.7908743023872375:  39%|███▉      | 2552/6470 [31:57<45:41,  1.43it/s]
2|2553|Loss: 0.7908743023872375:  39%|███▉      | 2553/6470 [31:58<44:32,  1.47it/s]
2|2554|Loss: 0.7555786967277527:  39%|███▉      | 2553/6470 [31:58<44:32,  1.47it/s]
2|2554|Loss: 0.7555786967277527:  39%|███▉      | 2554/6470 [31:58<44:03,  1.48it/s]
2|2555|Loss: 0.807502031326294:  39%|███▉      | 2554/6470 [31:59<44:03,  1.48it/s] 
2|2555|Loss: 0.807502031326294:  39%|███▉      | 2555/6470 [31:59<43:45,  1.49it/s]
2|2556|Loss: 0.8626778721809387:  39%|███▉      | 2555/6470 [31:59<43:45,  1.49it/s]
2|2556|Loss: 0.8626778721809387:  40%|███▉      | 2556/6470 [31:59<41:42,  1.56it/s]
2|2557|Loss: 0.7860893607139587:  40%|███▉      | 2556/6470 [32:00<41:42,  1.56it/s]
2|2557|Loss: 0.7860893607139587:  40%|███▉      | 2557/6470 [32:00<44:02,  1.48it/s]
2|2558|Loss: 0.6802350282669067:  40%|███▉      | 2557/6470 [32:01<44:02,  1.48it/s]
2|2558|Loss: 0.6802350282669067:  40%|███▉      | 2558/6470 [32:01<46:30,  1.40it/s]
2|2559|Loss: 0.8260001540184021:  40%|███▉      | 2558/6470 [32:01<46:30,  1.40it/s]
2|2559|Loss: 0.8260001540184021:  40%|███▉      | 2559/6470 [32:02<46:12,  1.41it/s]
2|2560|Loss: 0.6916517615318298:  40%|███▉      | 2559/6470 [32:02<46:12,  1.41it/s]
2|2560|Loss: 0.6916517615318298:  40%|███▉      | 2560/6470 [32:03<48:19,  1.35it/s]
2|2561|Loss: 0.8090490102767944:  40%|███▉      | 2560/6470 [32:03<48:19,  1.35it/s]
2|2561|Loss: 0.8090490102767944:  40%|███▉      | 2561/6470 [32:03<47:39,  1.37it/s]
2|2562|Loss: 0.8310542106628418:  40%|███▉      | 2561/6470 [32:04<47:39,  1.37it/s]
2|2562|Loss: 0.8310542106628418:  40%|███▉      | 2562/6470 [32:04<48:22,  1.35it/s]
2|2563|Loss: 0.7099248766899109:  40%|███▉      | 2562/6470 [32:04<48:22,  1.35it/s]
2|2563|Loss: 0.7099248766899109:  40%|███▉      | 2563/6470 [32:05<47:55,  1.36it/s]
2|2564|Loss: 0.7847817540168762:  40%|███▉      | 2563/6470 [32:05<47:55,  1.36it/s]
2|2564|Loss: 0.7847817540168762:  40%|███▉      | 2564/6470 [32:05<44:44,  1.45it/s]
2|2565|Loss: 0.8613244295120239:  40%|███▉      | 2564/6470 [32:06<44:44,  1.45it/s]
2|2565|Loss: 0.8613244295120239:  40%|███▉      | 2565/6470 [32:06<39:48,  1.63it/s]
2|2566|Loss: 0.8388901948928833:  40%|███▉      | 2565/6470 [32:06<39:48,  1.63it/s]
2|2566|Loss: 0.8388901948928833:  40%|███▉      | 2566/6470 [32:07<44:49,  1.45it/s]
2|2567|Loss: 0.7793307304382324:  40%|███▉      | 2566/6470 [32:07<44:49,  1.45it/s]
2|2567|Loss: 0.7793307304382324:  40%|███▉      | 2567/6470 [32:08<49:42,  1.31it/s]
2|2568|Loss: 0.877815306186676:  40%|███▉      | 2567/6470 [32:08<49:42,  1.31it/s] 
2|2568|Loss: 0.877815306186676:  40%|███▉      | 2568/6470 [32:08<50:46,  1.28it/s]
2|2569|Loss: 0.7923520803451538:  40%|███▉      | 2568/6470 [32:09<50:46,  1.28it/s]
2|2569|Loss: 0.7923520803451538:  40%|███▉      | 2569/6470 [32:09<47:03,  1.38it/s]
2|2570|Loss: 0.8192911148071289:  40%|███▉      | 2569/6470 [32:09<47:03,  1.38it/s]
2|2570|Loss: 0.8192911148071289:  40%|███▉      | 2570/6470 [32:10<48:21,  1.34it/s]
2|2571|Loss: 0.8782679438591003:  40%|███▉      | 2570/6470 [32:10<48:21,  1.34it/s]
2|2571|Loss: 0.8782679438591003:  40%|███▉      | 2571/6470 [32:10<47:47,  1.36it/s]
2|2572|Loss: 0.841005802154541:  40%|███▉      | 2571/6470 [32:11<47:47,  1.36it/s] 
2|2572|Loss: 0.841005802154541:  40%|███▉      | 2572/6470 [32:11<48:50,  1.33it/s]
2|2573|Loss: 0.8692091107368469:  40%|███▉      | 2572/6470 [32:12<48:50,  1.33it/s]
2|2573|Loss: 0.8692091107368469:  40%|███▉      | 2573/6470 [32:12<51:16,  1.27it/s]
2|2574|Loss: 0.8507159352302551:  40%|███▉      | 2573/6470 [32:13<51:16,  1.27it/s]
2|2574|Loss: 0.8507159352302551:  40%|███▉      | 2574/6470 [32:13<51:47,  1.25it/s]
2|2575|Loss: 1.0090943574905396:  40%|███▉      | 2574/6470 [32:13<51:47,  1.25it/s]
2|2575|Loss: 1.0090943574905396:  40%|███▉      | 2575/6470 [32:14<53:42,  1.21it/s]
2|2576|Loss: 0.6690576672554016:  40%|███▉      | 2575/6470 [32:14<53:42,  1.21it/s]
2|2576|Loss: 0.6690576672554016:  40%|███▉      | 2576/6470 [32:15<51:58,  1.25it/s]
2|2577|Loss: 1.0024100542068481:  40%|███▉      | 2576/6470 [32:15<51:58,  1.25it/s]
2|2577|Loss: 1.0024100542068481:  40%|███▉      | 2577/6470 [32:15<50:52,  1.28it/s]
2|2578|Loss: 0.7133058309555054:  40%|███▉      | 2577/6470 [32:16<50:52,  1.28it/s]
2|2578|Loss: 0.7133058309555054:  40%|███▉      | 2578/6470 [32:16<51:13,  1.27it/s]
2|2579|Loss: 0.8383733630180359:  40%|███▉      | 2578/6470 [32:17<51:13,  1.27it/s]
2|2579|Loss: 0.8383733630180359:  40%|███▉      | 2579/6470 [32:17<52:50,  1.23it/s]
2|2580|Loss: 1.0155162811279297:  40%|███▉      | 2579/6470 [32:17<52:50,  1.23it/s]
2|2580|Loss: 1.0155162811279297:  40%|███▉      | 2580/6470 [32:18<48:51,  1.33it/s]
2|2581|Loss: 0.809329628944397:  40%|███▉      | 2580/6470 [32:18<48:51,  1.33it/s] 
2|2581|Loss: 0.809329628944397:  40%|███▉      | 2581/6470 [32:18<48:24,  1.34it/s]
2|2582|Loss: 0.7991759777069092:  40%|███▉      | 2581/6470 [32:19<48:24,  1.34it/s]
2|2582|Loss: 0.7991759777069092:  40%|███▉      | 2582/6470 [32:19<50:59,  1.27it/s]
2|2583|Loss: 0.7414984107017517:  40%|███▉      | 2582/6470 [32:20<50:59,  1.27it/s]
2|2583|Loss: 0.7414984107017517:  40%|███▉      | 2583/6470 [32:20<52:05,  1.24it/s]
2|2584|Loss: 0.948566734790802:  40%|███▉      | 2583/6470 [32:21<52:05,  1.24it/s] 
2|2584|Loss: 0.948566734790802:  40%|███▉      | 2584/6470 [32:21<53:11,  1.22it/s]
2|2585|Loss: 0.8962486386299133:  40%|███▉      | 2584/6470 [32:21<53:11,  1.22it/s]
2|2585|Loss: 0.8962486386299133:  40%|███▉      | 2585/6470 [32:22<52:55,  1.22it/s]
2|2586|Loss: 0.9317911267280579:  40%|███▉      | 2585/6470 [32:22<52:55,  1.22it/s]
2|2586|Loss: 0.9317911267280579:  40%|███▉      | 2586/6470 [32:23<55:26,  1.17it/s]
2|2587|Loss: 0.925179123878479:  40%|███▉      | 2586/6470 [32:23<55:26,  1.17it/s] 
2|2587|Loss: 0.925179123878479:  40%|███▉      | 2587/6470 [32:23<51:43,  1.25it/s]
2|2588|Loss: 0.7755634188652039:  40%|███▉      | 2587/6470 [32:24<51:43,  1.25it/s]
2|2588|Loss: 0.7755634188652039:  40%|████      | 2588/6470 [32:24<50:52,  1.27it/s]
2|2589|Loss: 0.8490119576454163:  40%|████      | 2588/6470 [32:25<50:52,  1.27it/s]
2|2589|Loss: 0.8490119576454163:  40%|████      | 2589/6470 [32:25<50:57,  1.27it/s]
2|2590|Loss: 1.0469437837600708:  40%|████      | 2589/6470 [32:25<50:57,  1.27it/s]
2|2590|Loss: 1.0469437837600708:  40%|████      | 2590/6470 [32:26<51:14,  1.26it/s]
2|2591|Loss: 0.8402005434036255:  40%|████      | 2590/6470 [32:26<51:14,  1.26it/s]
2|2591|Loss: 0.8402005434036255:  40%|████      | 2591/6470 [32:26<49:21,  1.31it/s]
2|2592|Loss: 1.0140074491500854:  40%|████      | 2591/6470 [32:27<49:21,  1.31it/s]
2|2592|Loss: 1.0140074491500854:  40%|████      | 2592/6470 [32:27<48:57,  1.32it/s]
2|2593|Loss: 0.950171709060669:  40%|████      | 2592/6470 [32:28<48:57,  1.32it/s] 
2|2593|Loss: 0.950171709060669:  40%|████      | 2593/6470 [32:28<50:31,  1.28it/s]
2|2594|Loss: 0.7433851957321167:  40%|████      | 2593/6470 [32:28<50:31,  1.28it/s]
2|2594|Loss: 0.7433851957321167:  40%|████      | 2594/6470 [32:29<50:39,  1.28it/s]
2|2595|Loss: 0.8970215320587158:  40%|████      | 2594/6470 [32:29<50:39,  1.28it/s]
2|2595|Loss: 0.8970215320587158:  40%|████      | 2595/6470 [32:30<50:04,  1.29it/s]
2|2596|Loss: 0.7878063917160034:  40%|████      | 2595/6470 [32:30<50:04,  1.29it/s]
2|2596|Loss: 0.7878063917160034:  40%|████      | 2596/6470 [32:30<49:33,  1.30it/s]
2|2597|Loss: 0.830086350440979:  40%|████      | 2596/6470 [32:31<49:33,  1.30it/s] 
2|2597|Loss: 0.830086350440979:  40%|████      | 2597/6470 [32:31<49:52,  1.29it/s]
2|2598|Loss: 0.7939133644104004:  40%|████      | 2597/6470 [32:31<49:52,  1.29it/s]
2|2598|Loss: 0.7939133644104004:  40%|████      | 2598/6470 [32:32<48:58,  1.32it/s]
2|2599|Loss: 0.8569689393043518:  40%|████      | 2598/6470 [32:32<48:58,  1.32it/s]
2|2599|Loss: 0.8569689393043518:  40%|████      | 2599/6470 [32:32<47:34,  1.36it/s]
2|2600|Loss: 0.7201077938079834:  40%|████      | 2599/6470 [32:33<47:34,  1.36it/s]
2|2600|Loss: 0.7201077938079834:  40%|████      | 2600/6470 [32:33<47:58,  1.34it/s]
2|2601|Loss: 0.7936331629753113:  40%|████      | 2600/6470 [32:34<47:58,  1.34it/s]
2|2601|Loss: 0.7936331629753113:  40%|████      | 2601/6470 [32:34<49:19,  1.31it/s]
2|2602|Loss: 0.9297714233398438:  40%|████      | 2601/6470 [32:34<49:19,  1.31it/s]
2|2602|Loss: 0.9297714233398438:  40%|████      | 2602/6470 [32:35<48:26,  1.33it/s]
2|2603|Loss: 0.8192902207374573:  40%|████      | 2602/6470 [32:35<48:26,  1.33it/s]
2|2603|Loss: 0.8192902207374573:  40%|████      | 2603/6470 [32:36<50:14,  1.28it/s]
2|2604|Loss: 0.923603355884552:  40%|████      | 2603/6470 [32:36<50:14,  1.28it/s] 
2|2604|Loss: 0.923603355884552:  40%|████      | 2604/6470 [32:37<52:50,  1.22it/s]
2|2605|Loss: 0.8628670573234558:  40%|████      | 2604/6470 [32:37<52:50,  1.22it/s]
2|2605|Loss: 0.8628670573234558:  40%|████      | 2605/6470 [32:37<51:11,  1.26it/s]
2|2606|Loss: 0.883405864238739:  40%|████      | 2605/6470 [32:38<51:11,  1.26it/s] 
2|2606|Loss: 0.883405864238739:  40%|████      | 2606/6470 [32:38<45:51,  1.40it/s]
2|2607|Loss: 0.7763108015060425:  40%|████      | 2606/6470 [32:38<45:51,  1.40it/s]
2|2607|Loss: 0.7763108015060425:  40%|████      | 2607/6470 [32:38<46:02,  1.40it/s]
2|2608|Loss: 0.7484098076820374:  40%|████      | 2607/6470 [32:39<46:02,  1.40it/s]
2|2608|Loss: 0.7484098076820374:  40%|████      | 2608/6470 [32:39<45:27,  1.42it/s]
2|2609|Loss: 0.9109675288200378:  40%|████      | 2608/6470 [32:40<45:27,  1.42it/s]
2|2609|Loss: 0.9109675288200378:  40%|████      | 2609/6470 [32:40<48:41,  1.32it/s]
2|2610|Loss: 0.7767102718353271:  40%|████      | 2609/6470 [32:40<48:41,  1.32it/s]
2|2610|Loss: 0.7767102718353271:  40%|████      | 2610/6470 [32:41<47:59,  1.34it/s]
2|2611|Loss: 0.7686362266540527:  40%|████      | 2610/6470 [32:41<47:59,  1.34it/s]
2|2611|Loss: 0.7686362266540527:  40%|████      | 2611/6470 [32:42<48:19,  1.33it/s]
2|2612|Loss: 0.9864674806594849:  40%|████      | 2611/6470 [32:42<48:19,  1.33it/s]
2|2612|Loss: 0.9864674806594849:  40%|████      | 2612/6470 [32:42<47:45,  1.35it/s]
2|2613|Loss: 0.8329131007194519:  40%|████      | 2612/6470 [32:43<47:45,  1.35it/s]
2|2613|Loss: 0.8329131007194519:  40%|████      | 2613/6470 [32:43<49:07,  1.31it/s]
2|2614|Loss: 0.7996407151222229:  40%|████      | 2613/6470 [32:43<49:07,  1.31it/s]
2|2614|Loss: 0.7996407151222229:  40%|████      | 2614/6470 [32:44<47:09,  1.36it/s]
2|2615|Loss: 0.8843039274215698:  40%|████      | 2614/6470 [32:44<47:09,  1.36it/s]
2|2615|Loss: 0.8843039274215698:  40%|████      | 2615/6470 [32:44<45:26,  1.41it/s]
2|2616|Loss: 0.9448022842407227:  40%|████      | 2615/6470 [32:45<45:26,  1.41it/s]
2|2616|Loss: 0.9448022842407227:  40%|████      | 2616/6470 [32:45<48:52,  1.31it/s]
2|2617|Loss: 0.9431662559509277:  40%|████      | 2616/6470 [32:46<48:52,  1.31it/s]
2|2617|Loss: 0.9431662559509277:  40%|████      | 2617/6470 [32:46<49:50,  1.29it/s]
2|2618|Loss: 0.9345481395721436:  40%|████      | 2617/6470 [32:46<49:50,  1.29it/s]
2|2618|Loss: 0.9345481395721436:  40%|████      | 2618/6470 [32:47<46:22,  1.38it/s]
2|2619|Loss: 0.7586404085159302:  40%|████      | 2618/6470 [32:47<46:22,  1.38it/s]
2|2619|Loss: 0.7586404085159302:  40%|████      | 2619/6470 [32:47<43:56,  1.46it/s]
2|2620|Loss: 0.6583448052406311:  40%|████      | 2619/6470 [32:48<43:56,  1.46it/s]
2|2620|Loss: 0.6583448052406311:  40%|████      | 2620/6470 [32:48<47:37,  1.35it/s]
2|2621|Loss: 0.7564032077789307:  40%|████      | 2620/6470 [32:49<47:37,  1.35it/s]
2|2621|Loss: 0.7564032077789307:  41%|████      | 2621/6470 [32:49<47:47,  1.34it/s]
2|2622|Loss: 0.8146864771842957:  41%|████      | 2621/6470 [32:49<47:47,  1.34it/s]
2|2622|Loss: 0.8146864771842957:  41%|████      | 2622/6470 [32:49<44:29,  1.44it/s]
2|2623|Loss: 0.8173155784606934:  41%|████      | 2622/6470 [32:50<44:29,  1.44it/s]
2|2623|Loss: 0.8173155784606934:  41%|████      | 2623/6470 [32:50<42:01,  1.53it/s]
2|2624|Loss: 0.7478535771369934:  41%|████      | 2623/6470 [32:50<42:01,  1.53it/s]
2|2624|Loss: 0.7478535771369934:  41%|████      | 2624/6470 [32:51<39:56,  1.60it/s]
2|2625|Loss: 0.8329968452453613:  41%|████      | 2624/6470 [32:51<39:56,  1.60it/s]
2|2625|Loss: 0.8329968452453613:  41%|████      | 2625/6470 [32:51<41:44,  1.54it/s]
2|2626|Loss: 0.9023215770721436:  41%|████      | 2625/6470 [32:52<41:44,  1.54it/s]
2|2626|Loss: 0.9023215770721436:  41%|████      | 2626/6470 [32:52<42:31,  1.51it/s]
2|2627|Loss: 0.9986656904220581:  41%|████      | 2626/6470 [32:52<42:31,  1.51it/s]
2|2627|Loss: 0.9986656904220581:  41%|████      | 2627/6470 [32:53<42:39,  1.50it/s]
2|2628|Loss: 0.9455782175064087:  41%|████      | 2627/6470 [32:53<42:39,  1.50it/s]
2|2628|Loss: 0.9455782175064087:  41%|████      | 2628/6470 [32:53<44:34,  1.44it/s]
2|2629|Loss: 0.7955466508865356:  41%|████      | 2628/6470 [32:54<44:34,  1.44it/s]
2|2629|Loss: 0.7955466508865356:  41%|████      | 2629/6470 [32:54<49:00,  1.31it/s]
2|2630|Loss: 0.7505469918251038:  41%|████      | 2629/6470 [32:55<49:00,  1.31it/s]
2|2630|Loss: 0.7505469918251038:  41%|████      | 2630/6470 [32:55<48:19,  1.32it/s]
2|2631|Loss: 0.8477030396461487:  41%|████      | 2630/6470 [32:55<48:19,  1.32it/s]
2|2631|Loss: 0.8477030396461487:  41%|████      | 2631/6470 [32:56<46:49,  1.37it/s]
2|2632|Loss: 0.7735233306884766:  41%|████      | 2631/6470 [32:56<46:49,  1.37it/s]
2|2632|Loss: 0.7735233306884766:  41%|████      | 2632/6470 [32:57<47:03,  1.36it/s]
2|2633|Loss: 0.9045653343200684:  41%|████      | 2632/6470 [32:57<47:03,  1.36it/s]
2|2633|Loss: 0.9045653343200684:  41%|████      | 2633/6470 [32:57<48:01,  1.33it/s]
2|2634|Loss: 0.7424667477607727:  41%|████      | 2633/6470 [32:58<48:01,  1.33it/s]
2|2634|Loss: 0.7424667477607727:  41%|████      | 2634/6470 [32:58<49:52,  1.28it/s]
2|2635|Loss: 0.8269736766815186:  41%|████      | 2634/6470 [32:59<49:52,  1.28it/s]
2|2635|Loss: 0.8269736766815186:  41%|████      | 2635/6470 [32:59<50:34,  1.26it/s]
2|2636|Loss: 0.8610004782676697:  41%|████      | 2635/6470 [32:59<50:34,  1.26it/s]
2|2636|Loss: 0.8610004782676697:  41%|████      | 2636/6470 [33:00<46:26,  1.38it/s]
2|2637|Loss: 0.8842964172363281:  41%|████      | 2636/6470 [33:00<46:26,  1.38it/s]
2|2637|Loss: 0.8842964172363281:  41%|████      | 2637/6470 [33:00<45:53,  1.39it/s]
2|2638|Loss: 0.9802401065826416:  41%|████      | 2637/6470 [33:01<45:53,  1.39it/s]
2|2638|Loss: 0.9802401065826416:  41%|████      | 2638/6470 [33:01<45:03,  1.42it/s]
2|2639|Loss: 0.846581220626831:  41%|████      | 2638/6470 [33:01<45:03,  1.42it/s] 
2|2639|Loss: 0.846581220626831:  41%|████      | 2639/6470 [33:02<45:47,  1.39it/s]
2|2640|Loss: 0.8531191349029541:  41%|████      | 2639/6470 [33:02<45:47,  1.39it/s]
2|2640|Loss: 0.8531191349029541:  41%|████      | 2640/6470 [33:02<43:58,  1.45it/s]
2|2641|Loss: 0.9270364046096802:  41%|████      | 2640/6470 [33:03<43:58,  1.45it/s]
2|2641|Loss: 0.9270364046096802:  41%|████      | 2641/6470 [33:03<47:15,  1.35it/s]
2|2642|Loss: 0.9769530296325684:  41%|████      | 2641/6470 [33:04<47:15,  1.35it/s]
2|2642|Loss: 0.9769530296325684:  41%|████      | 2642/6470 [33:04<47:10,  1.35it/s]
2|2643|Loss: 0.8584297299385071:  41%|████      | 2642/6470 [33:04<47:10,  1.35it/s]
2|2643|Loss: 0.8584297299385071:  41%|████      | 2643/6470 [33:05<46:18,  1.38it/s]
2|2644|Loss: 0.7001561522483826:  41%|████      | 2643/6470 [33:05<46:18,  1.38it/s]
2|2644|Loss: 0.7001561522483826:  41%|████      | 2644/6470 [33:05<42:18,  1.51it/s]
2|2645|Loss: 0.8129216432571411:  41%|████      | 2644/6470 [33:06<42:18,  1.51it/s]
2|2645|Loss: 0.8129216432571411:  41%|████      | 2645/6470 [33:06<45:08,  1.41it/s]
2|2646|Loss: 0.6702628135681152:  41%|████      | 2645/6470 [33:06<45:08,  1.41it/s]
2|2646|Loss: 0.6702628135681152:  41%|████      | 2646/6470 [33:07<45:05,  1.41it/s]
2|2647|Loss: 0.9385280013084412:  41%|████      | 2646/6470 [33:07<45:05,  1.41it/s]
2|2647|Loss: 0.9385280013084412:  41%|████      | 2647/6470 [33:07<44:53,  1.42it/s]
2|2648|Loss: 0.7798981666564941:  41%|████      | 2647/6470 [33:08<44:53,  1.42it/s]
2|2648|Loss: 0.7798981666564941:  41%|████      | 2648/6470 [33:08<45:47,  1.39it/s]
2|2649|Loss: 0.8401638269424438:  41%|████      | 2648/6470 [33:08<45:47,  1.39it/s]
2|2649|Loss: 0.8401638269424438:  41%|████      | 2649/6470 [33:09<47:03,  1.35it/s]
2|2650|Loss: 0.6912139058113098:  41%|████      | 2649/6470 [33:09<47:03,  1.35it/s]
2|2650|Loss: 0.6912139058113098:  41%|████      | 2650/6470 [33:10<45:23,  1.40it/s]
2|2651|Loss: 0.7841721773147583:  41%|████      | 2650/6470 [33:10<45:23,  1.40it/s]
2|2651|Loss: 0.7841721773147583:  41%|████      | 2651/6470 [33:10<43:42,  1.46it/s]
2|2652|Loss: 0.7841987609863281:  41%|████      | 2651/6470 [33:11<43:42,  1.46it/s]
2|2652|Loss: 0.7841987609863281:  41%|████      | 2652/6470 [33:11<43:12,  1.47it/s]
2|2653|Loss: 0.6866118907928467:  41%|████      | 2652/6470 [33:11<43:12,  1.47it/s]
2|2653|Loss: 0.6866118907928467:  41%|████      | 2653/6470 [33:11<43:08,  1.47it/s]
2|2654|Loss: 0.9165433049201965:  41%|████      | 2653/6470 [33:12<43:08,  1.47it/s]
2|2654|Loss: 0.9165433049201965:  41%|████      | 2654/6470 [33:12<45:23,  1.40it/s]
2|2655|Loss: 0.8573569059371948:  41%|████      | 2654/6470 [33:13<45:23,  1.40it/s]
2|2655|Loss: 0.8573569059371948:  41%|████      | 2655/6470 [33:13<45:40,  1.39it/s]
2|2656|Loss: 0.897835373878479:  41%|████      | 2655/6470 [33:13<45:40,  1.39it/s] 
2|2656|Loss: 0.897835373878479:  41%|████      | 2656/6470 [33:14<48:27,  1.31it/s]
2|2657|Loss: 0.7487014532089233:  41%|████      | 2656/6470 [33:14<48:27,  1.31it/s]
2|2657|Loss: 0.7487014532089233:  41%|████      | 2657/6470 [33:15<49:32,  1.28it/s]
2|2658|Loss: 0.9929226040840149:  41%|████      | 2657/6470 [33:15<49:32,  1.28it/s]
2|2658|Loss: 0.9929226040840149:  41%|████      | 2658/6470 [33:15<50:15,  1.26it/s]
2|2659|Loss: 0.7897927165031433:  41%|████      | 2658/6470 [33:16<50:15,  1.26it/s]
2|2659|Loss: 0.7897927165031433:  41%|████      | 2659/6470 [33:16<48:32,  1.31it/s]
2|2660|Loss: 0.6921161413192749:  41%|████      | 2659/6470 [33:17<48:32,  1.31it/s]
2|2660|Loss: 0.6921161413192749:  41%|████      | 2660/6470 [33:17<46:40,  1.36it/s]
2|2661|Loss: 0.5952879786491394:  41%|████      | 2660/6470 [33:17<46:40,  1.36it/s]
2|2661|Loss: 0.5952879786491394:  41%|████      | 2661/6470 [33:18<52:48,  1.20it/s]
2|2662|Loss: 0.9041420221328735:  41%|████      | 2661/6470 [33:18<52:48,  1.20it/s]
2|2662|Loss: 0.9041420221328735:  41%|████      | 2662/6470 [33:19<53:38,  1.18it/s]
2|2663|Loss: 0.7314499020576477:  41%|████      | 2662/6470 [33:19<53:38,  1.18it/s]
2|2663|Loss: 0.7314499020576477:  41%|████      | 2663/6470 [33:20<52:54,  1.20it/s]
2|2664|Loss: 0.6805112361907959:  41%|████      | 2663/6470 [33:20<52:54,  1.20it/s]
2|2664|Loss: 0.6805112361907959:  41%|████      | 2664/6470 [33:21<54:23,  1.17it/s]
2|2665|Loss: 0.7385293245315552:  41%|████      | 2664/6470 [33:21<54:23,  1.17it/s]
2|2665|Loss: 0.7385293245315552:  41%|████      | 2665/6470 [33:21<54:10,  1.17it/s]
2|2666|Loss: 0.9261183142662048:  41%|████      | 2665/6470 [33:22<54:10,  1.17it/s]
2|2666|Loss: 0.9261183142662048:  41%|████      | 2666/6470 [33:22<50:11,  1.26it/s]
2|2667|Loss: 0.7389162182807922:  41%|████      | 2666/6470 [33:22<50:11,  1.26it/s]
2|2667|Loss: 0.7389162182807922:  41%|████      | 2667/6470 [33:23<46:39,  1.36it/s]
2|2668|Loss: 0.846364438533783:  41%|████      | 2667/6470 [33:23<46:39,  1.36it/s] 
2|2668|Loss: 0.846364438533783:  41%|████      | 2668/6470 [33:23<45:16,  1.40it/s]
2|2669|Loss: 0.7912585139274597:  41%|████      | 2668/6470 [33:24<45:16,  1.40it/s]
2|2669|Loss: 0.7912585139274597:  41%|████▏     | 2669/6470 [33:24<49:02,  1.29it/s]
2|2670|Loss: 0.7816913723945618:  41%|████▏     | 2669/6470 [33:25<49:02,  1.29it/s]
2|2670|Loss: 0.7816913723945618:  41%|████▏     | 2670/6470 [33:25<50:54,  1.24it/s]
2|2671|Loss: 0.8826116323471069:  41%|████▏     | 2670/6470 [33:26<50:54,  1.24it/s]
2|2671|Loss: 0.8826116323471069:  41%|████▏     | 2671/6470 [33:26<53:00,  1.19it/s]
2|2672|Loss: 0.9453744292259216:  41%|████▏     | 2671/6470 [33:26<53:00,  1.19it/s]
2|2672|Loss: 0.9453744292259216:  41%|████▏     | 2672/6470 [33:27<50:49,  1.25it/s]
2|2673|Loss: 0.9178757071495056:  41%|████▏     | 2672/6470 [33:27<50:49,  1.25it/s]
2|2673|Loss: 0.9178757071495056:  41%|████▏     | 2673/6470 [33:27<50:27,  1.25it/s]
2|2674|Loss: 0.9534931182861328:  41%|████▏     | 2673/6470 [33:28<50:27,  1.25it/s]
2|2674|Loss: 0.9534931182861328:  41%|████▏     | 2674/6470 [33:28<44:24,  1.42it/s]
2|2675|Loss: 0.866820752620697:  41%|████▏     | 2674/6470 [33:28<44:24,  1.42it/s] 
2|2675|Loss: 0.866820752620697:  41%|████▏     | 2675/6470 [33:29<43:47,  1.44it/s]
2|2676|Loss: 0.9324188828468323:  41%|████▏     | 2675/6470 [33:29<43:47,  1.44it/s]
2|2676|Loss: 0.9324188828468323:  41%|████▏     | 2676/6470 [33:29<43:49,  1.44it/s]
2|2677|Loss: 0.8439338207244873:  41%|████▏     | 2676/6470 [33:30<43:49,  1.44it/s]
2|2677|Loss: 0.8439338207244873:  41%|████▏     | 2677/6470 [33:30<40:47,  1.55it/s]
2|2678|Loss: 0.7765552997589111:  41%|████▏     | 2677/6470 [33:30<40:47,  1.55it/s]
2|2678|Loss: 0.7765552997589111:  41%|████▏     | 2678/6470 [33:31<44:51,  1.41it/s]
2|2679|Loss: 0.874502956867218:  41%|████▏     | 2678/6470 [33:31<44:51,  1.41it/s] 
2|2679|Loss: 0.874502956867218:  41%|████▏     | 2679/6470 [33:32<49:06,  1.29it/s]
2|2680|Loss: 0.8474643230438232:  41%|████▏     | 2679/6470 [33:32<49:06,  1.29it/s]
2|2680|Loss: 0.8474643230438232:  41%|████▏     | 2680/6470 [33:33<51:41,  1.22it/s]
2|2681|Loss: 1.0027647018432617:  41%|████▏     | 2680/6470 [33:33<51:41,  1.22it/s]
2|2681|Loss: 1.0027647018432617:  41%|████▏     | 2681/6470 [33:33<52:08,  1.21it/s]
2|2682|Loss: 0.870256245136261:  41%|████▏     | 2681/6470 [33:34<52:08,  1.21it/s] 
2|2682|Loss: 0.870256245136261:  41%|████▏     | 2682/6470 [33:34<51:24,  1.23it/s]
2|2683|Loss: 0.8693328499794006:  41%|████▏     | 2682/6470 [33:35<51:24,  1.23it/s]
2|2683|Loss: 0.8693328499794006:  41%|████▏     | 2683/6470 [33:35<48:58,  1.29it/s]
2|2684|Loss: 0.8692505955696106:  41%|████▏     | 2683/6470 [33:35<48:58,  1.29it/s]
2|2684|Loss: 0.8692505955696106:  41%|████▏     | 2684/6470 [33:36<48:27,  1.30it/s]
2|2685|Loss: 0.7800098657608032:  41%|████▏     | 2684/6470 [33:36<48:27,  1.30it/s]
2|2685|Loss: 0.7800098657608032:  41%|████▏     | 2685/6470 [33:36<45:15,  1.39it/s]
2|2686|Loss: 0.7995501756668091:  41%|████▏     | 2685/6470 [33:37<45:15,  1.39it/s]
2|2686|Loss: 0.7995501756668091:  42%|████▏     | 2686/6470 [33:37<42:31,  1.48it/s]
2|2687|Loss: 0.8529927730560303:  42%|████▏     | 2686/6470 [33:37<42:31,  1.48it/s]
2|2687|Loss: 0.8529927730560303:  42%|████▏     | 2687/6470 [33:38<43:26,  1.45it/s]
2|2688|Loss: 0.8085936903953552:  42%|████▏     | 2687/6470 [33:38<43:26,  1.45it/s]
2|2688|Loss: 0.8085936903953552:  42%|████▏     | 2688/6470 [33:38<46:28,  1.36it/s]
2|2689|Loss: 0.8737971186637878:  42%|████▏     | 2688/6470 [33:39<46:28,  1.36it/s]
2|2689|Loss: 0.8737971186637878:  42%|████▏     | 2689/6470 [33:39<47:41,  1.32it/s]
2|2690|Loss: 0.9565926194190979:  42%|████▏     | 2689/6470 [33:40<47:41,  1.32it/s]
2|2690|Loss: 0.9565926194190979:  42%|████▏     | 2690/6470 [33:40<48:41,  1.29it/s]
2|2691|Loss: 0.8058895468711853:  42%|████▏     | 2690/6470 [33:40<48:41,  1.29it/s]
2|2691|Loss: 0.8058895468711853:  42%|████▏     | 2691/6470 [33:41<47:40,  1.32it/s]
2|2692|Loss: 0.941744863986969:  42%|████▏     | 2691/6470 [33:41<47:40,  1.32it/s] 
2|2692|Loss: 0.941744863986969:  42%|████▏     | 2692/6470 [33:42<49:12,  1.28it/s]
2|2693|Loss: 0.8858334422111511:  42%|████▏     | 2692/6470 [33:42<49:12,  1.28it/s]
2|2693|Loss: 0.8858334422111511:  42%|████▏     | 2693/6470 [33:42<45:08,  1.39it/s]
2|2694|Loss: 0.7942675352096558:  42%|████▏     | 2693/6470 [33:42<45:08,  1.39it/s]
2|2694|Loss: 0.7942675352096558:  42%|████▏     | 2694/6470 [33:43<42:22,  1.49it/s]
2|2695|Loss: 0.847804844379425:  42%|████▏     | 2694/6470 [33:43<42:22,  1.49it/s] 
2|2695|Loss: 0.847804844379425:  42%|████▏     | 2695/6470 [33:43<43:27,  1.45it/s]
2|2696|Loss: 0.7983144521713257:  42%|████▏     | 2695/6470 [33:44<43:27,  1.45it/s]
2|2696|Loss: 0.7983144521713257:  42%|████▏     | 2696/6470 [33:44<44:52,  1.40it/s]
2|2697|Loss: 0.9364535808563232:  42%|████▏     | 2696/6470 [33:45<44:52,  1.40it/s]
2|2697|Loss: 0.9364535808563232:  42%|████▏     | 2697/6470 [33:45<46:43,  1.35it/s]
2|2698|Loss: 0.6089819669723511:  42%|████▏     | 2697/6470 [33:45<46:43,  1.35it/s]
2|2698|Loss: 0.6089819669723511:  42%|████▏     | 2698/6470 [33:46<48:13,  1.30it/s]
2|2699|Loss: 0.7944111227989197:  42%|████▏     | 2698/6470 [33:46<48:13,  1.30it/s]
2|2699|Loss: 0.7944111227989197:  42%|████▏     | 2699/6470 [33:47<48:45,  1.29it/s]
2|2700|Loss: 0.7186250686645508:  42%|████▏     | 2699/6470 [33:47<48:45,  1.29it/s]
2|2700|Loss: 0.7186250686645508:  42%|████▏     | 2700/6470 [33:47<47:49,  1.31it/s]
2|2701|Loss: 0.9586020708084106:  42%|████▏     | 2700/6470 [33:48<47:49,  1.31it/s]
2|2701|Loss: 0.9586020708084106:  42%|████▏     | 2701/6470 [33:48<46:37,  1.35it/s]
2|2702|Loss: 0.8551023006439209:  42%|████▏     | 2701/6470 [33:48<46:37,  1.35it/s]
2|2702|Loss: 0.8551023006439209:  42%|████▏     | 2702/6470 [33:49<46:53,  1.34it/s]
2|2703|Loss: 0.7795141935348511:  42%|████▏     | 2702/6470 [33:49<46:53,  1.34it/s]
2|2703|Loss: 0.7795141935348511:  42%|████▏     | 2703/6470 [33:49<41:22,  1.52it/s]
2|2704|Loss: 0.6933826804161072:  42%|████▏     | 2703/6470 [33:50<41:22,  1.52it/s]
2|2704|Loss: 0.6933826804161072:  42%|████▏     | 2704/6470 [33:50<48:04,  1.31it/s]
2|2705|Loss: 0.7583995461463928:  42%|████▏     | 2704/6470 [33:51<48:04,  1.31it/s]
2|2705|Loss: 0.7583995461463928:  42%|████▏     | 2705/6470 [33:51<49:36,  1.26it/s]
2|2706|Loss: 0.8786928057670593:  42%|████▏     | 2705/6470 [33:51<49:36,  1.26it/s]
2|2706|Loss: 0.8786928057670593:  42%|████▏     | 2706/6470 [33:52<46:11,  1.36it/s]
2|2707|Loss: 0.8766922354698181:  42%|████▏     | 2706/6470 [33:52<46:11,  1.36it/s]
2|2707|Loss: 0.8766922354698181:  42%|████▏     | 2707/6470 [33:53<48:16,  1.30it/s]
2|2708|Loss: 0.8921326994895935:  42%|████▏     | 2707/6470 [33:53<48:16,  1.30it/s]
2|2708|Loss: 0.8921326994895935:  42%|████▏     | 2708/6470 [33:54<51:30,  1.22it/s]
2|2709|Loss: 0.8873884677886963:  42%|████▏     | 2708/6470 [33:54<51:30,  1.22it/s]
2|2709|Loss: 0.8873884677886963:  42%|████▏     | 2709/6470 [33:54<50:16,  1.25it/s]
2|2710|Loss: 1.0111479759216309:  42%|████▏     | 2709/6470 [33:55<50:16,  1.25it/s]
2|2710|Loss: 1.0111479759216309:  42%|████▏     | 2710/6470 [33:55<45:50,  1.37it/s]
2|2711|Loss: 1.0933148860931396:  42%|████▏     | 2710/6470 [33:55<45:50,  1.37it/s]
2|2711|Loss: 1.0933148860931396:  42%|████▏     | 2711/6470 [33:56<45:46,  1.37it/s]
2|2712|Loss: 0.8971777558326721:  42%|████▏     | 2711/6470 [33:56<45:46,  1.37it/s]
2|2712|Loss: 0.8971777558326721:  42%|████▏     | 2712/6470 [33:56<44:25,  1.41it/s]
2|2713|Loss: 0.7596485614776611:  42%|████▏     | 2712/6470 [33:57<44:25,  1.41it/s]
2|2713|Loss: 0.7596485614776611:  42%|████▏     | 2713/6470 [33:57<43:48,  1.43it/s]
2|2714|Loss: 0.8033268451690674:  42%|████▏     | 2713/6470 [33:57<43:48,  1.43it/s]
2|2714|Loss: 0.8033268451690674:  42%|████▏     | 2714/6470 [33:58<44:43,  1.40it/s]
2|2715|Loss: 0.9544413089752197:  42%|████▏     | 2714/6470 [33:58<44:43,  1.40it/s]
2|2715|Loss: 0.9544413089752197:  42%|████▏     | 2715/6470 [33:58<46:11,  1.36it/s]
2|2716|Loss: 0.7294102311134338:  42%|████▏     | 2715/6470 [33:59<46:11,  1.36it/s]
2|2716|Loss: 0.7294102311134338:  42%|████▏     | 2716/6470 [33:59<49:14,  1.27it/s]
2|2717|Loss: 0.9701139330863953:  42%|████▏     | 2716/6470 [34:00<49:14,  1.27it/s]
2|2717|Loss: 0.9701139330863953:  42%|████▏     | 2717/6470 [34:00<46:14,  1.35it/s]
2|2718|Loss: 0.8686715960502625:  42%|████▏     | 2717/6470 [34:00<46:14,  1.35it/s]
2|2718|Loss: 0.8686715960502625:  42%|████▏     | 2718/6470 [34:01<47:30,  1.32it/s]
2|2719|Loss: 0.8246300220489502:  42%|████▏     | 2718/6470 [34:01<47:30,  1.32it/s]
2|2719|Loss: 0.8246300220489502:  42%|████▏     | 2719/6470 [34:01<45:53,  1.36it/s]
2|2720|Loss: 0.83738112449646:  42%|████▏     | 2719/6470 [34:02<45:53,  1.36it/s]  
2|2720|Loss: 0.83738112449646:  42%|████▏     | 2720/6470 [34:02<43:44,  1.43it/s]
2|2721|Loss: 0.8421091437339783:  42%|████▏     | 2720/6470 [34:02<43:44,  1.43it/s]
2|2721|Loss: 0.8421091437339783:  42%|████▏     | 2721/6470 [34:03<45:00,  1.39it/s]
2|2722|Loss: 0.7959225177764893:  42%|████▏     | 2721/6470 [34:03<45:00,  1.39it/s]
2|2722|Loss: 0.7959225177764893:  42%|████▏     | 2722/6470 [34:04<47:43,  1.31it/s]
2|2723|Loss: 0.7944429516792297:  42%|████▏     | 2722/6470 [34:04<47:43,  1.31it/s]
2|2723|Loss: 0.7944429516792297:  42%|████▏     | 2723/6470 [34:05<49:01,  1.27it/s]
2|2724|Loss: 0.9561706185340881:  42%|████▏     | 2723/6470 [34:05<49:01,  1.27it/s]
2|2724|Loss: 0.9561706185340881:  42%|████▏     | 2724/6470 [34:05<48:49,  1.28it/s]
2|2725|Loss: 0.8893594145774841:  42%|████▏     | 2724/6470 [34:06<48:49,  1.28it/s]
2|2725|Loss: 0.8893594145774841:  42%|████▏     | 2725/6470 [34:06<46:17,  1.35it/s]
2|2726|Loss: 0.7319092750549316:  42%|████▏     | 2725/6470 [34:06<46:17,  1.35it/s]
2|2726|Loss: 0.7319092750549316:  42%|████▏     | 2726/6470 [34:07<46:38,  1.34it/s]
2|2727|Loss: 0.8508784174919128:  42%|████▏     | 2726/6470 [34:07<46:38,  1.34it/s]
2|2727|Loss: 0.8508784174919128:  42%|████▏     | 2727/6470 [34:07<45:34,  1.37it/s]
2|2728|Loss: 0.9497177004814148:  42%|████▏     | 2727/6470 [34:08<45:34,  1.37it/s]
2|2728|Loss: 0.9497177004814148:  42%|████▏     | 2728/6470 [34:08<42:20,  1.47it/s]
2|2729|Loss: 0.6337331533432007:  42%|████▏     | 2728/6470 [34:08<42:20,  1.47it/s]
2|2729|Loss: 0.6337331533432007:  42%|████▏     | 2729/6470 [34:09<43:01,  1.45it/s]
2|2730|Loss: 0.748360812664032:  42%|████▏     | 2729/6470 [34:09<43:01,  1.45it/s] 
2|2730|Loss: 0.748360812664032:  42%|████▏     | 2730/6470 [34:09<44:17,  1.41it/s]
2|2731|Loss: 0.9853506088256836:  42%|████▏     | 2730/6470 [34:10<44:17,  1.41it/s]
2|2731|Loss: 0.9853506088256836:  42%|████▏     | 2731/6470 [34:10<47:02,  1.32it/s]
2|2732|Loss: 0.7103806138038635:  42%|████▏     | 2731/6470 [34:11<47:02,  1.32it/s]
2|2732|Loss: 0.7103806138038635:  42%|████▏     | 2732/6470 [34:11<49:15,  1.26it/s]
2|2733|Loss: 0.8022908568382263:  42%|████▏     | 2732/6470 [34:12<49:15,  1.26it/s]
2|2733|Loss: 0.8022908568382263:  42%|████▏     | 2733/6470 [34:12<49:05,  1.27it/s]
2|2734|Loss: 0.7162850499153137:  42%|████▏     | 2733/6470 [34:12<49:05,  1.27it/s]
2|2734|Loss: 0.7162850499153137:  42%|████▏     | 2734/6470 [34:13<52:50,  1.18it/s]
2|2735|Loss: 0.81599360704422:  42%|████▏     | 2734/6470 [34:13<52:50,  1.18it/s]  
2|2735|Loss: 0.81599360704422:  42%|████▏     | 2735/6470 [34:14<51:21,  1.21it/s]
2|2736|Loss: 0.8063848614692688:  42%|████▏     | 2735/6470 [34:14<51:21,  1.21it/s]
2|2736|Loss: 0.8063848614692688:  42%|████▏     | 2736/6470 [34:15<52:06,  1.19it/s]
2|2737|Loss: 0.7864094972610474:  42%|████▏     | 2736/6470 [34:15<52:06,  1.19it/s]
2|2737|Loss: 0.7864094972610474:  42%|████▏     | 2737/6470 [34:15<49:25,  1.26it/s]
2|2738|Loss: 0.8309313654899597:  42%|████▏     | 2737/6470 [34:16<49:25,  1.26it/s]
2|2738|Loss: 0.8309313654899597:  42%|████▏     | 2738/6470 [34:16<48:28,  1.28it/s]
2|2739|Loss: 0.8977664113044739:  42%|████▏     | 2738/6470 [34:16<48:28,  1.28it/s]
2|2739|Loss: 0.8977664113044739:  42%|████▏     | 2739/6470 [34:17<49:07,  1.27it/s]
2|2740|Loss: 0.8892896771430969:  42%|████▏     | 2739/6470 [34:17<49:07,  1.27it/s]
2|2740|Loss: 0.8892896771430969:  42%|████▏     | 2740/6470 [34:18<49:36,  1.25it/s]
2|2741|Loss: 0.8447145819664001:  42%|████▏     | 2740/6470 [34:18<49:36,  1.25it/s]
2|2741|Loss: 0.8447145819664001:  42%|████▏     | 2741/6470 [34:18<47:45,  1.30it/s]
2|2742|Loss: 0.7912328243255615:  42%|████▏     | 2741/6470 [34:19<47:45,  1.30it/s]
2|2742|Loss: 0.7912328243255615:  42%|████▏     | 2742/6470 [34:19<44:53,  1.38it/s]
2|2743|Loss: 0.7254739999771118:  42%|████▏     | 2742/6470 [34:19<44:53,  1.38it/s]
2|2743|Loss: 0.7254739999771118:  42%|████▏     | 2743/6470 [34:20<44:54,  1.38it/s]
2|2744|Loss: 0.8530929684638977:  42%|████▏     | 2743/6470 [34:20<44:54,  1.38it/s]
2|2744|Loss: 0.8530929684638977:  42%|████▏     | 2744/6470 [34:20<44:19,  1.40it/s]
2|2745|Loss: 0.8347623944282532:  42%|████▏     | 2744/6470 [34:21<44:19,  1.40it/s]
2|2745|Loss: 0.8347623944282532:  42%|████▏     | 2745/6470 [34:21<48:28,  1.28it/s]
2|2746|Loss: 0.8190727233886719:  42%|████▏     | 2745/6470 [34:22<48:28,  1.28it/s]
2|2746|Loss: 0.8190727233886719:  42%|████▏     | 2746/6470 [34:22<53:15,  1.17it/s]
2|2747|Loss: 0.8279314041137695:  42%|████▏     | 2746/6470 [34:23<53:15,  1.17it/s]
2|2747|Loss: 0.8279314041137695:  42%|████▏     | 2747/6470 [34:23<52:40,  1.18it/s]
2|2748|Loss: 0.7511749267578125:  42%|████▏     | 2747/6470 [34:24<52:40,  1.18it/s]
2|2748|Loss: 0.7511749267578125:  42%|████▏     | 2748/6470 [34:24<54:27,  1.14it/s]
2|2749|Loss: 0.830064058303833:  42%|████▏     | 2748/6470 [34:25<54:27,  1.14it/s] 
2|2749|Loss: 0.830064058303833:  42%|████▏     | 2749/6470 [34:25<56:19,  1.10it/s]
2|2750|Loss: 1.0274070501327515:  42%|████▏     | 2749/6470 [34:26<56:19,  1.10it/s]
2|2750|Loss: 1.0274070501327515:  43%|████▎     | 2750/6470 [34:26<56:54,  1.09it/s]
2|2751|Loss: 0.7851302623748779:  43%|████▎     | 2750/6470 [34:26<56:54,  1.09it/s]
2|2751|Loss: 0.7851302623748779:  43%|████▎     | 2751/6470 [34:27<53:14,  1.16it/s]
2|2752|Loss: 0.7867007255554199:  43%|████▎     | 2751/6470 [34:27<53:14,  1.16it/s]
2|2752|Loss: 0.7867007255554199:  43%|████▎     | 2752/6470 [34:28<51:55,  1.19it/s]
2|2753|Loss: 0.9093145728111267:  43%|████▎     | 2752/6470 [34:28<51:55,  1.19it/s]
2|2753|Loss: 0.9093145728111267:  43%|████▎     | 2753/6470 [34:28<51:02,  1.21it/s]
2|2754|Loss: 0.8895312547683716:  43%|████▎     | 2753/6470 [34:29<51:02,  1.21it/s]
2|2754|Loss: 0.8895312547683716:  43%|████▎     | 2754/6470 [34:29<46:59,  1.32it/s]
2|2755|Loss: 0.8332981467247009:  43%|████▎     | 2754/6470 [34:29<46:59,  1.32it/s]
2|2755|Loss: 0.8332981467247009:  43%|████▎     | 2755/6470 [34:30<43:39,  1.42it/s]
2|2756|Loss: 0.9056747555732727:  43%|████▎     | 2755/6470 [34:30<43:39,  1.42it/s]
2|2756|Loss: 0.9056747555732727:  43%|████▎     | 2756/6470 [34:30<44:42,  1.38it/s]
2|2757|Loss: 0.9683197140693665:  43%|████▎     | 2756/6470 [34:31<44:42,  1.38it/s]
2|2757|Loss: 0.9683197140693665:  43%|████▎     | 2757/6470 [34:31<45:31,  1.36it/s]
2|2758|Loss: 0.8726411461830139:  43%|████▎     | 2757/6470 [34:32<45:31,  1.36it/s]
2|2758|Loss: 0.8726411461830139:  43%|████▎     | 2758/6470 [34:32<46:00,  1.34it/s]
2|2759|Loss: 0.8900680541992188:  43%|████▎     | 2758/6470 [34:32<46:00,  1.34it/s]
2|2759|Loss: 0.8900680541992188:  43%|████▎     | 2759/6470 [34:33<48:42,  1.27it/s]
2|2760|Loss: 0.7831383347511292:  43%|████▎     | 2759/6470 [34:33<48:42,  1.27it/s]
2|2760|Loss: 0.7831383347511292:  43%|████▎     | 2760/6470 [34:34<51:30,  1.20it/s]
2|2761|Loss: 0.780802309513092:  43%|████▎     | 2760/6470 [34:34<51:30,  1.20it/s] 
2|2761|Loss: 0.780802309513092:  43%|████▎     | 2761/6470 [34:34<50:16,  1.23it/s]
2|2762|Loss: 0.8066139817237854:  43%|████▎     | 2761/6470 [34:35<50:16,  1.23it/s]
2|2762|Loss: 0.8066139817237854:  43%|████▎     | 2762/6470 [34:35<45:23,  1.36it/s]
2|2763|Loss: 0.8961470127105713:  43%|████▎     | 2762/6470 [34:35<45:23,  1.36it/s]
2|2763|Loss: 0.8961470127105713:  43%|████▎     | 2763/6470 [34:36<43:30,  1.42it/s]
2|2764|Loss: 0.916737973690033:  43%|████▎     | 2763/6470 [34:36<43:30,  1.42it/s] 
2|2764|Loss: 0.916737973690033:  43%|████▎     | 2764/6470 [34:36<40:10,  1.54it/s]
2|2765|Loss: 0.8054017424583435:  43%|████▎     | 2764/6470 [34:37<40:10,  1.54it/s]
2|2765|Loss: 0.8054017424583435:  43%|████▎     | 2765/6470 [34:37<41:42,  1.48it/s]
2|2766|Loss: 0.9628016948699951:  43%|████▎     | 2765/6470 [34:37<41:42,  1.48it/s]
2|2766|Loss: 0.9628016948699951:  43%|████▎     | 2766/6470 [34:38<43:38,  1.41it/s]
2|2767|Loss: 0.8978666067123413:  43%|████▎     | 2766/6470 [34:38<43:38,  1.41it/s]
2|2767|Loss: 0.8978666067123413:  43%|████▎     | 2767/6470 [34:38<43:44,  1.41it/s]
2|2768|Loss: 0.8748146295547485:  43%|████▎     | 2767/6470 [34:39<43:44,  1.41it/s]
2|2768|Loss: 0.8748146295547485:  43%|████▎     | 2768/6470 [34:39<46:46,  1.32it/s]
2|2769|Loss: 0.7425674200057983:  43%|████▎     | 2768/6470 [34:40<46:46,  1.32it/s]
2|2769|Loss: 0.7425674200057983:  43%|████▎     | 2769/6470 [34:40<49:00,  1.26it/s]
2|2770|Loss: 0.9238841533660889:  43%|████▎     | 2769/6470 [34:41<49:00,  1.26it/s]
2|2770|Loss: 0.9238841533660889:  43%|████▎     | 2770/6470 [34:41<48:49,  1.26it/s]
2|2771|Loss: 1.05918288230896:  43%|████▎     | 2770/6470 [34:41<48:49,  1.26it/s]  
2|2771|Loss: 1.05918288230896:  43%|████▎     | 2771/6470 [34:42<50:15,  1.23it/s]
2|2772|Loss: 0.9319120645523071:  43%|████▎     | 2771/6470 [34:42<50:15,  1.23it/s]
2|2772|Loss: 0.9319120645523071:  43%|████▎     | 2772/6470 [34:43<52:26,  1.18it/s]
2|2773|Loss: 0.9548967480659485:  43%|████▎     | 2772/6470 [34:43<52:26,  1.18it/s]
2|2773|Loss: 0.9548967480659485:  43%|████▎     | 2773/6470 [34:43<48:28,  1.27it/s]
2|2774|Loss: 0.8249436616897583:  43%|████▎     | 2773/6470 [34:44<48:28,  1.27it/s]
2|2774|Loss: 0.8249436616897583:  43%|████▎     | 2774/6470 [34:44<47:59,  1.28it/s]
2|2775|Loss: 0.8605849742889404:  43%|████▎     | 2774/6470 [34:44<47:59,  1.28it/s]
2|2775|Loss: 0.8605849742889404:  43%|████▎     | 2775/6470 [34:45<42:29,  1.45it/s]
2|2776|Loss: 0.6690375804901123:  43%|████▎     | 2775/6470 [34:45<42:29,  1.45it/s]
2|2776|Loss: 0.6690375804901123:  43%|████▎     | 2776/6470 [34:45<43:37,  1.41it/s]
2|2777|Loss: 0.8088735342025757:  43%|████▎     | 2776/6470 [34:46<43:37,  1.41it/s]
2|2777|Loss: 0.8088735342025757:  43%|████▎     | 2777/6470 [34:46<43:39,  1.41it/s]
2|2778|Loss: 0.9028226137161255:  43%|████▎     | 2777/6470 [34:46<43:39,  1.41it/s]
2|2778|Loss: 0.9028226137161255:  43%|████▎     | 2778/6470 [34:47<41:55,  1.47it/s]
2|2779|Loss: 0.8700154423713684:  43%|████▎     | 2778/6470 [34:47<41:55,  1.47it/s]
2|2779|Loss: 0.8700154423713684:  43%|████▎     | 2779/6470 [34:48<44:29,  1.38it/s]
2|2780|Loss: 0.8490594625473022:  43%|████▎     | 2779/6470 [34:48<44:29,  1.38it/s]
2|2780|Loss: 0.8490594625473022:  43%|████▎     | 2780/6470 [34:48<46:22,  1.33it/s]
2|2781|Loss: 0.9317013621330261:  43%|████▎     | 2780/6470 [34:49<46:22,  1.33it/s]
2|2781|Loss: 0.9317013621330261:  43%|████▎     | 2781/6470 [34:49<47:37,  1.29it/s]
2|2782|Loss: 0.8068050146102905:  43%|████▎     | 2781/6470 [34:50<47:37,  1.29it/s]
2|2782|Loss: 0.8068050146102905:  43%|████▎     | 2782/6470 [34:50<48:25,  1.27it/s]
2|2783|Loss: 0.8804598450660706:  43%|████▎     | 2782/6470 [34:50<48:25,  1.27it/s]
2|2783|Loss: 0.8804598450660706:  43%|████▎     | 2783/6470 [34:51<47:15,  1.30it/s]
2|2784|Loss: 0.8480333685874939:  43%|████▎     | 2783/6470 [34:51<47:15,  1.30it/s]
2|2784|Loss: 0.8480333685874939:  43%|████▎     | 2784/6470 [34:51<47:05,  1.30it/s]
2|2785|Loss: 0.8355243802070618:  43%|████▎     | 2784/6470 [34:52<47:05,  1.30it/s]
2|2785|Loss: 0.8355243802070618:  43%|████▎     | 2785/6470 [34:52<49:25,  1.24it/s]
2|2786|Loss: 0.7506354451179504:  43%|████▎     | 2785/6470 [34:53<49:25,  1.24it/s]
2|2786|Loss: 0.7506354451179504:  43%|████▎     | 2786/6470 [34:53<51:24,  1.19it/s]
2|2787|Loss: 0.8190377354621887:  43%|████▎     | 2786/6470 [34:54<51:24,  1.19it/s]
2|2787|Loss: 0.8190377354621887:  43%|████▎     | 2787/6470 [34:54<48:03,  1.28it/s]
2|2788|Loss: 0.8130574226379395:  43%|████▎     | 2787/6470 [34:54<48:03,  1.28it/s]
2|2788|Loss: 0.8130574226379395:  43%|████▎     | 2788/6470 [34:55<47:21,  1.30it/s]
2|2789|Loss: 0.7994791269302368:  43%|████▎     | 2788/6470 [34:55<47:21,  1.30it/s]
2|2789|Loss: 0.7994791269302368:  43%|████▎     | 2789/6470 [34:55<47:36,  1.29it/s]
2|2790|Loss: 0.822456955909729:  43%|████▎     | 2789/6470 [34:56<47:36,  1.29it/s] 
2|2790|Loss: 0.822456955909729:  43%|████▎     | 2790/6470 [34:56<47:34,  1.29it/s]
2|2791|Loss: 0.9194245934486389:  43%|████▎     | 2790/6470 [34:57<47:34,  1.29it/s]
2|2791|Loss: 0.9194245934486389:  43%|████▎     | 2791/6470 [34:57<50:14,  1.22it/s]
2|2792|Loss: 0.9078549742698669:  43%|████▎     | 2791/6470 [34:58<50:14,  1.22it/s]
2|2792|Loss: 0.9078549742698669:  43%|████▎     | 2792/6470 [34:58<49:57,  1.23it/s]
2|2793|Loss: 0.9250771999359131:  43%|████▎     | 2792/6470 [34:58<49:57,  1.23it/s]
2|2793|Loss: 0.9250771999359131:  43%|████▎     | 2793/6470 [34:59<49:04,  1.25it/s]
2|2794|Loss: 0.8856421709060669:  43%|████▎     | 2793/6470 [34:59<49:04,  1.25it/s]
2|2794|Loss: 0.8856421709060669:  43%|████▎     | 2794/6470 [35:00<49:15,  1.24it/s]
2|2795|Loss: 0.8204815983772278:  43%|████▎     | 2794/6470 [35:00<49:15,  1.24it/s]
2|2795|Loss: 0.8204815983772278:  43%|████▎     | 2795/6470 [35:00<50:32,  1.21it/s]
2|2796|Loss: 0.8109888434410095:  43%|████▎     | 2795/6470 [35:01<50:32,  1.21it/s]
2|2796|Loss: 0.8109888434410095:  43%|████▎     | 2796/6470 [35:01<48:46,  1.26it/s]
2|2797|Loss: 0.7002300024032593:  43%|████▎     | 2796/6470 [35:02<48:46,  1.26it/s]
2|2797|Loss: 0.7002300024032593:  43%|████▎     | 2797/6470 [35:02<46:25,  1.32it/s]
2|2798|Loss: 0.8293851613998413:  43%|████▎     | 2797/6470 [35:02<46:25,  1.32it/s]
2|2798|Loss: 0.8293851613998413:  43%|████▎     | 2798/6470 [35:03<45:19,  1.35it/s]
2|2799|Loss: 0.7253962159156799:  43%|████▎     | 2798/6470 [35:03<45:19,  1.35it/s]
2|2799|Loss: 0.7253962159156799:  43%|████▎     | 2799/6470 [35:03<45:36,  1.34it/s]
2|2800|Loss: 0.8196895122528076:  43%|████▎     | 2799/6470 [35:04<45:36,  1.34it/s]
2|2800|Loss: 0.8196895122528076:  43%|████▎     | 2800/6470 [35:04<47:59,  1.27it/s]
2|2801|Loss: 0.7050225734710693:  43%|████▎     | 2800/6470 [35:05<47:59,  1.27it/s]
2|2801|Loss: 0.7050225734710693:  43%|████▎     | 2801/6470 [35:05<43:54,  1.39it/s]
2|2802|Loss: 0.9944414496421814:  43%|████▎     | 2801/6470 [35:05<43:54,  1.39it/s]
2|2802|Loss: 0.9944414496421814:  43%|████▎     | 2802/6470 [35:06<46:01,  1.33it/s]
2|2803|Loss: 0.8236949443817139:  43%|████▎     | 2802/6470 [35:06<46:01,  1.33it/s]
2|2803|Loss: 0.8236949443817139:  43%|████▎     | 2803/6470 [35:06<49:18,  1.24it/s]
2|2804|Loss: 0.8593961000442505:  43%|████▎     | 2803/6470 [35:07<49:18,  1.24it/s]
2|2804|Loss: 0.8593961000442505:  43%|████▎     | 2804/6470 [35:07<47:18,  1.29it/s]
2|2805|Loss: 0.6914647221565247:  43%|████▎     | 2804/6470 [35:08<47:18,  1.29it/s]
2|2805|Loss: 0.6914647221565247:  43%|████▎     | 2805/6470 [35:08<44:26,  1.37it/s]
2|2806|Loss: 0.87709641456604:  43%|████▎     | 2805/6470 [35:08<44:26,  1.37it/s]  
2|2806|Loss: 0.87709641456604:  43%|████▎     | 2806/6470 [35:09<44:32,  1.37it/s]
2|2807|Loss: 0.9132852554321289:  43%|████▎     | 2806/6470 [35:09<44:32,  1.37it/s]
2|2807|Loss: 0.9132852554321289:  43%|████▎     | 2807/6470 [35:09<44:23,  1.38it/s]
2|2808|Loss: 0.823706865310669:  43%|████▎     | 2807/6470 [35:10<44:23,  1.38it/s] 
2|2808|Loss: 0.823706865310669:  43%|████▎     | 2808/6470 [35:10<43:34,  1.40it/s]
2|2809|Loss: 0.7697693705558777:  43%|████▎     | 2808/6470 [35:10<43:34,  1.40it/s]
2|2809|Loss: 0.7697693705558777:  43%|████▎     | 2809/6470 [35:11<46:08,  1.32it/s]
2|2810|Loss: 0.8644096255302429:  43%|████▎     | 2809/6470 [35:11<46:08,  1.32it/s]
2|2810|Loss: 0.8644096255302429:  43%|████▎     | 2810/6470 [35:11<42:40,  1.43it/s]
2|2811|Loss: 0.8149145841598511:  43%|████▎     | 2810/6470 [35:12<42:40,  1.43it/s]
2|2811|Loss: 0.8149145841598511:  43%|████▎     | 2811/6470 [35:12<43:14,  1.41it/s]
2|2812|Loss: 0.8355368375778198:  43%|████▎     | 2811/6470 [35:13<43:14,  1.41it/s]
2|2812|Loss: 0.8355368375778198:  43%|████▎     | 2812/6470 [35:13<48:16,  1.26it/s]
2|2813|Loss: 0.810326874256134:  43%|████▎     | 2812/6470 [35:14<48:16,  1.26it/s] 
2|2813|Loss: 0.810326874256134:  43%|████▎     | 2813/6470 [35:14<49:55,  1.22it/s]
2|2814|Loss: 1.0521079301834106:  43%|████▎     | 2813/6470 [35:15<49:55,  1.22it/s]
2|2814|Loss: 1.0521079301834106:  43%|████▎     | 2814/6470 [35:15<51:51,  1.17it/s]
2|2815|Loss: 0.8390067219734192:  43%|████▎     | 2814/6470 [35:15<51:51,  1.17it/s]
2|2815|Loss: 0.8390067219734192:  44%|████▎     | 2815/6470 [35:16<51:06,  1.19it/s]
2|2816|Loss: 0.7897122502326965:  44%|████▎     | 2815/6470 [35:16<51:06,  1.19it/s]
2|2816|Loss: 0.7897122502326965:  44%|████▎     | 2816/6470 [35:16<48:53,  1.25it/s]
2|2817|Loss: 0.7816152572631836:  44%|████▎     | 2816/6470 [35:17<48:53,  1.25it/s]
2|2817|Loss: 0.7816152572631836:  44%|████▎     | 2817/6470 [35:17<45:26,  1.34it/s]
2|2818|Loss: 0.6841517686843872:  44%|████▎     | 2817/6470 [35:17<45:26,  1.34it/s]
2|2818|Loss: 0.6841517686843872:  44%|████▎     | 2818/6470 [35:18<45:28,  1.34it/s]
2|2819|Loss: 0.8908239006996155:  44%|████▎     | 2818/6470 [35:18<45:28,  1.34it/s]
2|2819|Loss: 0.8908239006996155:  44%|████▎     | 2819/6470 [35:19<47:19,  1.29it/s]
2|2820|Loss: 0.7690418362617493:  44%|████▎     | 2819/6470 [35:19<47:19,  1.29it/s]
2|2820|Loss: 0.7690418362617493:  44%|████▎     | 2820/6470 [35:19<46:46,  1.30it/s]
2|2821|Loss: 1.0195258855819702:  44%|████▎     | 2820/6470 [35:20<46:46,  1.30it/s]
2|2821|Loss: 1.0195258855819702:  44%|████▎     | 2821/6470 [35:20<47:12,  1.29it/s]
2|2822|Loss: 0.7333080172538757:  44%|████▎     | 2821/6470 [35:21<47:12,  1.29it/s]
2|2822|Loss: 0.7333080172538757:  44%|████▎     | 2822/6470 [35:21<46:15,  1.31it/s]
2|2823|Loss: 0.8758575916290283:  44%|████▎     | 2822/6470 [35:21<46:15,  1.31it/s]
2|2823|Loss: 0.8758575916290283:  44%|████▎     | 2823/6470 [35:22<48:25,  1.26it/s]
2|2824|Loss: 1.00226891040802:  44%|████▎     | 2823/6470 [35:22<48:25,  1.26it/s]  
2|2824|Loss: 1.00226891040802:  44%|████▎     | 2824/6470 [35:23<48:33,  1.25it/s]
2|2825|Loss: 0.8856128454208374:  44%|████▎     | 2824/6470 [35:23<48:33,  1.25it/s]
2|2825|Loss: 0.8856128454208374:  44%|████▎     | 2825/6470 [35:23<48:28,  1.25it/s]
2|2826|Loss: 0.9626739621162415:  44%|████▎     | 2825/6470 [35:24<48:28,  1.25it/s]
2|2826|Loss: 0.9626739621162415:  44%|████▎     | 2826/6470 [35:24<49:27,  1.23it/s]
2|2827|Loss: 0.8365128636360168:  44%|████▎     | 2826/6470 [35:25<49:27,  1.23it/s]
2|2827|Loss: 0.8365128636360168:  44%|████▎     | 2827/6470 [35:25<50:50,  1.19it/s]
2|2828|Loss: 0.9018791913986206:  44%|████▎     | 2827/6470 [35:26<50:50,  1.19it/s]
2|2828|Loss: 0.9018791913986206:  44%|████▎     | 2828/6470 [35:26<49:04,  1.24it/s]
2|2829|Loss: 0.8934071063995361:  44%|████▎     | 2828/6470 [35:26<49:04,  1.24it/s]
2|2829|Loss: 0.8934071063995361:  44%|████▎     | 2829/6470 [35:27<49:48,  1.22it/s]
2|2830|Loss: 0.8247120976448059:  44%|████▎     | 2829/6470 [35:27<49:48,  1.22it/s]
2|2830|Loss: 0.8247120976448059:  44%|████▎     | 2830/6470 [35:27<48:25,  1.25it/s]
2|2831|Loss: 0.6380696892738342:  44%|████▎     | 2830/6470 [35:28<48:25,  1.25it/s]
2|2831|Loss: 0.6380696892738342:  44%|████▍     | 2831/6470 [35:28<47:30,  1.28it/s]
2|2832|Loss: 0.7879679799079895:  44%|████▍     | 2831/6470 [35:29<47:30,  1.28it/s]
2|2832|Loss: 0.7879679799079895:  44%|████▍     | 2832/6470 [35:29<49:41,  1.22it/s]
2|2833|Loss: 0.8209461569786072:  44%|████▍     | 2832/6470 [35:30<49:41,  1.22it/s]
2|2833|Loss: 0.8209461569786072:  44%|████▍     | 2833/6470 [35:30<46:37,  1.30it/s]
2|2834|Loss: 0.6940362453460693:  44%|████▍     | 2833/6470 [35:30<46:37,  1.30it/s]
2|2834|Loss: 0.6940362453460693:  44%|████▍     | 2834/6470 [35:30<44:14,  1.37it/s]
2|2835|Loss: 0.8808466196060181:  44%|████▍     | 2834/6470 [35:31<44:14,  1.37it/s]
2|2835|Loss: 0.8808466196060181:  44%|████▍     | 2835/6470 [35:31<42:29,  1.43it/s]
2|2836|Loss: 0.7821950316429138:  44%|████▍     | 2835/6470 [35:31<42:29,  1.43it/s]
2|2836|Loss: 0.7821950316429138:  44%|████▍     | 2836/6470 [35:31<37:31,  1.61it/s]
2|2837|Loss: 0.8833821415901184:  44%|████▍     | 2836/6470 [35:32<37:31,  1.61it/s]
2|2837|Loss: 0.8833821415901184:  44%|████▍     | 2837/6470 [35:32<37:16,  1.62it/s]
2|2838|Loss: 0.9184666872024536:  44%|████▍     | 2837/6470 [35:32<37:16,  1.62it/s]
2|2838|Loss: 0.9184666872024536:  44%|████▍     | 2838/6470 [35:33<38:35,  1.57it/s]
2|2839|Loss: 0.8531716465950012:  44%|████▍     | 2838/6470 [35:33<38:35,  1.57it/s]
2|2839|Loss: 0.8531716465950012:  44%|████▍     | 2839/6470 [35:34<40:55,  1.48it/s]
2|2840|Loss: 0.8426102995872498:  44%|████▍     | 2839/6470 [35:34<40:55,  1.48it/s]
2|2840|Loss: 0.8426102995872498:  44%|████▍     | 2840/6470 [35:34<44:05,  1.37it/s]
2|2841|Loss: 0.8820822238922119:  44%|████▍     | 2840/6470 [35:35<44:05,  1.37it/s]
2|2841|Loss: 0.8820822238922119:  44%|████▍     | 2841/6470 [35:35<45:42,  1.32it/s]
2|2842|Loss: 0.9026210308074951:  44%|████▍     | 2841/6470 [35:36<45:42,  1.32it/s]
2|2842|Loss: 0.9026210308074951:  44%|████▍     | 2842/6470 [35:36<42:58,  1.41it/s]
2|2843|Loss: 0.7296303510665894:  44%|████▍     | 2842/6470 [35:36<42:58,  1.41it/s]
2|2843|Loss: 0.7296303510665894:  44%|████▍     | 2843/6470 [35:36<39:26,  1.53it/s]
2|2844|Loss: 0.8401710987091064:  44%|████▍     | 2843/6470 [35:37<39:26,  1.53it/s]
2|2844|Loss: 0.8401710987091064:  44%|████▍     | 2844/6470 [35:37<42:28,  1.42it/s]
2|2845|Loss: 0.9584353566169739:  44%|████▍     | 2844/6470 [35:38<42:28,  1.42it/s]
2|2845|Loss: 0.9584353566169739:  44%|████▍     | 2845/6470 [35:38<44:22,  1.36it/s]
2|2846|Loss: 0.9823551177978516:  44%|████▍     | 2845/6470 [35:38<44:22,  1.36it/s]
2|2846|Loss: 0.9823551177978516:  44%|████▍     | 2846/6470 [35:39<44:58,  1.34it/s]
2|2847|Loss: 0.7883519530296326:  44%|████▍     | 2846/6470 [35:39<44:58,  1.34it/s]
2|2847|Loss: 0.7883519530296326:  44%|████▍     | 2847/6470 [35:40<46:54,  1.29it/s]
2|2848|Loss: 0.8266656398773193:  44%|████▍     | 2847/6470 [35:40<46:54,  1.29it/s]
2|2848|Loss: 0.8266656398773193:  44%|████▍     | 2848/6470 [35:40<49:38,  1.22it/s]
2|2849|Loss: 0.6779741644859314:  44%|████▍     | 2848/6470 [35:41<49:38,  1.22it/s]
2|2849|Loss: 0.6779741644859314:  44%|████▍     | 2849/6470 [35:41<43:49,  1.38it/s]
2|2850|Loss: 0.8079585433006287:  44%|████▍     | 2849/6470 [35:41<43:49,  1.38it/s]
2|2850|Loss: 0.8079585433006287:  44%|████▍     | 2850/6470 [35:42<43:24,  1.39it/s]
2|2851|Loss: 0.7565641403198242:  44%|████▍     | 2850/6470 [35:42<43:24,  1.39it/s]
2|2851|Loss: 0.7565641403198242:  44%|████▍     | 2851/6470 [35:42<41:43,  1.45it/s]
2|2852|Loss: 0.8405875563621521:  44%|████▍     | 2851/6470 [35:43<41:43,  1.45it/s]
2|2852|Loss: 0.8405875563621521:  44%|████▍     | 2852/6470 [35:43<39:43,  1.52it/s]
2|2853|Loss: 0.9803395867347717:  44%|████▍     | 2852/6470 [35:43<39:43,  1.52it/s]
2|2853|Loss: 0.9803395867347717:  44%|████▍     | 2853/6470 [35:44<39:11,  1.54it/s]
2|2854|Loss: 0.8402279615402222:  44%|████▍     | 2853/6470 [35:44<39:11,  1.54it/s]
2|2854|Loss: 0.8402279615402222:  44%|████▍     | 2854/6470 [35:44<39:56,  1.51it/s]
2|2855|Loss: 0.8440920114517212:  44%|████▍     | 2854/6470 [35:45<39:56,  1.51it/s]
2|2855|Loss: 0.8440920114517212:  44%|████▍     | 2855/6470 [35:45<43:41,  1.38it/s]
2|2856|Loss: 0.8917284607887268:  44%|████▍     | 2855/6470 [35:46<43:41,  1.38it/s]
2|2856|Loss: 0.8917284607887268:  44%|████▍     | 2856/6470 [35:46<43:34,  1.38it/s]
2|2857|Loss: 0.9801924824714661:  44%|████▍     | 2856/6470 [35:46<43:34,  1.38it/s]
2|2857|Loss: 0.9801924824714661:  44%|████▍     | 2857/6470 [35:47<43:49,  1.37it/s]
2|2858|Loss: 0.8488216996192932:  44%|████▍     | 2857/6470 [35:47<43:49,  1.37it/s]
2|2858|Loss: 0.8488216996192932:  44%|████▍     | 2858/6470 [35:47<44:20,  1.36it/s]
2|2859|Loss: 0.9431436061859131:  44%|████▍     | 2858/6470 [35:48<44:20,  1.36it/s]
2|2859|Loss: 0.9431436061859131:  44%|████▍     | 2859/6470 [35:48<45:33,  1.32it/s]
2|2860|Loss: 0.808387279510498:  44%|████▍     | 2859/6470 [35:49<45:33,  1.32it/s] 
2|2860|Loss: 0.808387279510498:  44%|████▍     | 2860/6470 [35:49<48:08,  1.25it/s]
2|2861|Loss: 0.8224170804023743:  44%|████▍     | 2860/6470 [35:49<48:08,  1.25it/s]
2|2861|Loss: 0.8224170804023743:  44%|████▍     | 2861/6470 [35:50<46:13,  1.30it/s]
2|2862|Loss: 0.712065577507019:  44%|████▍     | 2861/6470 [35:50<46:13,  1.30it/s] 
2|2862|Loss: 0.712065577507019:  44%|████▍     | 2862/6470 [35:51<48:07,  1.25it/s]
2|2863|Loss: 1.0193432569503784:  44%|████▍     | 2862/6470 [35:51<48:07,  1.25it/s]
2|2863|Loss: 1.0193432569503784:  44%|████▍     | 2863/6470 [35:51<46:45,  1.29it/s]
2|2864|Loss: 0.7714579105377197:  44%|████▍     | 2863/6470 [35:52<46:45,  1.29it/s]
2|2864|Loss: 0.7714579105377197:  44%|████▍     | 2864/6470 [35:52<46:15,  1.30it/s]
2|2865|Loss: 0.9811045527458191:  44%|████▍     | 2864/6470 [35:52<46:15,  1.30it/s]
2|2865|Loss: 0.9811045527458191:  44%|████▍     | 2865/6470 [35:53<45:59,  1.31it/s]
2|2866|Loss: 0.7767020463943481:  44%|████▍     | 2865/6470 [35:53<45:59,  1.31it/s]
2|2866|Loss: 0.7767020463943481:  44%|████▍     | 2866/6470 [35:54<46:32,  1.29it/s]
2|2867|Loss: 0.7841570377349854:  44%|████▍     | 2866/6470 [35:54<46:32,  1.29it/s]
2|2867|Loss: 0.7841570377349854:  44%|████▍     | 2867/6470 [35:54<46:53,  1.28it/s]
2|2868|Loss: 0.9472932815551758:  44%|████▍     | 2867/6470 [35:55<46:53,  1.28it/s]
2|2868|Loss: 0.9472932815551758:  44%|████▍     | 2868/6470 [35:55<47:18,  1.27it/s]
2|2869|Loss: 0.8341941237449646:  44%|████▍     | 2868/6470 [35:56<47:18,  1.27it/s]
2|2869|Loss: 0.8341941237449646:  44%|████▍     | 2869/6470 [35:56<46:04,  1.30it/s]
2|2870|Loss: 0.7979208827018738:  44%|████▍     | 2869/6470 [35:56<46:04,  1.30it/s]
2|2870|Loss: 0.7979208827018738:  44%|████▍     | 2870/6470 [35:57<43:53,  1.37it/s]
2|2871|Loss: 1.0033841133117676:  44%|████▍     | 2870/6470 [35:57<43:53,  1.37it/s]
2|2871|Loss: 1.0033841133117676:  44%|████▍     | 2871/6470 [35:57<46:09,  1.30it/s]
2|2872|Loss: 0.831601083278656:  44%|████▍     | 2871/6470 [35:58<46:09,  1.30it/s] 
2|2872|Loss: 0.831601083278656:  44%|████▍     | 2872/6470 [35:58<46:22,  1.29it/s]
2|2873|Loss: 0.7669366002082825:  44%|████▍     | 2872/6470 [35:59<46:22,  1.29it/s]
2|2873|Loss: 0.7669366002082825:  44%|████▍     | 2873/6470 [35:59<48:03,  1.25it/s]
2|2874|Loss: 0.8205599188804626:  44%|████▍     | 2873/6470 [36:00<48:03,  1.25it/s]
2|2874|Loss: 0.8205599188804626:  44%|████▍     | 2874/6470 [36:00<48:39,  1.23it/s]
2|2875|Loss: 0.7291136980056763:  44%|████▍     | 2874/6470 [36:00<48:39,  1.23it/s]
2|2875|Loss: 0.7291136980056763:  44%|████▍     | 2875/6470 [36:01<47:08,  1.27it/s]
2|2876|Loss: 0.7916865348815918:  44%|████▍     | 2875/6470 [36:01<47:08,  1.27it/s]
2|2876|Loss: 0.7916865348815918:  44%|████▍     | 2876/6470 [36:01<45:10,  1.33it/s]
2|2877|Loss: 0.785570502281189:  44%|████▍     | 2876/6470 [36:02<45:10,  1.33it/s] 
2|2877|Loss: 0.785570502281189:  44%|████▍     | 2877/6470 [36:02<43:31,  1.38it/s]
2|2878|Loss: 0.7815192937850952:  44%|████▍     | 2877/6470 [36:02<43:31,  1.38it/s]
2|2878|Loss: 0.7815192937850952:  44%|████▍     | 2878/6470 [36:03<45:59,  1.30it/s]
2|2879|Loss: 0.8882957696914673:  44%|████▍     | 2878/6470 [36:03<45:59,  1.30it/s]
2|2879|Loss: 0.8882957696914673:  44%|████▍     | 2879/6470 [36:04<47:54,  1.25it/s]
2|2880|Loss: 0.825157105922699:  44%|████▍     | 2879/6470 [36:04<47:54,  1.25it/s] 
2|2880|Loss: 0.825157105922699:  45%|████▍     | 2880/6470 [36:04<46:28,  1.29it/s]
2|2881|Loss: 0.908750593662262:  45%|████▍     | 2880/6470 [36:05<46:28,  1.29it/s]
2|2881|Loss: 0.908750593662262:  45%|████▍     | 2881/6470 [36:05<44:15,  1.35it/s]
2|2882|Loss: 0.8918308615684509:  45%|████▍     | 2881/6470 [36:05<44:15,  1.35it/s]
2|2882|Loss: 0.8918308615684509:  45%|████▍     | 2882/6470 [36:06<43:35,  1.37it/s]
2|2883|Loss: 0.9340453743934631:  45%|████▍     | 2882/6470 [36:06<43:35,  1.37it/s]
2|2883|Loss: 0.9340453743934631:  45%|████▍     | 2883/6470 [36:07<44:02,  1.36it/s]
2|2884|Loss: 0.6992467045783997:  45%|████▍     | 2883/6470 [36:07<44:02,  1.36it/s]
2|2884|Loss: 0.6992467045783997:  45%|████▍     | 2884/6470 [36:07<47:14,  1.27it/s]
2|2885|Loss: 0.949958860874176:  45%|████▍     | 2884/6470 [36:08<47:14,  1.27it/s] 
2|2885|Loss: 0.949958860874176:  45%|████▍     | 2885/6470 [36:08<45:34,  1.31it/s]
2|2886|Loss: 0.8579318523406982:  45%|████▍     | 2885/6470 [36:09<45:34,  1.31it/s]
2|2886|Loss: 0.8579318523406982:  45%|████▍     | 2886/6470 [36:09<44:20,  1.35it/s]
2|2887|Loss: 0.7988681793212891:  45%|████▍     | 2886/6470 [36:09<44:20,  1.35it/s]
2|2887|Loss: 0.7988681793212891:  45%|████▍     | 2887/6470 [36:10<45:59,  1.30it/s]
2|2888|Loss: 0.6957948803901672:  45%|████▍     | 2887/6470 [36:10<45:59,  1.30it/s]
2|2888|Loss: 0.6957948803901672:  45%|████▍     | 2888/6470 [36:10<43:16,  1.38it/s]
2|2889|Loss: 0.7514248490333557:  45%|████▍     | 2888/6470 [36:11<43:16,  1.38it/s]
2|2889|Loss: 0.7514248490333557:  45%|████▍     | 2889/6470 [36:11<43:30,  1.37it/s]
2|2890|Loss: 0.8889344334602356:  45%|████▍     | 2889/6470 [36:12<43:30,  1.37it/s]
2|2890|Loss: 0.8889344334602356:  45%|████▍     | 2890/6470 [36:12<46:41,  1.28it/s]
2|2891|Loss: 0.8250023722648621:  45%|████▍     | 2890/6470 [36:13<46:41,  1.28it/s]
2|2891|Loss: 0.8250023722648621:  45%|████▍     | 2891/6470 [36:13<49:52,  1.20it/s]
2|2892|Loss: 0.8023968935012817:  45%|████▍     | 2891/6470 [36:13<49:52,  1.20it/s]
2|2892|Loss: 0.8023968935012817:  45%|████▍     | 2892/6470 [36:14<51:29,  1.16it/s]
2|2893|Loss: 0.8822062611579895:  45%|████▍     | 2892/6470 [36:14<51:29,  1.16it/s]
2|2893|Loss: 0.8822062611579895:  45%|████▍     | 2893/6470 [36:15<51:42,  1.15it/s]
2|2894|Loss: 0.6535736918449402:  45%|████▍     | 2893/6470 [36:15<51:42,  1.15it/s]
2|2894|Loss: 0.6535736918449402:  45%|████▍     | 2894/6470 [36:16<51:01,  1.17it/s]
2|2895|Loss: 0.7592884302139282:  45%|████▍     | 2894/6470 [36:16<51:01,  1.17it/s]
2|2895|Loss: 0.7592884302139282:  45%|████▍     | 2895/6470 [36:16<49:58,  1.19it/s]
2|2896|Loss: 0.9363657832145691:  45%|████▍     | 2895/6470 [36:17<49:58,  1.19it/s]
2|2896|Loss: 0.9363657832145691:  45%|████▍     | 2896/6470 [36:17<46:50,  1.27it/s]
2|2897|Loss: 0.8199126720428467:  45%|████▍     | 2896/6470 [36:17<46:50,  1.27it/s]
2|2897|Loss: 0.8199126720428467:  45%|████▍     | 2897/6470 [36:18<46:35,  1.28it/s]
2|2898|Loss: 0.7458052635192871:  45%|████▍     | 2897/6470 [36:18<46:35,  1.28it/s]
2|2898|Loss: 0.7458052635192871:  45%|████▍     | 2898/6470 [36:18<42:53,  1.39it/s]
2|2899|Loss: 0.8855581879615784:  45%|████▍     | 2898/6470 [36:19<42:53,  1.39it/s]
2|2899|Loss: 0.8855581879615784:  45%|████▍     | 2899/6470 [36:19<40:54,  1.45it/s]
2|2900|Loss: 0.8776937127113342:  45%|████▍     | 2899/6470 [36:19<40:54,  1.45it/s]
2|2900|Loss: 0.8776937127113342:  45%|████▍     | 2900/6470 [36:20<43:59,  1.35it/s]
2|2901|Loss: 0.8547259569168091:  45%|████▍     | 2900/6470 [36:20<43:59,  1.35it/s]
2|2901|Loss: 0.8547259569168091:  45%|████▍     | 2901/6470 [36:21<42:56,  1.39it/s]
2|2902|Loss: 0.7415640950202942:  45%|████▍     | 2901/6470 [36:21<42:56,  1.39it/s]
2|2902|Loss: 0.7415640950202942:  45%|████▍     | 2902/6470 [36:21<43:48,  1.36it/s]
2|2903|Loss: 0.9073889255523682:  45%|████▍     | 2902/6470 [36:22<43:48,  1.36it/s]
2|2903|Loss: 0.9073889255523682:  45%|████▍     | 2903/6470 [36:22<44:29,  1.34it/s]
2|2904|Loss: 0.6339467167854309:  45%|████▍     | 2903/6470 [36:23<44:29,  1.34it/s]
2|2904|Loss: 0.6339467167854309:  45%|████▍     | 2904/6470 [36:23<45:12,  1.31it/s]
2|2905|Loss: 0.9207050204277039:  45%|████▍     | 2904/6470 [36:23<45:12,  1.31it/s]
2|2905|Loss: 0.9207050204277039:  45%|████▍     | 2905/6470 [36:24<45:16,  1.31it/s]
2|2906|Loss: 0.9204962253570557:  45%|████▍     | 2905/6470 [36:24<45:16,  1.31it/s]
2|2906|Loss: 0.9204962253570557:  45%|████▍     | 2906/6470 [36:25<48:08,  1.23it/s]
2|2907|Loss: 0.8586062788963318:  45%|████▍     | 2906/6470 [36:25<48:08,  1.23it/s]
2|2907|Loss: 0.8586062788963318:  45%|████▍     | 2907/6470 [36:25<49:41,  1.20it/s]
2|2908|Loss: 0.8681667447090149:  45%|████▍     | 2907/6470 [36:26<49:41,  1.20it/s]
2|2908|Loss: 0.8681667447090149:  45%|████▍     | 2908/6470 [36:26<51:47,  1.15it/s]
2|2909|Loss: 1.0058537721633911:  45%|████▍     | 2908/6470 [36:27<51:47,  1.15it/s]
2|2909|Loss: 1.0058537721633911:  45%|████▍     | 2909/6470 [36:27<53:01,  1.12it/s]
2|2910|Loss: 0.917582094669342:  45%|████▍     | 2909/6470 [36:28<53:01,  1.12it/s] 
2|2910|Loss: 0.917582094669342:  45%|████▍     | 2910/6470 [36:28<49:46,  1.19it/s]
2|2911|Loss: 0.7451059818267822:  45%|████▍     | 2910/6470 [36:28<49:46,  1.19it/s]
2|2911|Loss: 0.7451059818267822:  45%|████▍     | 2911/6470 [36:29<49:41,  1.19it/s]
2|2912|Loss: 0.8172954320907593:  45%|████▍     | 2911/6470 [36:29<49:41,  1.19it/s]
2|2912|Loss: 0.8172954320907593:  45%|████▌     | 2912/6470 [36:30<50:22,  1.18it/s]
2|2913|Loss: 0.7547810077667236:  45%|████▌     | 2912/6470 [36:30<50:22,  1.18it/s]
2|2913|Loss: 0.7547810077667236:  45%|████▌     | 2913/6470 [36:30<47:56,  1.24it/s]
2|2914|Loss: 0.9021387100219727:  45%|████▌     | 2913/6470 [36:31<47:56,  1.24it/s]
2|2914|Loss: 0.9021387100219727:  45%|████▌     | 2914/6470 [36:31<47:21,  1.25it/s]
2|2915|Loss: 0.884526789188385:  45%|████▌     | 2914/6470 [36:32<47:21,  1.25it/s] 
2|2915|Loss: 0.884526789188385:  45%|████▌     | 2915/6470 [36:32<48:10,  1.23it/s]
2|2916|Loss: 0.7991433143615723:  45%|████▌     | 2915/6470 [36:33<48:10,  1.23it/s]
2|2916|Loss: 0.7991433143615723:  45%|████▌     | 2916/6470 [36:33<45:46,  1.29it/s]
2|2917|Loss: 0.9329315423965454:  45%|████▌     | 2916/6470 [36:33<45:46,  1.29it/s]
2|2917|Loss: 0.9329315423965454:  45%|████▌     | 2917/6470 [36:34<44:58,  1.32it/s]
2|2918|Loss: 0.9245195984840393:  45%|████▌     | 2917/6470 [36:34<44:58,  1.32it/s]
2|2918|Loss: 0.9245195984840393:  45%|████▌     | 2918/6470 [36:34<47:25,  1.25it/s]
2|2919|Loss: 0.9047479033470154:  45%|████▌     | 2918/6470 [36:35<47:25,  1.25it/s]
2|2919|Loss: 0.9047479033470154:  45%|████▌     | 2919/6470 [36:35<45:28,  1.30it/s]
2|2920|Loss: 0.8087955117225647:  45%|████▌     | 2919/6470 [36:36<45:28,  1.30it/s]
2|2920|Loss: 0.8087955117225647:  45%|████▌     | 2920/6470 [36:36<45:21,  1.30it/s]
2|2921|Loss: 0.7845008373260498:  45%|████▌     | 2920/6470 [36:36<45:21,  1.30it/s]
2|2921|Loss: 0.7845008373260498:  45%|████▌     | 2921/6470 [36:37<46:52,  1.26it/s]
2|2922|Loss: 1.015036940574646:  45%|████▌     | 2921/6470 [36:37<46:52,  1.26it/s] 
2|2922|Loss: 1.015036940574646:  45%|████▌     | 2922/6470 [36:38<47:28,  1.25it/s]
2|2923|Loss: 0.6891356110572815:  45%|████▌     | 2922/6470 [36:38<47:28,  1.25it/s]
2|2923|Loss: 0.6891356110572815:  45%|████▌     | 2923/6470 [36:38<46:11,  1.28it/s]
2|2924|Loss: 0.7277446389198303:  45%|████▌     | 2923/6470 [36:39<46:11,  1.28it/s]
2|2924|Loss: 0.7277446389198303:  45%|████▌     | 2924/6470 [36:39<46:10,  1.28it/s]
2|2925|Loss: 0.9281729459762573:  45%|████▌     | 2924/6470 [36:39<46:10,  1.28it/s]
2|2925|Loss: 0.9281729459762573:  45%|████▌     | 2925/6470 [36:40<44:37,  1.32it/s]
2|2926|Loss: 0.8979378938674927:  45%|████▌     | 2925/6470 [36:40<44:37,  1.32it/s]
2|2926|Loss: 0.8979378938674927:  45%|████▌     | 2926/6470 [36:40<43:03,  1.37it/s]
2|2927|Loss: 0.7847203016281128:  45%|████▌     | 2926/6470 [36:41<43:03,  1.37it/s]
2|2927|Loss: 0.7847203016281128:  45%|████▌     | 2927/6470 [36:41<40:06,  1.47it/s]
2|2928|Loss: 0.8530629277229309:  45%|████▌     | 2927/6470 [36:41<40:06,  1.47it/s]
2|2928|Loss: 0.8530629277229309:  45%|████▌     | 2928/6470 [36:42<39:32,  1.49it/s]
2|2929|Loss: 0.6991965174674988:  45%|████▌     | 2928/6470 [36:42<39:32,  1.49it/s]
2|2929|Loss: 0.6991965174674988:  45%|████▌     | 2929/6470 [36:42<41:12,  1.43it/s]
2|2930|Loss: 0.8229027986526489:  45%|████▌     | 2929/6470 [36:43<41:12,  1.43it/s]
2|2930|Loss: 0.8229027986526489:  45%|████▌     | 2930/6470 [36:43<39:18,  1.50it/s]
2|2931|Loss: 0.7907609939575195:  45%|████▌     | 2930/6470 [36:43<39:18,  1.50it/s]
2|2931|Loss: 0.7907609939575195:  45%|████▌     | 2931/6470 [36:44<39:14,  1.50it/s]
2|2932|Loss: 0.8447447419166565:  45%|████▌     | 2931/6470 [36:44<39:14,  1.50it/s]
2|2932|Loss: 0.8447447419166565:  45%|████▌     | 2932/6470 [36:45<42:50,  1.38it/s]
2|2933|Loss: 0.844020426273346:  45%|████▌     | 2932/6470 [36:45<42:50,  1.38it/s] 
2|2933|Loss: 0.844020426273346:  45%|████▌     | 2933/6470 [36:46<48:09,  1.22it/s]
2|2934|Loss: 0.7758563160896301:  45%|████▌     | 2933/6470 [36:46<48:09,  1.22it/s]
2|2934|Loss: 0.7758563160896301:  45%|████▌     | 2934/6470 [36:46<47:35,  1.24it/s]
2|2935|Loss: 0.8029736280441284:  45%|████▌     | 2934/6470 [36:47<47:35,  1.24it/s]
2|2935|Loss: 0.8029736280441284:  45%|████▌     | 2935/6470 [36:47<48:57,  1.20it/s]
2|2936|Loss: 0.8623063564300537:  45%|████▌     | 2935/6470 [36:48<48:57,  1.20it/s]
2|2936|Loss: 0.8623063564300537:  45%|████▌     | 2936/6470 [36:48<48:21,  1.22it/s]
2|2937|Loss: 0.8198817372322083:  45%|████▌     | 2936/6470 [36:48<48:21,  1.22it/s]
2|2937|Loss: 0.8198817372322083:  45%|████▌     | 2937/6470 [36:49<46:35,  1.26it/s]
2|2938|Loss: 0.8862355947494507:  45%|████▌     | 2937/6470 [36:49<46:35,  1.26it/s]
2|2938|Loss: 0.8862355947494507:  45%|████▌     | 2938/6470 [36:50<46:31,  1.27it/s]
2|2939|Loss: 0.8339770436286926:  45%|████▌     | 2938/6470 [36:50<46:31,  1.27it/s]
2|2939|Loss: 0.8339770436286926:  45%|████▌     | 2939/6470 [36:50<48:15,  1.22it/s]
2|2940|Loss: 0.8923728466033936:  45%|████▌     | 2939/6470 [36:51<48:15,  1.22it/s]
2|2940|Loss: 0.8923728466033936:  45%|████▌     | 2940/6470 [36:51<47:29,  1.24it/s]
2|2941|Loss: 0.7701861262321472:  45%|████▌     | 2940/6470 [36:52<47:29,  1.24it/s]
2|2941|Loss: 0.7701861262321472:  45%|████▌     | 2941/6470 [36:52<44:49,  1.31it/s]
2|2942|Loss: 0.7194007635116577:  45%|████▌     | 2941/6470 [36:52<44:49,  1.31it/s]
2|2942|Loss: 0.7194007635116577:  45%|████▌     | 2942/6470 [36:53<45:48,  1.28it/s]
2|2943|Loss: 0.7982948422431946:  45%|████▌     | 2942/6470 [36:53<45:48,  1.28it/s]
2|2943|Loss: 0.7982948422431946:  45%|████▌     | 2943/6470 [36:53<43:44,  1.34it/s]
2|2944|Loss: 0.9980201721191406:  45%|████▌     | 2943/6470 [36:54<43:44,  1.34it/s]
2|2944|Loss: 0.9980201721191406:  46%|████▌     | 2944/6470 [36:54<45:27,  1.29it/s]
2|2945|Loss: 0.9375735521316528:  46%|████▌     | 2944/6470 [36:55<45:27,  1.29it/s]
2|2945|Loss: 0.9375735521316528:  46%|████▌     | 2945/6470 [36:55<47:39,  1.23it/s]
2|2946|Loss: 0.8204453587532043:  46%|████▌     | 2945/6470 [36:55<47:39,  1.23it/s]
2|2946|Loss: 0.8204453587532043:  46%|████▌     | 2946/6470 [36:56<44:15,  1.33it/s]
2|2947|Loss: 0.8573306798934937:  46%|████▌     | 2946/6470 [36:56<44:15,  1.33it/s]
2|2947|Loss: 0.8573306798934937:  46%|████▌     | 2947/6470 [36:56<42:28,  1.38it/s]
2|2948|Loss: 0.8972834348678589:  46%|████▌     | 2947/6470 [36:57<42:28,  1.38it/s]
2|2948|Loss: 0.8972834348678589:  46%|████▌     | 2948/6470 [36:57<41:46,  1.41it/s]
2|2949|Loss: 0.7234377861022949:  46%|████▌     | 2948/6470 [36:57<41:46,  1.41it/s]
2|2949|Loss: 0.7234377861022949:  46%|████▌     | 2949/6470 [36:58<39:37,  1.48it/s]
2|2950|Loss: 0.7639553546905518:  46%|████▌     | 2949/6470 [36:58<39:37,  1.48it/s]
2|2950|Loss: 0.7639553546905518:  46%|████▌     | 2950/6470 [36:58<41:58,  1.40it/s]
2|2951|Loss: 0.8487988710403442:  46%|████▌     | 2950/6470 [36:59<41:58,  1.40it/s]
2|2951|Loss: 0.8487988710403442:  46%|████▌     | 2951/6470 [36:59<44:34,  1.32it/s]
2|2952|Loss: 0.9241523146629333:  46%|████▌     | 2951/6470 [37:00<44:34,  1.32it/s]
2|2952|Loss: 0.9241523146629333:  46%|████▌     | 2952/6470 [37:00<44:52,  1.31it/s]
2|2953|Loss: 0.8908030390739441:  46%|████▌     | 2952/6470 [37:00<44:52,  1.31it/s]
2|2953|Loss: 0.8908030390739441:  46%|████▌     | 2953/6470 [37:01<43:54,  1.33it/s]
2|2954|Loss: 0.8544647693634033:  46%|████▌     | 2953/6470 [37:01<43:54,  1.33it/s]
2|2954|Loss: 0.8544647693634033:  46%|████▌     | 2954/6470 [37:01<41:58,  1.40it/s]
2|2955|Loss: 0.8357505202293396:  46%|████▌     | 2954/6470 [37:02<41:58,  1.40it/s]
2|2955|Loss: 0.8357505202293396:  46%|████▌     | 2955/6470 [37:02<44:46,  1.31it/s]
2|2956|Loss: 0.9338122606277466:  46%|████▌     | 2955/6470 [37:03<44:46,  1.31it/s]
2|2956|Loss: 0.9338122606277466:  46%|████▌     | 2956/6470 [37:03<42:30,  1.38it/s]
2|2957|Loss: 0.893315315246582:  46%|████▌     | 2956/6470 [37:03<42:30,  1.38it/s] 
2|2957|Loss: 0.893315315246582:  46%|████▌     | 2957/6470 [37:04<42:29,  1.38it/s]
2|2958|Loss: 0.7329397201538086:  46%|████▌     | 2957/6470 [37:04<42:29,  1.38it/s]
2|2958|Loss: 0.7329397201538086:  46%|████▌     | 2958/6470 [37:04<43:58,  1.33it/s]
2|2959|Loss: 0.7727270722389221:  46%|████▌     | 2958/6470 [37:05<43:58,  1.33it/s]
2|2959|Loss: 0.7727270722389221:  46%|████▌     | 2959/6470 [37:05<43:25,  1.35it/s]
2|2960|Loss: 0.9802061319351196:  46%|████▌     | 2959/6470 [37:06<43:25,  1.35it/s]
2|2960|Loss: 0.9802061319351196:  46%|████▌     | 2960/6470 [37:06<40:40,  1.44it/s]
2|2961|Loss: 0.7993670701980591:  46%|████▌     | 2960/6470 [37:06<40:40,  1.44it/s]
2|2961|Loss: 0.7993670701980591:  46%|████▌     | 2961/6470 [37:06<39:53,  1.47it/s]
2|2962|Loss: 0.7364354133605957:  46%|████▌     | 2961/6470 [37:07<39:53,  1.47it/s]
2|2962|Loss: 0.7364354133605957:  46%|████▌     | 2962/6470 [37:07<39:40,  1.47it/s]
2|2963|Loss: 0.8548652529716492:  46%|████▌     | 2962/6470 [37:08<39:40,  1.47it/s]
2|2963|Loss: 0.8548652529716492:  46%|████▌     | 2963/6470 [37:08<41:29,  1.41it/s]
2|2964|Loss: 0.882832944393158:  46%|████▌     | 2963/6470 [37:08<41:29,  1.41it/s] 
2|2964|Loss: 0.882832944393158:  46%|████▌     | 2964/6470 [37:09<41:25,  1.41it/s]
2|2965|Loss: 0.8298065662384033:  46%|████▌     | 2964/6470 [37:09<41:25,  1.41it/s]
2|2965|Loss: 0.8298065662384033:  46%|████▌     | 2965/6470 [37:09<40:00,  1.46it/s]
2|2966|Loss: 0.8993212580680847:  46%|████▌     | 2965/6470 [37:10<40:00,  1.46it/s]
2|2966|Loss: 0.8993212580680847:  46%|████▌     | 2966/6470 [37:10<41:19,  1.41it/s]
2|2967|Loss: 0.8373765349388123:  46%|████▌     | 2966/6470 [37:10<41:19,  1.41it/s]
2|2967|Loss: 0.8373765349388123:  46%|████▌     | 2967/6470 [37:11<42:42,  1.37it/s]
2|2968|Loss: 0.9561005234718323:  46%|████▌     | 2967/6470 [37:11<42:42,  1.37it/s]
2|2968|Loss: 0.9561005234718323:  46%|████▌     | 2968/6470 [37:12<44:06,  1.32it/s]
2|2969|Loss: 0.6593389511108398:  46%|████▌     | 2968/6470 [37:12<44:06,  1.32it/s]
2|2969|Loss: 0.6593389511108398:  46%|████▌     | 2969/6470 [37:12<42:37,  1.37it/s]
2|2970|Loss: 0.7459248900413513:  46%|████▌     | 2969/6470 [37:13<42:37,  1.37it/s]
2|2970|Loss: 0.7459248900413513:  46%|████▌     | 2970/6470 [37:13<37:23,  1.56it/s]
2|2971|Loss: 0.8065412640571594:  46%|████▌     | 2970/6470 [37:13<37:23,  1.56it/s]
2|2971|Loss: 0.8065412640571594:  46%|████▌     | 2971/6470 [37:13<39:51,  1.46it/s]
2|2972|Loss: 0.9403169751167297:  46%|████▌     | 2971/6470 [37:14<39:51,  1.46it/s]
2|2972|Loss: 0.9403169751167297:  46%|████▌     | 2972/6470 [37:15<46:24,  1.26it/s]
2|2973|Loss: 0.8572714328765869:  46%|████▌     | 2972/6470 [37:15<46:24,  1.26it/s]
2|2973|Loss: 0.8572714328765869:  46%|████▌     | 2973/6470 [37:15<49:18,  1.18it/s]
2|2974|Loss: 0.9638291597366333:  46%|████▌     | 2973/6470 [37:16<49:18,  1.18it/s]
2|2974|Loss: 0.9638291597366333:  46%|████▌     | 2974/6470 [37:16<47:41,  1.22it/s]
2|2975|Loss: 0.8789072036743164:  46%|████▌     | 2974/6470 [37:17<47:41,  1.22it/s]
2|2975|Loss: 0.8789072036743164:  46%|████▌     | 2975/6470 [37:17<45:35,  1.28it/s]
2|2976|Loss: 1.008695363998413:  46%|████▌     | 2975/6470 [37:17<45:35,  1.28it/s] 
2|2976|Loss: 1.008695363998413:  46%|████▌     | 2976/6470 [37:18<46:23,  1.26it/s]
2|2977|Loss: 0.768835186958313:  46%|████▌     | 2976/6470 [37:18<46:23,  1.26it/s]
2|2977|Loss: 0.768835186958313:  46%|████▌     | 2977/6470 [37:18<42:44,  1.36it/s]
2|2978|Loss: 0.8313040733337402:  46%|████▌     | 2977/6470 [37:19<42:44,  1.36it/s]
2|2978|Loss: 0.8313040733337402:  46%|████▌     | 2978/6470 [37:19<49:34,  1.17it/s]
2|2979|Loss: 0.8649181127548218:  46%|████▌     | 2978/6470 [37:20<49:34,  1.17it/s]
2|2979|Loss: 0.8649181127548218:  46%|████▌     | 2979/6470 [37:20<47:42,  1.22it/s]
2|2980|Loss: 0.8041761517524719:  46%|████▌     | 2979/6470 [37:21<47:42,  1.22it/s]
2|2980|Loss: 0.8041761517524719:  46%|████▌     | 2980/6470 [37:21<45:17,  1.28it/s]
2|2981|Loss: 0.9140810370445251:  46%|████▌     | 2980/6470 [37:21<45:17,  1.28it/s]
2|2981|Loss: 0.9140810370445251:  46%|████▌     | 2981/6470 [37:22<45:35,  1.28it/s]
2|2982|Loss: 0.8822025656700134:  46%|████▌     | 2981/6470 [37:22<45:35,  1.28it/s]
2|2982|Loss: 0.8822025656700134:  46%|████▌     | 2982/6470 [37:23<45:52,  1.27it/s]
2|2983|Loss: 0.8370322585105896:  46%|████▌     | 2982/6470 [37:23<45:52,  1.27it/s]
2|2983|Loss: 0.8370322585105896:  46%|████▌     | 2983/6470 [37:23<43:06,  1.35it/s]
2|2984|Loss: 0.8773130774497986:  46%|████▌     | 2983/6470 [37:24<43:06,  1.35it/s]
2|2984|Loss: 0.8773130774497986:  46%|████▌     | 2984/6470 [37:24<43:01,  1.35it/s]
2|2985|Loss: 0.9242034554481506:  46%|████▌     | 2984/6470 [37:24<43:01,  1.35it/s]
2|2985|Loss: 0.9242034554481506:  46%|████▌     | 2985/6470 [37:25<43:19,  1.34it/s]
2|2986|Loss: 0.7400840520858765:  46%|████▌     | 2985/6470 [37:25<43:19,  1.34it/s]
2|2986|Loss: 0.7400840520858765:  46%|████▌     | 2986/6470 [37:25<43:23,  1.34it/s]
2|2987|Loss: 0.7780347466468811:  46%|████▌     | 2986/6470 [37:26<43:23,  1.34it/s]
2|2987|Loss: 0.7780347466468811:  46%|████▌     | 2987/6470 [37:26<40:45,  1.42it/s]
2|2988|Loss: 0.8771743774414062:  46%|████▌     | 2987/6470 [37:26<40:45,  1.42it/s]
2|2988|Loss: 0.8771743774414062:  46%|████▌     | 2988/6470 [37:27<39:48,  1.46it/s]
2|2989|Loss: 0.6214925050735474:  46%|████▌     | 2988/6470 [37:27<39:48,  1.46it/s]
2|2989|Loss: 0.6214925050735474:  46%|████▌     | 2989/6470 [37:28<43:16,  1.34it/s]
2|2990|Loss: 0.7704678177833557:  46%|████▌     | 2989/6470 [37:28<43:16,  1.34it/s]
2|2990|Loss: 0.7704678177833557:  46%|████▌     | 2990/6470 [37:28<42:57,  1.35it/s]
2|2991|Loss: 0.7767917513847351:  46%|████▌     | 2990/6470 [37:29<42:57,  1.35it/s]
2|2991|Loss: 0.7767917513847351:  46%|████▌     | 2991/6470 [37:29<41:41,  1.39it/s]
2|2992|Loss: 0.7715020775794983:  46%|████▌     | 2991/6470 [37:29<41:41,  1.39it/s]
2|2992|Loss: 0.7715020775794983:  46%|████▌     | 2992/6470 [37:30<43:00,  1.35it/s]
2|2993|Loss: 0.7672829031944275:  46%|████▌     | 2992/6470 [37:30<43:00,  1.35it/s]
2|2993|Loss: 0.7672829031944275:  46%|████▋     | 2993/6470 [37:31<44:19,  1.31it/s]
2|2994|Loss: 0.6584793925285339:  46%|████▋     | 2993/6470 [37:31<44:19,  1.31it/s]
2|2994|Loss: 0.6584793925285339:  46%|████▋     | 2994/6470 [37:31<46:50,  1.24it/s]
2|2995|Loss: 0.9038938879966736:  46%|████▋     | 2994/6470 [37:32<46:50,  1.24it/s]
2|2995|Loss: 0.9038938879966736:  46%|████▋     | 2995/6470 [37:32<46:38,  1.24it/s]
2|2996|Loss: 0.8589898347854614:  46%|████▋     | 2995/6470 [37:33<46:38,  1.24it/s]
2|2996|Loss: 0.8589898347854614:  46%|████▋     | 2996/6470 [37:33<42:57,  1.35it/s]
2|2997|Loss: 0.8334141373634338:  46%|████▋     | 2996/6470 [37:33<42:57,  1.35it/s]
2|2997|Loss: 0.8334141373634338:  46%|████▋     | 2997/6470 [37:33<41:05,  1.41it/s]
2|2998|Loss: 0.8854544162750244:  46%|████▋     | 2997/6470 [37:34<41:05,  1.41it/s]
2|2998|Loss: 0.8854544162750244:  46%|████▋     | 2998/6470 [37:34<38:57,  1.49it/s]
2|2999|Loss: 0.8111473917961121:  46%|████▋     | 2998/6470 [37:34<38:57,  1.49it/s]
2|2999|Loss: 0.8111473917961121:  46%|████▋     | 2999/6470 [37:35<41:17,  1.40it/s]
2|3000|Loss: 0.8539608120918274:  46%|████▋     | 2999/6470 [37:35<41:17,  1.40it/s]
2|3000|Loss: 0.8539608120918274:  46%|████▋     | 3000/6470 [37:36<42:16,  1.37it/s]
2|3001|Loss: 0.7919455170631409:  46%|████▋     | 3000/6470 [37:36<42:16,  1.37it/s]
2|3001|Loss: 0.7919455170631409:  46%|████▋     | 3001/6470 [37:37<50:20,  1.15it/s]
2|3002|Loss: 0.9338723421096802:  46%|████▋     | 3001/6470 [37:37<50:20,  1.15it/s]
2|3002|Loss: 0.9338723421096802:  46%|████▋     | 3002/6470 [37:38<50:19,  1.15it/s]
2|3003|Loss: 0.8937342166900635:  46%|████▋     | 3002/6470 [37:38<50:19,  1.15it/s]
2|3003|Loss: 0.8937342166900635:  46%|████▋     | 3003/6470 [37:38<47:15,  1.22it/s]
2|3004|Loss: 0.7873486876487732:  46%|████▋     | 3003/6470 [37:39<47:15,  1.22it/s]
2|3004|Loss: 0.7873486876487732:  46%|████▋     | 3004/6470 [37:39<46:10,  1.25it/s]
2|3005|Loss: 0.7840842604637146:  46%|████▋     | 3004/6470 [37:40<46:10,  1.25it/s]
2|3005|Loss: 0.7840842604637146:  46%|████▋     | 3005/6470 [37:40<46:10,  1.25it/s]
2|3006|Loss: 0.864514946937561:  46%|████▋     | 3005/6470 [37:40<46:10,  1.25it/s] 
2|3006|Loss: 0.864514946937561:  46%|████▋     | 3006/6470 [37:41<45:00,  1.28it/s]
2|3007|Loss: 0.7248046398162842:  46%|████▋     | 3006/6470 [37:41<45:00,  1.28it/s]
2|3007|Loss: 0.7248046398162842:  46%|████▋     | 3007/6470 [37:42<45:54,  1.26it/s]
2|3008|Loss: 0.8896788358688354:  46%|████▋     | 3007/6470 [37:42<45:54,  1.26it/s]
2|3008|Loss: 0.8896788358688354:  46%|████▋     | 3008/6470 [37:42<44:49,  1.29it/s]
2|3009|Loss: 0.6843999624252319:  46%|████▋     | 3008/6470 [37:43<44:49,  1.29it/s]
2|3009|Loss: 0.6843999624252319:  47%|████▋     | 3009/6470 [37:43<43:03,  1.34it/s]
2|3010|Loss: 0.9464557766914368:  47%|████▋     | 3009/6470 [37:43<43:03,  1.34it/s]
2|3010|Loss: 0.9464557766914368:  47%|████▋     | 3010/6470 [37:43<39:58,  1.44it/s]
2|3011|Loss: 0.6372305750846863:  47%|████▋     | 3010/6470 [37:44<39:58,  1.44it/s]
2|3011|Loss: 0.6372305750846863:  47%|████▋     | 3011/6470 [37:44<35:36,  1.62it/s]
2|3012|Loss: 0.7271378040313721:  47%|████▋     | 3011/6470 [37:44<35:36,  1.62it/s]
2|3012|Loss: 0.7271378040313721:  47%|████▋     | 3012/6470 [37:45<35:36,  1.62it/s]
2|3013|Loss: 0.9013983011245728:  47%|████▋     | 3012/6470 [37:45<35:36,  1.62it/s]
2|3013|Loss: 0.9013983011245728:  47%|████▋     | 3013/6470 [37:45<39:04,  1.47it/s]
2|3014|Loss: 0.9520935416221619:  47%|████▋     | 3013/6470 [37:46<39:04,  1.47it/s]
2|3014|Loss: 0.9520935416221619:  47%|████▋     | 3014/6470 [37:46<38:41,  1.49it/s]
2|3015|Loss: 0.7602168321609497:  47%|████▋     | 3014/6470 [37:46<38:41,  1.49it/s]
2|3015|Loss: 0.7602168321609497:  47%|████▋     | 3015/6470 [37:47<39:26,  1.46it/s]
2|3016|Loss: 0.8274810910224915:  47%|████▋     | 3015/6470 [37:47<39:26,  1.46it/s]
2|3016|Loss: 0.8274810910224915:  47%|████▋     | 3016/6470 [37:47<39:52,  1.44it/s]
2|3017|Loss: 0.9305441975593567:  47%|████▋     | 3016/6470 [37:48<39:52,  1.44it/s]
2|3017|Loss: 0.9305441975593567:  47%|████▋     | 3017/6470 [37:48<40:50,  1.41it/s]
2|3018|Loss: 0.8478024005889893:  47%|████▋     | 3017/6470 [37:49<40:50,  1.41it/s]
2|3018|Loss: 0.8478024005889893:  47%|████▋     | 3018/6470 [37:49<38:46,  1.48it/s]
2|3019|Loss: 0.885130763053894:  47%|████▋     | 3018/6470 [37:49<38:46,  1.48it/s] 
2|3019|Loss: 0.885130763053894:  47%|████▋     | 3019/6470 [37:50<41:07,  1.40it/s]
2|3020|Loss: 0.7131845355033875:  47%|████▋     | 3019/6470 [37:50<41:07,  1.40it/s]
2|3020|Loss: 0.7131845355033875:  47%|████▋     | 3020/6470 [37:50<40:58,  1.40it/s]
2|3021|Loss: 0.8012218475341797:  47%|████▋     | 3020/6470 [37:51<40:58,  1.40it/s]
2|3021|Loss: 0.8012218475341797:  47%|████▋     | 3021/6470 [37:51<41:27,  1.39it/s]
2|3022|Loss: 0.8304641246795654:  47%|████▋     | 3021/6470 [37:51<41:27,  1.39it/s]
2|3022|Loss: 0.8304641246795654:  47%|████▋     | 3022/6470 [37:52<42:29,  1.35it/s]
2|3023|Loss: 1.0661743879318237:  47%|████▋     | 3022/6470 [37:52<42:29,  1.35it/s]
2|3023|Loss: 1.0661743879318237:  47%|████▋     | 3023/6470 [37:53<43:31,  1.32it/s]
2|3024|Loss: 0.7588369846343994:  47%|████▋     | 3023/6470 [37:53<43:31,  1.32it/s]
2|3024|Loss: 0.7588369846343994:  47%|████▋     | 3024/6470 [37:53<42:51,  1.34it/s]
2|3025|Loss: 0.8280948996543884:  47%|████▋     | 3024/6470 [37:54<42:51,  1.34it/s]
2|3025|Loss: 0.8280948996543884:  47%|████▋     | 3025/6470 [37:54<40:58,  1.40it/s]
2|3026|Loss: 0.790204644203186:  47%|████▋     | 3025/6470 [37:54<40:58,  1.40it/s] 
2|3026|Loss: 0.790204644203186:  47%|████▋     | 3026/6470 [37:55<39:49,  1.44it/s]
2|3027|Loss: 0.8592367172241211:  47%|████▋     | 3026/6470 [37:55<39:49,  1.44it/s]
2|3027|Loss: 0.8592367172241211:  47%|████▋     | 3027/6470 [37:56<44:02,  1.30it/s]
2|3028|Loss: 0.8429741263389587:  47%|████▋     | 3027/6470 [37:56<44:02,  1.30it/s]
2|3028|Loss: 0.8429741263389587:  47%|████▋     | 3028/6470 [37:57<47:22,  1.21it/s]
2|3029|Loss: 0.8780850768089294:  47%|████▋     | 3028/6470 [37:57<47:22,  1.21it/s]
2|3029|Loss: 0.8780850768089294:  47%|████▋     | 3029/6470 [37:57<45:08,  1.27it/s]
2|3030|Loss: 0.8471336364746094:  47%|████▋     | 3029/6470 [37:58<45:08,  1.27it/s]
2|3030|Loss: 0.8471336364746094:  47%|████▋     | 3030/6470 [37:58<43:06,  1.33it/s]
2|3031|Loss: 0.810860812664032:  47%|████▋     | 3030/6470 [37:58<43:06,  1.33it/s] 
2|3031|Loss: 0.810860812664032:  47%|████▋     | 3031/6470 [37:59<43:27,  1.32it/s]
2|3032|Loss: 0.8316042423248291:  47%|████▋     | 3031/6470 [37:59<43:27,  1.32it/s]
2|3032|Loss: 0.8316042423248291:  47%|████▋     | 3032/6470 [38:00<44:58,  1.27it/s]
2|3033|Loss: 0.962857723236084:  47%|████▋     | 3032/6470 [38:00<44:58,  1.27it/s] 
2|3033|Loss: 0.962857723236084:  47%|████▋     | 3033/6470 [38:00<42:46,  1.34it/s]
2|3034|Loss: 0.9371207356452942:  47%|████▋     | 3033/6470 [38:01<42:46,  1.34it/s]
2|3034|Loss: 0.9371207356452942:  47%|████▋     | 3034/6470 [38:01<42:54,  1.33it/s]
2|3035|Loss: 0.7302168607711792:  47%|████▋     | 3034/6470 [38:01<42:54,  1.33it/s]
2|3035|Loss: 0.7302168607711792:  47%|████▋     | 3035/6470 [38:02<41:23,  1.38it/s]
2|3036|Loss: 0.9690898060798645:  47%|████▋     | 3035/6470 [38:02<41:23,  1.38it/s]
2|3036|Loss: 0.9690898060798645:  47%|████▋     | 3036/6470 [38:02<43:11,  1.33it/s]
2|3037|Loss: 0.838547945022583:  47%|████▋     | 3036/6470 [38:03<43:11,  1.33it/s] 
2|3037|Loss: 0.838547945022583:  47%|████▋     | 3037/6470 [38:03<45:55,  1.25it/s]
2|3038|Loss: 0.9446370005607605:  47%|████▋     | 3037/6470 [38:04<45:55,  1.25it/s]
2|3038|Loss: 0.9446370005607605:  47%|████▋     | 3038/6470 [38:04<45:06,  1.27it/s]
2|3039|Loss: 0.744784951210022:  47%|████▋     | 3038/6470 [38:05<45:06,  1.27it/s] 
2|3039|Loss: 0.744784951210022:  47%|████▋     | 3039/6470 [38:05<46:01,  1.24it/s]
2|3040|Loss: 0.7110952734947205:  47%|████▋     | 3039/6470 [38:05<46:01,  1.24it/s]
2|3040|Loss: 0.7110952734947205:  47%|████▋     | 3040/6470 [38:05<40:01,  1.43it/s]
2|3041|Loss: 0.8341765999794006:  47%|████▋     | 3040/6470 [38:06<40:01,  1.43it/s]
2|3041|Loss: 0.8341765999794006:  47%|████▋     | 3041/6470 [38:06<39:04,  1.46it/s]
2|3042|Loss: 0.9315714240074158:  47%|████▋     | 3041/6470 [38:06<39:04,  1.46it/s]
2|3042|Loss: 0.9315714240074158:  47%|████▋     | 3042/6470 [38:07<37:39,  1.52it/s]
2|3043|Loss: 0.7047842144966125:  47%|████▋     | 3042/6470 [38:07<37:39,  1.52it/s]
2|3043|Loss: 0.7047842144966125:  47%|████▋     | 3043/6470 [38:07<39:51,  1.43it/s]
2|3044|Loss: 0.7913374304771423:  47%|████▋     | 3043/6470 [38:08<39:51,  1.43it/s]
2|3044|Loss: 0.7913374304771423:  47%|████▋     | 3044/6470 [38:08<42:41,  1.34it/s]
2|3045|Loss: 0.8511266708374023:  47%|████▋     | 3044/6470 [38:09<42:41,  1.34it/s]
2|3045|Loss: 0.8511266708374023:  47%|████▋     | 3045/6470 [38:09<42:24,  1.35it/s]
2|3046|Loss: 0.8120147585868835:  47%|████▋     | 3045/6470 [38:09<42:24,  1.35it/s]
2|3046|Loss: 0.8120147585868835:  47%|████▋     | 3046/6470 [38:10<38:29,  1.48it/s]
2|3047|Loss: 0.8310597538948059:  47%|████▋     | 3046/6470 [38:10<38:29,  1.48it/s]
2|3047|Loss: 0.8310597538948059:  47%|████▋     | 3047/6470 [38:10<41:45,  1.37it/s]
2|3048|Loss: 0.841395914554596:  47%|████▋     | 3047/6470 [38:11<41:45,  1.37it/s] 
2|3048|Loss: 0.841395914554596:  47%|████▋     | 3048/6470 [38:11<39:58,  1.43it/s]
2|3049|Loss: 0.8482387065887451:  47%|████▋     | 3048/6470 [38:11<39:58,  1.43it/s]
2|3049|Loss: 0.8482387065887451:  47%|████▋     | 3049/6470 [38:12<41:24,  1.38it/s]
2|3050|Loss: 0.7884994745254517:  47%|████▋     | 3049/6470 [38:12<41:24,  1.38it/s]
2|3050|Loss: 0.7884994745254517:  47%|████▋     | 3050/6470 [38:13<42:49,  1.33it/s]
2|3051|Loss: 0.843082070350647:  47%|████▋     | 3050/6470 [38:13<42:49,  1.33it/s] 
2|3051|Loss: 0.843082070350647:  47%|████▋     | 3051/6470 [38:13<41:53,  1.36it/s]
2|3052|Loss: 0.7542927861213684:  47%|████▋     | 3051/6470 [38:14<41:53,  1.36it/s]
2|3052|Loss: 0.7542927861213684:  47%|████▋     | 3052/6470 [38:14<39:10,  1.45it/s]
2|3053|Loss: 0.8230321407318115:  47%|████▋     | 3052/6470 [38:14<39:10,  1.45it/s]
2|3053|Loss: 0.8230321407318115:  47%|████▋     | 3053/6470 [38:15<39:47,  1.43it/s]
2|3054|Loss: 0.8133735060691833:  47%|████▋     | 3053/6470 [38:15<39:47,  1.43it/s]
2|3054|Loss: 0.8133735060691833:  47%|████▋     | 3054/6470 [38:15<40:02,  1.42it/s]
2|3055|Loss: 0.7194865345954895:  47%|████▋     | 3054/6470 [38:16<40:02,  1.42it/s]
2|3055|Loss: 0.7194865345954895:  47%|████▋     | 3055/6470 [38:16<39:45,  1.43it/s]
2|3056|Loss: 0.766018271446228:  47%|████▋     | 3055/6470 [38:16<39:45,  1.43it/s] 
2|3056|Loss: 0.766018271446228:  47%|████▋     | 3056/6470 [38:17<39:23,  1.44it/s]
2|3057|Loss: 0.8519876003265381:  47%|████▋     | 3056/6470 [38:17<39:23,  1.44it/s]
2|3057|Loss: 0.8519876003265381:  47%|████▋     | 3057/6470 [38:17<39:33,  1.44it/s]
2|3058|Loss: 0.722162663936615:  47%|████▋     | 3057/6470 [38:18<39:33,  1.44it/s] 
2|3058|Loss: 0.722162663936615:  47%|████▋     | 3058/6470 [38:18<40:57,  1.39it/s]
2|3059|Loss: 0.9108320474624634:  47%|████▋     | 3058/6470 [38:19<40:57,  1.39it/s]
2|3059|Loss: 0.9108320474624634:  47%|████▋     | 3059/6470 [38:19<40:56,  1.39it/s]
2|3060|Loss: 0.8840370774269104:  47%|████▋     | 3059/6470 [38:19<40:56,  1.39it/s]
2|3060|Loss: 0.8840370774269104:  47%|████▋     | 3060/6470 [38:20<41:05,  1.38it/s]
2|3061|Loss: 0.8187841773033142:  47%|████▋     | 3060/6470 [38:20<41:05,  1.38it/s]
2|3061|Loss: 0.8187841773033142:  47%|████▋     | 3061/6470 [38:20<40:20,  1.41it/s]
2|3062|Loss: 0.9971247315406799:  47%|████▋     | 3061/6470 [38:21<40:20,  1.41it/s]
2|3062|Loss: 0.9971247315406799:  47%|████▋     | 3062/6470 [38:21<41:35,  1.37it/s]
2|3063|Loss: 0.8567296862602234:  47%|████▋     | 3062/6470 [38:22<41:35,  1.37it/s]
2|3063|Loss: 0.8567296862602234:  47%|████▋     | 3063/6470 [38:22<43:52,  1.29it/s]
2|3064|Loss: 0.8873267769813538:  47%|████▋     | 3063/6470 [38:22<43:52,  1.29it/s]
2|3064|Loss: 0.8873267769813538:  47%|████▋     | 3064/6470 [38:23<44:45,  1.27it/s]
2|3065|Loss: 0.8804604411125183:  47%|████▋     | 3064/6470 [38:23<44:45,  1.27it/s]
2|3065|Loss: 0.8804604411125183:  47%|████▋     | 3065/6470 [38:24<44:30,  1.28it/s]
2|3066|Loss: 1.0069479942321777:  47%|████▋     | 3065/6470 [38:24<44:30,  1.28it/s]
2|3066|Loss: 1.0069479942321777:  47%|████▋     | 3066/6470 [38:24<43:54,  1.29it/s]
2|3067|Loss: 0.9228335022926331:  47%|████▋     | 3066/6470 [38:25<43:54,  1.29it/s]
2|3067|Loss: 0.9228335022926331:  47%|████▋     | 3067/6470 [38:25<41:56,  1.35it/s]
2|3068|Loss: 0.9749751091003418:  47%|████▋     | 3067/6470 [38:25<41:56,  1.35it/s]
2|3068|Loss: 0.9749751091003418:  47%|████▋     | 3068/6470 [38:26<44:31,  1.27it/s]
2|3069|Loss: 0.8711259961128235:  47%|████▋     | 3068/6470 [38:26<44:31,  1.27it/s]
2|3069|Loss: 0.8711259961128235:  47%|████▋     | 3069/6470 [38:27<45:42,  1.24it/s]
2|3070|Loss: 0.907628059387207:  47%|████▋     | 3069/6470 [38:27<45:42,  1.24it/s] 
2|3070|Loss: 0.907628059387207:  47%|████▋     | 3070/6470 [38:28<46:56,  1.21it/s]
2|3071|Loss: 0.8284257054328918:  47%|████▋     | 3070/6470 [38:28<46:56,  1.21it/s]
2|3071|Loss: 0.8284257054328918:  47%|████▋     | 3071/6470 [38:28<45:25,  1.25it/s]
2|3072|Loss: 0.804601788520813:  47%|████▋     | 3071/6470 [38:29<45:25,  1.25it/s] 
2|3072|Loss: 0.804601788520813:  47%|████▋     | 3072/6470 [38:29<46:11,  1.23it/s]
2|3073|Loss: 0.7804935574531555:  47%|████▋     | 3072/6470 [38:30<46:11,  1.23it/s]
2|3073|Loss: 0.7804935574531555:  47%|████▋     | 3073/6470 [38:30<46:10,  1.23it/s]
2|3074|Loss: 0.7079985737800598:  47%|████▋     | 3073/6470 [38:31<46:10,  1.23it/s]
2|3074|Loss: 0.7079985737800598:  48%|████▊     | 3074/6470 [38:31<47:52,  1.18it/s]
2|3075|Loss: 0.8664683103561401:  48%|████▊     | 3074/6470 [38:31<47:52,  1.18it/s]
2|3075|Loss: 0.8664683103561401:  48%|████▊     | 3075/6470 [38:32<45:23,  1.25it/s]
2|3076|Loss: 0.7833300828933716:  48%|████▊     | 3075/6470 [38:32<45:23,  1.25it/s]
2|3076|Loss: 0.7833300828933716:  48%|████▊     | 3076/6470 [38:32<40:20,  1.40it/s]
2|3077|Loss: 0.9811713099479675:  48%|████▊     | 3076/6470 [38:33<40:20,  1.40it/s]
2|3077|Loss: 0.9811713099479675:  48%|████▊     | 3077/6470 [38:33<40:48,  1.39it/s]
2|3078|Loss: 0.9427588582038879:  48%|████▊     | 3077/6470 [38:33<40:48,  1.39it/s]
2|3078|Loss: 0.9427588582038879:  48%|████▊     | 3078/6470 [38:34<40:25,  1.40it/s]
2|3079|Loss: 0.9137620329856873:  48%|████▊     | 3078/6470 [38:34<40:25,  1.40it/s]
2|3079|Loss: 0.9137620329856873:  48%|████▊     | 3079/6470 [38:34<39:29,  1.43it/s]
2|3080|Loss: 0.9173009991645813:  48%|████▊     | 3079/6470 [38:35<39:29,  1.43it/s]
2|3080|Loss: 0.9173009991645813:  48%|████▊     | 3080/6470 [38:35<37:17,  1.51it/s]
2|3081|Loss: 0.8385103344917297:  48%|████▊     | 3080/6470 [38:35<37:17,  1.51it/s]
2|3081|Loss: 0.8385103344917297:  48%|████▊     | 3081/6470 [38:36<40:17,  1.40it/s]
2|3082|Loss: 0.9450731873512268:  48%|████▊     | 3081/6470 [38:36<40:17,  1.40it/s]
2|3082|Loss: 0.9450731873512268:  48%|████▊     | 3082/6470 [38:37<44:02,  1.28it/s]
2|3083|Loss: 0.6887654066085815:  48%|████▊     | 3082/6470 [38:37<44:02,  1.28it/s]
2|3083|Loss: 0.6887654066085815:  48%|████▊     | 3083/6470 [38:37<43:31,  1.30it/s]
2|3084|Loss: 0.9249918460845947:  48%|████▊     | 3083/6470 [38:38<43:31,  1.30it/s]
2|3084|Loss: 0.9249918460845947:  48%|████▊     | 3084/6470 [38:38<42:08,  1.34it/s]
2|3085|Loss: 0.8621571660041809:  48%|████▊     | 3084/6470 [38:38<42:08,  1.34it/s]
2|3085|Loss: 0.8621571660041809:  48%|████▊     | 3085/6470 [38:39<39:30,  1.43it/s]
2|3086|Loss: 0.9369348287582397:  48%|████▊     | 3085/6470 [38:39<39:30,  1.43it/s]
2|3086|Loss: 0.9369348287582397:  48%|████▊     | 3086/6470 [38:39<39:06,  1.44it/s]
2|3087|Loss: 1.0018161535263062:  48%|████▊     | 3086/6470 [38:40<39:06,  1.44it/s]
2|3087|Loss: 1.0018161535263062:  48%|████▊     | 3087/6470 [38:40<40:19,  1.40it/s]
2|3088|Loss: 0.8116524815559387:  48%|████▊     | 3087/6470 [38:41<40:19,  1.40it/s]
2|3088|Loss: 0.8116524815559387:  48%|████▊     | 3088/6470 [38:41<42:00,  1.34it/s]
2|3089|Loss: 0.8843644261360168:  48%|████▊     | 3088/6470 [38:41<42:00,  1.34it/s]
2|3089|Loss: 0.8843644261360168:  48%|████▊     | 3089/6470 [38:42<42:52,  1.31it/s]
2|3090|Loss: 0.7856127619743347:  48%|████▊     | 3089/6470 [38:42<42:52,  1.31it/s]
2|3090|Loss: 0.7856127619743347:  48%|████▊     | 3090/6470 [38:42<42:50,  1.31it/s]
2|3091|Loss: 0.7736883759498596:  48%|████▊     | 3090/6470 [38:43<42:50,  1.31it/s]
2|3091|Loss: 0.7736883759498596:  48%|████▊     | 3091/6470 [38:43<42:02,  1.34it/s]
2|3092|Loss: 0.7529793381690979:  48%|████▊     | 3091/6470 [38:44<42:02,  1.34it/s]
2|3092|Loss: 0.7529793381690979:  48%|████▊     | 3092/6470 [38:44<44:07,  1.28it/s]
2|3093|Loss: 1.011892557144165:  48%|████▊     | 3092/6470 [38:44<44:07,  1.28it/s] 
2|3093|Loss: 1.011892557144165:  48%|████▊     | 3093/6470 [38:45<42:28,  1.33it/s]
2|3094|Loss: 0.8852944374084473:  48%|████▊     | 3093/6470 [38:45<42:28,  1.33it/s]
2|3094|Loss: 0.8852944374084473:  48%|████▊     | 3094/6470 [38:45<40:43,  1.38it/s]
2|3095|Loss: 0.7237513065338135:  48%|████▊     | 3094/6470 [38:46<40:43,  1.38it/s]
2|3095|Loss: 0.7237513065338135:  48%|████▊     | 3095/6470 [38:46<43:40,  1.29it/s]
2|3096|Loss: 0.9025545716285706:  48%|████▊     | 3095/6470 [38:47<43:40,  1.29it/s]
2|3096|Loss: 0.9025545716285706:  48%|████▊     | 3096/6470 [38:47<43:38,  1.29it/s]
2|3097|Loss: 0.7814632058143616:  48%|████▊     | 3096/6470 [38:47<43:38,  1.29it/s]
2|3097|Loss: 0.7814632058143616:  48%|████▊     | 3097/6470 [38:48<43:39,  1.29it/s]
2|3098|Loss: 0.7096689939498901:  48%|████▊     | 3097/6470 [38:48<43:39,  1.29it/s]
2|3098|Loss: 0.7096689939498901:  48%|████▊     | 3098/6470 [38:49<46:05,  1.22it/s]
2|3099|Loss: 0.872295618057251:  48%|████▊     | 3098/6470 [38:49<46:05,  1.22it/s] 
2|3099|Loss: 0.872295618057251:  48%|████▊     | 3099/6470 [38:49<44:36,  1.26it/s]
2|3100|Loss: 0.806640625:  48%|████▊     | 3099/6470 [38:50<44:36,  1.26it/s]      
2|3100|Loss: 0.806640625:  48%|████▊     | 3100/6470 [38:50<43:53,  1.28it/s]
2|3101|Loss: 0.9259107708930969:  48%|████▊     | 3100/6470 [38:51<43:53,  1.28it/s]
2|3101|Loss: 0.9259107708930969:  48%|████▊     | 3101/6470 [38:51<41:37,  1.35it/s]
2|3102|Loss: 0.7833245992660522:  48%|████▊     | 3101/6470 [38:51<41:37,  1.35it/s]
2|3102|Loss: 0.7833245992660522:  48%|████▊     | 3102/6470 [38:52<41:27,  1.35it/s]
2|3103|Loss: 0.857718825340271:  48%|████▊     | 3102/6470 [38:52<41:27,  1.35it/s] 
2|3103|Loss: 0.857718825340271:  48%|████▊     | 3103/6470 [38:52<38:00,  1.48it/s]
2|3104|Loss: 0.926676332950592:  48%|████▊     | 3103/6470 [38:52<38:00,  1.48it/s]
2|3104|Loss: 0.926676332950592:  48%|████▊     | 3104/6470 [38:53<36:05,  1.55it/s]
2|3105|Loss: 0.7808278799057007:  48%|████▊     | 3104/6470 [38:53<36:05,  1.55it/s]
2|3105|Loss: 0.7808278799057007:  48%|████▊     | 3105/6470 [38:54<39:56,  1.40it/s]
2|3106|Loss: 0.9122216105461121:  48%|████▊     | 3105/6470 [38:54<39:56,  1.40it/s]
2|3106|Loss: 0.9122216105461121:  48%|████▊     | 3106/6470 [38:54<40:19,  1.39it/s]
2|3107|Loss: 0.7299022078514099:  48%|████▊     | 3106/6470 [38:55<40:19,  1.39it/s]
2|3107|Loss: 0.7299022078514099:  48%|████▊     | 3107/6470 [38:55<40:58,  1.37it/s]
2|3108|Loss: 0.9260233640670776:  48%|████▊     | 3107/6470 [38:55<40:58,  1.37it/s]
2|3108|Loss: 0.9260233640670776:  48%|████▊     | 3108/6470 [38:56<42:07,  1.33it/s]
2|3109|Loss: 0.889992356300354:  48%|████▊     | 3108/6470 [38:56<42:07,  1.33it/s] 
2|3109|Loss: 0.889992356300354:  48%|████▊     | 3109/6470 [38:57<41:26,  1.35it/s]
2|3110|Loss: 0.7761391997337341:  48%|████▊     | 3109/6470 [38:57<41:26,  1.35it/s]
2|3110|Loss: 0.7761391997337341:  48%|████▊     | 3110/6470 [38:57<42:25,  1.32it/s]
2|3111|Loss: 0.8581823110580444:  48%|████▊     | 3110/6470 [38:58<42:25,  1.32it/s]
2|3111|Loss: 0.8581823110580444:  48%|████▊     | 3111/6470 [38:58<41:34,  1.35it/s]
2|3112|Loss: 0.7622309923171997:  48%|████▊     | 3111/6470 [38:58<41:34,  1.35it/s]
2|3112|Loss: 0.7622309923171997:  48%|████▊     | 3112/6470 [38:59<39:51,  1.40it/s]
2|3113|Loss: 1.0940028429031372:  48%|████▊     | 3112/6470 [38:59<39:51,  1.40it/s]
2|3113|Loss: 1.0940028429031372:  48%|████▊     | 3113/6470 [39:00<42:41,  1.31it/s]
2|3114|Loss: 0.8473758101463318:  48%|████▊     | 3113/6470 [39:00<42:41,  1.31it/s]
2|3114|Loss: 0.8473758101463318:  48%|████▊     | 3114/6470 [39:00<44:56,  1.24it/s]
2|3115|Loss: 0.8025820255279541:  48%|████▊     | 3114/6470 [39:01<44:56,  1.24it/s]
2|3115|Loss: 0.8025820255279541:  48%|████▊     | 3115/6470 [39:01<43:20,  1.29it/s]
2|3116|Loss: 0.8917027115821838:  48%|████▊     | 3115/6470 [39:02<43:20,  1.29it/s]
2|3116|Loss: 0.8917027115821838:  48%|████▊     | 3116/6470 [39:02<42:35,  1.31it/s]
2|3117|Loss: 1.0001158714294434:  48%|████▊     | 3116/6470 [39:02<42:35,  1.31it/s]
2|3117|Loss: 1.0001158714294434:  48%|████▊     | 3117/6470 [39:03<41:27,  1.35it/s]
2|3118|Loss: 0.8497602939605713:  48%|████▊     | 3117/6470 [39:03<41:27,  1.35it/s]
2|3118|Loss: 0.8497602939605713:  48%|████▊     | 3118/6470 [39:03<40:22,  1.38it/s]
2|3119|Loss: 0.8400323390960693:  48%|████▊     | 3118/6470 [39:04<40:22,  1.38it/s]
2|3119|Loss: 0.8400323390960693:  48%|████▊     | 3119/6470 [39:04<40:49,  1.37it/s]
2|3120|Loss: 0.8547377586364746:  48%|████▊     | 3119/6470 [39:04<40:49,  1.37it/s]
2|3120|Loss: 0.8547377586364746:  48%|████▊     | 3120/6470 [39:05<41:50,  1.33it/s]
2|3121|Loss: 0.7971201539039612:  48%|████▊     | 3120/6470 [39:05<41:50,  1.33it/s]
2|3121|Loss: 0.7971201539039612:  48%|████▊     | 3121/6470 [39:06<42:46,  1.30it/s]
2|3122|Loss: 0.9401512742042542:  48%|████▊     | 3121/6470 [39:06<42:46,  1.30it/s]
2|3122|Loss: 0.9401512742042542:  48%|████▊     | 3122/6470 [39:06<40:38,  1.37it/s]
2|3123|Loss: 0.757144570350647:  48%|████▊     | 3122/6470 [39:07<40:38,  1.37it/s] 
2|3123|Loss: 0.757144570350647:  48%|████▊     | 3123/6470 [39:07<39:49,  1.40it/s]
2|3124|Loss: 0.9516433477401733:  48%|████▊     | 3123/6470 [39:07<39:49,  1.40it/s]
2|3124|Loss: 0.9516433477401733:  48%|████▊     | 3124/6470 [39:08<41:10,  1.35it/s]
2|3125|Loss: 0.7719578146934509:  48%|████▊     | 3124/6470 [39:08<41:10,  1.35it/s]
2|3125|Loss: 0.7719578146934509:  48%|████▊     | 3125/6470 [39:09<44:10,  1.26it/s]
2|3126|Loss: 0.7948599457740784:  48%|████▊     | 3125/6470 [39:09<44:10,  1.26it/s]
2|3126|Loss: 0.7948599457740784:  48%|████▊     | 3126/6470 [39:10<45:14,  1.23it/s]
2|3127|Loss: 0.8772988319396973:  48%|████▊     | 3126/6470 [39:10<45:14,  1.23it/s]
2|3127|Loss: 0.8772988319396973:  48%|████▊     | 3127/6470 [39:10<45:12,  1.23it/s]
2|3128|Loss: 0.9685507416725159:  48%|████▊     | 3127/6470 [39:11<45:12,  1.23it/s]
2|3128|Loss: 0.9685507416725159:  48%|████▊     | 3128/6470 [39:11<44:23,  1.25it/s]
2|3129|Loss: 0.7971281409263611:  48%|████▊     | 3128/6470 [39:12<44:23,  1.25it/s]
2|3129|Loss: 0.7971281409263611:  48%|████▊     | 3129/6470 [39:12<43:43,  1.27it/s]
2|3130|Loss: 0.8650282621383667:  48%|████▊     | 3129/6470 [39:12<43:43,  1.27it/s]
2|3130|Loss: 0.8650282621383667:  48%|████▊     | 3130/6470 [39:13<43:10,  1.29it/s]
2|3131|Loss: 0.8282058835029602:  48%|████▊     | 3130/6470 [39:13<43:10,  1.29it/s]
2|3131|Loss: 0.8282058835029602:  48%|████▊     | 3131/6470 [39:13<44:13,  1.26it/s]
2|3132|Loss: 0.7956780791282654:  48%|████▊     | 3131/6470 [39:14<44:13,  1.26it/s]
2|3132|Loss: 0.7956780791282654:  48%|████▊     | 3132/6470 [39:14<42:45,  1.30it/s]
2|3133|Loss: 0.9240465760231018:  48%|████▊     | 3132/6470 [39:15<42:45,  1.30it/s]
2|3133|Loss: 0.9240465760231018:  48%|████▊     | 3133/6470 [39:15<42:12,  1.32it/s]
2|3134|Loss: 0.796257734298706:  48%|████▊     | 3133/6470 [39:15<42:12,  1.32it/s] 
2|3134|Loss: 0.796257734298706:  48%|████▊     | 3134/6470 [39:16<39:15,  1.42it/s]
2|3135|Loss: 0.8158632516860962:  48%|████▊     | 3134/6470 [39:16<39:15,  1.42it/s]
2|3135|Loss: 0.8158632516860962:  48%|████▊     | 3135/6470 [39:16<39:47,  1.40it/s]
2|3136|Loss: 0.8367010951042175:  48%|████▊     | 3135/6470 [39:17<39:47,  1.40it/s]
2|3136|Loss: 0.8367010951042175:  48%|████▊     | 3136/6470 [39:17<41:14,  1.35it/s]
2|3137|Loss: 0.7667190432548523:  48%|████▊     | 3136/6470 [39:17<41:14,  1.35it/s]
2|3137|Loss: 0.7667190432548523:  48%|████▊     | 3137/6470 [39:18<40:19,  1.38it/s]
2|3138|Loss: 0.869073748588562:  48%|████▊     | 3137/6470 [39:18<40:19,  1.38it/s] 
2|3138|Loss: 0.869073748588562:  49%|████▊     | 3138/6470 [39:18<40:21,  1.38it/s]
2|3139|Loss: 0.8185216188430786:  49%|████▊     | 3138/6470 [39:19<40:21,  1.38it/s]
2|3139|Loss: 0.8185216188430786:  49%|████▊     | 3139/6470 [39:19<43:08,  1.29it/s]
2|3140|Loss: 0.906065821647644:  49%|████▊     | 3139/6470 [39:20<43:08,  1.29it/s] 
2|3140|Loss: 0.906065821647644:  49%|████▊     | 3140/6470 [39:20<43:58,  1.26it/s]
2|3141|Loss: 0.8668791651725769:  49%|████▊     | 3140/6470 [39:21<43:58,  1.26it/s]
2|3141|Loss: 0.8668791651725769:  49%|████▊     | 3141/6470 [39:21<44:08,  1.26it/s]
2|3142|Loss: 0.8267125487327576:  49%|████▊     | 3141/6470 [39:21<44:08,  1.26it/s]
2|3142|Loss: 0.8267125487327576:  49%|████▊     | 3142/6470 [39:22<41:50,  1.33it/s]
2|3143|Loss: 0.9643663763999939:  49%|████▊     | 3142/6470 [39:22<41:50,  1.33it/s]
2|3143|Loss: 0.9643663763999939:  49%|████▊     | 3143/6470 [39:22<39:39,  1.40it/s]
2|3144|Loss: 0.8487963080406189:  49%|████▊     | 3143/6470 [39:23<39:39,  1.40it/s]
2|3144|Loss: 0.8487963080406189:  49%|████▊     | 3144/6470 [39:23<37:44,  1.47it/s]
2|3145|Loss: 0.8117210268974304:  49%|████▊     | 3144/6470 [39:23<37:44,  1.47it/s]
2|3145|Loss: 0.8117210268974304:  49%|████▊     | 3145/6470 [39:24<40:17,  1.38it/s]
2|3146|Loss: 0.8221279382705688:  49%|████▊     | 3145/6470 [39:24<40:17,  1.38it/s]
2|3146|Loss: 0.8221279382705688:  49%|████▊     | 3146/6470 [39:24<38:39,  1.43it/s]
2|3147|Loss: 0.7902748584747314:  49%|████▊     | 3146/6470 [39:25<38:39,  1.43it/s]
2|3147|Loss: 0.7902748584747314:  49%|████▊     | 3147/6470 [39:25<39:25,  1.40it/s]
2|3148|Loss: 0.8433917164802551:  49%|████▊     | 3147/6470 [39:25<39:25,  1.40it/s]
2|3148|Loss: 0.8433917164802551:  49%|████▊     | 3148/6470 [39:26<39:40,  1.40it/s]
2|3149|Loss: 0.7699620127677917:  49%|████▊     | 3148/6470 [39:26<39:40,  1.40it/s]
2|3149|Loss: 0.7699620127677917:  49%|████▊     | 3149/6470 [39:27<40:03,  1.38it/s]
2|3150|Loss: 1.0565351247787476:  49%|████▊     | 3149/6470 [39:27<40:03,  1.38it/s]
2|3150|Loss: 1.0565351247787476:  49%|████▊     | 3150/6470 [39:27<43:13,  1.28it/s]
2|3151|Loss: 0.7677725553512573:  49%|████▊     | 3150/6470 [39:28<43:13,  1.28it/s]
2|3151|Loss: 0.7677725553512573:  49%|████▊     | 3151/6470 [39:28<43:32,  1.27it/s]
2|3152|Loss: 0.805472195148468:  49%|████▊     | 3151/6470 [39:29<43:32,  1.27it/s] 
2|3152|Loss: 0.805472195148468:  49%|████▊     | 3152/6470 [39:29<42:23,  1.30it/s]
2|3153|Loss: 0.8544432520866394:  49%|████▊     | 3152/6470 [39:29<42:23,  1.30it/s]
2|3153|Loss: 0.8544432520866394:  49%|████▊     | 3153/6470 [39:30<42:32,  1.30it/s]
2|3154|Loss: 0.7513978481292725:  49%|████▊     | 3153/6470 [39:30<42:32,  1.30it/s]
2|3154|Loss: 0.7513978481292725:  49%|████▊     | 3154/6470 [39:30<41:41,  1.33it/s]
2|3155|Loss: 0.6795982718467712:  49%|████▊     | 3154/6470 [39:31<41:41,  1.33it/s]
2|3155|Loss: 0.6795982718467712:  49%|████▉     | 3155/6470 [39:31<38:11,  1.45it/s]
2|3156|Loss: 0.9026357531547546:  49%|████▉     | 3155/6470 [39:31<38:11,  1.45it/s]
2|3156|Loss: 0.9026357531547546:  49%|████▉     | 3156/6470 [39:32<37:08,  1.49it/s]
2|3157|Loss: 0.8647111654281616:  49%|████▉     | 3156/6470 [39:32<37:08,  1.49it/s]
2|3157|Loss: 0.8647111654281616:  49%|████▉     | 3157/6470 [39:32<40:01,  1.38it/s]
2|3158|Loss: 0.8250405788421631:  49%|████▉     | 3157/6470 [39:33<40:01,  1.38it/s]
2|3158|Loss: 0.8250405788421631:  49%|████▉     | 3158/6470 [39:33<37:12,  1.48it/s]
2|3159|Loss: 0.7861433029174805:  49%|████▉     | 3158/6470 [39:33<37:12,  1.48it/s]
2|3159|Loss: 0.7861433029174805:  49%|████▉     | 3159/6470 [39:34<38:56,  1.42it/s]
2|3160|Loss: 0.905521810054779:  49%|████▉     | 3159/6470 [39:34<38:56,  1.42it/s] 
2|3160|Loss: 0.905521810054779:  49%|████▉     | 3160/6470 [39:35<39:59,  1.38it/s]
2|3161|Loss: 0.7555012702941895:  49%|████▉     | 3160/6470 [39:35<39:59,  1.38it/s]
2|3161|Loss: 0.7555012702941895:  49%|████▉     | 3161/6470 [39:35<41:07,  1.34it/s]
2|3162|Loss: 0.8115153908729553:  49%|████▉     | 3161/6470 [39:36<41:07,  1.34it/s]
2|3162|Loss: 0.8115153908729553:  49%|████▉     | 3162/6470 [39:36<42:25,  1.30it/s]
2|3163|Loss: 0.9170030951499939:  49%|████▉     | 3162/6470 [39:37<42:25,  1.30it/s]
2|3163|Loss: 0.9170030951499939:  49%|████▉     | 3163/6470 [39:37<45:59,  1.20it/s]
2|3164|Loss: 0.8367142081260681:  49%|████▉     | 3163/6470 [39:38<45:59,  1.20it/s]
2|3164|Loss: 0.8367142081260681:  49%|████▉     | 3164/6470 [39:38<48:00,  1.15it/s]
2|3165|Loss: 0.6762362122535706:  49%|████▉     | 3164/6470 [39:39<48:00,  1.15it/s]
2|3165|Loss: 0.6762362122535706:  49%|████▉     | 3165/6470 [39:39<48:07,  1.14it/s]
2|3166|Loss: 0.6929370164871216:  49%|████▉     | 3165/6470 [39:40<48:07,  1.14it/s]
2|3166|Loss: 0.6929370164871216:  49%|████▉     | 3166/6470 [39:40<47:58,  1.15it/s]
2|3167|Loss: 0.8914571404457092:  49%|████▉     | 3166/6470 [39:40<47:58,  1.15it/s]
2|3167|Loss: 0.8914571404457092:  49%|████▉     | 3167/6470 [39:41<45:19,  1.21it/s]
2|3168|Loss: 0.7494788765907288:  49%|████▉     | 3167/6470 [39:41<45:19,  1.21it/s]
2|3168|Loss: 0.7494788765907288:  49%|████▉     | 3168/6470 [39:41<44:34,  1.23it/s]
2|3169|Loss: 0.892310380935669:  49%|████▉     | 3168/6470 [39:42<44:34,  1.23it/s] 
2|3169|Loss: 0.892310380935669:  49%|████▉     | 3169/6470 [39:42<46:17,  1.19it/s]
2|3170|Loss: 0.9438555240631104:  49%|████▉     | 3169/6470 [39:43<46:17,  1.19it/s]
2|3170|Loss: 0.9438555240631104:  49%|████▉     | 3170/6470 [39:43<46:06,  1.19it/s]
2|3171|Loss: 0.7438362240791321:  49%|████▉     | 3170/6470 [39:44<46:06,  1.19it/s]
2|3171|Loss: 0.7438362240791321:  49%|████▉     | 3171/6470 [39:44<41:40,  1.32it/s]
2|3172|Loss: 0.9236338138580322:  49%|████▉     | 3171/6470 [39:44<41:40,  1.32it/s]
2|3172|Loss: 0.9236338138580322:  49%|████▉     | 3172/6470 [39:44<38:15,  1.44it/s]
2|3173|Loss: 0.7500320076942444:  49%|████▉     | 3172/6470 [39:45<38:15,  1.44it/s]
2|3173|Loss: 0.7500320076942444:  49%|████▉     | 3173/6470 [39:45<38:40,  1.42it/s]
2|3174|Loss: 0.9566775560379028:  49%|████▉     | 3173/6470 [39:45<38:40,  1.42it/s]
2|3174|Loss: 0.9566775560379028:  49%|████▉     | 3174/6470 [39:46<40:03,  1.37it/s]
2|3175|Loss: 0.8551380038261414:  49%|████▉     | 3174/6470 [39:46<40:03,  1.37it/s]
2|3175|Loss: 0.8551380038261414:  49%|████▉     | 3175/6470 [39:46<39:30,  1.39it/s]
2|3176|Loss: 1.0183074474334717:  49%|████▉     | 3175/6470 [39:47<39:30,  1.39it/s]
2|3176|Loss: 1.0183074474334717:  49%|████▉     | 3176/6470 [39:47<40:26,  1.36it/s]
2|3177|Loss: 0.9616273045539856:  49%|████▉     | 3176/6470 [39:48<40:26,  1.36it/s]
2|3177|Loss: 0.9616273045539856:  49%|████▉     | 3177/6470 [39:48<39:10,  1.40it/s]
2|3178|Loss: 0.8561024069786072:  49%|████▉     | 3177/6470 [39:48<39:10,  1.40it/s]
2|3178|Loss: 0.8561024069786072:  49%|████▉     | 3178/6470 [39:49<40:15,  1.36it/s]
2|3179|Loss: 0.7368428707122803:  49%|████▉     | 3178/6470 [39:49<40:15,  1.36it/s]
2|3179|Loss: 0.7368428707122803:  49%|████▉     | 3179/6470 [39:50<44:35,  1.23it/s]
2|3180|Loss: 0.836539089679718:  49%|████▉     | 3179/6470 [39:50<44:35,  1.23it/s] 
2|3180|Loss: 0.836539089679718:  49%|████▉     | 3180/6470 [39:50<42:04,  1.30it/s]
2|3181|Loss: 0.8109287023544312:  49%|████▉     | 3180/6470 [39:51<42:04,  1.30it/s]
2|3181|Loss: 0.8109287023544312:  49%|████▉     | 3181/6470 [39:51<43:28,  1.26it/s]
2|3182|Loss: 0.7714389562606812:  49%|████▉     | 3181/6470 [39:52<43:28,  1.26it/s]
2|3182|Loss: 0.7714389562606812:  49%|████▉     | 3182/6470 [39:52<41:15,  1.33it/s]
2|3183|Loss: 0.6736518144607544:  49%|████▉     | 3182/6470 [39:52<41:15,  1.33it/s]
2|3183|Loss: 0.6736518144607544:  49%|████▉     | 3183/6470 [39:52<38:50,  1.41it/s]
2|3184|Loss: 0.8076603412628174:  49%|████▉     | 3183/6470 [39:53<38:50,  1.41it/s]
2|3184|Loss: 0.8076603412628174:  49%|████▉     | 3184/6470 [39:53<39:17,  1.39it/s]
2|3185|Loss: 0.8347271084785461:  49%|████▉     | 3184/6470 [39:54<39:17,  1.39it/s]
2|3185|Loss: 0.8347271084785461:  49%|████▉     | 3185/6470 [39:54<41:46,  1.31it/s]
2|3186|Loss: 0.7430376410484314:  49%|████▉     | 3185/6470 [39:54<41:46,  1.31it/s]
2|3186|Loss: 0.7430376410484314:  49%|████▉     | 3186/6470 [39:55<40:54,  1.34it/s]
2|3187|Loss: 0.8916452527046204:  49%|████▉     | 3186/6470 [39:55<40:54,  1.34it/s]
2|3187|Loss: 0.8916452527046204:  49%|████▉     | 3187/6470 [39:56<42:22,  1.29it/s]
2|3188|Loss: 0.6900907158851624:  49%|████▉     | 3187/6470 [39:56<42:22,  1.29it/s]
2|3188|Loss: 0.6900907158851624:  49%|████▉     | 3188/6470 [39:57<46:08,  1.19it/s]
2|3189|Loss: 0.901577353477478:  49%|████▉     | 3188/6470 [39:57<46:08,  1.19it/s] 
2|3189|Loss: 0.901577353477478:  49%|████▉     | 3189/6470 [39:57<43:57,  1.24it/s]
2|3190|Loss: 0.9005257487297058:  49%|████▉     | 3189/6470 [39:58<43:57,  1.24it/s]
2|3190|Loss: 0.9005257487297058:  49%|████▉     | 3190/6470 [39:58<43:20,  1.26it/s]
2|3191|Loss: 0.77580326795578:  49%|████▉     | 3190/6470 [39:59<43:20,  1.26it/s]  
2|3191|Loss: 0.77580326795578:  49%|████▉     | 3191/6470 [39:59<44:10,  1.24it/s]
2|3192|Loss: 0.7571703195571899:  49%|████▉     | 3191/6470 [39:59<44:10,  1.24it/s]
2|3192|Loss: 0.7571703195571899:  49%|████▉     | 3192/6470 [40:00<44:06,  1.24it/s]
2|3193|Loss: 0.8490284085273743:  49%|████▉     | 3192/6470 [40:00<44:06,  1.24it/s]
2|3193|Loss: 0.8490284085273743:  49%|████▉     | 3193/6470 [40:00<42:37,  1.28it/s]
2|3194|Loss: 0.9456899762153625:  49%|████▉     | 3193/6470 [40:01<42:37,  1.28it/s]
2|3194|Loss: 0.9456899762153625:  49%|████▉     | 3194/6470 [40:01<41:00,  1.33it/s]
2|3195|Loss: 0.8757115602493286:  49%|████▉     | 3194/6470 [40:02<41:00,  1.33it/s]
2|3195|Loss: 0.8757115602493286:  49%|████▉     | 3195/6470 [40:02<42:33,  1.28it/s]
2|3196|Loss: 0.8762490749359131:  49%|████▉     | 3195/6470 [40:02<42:33,  1.28it/s]
2|3196|Loss: 0.8762490749359131:  49%|████▉     | 3196/6470 [40:03<41:25,  1.32it/s]
2|3197|Loss: 0.9203164577484131:  49%|████▉     | 3196/6470 [40:03<41:25,  1.32it/s]
2|3197|Loss: 0.9203164577484131:  49%|████▉     | 3197/6470 [40:04<42:17,  1.29it/s]
2|3198|Loss: 0.7473824620246887:  49%|████▉     | 3197/6470 [40:04<42:17,  1.29it/s]
2|3198|Loss: 0.7473824620246887:  49%|████▉     | 3198/6470 [40:04<42:18,  1.29it/s]
2|3199|Loss: 0.9738913178443909:  49%|████▉     | 3198/6470 [40:05<42:18,  1.29it/s]
2|3199|Loss: 0.9738913178443909:  49%|████▉     | 3199/6470 [40:05<41:03,  1.33it/s]
2|3200|Loss: 0.9267938137054443:  49%|████▉     | 3199/6470 [40:05<41:03,  1.33it/s]
2|3200|Loss: 0.9267938137054443:  49%|████▉     | 3200/6470 [40:06<41:15,  1.32it/s]
2|3201|Loss: 0.9418939352035522:  49%|████▉     | 3200/6470 [40:06<41:15,  1.32it/s]
2|3201|Loss: 0.9418939352035522:  49%|████▉     | 3201/6470 [40:07<43:33,  1.25it/s]
2|3202|Loss: 0.9546789526939392:  49%|████▉     | 3201/6470 [40:07<43:33,  1.25it/s]
2|3202|Loss: 0.9546789526939392:  49%|████▉     | 3202/6470 [40:08<45:43,  1.19it/s]
2|3203|Loss: 0.7341136336326599:  49%|████▉     | 3202/6470 [40:08<45:43,  1.19it/s]
2|3203|Loss: 0.7341136336326599:  50%|████▉     | 3203/6470 [40:08<41:32,  1.31it/s]
2|3204|Loss: 0.8043912053108215:  50%|████▉     | 3203/6470 [40:08<41:32,  1.31it/s]
2|3204|Loss: 0.8043912053108215:  50%|████▉     | 3204/6470 [40:09<37:33,  1.45it/s]
2|3205|Loss: 0.7870638966560364:  50%|████▉     | 3204/6470 [40:09<37:33,  1.45it/s]
2|3205|Loss: 0.7870638966560364:  50%|████▉     | 3205/6470 [40:09<37:17,  1.46it/s]
2|3206|Loss: 0.6744946837425232:  50%|████▉     | 3205/6470 [40:10<37:17,  1.46it/s]
2|3206|Loss: 0.6744946837425232:  50%|████▉     | 3206/6470 [40:10<38:26,  1.42it/s]
2|3207|Loss: 0.800689697265625:  50%|████▉     | 3206/6470 [40:11<38:26,  1.42it/s] 
2|3207|Loss: 0.800689697265625:  50%|████▉     | 3207/6470 [40:11<39:44,  1.37it/s]
2|3208|Loss: 0.7330617904663086:  50%|████▉     | 3207/6470 [40:11<39:44,  1.37it/s]
2|3208|Loss: 0.7330617904663086:  50%|████▉     | 3208/6470 [40:12<42:55,  1.27it/s]
2|3209|Loss: 0.7038156986236572:  50%|████▉     | 3208/6470 [40:12<42:55,  1.27it/s]
2|3209|Loss: 0.7038156986236572:  50%|████▉     | 3209/6470 [40:13<43:37,  1.25it/s]
2|3210|Loss: 0.814958930015564:  50%|████▉     | 3209/6470 [40:13<43:37,  1.25it/s] 
2|3210|Loss: 0.814958930015564:  50%|████▉     | 3210/6470 [40:13<39:21,  1.38it/s]
2|3211|Loss: 0.7961643934249878:  50%|████▉     | 3210/6470 [40:14<39:21,  1.38it/s]
2|3211|Loss: 0.7961643934249878:  50%|████▉     | 3211/6470 [40:14<38:00,  1.43it/s]
2|3212|Loss: 0.9128645062446594:  50%|████▉     | 3211/6470 [40:14<38:00,  1.43it/s]
2|3212|Loss: 0.9128645062446594:  50%|████▉     | 3212/6470 [40:15<39:16,  1.38it/s]
2|3213|Loss: 0.8415515422821045:  50%|████▉     | 3212/6470 [40:15<39:16,  1.38it/s]
2|3213|Loss: 0.8415515422821045:  50%|████▉     | 3213/6470 [40:15<38:25,  1.41it/s]
2|3214|Loss: 0.9613751769065857:  50%|████▉     | 3213/6470 [40:16<38:25,  1.41it/s]
2|3214|Loss: 0.9613751769065857:  50%|████▉     | 3214/6470 [40:16<39:13,  1.38it/s]
2|3215|Loss: 0.8485937118530273:  50%|████▉     | 3214/6470 [40:16<39:13,  1.38it/s]
2|3215|Loss: 0.8485937118530273:  50%|████▉     | 3215/6470 [40:17<38:41,  1.40it/s]
2|3216|Loss: 0.741202175617218:  50%|████▉     | 3215/6470 [40:17<38:41,  1.40it/s] 
2|3216|Loss: 0.741202175617218:  50%|████▉     | 3216/6470 [40:18<39:45,  1.36it/s]
2|3217|Loss: 0.9009686708450317:  50%|████▉     | 3216/6470 [40:18<39:45,  1.36it/s]
2|3217|Loss: 0.9009686708450317:  50%|████▉     | 3217/6470 [40:18<42:56,  1.26it/s]
2|3218|Loss: 0.8480435013771057:  50%|████▉     | 3217/6470 [40:19<42:56,  1.26it/s]
2|3218|Loss: 0.8480435013771057:  50%|████▉     | 3218/6470 [40:19<42:33,  1.27it/s]
2|3219|Loss: 0.89810711145401:  50%|████▉     | 3218/6470 [40:20<42:33,  1.27it/s]  
2|3219|Loss: 0.89810711145401:  50%|████▉     | 3219/6470 [40:20<40:52,  1.33it/s]
2|3220|Loss: 0.8805598616600037:  50%|████▉     | 3219/6470 [40:20<40:52,  1.33it/s]
2|3220|Loss: 0.8805598616600037:  50%|████▉     | 3220/6470 [40:21<39:52,  1.36it/s]
2|3221|Loss: 0.9236072301864624:  50%|████▉     | 3220/6470 [40:21<39:52,  1.36it/s]
2|3221|Loss: 0.9236072301864624:  50%|████▉     | 3221/6470 [40:21<37:08,  1.46it/s]
2|3222|Loss: 0.696894109249115:  50%|████▉     | 3221/6470 [40:21<37:08,  1.46it/s] 
2|3222|Loss: 0.696894109249115:  50%|████▉     | 3222/6470 [40:22<33:58,  1.59it/s]
2|3223|Loss: 0.6552956104278564:  50%|████▉     | 3222/6470 [40:22<33:58,  1.59it/s]
2|3223|Loss: 0.6552956104278564:  50%|████▉     | 3223/6470 [40:22<32:13,  1.68it/s]
2|3224|Loss: 0.8092342615127563:  50%|████▉     | 3223/6470 [40:23<32:13,  1.68it/s]
2|3224|Loss: 0.8092342615127563:  50%|████▉     | 3224/6470 [40:23<34:32,  1.57it/s]
2|3225|Loss: 0.9124386310577393:  50%|████▉     | 3224/6470 [40:23<34:32,  1.57it/s]
2|3225|Loss: 0.9124386310577393:  50%|████▉     | 3225/6470 [40:24<34:59,  1.55it/s]
2|3226|Loss: 0.8326098322868347:  50%|████▉     | 3225/6470 [40:24<34:59,  1.55it/s]
2|3226|Loss: 0.8326098322868347:  50%|████▉     | 3226/6470 [40:24<34:29,  1.57it/s]
2|3227|Loss: 0.8196218609809875:  50%|████▉     | 3226/6470 [40:25<34:29,  1.57it/s]
2|3227|Loss: 0.8196218609809875:  50%|████▉     | 3227/6470 [40:25<35:57,  1.50it/s]
2|3228|Loss: 0.8889658451080322:  50%|████▉     | 3227/6470 [40:25<35:57,  1.50it/s]
2|3228|Loss: 0.8889658451080322:  50%|████▉     | 3228/6470 [40:26<38:02,  1.42it/s]
2|3229|Loss: 0.7233832478523254:  50%|████▉     | 3228/6470 [40:26<38:02,  1.42it/s]
2|3229|Loss: 0.7233832478523254:  50%|████▉     | 3229/6470 [40:26<37:50,  1.43it/s]
2|3230|Loss: 0.7189329862594604:  50%|████▉     | 3229/6470 [40:27<37:50,  1.43it/s]
2|3230|Loss: 0.7189329862594604:  50%|████▉     | 3230/6470 [40:27<39:52,  1.35it/s]
2|3231|Loss: 0.8854213356971741:  50%|████▉     | 3230/6470 [40:28<39:52,  1.35it/s]
2|3231|Loss: 0.8854213356971741:  50%|████▉     | 3231/6470 [40:28<37:15,  1.45it/s]
2|3232|Loss: 0.9467525482177734:  50%|████▉     | 3231/6470 [40:28<37:15,  1.45it/s]
2|3232|Loss: 0.9467525482177734:  50%|████▉     | 3232/6470 [40:28<36:06,  1.49it/s]
2|3233|Loss: 0.8176830410957336:  50%|████▉     | 3232/6470 [40:29<36:06,  1.49it/s]
2|3233|Loss: 0.8176830410957336:  50%|████▉     | 3233/6470 [40:29<37:43,  1.43it/s]
2|3234|Loss: 0.8540877103805542:  50%|████▉     | 3233/6470 [40:30<37:43,  1.43it/s]
2|3234|Loss: 0.8540877103805542:  50%|████▉     | 3234/6470 [40:30<38:36,  1.40it/s]
2|3235|Loss: 0.816394031047821:  50%|████▉     | 3234/6470 [40:30<38:36,  1.40it/s] 
2|3235|Loss: 0.816394031047821:  50%|█████     | 3235/6470 [40:30<34:34,  1.56it/s]
2|3236|Loss: 0.9493071436882019:  50%|█████     | 3235/6470 [40:31<34:34,  1.56it/s]
2|3236|Loss: 0.9493071436882019:  50%|█████     | 3236/6470 [40:31<35:41,  1.51it/s]
2|3237|Loss: 0.8832440972328186:  50%|█████     | 3236/6470 [40:32<35:41,  1.51it/s]
2|3237|Loss: 0.8832440972328186:  50%|█████     | 3237/6470 [40:32<36:42,  1.47it/s]
2|3238|Loss: 0.8715702891349792:  50%|█████     | 3237/6470 [40:32<36:42,  1.47it/s]
2|3238|Loss: 0.8715702891349792:  50%|█████     | 3238/6470 [40:33<38:01,  1.42it/s]
2|3239|Loss: 0.8554583787918091:  50%|█████     | 3238/6470 [40:33<38:01,  1.42it/s]
2|3239|Loss: 0.8554583787918091:  50%|█████     | 3239/6470 [40:33<38:01,  1.42it/s]
2|3240|Loss: 0.936567485332489:  50%|█████     | 3239/6470 [40:34<38:01,  1.42it/s] 
2|3240|Loss: 0.936567485332489:  50%|█████     | 3240/6470 [40:34<35:28,  1.52it/s]
2|3241|Loss: 0.8545361757278442:  50%|█████     | 3240/6470 [40:34<35:28,  1.52it/s]
2|3241|Loss: 0.8545361757278442:  50%|█████     | 3241/6470 [40:35<36:52,  1.46it/s]
2|3242|Loss: 0.8265553712844849:  50%|█████     | 3241/6470 [40:35<36:52,  1.46it/s]
2|3242|Loss: 0.8265553712844849:  50%|█████     | 3242/6470 [40:35<35:46,  1.50it/s]
2|3243|Loss: 0.9176554083824158:  50%|█████     | 3242/6470 [40:36<35:46,  1.50it/s]
2|3243|Loss: 0.9176554083824158:  50%|█████     | 3243/6470 [40:36<34:27,  1.56it/s]
2|3244|Loss: 0.9777914881706238:  50%|█████     | 3243/6470 [40:36<34:27,  1.56it/s]
2|3244|Loss: 0.9777914881706238:  50%|█████     | 3244/6470 [40:37<36:25,  1.48it/s]
2|3245|Loss: 0.7855663299560547:  50%|█████     | 3244/6470 [40:37<36:25,  1.48it/s]
2|3245|Loss: 0.7855663299560547:  50%|█████     | 3245/6470 [40:37<38:13,  1.41it/s]
2|3246|Loss: 0.8340111374855042:  50%|█████     | 3245/6470 [40:38<38:13,  1.41it/s]
2|3246|Loss: 0.8340111374855042:  50%|█████     | 3246/6470 [40:38<36:52,  1.46it/s]
2|3247|Loss: 0.8786463141441345:  50%|█████     | 3246/6470 [40:38<36:52,  1.46it/s]
2|3247|Loss: 0.8786463141441345:  50%|█████     | 3247/6470 [40:39<37:51,  1.42it/s]
2|3248|Loss: 0.8183200359344482:  50%|█████     | 3247/6470 [40:39<37:51,  1.42it/s]
2|3248|Loss: 0.8183200359344482:  50%|█████     | 3248/6470 [40:40<39:12,  1.37it/s]
2|3249|Loss: 0.8091182112693787:  50%|█████     | 3248/6470 [40:40<39:12,  1.37it/s]
2|3249|Loss: 0.8091182112693787:  50%|█████     | 3249/6470 [40:40<39:47,  1.35it/s]
2|3250|Loss: 0.60456383228302:  50%|█████     | 3249/6470 [40:41<39:47,  1.35it/s]  
2|3250|Loss: 0.60456383228302:  50%|█████     | 3250/6470 [40:42<57:41,  1.08s/it]
2|3251|Loss: 0.8293184638023376:  50%|█████     | 3250/6470 [40:43<57:41,  1.08s/it]
2|3251|Loss: 0.8293184638023376:  50%|█████     | 3251/6470 [40:43<57:09,  1.07s/it]
2|3252|Loss: 0.8906498551368713:  50%|█████     | 3251/6470 [40:44<57:09,  1.07s/it]
2|3252|Loss: 0.8906498551368713:  50%|█████     | 3252/6470 [40:44<53:54,  1.01s/it]
2|3253|Loss: 0.8982583284378052:  50%|█████     | 3252/6470 [40:45<53:54,  1.01s/it]
2|3253|Loss: 0.8982583284378052:  50%|█████     | 3253/6470 [40:45<52:18,  1.02it/s]
2|3254|Loss: 0.8478657007217407:  50%|█████     | 3253/6470 [40:45<52:18,  1.02it/s]
2|3254|Loss: 0.8478657007217407:  50%|█████     | 3254/6470 [40:46<49:48,  1.08it/s]
2|3255|Loss: 0.8406980633735657:  50%|█████     | 3254/6470 [40:46<49:48,  1.08it/s]
2|3255|Loss: 0.8406980633735657:  50%|█████     | 3255/6470 [40:46<45:29,  1.18it/s]
2|3256|Loss: 0.8394158482551575:  50%|█████     | 3255/6470 [40:47<45:29,  1.18it/s]
2|3256|Loss: 0.8394158482551575:  50%|█████     | 3256/6470 [40:47<43:55,  1.22it/s]
2|3257|Loss: 0.8118407726287842:  50%|█████     | 3256/6470 [40:48<43:55,  1.22it/s]
2|3257|Loss: 0.8118407726287842:  50%|█████     | 3257/6470 [40:48<41:45,  1.28it/s]
2|3258|Loss: 0.8999704122543335:  50%|█████     | 3257/6470 [40:48<41:45,  1.28it/s]
2|3258|Loss: 0.8999704122543335:  50%|█████     | 3258/6470 [40:49<41:12,  1.30it/s]
2|3259|Loss: 0.778689444065094:  50%|█████     | 3258/6470 [40:49<41:12,  1.30it/s] 
2|3259|Loss: 0.778689444065094:  50%|█████     | 3259/6470 [40:49<41:00,  1.30it/s]
2|3260|Loss: 0.9196786880493164:  50%|█████     | 3259/6470 [40:50<41:00,  1.30it/s]
2|3260|Loss: 0.9196786880493164:  50%|█████     | 3260/6470 [40:50<40:45,  1.31it/s]
2|3261|Loss: 0.8047707080841064:  50%|█████     | 3260/6470 [40:51<40:45,  1.31it/s]
2|3261|Loss: 0.8047707080841064:  50%|█████     | 3261/6470 [40:51<39:55,  1.34it/s]
2|3262|Loss: 0.8748811483383179:  50%|█████     | 3261/6470 [40:51<39:55,  1.34it/s]
2|3262|Loss: 0.8748811483383179:  50%|█████     | 3262/6470 [40:52<42:20,  1.26it/s]
2|3263|Loss: 0.9274736642837524:  50%|█████     | 3262/6470 [40:52<42:20,  1.26it/s]
2|3263|Loss: 0.9274736642837524:  50%|█████     | 3263/6470 [40:52<40:34,  1.32it/s]
2|3264|Loss: 0.9814270734786987:  50%|█████     | 3263/6470 [40:53<40:34,  1.32it/s]
2|3264|Loss: 0.9814270734786987:  50%|█████     | 3264/6470 [40:53<42:27,  1.26it/s]
2|3265|Loss: 0.804032564163208:  50%|█████     | 3264/6470 [40:54<42:27,  1.26it/s] 
2|3265|Loss: 0.804032564163208:  50%|█████     | 3265/6470 [40:54<43:48,  1.22it/s]
2|3266|Loss: 0.8829849362373352:  50%|█████     | 3265/6470 [40:55<43:48,  1.22it/s]
2|3266|Loss: 0.8829849362373352:  50%|█████     | 3266/6470 [40:55<42:07,  1.27it/s]
2|3267|Loss: 0.7660049200057983:  50%|█████     | 3266/6470 [40:55<42:07,  1.27it/s]
2|3267|Loss: 0.7660049200057983:  50%|█████     | 3267/6470 [40:56<41:42,  1.28it/s]
2|3268|Loss: 0.8857470154762268:  50%|█████     | 3267/6470 [40:56<41:42,  1.28it/s]
2|3268|Loss: 0.8857470154762268:  51%|█████     | 3268/6470 [40:57<41:41,  1.28it/s]
2|3269|Loss: 0.8660047650337219:  51%|█████     | 3268/6470 [40:57<41:41,  1.28it/s]
2|3269|Loss: 0.8660047650337219:  51%|█████     | 3269/6470 [40:57<43:57,  1.21it/s]
2|3270|Loss: 0.9816756844520569:  51%|█████     | 3269/6470 [40:58<43:57,  1.21it/s]
2|3270|Loss: 0.9816756844520569:  51%|█████     | 3270/6470 [40:58<45:58,  1.16it/s]
2|3271|Loss: 0.7712152600288391:  51%|█████     | 3270/6470 [40:59<45:58,  1.16it/s]
2|3271|Loss: 0.7712152600288391:  51%|█████     | 3271/6470 [40:59<48:07,  1.11it/s]
2|3272|Loss: 0.8666290640830994:  51%|█████     | 3271/6470 [41:00<48:07,  1.11it/s]
2|3272|Loss: 0.8666290640830994:  51%|█████     | 3272/6470 [41:00<44:00,  1.21it/s]
2|3273|Loss: 0.9032139778137207:  51%|█████     | 3272/6470 [41:00<44:00,  1.21it/s]
2|3273|Loss: 0.9032139778137207:  51%|█████     | 3273/6470 [41:01<44:16,  1.20it/s]
2|3274|Loss: 0.751211404800415:  51%|█████     | 3273/6470 [41:01<44:16,  1.20it/s] 
2|3274|Loss: 0.751211404800415:  51%|█████     | 3274/6470 [41:02<44:01,  1.21it/s]
2|3275|Loss: 0.7476316094398499:  51%|█████     | 3274/6470 [41:02<44:01,  1.21it/s]
2|3275|Loss: 0.7476316094398499:  51%|█████     | 3275/6470 [41:02<43:28,  1.22it/s]
2|3276|Loss: 0.7648001313209534:  51%|█████     | 3275/6470 [41:03<43:28,  1.22it/s]
2|3276|Loss: 0.7648001313209534:  51%|█████     | 3276/6470 [41:03<45:03,  1.18it/s]
2|3277|Loss: 0.6390604376792908:  51%|█████     | 3276/6470 [41:04<45:03,  1.18it/s]
2|3277|Loss: 0.6390604376792908:  51%|█████     | 3277/6470 [41:04<41:35,  1.28it/s]
2|3278|Loss: 0.8397644758224487:  51%|█████     | 3277/6470 [41:04<41:35,  1.28it/s]
2|3278|Loss: 0.8397644758224487:  51%|█████     | 3278/6470 [41:05<39:41,  1.34it/s]
2|3279|Loss: 0.745968222618103:  51%|█████     | 3278/6470 [41:05<39:41,  1.34it/s] 
2|3279|Loss: 0.745968222618103:  51%|█████     | 3279/6470 [41:05<38:03,  1.40it/s]
2|3280|Loss: 0.8047724962234497:  51%|█████     | 3279/6470 [41:06<38:03,  1.40it/s]
2|3280|Loss: 0.8047724962234497:  51%|█████     | 3280/6470 [41:06<35:56,  1.48it/s]
2|3281|Loss: 0.9716241359710693:  51%|█████     | 3280/6470 [41:06<35:56,  1.48it/s]
2|3281|Loss: 0.9716241359710693:  51%|█████     | 3281/6470 [41:07<39:18,  1.35it/s]
2|3282|Loss: 0.9009923338890076:  51%|█████     | 3281/6470 [41:07<39:18,  1.35it/s]
2|3282|Loss: 0.9009923338890076:  51%|█████     | 3282/6470 [41:07<36:02,  1.47it/s]
2|3283|Loss: 0.907480776309967:  51%|█████     | 3282/6470 [41:08<36:02,  1.47it/s] 
2|3283|Loss: 0.907480776309967:  51%|█████     | 3283/6470 [41:08<36:19,  1.46it/s]
2|3284|Loss: 0.851524293422699:  51%|█████     | 3283/6470 [41:09<36:19,  1.46it/s]
2|3284|Loss: 0.851524293422699:  51%|█████     | 3284/6470 [41:09<39:28,  1.35it/s]
2|3285|Loss: 0.7971097230911255:  51%|█████     | 3284/6470 [41:09<39:28,  1.35it/s]
2|3285|Loss: 0.7971097230911255:  51%|█████     | 3285/6470 [41:10<39:33,  1.34it/s]
2|3286|Loss: 0.8892019391059875:  51%|█████     | 3285/6470 [41:10<39:33,  1.34it/s]
2|3286|Loss: 0.8892019391059875:  51%|█████     | 3286/6470 [41:10<38:49,  1.37it/s]
2|3287|Loss: 0.9392011761665344:  51%|█████     | 3286/6470 [41:11<38:49,  1.37it/s]
2|3287|Loss: 0.9392011761665344:  51%|█████     | 3287/6470 [41:11<40:33,  1.31it/s]
2|3288|Loss: 0.8787435293197632:  51%|█████     | 3287/6470 [41:12<40:33,  1.31it/s]
2|3288|Loss: 0.8787435293197632:  51%|█████     | 3288/6470 [41:12<42:00,  1.26it/s]
2|3289|Loss: 0.8034998774528503:  51%|█████     | 3288/6470 [41:13<42:00,  1.26it/s]
2|3289|Loss: 0.8034998774528503:  51%|█████     | 3289/6470 [41:13<42:22,  1.25it/s]
2|3290|Loss: 0.8079915642738342:  51%|█████     | 3289/6470 [41:13<42:22,  1.25it/s]
2|3290|Loss: 0.8079915642738342:  51%|█████     | 3290/6470 [41:14<41:03,  1.29it/s]
2|3291|Loss: 0.9844037890434265:  51%|█████     | 3290/6470 [41:14<41:03,  1.29it/s]
2|3291|Loss: 0.9844037890434265:  51%|█████     | 3291/6470 [41:14<41:09,  1.29it/s]
2|3292|Loss: 0.8323153257369995:  51%|█████     | 3291/6470 [41:15<41:09,  1.29it/s]
2|3292|Loss: 0.8323153257369995:  51%|█████     | 3292/6470 [41:15<41:23,  1.28it/s]
2|3293|Loss: 0.8639047145843506:  51%|█████     | 3292/6470 [41:16<41:23,  1.28it/s]
2|3293|Loss: 0.8639047145843506:  51%|█████     | 3293/6470 [41:16<40:13,  1.32it/s]
2|3294|Loss: 0.8529943823814392:  51%|█████     | 3293/6470 [41:16<40:13,  1.32it/s]
2|3294|Loss: 0.8529943823814392:  51%|█████     | 3294/6470 [41:17<40:06,  1.32it/s]
2|3295|Loss: 0.8160226941108704:  51%|█████     | 3294/6470 [41:17<40:06,  1.32it/s]
2|3295|Loss: 0.8160226941108704:  51%|█████     | 3295/6470 [41:17<39:20,  1.35it/s]
2|3296|Loss: 0.8504946827888489:  51%|█████     | 3295/6470 [41:18<39:20,  1.35it/s]
2|3296|Loss: 0.8504946827888489:  51%|█████     | 3296/6470 [41:18<38:17,  1.38it/s]
2|3297|Loss: 0.7826137542724609:  51%|█████     | 3296/6470 [41:18<38:17,  1.38it/s]
2|3297|Loss: 0.7826137542724609:  51%|█████     | 3297/6470 [41:19<38:21,  1.38it/s]
2|3298|Loss: 0.7373643517494202:  51%|█████     | 3297/6470 [41:19<38:21,  1.38it/s]
2|3298|Loss: 0.7373643517494202:  51%|█████     | 3298/6470 [41:19<35:13,  1.50it/s]
2|3299|Loss: 1.0839099884033203:  51%|█████     | 3298/6470 [41:20<35:13,  1.50it/s]
2|3299|Loss: 1.0839099884033203:  51%|█████     | 3299/6470 [41:20<36:01,  1.47it/s]
2|3300|Loss: 0.8760188221931458:  51%|█████     | 3299/6470 [41:20<36:01,  1.47it/s]
2|3300|Loss: 0.8760188221931458:  51%|█████     | 3300/6470 [41:21<37:43,  1.40it/s]
2|3301|Loss: 0.8966047167778015:  51%|█████     | 3300/6470 [41:21<37:43,  1.40it/s]
2|3301|Loss: 0.8966047167778015:  51%|█████     | 3301/6470 [41:22<38:56,  1.36it/s]
2|3302|Loss: 0.8296268582344055:  51%|█████     | 3301/6470 [41:22<38:56,  1.36it/s]
2|3302|Loss: 0.8296268582344055:  51%|█████     | 3302/6470 [41:22<40:44,  1.30it/s]
2|3303|Loss: 0.7321118712425232:  51%|█████     | 3302/6470 [41:23<40:44,  1.30it/s]
2|3303|Loss: 0.7321118712425232:  51%|█████     | 3303/6470 [41:23<37:41,  1.40it/s]
2|3304|Loss: 0.6636724472045898:  51%|█████     | 3303/6470 [41:23<37:41,  1.40it/s]
2|3304|Loss: 0.6636724472045898:  51%|█████     | 3304/6470 [41:24<40:34,  1.30it/s]
2|3305|Loss: 0.8944273591041565:  51%|█████     | 3304/6470 [41:24<40:34,  1.30it/s]
2|3305|Loss: 0.8944273591041565:  51%|█████     | 3305/6470 [41:25<41:25,  1.27it/s]
2|3306|Loss: 0.8461681008338928:  51%|█████     | 3305/6470 [41:25<41:25,  1.27it/s]
2|3306|Loss: 0.8461681008338928:  51%|█████     | 3306/6470 [41:25<41:04,  1.28it/s]
2|3307|Loss: 0.9210329651832581:  51%|█████     | 3306/6470 [41:26<41:04,  1.28it/s]
2|3307|Loss: 0.9210329651832581:  51%|█████     | 3307/6470 [41:26<39:12,  1.34it/s]
2|3308|Loss: 0.9811446666717529:  51%|█████     | 3307/6470 [41:26<39:12,  1.34it/s]
2|3308|Loss: 0.9811446666717529:  51%|█████     | 3308/6470 [41:27<36:49,  1.43it/s]
2|3309|Loss: 0.7972304821014404:  51%|█████     | 3308/6470 [41:27<36:49,  1.43it/s]
2|3309|Loss: 0.7972304821014404:  51%|█████     | 3309/6470 [41:27<37:32,  1.40it/s]
2|3310|Loss: 0.8462677597999573:  51%|█████     | 3309/6470 [41:28<37:32,  1.40it/s]
2|3310|Loss: 0.8462677597999573:  51%|█████     | 3310/6470 [41:28<36:42,  1.43it/s]
2|3311|Loss: 0.9995913505554199:  51%|█████     | 3310/6470 [41:29<36:42,  1.43it/s]
2|3311|Loss: 0.9995913505554199:  51%|█████     | 3311/6470 [41:29<38:47,  1.36it/s]
2|3312|Loss: 0.7346533536911011:  51%|█████     | 3311/6470 [41:29<38:47,  1.36it/s]
2|3312|Loss: 0.7346533536911011:  51%|█████     | 3312/6470 [41:30<39:57,  1.32it/s]
2|3313|Loss: 0.8671356439590454:  51%|█████     | 3312/6470 [41:30<39:57,  1.32it/s]
2|3313|Loss: 0.8671356439590454:  51%|█████     | 3313/6470 [41:31<40:46,  1.29it/s]
2|3314|Loss: 0.8571712374687195:  51%|█████     | 3313/6470 [41:31<40:46,  1.29it/s]
2|3314|Loss: 0.8571712374687195:  51%|█████     | 3314/6470 [41:31<42:12,  1.25it/s]
2|3315|Loss: 0.9232780337333679:  51%|█████     | 3314/6470 [41:32<42:12,  1.25it/s]
2|3315|Loss: 0.9232780337333679:  51%|█████     | 3315/6470 [41:32<38:43,  1.36it/s]
2|3316|Loss: 0.9481373429298401:  51%|█████     | 3315/6470 [41:32<38:43,  1.36it/s]
2|3316|Loss: 0.9481373429298401:  51%|█████▏    | 3316/6470 [41:33<39:27,  1.33it/s]
2|3317|Loss: 0.8333032727241516:  51%|█████▏    | 3316/6470 [41:33<39:27,  1.33it/s]
2|3317|Loss: 0.8333032727241516:  51%|█████▏    | 3317/6470 [41:33<37:48,  1.39it/s]
2|3318|Loss: 0.8394607305526733:  51%|█████▏    | 3317/6470 [41:34<37:48,  1.39it/s]
2|3318|Loss: 0.8394607305526733:  51%|█████▏    | 3318/6470 [41:34<38:17,  1.37it/s]
2|3319|Loss: 0.7650938630104065:  51%|█████▏    | 3318/6470 [41:35<38:17,  1.37it/s]
2|3319|Loss: 0.7650938630104065:  51%|█████▏    | 3319/6470 [41:35<39:29,  1.33it/s]
2|3320|Loss: 0.7667162418365479:  51%|█████▏    | 3319/6470 [41:35<39:29,  1.33it/s]
2|3320|Loss: 0.7667162418365479:  51%|█████▏    | 3320/6470 [41:36<39:09,  1.34it/s]
2|3321|Loss: 0.9246698021888733:  51%|█████▏    | 3320/6470 [41:36<39:09,  1.34it/s]
2|3321|Loss: 0.9246698021888733:  51%|█████▏    | 3321/6470 [41:37<39:50,  1.32it/s]
2|3322|Loss: 0.8139379620552063:  51%|█████▏    | 3321/6470 [41:37<39:50,  1.32it/s]
2|3322|Loss: 0.8139379620552063:  51%|█████▏    | 3322/6470 [41:37<39:18,  1.33it/s]
2|3323|Loss: 0.8652793765068054:  51%|█████▏    | 3322/6470 [41:38<39:18,  1.33it/s]
2|3323|Loss: 0.8652793765068054:  51%|█████▏    | 3323/6470 [41:38<39:18,  1.33it/s]
2|3324|Loss: 0.9585777521133423:  51%|█████▏    | 3323/6470 [41:38<39:18,  1.33it/s]
2|3324|Loss: 0.9585777521133423:  51%|█████▏    | 3324/6470 [41:39<37:56,  1.38it/s]
2|3325|Loss: 0.8280736804008484:  51%|█████▏    | 3324/6470 [41:39<37:56,  1.38it/s]
2|3325|Loss: 0.8280736804008484:  51%|█████▏    | 3325/6470 [41:39<35:27,  1.48it/s]
2|3326|Loss: 0.8727846145629883:  51%|█████▏    | 3325/6470 [41:40<35:27,  1.48it/s]
2|3326|Loss: 0.8727846145629883:  51%|█████▏    | 3326/6470 [41:40<36:22,  1.44it/s]
2|3327|Loss: 0.8621281385421753:  51%|█████▏    | 3326/6470 [41:40<36:22,  1.44it/s]
2|3327|Loss: 0.8621281385421753:  51%|█████▏    | 3327/6470 [41:41<39:32,  1.32it/s]
2|3328|Loss: 0.8635669946670532:  51%|█████▏    | 3327/6470 [41:41<39:32,  1.32it/s]
2|3328|Loss: 0.8635669946670532:  51%|█████▏    | 3328/6470 [41:42<39:11,  1.34it/s]
2|3329|Loss: 0.7489619255065918:  51%|█████▏    | 3328/6470 [41:42<39:11,  1.34it/s]
2|3329|Loss: 0.7489619255065918:  51%|█████▏    | 3329/6470 [41:42<40:47,  1.28it/s]
2|3330|Loss: 0.8242742419242859:  51%|█████▏    | 3329/6470 [41:43<40:47,  1.28it/s]
2|3330|Loss: 0.8242742419242859:  51%|█████▏    | 3330/6470 [41:43<39:51,  1.31it/s]
2|3331|Loss: 0.907632052898407:  51%|█████▏    | 3330/6470 [41:44<39:51,  1.31it/s] 
2|3331|Loss: 0.907632052898407:  51%|█████▏    | 3331/6470 [41:44<39:47,  1.31it/s]
2|3332|Loss: 0.8197236061096191:  51%|█████▏    | 3331/6470 [41:44<39:47,  1.31it/s]
2|3332|Loss: 0.8197236061096191:  51%|█████▏    | 3332/6470 [41:45<41:37,  1.26it/s]
2|3333|Loss: 1.0339876413345337:  51%|█████▏    | 3332/6470 [41:45<41:37,  1.26it/s]
2|3333|Loss: 1.0339876413345337:  52%|█████▏    | 3333/6470 [41:46<40:13,  1.30it/s]
2|3334|Loss: 0.9380776286125183:  52%|█████▏    | 3333/6470 [41:46<40:13,  1.30it/s]
2|3334|Loss: 0.9380776286125183:  52%|█████▏    | 3334/6470 [41:46<40:12,  1.30it/s]
2|3335|Loss: 0.8697565197944641:  52%|█████▏    | 3334/6470 [41:47<40:12,  1.30it/s]
2|3335|Loss: 0.8697565197944641:  52%|█████▏    | 3335/6470 [41:47<41:55,  1.25it/s]
2|3336|Loss: 0.9306448101997375:  52%|█████▏    | 3335/6470 [41:48<41:55,  1.25it/s]
2|3336|Loss: 0.9306448101997375:  52%|█████▏    | 3336/6470 [41:48<43:54,  1.19it/s]
2|3337|Loss: 0.8130590915679932:  52%|█████▏    | 3336/6470 [41:49<43:54,  1.19it/s]
2|3337|Loss: 0.8130590915679932:  52%|█████▏    | 3337/6470 [41:49<44:00,  1.19it/s]
2|3338|Loss: 0.814842939376831:  52%|█████▏    | 3337/6470 [41:49<44:00,  1.19it/s] 
2|3338|Loss: 0.814842939376831:  52%|█████▏    | 3338/6470 [41:50<40:54,  1.28it/s]
2|3339|Loss: 0.8441703915596008:  52%|█████▏    | 3338/6470 [41:50<40:54,  1.28it/s]
2|3339|Loss: 0.8441703915596008:  52%|█████▏    | 3339/6470 [41:50<40:17,  1.30it/s]
2|3340|Loss: 0.9714356660842896:  52%|█████▏    | 3339/6470 [41:51<40:17,  1.30it/s]
2|3340|Loss: 0.9714356660842896:  52%|█████▏    | 3340/6470 [41:51<36:42,  1.42it/s]
2|3341|Loss: 0.8899754285812378:  52%|█████▏    | 3340/6470 [41:51<36:42,  1.42it/s]
2|3341|Loss: 0.8899754285812378:  52%|█████▏    | 3341/6470 [41:51<34:28,  1.51it/s]
2|3342|Loss: 0.873992383480072:  52%|█████▏    | 3341/6470 [41:52<34:28,  1.51it/s] 
2|3342|Loss: 0.873992383480072:  52%|█████▏    | 3342/6470 [41:52<38:19,  1.36it/s]
2|3343|Loss: 0.8321599364280701:  52%|█████▏    | 3342/6470 [41:53<38:19,  1.36it/s]
2|3343|Loss: 0.8321599364280701:  52%|█████▏    | 3343/6470 [41:53<38:46,  1.34it/s]
2|3344|Loss: 0.9665490984916687:  52%|█████▏    | 3343/6470 [41:54<38:46,  1.34it/s]
2|3344|Loss: 0.9665490984916687:  52%|█████▏    | 3344/6470 [41:54<38:06,  1.37it/s]
2|3345|Loss: 0.9586136341094971:  52%|█████▏    | 3344/6470 [41:54<38:06,  1.37it/s]
2|3345|Loss: 0.9586136341094971:  52%|█████▏    | 3345/6470 [41:55<40:34,  1.28it/s]
2|3346|Loss: 0.83348548412323:  52%|█████▏    | 3345/6470 [41:55<40:34,  1.28it/s]  
2|3346|Loss: 0.83348548412323:  52%|█████▏    | 3346/6470 [41:56<41:07,  1.27it/s]
2|3347|Loss: 0.7345671653747559:  52%|█████▏    | 3346/6470 [41:56<41:07,  1.27it/s]
2|3347|Loss: 0.7345671653747559:  52%|█████▏    | 3347/6470 [41:56<41:03,  1.27it/s]
2|3348|Loss: 0.9171323180198669:  52%|█████▏    | 3347/6470 [41:57<41:03,  1.27it/s]
2|3348|Loss: 0.9171323180198669:  52%|█████▏    | 3348/6470 [41:57<41:04,  1.27it/s]
2|3349|Loss: 0.8454156517982483:  52%|█████▏    | 3348/6470 [41:58<41:04,  1.27it/s]
2|3349|Loss: 0.8454156517982483:  52%|█████▏    | 3349/6470 [41:58<43:07,  1.21it/s]
2|3350|Loss: 0.8017346262931824:  52%|█████▏    | 3349/6470 [41:59<43:07,  1.21it/s]
2|3350|Loss: 0.8017346262931824:  52%|█████▏    | 3350/6470 [41:59<45:18,  1.15it/s]
2|3351|Loss: 0.796340823173523:  52%|█████▏    | 3350/6470 [41:59<45:18,  1.15it/s] 
2|3351|Loss: 0.796340823173523:  52%|█████▏    | 3351/6470 [42:00<43:15,  1.20it/s]
2|3352|Loss: 0.8615816831588745:  52%|█████▏    | 3351/6470 [42:00<43:15,  1.20it/s]
2|3352|Loss: 0.8615816831588745:  52%|█████▏    | 3352/6470 [42:01<42:14,  1.23it/s]
2|3353|Loss: 0.9122878313064575:  52%|█████▏    | 3352/6470 [42:01<42:14,  1.23it/s]
2|3353|Loss: 0.9122878313064575:  52%|█████▏    | 3353/6470 [42:01<39:41,  1.31it/s]
2|3354|Loss: 0.8846845626831055:  52%|█████▏    | 3353/6470 [42:02<39:41,  1.31it/s]
2|3354|Loss: 0.8846845626831055:  52%|█████▏    | 3354/6470 [42:02<39:09,  1.33it/s]
2|3355|Loss: 0.8107448220252991:  52%|█████▏    | 3354/6470 [42:02<39:09,  1.33it/s]
2|3355|Loss: 0.8107448220252991:  52%|█████▏    | 3355/6470 [42:03<39:36,  1.31it/s]
2|3356|Loss: 0.8145032525062561:  52%|█████▏    | 3355/6470 [42:03<39:36,  1.31it/s]
2|3356|Loss: 0.8145032525062561:  52%|█████▏    | 3356/6470 [42:04<41:28,  1.25it/s]
2|3357|Loss: 0.827389657497406:  52%|█████▏    | 3356/6470 [42:04<41:28,  1.25it/s] 
2|3357|Loss: 0.827389657497406:  52%|█████▏    | 3357/6470 [42:04<41:08,  1.26it/s]
2|3358|Loss: 0.7484537959098816:  52%|█████▏    | 3357/6470 [42:05<41:08,  1.26it/s]
2|3358|Loss: 0.7484537959098816:  52%|█████▏    | 3358/6470 [42:05<42:19,  1.23it/s]
2|3359|Loss: 0.8614770174026489:  52%|█████▏    | 3358/6470 [42:06<42:19,  1.23it/s]
2|3359|Loss: 0.8614770174026489:  52%|█████▏    | 3359/6470 [42:06<39:18,  1.32it/s]
2|3360|Loss: 0.9413753747940063:  52%|█████▏    | 3359/6470 [42:06<39:18,  1.32it/s]
2|3360|Loss: 0.9413753747940063:  52%|█████▏    | 3360/6470 [42:07<38:38,  1.34it/s]
2|3361|Loss: 0.6928098201751709:  52%|█████▏    | 3360/6470 [42:07<38:38,  1.34it/s]
2|3361|Loss: 0.6928098201751709:  52%|█████▏    | 3361/6470 [42:07<36:49,  1.41it/s]
2|3362|Loss: 0.8483573794364929:  52%|█████▏    | 3361/6470 [42:08<36:49,  1.41it/s]
2|3362|Loss: 0.8483573794364929:  52%|█████▏    | 3362/6470 [42:08<37:14,  1.39it/s]
2|3363|Loss: 0.8404274582862854:  52%|█████▏    | 3362/6470 [42:08<37:14,  1.39it/s]
2|3363|Loss: 0.8404274582862854:  52%|█████▏    | 3363/6470 [42:09<38:03,  1.36it/s]
2|3364|Loss: 0.7929295897483826:  52%|█████▏    | 3363/6470 [42:09<38:03,  1.36it/s]
2|3364|Loss: 0.7929295897483826:  52%|█████▏    | 3364/6470 [42:10<40:35,  1.28it/s]
2|3365|Loss: 0.793444812297821:  52%|█████▏    | 3364/6470 [42:10<40:35,  1.28it/s] 
2|3365|Loss: 0.793444812297821:  52%|█████▏    | 3365/6470 [42:10<39:57,  1.29it/s]
2|3366|Loss: 0.8688530921936035:  52%|█████▏    | 3365/6470 [42:11<39:57,  1.29it/s]
2|3366|Loss: 0.8688530921936035:  52%|█████▏    | 3366/6470 [42:11<38:26,  1.35it/s]
2|3367|Loss: 0.8845910429954529:  52%|█████▏    | 3366/6470 [42:11<38:26,  1.35it/s]
2|3367|Loss: 0.8845910429954529:  52%|█████▏    | 3367/6470 [42:12<37:45,  1.37it/s]
2|3368|Loss: 0.857175350189209:  52%|█████▏    | 3367/6470 [42:12<37:45,  1.37it/s] 
2|3368|Loss: 0.857175350189209:  52%|█████▏    | 3368/6470 [42:12<35:41,  1.45it/s]
2|3369|Loss: 0.8712804913520813:  52%|█████▏    | 3368/6470 [42:13<35:41,  1.45it/s]
2|3369|Loss: 0.8712804913520813:  52%|█████▏    | 3369/6470 [42:13<38:17,  1.35it/s]
2|3370|Loss: 0.7581599950790405:  52%|█████▏    | 3369/6470 [42:14<38:17,  1.35it/s]
2|3370|Loss: 0.7581599950790405:  52%|█████▏    | 3370/6470 [42:14<37:58,  1.36it/s]
2|3371|Loss: 0.9119419455528259:  52%|█████▏    | 3370/6470 [42:14<37:58,  1.36it/s]
2|3371|Loss: 0.9119419455528259:  52%|█████▏    | 3371/6470 [42:15<36:25,  1.42it/s]
2|3372|Loss: 0.7444432973861694:  52%|█████▏    | 3371/6470 [42:15<36:25,  1.42it/s]
2|3372|Loss: 0.7444432973861694:  52%|█████▏    | 3372/6470 [42:15<38:06,  1.35it/s]
2|3373|Loss: 0.9668271541595459:  52%|█████▏    | 3372/6470 [42:16<38:06,  1.35it/s]
2|3373|Loss: 0.9668271541595459:  52%|█████▏    | 3373/6470 [42:16<40:47,  1.27it/s]
2|3374|Loss: 0.696409285068512:  52%|█████▏    | 3373/6470 [42:17<40:47,  1.27it/s] 
2|3374|Loss: 0.696409285068512:  52%|█████▏    | 3374/6470 [42:17<39:06,  1.32it/s]
2|3375|Loss: 0.8471423387527466:  52%|█████▏    | 3374/6470 [42:17<39:06,  1.32it/s]
2|3375|Loss: 0.8471423387527466:  52%|█████▏    | 3375/6470 [42:18<37:59,  1.36it/s]
2|3376|Loss: 0.7689799070358276:  52%|█████▏    | 3375/6470 [42:18<37:59,  1.36it/s]
2|3376|Loss: 0.7689799070358276:  52%|█████▏    | 3376/6470 [42:18<39:38,  1.30it/s]
2|3377|Loss: 0.8537223935127258:  52%|█████▏    | 3376/6470 [42:19<39:38,  1.30it/s]
2|3377|Loss: 0.8537223935127258:  52%|█████▏    | 3377/6470 [42:19<42:05,  1.22it/s]
2|3378|Loss: 0.771628201007843:  52%|█████▏    | 3377/6470 [42:20<42:05,  1.22it/s] 
2|3378|Loss: 0.771628201007843:  52%|█████▏    | 3378/6470 [42:20<42:35,  1.21it/s]
2|3379|Loss: 0.942145586013794:  52%|█████▏    | 3378/6470 [42:21<42:35,  1.21it/s]
2|3379|Loss: 0.942145586013794:  52%|█████▏    | 3379/6470 [42:21<43:44,  1.18it/s]
2|3380|Loss: 0.8214284777641296:  52%|█████▏    | 3379/6470 [42:22<43:44,  1.18it/s]
2|3380|Loss: 0.8214284777641296:  52%|█████▏    | 3380/6470 [42:22<44:07,  1.17it/s]
2|3381|Loss: 0.8256292343139648:  52%|█████▏    | 3380/6470 [42:23<44:07,  1.17it/s]
2|3381|Loss: 0.8256292343139648:  52%|█████▏    | 3381/6470 [42:23<45:13,  1.14it/s]
2|3382|Loss: 0.8312983512878418:  52%|█████▏    | 3381/6470 [42:23<45:13,  1.14it/s]
2|3382|Loss: 0.8312983512878418:  52%|█████▏    | 3382/6470 [42:24<44:31,  1.16it/s]
2|3383|Loss: 0.827818751335144:  52%|█████▏    | 3382/6470 [42:24<44:31,  1.16it/s] 
2|3383|Loss: 0.827818751335144:  52%|█████▏    | 3383/6470 [42:25<44:35,  1.15it/s]
2|3384|Loss: 0.959354817867279:  52%|█████▏    | 3383/6470 [42:25<44:35,  1.15it/s]
2|3384|Loss: 0.959354817867279:  52%|█████▏    | 3384/6470 [42:25<43:45,  1.18it/s]
2|3385|Loss: 0.723335325717926:  52%|█████▏    | 3384/6470 [42:26<43:45,  1.18it/s]
2|3385|Loss: 0.723335325717926:  52%|█████▏    | 3385/6470 [42:26<43:09,  1.19it/s]
2|3386|Loss: 0.8982905745506287:  52%|█████▏    | 3385/6470 [42:27<43:09,  1.19it/s]
2|3386|Loss: 0.8982905745506287:  52%|█████▏    | 3386/6470 [42:27<43:24,  1.18it/s]
2|3387|Loss: 0.9455135464668274:  52%|█████▏    | 3386/6470 [42:28<43:24,  1.18it/s]
2|3387|Loss: 0.9455135464668274:  52%|█████▏    | 3387/6470 [42:28<42:55,  1.20it/s]
2|3388|Loss: 0.7785659432411194:  52%|█████▏    | 3387/6470 [42:28<42:55,  1.20it/s]
2|3388|Loss: 0.7785659432411194:  52%|█████▏    | 3388/6470 [42:29<42:19,  1.21it/s]
2|3389|Loss: 0.8112557530403137:  52%|█████▏    | 3388/6470 [42:29<42:19,  1.21it/s]
2|3389|Loss: 0.8112557530403137:  52%|█████▏    | 3389/6470 [42:29<40:40,  1.26it/s]
2|3390|Loss: 0.9195300936698914:  52%|█████▏    | 3389/6470 [42:30<40:40,  1.26it/s]
2|3390|Loss: 0.9195300936698914:  52%|█████▏    | 3390/6470 [42:30<38:40,  1.33it/s]
2|3391|Loss: 0.8511621952056885:  52%|█████▏    | 3390/6470 [42:31<38:40,  1.33it/s]
2|3391|Loss: 0.8511621952056885:  52%|█████▏    | 3391/6470 [42:31<37:49,  1.36it/s]
2|3392|Loss: 0.9184963703155518:  52%|█████▏    | 3391/6470 [42:31<37:49,  1.36it/s]
2|3392|Loss: 0.9184963703155518:  52%|█████▏    | 3392/6470 [42:32<37:22,  1.37it/s]
2|3393|Loss: 0.7777707576751709:  52%|█████▏    | 3392/6470 [42:32<37:22,  1.37it/s]
2|3393|Loss: 0.7777707576751709:  52%|█████▏    | 3393/6470 [42:32<36:03,  1.42it/s]
2|3394|Loss: 0.7915124893188477:  52%|█████▏    | 3393/6470 [42:33<36:03,  1.42it/s]
2|3394|Loss: 0.7915124893188477:  52%|█████▏    | 3394/6470 [42:33<37:32,  1.37it/s]
2|3395|Loss: 1.0077617168426514:  52%|█████▏    | 3394/6470 [42:33<37:32,  1.37it/s]
2|3395|Loss: 1.0077617168426514:  52%|█████▏    | 3395/6470 [42:34<39:24,  1.30it/s]
2|3396|Loss: 0.8994025588035583:  52%|█████▏    | 3395/6470 [42:34<39:24,  1.30it/s]
2|3396|Loss: 0.8994025588035583:  52%|█████▏    | 3396/6470 [42:35<40:47,  1.26it/s]
2|3397|Loss: 0.8617708086967468:  52%|█████▏    | 3396/6470 [42:35<40:47,  1.26it/s]
2|3397|Loss: 0.8617708086967468:  53%|█████▎    | 3397/6470 [42:35<40:33,  1.26it/s]
2|3398|Loss: 0.8727908730506897:  53%|█████▎    | 3397/6470 [42:36<40:33,  1.26it/s]
2|3398|Loss: 0.8727908730506897:  53%|█████▎    | 3398/6470 [42:36<40:24,  1.27it/s]
2|3399|Loss: 0.8530753254890442:  53%|█████▎    | 3398/6470 [42:37<40:24,  1.27it/s]
2|3399|Loss: 0.8530753254890442:  53%|█████▎    | 3399/6470 [42:37<39:47,  1.29it/s]
2|3400|Loss: 1.0413347482681274:  53%|█████▎    | 3399/6470 [42:38<39:47,  1.29it/s]
2|3400|Loss: 1.0413347482681274:  53%|█████▎    | 3400/6470 [42:38<42:25,  1.21it/s]
2|3401|Loss: 0.980842649936676:  53%|█████▎    | 3400/6470 [42:38<42:25,  1.21it/s] 
2|3401|Loss: 0.980842649936676:  53%|█████▎    | 3401/6470 [42:39<41:07,  1.24it/s]
2|3402|Loss: 0.8581752181053162:  53%|█████▎    | 3401/6470 [42:39<41:07,  1.24it/s]
2|3402|Loss: 0.8581752181053162:  53%|█████▎    | 3402/6470 [42:40<41:58,  1.22it/s]
2|3403|Loss: 0.8809353709220886:  53%|█████▎    | 3402/6470 [42:40<41:58,  1.22it/s]
2|3403|Loss: 0.8809353709220886:  53%|█████▎    | 3403/6470 [42:40<43:13,  1.18it/s]
2|3404|Loss: 0.8892590999603271:  53%|█████▎    | 3403/6470 [42:41<43:13,  1.18it/s]
2|3404|Loss: 0.8892590999603271:  53%|█████▎    | 3404/6470 [42:41<42:01,  1.22it/s]
2|3405|Loss: 0.8418029546737671:  53%|█████▎    | 3404/6470 [42:42<42:01,  1.22it/s]
2|3405|Loss: 0.8418029546737671:  53%|█████▎    | 3405/6470 [42:42<40:57,  1.25it/s]
2|3406|Loss: 0.9128715395927429:  53%|█████▎    | 3405/6470 [42:42<40:57,  1.25it/s]
2|3406|Loss: 0.9128715395927429:  53%|█████▎    | 3406/6470 [42:43<38:16,  1.33it/s]
2|3407|Loss: 0.9461609721183777:  53%|█████▎    | 3406/6470 [42:43<38:16,  1.33it/s]
2|3407|Loss: 0.9461609721183777:  53%|█████▎    | 3407/6470 [42:43<39:17,  1.30it/s]
2|3408|Loss: 0.8300018906593323:  53%|█████▎    | 3407/6470 [42:44<39:17,  1.30it/s]
2|3408|Loss: 0.8300018906593323:  53%|█████▎    | 3408/6470 [42:44<37:41,  1.35it/s]
2|3409|Loss: 0.8465448021888733:  53%|█████▎    | 3408/6470 [42:44<37:41,  1.35it/s]
2|3409|Loss: 0.8465448021888733:  53%|█████▎    | 3409/6470 [42:45<37:42,  1.35it/s]
2|3410|Loss: 0.9027984738349915:  53%|█████▎    | 3409/6470 [42:45<37:42,  1.35it/s]
2|3410|Loss: 0.9027984738349915:  53%|█████▎    | 3410/6470 [42:46<36:38,  1.39it/s]
2|3411|Loss: 0.8017110228538513:  53%|█████▎    | 3410/6470 [42:46<36:38,  1.39it/s]
2|3411|Loss: 0.8017110228538513:  53%|█████▎    | 3411/6470 [42:46<38:55,  1.31it/s]
2|3412|Loss: 0.8551187515258789:  53%|█████▎    | 3411/6470 [42:47<38:55,  1.31it/s]
2|3412|Loss: 0.8551187515258789:  53%|█████▎    | 3412/6470 [42:47<37:56,  1.34it/s]
2|3413|Loss: 0.812696635723114:  53%|█████▎    | 3412/6470 [42:47<37:56,  1.34it/s] 
2|3413|Loss: 0.812696635723114:  53%|█████▎    | 3413/6470 [42:48<36:56,  1.38it/s]
2|3414|Loss: 0.9138527512550354:  53%|█████▎    | 3413/6470 [42:48<36:56,  1.38it/s]
2|3414|Loss: 0.9138527512550354:  53%|█████▎    | 3414/6470 [42:48<34:33,  1.47it/s]
2|3415|Loss: 0.889835774898529:  53%|█████▎    | 3414/6470 [42:49<34:33,  1.47it/s] 
2|3415|Loss: 0.889835774898529:  53%|█████▎    | 3415/6470 [42:49<32:43,  1.56it/s]
2|3416|Loss: 0.872294008731842:  53%|█████▎    | 3415/6470 [42:49<32:43,  1.56it/s]
2|3416|Loss: 0.872294008731842:  53%|█████▎    | 3416/6470 [42:50<34:10,  1.49it/s]
2|3417|Loss: 0.9035859704017639:  53%|█████▎    | 3416/6470 [42:50<34:10,  1.49it/s]
2|3417|Loss: 0.9035859704017639:  53%|█████▎    | 3417/6470 [42:51<38:02,  1.34it/s]
2|3418|Loss: 0.8605318069458008:  53%|█████▎    | 3417/6470 [42:51<38:02,  1.34it/s]
2|3418|Loss: 0.8605318069458008:  53%|█████▎    | 3418/6470 [42:51<38:18,  1.33it/s]
2|3419|Loss: 1.0787376165390015:  53%|█████▎    | 3418/6470 [42:52<38:18,  1.33it/s]
2|3419|Loss: 1.0787376165390015:  53%|█████▎    | 3419/6470 [42:52<38:13,  1.33it/s]
2|3420|Loss: 0.7984763979911804:  53%|█████▎    | 3419/6470 [42:52<38:13,  1.33it/s]
2|3420|Loss: 0.7984763979911804:  53%|█████▎    | 3420/6470 [42:53<35:34,  1.43it/s]
2|3421|Loss: 0.7881717085838318:  53%|█████▎    | 3420/6470 [42:53<35:34,  1.43it/s]
2|3421|Loss: 0.7881717085838318:  53%|█████▎    | 3421/6470 [42:53<34:00,  1.49it/s]
2|3422|Loss: 0.7810057997703552:  53%|█████▎    | 3421/6470 [42:54<34:00,  1.49it/s]
2|3422|Loss: 0.7810057997703552:  53%|█████▎    | 3422/6470 [42:54<35:04,  1.45it/s]
2|3423|Loss: 0.8861725926399231:  53%|█████▎    | 3422/6470 [42:54<35:04,  1.45it/s]
2|3423|Loss: 0.8861725926399231:  53%|█████▎    | 3423/6470 [42:55<35:21,  1.44it/s]
2|3424|Loss: 0.8838126063346863:  53%|█████▎    | 3423/6470 [42:55<35:21,  1.44it/s]
2|3424|Loss: 0.8838126063346863:  53%|█████▎    | 3424/6470 [42:56<38:47,  1.31it/s]
2|3425|Loss: 0.819149374961853:  53%|█████▎    | 3424/6470 [42:56<38:47,  1.31it/s] 
2|3425|Loss: 0.819149374961853:  53%|█████▎    | 3425/6470 [42:56<38:58,  1.30it/s]
2|3426|Loss: 0.7435017228126526:  53%|█████▎    | 3425/6470 [42:57<38:58,  1.30it/s]
2|3426|Loss: 0.7435017228126526:  53%|█████▎    | 3426/6470 [42:57<35:09,  1.44it/s]
2|3427|Loss: 0.818516194820404:  53%|█████▎    | 3426/6470 [42:57<35:09,  1.44it/s] 
2|3427|Loss: 0.818516194820404:  53%|█████▎    | 3427/6470 [42:58<38:06,  1.33it/s]
2|3428|Loss: 0.8384395241737366:  53%|█████▎    | 3427/6470 [42:58<38:06,  1.33it/s]
2|3428|Loss: 0.8384395241737366:  53%|█████▎    | 3428/6470 [42:59<40:52,  1.24it/s]
2|3429|Loss: 0.975554883480072:  53%|█████▎    | 3428/6470 [42:59<40:52,  1.24it/s] 
2|3429|Loss: 0.975554883480072:  53%|█████▎    | 3429/6470 [43:00<44:15,  1.15it/s]
2|3430|Loss: 0.788818359375:  53%|█████▎    | 3429/6470 [43:00<44:15,  1.15it/s]   
2|3430|Loss: 0.788818359375:  53%|█████▎    | 3430/6470 [43:01<45:17,  1.12it/s]
2|3431|Loss: 0.8491849899291992:  53%|█████▎    | 3430/6470 [43:01<45:17,  1.12it/s]
2|3431|Loss: 0.8491849899291992:  53%|█████▎    | 3431/6470 [43:02<45:14,  1.12it/s]
2|3432|Loss: 0.813238799571991:  53%|█████▎    | 3431/6470 [43:02<45:14,  1.12it/s] 
2|3432|Loss: 0.813238799571991:  53%|█████▎    | 3432/6470 [43:02<40:37,  1.25it/s]
2|3433|Loss: 0.8733090162277222:  53%|█████▎    | 3432/6470 [43:03<40:37,  1.25it/s]
2|3433|Loss: 0.8733090162277222:  53%|█████▎    | 3433/6470 [43:03<38:18,  1.32it/s]
2|3434|Loss: 0.8483565449714661:  53%|█████▎    | 3433/6470 [43:03<38:18,  1.32it/s]
2|3434|Loss: 0.8483565449714661:  53%|█████▎    | 3434/6470 [43:03<36:30,  1.39it/s]
2|3435|Loss: 0.887514054775238:  53%|█████▎    | 3434/6470 [43:04<36:30,  1.39it/s] 
2|3435|Loss: 0.887514054775238:  53%|█████▎    | 3435/6470 [43:04<37:12,  1.36it/s]
2|3436|Loss: 0.7337616682052612:  53%|█████▎    | 3435/6470 [43:05<37:12,  1.36it/s]
2|3436|Loss: 0.7337616682052612:  53%|█████▎    | 3436/6470 [43:05<36:05,  1.40it/s]
2|3437|Loss: 0.8688478469848633:  53%|█████▎    | 3436/6470 [43:05<36:05,  1.40it/s]
2|3437|Loss: 0.8688478469848633:  53%|█████▎    | 3437/6470 [43:06<35:15,  1.43it/s]
2|3438|Loss: 0.850041925907135:  53%|█████▎    | 3437/6470 [43:06<35:15,  1.43it/s] 
2|3438|Loss: 0.850041925907135:  53%|█████▎    | 3438/6470 [43:06<35:25,  1.43it/s]
2|3439|Loss: 0.788162112236023:  53%|█████▎    | 3438/6470 [43:07<35:25,  1.43it/s]
2|3439|Loss: 0.788162112236023:  53%|█████▎    | 3439/6470 [43:07<36:15,  1.39it/s]
2|3440|Loss: 0.8473768830299377:  53%|█████▎    | 3439/6470 [43:07<36:15,  1.39it/s]
2|3440|Loss: 0.8473768830299377:  53%|█████▎    | 3440/6470 [43:08<35:49,  1.41it/s]
2|3441|Loss: 0.7982839941978455:  53%|█████▎    | 3440/6470 [43:08<35:49,  1.41it/s]
2|3441|Loss: 0.7982839941978455:  53%|█████▎    | 3441/6470 [43:09<39:10,  1.29it/s]
2|3442|Loss: 0.8577642440795898:  53%|█████▎    | 3441/6470 [43:09<39:10,  1.29it/s]
2|3442|Loss: 0.8577642440795898:  53%|█████▎    | 3442/6470 [43:09<36:38,  1.38it/s]
2|3443|Loss: 0.8010669350624084:  53%|█████▎    | 3442/6470 [43:10<36:38,  1.38it/s]
2|3443|Loss: 0.8010669350624084:  53%|█████▎    | 3443/6470 [43:10<35:18,  1.43it/s]
2|3444|Loss: 0.7643144726753235:  53%|█████▎    | 3443/6470 [43:10<35:18,  1.43it/s]
2|3444|Loss: 0.7643144726753235:  53%|█████▎    | 3444/6470 [43:11<34:57,  1.44it/s]
2|3445|Loss: 0.9036940336227417:  53%|█████▎    | 3444/6470 [43:11<34:57,  1.44it/s]
2|3445|Loss: 0.9036940336227417:  53%|█████▎    | 3445/6470 [43:11<35:41,  1.41it/s]
2|3446|Loss: 0.8531680107116699:  53%|█████▎    | 3445/6470 [43:12<35:41,  1.41it/s]
2|3446|Loss: 0.8531680107116699:  53%|█████▎    | 3446/6470 [43:12<36:20,  1.39it/s]
2|3447|Loss: 0.7594820261001587:  53%|█████▎    | 3446/6470 [43:12<36:20,  1.39it/s]
2|3447|Loss: 0.7594820261001587:  53%|█████▎    | 3447/6470 [43:13<35:51,  1.41it/s]
2|3448|Loss: 0.8721131682395935:  53%|█████▎    | 3447/6470 [43:13<35:51,  1.41it/s]
2|3448|Loss: 0.8721131682395935:  53%|█████▎    | 3448/6470 [43:14<36:16,  1.39it/s]
2|3449|Loss: 0.7311255931854248:  53%|█████▎    | 3448/6470 [43:14<36:16,  1.39it/s]
2|3449|Loss: 0.7311255931854248:  53%|█████▎    | 3449/6470 [43:14<35:35,  1.41it/s]
2|3450|Loss: 0.8065145611763:  53%|█████▎    | 3449/6470 [43:15<35:35,  1.41it/s]   
2|3450|Loss: 0.8065145611763:  53%|█████▎    | 3450/6470 [43:15<36:00,  1.40it/s]
2|3451|Loss: 0.8031440377235413:  53%|█████▎    | 3450/6470 [43:15<36:00,  1.40it/s]
2|3451|Loss: 0.8031440377235413:  53%|█████▎    | 3451/6470 [43:16<39:27,  1.28it/s]
2|3452|Loss: 0.7173132300376892:  53%|█████▎    | 3451/6470 [43:16<39:27,  1.28it/s]
2|3452|Loss: 0.7173132300376892:  53%|█████▎    | 3452/6470 [43:17<39:06,  1.29it/s]
2|3453|Loss: 1.0578449964523315:  53%|█████▎    | 3452/6470 [43:17<39:06,  1.29it/s]
2|3453|Loss: 1.0578449964523315:  53%|█████▎    | 3453/6470 [43:17<39:13,  1.28it/s]
2|3454|Loss: 0.7632278800010681:  53%|█████▎    | 3453/6470 [43:18<39:13,  1.28it/s]
2|3454|Loss: 0.7632278800010681:  53%|█████▎    | 3454/6470 [43:18<40:46,  1.23it/s]
2|3455|Loss: 1.0162601470947266:  53%|█████▎    | 3454/6470 [43:19<40:46,  1.23it/s]
2|3455|Loss: 1.0162601470947266:  53%|█████▎    | 3455/6470 [43:19<40:08,  1.25it/s]
2|3456|Loss: 0.8815163373947144:  53%|█████▎    | 3455/6470 [43:19<40:08,  1.25it/s]
2|3456|Loss: 0.8815163373947144:  53%|█████▎    | 3456/6470 [43:20<39:30,  1.27it/s]
2|3457|Loss: 0.8909673690795898:  53%|█████▎    | 3456/6470 [43:20<39:30,  1.27it/s]
2|3457|Loss: 0.8909673690795898:  53%|█████▎    | 3457/6470 [43:21<40:48,  1.23it/s]
2|3458|Loss: 0.9479394555091858:  53%|█████▎    | 3457/6470 [43:21<40:48,  1.23it/s]
2|3458|Loss: 0.9479394555091858:  53%|█████▎    | 3458/6470 [43:21<37:24,  1.34it/s]
2|3459|Loss: 0.8933594226837158:  53%|█████▎    | 3458/6470 [43:22<37:24,  1.34it/s]
2|3459|Loss: 0.8933594226837158:  53%|█████▎    | 3459/6470 [43:22<33:22,  1.50it/s]
2|3460|Loss: 1.0487470626831055:  53%|█████▎    | 3459/6470 [43:22<33:22,  1.50it/s]
2|3460|Loss: 1.0487470626831055:  53%|█████▎    | 3460/6470 [43:23<35:51,  1.40it/s]
2|3461|Loss: 0.84181147813797:  53%|█████▎    | 3460/6470 [43:23<35:51,  1.40it/s]  
2|3461|Loss: 0.84181147813797:  53%|█████▎    | 3461/6470 [43:23<36:55,  1.36it/s]
2|3462|Loss: 0.9115146398544312:  53%|█████▎    | 3461/6470 [43:24<36:55,  1.36it/s]
2|3462|Loss: 0.9115146398544312:  54%|█████▎    | 3462/6470 [43:24<37:39,  1.33it/s]
2|3463|Loss: 0.8504135012626648:  54%|█████▎    | 3462/6470 [43:25<37:39,  1.33it/s]
2|3463|Loss: 0.8504135012626648:  54%|█████▎    | 3463/6470 [43:25<38:37,  1.30it/s]
2|3464|Loss: 0.7364070415496826:  54%|█████▎    | 3463/6470 [43:25<38:37,  1.30it/s]
2|3464|Loss: 0.7364070415496826:  54%|█████▎    | 3464/6470 [43:26<39:11,  1.28it/s]
2|3465|Loss: 0.960883378982544:  54%|█████▎    | 3464/6470 [43:26<39:11,  1.28it/s] 
2|3465|Loss: 0.960883378982544:  54%|█████▎    | 3465/6470 [43:26<35:31,  1.41it/s]
2|3466|Loss: 0.7699315547943115:  54%|█████▎    | 3465/6470 [43:27<35:31,  1.41it/s]
2|3466|Loss: 0.7699315547943115:  54%|█████▎    | 3466/6470 [43:27<32:42,  1.53it/s]
2|3467|Loss: 0.8078410625457764:  54%|█████▎    | 3466/6470 [43:27<32:42,  1.53it/s]
2|3467|Loss: 0.8078410625457764:  54%|█████▎    | 3467/6470 [43:28<35:21,  1.42it/s]
2|3468|Loss: 0.7677832841873169:  54%|█████▎    | 3467/6470 [43:28<35:21,  1.42it/s]
2|3468|Loss: 0.7677832841873169:  54%|█████▎    | 3468/6470 [43:28<36:22,  1.38it/s]
2|3469|Loss: 0.861977219581604:  54%|█████▎    | 3468/6470 [43:29<36:22,  1.38it/s] 
2|3469|Loss: 0.861977219581604:  54%|█████▎    | 3469/6470 [43:29<37:19,  1.34it/s]
2|3470|Loss: 0.8920404314994812:  54%|█████▎    | 3469/6470 [43:30<37:19,  1.34it/s]
2|3470|Loss: 0.8920404314994812:  54%|█████▎    | 3470/6470 [43:30<38:57,  1.28it/s]
2|3471|Loss: 0.9968900084495544:  54%|█████▎    | 3470/6470 [43:31<38:57,  1.28it/s]
2|3471|Loss: 0.9968900084495544:  54%|█████▎    | 3471/6470 [43:31<41:06,  1.22it/s]
2|3472|Loss: 0.7090516686439514:  54%|█████▎    | 3471/6470 [43:32<41:06,  1.22it/s]
2|3472|Loss: 0.7090516686439514:  54%|█████▎    | 3472/6470 [43:32<41:16,  1.21it/s]
2|3473|Loss: 0.9309130907058716:  54%|█████▎    | 3472/6470 [43:32<41:16,  1.21it/s]
2|3473|Loss: 0.9309130907058716:  54%|█████▎    | 3473/6470 [43:33<40:26,  1.23it/s]
2|3474|Loss: 0.8071344494819641:  54%|█████▎    | 3473/6470 [43:33<40:26,  1.23it/s]
2|3474|Loss: 0.8071344494819641:  54%|█████▎    | 3474/6470 [43:34<42:04,  1.19it/s]
2|3475|Loss: 0.9545893669128418:  54%|█████▎    | 3474/6470 [43:34<42:04,  1.19it/s]
2|3475|Loss: 0.9545893669128418:  54%|█████▎    | 3475/6470 [43:34<40:00,  1.25it/s]
2|3476|Loss: 0.9830708503723145:  54%|█████▎    | 3475/6470 [43:35<40:00,  1.25it/s]
2|3476|Loss: 0.9830708503723145:  54%|█████▎    | 3476/6470 [43:35<38:08,  1.31it/s]
2|3477|Loss: 0.8810756802558899:  54%|█████▎    | 3476/6470 [43:35<38:08,  1.31it/s]
2|3477|Loss: 0.8810756802558899:  54%|█████▎    | 3477/6470 [43:36<40:25,  1.23it/s]
2|3478|Loss: 0.806202232837677:  54%|█████▎    | 3477/6470 [43:36<40:25,  1.23it/s] 
2|3478|Loss: 0.806202232837677:  54%|█████▍    | 3478/6470 [43:37<39:05,  1.28it/s]
2|3479|Loss: 0.7715927958488464:  54%|█████▍    | 3478/6470 [43:37<39:05,  1.28it/s]
2|3479|Loss: 0.7715927958488464:  54%|█████▍    | 3479/6470 [43:37<35:49,  1.39it/s]
2|3480|Loss: 0.8896426558494568:  54%|█████▍    | 3479/6470 [43:38<35:49,  1.39it/s]
2|3480|Loss: 0.8896426558494568:  54%|█████▍    | 3480/6470 [43:38<35:07,  1.42it/s]
2|3481|Loss: 0.9723075032234192:  54%|█████▍    | 3480/6470 [43:38<35:07,  1.42it/s]
2|3481|Loss: 0.9723075032234192:  54%|█████▍    | 3481/6470 [43:39<35:20,  1.41it/s]
2|3482|Loss: 0.991811215877533:  54%|█████▍    | 3481/6470 [43:39<35:20,  1.41it/s] 
2|3482|Loss: 0.991811215877533:  54%|█████▍    | 3482/6470 [43:39<36:03,  1.38it/s]
2|3483|Loss: 0.8941785097122192:  54%|█████▍    | 3482/6470 [43:40<36:03,  1.38it/s]
2|3483|Loss: 0.8941785097122192:  54%|█████▍    | 3483/6470 [43:40<36:50,  1.35it/s]
2|3484|Loss: 0.9000853896141052:  54%|█████▍    | 3483/6470 [43:40<36:50,  1.35it/s]
2|3484|Loss: 0.9000853896141052:  54%|█████▍    | 3484/6470 [43:41<35:56,  1.38it/s]
2|3485|Loss: 0.8952436447143555:  54%|█████▍    | 3484/6470 [43:41<35:56,  1.38it/s]
2|3485|Loss: 0.8952436447143555:  54%|█████▍    | 3485/6470 [43:41<35:29,  1.40it/s]
2|3486|Loss: 0.9722765684127808:  54%|█████▍    | 3485/6470 [43:42<35:29,  1.40it/s]
2|3486|Loss: 0.9722765684127808:  54%|█████▍    | 3486/6470 [43:42<34:51,  1.43it/s]
2|3487|Loss: 0.8052198886871338:  54%|█████▍    | 3486/6470 [43:42<34:51,  1.43it/s]
2|3487|Loss: 0.8052198886871338:  54%|█████▍    | 3487/6470 [43:43<33:36,  1.48it/s]
2|3488|Loss: 0.8284292221069336:  54%|█████▍    | 3487/6470 [43:43<33:36,  1.48it/s]
2|3488|Loss: 0.8284292221069336:  54%|█████▍    | 3488/6470 [43:44<36:32,  1.36it/s]
2|3489|Loss: 0.7744797468185425:  54%|█████▍    | 3488/6470 [43:44<36:32,  1.36it/s]
2|3489|Loss: 0.7744797468185425:  54%|█████▍    | 3489/6470 [43:44<35:32,  1.40it/s]
2|3490|Loss: 0.9146833419799805:  54%|█████▍    | 3489/6470 [43:45<35:32,  1.40it/s]
2|3490|Loss: 0.9146833419799805:  54%|█████▍    | 3490/6470 [43:45<33:54,  1.46it/s]
2|3491|Loss: 0.9373133182525635:  54%|█████▍    | 3490/6470 [43:45<33:54,  1.46it/s]
2|3491|Loss: 0.9373133182525635:  54%|█████▍    | 3491/6470 [43:46<33:58,  1.46it/s]
2|3492|Loss: 0.8141577839851379:  54%|█████▍    | 3491/6470 [43:46<33:58,  1.46it/s]
2|3492|Loss: 0.8141577839851379:  54%|█████▍    | 3492/6470 [43:46<34:13,  1.45it/s]
2|3493|Loss: 0.7793238162994385:  54%|█████▍    | 3492/6470 [43:47<34:13,  1.45it/s]
2|3493|Loss: 0.7793238162994385:  54%|█████▍    | 3493/6470 [43:47<32:27,  1.53it/s]
2|3494|Loss: 0.8706564903259277:  54%|█████▍    | 3493/6470 [43:47<32:27,  1.53it/s]
2|3494|Loss: 0.8706564903259277:  54%|█████▍    | 3494/6470 [43:47<32:16,  1.54it/s]
2|3495|Loss: 0.934220016002655:  54%|█████▍    | 3494/6470 [43:48<32:16,  1.54it/s] 
2|3495|Loss: 0.934220016002655:  54%|█████▍    | 3495/6470 [43:48<31:56,  1.55it/s]
2|3496|Loss: 0.9271569848060608:  54%|█████▍    | 3495/6470 [43:49<31:56,  1.55it/s]
2|3496|Loss: 0.9271569848060608:  54%|█████▍    | 3496/6470 [43:49<35:07,  1.41it/s]
2|3497|Loss: 0.7388480305671692:  54%|█████▍    | 3496/6470 [43:49<35:07,  1.41it/s]
2|3497|Loss: 0.7388480305671692:  54%|█████▍    | 3497/6470 [43:50<36:39,  1.35it/s]
2|3498|Loss: 0.7706217169761658:  54%|█████▍    | 3497/6470 [43:50<36:39,  1.35it/s]
2|3498|Loss: 0.7706217169761658:  54%|█████▍    | 3498/6470 [43:50<35:19,  1.40it/s]
2|3499|Loss: 0.9035485982894897:  54%|█████▍    | 3498/6470 [43:51<35:19,  1.40it/s]
2|3499|Loss: 0.9035485982894897:  54%|█████▍    | 3499/6470 [43:51<35:30,  1.39it/s]
2|3500|Loss: 0.8264634013175964:  54%|█████▍    | 3499/6470 [43:52<35:30,  1.39it/s]
2|3500|Loss: 0.8264634013175964:  54%|█████▍    | 3500/6470 [43:52<35:00,  1.41it/s]
2|3501|Loss: 1.0056663751602173:  54%|█████▍    | 3500/6470 [43:52<35:00,  1.41it/s]
2|3501|Loss: 1.0056663751602173:  54%|█████▍    | 3501/6470 [43:52<33:54,  1.46it/s]
2|3502|Loss: 0.8127274513244629:  54%|█████▍    | 3501/6470 [43:53<33:54,  1.46it/s]
2|3502|Loss: 0.8127274513244629:  54%|█████▍    | 3502/6470 [43:53<33:57,  1.46it/s]
2|3503|Loss: 0.8169528245925903:  54%|█████▍    | 3502/6470 [43:54<33:57,  1.46it/s]
2|3503|Loss: 0.8169528245925903:  54%|█████▍    | 3503/6470 [43:54<35:36,  1.39it/s]
2|3504|Loss: 0.8209805488586426:  54%|█████▍    | 3503/6470 [43:54<35:36,  1.39it/s]
2|3504|Loss: 0.8209805488586426:  54%|█████▍    | 3504/6470 [43:55<36:26,  1.36it/s]
2|3505|Loss: 0.87278151512146:  54%|█████▍    | 3504/6470 [43:55<36:26,  1.36it/s]  
2|3505|Loss: 0.87278151512146:  54%|█████▍    | 3505/6470 [43:55<35:34,  1.39it/s]
2|3506|Loss: 0.7753128409385681:  54%|█████▍    | 3505/6470 [43:56<35:34,  1.39it/s]
2|3506|Loss: 0.7753128409385681:  54%|█████▍    | 3506/6470 [43:56<37:39,  1.31it/s]
2|3507|Loss: 0.6972624063491821:  54%|█████▍    | 3506/6470 [43:57<37:39,  1.31it/s]
2|3507|Loss: 0.6972624063491821:  54%|█████▍    | 3507/6470 [43:57<37:51,  1.30it/s]
2|3508|Loss: 0.6400640606880188:  54%|█████▍    | 3507/6470 [43:58<37:51,  1.30it/s]
2|3508|Loss: 0.6400640606880188:  54%|█████▍    | 3508/6470 [43:58<38:55,  1.27it/s]
2|3509|Loss: 0.6859982013702393:  54%|█████▍    | 3508/6470 [43:58<38:55,  1.27it/s]
2|3509|Loss: 0.6859982013702393:  54%|█████▍    | 3509/6470 [43:59<40:53,  1.21it/s]
2|3510|Loss: 0.8121305704116821:  54%|█████▍    | 3509/6470 [43:59<40:53,  1.21it/s]
2|3510|Loss: 0.8121305704116821:  54%|█████▍    | 3510/6470 [43:59<35:56,  1.37it/s]
2|3511|Loss: 0.8380961418151855:  54%|█████▍    | 3510/6470 [44:00<35:56,  1.37it/s]
2|3511|Loss: 0.8380961418151855:  54%|█████▍    | 3511/6470 [44:00<36:55,  1.34it/s]
2|3512|Loss: 0.7601566314697266:  54%|█████▍    | 3511/6470 [44:01<36:55,  1.34it/s]
2|3512|Loss: 0.7601566314697266:  54%|█████▍    | 3512/6470 [44:01<34:17,  1.44it/s]
2|3513|Loss: 0.9055675864219666:  54%|█████▍    | 3512/6470 [44:01<34:17,  1.44it/s]
2|3513|Loss: 0.9055675864219666:  54%|█████▍    | 3513/6470 [44:01<33:30,  1.47it/s]
2|3514|Loss: 0.8827476501464844:  54%|█████▍    | 3513/6470 [44:02<33:30,  1.47it/s]
2|3514|Loss: 0.8827476501464844:  54%|█████▍    | 3514/6470 [44:02<33:53,  1.45it/s]
2|3515|Loss: 0.8549280166625977:  54%|█████▍    | 3514/6470 [44:02<33:53,  1.45it/s]
2|3515|Loss: 0.8549280166625977:  54%|█████▍    | 3515/6470 [44:03<35:24,  1.39it/s]
2|3516|Loss: 0.7969408631324768:  54%|█████▍    | 3515/6470 [44:03<35:24,  1.39it/s]
2|3516|Loss: 0.7969408631324768:  54%|█████▍    | 3516/6470 [44:03<33:39,  1.46it/s]
2|3517|Loss: 0.9661361575126648:  54%|█████▍    | 3516/6470 [44:04<33:39,  1.46it/s]
2|3517|Loss: 0.9661361575126648:  54%|█████▍    | 3517/6470 [44:04<35:48,  1.37it/s]
2|3518|Loss: 0.8909345269203186:  54%|█████▍    | 3517/6470 [44:05<35:48,  1.37it/s]
2|3518|Loss: 0.8909345269203186:  54%|█████▍    | 3518/6470 [44:05<39:28,  1.25it/s]
2|3519|Loss: 0.8145354986190796:  54%|█████▍    | 3518/6470 [44:06<39:28,  1.25it/s]
2|3519|Loss: 0.8145354986190796:  54%|█████▍    | 3519/6470 [44:06<39:32,  1.24it/s]
2|3520|Loss: 0.8121820092201233:  54%|█████▍    | 3519/6470 [44:06<39:32,  1.24it/s]
2|3520|Loss: 0.8121820092201233:  54%|█████▍    | 3520/6470 [44:07<36:55,  1.33it/s]
2|3521|Loss: 0.7779998779296875:  54%|█████▍    | 3520/6470 [44:07<36:55,  1.33it/s]
2|3521|Loss: 0.7779998779296875:  54%|█████▍    | 3521/6470 [44:08<38:19,  1.28it/s]
2|3522|Loss: 1.1141655445098877:  54%|█████▍    | 3521/6470 [44:08<38:19,  1.28it/s]
2|3522|Loss: 1.1141655445098877:  54%|█████▍    | 3522/6470 [44:08<40:31,  1.21it/s]
2|3523|Loss: 0.7791329622268677:  54%|█████▍    | 3522/6470 [44:09<40:31,  1.21it/s]
2|3523|Loss: 0.7791329622268677:  54%|█████▍    | 3523/6470 [44:09<38:07,  1.29it/s]
2|3524|Loss: 0.7285348773002625:  54%|█████▍    | 3523/6470 [44:10<38:07,  1.29it/s]
2|3524|Loss: 0.7285348773002625:  54%|█████▍    | 3524/6470 [44:10<39:35,  1.24it/s]
2|3525|Loss: 0.8251715898513794:  54%|█████▍    | 3524/6470 [44:10<39:35,  1.24it/s]
2|3525|Loss: 0.8251715898513794:  54%|█████▍    | 3525/6470 [44:11<39:32,  1.24it/s]
2|3526|Loss: 0.8667214512825012:  54%|█████▍    | 3525/6470 [44:11<39:32,  1.24it/s]
2|3526|Loss: 0.8667214512825012:  54%|█████▍    | 3526/6470 [44:12<40:38,  1.21it/s]
2|3527|Loss: 0.993783712387085:  54%|█████▍    | 3526/6470 [44:12<40:38,  1.21it/s] 
2|3527|Loss: 0.993783712387085:  55%|█████▍    | 3527/6470 [44:12<38:21,  1.28it/s]
2|3528|Loss: 0.7934656143188477:  55%|█████▍    | 3527/6470 [44:13<38:21,  1.28it/s]
2|3528|Loss: 0.7934656143188477:  55%|█████▍    | 3528/6470 [44:13<36:23,  1.35it/s]
2|3529|Loss: 0.8013753294944763:  55%|█████▍    | 3528/6470 [44:13<36:23,  1.35it/s]
2|3529|Loss: 0.8013753294944763:  55%|█████▍    | 3529/6470 [44:14<33:31,  1.46it/s]
2|3530|Loss: 0.9627903699874878:  55%|█████▍    | 3529/6470 [44:14<33:31,  1.46it/s]
2|3530|Loss: 0.9627903699874878:  55%|█████▍    | 3530/6470 [44:14<31:13,  1.57it/s]
2|3531|Loss: 0.8864081501960754:  55%|█████▍    | 3530/6470 [44:14<31:13,  1.57it/s]
2|3531|Loss: 0.8864081501960754:  55%|█████▍    | 3531/6470 [44:15<29:42,  1.65it/s]
2|3532|Loss: 1.0060242414474487:  55%|█████▍    | 3531/6470 [44:15<29:42,  1.65it/s]
2|3532|Loss: 1.0060242414474487:  55%|█████▍    | 3532/6470 [44:15<31:04,  1.58it/s]
2|3533|Loss: 0.8716263771057129:  55%|█████▍    | 3532/6470 [44:16<31:04,  1.58it/s]
2|3533|Loss: 0.8716263771057129:  55%|█████▍    | 3533/6470 [44:16<31:42,  1.54it/s]
2|3534|Loss: 0.833638608455658:  55%|█████▍    | 3533/6470 [44:16<31:42,  1.54it/s] 
2|3534|Loss: 0.833638608455658:  55%|█████▍    | 3534/6470 [44:17<31:24,  1.56it/s]
2|3535|Loss: 0.8500353693962097:  55%|█████▍    | 3534/6470 [44:17<31:24,  1.56it/s]
2|3535|Loss: 0.8500353693962097:  55%|█████▍    | 3535/6470 [44:17<32:55,  1.49it/s]
2|3536|Loss: 0.868197500705719:  55%|█████▍    | 3535/6470 [44:18<32:55,  1.49it/s] 
2|3536|Loss: 0.868197500705719:  55%|█████▍    | 3536/6470 [44:18<36:08,  1.35it/s]
2|3537|Loss: 0.7827826142311096:  55%|█████▍    | 3536/6470 [44:19<36:08,  1.35it/s]
2|3537|Loss: 0.7827826142311096:  55%|█████▍    | 3537/6470 [44:19<35:22,  1.38it/s]
2|3538|Loss: 1.0403271913528442:  55%|█████▍    | 3537/6470 [44:19<35:22,  1.38it/s]
2|3538|Loss: 1.0403271913528442:  55%|█████▍    | 3538/6470 [44:20<35:48,  1.36it/s]
2|3539|Loss: 0.8478795289993286:  55%|█████▍    | 3538/6470 [44:20<35:48,  1.36it/s]
2|3539|Loss: 0.8478795289993286:  55%|█████▍    | 3539/6470 [44:20<35:08,  1.39it/s]
2|3540|Loss: 0.7892773747444153:  55%|█████▍    | 3539/6470 [44:21<35:08,  1.39it/s]
2|3540|Loss: 0.7892773747444153:  55%|█████▍    | 3540/6470 [44:21<34:46,  1.40it/s]
2|3541|Loss: 0.838895857334137:  55%|█████▍    | 3540/6470 [44:22<34:46,  1.40it/s] 
2|3541|Loss: 0.838895857334137:  55%|█████▍    | 3541/6470 [44:22<37:18,  1.31it/s]
2|3542|Loss: 0.8418722748756409:  55%|█████▍    | 3541/6470 [44:22<37:18,  1.31it/s]
2|3542|Loss: 0.8418722748756409:  55%|█████▍    | 3542/6470 [44:23<36:14,  1.35it/s]
2|3543|Loss: 0.820989191532135:  55%|█████▍    | 3542/6470 [44:23<36:14,  1.35it/s] 
2|3543|Loss: 0.820989191532135:  55%|█████▍    | 3543/6470 [44:23<35:19,  1.38it/s]
2|3544|Loss: 0.8670427203178406:  55%|█████▍    | 3543/6470 [44:24<35:19,  1.38it/s]
2|3544|Loss: 0.8670427203178406:  55%|█████▍    | 3544/6470 [44:24<34:56,  1.40it/s]
2|3545|Loss: 0.6576805710792542:  55%|█████▍    | 3544/6470 [44:24<34:56,  1.40it/s]
2|3545|Loss: 0.6576805710792542:  55%|█████▍    | 3545/6470 [44:25<36:21,  1.34it/s]
2|3546|Loss: 0.9003731608390808:  55%|█████▍    | 3545/6470 [44:25<36:21,  1.34it/s]
2|3546|Loss: 0.9003731608390808:  55%|█████▍    | 3546/6470 [44:25<34:21,  1.42it/s]
2|3547|Loss: 0.6059948801994324:  55%|█████▍    | 3546/6470 [44:26<34:21,  1.42it/s]
2|3547|Loss: 0.6059948801994324:  55%|█████▍    | 3547/6470 [44:26<34:42,  1.40it/s]
2|3548|Loss: 0.8213919401168823:  55%|█████▍    | 3547/6470 [44:27<34:42,  1.40it/s]
2|3548|Loss: 0.8213919401168823:  55%|█████▍    | 3548/6470 [44:27<31:16,  1.56it/s]
2|3549|Loss: 0.9101644158363342:  55%|█████▍    | 3548/6470 [44:27<31:16,  1.56it/s]
2|3549|Loss: 0.9101644158363342:  55%|█████▍    | 3549/6470 [44:27<32:31,  1.50it/s]
2|3550|Loss: 0.8122743368148804:  55%|█████▍    | 3549/6470 [44:28<32:31,  1.50it/s]
2|3550|Loss: 0.8122743368148804:  55%|█████▍    | 3550/6470 [44:28<34:05,  1.43it/s]
2|3551|Loss: 0.8768783807754517:  55%|█████▍    | 3550/6470 [44:29<34:05,  1.43it/s]
2|3551|Loss: 0.8768783807754517:  55%|█████▍    | 3551/6470 [44:29<32:03,  1.52it/s]
2|3552|Loss: 0.8396384716033936:  55%|█████▍    | 3551/6470 [44:29<32:03,  1.52it/s]
2|3552|Loss: 0.8396384716033936:  55%|█████▍    | 3552/6470 [44:30<33:49,  1.44it/s]
2|3553|Loss: 0.635360598564148:  55%|█████▍    | 3552/6470 [44:30<33:49,  1.44it/s] 
2|3553|Loss: 0.635360598564148:  55%|█████▍    | 3553/6470 [44:30<34:13,  1.42it/s]
2|3554|Loss: 0.6558486223220825:  55%|█████▍    | 3553/6470 [44:31<34:13,  1.42it/s]
2|3554|Loss: 0.6558486223220825:  55%|█████▍    | 3554/6470 [44:31<33:22,  1.46it/s]
2|3555|Loss: 0.9045486450195312:  55%|█████▍    | 3554/6470 [44:31<33:22,  1.46it/s]
2|3555|Loss: 0.9045486450195312:  55%|█████▍    | 3555/6470 [44:32<35:37,  1.36it/s]
2|3556|Loss: 0.8078433871269226:  55%|█████▍    | 3555/6470 [44:32<35:37,  1.36it/s]
2|3556|Loss: 0.8078433871269226:  55%|█████▍    | 3556/6470 [44:33<36:41,  1.32it/s]
2|3557|Loss: 0.6973251700401306:  55%|█████▍    | 3556/6470 [44:33<36:41,  1.32it/s]
2|3557|Loss: 0.6973251700401306:  55%|█████▍    | 3557/6470 [44:33<36:06,  1.34it/s]
2|3558|Loss: 0.8121613264083862:  55%|█████▍    | 3557/6470 [44:34<36:06,  1.34it/s]
2|3558|Loss: 0.8121613264083862:  55%|█████▍    | 3558/6470 [44:34<35:32,  1.37it/s]
2|3559|Loss: 0.7745112180709839:  55%|█████▍    | 3558/6470 [44:34<35:32,  1.37it/s]
2|3559|Loss: 0.7745112180709839:  55%|█████▌    | 3559/6470 [44:35<35:50,  1.35it/s]
2|3560|Loss: 0.7190923094749451:  55%|█████▌    | 3559/6470 [44:35<35:50,  1.35it/s]
2|3560|Loss: 0.7190923094749451:  55%|█████▌    | 3560/6470 [44:36<37:51,  1.28it/s]
2|3561|Loss: 0.8256898522377014:  55%|█████▌    | 3560/6470 [44:36<37:51,  1.28it/s]
2|3561|Loss: 0.8256898522377014:  55%|█████▌    | 3561/6470 [44:36<36:47,  1.32it/s]
2|3562|Loss: 0.9765869975090027:  55%|█████▌    | 3561/6470 [44:37<36:47,  1.32it/s]
2|3562|Loss: 0.9765869975090027:  55%|█████▌    | 3562/6470 [44:37<35:28,  1.37it/s]
2|3563|Loss: 1.0356361865997314:  55%|█████▌    | 3562/6470 [44:37<35:28,  1.37it/s]
2|3563|Loss: 1.0356361865997314:  55%|█████▌    | 3563/6470 [44:38<35:29,  1.37it/s]
2|3564|Loss: 0.9055148959159851:  55%|█████▌    | 3563/6470 [44:38<35:29,  1.37it/s]
2|3564|Loss: 0.9055148959159851:  55%|█████▌    | 3564/6470 [44:39<36:26,  1.33it/s]
2|3565|Loss: 0.9353455901145935:  55%|█████▌    | 3564/6470 [44:39<36:26,  1.33it/s]
2|3565|Loss: 0.9353455901145935:  55%|█████▌    | 3565/6470 [44:39<37:55,  1.28it/s]
2|3566|Loss: 0.8456777930259705:  55%|█████▌    | 3565/6470 [44:40<37:55,  1.28it/s]
2|3566|Loss: 0.8456777930259705:  55%|█████▌    | 3566/6470 [44:40<36:55,  1.31it/s]
2|3567|Loss: 0.8399454951286316:  55%|█████▌    | 3566/6470 [44:40<36:55,  1.31it/s]
2|3567|Loss: 0.8399454951286316:  55%|█████▌    | 3567/6470 [44:41<37:03,  1.31it/s]
2|3568|Loss: 0.8613039255142212:  55%|█████▌    | 3567/6470 [44:41<37:03,  1.31it/s]
2|3568|Loss: 0.8613039255142212:  55%|█████▌    | 3568/6470 [44:42<35:59,  1.34it/s]
2|3569|Loss: 0.9075208902359009:  55%|█████▌    | 3568/6470 [44:42<35:59,  1.34it/s]
2|3569|Loss: 0.9075208902359009:  55%|█████▌    | 3569/6470 [44:42<37:36,  1.29it/s]
2|3570|Loss: 0.8969287872314453:  55%|█████▌    | 3569/6470 [44:43<37:36,  1.29it/s]
2|3570|Loss: 0.8969287872314453:  55%|█████▌    | 3570/6470 [44:43<36:58,  1.31it/s]
2|3571|Loss: 0.9253426194190979:  55%|█████▌    | 3570/6470 [44:43<36:58,  1.31it/s]
2|3571|Loss: 0.9253426194190979:  55%|█████▌    | 3571/6470 [44:44<33:46,  1.43it/s]
2|3572|Loss: 0.761333167552948:  55%|█████▌    | 3571/6470 [44:44<33:46,  1.43it/s] 
2|3572|Loss: 0.761333167552948:  55%|█████▌    | 3572/6470 [44:44<32:44,  1.48it/s]
2|3573|Loss: 0.8845089077949524:  55%|█████▌    | 3572/6470 [44:45<32:44,  1.48it/s]
2|3573|Loss: 0.8845089077949524:  55%|█████▌    | 3573/6470 [44:45<35:49,  1.35it/s]
2|3574|Loss: 0.901647686958313:  55%|█████▌    | 3573/6470 [44:46<35:49,  1.35it/s] 
2|3574|Loss: 0.901647686958313:  55%|█████▌    | 3574/6470 [44:46<36:38,  1.32it/s]
2|3575|Loss: 0.7982490658760071:  55%|█████▌    | 3574/6470 [44:46<36:38,  1.32it/s]
2|3575|Loss: 0.7982490658760071:  55%|█████▌    | 3575/6470 [44:47<35:45,  1.35it/s]
2|3576|Loss: 0.8834853172302246:  55%|█████▌    | 3575/6470 [44:47<35:45,  1.35it/s]
2|3576|Loss: 0.8834853172302246:  55%|█████▌    | 3576/6470 [44:47<36:29,  1.32it/s]
2|3577|Loss: 0.7901223301887512:  55%|█████▌    | 3576/6470 [44:48<36:29,  1.32it/s]
2|3577|Loss: 0.7901223301887512:  55%|█████▌    | 3577/6470 [44:48<37:58,  1.27it/s]
2|3578|Loss: 0.8285238146781921:  55%|█████▌    | 3577/6470 [44:49<37:58,  1.27it/s]
2|3578|Loss: 0.8285238146781921:  55%|█████▌    | 3578/6470 [44:49<37:59,  1.27it/s]
2|3579|Loss: 0.8443784713745117:  55%|█████▌    | 3578/6470 [44:50<37:59,  1.27it/s]
2|3579|Loss: 0.8443784713745117:  55%|█████▌    | 3579/6470 [44:50<39:31,  1.22it/s]
2|3580|Loss: 0.7621660828590393:  55%|█████▌    | 3579/6470 [44:51<39:31,  1.22it/s]
2|3580|Loss: 0.7621660828590393:  55%|█████▌    | 3580/6470 [44:51<43:09,  1.12it/s]
2|3581|Loss: 0.8506096005439758:  55%|█████▌    | 3580/6470 [44:52<43:09,  1.12it/s]
2|3581|Loss: 0.8506096005439758:  55%|█████▌    | 3581/6470 [44:52<44:41,  1.08it/s]
2|3582|Loss: 0.8853750824928284:  55%|█████▌    | 3581/6470 [44:53<44:41,  1.08it/s]
2|3582|Loss: 0.8853750824928284:  55%|█████▌    | 3582/6470 [44:53<43:15,  1.11it/s]
2|3583|Loss: 0.7958756685256958:  55%|█████▌    | 3582/6470 [44:53<43:15,  1.11it/s]
2|3583|Loss: 0.7958756685256958:  55%|█████▌    | 3583/6470 [44:54<42:24,  1.13it/s]
2|3584|Loss: 0.9445485472679138:  55%|█████▌    | 3583/6470 [44:54<42:24,  1.13it/s]
2|3584|Loss: 0.9445485472679138:  55%|█████▌    | 3584/6470 [44:54<39:50,  1.21it/s]
2|3585|Loss: 0.9743107557296753:  55%|█████▌    | 3584/6470 [44:55<39:50,  1.21it/s]
2|3585|Loss: 0.9743107557296753:  55%|█████▌    | 3585/6470 [44:55<38:52,  1.24it/s]
2|3586|Loss: 0.8927950263023376:  55%|█████▌    | 3585/6470 [44:56<38:52,  1.24it/s]
2|3586|Loss: 0.8927950263023376:  55%|█████▌    | 3586/6470 [44:56<35:39,  1.35it/s]
2|3587|Loss: 0.8831505179405212:  55%|█████▌    | 3586/6470 [44:56<35:39,  1.35it/s]
2|3587|Loss: 0.8831505179405212:  55%|█████▌    | 3587/6470 [44:57<35:36,  1.35it/s]
2|3588|Loss: 0.620692789554596:  55%|█████▌    | 3587/6470 [44:57<35:36,  1.35it/s] 
2|3588|Loss: 0.620692789554596:  55%|█████▌    | 3588/6470 [44:57<37:14,  1.29it/s]
2|3589|Loss: 0.9368196129798889:  55%|█████▌    | 3588/6470 [44:58<37:14,  1.29it/s]
2|3589|Loss: 0.9368196129798889:  55%|█████▌    | 3589/6470 [44:58<36:37,  1.31it/s]
2|3590|Loss: 0.7535485625267029:  55%|█████▌    | 3589/6470 [44:59<36:37,  1.31it/s]
2|3590|Loss: 0.7535485625267029:  55%|█████▌    | 3590/6470 [44:59<39:08,  1.23it/s]
2|3591|Loss: 0.8866817355155945:  55%|█████▌    | 3590/6470 [45:00<39:08,  1.23it/s]
2|3591|Loss: 0.8866817355155945:  56%|█████▌    | 3591/6470 [45:00<40:49,  1.18it/s]
2|3592|Loss: 1.107992172241211:  56%|█████▌    | 3591/6470 [45:01<40:49,  1.18it/s] 
2|3592|Loss: 1.107992172241211:  56%|█████▌    | 3592/6470 [45:01<41:54,  1.14it/s]
2|3593|Loss: 0.8354628682136536:  56%|█████▌    | 3592/6470 [45:01<41:54,  1.14it/s]
2|3593|Loss: 0.8354628682136536:  56%|█████▌    | 3593/6470 [45:02<40:12,  1.19it/s]
2|3594|Loss: 0.8052151203155518:  56%|█████▌    | 3593/6470 [45:02<40:12,  1.19it/s]
2|3594|Loss: 0.8052151203155518:  56%|█████▌    | 3594/6470 [45:03<40:29,  1.18it/s]
2|3595|Loss: 0.9233877062797546:  56%|█████▌    | 3594/6470 [45:03<40:29,  1.18it/s]
2|3595|Loss: 0.9233877062797546:  56%|█████▌    | 3595/6470 [45:03<40:44,  1.18it/s]
2|3596|Loss: 0.8553079962730408:  56%|█████▌    | 3595/6470 [45:04<40:44,  1.18it/s]
2|3596|Loss: 0.8553079962730408:  56%|█████▌    | 3596/6470 [45:04<40:14,  1.19it/s]
2|3597|Loss: 0.8282642960548401:  56%|█████▌    | 3596/6470 [45:05<40:14,  1.19it/s]
2|3597|Loss: 0.8282642960548401:  56%|█████▌    | 3597/6470 [45:05<38:14,  1.25it/s]
2|3598|Loss: 0.8352925777435303:  56%|█████▌    | 3597/6470 [45:05<38:14,  1.25it/s]
2|3598|Loss: 0.8352925777435303:  56%|█████▌    | 3598/6470 [45:06<35:11,  1.36it/s]
2|3599|Loss: 0.7687249183654785:  56%|█████▌    | 3598/6470 [45:06<35:11,  1.36it/s]
2|3599|Loss: 0.7687249183654785:  56%|█████▌    | 3599/6470 [45:06<35:08,  1.36it/s]
2|3600|Loss: 0.8601697683334351:  56%|█████▌    | 3599/6470 [45:07<35:08,  1.36it/s]
2|3600|Loss: 0.8601697683334351:  56%|█████▌    | 3600/6470 [45:07<34:28,  1.39it/s]
2|3601|Loss: 0.8601315021514893:  56%|█████▌    | 3600/6470 [45:07<34:28,  1.39it/s]
2|3601|Loss: 0.8601315021514893:  56%|█████▌    | 3601/6470 [45:08<33:35,  1.42it/s]
2|3602|Loss: 0.8459482192993164:  56%|█████▌    | 3601/6470 [45:08<33:35,  1.42it/s]
2|3602|Loss: 0.8459482192993164:  56%|█████▌    | 3602/6470 [45:08<34:23,  1.39it/s]
2|3603|Loss: 0.8392891883850098:  56%|█████▌    | 3602/6470 [45:09<34:23,  1.39it/s]
2|3603|Loss: 0.8392891883850098:  56%|█████▌    | 3603/6470 [45:09<35:19,  1.35it/s]
2|3604|Loss: 0.743760347366333:  56%|█████▌    | 3603/6470 [45:10<35:19,  1.35it/s] 
2|3604|Loss: 0.743760347366333:  56%|█████▌    | 3604/6470 [45:10<35:37,  1.34it/s]
2|3605|Loss: 0.9194656014442444:  56%|█████▌    | 3604/6470 [45:10<35:37,  1.34it/s]
2|3605|Loss: 0.9194656014442444:  56%|█████▌    | 3605/6470 [45:11<33:45,  1.41it/s]
2|3606|Loss: 0.8255481719970703:  56%|█████▌    | 3605/6470 [45:11<33:45,  1.41it/s]
2|3606|Loss: 0.8255481719970703:  56%|█████▌    | 3606/6470 [45:11<31:24,  1.52it/s]
2|3607|Loss: 0.779175341129303:  56%|█████▌    | 3606/6470 [45:11<31:24,  1.52it/s] 
2|3607|Loss: 0.779175341129303:  56%|█████▌    | 3607/6470 [45:12<29:56,  1.59it/s]
2|3608|Loss: 0.8654699921607971:  56%|█████▌    | 3607/6470 [45:12<29:56,  1.59it/s]
2|3608|Loss: 0.8654699921607971:  56%|█████▌    | 3608/6470 [45:12<30:03,  1.59it/s]
2|3609|Loss: 0.8523043394088745:  56%|█████▌    | 3608/6470 [45:13<30:03,  1.59it/s]
2|3609|Loss: 0.8523043394088745:  56%|█████▌    | 3609/6470 [45:13<33:13,  1.44it/s]
2|3610|Loss: 0.8129134178161621:  56%|█████▌    | 3609/6470 [45:14<33:13,  1.44it/s]
2|3610|Loss: 0.8129134178161621:  56%|█████▌    | 3610/6470 [45:14<34:32,  1.38it/s]
2|3611|Loss: 0.8776872158050537:  56%|█████▌    | 3610/6470 [45:14<34:32,  1.38it/s]
2|3611|Loss: 0.8776872158050537:  56%|█████▌    | 3611/6470 [45:15<36:42,  1.30it/s]
2|3612|Loss: 0.741870641708374:  56%|█████▌    | 3611/6470 [45:15<36:42,  1.30it/s] 
2|3612|Loss: 0.741870641708374:  56%|█████▌    | 3612/6470 [45:15<34:14,  1.39it/s]
2|3613|Loss: 0.8701037168502808:  56%|█████▌    | 3612/6470 [45:16<34:14,  1.39it/s]
2|3613|Loss: 0.8701037168502808:  56%|█████▌    | 3613/6470 [45:16<34:36,  1.38it/s]
2|3614|Loss: 0.6579886674880981:  56%|█████▌    | 3613/6470 [45:17<34:36,  1.38it/s]
2|3614|Loss: 0.6579886674880981:  56%|█████▌    | 3614/6470 [45:17<34:14,  1.39it/s]
2|3615|Loss: 0.6683555245399475:  56%|█████▌    | 3614/6470 [45:17<34:14,  1.39it/s]
2|3615|Loss: 0.6683555245399475:  56%|█████▌    | 3615/6470 [45:18<37:27,  1.27it/s]
2|3616|Loss: 0.6945996284484863:  56%|█████▌    | 3615/6470 [45:18<37:27,  1.27it/s]
2|3616|Loss: 0.6945996284484863:  56%|█████▌    | 3616/6470 [45:19<37:44,  1.26it/s]
2|3617|Loss: 0.8319088220596313:  56%|█████▌    | 3616/6470 [45:19<37:44,  1.26it/s]
2|3617|Loss: 0.8319088220596313:  56%|█████▌    | 3617/6470 [45:19<36:08,  1.32it/s]
2|3618|Loss: 0.824396014213562:  56%|█████▌    | 3617/6470 [45:20<36:08,  1.32it/s] 
2|3618|Loss: 0.824396014213562:  56%|█████▌    | 3618/6470 [45:20<31:28,  1.51it/s]
2|3619|Loss: 0.7060801982879639:  56%|█████▌    | 3618/6470 [45:20<31:28,  1.51it/s]
2|3619|Loss: 0.7060801982879639:  56%|█████▌    | 3619/6470 [45:20<32:08,  1.48it/s]
2|3620|Loss: 0.8721408843994141:  56%|█████▌    | 3619/6470 [45:21<32:08,  1.48it/s]
2|3620|Loss: 0.8721408843994141:  56%|█████▌    | 3620/6470 [45:21<34:57,  1.36it/s]
2|3621|Loss: 0.9382529258728027:  56%|█████▌    | 3620/6470 [45:22<34:57,  1.36it/s]
2|3621|Loss: 0.9382529258728027:  56%|█████▌    | 3621/6470 [45:22<37:28,  1.27it/s]
2|3622|Loss: 0.9291444420814514:  56%|█████▌    | 3621/6470 [45:23<37:28,  1.27it/s]
2|3622|Loss: 0.9291444420814514:  56%|█████▌    | 3622/6470 [45:23<38:05,  1.25it/s]
2|3623|Loss: 0.8278945684432983:  56%|█████▌    | 3622/6470 [45:23<38:05,  1.25it/s]
2|3623|Loss: 0.8278945684432983:  56%|█████▌    | 3623/6470 [45:24<37:42,  1.26it/s]
2|3624|Loss: 0.9144718050956726:  56%|█████▌    | 3623/6470 [45:24<37:42,  1.26it/s]
2|3624|Loss: 0.9144718050956726:  56%|█████▌    | 3624/6470 [45:25<36:35,  1.30it/s]
2|3625|Loss: 0.9860532283782959:  56%|█████▌    | 3624/6470 [45:25<36:35,  1.30it/s]
2|3625|Loss: 0.9860532283782959:  56%|█████▌    | 3625/6470 [45:25<38:01,  1.25it/s]
2|3626|Loss: 0.7433176040649414:  56%|█████▌    | 3625/6470 [45:26<38:01,  1.25it/s]
2|3626|Loss: 0.7433176040649414:  56%|█████▌    | 3626/6470 [45:26<39:52,  1.19it/s]
2|3627|Loss: 0.7334121465682983:  56%|█████▌    | 3626/6470 [45:27<39:52,  1.19it/s]
2|3627|Loss: 0.7334121465682983:  56%|█████▌    | 3627/6470 [45:27<37:53,  1.25it/s]
2|3628|Loss: 0.7388063669204712:  56%|█████▌    | 3627/6470 [45:28<37:53,  1.25it/s]
2|3628|Loss: 0.7388063669204712:  56%|█████▌    | 3628/6470 [45:28<40:06,  1.18it/s]
2|3629|Loss: 0.9042576551437378:  56%|█████▌    | 3628/6470 [45:29<40:06,  1.18it/s]
2|3629|Loss: 0.9042576551437378:  56%|█████▌    | 3629/6470 [45:29<40:00,  1.18it/s]
2|3630|Loss: 0.852575421333313:  56%|█████▌    | 3629/6470 [45:29<40:00,  1.18it/s] 
2|3630|Loss: 0.852575421333313:  56%|█████▌    | 3630/6470 [45:30<40:44,  1.16it/s]
2|3631|Loss: 0.8455990552902222:  56%|█████▌    | 3630/6470 [45:30<40:44,  1.16it/s]
2|3631|Loss: 0.8455990552902222:  56%|█████▌    | 3631/6470 [45:31<42:25,  1.12it/s]
2|3632|Loss: 0.8947803378105164:  56%|█████▌    | 3631/6470 [45:31<42:25,  1.12it/s]
2|3632|Loss: 0.8947803378105164:  56%|█████▌    | 3632/6470 [45:32<40:35,  1.17it/s]
2|3633|Loss: 0.8842083215713501:  56%|█████▌    | 3632/6470 [45:32<40:35,  1.17it/s]
2|3633|Loss: 0.8842083215713501:  56%|█████▌    | 3633/6470 [45:32<40:58,  1.15it/s]
2|3634|Loss: 0.64641934633255:  56%|█████▌    | 3633/6470 [45:33<40:58,  1.15it/s]  
2|3634|Loss: 0.64641934633255:  56%|█████▌    | 3634/6470 [45:33<35:44,  1.32it/s]
2|3635|Loss: 0.7766838073730469:  56%|█████▌    | 3634/6470 [45:33<35:44,  1.32it/s]
2|3635|Loss: 0.7766838073730469:  56%|█████▌    | 3635/6470 [45:34<36:06,  1.31it/s]
2|3636|Loss: 0.8041117787361145:  56%|█████▌    | 3635/6470 [45:34<36:06,  1.31it/s]
2|3636|Loss: 0.8041117787361145:  56%|█████▌    | 3636/6470 [45:34<36:19,  1.30it/s]
2|3637|Loss: 0.8223019242286682:  56%|█████▌    | 3636/6470 [45:35<36:19,  1.30it/s]
2|3637|Loss: 0.8223019242286682:  56%|█████▌    | 3637/6470 [45:35<38:34,  1.22it/s]
2|3638|Loss: 0.7394556403160095:  56%|█████▌    | 3637/6470 [45:36<38:34,  1.22it/s]
2|3638|Loss: 0.7394556403160095:  56%|█████▌    | 3638/6470 [45:36<38:56,  1.21it/s]
2|3639|Loss: 0.736566960811615:  56%|█████▌    | 3638/6470 [45:37<38:56,  1.21it/s] 
2|3639|Loss: 0.736566960811615:  56%|█████▌    | 3639/6470 [45:37<37:15,  1.27it/s]
2|3640|Loss: 0.8641290068626404:  56%|█████▌    | 3639/6470 [45:37<37:15,  1.27it/s]
2|3640|Loss: 0.8641290068626404:  56%|█████▋    | 3640/6470 [45:38<38:04,  1.24it/s]
2|3641|Loss: 0.7266995906829834:  56%|█████▋    | 3640/6470 [45:38<38:04,  1.24it/s]
2|3641|Loss: 0.7266995906829834:  56%|█████▋    | 3641/6470 [45:39<37:31,  1.26it/s]
2|3642|Loss: 0.773061215877533:  56%|█████▋    | 3641/6470 [45:39<37:31,  1.26it/s] 
2|3642|Loss: 0.773061215877533:  56%|█████▋    | 3642/6470 [45:39<38:57,  1.21it/s]
2|3643|Loss: 0.88566654920578:  56%|█████▋    | 3642/6470 [45:40<38:57,  1.21it/s] 
2|3643|Loss: 0.88566654920578:  56%|█████▋    | 3643/6470 [45:40<37:05,  1.27it/s]
2|3644|Loss: 0.8501297235488892:  56%|█████▋    | 3643/6470 [45:41<37:05,  1.27it/s]
2|3644|Loss: 0.8501297235488892:  56%|█████▋    | 3644/6470 [45:41<39:04,  1.21it/s]
2|3645|Loss: 0.9366329312324524:  56%|█████▋    | 3644/6470 [45:42<39:04,  1.21it/s]
2|3645|Loss: 0.9366329312324524:  56%|█████▋    | 3645/6470 [45:42<39:01,  1.21it/s]
2|3646|Loss: 0.7178805470466614:  56%|█████▋    | 3645/6470 [45:42<39:01,  1.21it/s]
2|3646|Loss: 0.7178805470466614:  56%|█████▋    | 3646/6470 [45:43<36:34,  1.29it/s]
2|3647|Loss: 0.6218761801719666:  56%|█████▋    | 3646/6470 [45:43<36:34,  1.29it/s]
2|3647|Loss: 0.6218761801719666:  56%|█████▋    | 3647/6470 [45:43<36:24,  1.29it/s]
2|3648|Loss: 0.9532164335250854:  56%|█████▋    | 3647/6470 [45:44<36:24,  1.29it/s]
2|3648|Loss: 0.9532164335250854:  56%|█████▋    | 3648/6470 [45:44<37:29,  1.25it/s]
2|3649|Loss: 0.7610759139060974:  56%|█████▋    | 3648/6470 [45:45<37:29,  1.25it/s]
2|3649|Loss: 0.7610759139060974:  56%|█████▋    | 3649/6470 [45:45<37:42,  1.25it/s]
2|3650|Loss: 0.8542605042457581:  56%|█████▋    | 3649/6470 [45:45<37:42,  1.25it/s]
2|3650|Loss: 0.8542605042457581:  56%|█████▋    | 3650/6470 [45:46<39:57,  1.18it/s]
2|3651|Loss: 0.8108749985694885:  56%|█████▋    | 3650/6470 [45:46<39:57,  1.18it/s]
2|3651|Loss: 0.8108749985694885:  56%|█████▋    | 3651/6470 [45:47<40:39,  1.16it/s]
2|3652|Loss: 0.8138022422790527:  56%|█████▋    | 3651/6470 [45:47<40:39,  1.16it/s]
2|3652|Loss: 0.8138022422790527:  56%|█████▋    | 3652/6470 [45:48<39:53,  1.18it/s]
2|3653|Loss: 0.9491909146308899:  56%|█████▋    | 3652/6470 [45:48<39:53,  1.18it/s]
2|3653|Loss: 0.9491909146308899:  56%|█████▋    | 3653/6470 [45:48<39:09,  1.20it/s]
2|3654|Loss: 0.8473499417304993:  56%|█████▋    | 3653/6470 [45:49<39:09,  1.20it/s]
2|3654|Loss: 0.8473499417304993:  56%|█████▋    | 3654/6470 [45:49<37:22,  1.26it/s]
2|3655|Loss: 0.8399414420127869:  56%|█████▋    | 3654/6470 [45:50<37:22,  1.26it/s]
2|3655|Loss: 0.8399414420127869:  56%|█████▋    | 3655/6470 [45:50<35:51,  1.31it/s]
2|3656|Loss: 0.7529364824295044:  56%|█████▋    | 3655/6470 [45:50<35:51,  1.31it/s]
2|3656|Loss: 0.7529364824295044:  57%|█████▋    | 3656/6470 [45:50<32:59,  1.42it/s]
2|3657|Loss: 0.821690022945404:  57%|█████▋    | 3656/6470 [45:51<32:59,  1.42it/s] 
2|3657|Loss: 0.821690022945404:  57%|█████▋    | 3657/6470 [45:51<30:23,  1.54it/s]
2|3658|Loss: 0.988637387752533:  57%|█████▋    | 3657/6470 [45:51<30:23,  1.54it/s]
2|3658|Loss: 0.988637387752533:  57%|█████▋    | 3658/6470 [45:52<31:07,  1.51it/s]
2|3659|Loss: 0.822019636631012:  57%|█████▋    | 3658/6470 [45:52<31:07,  1.51it/s]
2|3659|Loss: 0.822019636631012:  57%|█████▋    | 3659/6470 [45:52<32:57,  1.42it/s]
2|3660|Loss: 0.7846243977546692:  57%|█████▋    | 3659/6470 [45:53<32:57,  1.42it/s]
2|3660|Loss: 0.7846243977546692:  57%|█████▋    | 3660/6470 [45:53<34:22,  1.36it/s]
2|3661|Loss: 0.9147054553031921:  57%|█████▋    | 3660/6470 [45:54<34:22,  1.36it/s]
2|3661|Loss: 0.9147054553031921:  57%|█████▋    | 3661/6470 [45:54<32:33,  1.44it/s]
2|3662|Loss: 0.7933629155158997:  57%|█████▋    | 3661/6470 [45:54<32:33,  1.44it/s]
2|3662|Loss: 0.7933629155158997:  57%|█████▋    | 3662/6470 [45:54<31:29,  1.49it/s]
2|3663|Loss: 0.7523185014724731:  57%|█████▋    | 3662/6470 [45:55<31:29,  1.49it/s]
2|3663|Loss: 0.7523185014724731:  57%|█████▋    | 3663/6470 [45:55<35:38,  1.31it/s]
2|3664|Loss: 0.7325813174247742:  57%|█████▋    | 3663/6470 [45:56<35:38,  1.31it/s]
2|3664|Loss: 0.7325813174247742:  57%|█████▋    | 3664/6470 [45:56<34:45,  1.35it/s]
2|3665|Loss: 0.8593278527259827:  57%|█████▋    | 3664/6470 [45:56<34:45,  1.35it/s]
2|3665|Loss: 0.8593278527259827:  57%|█████▋    | 3665/6470 [45:57<33:39,  1.39it/s]
2|3666|Loss: 1.0528430938720703:  57%|█████▋    | 3665/6470 [45:57<33:39,  1.39it/s]
2|3666|Loss: 1.0528430938720703:  57%|█████▋    | 3666/6470 [45:58<35:29,  1.32it/s]
2|3667|Loss: 0.7832450866699219:  57%|█████▋    | 3666/6470 [45:58<35:29,  1.32it/s]
2|3667|Loss: 0.7832450866699219:  57%|█████▋    | 3667/6470 [45:59<37:06,  1.26it/s]
2|3668|Loss: 0.8994954228401184:  57%|█████▋    | 3667/6470 [45:59<37:06,  1.26it/s]
2|3668|Loss: 0.8994954228401184:  57%|█████▋    | 3668/6470 [45:59<37:23,  1.25it/s]
2|3669|Loss: 0.788506805896759:  57%|█████▋    | 3668/6470 [46:00<37:23,  1.25it/s] 
2|3669|Loss: 0.788506805896759:  57%|█████▋    | 3669/6470 [46:00<35:13,  1.33it/s]
2|3670|Loss: 0.8348469138145447:  57%|█████▋    | 3669/6470 [46:00<35:13,  1.33it/s]
2|3670|Loss: 0.8348469138145447:  57%|█████▋    | 3670/6470 [46:01<35:02,  1.33it/s]
2|3671|Loss: 1.0358731746673584:  57%|█████▋    | 3670/6470 [46:01<35:02,  1.33it/s]
2|3671|Loss: 1.0358731746673584:  57%|█████▋    | 3671/6470 [46:01<32:05,  1.45it/s]
2|3672|Loss: 1.1292896270751953:  57%|█████▋    | 3671/6470 [46:02<32:05,  1.45it/s]
2|3672|Loss: 1.1292896270751953:  57%|█████▋    | 3672/6470 [46:02<33:43,  1.38it/s]
2|3673|Loss: 0.938527524471283:  57%|█████▋    | 3672/6470 [46:02<33:43,  1.38it/s] 
2|3673|Loss: 0.938527524471283:  57%|█████▋    | 3673/6470 [46:03<33:29,  1.39it/s]
2|3674|Loss: 0.7904676198959351:  57%|█████▋    | 3673/6470 [46:03<33:29,  1.39it/s]
2|3674|Loss: 0.7904676198959351:  57%|█████▋    | 3674/6470 [46:04<35:26,  1.32it/s]
2|3675|Loss: 0.8116588592529297:  57%|█████▋    | 3674/6470 [46:04<35:26,  1.32it/s]
2|3675|Loss: 0.8116588592529297:  57%|█████▋    | 3675/6470 [46:04<35:40,  1.31it/s]
2|3676|Loss: 0.7458739876747131:  57%|█████▋    | 3675/6470 [46:05<35:40,  1.31it/s]
2|3676|Loss: 0.7458739876747131:  57%|█████▋    | 3676/6470 [46:05<33:15,  1.40it/s]
2|3677|Loss: 0.8451880812644958:  57%|█████▋    | 3676/6470 [46:05<33:15,  1.40it/s]
2|3677|Loss: 0.8451880812644958:  57%|█████▋    | 3677/6470 [46:06<33:34,  1.39it/s]
2|3678|Loss: 0.8678469657897949:  57%|█████▋    | 3677/6470 [46:06<33:34,  1.39it/s]
2|3678|Loss: 0.8678469657897949:  57%|█████▋    | 3678/6470 [46:06<33:20,  1.40it/s]
2|3679|Loss: 0.6819442510604858:  57%|█████▋    | 3678/6470 [46:07<33:20,  1.40it/s]
2|3679|Loss: 0.6819442510604858:  57%|█████▋    | 3679/6470 [46:07<30:34,  1.52it/s]
2|3680|Loss: 0.8503239750862122:  57%|█████▋    | 3679/6470 [46:07<30:34,  1.52it/s]
2|3680|Loss: 0.8503239750862122:  57%|█████▋    | 3680/6470 [46:08<30:05,  1.55it/s]
2|3681|Loss: 0.6740872263908386:  57%|█████▋    | 3680/6470 [46:08<30:05,  1.55it/s]
2|3681|Loss: 0.6740872263908386:  57%|█████▋    | 3681/6470 [46:08<30:47,  1.51it/s]
2|3682|Loss: 0.9112390279769897:  57%|█████▋    | 3681/6470 [46:09<30:47,  1.51it/s]
2|3682|Loss: 0.9112390279769897:  57%|█████▋    | 3682/6470 [46:09<32:26,  1.43it/s]
2|3683|Loss: 0.8046967387199402:  57%|█████▋    | 3682/6470 [46:09<32:26,  1.43it/s]
2|3683|Loss: 0.8046967387199402:  57%|█████▋    | 3683/6470 [46:10<30:45,  1.51it/s]
2|3684|Loss: 0.9596741199493408:  57%|█████▋    | 3683/6470 [46:10<30:45,  1.51it/s]
2|3684|Loss: 0.9596741199493408:  57%|█████▋    | 3684/6470 [46:10<31:18,  1.48it/s]
2|3685|Loss: 0.8038347959518433:  57%|█████▋    | 3684/6470 [46:11<31:18,  1.48it/s]
2|3685|Loss: 0.8038347959518433:  57%|█████▋    | 3685/6470 [46:11<31:17,  1.48it/s]
2|3686|Loss: 0.8815663456916809:  57%|█████▋    | 3685/6470 [46:11<31:17,  1.48it/s]
2|3686|Loss: 0.8815663456916809:  57%|█████▋    | 3686/6470 [46:12<29:10,  1.59it/s]
2|3687|Loss: 0.8857045769691467:  57%|█████▋    | 3686/6470 [46:12<29:10,  1.59it/s]
2|3687|Loss: 0.8857045769691467:  57%|█████▋    | 3687/6470 [46:12<30:42,  1.51it/s]
2|3688|Loss: 0.7295597791671753:  57%|█████▋    | 3687/6470 [46:13<30:42,  1.51it/s]
2|3688|Loss: 0.7295597791671753:  57%|█████▋    | 3688/6470 [46:13<29:50,  1.55it/s]
2|3689|Loss: 0.8226482272148132:  57%|█████▋    | 3688/6470 [46:13<29:50,  1.55it/s]
2|3689|Loss: 0.8226482272148132:  57%|█████▋    | 3689/6470 [46:14<30:07,  1.54it/s]
2|3690|Loss: 0.715200662612915:  57%|█████▋    | 3689/6470 [46:14<30:07,  1.54it/s] 
2|3690|Loss: 0.715200662612915:  57%|█████▋    | 3690/6470 [46:14<30:32,  1.52it/s]
2|3691|Loss: 0.826289713382721:  57%|█████▋    | 3690/6470 [46:15<30:32,  1.52it/s]
2|3691|Loss: 0.826289713382721:  57%|█████▋    | 3691/6470 [46:15<31:37,  1.46it/s]
2|3692|Loss: 0.8261313438415527:  57%|█████▋    | 3691/6470 [46:15<31:37,  1.46it/s]
2|3692|Loss: 0.8261313438415527:  57%|█████▋    | 3692/6470 [46:16<31:02,  1.49it/s]
2|3693|Loss: 0.8434827923774719:  57%|█████▋    | 3692/6470 [46:16<31:02,  1.49it/s]
2|3693|Loss: 0.8434827923774719:  57%|█████▋    | 3693/6470 [46:16<30:10,  1.53it/s]
2|3694|Loss: 0.9929824471473694:  57%|█████▋    | 3693/6470 [46:17<30:10,  1.53it/s]
2|3694|Loss: 0.9929824471473694:  57%|█████▋    | 3694/6470 [46:17<32:45,  1.41it/s]
2|3695|Loss: 0.8375384211540222:  57%|█████▋    | 3694/6470 [46:17<32:45,  1.41it/s]
2|3695|Loss: 0.8375384211540222:  57%|█████▋    | 3695/6470 [46:18<31:50,  1.45it/s]
2|3696|Loss: 0.7726544737815857:  57%|█████▋    | 3695/6470 [46:18<31:50,  1.45it/s]
2|3696|Loss: 0.7726544737815857:  57%|█████▋    | 3696/6470 [46:19<34:09,  1.35it/s]
2|3697|Loss: 0.8363421559333801:  57%|█████▋    | 3696/6470 [46:19<34:09,  1.35it/s]
2|3697|Loss: 0.8363421559333801:  57%|█████▋    | 3697/6470 [46:19<36:46,  1.26it/s]
2|3698|Loss: 0.8627633452415466:  57%|█████▋    | 3697/6470 [46:20<36:46,  1.26it/s]
2|3698|Loss: 0.8627633452415466:  57%|█████▋    | 3698/6470 [46:20<36:49,  1.25it/s]
2|3699|Loss: 0.7325204610824585:  57%|█████▋    | 3698/6470 [46:21<36:49,  1.25it/s]
2|3699|Loss: 0.7325204610824585:  57%|█████▋    | 3699/6470 [46:21<38:31,  1.20it/s]
2|3700|Loss: 0.8933078050613403:  57%|█████▋    | 3699/6470 [46:22<38:31,  1.20it/s]
2|3700|Loss: 0.8933078050613403:  57%|█████▋    | 3700/6470 [46:22<38:22,  1.20it/s]
2|3701|Loss: 0.6420716047286987:  57%|█████▋    | 3700/6470 [46:23<38:22,  1.20it/s]
2|3701|Loss: 0.6420716047286987:  57%|█████▋    | 3701/6470 [46:23<40:44,  1.13it/s]
2|3702|Loss: 0.761440098285675:  57%|█████▋    | 3701/6470 [46:24<40:44,  1.13it/s] 
2|3702|Loss: 0.761440098285675:  57%|█████▋    | 3702/6470 [46:24<39:54,  1.16it/s]
2|3703|Loss: 0.7908402681350708:  57%|█████▋    | 3702/6470 [46:24<39:54,  1.16it/s]
2|3703|Loss: 0.7908402681350708:  57%|█████▋    | 3703/6470 [46:25<39:29,  1.17it/s]
2|3704|Loss: 0.7884778380393982:  57%|█████▋    | 3703/6470 [46:25<39:29,  1.17it/s]
2|3704|Loss: 0.7884778380393982:  57%|█████▋    | 3704/6470 [46:26<40:17,  1.14it/s]
2|3705|Loss: 0.8265674710273743:  57%|█████▋    | 3704/6470 [46:26<40:17,  1.14it/s]
2|3705|Loss: 0.8265674710273743:  57%|█████▋    | 3705/6470 [46:26<39:26,  1.17it/s]
2|3706|Loss: 0.799996018409729:  57%|█████▋    | 3705/6470 [46:27<39:26,  1.17it/s] 
2|3706|Loss: 0.799996018409729:  57%|█████▋    | 3706/6470 [46:27<37:04,  1.24it/s]
2|3707|Loss: 0.8106439709663391:  57%|█████▋    | 3706/6470 [46:28<37:04,  1.24it/s]
2|3707|Loss: 0.8106439709663391:  57%|█████▋    | 3707/6470 [46:28<36:39,  1.26it/s]
2|3708|Loss: 0.7346338629722595:  57%|█████▋    | 3707/6470 [46:28<36:39,  1.26it/s]
2|3708|Loss: 0.7346338629722595:  57%|█████▋    | 3708/6470 [46:29<35:42,  1.29it/s]
2|3709|Loss: 0.885219156742096:  57%|█████▋    | 3708/6470 [46:29<35:42,  1.29it/s] 
2|3709|Loss: 0.885219156742096:  57%|█████▋    | 3709/6470 [46:29<35:36,  1.29it/s]
2|3710|Loss: 0.8876779079437256:  57%|█████▋    | 3709/6470 [46:30<35:36,  1.29it/s]
2|3710|Loss: 0.8876779079437256:  57%|█████▋    | 3710/6470 [46:30<35:31,  1.29it/s]
2|3711|Loss: 0.9281511306762695:  57%|█████▋    | 3710/6470 [46:31<35:31,  1.29it/s]
2|3711|Loss: 0.9281511306762695:  57%|█████▋    | 3711/6470 [46:31<36:29,  1.26it/s]
2|3712|Loss: 0.9615576267242432:  57%|█████▋    | 3711/6470 [46:31<36:29,  1.26it/s]
2|3712|Loss: 0.9615576267242432:  57%|█████▋    | 3712/6470 [46:32<37:16,  1.23it/s]
2|3713|Loss: 0.6279134154319763:  57%|█████▋    | 3712/6470 [46:32<37:16,  1.23it/s]
2|3713|Loss: 0.6279134154319763:  57%|█████▋    | 3713/6470 [46:32<32:41,  1.41it/s]
2|3714|Loss: 0.705100417137146:  57%|█████▋    | 3713/6470 [46:33<32:41,  1.41it/s] 
2|3714|Loss: 0.705100417137146:  57%|█████▋    | 3714/6470 [46:33<30:09,  1.52it/s]
2|3715|Loss: 0.7843388319015503:  57%|█████▋    | 3714/6470 [46:33<30:09,  1.52it/s]
2|3715|Loss: 0.7843388319015503:  57%|█████▋    | 3715/6470 [46:34<30:13,  1.52it/s]
2|3716|Loss: 0.7907542586326599:  57%|█████▋    | 3715/6470 [46:34<30:13,  1.52it/s]
2|3716|Loss: 0.7907542586326599:  57%|█████▋    | 3716/6470 [46:34<31:27,  1.46it/s]
2|3717|Loss: 0.9912267923355103:  57%|█████▋    | 3716/6470 [46:35<31:27,  1.46it/s]
2|3717|Loss: 0.9912267923355103:  57%|█████▋    | 3717/6470 [46:35<33:38,  1.36it/s]
2|3718|Loss: 0.8390573859214783:  57%|█████▋    | 3717/6470 [46:36<33:38,  1.36it/s]
2|3718|Loss: 0.8390573859214783:  57%|█████▋    | 3718/6470 [46:36<33:18,  1.38it/s]
2|3719|Loss: 0.8258416056632996:  57%|█████▋    | 3718/6470 [46:36<33:18,  1.38it/s]
2|3719|Loss: 0.8258416056632996:  57%|█████▋    | 3719/6470 [46:37<32:52,  1.39it/s]
2|3720|Loss: 0.8284006714820862:  57%|█████▋    | 3719/6470 [46:37<32:52,  1.39it/s]
2|3720|Loss: 0.8284006714820862:  57%|█████▋    | 3720/6470 [46:37<33:51,  1.35it/s]
2|3721|Loss: 0.8908383250236511:  57%|█████▋    | 3720/6470 [46:38<33:51,  1.35it/s]
2|3721|Loss: 0.8908383250236511:  58%|█████▊    | 3721/6470 [46:38<34:21,  1.33it/s]
2|3722|Loss: 0.7989204525947571:  58%|█████▊    | 3721/6470 [46:39<34:21,  1.33it/s]
2|3722|Loss: 0.7989204525947571:  58%|█████▊    | 3722/6470 [46:39<34:26,  1.33it/s]
2|3723|Loss: 0.8172470927238464:  58%|█████▊    | 3722/6470 [46:39<34:26,  1.33it/s]
2|3723|Loss: 0.8172470927238464:  58%|█████▊    | 3723/6470 [46:40<36:08,  1.27it/s]
2|3724|Loss: 0.9018893241882324:  58%|█████▊    | 3723/6470 [46:40<36:08,  1.27it/s]
2|3724|Loss: 0.9018893241882324:  58%|█████▊    | 3724/6470 [46:40<33:05,  1.38it/s]
2|3725|Loss: 0.636623740196228:  58%|█████▊    | 3724/6470 [46:41<33:05,  1.38it/s] 
2|3725|Loss: 0.636623740196228:  58%|█████▊    | 3725/6470 [46:41<32:46,  1.40it/s]
2|3726|Loss: 0.8745917081832886:  58%|█████▊    | 3725/6470 [46:41<32:46,  1.40it/s]
2|3726|Loss: 0.8745917081832886:  58%|█████▊    | 3726/6470 [46:42<33:45,  1.35it/s]
2|3727|Loss: 0.7721759676933289:  58%|█████▊    | 3726/6470 [46:42<33:45,  1.35it/s]
2|3727|Loss: 0.7721759676933289:  58%|█████▊    | 3727/6470 [46:43<34:32,  1.32it/s]
2|3728|Loss: 0.7564020156860352:  58%|█████▊    | 3727/6470 [46:43<34:32,  1.32it/s]
2|3728|Loss: 0.7564020156860352:  58%|█████▊    | 3728/6470 [46:43<33:46,  1.35it/s]
2|3729|Loss: 0.6645574569702148:  58%|█████▊    | 3728/6470 [46:44<33:46,  1.35it/s]
2|3729|Loss: 0.6645574569702148:  58%|█████▊    | 3729/6470 [46:44<32:43,  1.40it/s]
2|3730|Loss: 0.8733119964599609:  58%|█████▊    | 3729/6470 [46:44<32:43,  1.40it/s]
2|3730|Loss: 0.8733119964599609:  58%|█████▊    | 3730/6470 [46:45<32:55,  1.39it/s]
2|3731|Loss: 0.918505072593689:  58%|█████▊    | 3730/6470 [46:45<32:55,  1.39it/s] 
2|3731|Loss: 0.918505072593689:  58%|█████▊    | 3731/6470 [46:45<33:48,  1.35it/s]
2|3732|Loss: 0.79563969373703:  58%|█████▊    | 3731/6470 [46:46<33:48,  1.35it/s] 
2|3732|Loss: 0.79563969373703:  58%|█████▊    | 3732/6470 [46:46<33:13,  1.37it/s]
2|3733|Loss: 0.863158643245697:  58%|█████▊    | 3732/6470 [46:47<33:13,  1.37it/s]
2|3733|Loss: 0.863158643245697:  58%|█████▊    | 3733/6470 [46:47<32:48,  1.39it/s]
2|3734|Loss: 0.790120542049408:  58%|█████▊    | 3733/6470 [46:47<32:48,  1.39it/s]
2|3734|Loss: 0.790120542049408:  58%|█████▊    | 3734/6470 [46:48<33:32,  1.36it/s]
2|3735|Loss: 0.8647472858428955:  58%|█████▊    | 3734/6470 [46:48<33:32,  1.36it/s]
2|3735|Loss: 0.8647472858428955:  58%|█████▊    | 3735/6470 [46:48<31:39,  1.44it/s]
2|3736|Loss: 0.8644912838935852:  58%|█████▊    | 3735/6470 [46:49<31:39,  1.44it/s]
2|3736|Loss: 0.8644912838935852:  58%|█████▊    | 3736/6470 [46:49<32:18,  1.41it/s]
2|3737|Loss: 0.9003945589065552:  58%|█████▊    | 3736/6470 [46:49<32:18,  1.41it/s]
2|3737|Loss: 0.9003945589065552:  58%|█████▊    | 3737/6470 [46:50<33:28,  1.36it/s]
2|3738|Loss: 0.8456001281738281:  58%|█████▊    | 3737/6470 [46:50<33:28,  1.36it/s]
2|3738|Loss: 0.8456001281738281:  58%|█████▊    | 3738/6470 [46:50<32:58,  1.38it/s]
2|3739|Loss: 0.920759379863739:  58%|█████▊    | 3738/6470 [46:51<32:58,  1.38it/s] 
2|3739|Loss: 0.920759379863739:  58%|█████▊    | 3739/6470 [46:51<31:40,  1.44it/s]
2|3740|Loss: 0.7921056747436523:  58%|█████▊    | 3739/6470 [46:51<31:40,  1.44it/s]
2|3740|Loss: 0.7921056747436523:  58%|█████▊    | 3740/6470 [46:52<31:28,  1.45it/s]
2|3741|Loss: 0.8417695164680481:  58%|█████▊    | 3740/6470 [46:52<31:28,  1.45it/s]
2|3741|Loss: 0.8417695164680481:  58%|█████▊    | 3741/6470 [46:52<30:37,  1.49it/s]
2|3742|Loss: 0.9031424522399902:  58%|█████▊    | 3741/6470 [46:53<30:37,  1.49it/s]
2|3742|Loss: 0.9031424522399902:  58%|█████▊    | 3742/6470 [46:53<31:37,  1.44it/s]
2|3743|Loss: 0.7808826565742493:  58%|█████▊    | 3742/6470 [46:54<31:37,  1.44it/s]
2|3743|Loss: 0.7808826565742493:  58%|█████▊    | 3743/6470 [46:54<34:22,  1.32it/s]
2|3744|Loss: 0.8562522530555725:  58%|█████▊    | 3743/6470 [46:55<34:22,  1.32it/s]
2|3744|Loss: 0.8562522530555725:  58%|█████▊    | 3744/6470 [46:55<36:46,  1.24it/s]
2|3745|Loss: 0.9572132229804993:  58%|█████▊    | 3744/6470 [46:55<36:46,  1.24it/s]
2|3745|Loss: 0.9572132229804993:  58%|█████▊    | 3745/6470 [46:56<37:44,  1.20it/s]
2|3746|Loss: 0.7423765659332275:  58%|█████▊    | 3745/6470 [46:56<37:44,  1.20it/s]
2|3746|Loss: 0.7423765659332275:  58%|█████▊    | 3746/6470 [46:57<38:26,  1.18it/s]
2|3747|Loss: 0.8296734094619751:  58%|█████▊    | 3746/6470 [46:57<38:26,  1.18it/s]
2|3747|Loss: 0.8296734094619751:  58%|█████▊    | 3747/6470 [46:57<36:31,  1.24it/s]
2|3748|Loss: 0.6918257474899292:  58%|█████▊    | 3747/6470 [46:58<36:31,  1.24it/s]
2|3748|Loss: 0.6918257474899292:  58%|█████▊    | 3748/6470 [46:58<34:35,  1.31it/s]
2|3749|Loss: 0.631195604801178:  58%|█████▊    | 3748/6470 [46:59<34:35,  1.31it/s] 
2|3749|Loss: 0.631195604801178:  58%|█████▊    | 3749/6470 [46:59<33:17,  1.36it/s]
2|3750|Loss: 0.7563297152519226:  58%|█████▊    | 3749/6470 [46:59<33:17,  1.36it/s]
2|3750|Loss: 0.7563297152519226:  58%|█████▊    | 3750/6470 [47:00<33:14,  1.36it/s]
2|3751|Loss: 0.894202470779419:  58%|█████▊    | 3750/6470 [47:00<33:14,  1.36it/s] 
2|3751|Loss: 0.894202470779419:  58%|█████▊    | 3751/6470 [47:00<31:57,  1.42it/s]
2|3752|Loss: 0.8550307750701904:  58%|█████▊    | 3751/6470 [47:01<31:57,  1.42it/s]
2|3752|Loss: 0.8550307750701904:  58%|█████▊    | 3752/6470 [47:01<31:05,  1.46it/s]
2|3753|Loss: 0.8232302665710449:  58%|█████▊    | 3752/6470 [47:01<31:05,  1.46it/s]
2|3753|Loss: 0.8232302665710449:  58%|█████▊    | 3753/6470 [47:02<32:50,  1.38it/s]
2|3754|Loss: 0.7794026732444763:  58%|█████▊    | 3753/6470 [47:02<32:50,  1.38it/s]
2|3754|Loss: 0.7794026732444763:  58%|█████▊    | 3754/6470 [47:03<36:05,  1.25it/s]
2|3755|Loss: 0.9096221923828125:  58%|█████▊    | 3754/6470 [47:03<36:05,  1.25it/s]
2|3755|Loss: 0.9096221923828125:  58%|█████▊    | 3755/6470 [47:04<38:21,  1.18it/s]
2|3756|Loss: 0.7525634169578552:  58%|█████▊    | 3755/6470 [47:04<38:21,  1.18it/s]
2|3756|Loss: 0.7525634169578552:  58%|█████▊    | 3756/6470 [47:04<38:49,  1.16it/s]
2|3757|Loss: 0.9950965046882629:  58%|█████▊    | 3756/6470 [47:05<38:49,  1.16it/s]
2|3757|Loss: 0.9950965046882629:  58%|█████▊    | 3757/6470 [47:05<38:08,  1.19it/s]
2|3758|Loss: 0.7972006797790527:  58%|█████▊    | 3757/6470 [47:06<38:08,  1.19it/s]
2|3758|Loss: 0.7972006797790527:  58%|█████▊    | 3758/6470 [47:06<34:40,  1.30it/s]
2|3759|Loss: 0.8457698822021484:  58%|█████▊    | 3758/6470 [47:06<34:40,  1.30it/s]
2|3759|Loss: 0.8457698822021484:  58%|█████▊    | 3759/6470 [47:07<34:22,  1.31it/s]
2|3760|Loss: 1.3792706727981567:  58%|█████▊    | 3759/6470 [47:07<34:22,  1.31it/s]
2|3760|Loss: 1.3792706727981567:  58%|█████▊    | 3760/6470 [47:08<41:42,  1.08it/s]
2|3761|Loss: 0.7936415672302246:  58%|█████▊    | 3760/6470 [47:08<41:42,  1.08it/s]
2|3761|Loss: 0.7936415672302246:  58%|█████▊    | 3761/6470 [47:09<40:43,  1.11it/s]
2|3762|Loss: 0.6987768411636353:  58%|█████▊    | 3761/6470 [47:09<40:43,  1.11it/s]
2|3762|Loss: 0.6987768411636353:  58%|█████▊    | 3762/6470 [47:09<33:55,  1.33it/s]
2|3763|Loss: 0.8483564257621765:  58%|█████▊    | 3762/6470 [47:10<33:55,  1.33it/s]
2|3763|Loss: 0.8483564257621765:  58%|█████▊    | 3763/6470 [47:10<34:35,  1.30it/s]
2|3764|Loss: 0.6645329594612122:  58%|█████▊    | 3763/6470 [47:10<34:35,  1.30it/s]
2|3764|Loss: 0.6645329594612122:  58%|█████▊    | 3764/6470 [47:11<34:17,  1.32it/s]
2|3765|Loss: 0.9399407505989075:  58%|█████▊    | 3764/6470 [47:11<34:17,  1.32it/s]
2|3765|Loss: 0.9399407505989075:  58%|█████▊    | 3765/6470 [47:11<31:22,  1.44it/s]
2|3766|Loss: 0.8513698577880859:  58%|█████▊    | 3765/6470 [47:12<31:22,  1.44it/s]
2|3766|Loss: 0.8513698577880859:  58%|█████▊    | 3766/6470 [47:12<31:21,  1.44it/s]
2|3767|Loss: 0.9259916543960571:  58%|█████▊    | 3766/6470 [47:12<31:21,  1.44it/s]
2|3767|Loss: 0.9259916543960571:  58%|█████▊    | 3767/6470 [47:13<31:16,  1.44it/s]
2|3768|Loss: 0.9651109576225281:  58%|█████▊    | 3767/6470 [47:13<31:16,  1.44it/s]
2|3768|Loss: 0.9651109576225281:  58%|█████▊    | 3768/6470 [47:13<30:19,  1.48it/s]
2|3769|Loss: 0.8734782338142395:  58%|█████▊    | 3768/6470 [47:14<30:19,  1.48it/s]
2|3769|Loss: 0.8734782338142395:  58%|█████▊    | 3769/6470 [47:14<31:25,  1.43it/s]
2|3770|Loss: 0.8645226359367371:  58%|█████▊    | 3769/6470 [47:14<31:25,  1.43it/s]
2|3770|Loss: 0.8645226359367371:  58%|█████▊    | 3770/6470 [47:15<32:18,  1.39it/s]
2|3771|Loss: 0.9421144723892212:  58%|█████▊    | 3770/6470 [47:15<32:18,  1.39it/s]
2|3771|Loss: 0.9421144723892212:  58%|█████▊    | 3771/6470 [47:16<34:48,  1.29it/s]
2|3772|Loss: 0.9185181260108948:  58%|█████▊    | 3771/6470 [47:16<34:48,  1.29it/s]
2|3772|Loss: 0.9185181260108948:  58%|█████▊    | 3772/6470 [47:17<37:00,  1.22it/s]
2|3773|Loss: 0.75982266664505:  58%|█████▊    | 3772/6470 [47:17<37:00,  1.22it/s]  
2|3773|Loss: 0.75982266664505:  58%|█████▊    | 3773/6470 [47:17<32:48,  1.37it/s]
2|3774|Loss: 0.8088667392730713:  58%|█████▊    | 3773/6470 [47:17<32:48,  1.37it/s]
2|3774|Loss: 0.8088667392730713:  58%|█████▊    | 3774/6470 [47:18<33:33,  1.34it/s]
2|3775|Loss: 0.8008125424385071:  58%|█████▊    | 3774/6470 [47:18<33:33,  1.34it/s]
2|3775|Loss: 0.8008125424385071:  58%|█████▊    | 3775/6470 [47:19<32:34,  1.38it/s]
2|3776|Loss: 0.9232994914054871:  58%|█████▊    | 3775/6470 [47:19<32:34,  1.38it/s]
2|3776|Loss: 0.9232994914054871:  58%|█████▊    | 3776/6470 [47:19<33:08,  1.35it/s]
2|3777|Loss: 0.8925899267196655:  58%|█████▊    | 3776/6470 [47:20<33:08,  1.35it/s]
2|3777|Loss: 0.8925899267196655:  58%|█████▊    | 3777/6470 [47:20<36:09,  1.24it/s]
2|3778|Loss: 0.8428134918212891:  58%|█████▊    | 3777/6470 [47:21<36:09,  1.24it/s]
2|3778|Loss: 0.8428134918212891:  58%|█████▊    | 3778/6470 [47:21<35:06,  1.28it/s]
2|3779|Loss: 1.0097429752349854:  58%|█████▊    | 3778/6470 [47:21<35:06,  1.28it/s]
2|3779|Loss: 1.0097429752349854:  58%|█████▊    | 3779/6470 [47:22<35:33,  1.26it/s]
2|3780|Loss: 0.8539695739746094:  58%|█████▊    | 3779/6470 [47:22<35:33,  1.26it/s]
2|3780|Loss: 0.8539695739746094:  58%|█████▊    | 3780/6470 [47:23<35:56,  1.25it/s]
2|3781|Loss: 0.7877591848373413:  58%|█████▊    | 3780/6470 [47:23<35:56,  1.25it/s]
2|3781|Loss: 0.7877591848373413:  58%|█████▊    | 3781/6470 [47:24<38:12,  1.17it/s]
2|3782|Loss: 0.7221484184265137:  58%|█████▊    | 3781/6470 [47:24<38:12,  1.17it/s]
2|3782|Loss: 0.7221484184265137:  58%|█████▊    | 3782/6470 [47:25<39:49,  1.13it/s]
2|3783|Loss: 0.9604153633117676:  58%|█████▊    | 3782/6470 [47:25<39:49,  1.13it/s]
2|3783|Loss: 0.9604153633117676:  58%|█████▊    | 3783/6470 [47:25<36:07,  1.24it/s]
2|3784|Loss: 0.7957978844642639:  58%|█████▊    | 3783/6470 [47:26<36:07,  1.24it/s]
2|3784|Loss: 0.7957978844642639:  58%|█████▊    | 3784/6470 [47:26<34:01,  1.32it/s]
2|3785|Loss: 0.7753966450691223:  58%|█████▊    | 3784/6470 [47:26<34:01,  1.32it/s]
2|3785|Loss: 0.7753966450691223:  59%|█████▊    | 3785/6470 [47:27<33:42,  1.33it/s]
2|3786|Loss: 0.8178543448448181:  59%|█████▊    | 3785/6470 [47:27<33:42,  1.33it/s]
2|3786|Loss: 0.8178543448448181:  59%|█████▊    | 3786/6470 [47:27<33:07,  1.35it/s]
2|3787|Loss: 0.8271610736846924:  59%|█████▊    | 3786/6470 [47:28<33:07,  1.35it/s]
2|3787|Loss: 0.8271610736846924:  59%|█████▊    | 3787/6470 [47:28<32:10,  1.39it/s]
2|3788|Loss: 0.9579148888587952:  59%|█████▊    | 3787/6470 [47:28<32:10,  1.39it/s]
2|3788|Loss: 0.9579148888587952:  59%|█████▊    | 3788/6470 [47:29<31:40,  1.41it/s]
2|3789|Loss: 0.7974367141723633:  59%|█████▊    | 3788/6470 [47:29<31:40,  1.41it/s]
2|3789|Loss: 0.7974367141723633:  59%|█████▊    | 3789/6470 [47:29<29:09,  1.53it/s]
2|3790|Loss: 0.7789932489395142:  59%|█████▊    | 3789/6470 [47:29<29:09,  1.53it/s]
2|3790|Loss: 0.7789932489395142:  59%|█████▊    | 3790/6470 [47:30<27:06,  1.65it/s]
2|3791|Loss: 0.8809418678283691:  59%|█████▊    | 3790/6470 [47:30<27:06,  1.65it/s]
2|3791|Loss: 0.8809418678283691:  59%|█████▊    | 3791/6470 [47:30<28:51,  1.55it/s]
2|3792|Loss: 0.8747692704200745:  59%|█████▊    | 3791/6470 [47:31<28:51,  1.55it/s]
2|3792|Loss: 0.8747692704200745:  59%|█████▊    | 3792/6470 [47:31<31:59,  1.39it/s]
2|3793|Loss: 0.8683846592903137:  59%|█████▊    | 3792/6470 [47:32<31:59,  1.39it/s]
2|3793|Loss: 0.8683846592903137:  59%|█████▊    | 3793/6470 [47:32<31:29,  1.42it/s]
2|3794|Loss: 0.7097216248512268:  59%|█████▊    | 3793/6470 [47:32<31:29,  1.42it/s]
2|3794|Loss: 0.7097216248512268:  59%|█████▊    | 3794/6470 [47:33<32:04,  1.39it/s]
2|3795|Loss: 0.8734778165817261:  59%|█████▊    | 3794/6470 [47:33<32:04,  1.39it/s]
2|3795|Loss: 0.8734778165817261:  59%|█████▊    | 3795/6470 [47:34<32:37,  1.37it/s]
2|3796|Loss: 0.9858695268630981:  59%|█████▊    | 3795/6470 [47:34<32:37,  1.37it/s]
2|3796|Loss: 0.9858695268630981:  59%|█████▊    | 3796/6470 [47:34<30:43,  1.45it/s]
2|3797|Loss: 0.9041537046432495:  59%|█████▊    | 3796/6470 [47:34<30:43,  1.45it/s]
2|3797|Loss: 0.9041537046432495:  59%|█████▊    | 3797/6470 [47:35<31:16,  1.42it/s]
2|3798|Loss: 1.1839988231658936:  59%|█████▊    | 3797/6470 [47:35<31:16,  1.42it/s]
2|3798|Loss: 1.1839988231658936:  59%|█████▊    | 3798/6470 [47:36<33:48,  1.32it/s]
2|3799|Loss: 1.0087330341339111:  59%|█████▊    | 3798/6470 [47:36<33:48,  1.32it/s]
2|3799|Loss: 1.0087330341339111:  59%|█████▊    | 3799/6470 [47:37<36:14,  1.23it/s]
2|3800|Loss: 0.8761730790138245:  59%|█████▊    | 3799/6470 [47:37<36:14,  1.23it/s]
2|3800|Loss: 0.8761730790138245:  59%|█████▊    | 3800/6470 [47:37<36:01,  1.24it/s]
2|3801|Loss: 0.844152569770813:  59%|█████▊    | 3800/6470 [47:38<36:01,  1.24it/s] 
2|3801|Loss: 0.844152569770813:  59%|█████▊    | 3801/6470 [47:38<34:19,  1.30it/s]
2|3802|Loss: 0.5392163395881653:  59%|█████▊    | 3801/6470 [47:39<34:19,  1.30it/s]
2|3802|Loss: 0.5392163395881653:  59%|█████▉    | 3802/6470 [47:39<33:09,  1.34it/s]
2|3803|Loss: 0.9064278602600098:  59%|█████▉    | 3802/6470 [47:39<33:09,  1.34it/s]
2|3803|Loss: 0.9064278602600098:  59%|█████▉    | 3803/6470 [47:40<34:57,  1.27it/s]
2|3804|Loss: 0.6616088151931763:  59%|█████▉    | 3803/6470 [47:40<34:57,  1.27it/s]
2|3804|Loss: 0.6616088151931763:  59%|█████▉    | 3804/6470 [47:41<35:25,  1.25it/s]
2|3805|Loss: 0.8454610705375671:  59%|█████▉    | 3804/6470 [47:41<35:25,  1.25it/s]
2|3805|Loss: 0.8454610705375671:  59%|█████▉    | 3805/6470 [47:41<36:35,  1.21it/s]
2|3806|Loss: 0.8031389117240906:  59%|█████▉    | 3805/6470 [47:42<36:35,  1.21it/s]
2|3806|Loss: 0.8031389117240906:  59%|█████▉    | 3806/6470 [47:42<35:07,  1.26it/s]
2|3807|Loss: 0.9623349905014038:  59%|█████▉    | 3806/6470 [47:43<35:07,  1.26it/s]
2|3807|Loss: 0.9623349905014038:  59%|█████▉    | 3807/6470 [47:43<33:27,  1.33it/s]
2|3808|Loss: 0.8824886679649353:  59%|█████▉    | 3807/6470 [47:43<33:27,  1.33it/s]
2|3808|Loss: 0.8824886679649353:  59%|█████▉    | 3808/6470 [47:44<33:33,  1.32it/s]
2|3809|Loss: 0.8275049924850464:  59%|█████▉    | 3808/6470 [47:44<33:33,  1.32it/s]
2|3809|Loss: 0.8275049924850464:  59%|█████▉    | 3809/6470 [47:44<30:41,  1.44it/s]
2|3810|Loss: 0.7120407819747925:  59%|█████▉    | 3809/6470 [47:44<30:41,  1.44it/s]
2|3810|Loss: 0.7120407819747925:  59%|█████▉    | 3810/6470 [47:45<29:56,  1.48it/s]
2|3811|Loss: 0.7781476974487305:  59%|█████▉    | 3810/6470 [47:45<29:56,  1.48it/s]
2|3811|Loss: 0.7781476974487305:  59%|█████▉    | 3811/6470 [47:46<31:29,  1.41it/s]
2|3812|Loss: 0.708198070526123:  59%|█████▉    | 3811/6470 [47:46<31:29,  1.41it/s] 
2|3812|Loss: 0.708198070526123:  59%|█████▉    | 3812/6470 [47:46<31:40,  1.40it/s]
2|3813|Loss: 0.6877135634422302:  59%|█████▉    | 3812/6470 [47:47<31:40,  1.40it/s]
2|3813|Loss: 0.6877135634422302:  59%|█████▉    | 3813/6470 [47:47<32:30,  1.36it/s]
2|3814|Loss: 0.7170162200927734:  59%|█████▉    | 3813/6470 [47:47<32:30,  1.36it/s]
2|3814|Loss: 0.7170162200927734:  59%|█████▉    | 3814/6470 [47:48<33:02,  1.34it/s]
2|3815|Loss: 0.8015041947364807:  59%|█████▉    | 3814/6470 [47:48<33:02,  1.34it/s]
2|3815|Loss: 0.8015041947364807:  59%|█████▉    | 3815/6470 [47:48<29:26,  1.50it/s]
2|3816|Loss: 0.8956791758537292:  59%|█████▉    | 3815/6470 [47:49<29:26,  1.50it/s]
2|3816|Loss: 0.8956791758537292:  59%|█████▉    | 3816/6470 [47:49<31:28,  1.41it/s]
2|3817|Loss: 0.8601828217506409:  59%|█████▉    | 3816/6470 [47:50<31:28,  1.41it/s]
2|3817|Loss: 0.8601828217506409:  59%|█████▉    | 3817/6470 [47:50<31:00,  1.43it/s]
2|3818|Loss: 0.8052949905395508:  59%|█████▉    | 3817/6470 [47:50<31:00,  1.43it/s]
2|3818|Loss: 0.8052949905395508:  59%|█████▉    | 3818/6470 [47:51<31:42,  1.39it/s]
2|3819|Loss: 0.930850625038147:  59%|█████▉    | 3818/6470 [47:51<31:42,  1.39it/s] 
2|3819|Loss: 0.930850625038147:  59%|█████▉    | 3819/6470 [47:51<31:40,  1.39it/s]
2|3820|Loss: 0.8006128668785095:  59%|█████▉    | 3819/6470 [47:52<31:40,  1.39it/s]
2|3820|Loss: 0.8006128668785095:  59%|█████▉    | 3820/6470 [47:52<30:51,  1.43it/s]
2|3821|Loss: 0.7671390175819397:  59%|█████▉    | 3820/6470 [47:52<30:51,  1.43it/s]
2|3821|Loss: 0.7671390175819397:  59%|█████▉    | 3821/6470 [47:53<31:38,  1.40it/s]
2|3822|Loss: 0.8825560808181763:  59%|█████▉    | 3821/6470 [47:53<31:38,  1.40it/s]
2|3822|Loss: 0.8825560808181763:  59%|█████▉    | 3822/6470 [47:53<30:11,  1.46it/s]
2|3823|Loss: 0.7029868364334106:  59%|█████▉    | 3822/6470 [47:54<30:11,  1.46it/s]
2|3823|Loss: 0.7029868364334106:  59%|█████▉    | 3823/6470 [47:54<29:09,  1.51it/s]
2|3824|Loss: 0.7091487646102905:  59%|█████▉    | 3823/6470 [47:54<29:09,  1.51it/s]
2|3824|Loss: 0.7091487646102905:  59%|█████▉    | 3824/6470 [47:55<30:06,  1.47it/s]
2|3825|Loss: 0.8259782195091248:  59%|█████▉    | 3824/6470 [47:55<30:06,  1.47it/s]
2|3825|Loss: 0.8259782195091248:  59%|█████▉    | 3825/6470 [47:56<33:06,  1.33it/s]
2|3826|Loss: 0.7105709314346313:  59%|█████▉    | 3825/6470 [47:56<33:06,  1.33it/s]
2|3826|Loss: 0.7105709314346313:  59%|█████▉    | 3826/6470 [47:56<30:51,  1.43it/s]
2|3827|Loss: 0.846520185470581:  59%|█████▉    | 3826/6470 [47:56<30:51,  1.43it/s] 
2|3827|Loss: 0.846520185470581:  59%|█████▉    | 3827/6470 [47:57<29:43,  1.48it/s]
2|3828|Loss: 0.8701900243759155:  59%|█████▉    | 3827/6470 [47:57<29:43,  1.48it/s]
2|3828|Loss: 0.8701900243759155:  59%|█████▉    | 3828/6470 [47:58<31:39,  1.39it/s]
2|3829|Loss: 0.8020238876342773:  59%|█████▉    | 3828/6470 [47:58<31:39,  1.39it/s]
2|3829|Loss: 0.8020238876342773:  59%|█████▉    | 3829/6470 [47:58<31:37,  1.39it/s]
2|3830|Loss: 0.773777425289154:  59%|█████▉    | 3829/6470 [47:59<31:37,  1.39it/s] 
2|3830|Loss: 0.773777425289154:  59%|█████▉    | 3830/6470 [47:59<30:56,  1.42it/s]
2|3831|Loss: 0.8208328485488892:  59%|█████▉    | 3830/6470 [47:59<30:56,  1.42it/s]
2|3831|Loss: 0.8208328485488892:  59%|█████▉    | 3831/6470 [48:00<33:40,  1.31it/s]
2|3832|Loss: 0.9220061302185059:  59%|█████▉    | 3831/6470 [48:00<33:40,  1.31it/s]
2|3832|Loss: 0.9220061302185059:  59%|█████▉    | 3832/6470 [48:01<33:27,  1.31it/s]
2|3833|Loss: 0.8368355631828308:  59%|█████▉    | 3832/6470 [48:01<33:27,  1.31it/s]
2|3833|Loss: 0.8368355631828308:  59%|█████▉    | 3833/6470 [48:01<31:24,  1.40it/s]
2|3834|Loss: 0.8502451181411743:  59%|█████▉    | 3833/6470 [48:02<31:24,  1.40it/s]
2|3834|Loss: 0.8502451181411743:  59%|█████▉    | 3834/6470 [48:02<30:38,  1.43it/s]
2|3835|Loss: 0.861461877822876:  59%|█████▉    | 3834/6470 [48:02<30:38,  1.43it/s] 
2|3835|Loss: 0.861461877822876:  59%|█████▉    | 3835/6470 [48:03<31:49,  1.38it/s]
2|3836|Loss: 0.7075169682502747:  59%|█████▉    | 3835/6470 [48:03<31:49,  1.38it/s]
2|3836|Loss: 0.7075169682502747:  59%|█████▉    | 3836/6470 [48:03<31:08,  1.41it/s]
2|3837|Loss: 0.9618713855743408:  59%|█████▉    | 3836/6470 [48:04<31:08,  1.41it/s]
2|3837|Loss: 0.9618713855743408:  59%|█████▉    | 3837/6470 [48:04<31:39,  1.39it/s]
2|3838|Loss: 1.0202306509017944:  59%|█████▉    | 3837/6470 [48:04<31:39,  1.39it/s]
2|3838|Loss: 1.0202306509017944:  59%|█████▉    | 3838/6470 [48:05<30:38,  1.43it/s]
2|3839|Loss: 0.9280471205711365:  59%|█████▉    | 3838/6470 [48:05<30:38,  1.43it/s]
2|3839|Loss: 0.9280471205711365:  59%|█████▉    | 3839/6470 [48:05<31:04,  1.41it/s]
2|3840|Loss: 0.8409715890884399:  59%|█████▉    | 3839/6470 [48:06<31:04,  1.41it/s]
2|3840|Loss: 0.8409715890884399:  59%|█████▉    | 3840/6470 [48:06<33:13,  1.32it/s]
2|3841|Loss: 0.7935406565666199:  59%|█████▉    | 3840/6470 [48:07<33:13,  1.32it/s]
2|3841|Loss: 0.7935406565666199:  59%|█████▉    | 3841/6470 [48:07<33:52,  1.29it/s]
2|3842|Loss: 0.7807546854019165:  59%|█████▉    | 3841/6470 [48:08<33:52,  1.29it/s]
2|3842|Loss: 0.7807546854019165:  59%|█████▉    | 3842/6470 [48:08<33:09,  1.32it/s]
2|3843|Loss: 0.8582290410995483:  59%|█████▉    | 3842/6470 [48:08<33:09,  1.32it/s]
2|3843|Loss: 0.8582290410995483:  59%|█████▉    | 3843/6470 [48:09<31:55,  1.37it/s]
2|3844|Loss: 0.7480892539024353:  59%|█████▉    | 3843/6470 [48:09<31:55,  1.37it/s]
2|3844|Loss: 0.7480892539024353:  59%|█████▉    | 3844/6470 [48:09<33:27,  1.31it/s]
2|3845|Loss: 0.701658308506012:  59%|█████▉    | 3844/6470 [48:10<33:27,  1.31it/s] 
2|3845|Loss: 0.701658308506012:  59%|█████▉    | 3845/6470 [48:10<33:32,  1.30it/s]
2|3846|Loss: 0.8085786700248718:  59%|█████▉    | 3845/6470 [48:11<33:32,  1.30it/s]
2|3846|Loss: 0.8085786700248718:  59%|█████▉    | 3846/6470 [48:11<32:02,  1.36it/s]
2|3847|Loss: 0.9146032929420471:  59%|█████▉    | 3846/6470 [48:11<32:02,  1.36it/s]
2|3847|Loss: 0.9146032929420471:  59%|█████▉    | 3847/6470 [48:11<29:55,  1.46it/s]
2|3848|Loss: 0.7817984223365784:  59%|█████▉    | 3847/6470 [48:12<29:55,  1.46it/s]
2|3848|Loss: 0.7817984223365784:  59%|█████▉    | 3848/6470 [48:12<30:53,  1.41it/s]
2|3849|Loss: 0.782197892665863:  59%|█████▉    | 3848/6470 [48:13<30:53,  1.41it/s] 
2|3849|Loss: 0.782197892665863:  59%|█████▉    | 3849/6470 [48:13<30:28,  1.43it/s]
2|3850|Loss: 0.8196952939033508:  59%|█████▉    | 3849/6470 [48:13<30:28,  1.43it/s]
2|3850|Loss: 0.8196952939033508:  60%|█████▉    | 3850/6470 [48:14<31:22,  1.39it/s]
2|3851|Loss: 0.7958506345748901:  60%|█████▉    | 3850/6470 [48:14<31:22,  1.39it/s]
2|3851|Loss: 0.7958506345748901:  60%|█████▉    | 3851/6470 [48:14<32:58,  1.32it/s]
2|3852|Loss: 0.7281426191329956:  60%|█████▉    | 3851/6470 [48:15<32:58,  1.32it/s]
2|3852|Loss: 0.7281426191329956:  60%|█████▉    | 3852/6470 [48:15<32:25,  1.35it/s]
2|3853|Loss: 0.8962281942367554:  60%|█████▉    | 3852/6470 [48:16<32:25,  1.35it/s]
2|3853|Loss: 0.8962281942367554:  60%|█████▉    | 3853/6470 [48:16<33:59,  1.28it/s]
2|3854|Loss: 0.7988051772117615:  60%|█████▉    | 3853/6470 [48:16<33:59,  1.28it/s]
2|3854|Loss: 0.7988051772117615:  60%|█████▉    | 3854/6470 [48:17<35:35,  1.23it/s]
2|3855|Loss: 0.8919204473495483:  60%|█████▉    | 3854/6470 [48:17<35:35,  1.23it/s]
2|3855|Loss: 0.8919204473495483:  60%|█████▉    | 3855/6470 [48:17<32:19,  1.35it/s]
2|3856|Loss: 0.86870938539505:  60%|█████▉    | 3855/6470 [48:18<32:19,  1.35it/s]  
2|3856|Loss: 0.86870938539505:  60%|█████▉    | 3856/6470 [48:18<30:48,  1.41it/s]
2|3857|Loss: 0.7238407135009766:  60%|█████▉    | 3856/6470 [48:18<30:48,  1.41it/s]
2|3857|Loss: 0.7238407135009766:  60%|█████▉    | 3857/6470 [48:19<30:45,  1.42it/s]
2|3858|Loss: 0.6507245302200317:  60%|█████▉    | 3857/6470 [48:19<30:45,  1.42it/s]
2|3858|Loss: 0.6507245302200317:  60%|█████▉    | 3858/6470 [48:19<28:16,  1.54it/s]
2|3859|Loss: 0.807035505771637:  60%|█████▉    | 3858/6470 [48:20<28:16,  1.54it/s] 
2|3859|Loss: 0.807035505771637:  60%|█████▉    | 3859/6470 [48:20<26:47,  1.62it/s]
2|3860|Loss: 0.9948870539665222:  60%|█████▉    | 3859/6470 [48:20<26:47,  1.62it/s]
2|3860|Loss: 0.9948870539665222:  60%|█████▉    | 3860/6470 [48:21<27:58,  1.55it/s]
2|3861|Loss: 0.6681120991706848:  60%|█████▉    | 3860/6470 [48:21<27:58,  1.55it/s]
2|3861|Loss: 0.6681120991706848:  60%|█████▉    | 3861/6470 [48:22<32:33,  1.34it/s]
2|3862|Loss: 0.7447299957275391:  60%|█████▉    | 3861/6470 [48:22<32:33,  1.34it/s]
2|3862|Loss: 0.7447299957275391:  60%|█████▉    | 3862/6470 [48:22<31:42,  1.37it/s]
2|3863|Loss: 0.8649652600288391:  60%|█████▉    | 3862/6470 [48:23<31:42,  1.37it/s]
2|3863|Loss: 0.8649652600288391:  60%|█████▉    | 3863/6470 [48:23<30:53,  1.41it/s]
2|3864|Loss: 0.9093489050865173:  60%|█████▉    | 3863/6470 [48:23<30:53,  1.41it/s]
2|3864|Loss: 0.9093489050865173:  60%|█████▉    | 3864/6470 [48:24<30:48,  1.41it/s]
2|3865|Loss: 0.8946883082389832:  60%|█████▉    | 3864/6470 [48:24<30:48,  1.41it/s]
2|3865|Loss: 0.8946883082389832:  60%|█████▉    | 3865/6470 [48:24<30:08,  1.44it/s]
2|3866|Loss: 0.7190533876419067:  60%|█████▉    | 3865/6470 [48:25<30:08,  1.44it/s]
2|3866|Loss: 0.7190533876419067:  60%|█████▉    | 3866/6470 [48:25<30:00,  1.45it/s]
2|3867|Loss: 0.8397347927093506:  60%|█████▉    | 3866/6470 [48:25<30:00,  1.45it/s]
2|3867|Loss: 0.8397347927093506:  60%|█████▉    | 3867/6470 [48:26<30:15,  1.43it/s]
2|3868|Loss: 0.8389564156532288:  60%|█████▉    | 3867/6470 [48:26<30:15,  1.43it/s]
2|3868|Loss: 0.8389564156532288:  60%|█████▉    | 3868/6470 [48:26<29:40,  1.46it/s]
2|3869|Loss: 0.8909887075424194:  60%|█████▉    | 3868/6470 [48:27<29:40,  1.46it/s]
2|3869|Loss: 0.8909887075424194:  60%|█████▉    | 3869/6470 [48:27<32:23,  1.34it/s]
2|3870|Loss: 0.7637144327163696:  60%|█████▉    | 3869/6470 [48:28<32:23,  1.34it/s]
2|3870|Loss: 0.7637144327163696:  60%|█████▉    | 3870/6470 [48:28<33:23,  1.30it/s]
2|3871|Loss: 0.9195530414581299:  60%|█████▉    | 3870/6470 [48:28<33:23,  1.30it/s]
2|3871|Loss: 0.9195530414581299:  60%|█████▉    | 3871/6470 [48:29<32:43,  1.32it/s]
2|3872|Loss: 0.8697084784507751:  60%|█████▉    | 3871/6470 [48:29<32:43,  1.32it/s]
2|3872|Loss: 0.8697084784507751:  60%|█████▉    | 3872/6470 [48:29<31:36,  1.37it/s]
2|3873|Loss: 0.8450968861579895:  60%|█████▉    | 3872/6470 [48:30<31:36,  1.37it/s]
2|3873|Loss: 0.8450968861579895:  60%|█████▉    | 3873/6470 [48:30<31:07,  1.39it/s]
2|3874|Loss: 0.9924965500831604:  60%|█████▉    | 3873/6470 [48:31<31:07,  1.39it/s]
2|3874|Loss: 0.9924965500831604:  60%|█████▉    | 3874/6470 [48:31<33:35,  1.29it/s]
2|3875|Loss: 0.8135203123092651:  60%|█████▉    | 3874/6470 [48:31<33:35,  1.29it/s]
2|3875|Loss: 0.8135203123092651:  60%|█████▉    | 3875/6470 [48:32<32:30,  1.33it/s]
2|3876|Loss: 0.8950003981590271:  60%|█████▉    | 3875/6470 [48:32<32:30,  1.33it/s]
2|3876|Loss: 0.8950003981590271:  60%|█████▉    | 3876/6470 [48:33<34:14,  1.26it/s]
2|3877|Loss: 0.9127358198165894:  60%|█████▉    | 3876/6470 [48:33<34:14,  1.26it/s]
2|3877|Loss: 0.9127358198165894:  60%|█████▉    | 3877/6470 [48:33<34:52,  1.24it/s]
2|3878|Loss: 0.8693547248840332:  60%|█████▉    | 3877/6470 [48:34<34:52,  1.24it/s]
2|3878|Loss: 0.8693547248840332:  60%|█████▉    | 3878/6470 [48:34<35:32,  1.22it/s]
2|3879|Loss: 0.9397954940795898:  60%|█████▉    | 3878/6470 [48:35<35:32,  1.22it/s]
2|3879|Loss: 0.9397954940795898:  60%|█████▉    | 3879/6470 [48:35<35:40,  1.21it/s]
2|3880|Loss: 0.9192584156990051:  60%|█████▉    | 3879/6470 [48:36<35:40,  1.21it/s]
2|3880|Loss: 0.9192584156990051:  60%|█████▉    | 3880/6470 [48:36<36:09,  1.19it/s]
2|3881|Loss: 0.7777788639068604:  60%|█████▉    | 3880/6470 [48:36<36:09,  1.19it/s]
2|3881|Loss: 0.7777788639068604:  60%|█████▉    | 3881/6470 [48:37<32:46,  1.32it/s]
2|3882|Loss: 0.7917205691337585:  60%|█████▉    | 3881/6470 [48:37<32:46,  1.32it/s]
2|3882|Loss: 0.7917205691337585:  60%|██████    | 3882/6470 [48:37<32:12,  1.34it/s]
2|3883|Loss: 0.8353666663169861:  60%|██████    | 3882/6470 [48:38<32:12,  1.34it/s]
2|3883|Loss: 0.8353666663169861:  60%|██████    | 3883/6470 [48:38<34:20,  1.26it/s]
2|3884|Loss: 0.7507159113883972:  60%|██████    | 3883/6470 [48:39<34:20,  1.26it/s]
2|3884|Loss: 0.7507159113883972:  60%|██████    | 3884/6470 [48:39<31:40,  1.36it/s]
2|3885|Loss: 0.7795619964599609:  60%|██████    | 3884/6470 [48:39<31:40,  1.36it/s]
2|3885|Loss: 0.7795619964599609:  60%|██████    | 3885/6470 [48:39<30:10,  1.43it/s]
2|3886|Loss: 0.7589806914329529:  60%|██████    | 3885/6470 [48:40<30:10,  1.43it/s]
2|3886|Loss: 0.7589806914329529:  60%|██████    | 3886/6470 [48:40<29:47,  1.45it/s]
2|3887|Loss: 1.023352026939392:  60%|██████    | 3886/6470 [48:40<29:47,  1.45it/s] 
2|3887|Loss: 1.023352026939392:  60%|██████    | 3887/6470 [48:41<28:52,  1.49it/s]
2|3888|Loss: 0.7823887467384338:  60%|██████    | 3887/6470 [48:41<28:52,  1.49it/s]
2|3888|Loss: 0.7823887467384338:  60%|██████    | 3888/6470 [48:41<30:05,  1.43it/s]
2|3889|Loss: 0.836567759513855:  60%|██████    | 3888/6470 [48:42<30:05,  1.43it/s] 
2|3889|Loss: 0.836567759513855:  60%|██████    | 3889/6470 [48:42<30:27,  1.41it/s]
2|3890|Loss: 0.781380832195282:  60%|██████    | 3889/6470 [48:43<30:27,  1.41it/s]
2|3890|Loss: 0.781380832195282:  60%|██████    | 3890/6470 [48:43<31:03,  1.38it/s]
2|3891|Loss: 0.9112778902053833:  60%|██████    | 3890/6470 [48:43<31:03,  1.38it/s]
2|3891|Loss: 0.9112778902053833:  60%|██████    | 3891/6470 [48:44<30:27,  1.41it/s]
2|3892|Loss: 0.8827440142631531:  60%|██████    | 3891/6470 [48:44<30:27,  1.41it/s]
2|3892|Loss: 0.8827440142631531:  60%|██████    | 3892/6470 [48:44<29:46,  1.44it/s]
2|3893|Loss: 0.8931354880332947:  60%|██████    | 3892/6470 [48:45<29:46,  1.44it/s]
2|3893|Loss: 0.8931354880332947:  60%|██████    | 3893/6470 [48:45<30:44,  1.40it/s]
2|3894|Loss: 0.7504281997680664:  60%|██████    | 3893/6470 [48:45<30:44,  1.40it/s]
2|3894|Loss: 0.7504281997680664:  60%|██████    | 3894/6470 [48:46<30:29,  1.41it/s]
2|3895|Loss: 0.8008518218994141:  60%|██████    | 3894/6470 [48:46<30:29,  1.41it/s]
2|3895|Loss: 0.8008518218994141:  60%|██████    | 3895/6470 [48:47<32:23,  1.32it/s]
2|3896|Loss: 0.7602384090423584:  60%|██████    | 3895/6470 [48:47<32:23,  1.32it/s]
2|3896|Loss: 0.7602384090423584:  60%|██████    | 3896/6470 [48:47<32:43,  1.31it/s]
2|3897|Loss: 0.758514404296875:  60%|██████    | 3896/6470 [48:48<32:43,  1.31it/s] 
2|3897|Loss: 0.758514404296875:  60%|██████    | 3897/6470 [48:48<34:13,  1.25it/s]
2|3898|Loss: 0.9574262499809265:  60%|██████    | 3897/6470 [48:49<34:13,  1.25it/s]
2|3898|Loss: 0.9574262499809265:  60%|██████    | 3898/6470 [48:49<35:26,  1.21it/s]
2|3899|Loss: 0.839470386505127:  60%|██████    | 3898/6470 [48:50<35:26,  1.21it/s] 
2|3899|Loss: 0.839470386505127:  60%|██████    | 3899/6470 [48:50<34:11,  1.25it/s]
2|3900|Loss: 0.7600791454315186:  60%|██████    | 3899/6470 [48:50<34:11,  1.25it/s]
2|3900|Loss: 0.7600791454315186:  60%|██████    | 3900/6470 [48:51<32:04,  1.34it/s]
2|3901|Loss: 0.700946569442749:  60%|██████    | 3900/6470 [48:51<32:04,  1.34it/s] 
2|3901|Loss: 0.700946569442749:  60%|██████    | 3901/6470 [48:52<40:33,  1.06it/s]
2|3902|Loss: 0.8413753509521484:  60%|██████    | 3901/6470 [48:53<40:33,  1.06it/s]
2|3902|Loss: 0.8413753509521484:  60%|██████    | 3902/6470 [48:53<40:49,  1.05it/s]
2|3903|Loss: 0.821794331073761:  60%|██████    | 3902/6470 [48:53<40:49,  1.05it/s] 
2|3903|Loss: 0.821794331073761:  60%|██████    | 3903/6470 [48:54<39:33,  1.08it/s]
2|3904|Loss: 0.7727340459823608:  60%|██████    | 3903/6470 [48:54<39:33,  1.08it/s]
2|3904|Loss: 0.7727340459823608:  60%|██████    | 3904/6470 [48:55<37:58,  1.13it/s]
2|3905|Loss: 0.8654208779335022:  60%|██████    | 3904/6470 [48:55<37:58,  1.13it/s]
2|3905|Loss: 0.8654208779335022:  60%|██████    | 3905/6470 [48:55<35:17,  1.21it/s]
2|3906|Loss: 1.0443930625915527:  60%|██████    | 3905/6470 [48:56<35:17,  1.21it/s]
2|3906|Loss: 1.0443930625915527:  60%|██████    | 3906/6470 [48:56<35:38,  1.20it/s]
2|3907|Loss: 0.8092002272605896:  60%|██████    | 3906/6470 [48:57<35:38,  1.20it/s]
2|3907|Loss: 0.8092002272605896:  60%|██████    | 3907/6470 [48:57<34:56,  1.22it/s]
2|3908|Loss: 0.9033061861991882:  60%|██████    | 3907/6470 [48:57<34:56,  1.22it/s]
2|3908|Loss: 0.9033061861991882:  60%|██████    | 3908/6470 [48:58<34:10,  1.25it/s]
2|3909|Loss: 0.8962917923927307:  60%|██████    | 3908/6470 [48:58<34:10,  1.25it/s]
2|3909|Loss: 0.8962917923927307:  60%|██████    | 3909/6470 [48:58<32:45,  1.30it/s]
2|3910|Loss: 0.8208169341087341:  60%|██████    | 3909/6470 [48:59<32:45,  1.30it/s]
2|3910|Loss: 0.8208169341087341:  60%|██████    | 3910/6470 [48:59<32:22,  1.32it/s]
2|3911|Loss: 0.7789959907531738:  60%|██████    | 3910/6470 [49:00<32:22,  1.32it/s]
2|3911|Loss: 0.7789959907531738:  60%|██████    | 3911/6470 [49:00<32:11,  1.33it/s]
2|3912|Loss: 0.8988609910011292:  60%|██████    | 3911/6470 [49:00<32:11,  1.33it/s]
2|3912|Loss: 0.8988609910011292:  60%|██████    | 3912/6470 [49:01<32:15,  1.32it/s]
2|3913|Loss: 0.8473916053771973:  60%|██████    | 3912/6470 [49:01<32:15,  1.32it/s]
2|3913|Loss: 0.8473916053771973:  60%|██████    | 3913/6470 [49:01<31:39,  1.35it/s]
2|3914|Loss: 0.9519570469856262:  60%|██████    | 3913/6470 [49:02<31:39,  1.35it/s]
2|3914|Loss: 0.9519570469856262:  60%|██████    | 3914/6470 [49:02<29:38,  1.44it/s]
2|3915|Loss: 0.687614917755127:  60%|██████    | 3914/6470 [49:02<29:38,  1.44it/s] 
2|3915|Loss: 0.687614917755127:  61%|██████    | 3915/6470 [49:03<31:37,  1.35it/s]
2|3916|Loss: 0.8801382780075073:  61%|██████    | 3915/6470 [49:03<31:37,  1.35it/s]
2|3916|Loss: 0.8801382780075073:  61%|██████    | 3916/6470 [49:04<33:42,  1.26it/s]
2|3917|Loss: 0.9055697321891785:  61%|██████    | 3916/6470 [49:04<33:42,  1.26it/s]
2|3917|Loss: 0.9055697321891785:  61%|██████    | 3917/6470 [49:04<33:40,  1.26it/s]
2|3918|Loss: 0.8402157425880432:  61%|██████    | 3917/6470 [49:05<33:40,  1.26it/s]
2|3918|Loss: 0.8402157425880432:  61%|██████    | 3918/6470 [49:05<36:18,  1.17it/s]
2|3919|Loss: 0.9153086543083191:  61%|██████    | 3918/6470 [49:06<36:18,  1.17it/s]
2|3919|Loss: 0.9153086543083191:  61%|██████    | 3919/6470 [49:06<37:38,  1.13it/s]
2|3920|Loss: 0.848366379737854:  61%|██████    | 3919/6470 [49:07<37:38,  1.13it/s] 
2|3920|Loss: 0.848366379737854:  61%|██████    | 3920/6470 [49:07<35:45,  1.19it/s]
2|3921|Loss: 0.8976150155067444:  61%|██████    | 3920/6470 [49:08<35:45,  1.19it/s]
2|3921|Loss: 0.8976150155067444:  61%|██████    | 3921/6470 [49:08<33:05,  1.28it/s]
2|3922|Loss: 0.8702710866928101:  61%|██████    | 3921/6470 [49:08<33:05,  1.28it/s]
2|3922|Loss: 0.8702710866928101:  61%|██████    | 3922/6470 [49:09<32:59,  1.29it/s]
2|3923|Loss: 0.8369342684745789:  61%|██████    | 3922/6470 [49:09<32:59,  1.29it/s]
2|3923|Loss: 0.8369342684745789:  61%|██████    | 3923/6470 [49:09<33:11,  1.28it/s]
2|3924|Loss: 0.6818389296531677:  61%|██████    | 3923/6470 [49:10<33:11,  1.28it/s]
2|3924|Loss: 0.6818389296531677:  61%|██████    | 3924/6470 [49:10<32:05,  1.32it/s]
2|3925|Loss: 0.7797223329544067:  61%|██████    | 3924/6470 [49:10<32:05,  1.32it/s]
2|3925|Loss: 0.7797223329544067:  61%|██████    | 3925/6470 [49:11<33:59,  1.25it/s]
2|3926|Loss: 0.8053073883056641:  61%|██████    | 3925/6470 [49:11<33:59,  1.25it/s]
2|3926|Loss: 0.8053073883056641:  61%|██████    | 3926/6470 [49:12<35:43,  1.19it/s]
2|3927|Loss: 0.8647811412811279:  61%|██████    | 3926/6470 [49:12<35:43,  1.19it/s]
2|3927|Loss: 0.8647811412811279:  61%|██████    | 3927/6470 [49:13<35:07,  1.21it/s]
2|3928|Loss: 0.8164393901824951:  61%|██████    | 3927/6470 [49:13<35:07,  1.21it/s]
2|3928|Loss: 0.8164393901824951:  61%|██████    | 3928/6470 [49:13<34:33,  1.23it/s]
2|3929|Loss: 0.7001545429229736:  61%|██████    | 3928/6470 [49:14<34:33,  1.23it/s]
2|3929|Loss: 0.7001545429229736:  61%|██████    | 3929/6470 [49:14<32:29,  1.30it/s]
2|3930|Loss: 0.9154645204544067:  61%|██████    | 3929/6470 [49:14<32:29,  1.30it/s]
2|3930|Loss: 0.9154645204544067:  61%|██████    | 3930/6470 [49:15<30:43,  1.38it/s]
2|3931|Loss: 0.9345241785049438:  61%|██████    | 3930/6470 [49:15<30:43,  1.38it/s]
2|3931|Loss: 0.9345241785049438:  61%|██████    | 3931/6470 [49:15<30:13,  1.40it/s]
2|3932|Loss: 0.8760057091712952:  61%|██████    | 3931/6470 [49:16<30:13,  1.40it/s]
2|3932|Loss: 0.8760057091712952:  61%|██████    | 3932/6470 [49:16<30:55,  1.37it/s]
2|3933|Loss: 0.88875812292099:  61%|██████    | 3932/6470 [49:17<30:55,  1.37it/s]  
2|3933|Loss: 0.88875812292099:  61%|██████    | 3933/6470 [49:17<30:23,  1.39it/s]
2|3934|Loss: 0.8064890503883362:  61%|██████    | 3933/6470 [49:17<30:23,  1.39it/s]
2|3934|Loss: 0.8064890503883362:  61%|██████    | 3934/6470 [49:18<31:41,  1.33it/s]
2|3935|Loss: 0.7968352437019348:  61%|██████    | 3934/6470 [49:18<31:41,  1.33it/s]
2|3935|Loss: 0.7968352437019348:  61%|██████    | 3935/6470 [49:19<33:55,  1.25it/s]
2|3936|Loss: 0.8018677234649658:  61%|██████    | 3935/6470 [49:19<33:55,  1.25it/s]
2|3936|Loss: 0.8018677234649658:  61%|██████    | 3936/6470 [49:19<33:02,  1.28it/s]
2|3937|Loss: 0.8946613073348999:  61%|██████    | 3936/6470 [49:20<33:02,  1.28it/s]
2|3937|Loss: 0.8946613073348999:  61%|██████    | 3937/6470 [49:20<30:44,  1.37it/s]
2|3938|Loss: 0.875391960144043:  61%|██████    | 3937/6470 [49:20<30:44,  1.37it/s] 
2|3938|Loss: 0.875391960144043:  61%|██████    | 3938/6470 [49:21<31:55,  1.32it/s]
2|3939|Loss: 0.8852782249450684:  61%|██████    | 3938/6470 [49:21<31:55,  1.32it/s]
2|3939|Loss: 0.8852782249450684:  61%|██████    | 3939/6470 [49:22<32:06,  1.31it/s]
2|3940|Loss: 0.9846776723861694:  61%|██████    | 3939/6470 [49:22<32:06,  1.31it/s]
2|3940|Loss: 0.9846776723861694:  61%|██████    | 3940/6470 [49:22<33:02,  1.28it/s]
2|3941|Loss: 0.594590961933136:  61%|██████    | 3940/6470 [49:23<33:02,  1.28it/s] 
2|3941|Loss: 0.594590961933136:  61%|██████    | 3941/6470 [49:23<34:03,  1.24it/s]
2|3942|Loss: 0.8278393745422363:  61%|██████    | 3941/6470 [49:24<34:03,  1.24it/s]
2|3942|Loss: 0.8278393745422363:  61%|██████    | 3942/6470 [49:24<35:42,  1.18it/s]
2|3943|Loss: 0.8678712844848633:  61%|██████    | 3942/6470 [49:25<35:42,  1.18it/s]
2|3943|Loss: 0.8678712844848633:  61%|██████    | 3943/6470 [49:25<35:05,  1.20it/s]
2|3944|Loss: 0.7431005835533142:  61%|██████    | 3943/6470 [49:25<35:05,  1.20it/s]
2|3944|Loss: 0.7431005835533142:  61%|██████    | 3944/6470 [49:26<34:05,  1.23it/s]
2|3945|Loss: 1.0328999757766724:  61%|██████    | 3944/6470 [49:26<34:05,  1.23it/s]
2|3945|Loss: 1.0328999757766724:  61%|██████    | 3945/6470 [49:26<32:47,  1.28it/s]
2|3946|Loss: 0.6541571617126465:  61%|██████    | 3945/6470 [49:27<32:47,  1.28it/s]
2|3946|Loss: 0.6541571617126465:  61%|██████    | 3946/6470 [49:27<33:56,  1.24it/s]
2|3947|Loss: 0.7032803297042847:  61%|██████    | 3946/6470 [49:28<33:56,  1.24it/s]
2|3947|Loss: 0.7032803297042847:  61%|██████    | 3947/6470 [49:28<31:32,  1.33it/s]
2|3948|Loss: 0.9898614287376404:  61%|██████    | 3947/6470 [49:28<31:32,  1.33it/s]
2|3948|Loss: 0.9898614287376404:  61%|██████    | 3948/6470 [49:29<30:18,  1.39it/s]
2|3949|Loss: 0.817509651184082:  61%|██████    | 3948/6470 [49:29<30:18,  1.39it/s] 
2|3949|Loss: 0.817509651184082:  61%|██████    | 3949/6470 [49:29<29:02,  1.45it/s]
2|3950|Loss: 0.8702799081802368:  61%|██████    | 3949/6470 [49:30<29:02,  1.45it/s]
2|3950|Loss: 0.8702799081802368:  61%|██████    | 3950/6470 [49:30<26:38,  1.58it/s]
2|3951|Loss: 0.7495037913322449:  61%|██████    | 3950/6470 [49:30<26:38,  1.58it/s]
2|3951|Loss: 0.7495037913322449:  61%|██████    | 3951/6470 [49:30<27:45,  1.51it/s]
2|3952|Loss: 0.7224746346473694:  61%|██████    | 3951/6470 [49:31<27:45,  1.51it/s]
2|3952|Loss: 0.7224746346473694:  61%|██████    | 3952/6470 [49:31<29:22,  1.43it/s]
2|3953|Loss: 0.884175717830658:  61%|██████    | 3952/6470 [49:32<29:22,  1.43it/s] 
2|3953|Loss: 0.884175717830658:  61%|██████    | 3953/6470 [49:32<27:58,  1.50it/s]
2|3954|Loss: 0.8957087397575378:  61%|██████    | 3953/6470 [49:32<27:58,  1.50it/s]
2|3954|Loss: 0.8957087397575378:  61%|██████    | 3954/6470 [49:33<27:57,  1.50it/s]
2|3955|Loss: 0.7703356742858887:  61%|██████    | 3954/6470 [49:33<27:57,  1.50it/s]
2|3955|Loss: 0.7703356742858887:  61%|██████    | 3955/6470 [49:33<27:28,  1.53it/s]
2|3956|Loss: 0.9948962926864624:  61%|██████    | 3955/6470 [49:33<27:28,  1.53it/s]
2|3956|Loss: 0.9948962926864624:  61%|██████    | 3956/6470 [49:34<26:08,  1.60it/s]
2|3957|Loss: 1.0149704217910767:  61%|██████    | 3956/6470 [49:34<26:08,  1.60it/s]
2|3957|Loss: 1.0149704217910767:  61%|██████    | 3957/6470 [49:34<25:36,  1.64it/s]
2|3958|Loss: 0.8412171006202698:  61%|██████    | 3957/6470 [49:35<25:36,  1.64it/s]
2|3958|Loss: 0.8412171006202698:  61%|██████    | 3958/6470 [49:35<26:28,  1.58it/s]
2|3959|Loss: 0.9106565713882446:  61%|██████    | 3958/6470 [49:35<26:28,  1.58it/s]
2|3959|Loss: 0.9106565713882446:  61%|██████    | 3959/6470 [49:36<26:56,  1.55it/s]
2|3960|Loss: 0.8277746438980103:  61%|██████    | 3959/6470 [49:36<26:56,  1.55it/s]
2|3960|Loss: 0.8277746438980103:  61%|██████    | 3960/6470 [49:36<28:21,  1.48it/s]
2|3961|Loss: 0.9500391483306885:  61%|██████    | 3960/6470 [49:37<28:21,  1.48it/s]
2|3961|Loss: 0.9500391483306885:  61%|██████    | 3961/6470 [49:37<27:12,  1.54it/s]
2|3962|Loss: 0.9523662328720093:  61%|██████    | 3961/6470 [49:37<27:12,  1.54it/s]
2|3962|Loss: 0.9523662328720093:  61%|██████    | 3962/6470 [49:38<28:20,  1.47it/s]
2|3963|Loss: 0.7359368205070496:  61%|██████    | 3962/6470 [49:38<28:20,  1.47it/s]
2|3963|Loss: 0.7359368205070496:  61%|██████▏   | 3963/6470 [49:39<31:46,  1.31it/s]
2|3964|Loss: 0.714903712272644:  61%|██████▏   | 3963/6470 [49:39<31:46,  1.31it/s] 
2|3964|Loss: 0.714903712272644:  61%|██████▏   | 3964/6470 [49:40<33:27,  1.25it/s]
2|3965|Loss: 0.9572544097900391:  61%|██████▏   | 3964/6470 [49:40<33:27,  1.25it/s]
2|3965|Loss: 0.9572544097900391:  61%|██████▏   | 3965/6470 [49:40<33:10,  1.26it/s]
2|3966|Loss: 0.8567890524864197:  61%|██████▏   | 3965/6470 [49:41<33:10,  1.26it/s]
2|3966|Loss: 0.8567890524864197:  61%|██████▏   | 3966/6470 [49:41<31:18,  1.33it/s]
2|3967|Loss: 0.8381152153015137:  61%|██████▏   | 3966/6470 [49:41<31:18,  1.33it/s]
2|3967|Loss: 0.8381152153015137:  61%|██████▏   | 3967/6470 [49:42<32:28,  1.28it/s]
2|3968|Loss: 0.8382681012153625:  61%|██████▏   | 3967/6470 [49:42<32:28,  1.28it/s]
2|3968|Loss: 0.8382681012153625:  61%|██████▏   | 3968/6470 [49:43<31:52,  1.31it/s]
2|3969|Loss: 0.803509533405304:  61%|██████▏   | 3968/6470 [49:43<31:52,  1.31it/s] 
2|3969|Loss: 0.803509533405304:  61%|██████▏   | 3969/6470 [49:43<30:15,  1.38it/s]
2|3970|Loss: 0.8875577449798584:  61%|██████▏   | 3969/6470 [49:44<30:15,  1.38it/s]
2|3970|Loss: 0.8875577449798584:  61%|██████▏   | 3970/6470 [49:44<29:37,  1.41it/s]
2|3971|Loss: 0.9456299543380737:  61%|██████▏   | 3970/6470 [49:44<29:37,  1.41it/s]
2|3971|Loss: 0.9456299543380737:  61%|██████▏   | 3971/6470 [49:44<28:27,  1.46it/s]
2|3972|Loss: 0.7716270685195923:  61%|██████▏   | 3971/6470 [49:45<28:27,  1.46it/s]
2|3972|Loss: 0.7716270685195923:  61%|██████▏   | 3972/6470 [49:45<28:58,  1.44it/s]
2|3973|Loss: 0.7603644728660583:  61%|██████▏   | 3972/6470 [49:46<28:58,  1.44it/s]
2|3973|Loss: 0.7603644728660583:  61%|██████▏   | 3973/6470 [49:46<27:33,  1.51it/s]
2|3974|Loss: 0.896360456943512:  61%|██████▏   | 3973/6470 [49:46<27:33,  1.51it/s] 
2|3974|Loss: 0.896360456943512:  61%|██████▏   | 3974/6470 [49:47<29:20,  1.42it/s]
2|3975|Loss: 0.9743649363517761:  61%|██████▏   | 3974/6470 [49:47<29:20,  1.42it/s]
2|3975|Loss: 0.9743649363517761:  61%|██████▏   | 3975/6470 [49:47<30:33,  1.36it/s]
2|3976|Loss: 0.9353679418563843:  61%|██████▏   | 3975/6470 [49:48<30:33,  1.36it/s]
2|3976|Loss: 0.9353679418563843:  61%|██████▏   | 3976/6470 [49:48<30:08,  1.38it/s]
2|3977|Loss: 0.6520177721977234:  61%|██████▏   | 3976/6470 [49:48<30:08,  1.38it/s]
2|3977|Loss: 0.6520177721977234:  61%|██████▏   | 3977/6470 [49:49<29:24,  1.41it/s]
2|3978|Loss: 0.7779774069786072:  61%|██████▏   | 3977/6470 [49:49<29:24,  1.41it/s]
2|3978|Loss: 0.7779774069786072:  61%|██████▏   | 3978/6470 [49:50<30:08,  1.38it/s]
2|3979|Loss: 0.7448307275772095:  61%|██████▏   | 3978/6470 [49:50<30:08,  1.38it/s]
2|3979|Loss: 0.7448307275772095:  61%|██████▏   | 3979/6470 [49:50<31:15,  1.33it/s]
2|3980|Loss: 0.8470396399497986:  61%|██████▏   | 3979/6470 [49:51<31:15,  1.33it/s]
2|3980|Loss: 0.8470396399497986:  62%|██████▏   | 3980/6470 [49:51<29:14,  1.42it/s]
2|3981|Loss: 0.8020987510681152:  62%|██████▏   | 3980/6470 [49:51<29:14,  1.42it/s]
2|3981|Loss: 0.8020987510681152:  62%|██████▏   | 3981/6470 [49:52<28:14,  1.47it/s]
2|3982|Loss: 0.8647423386573792:  62%|██████▏   | 3981/6470 [49:52<28:14,  1.47it/s]
2|3982|Loss: 0.8647423386573792:  62%|██████▏   | 3982/6470 [49:52<28:58,  1.43it/s]
2|3983|Loss: 0.9359543919563293:  62%|██████▏   | 3982/6470 [49:53<28:58,  1.43it/s]
2|3983|Loss: 0.9359543919563293:  62%|██████▏   | 3983/6470 [49:53<28:45,  1.44it/s]
2|3984|Loss: 0.7949327230453491:  62%|██████▏   | 3983/6470 [49:53<28:45,  1.44it/s]
2|3984|Loss: 0.7949327230453491:  62%|██████▏   | 3984/6470 [49:54<30:59,  1.34it/s]
2|3985|Loss: 0.9573705196380615:  62%|██████▏   | 3984/6470 [49:54<30:59,  1.34it/s]
2|3985|Loss: 0.9573705196380615:  62%|██████▏   | 3985/6470 [49:55<33:30,  1.24it/s]
2|3986|Loss: 1.2033604383468628:  62%|██████▏   | 3985/6470 [49:55<33:30,  1.24it/s]
2|3986|Loss: 1.2033604383468628:  62%|██████▏   | 3986/6470 [49:56<38:03,  1.09it/s]
2|3987|Loss: 0.9007266163825989:  62%|██████▏   | 3986/6470 [49:57<38:03,  1.09it/s]
2|3987|Loss: 0.9007266163825989:  62%|██████▏   | 3987/6470 [49:57<38:14,  1.08it/s]
2|3988|Loss: 0.9841848015785217:  62%|██████▏   | 3987/6470 [49:57<38:14,  1.08it/s]
2|3988|Loss: 0.9841848015785217:  62%|██████▏   | 3988/6470 [49:58<35:51,  1.15it/s]
2|3989|Loss: 1.03385329246521:  62%|██████▏   | 3988/6470 [49:58<35:51,  1.15it/s]  
2|3989|Loss: 1.03385329246521:  62%|██████▏   | 3989/6470 [49:58<34:41,  1.19it/s]
2|3990|Loss: 0.8288421630859375:  62%|██████▏   | 3989/6470 [49:59<34:41,  1.19it/s]
2|3990|Loss: 0.8288421630859375:  62%|██████▏   | 3990/6470 [49:59<32:35,  1.27it/s]
2|3991|Loss: 1.0129780769348145:  62%|██████▏   | 3990/6470 [50:00<32:35,  1.27it/s]
2|3991|Loss: 1.0129780769348145:  62%|██████▏   | 3991/6470 [50:00<33:39,  1.23it/s]
2|3992|Loss: 0.8181736469268799:  62%|██████▏   | 3991/6470 [50:00<33:39,  1.23it/s]
2|3992|Loss: 0.8181736469268799:  62%|██████▏   | 3992/6470 [50:01<32:26,  1.27it/s]
2|3993|Loss: 0.9159189462661743:  62%|██████▏   | 3992/6470 [50:01<32:26,  1.27it/s]
2|3993|Loss: 0.9159189462661743:  62%|██████▏   | 3993/6470 [50:02<32:50,  1.26it/s]
2|3994|Loss: 0.879944384098053:  62%|██████▏   | 3993/6470 [50:02<32:50,  1.26it/s] 
2|3994|Loss: 0.879944384098053:  62%|██████▏   | 3994/6470 [50:02<29:08,  1.42it/s]
2|3995|Loss: 0.7351816892623901:  62%|██████▏   | 3994/6470 [50:02<29:08,  1.42it/s]
2|3995|Loss: 0.7351816892623901:  62%|██████▏   | 3995/6470 [50:03<28:05,  1.47it/s]
2|3996|Loss: 0.91193026304245:  62%|██████▏   | 3995/6470 [50:03<28:05,  1.47it/s]  
2|3996|Loss: 0.91193026304245:  62%|██████▏   | 3996/6470 [50:03<29:46,  1.38it/s]
2|3997|Loss: 0.7971551418304443:  62%|██████▏   | 3996/6470 [50:04<29:46,  1.38it/s]
2|3997|Loss: 0.7971551418304443:  62%|██████▏   | 3997/6470 [50:04<29:33,  1.39it/s]
2|3998|Loss: 0.8650624752044678:  62%|██████▏   | 3997/6470 [50:05<29:33,  1.39it/s]
2|3998|Loss: 0.8650624752044678:  62%|██████▏   | 3998/6470 [50:05<28:07,  1.46it/s]
2|3999|Loss: 0.9458332061767578:  62%|██████▏   | 3998/6470 [50:05<28:07,  1.46it/s]
2|3999|Loss: 0.9458332061767578:  62%|██████▏   | 3999/6470 [50:05<27:15,  1.51it/s]
2|4000|Loss: 0.7021052241325378:  62%|██████▏   | 3999/6470 [50:06<27:15,  1.51it/s]
2|4000|Loss: 0.7021052241325378:  62%|██████▏   | 4000/6470 [50:06<27:24,  1.50it/s]
2|4001|Loss: 0.9026349782943726:  62%|██████▏   | 4000/6470 [50:07<27:24,  1.50it/s]
2|4001|Loss: 0.9026349782943726:  62%|██████▏   | 4001/6470 [50:07<29:33,  1.39it/s]
2|4002|Loss: 0.8432061076164246:  62%|██████▏   | 4001/6470 [50:07<29:33,  1.39it/s]
2|4002|Loss: 0.8432061076164246:  62%|██████▏   | 4002/6470 [50:08<28:15,  1.46it/s]
2|4003|Loss: 0.7718801498413086:  62%|██████▏   | 4002/6470 [50:08<28:15,  1.46it/s]
2|4003|Loss: 0.7718801498413086:  62%|██████▏   | 4003/6470 [50:08<28:51,  1.42it/s]
2|4004|Loss: 0.8602645993232727:  62%|██████▏   | 4003/6470 [50:09<28:51,  1.42it/s]
2|4004|Loss: 0.8602645993232727:  62%|██████▏   | 4004/6470 [50:09<28:45,  1.43it/s]
2|4005|Loss: 0.9116014242172241:  62%|██████▏   | 4004/6470 [50:09<28:45,  1.43it/s]
2|4005|Loss: 0.9116014242172241:  62%|██████▏   | 4005/6470 [50:10<29:44,  1.38it/s]
2|4006|Loss: 0.7847355008125305:  62%|██████▏   | 4005/6470 [50:10<29:44,  1.38it/s]
2|4006|Loss: 0.7847355008125305:  62%|██████▏   | 4006/6470 [50:11<31:32,  1.30it/s]
2|4007|Loss: 0.7491387724876404:  62%|██████▏   | 4006/6470 [50:11<31:32,  1.30it/s]
2|4007|Loss: 0.7491387724876404:  62%|██████▏   | 4007/6470 [50:11<30:46,  1.33it/s]
2|4008|Loss: 0.8389569520950317:  62%|██████▏   | 4007/6470 [50:12<30:46,  1.33it/s]
2|4008|Loss: 0.8389569520950317:  62%|██████▏   | 4008/6470 [50:12<31:09,  1.32it/s]
2|4009|Loss: 0.9375045895576477:  62%|██████▏   | 4008/6470 [50:13<31:09,  1.32it/s]
2|4009|Loss: 0.9375045895576477:  62%|██████▏   | 4009/6470 [50:13<33:20,  1.23it/s]
2|4010|Loss: 0.8316372632980347:  62%|██████▏   | 4009/6470 [50:14<33:20,  1.23it/s]
2|4010|Loss: 0.8316372632980347:  62%|██████▏   | 4010/6470 [50:14<34:27,  1.19it/s]
2|4011|Loss: 1.0802524089813232:  62%|██████▏   | 4010/6470 [50:14<34:27,  1.19it/s]
2|4011|Loss: 1.0802524089813232:  62%|██████▏   | 4011/6470 [50:15<34:11,  1.20it/s]
2|4012|Loss: 0.8102529048919678:  62%|██████▏   | 4011/6470 [50:15<34:11,  1.20it/s]
2|4012|Loss: 0.8102529048919678:  62%|██████▏   | 4012/6470 [50:16<34:46,  1.18it/s]
2|4013|Loss: 0.825361430644989:  62%|██████▏   | 4012/6470 [50:16<34:46,  1.18it/s] 
2|4013|Loss: 0.825361430644989:  62%|██████▏   | 4013/6470 [50:16<33:18,  1.23it/s]
2|4014|Loss: 0.9314870834350586:  62%|██████▏   | 4013/6470 [50:17<33:18,  1.23it/s]
2|4014|Loss: 0.9314870834350586:  62%|██████▏   | 4014/6470 [50:17<34:27,  1.19it/s]
2|4015|Loss: 0.6441112160682678:  62%|██████▏   | 4014/6470 [50:18<34:27,  1.19it/s]
2|4015|Loss: 0.6441112160682678:  62%|██████▏   | 4015/6470 [50:18<33:07,  1.24it/s]
2|4016|Loss: 0.818587601184845:  62%|██████▏   | 4015/6470 [50:18<33:07,  1.24it/s] 
2|4016|Loss: 0.818587601184845:  62%|██████▏   | 4016/6470 [50:19<32:15,  1.27it/s]
2|4017|Loss: 0.9386602640151978:  62%|██████▏   | 4016/6470 [50:19<32:15,  1.27it/s]
2|4017|Loss: 0.9386602640151978:  62%|██████▏   | 4017/6470 [50:19<30:56,  1.32it/s]
2|4018|Loss: 0.8179388642311096:  62%|██████▏   | 4017/6470 [50:20<30:56,  1.32it/s]
2|4018|Loss: 0.8179388642311096:  62%|██████▏   | 4018/6470 [50:20<32:36,  1.25it/s]
2|4019|Loss: 0.9853420853614807:  62%|██████▏   | 4018/6470 [50:21<32:36,  1.25it/s]
2|4019|Loss: 0.9853420853614807:  62%|██████▏   | 4019/6470 [50:21<34:16,  1.19it/s]
2|4020|Loss: 0.9705804586410522:  62%|██████▏   | 4019/6470 [50:22<34:16,  1.19it/s]
2|4020|Loss: 0.9705804586410522:  62%|██████▏   | 4020/6470 [50:22<34:18,  1.19it/s]
2|4021|Loss: 0.8967581987380981:  62%|██████▏   | 4020/6470 [50:22<34:18,  1.19it/s]
2|4021|Loss: 0.8967581987380981:  62%|██████▏   | 4021/6470 [50:23<31:42,  1.29it/s]
2|4022|Loss: 0.8274895548820496:  62%|██████▏   | 4021/6470 [50:23<31:42,  1.29it/s]
2|4022|Loss: 0.8274895548820496:  62%|██████▏   | 4022/6470 [50:23<30:29,  1.34it/s]
2|4023|Loss: 0.8291876316070557:  62%|██████▏   | 4022/6470 [50:24<30:29,  1.34it/s]
2|4023|Loss: 0.8291876316070557:  62%|██████▏   | 4023/6470 [50:24<29:39,  1.38it/s]
2|4024|Loss: 0.6064804792404175:  62%|██████▏   | 4023/6470 [50:25<29:39,  1.38it/s]
2|4024|Loss: 0.6064804792404175:  62%|██████▏   | 4024/6470 [50:25<33:04,  1.23it/s]
2|4025|Loss: 0.9221457839012146:  62%|██████▏   | 4024/6470 [50:26<33:04,  1.23it/s]
2|4025|Loss: 0.9221457839012146:  62%|██████▏   | 4025/6470 [50:26<32:42,  1.25it/s]
2|4026|Loss: 0.8519745469093323:  62%|██████▏   | 4025/6470 [50:26<32:42,  1.25it/s]
2|4026|Loss: 0.8519745469093323:  62%|██████▏   | 4026/6470 [50:27<32:25,  1.26it/s]
2|4027|Loss: 0.7887402176856995:  62%|██████▏   | 4026/6470 [50:27<32:25,  1.26it/s]
2|4027|Loss: 0.7887402176856995:  62%|██████▏   | 4027/6470 [50:27<32:19,  1.26it/s]
2|4028|Loss: 0.8139693140983582:  62%|██████▏   | 4027/6470 [50:28<32:19,  1.26it/s]
2|4028|Loss: 0.8139693140983582:  62%|██████▏   | 4028/6470 [50:28<30:49,  1.32it/s]
2|4029|Loss: 0.7979098558425903:  62%|██████▏   | 4028/6470 [50:28<30:49,  1.32it/s]
2|4029|Loss: 0.7979098558425903:  62%|██████▏   | 4029/6470 [50:29<29:42,  1.37it/s]
2|4030|Loss: 0.9454031586647034:  62%|██████▏   | 4029/6470 [50:29<29:42,  1.37it/s]
2|4030|Loss: 0.9454031586647034:  62%|██████▏   | 4030/6470 [50:29<28:34,  1.42it/s]
2|4031|Loss: 0.8604137897491455:  62%|██████▏   | 4030/6470 [50:30<28:34,  1.42it/s]
2|4031|Loss: 0.8604137897491455:  62%|██████▏   | 4031/6470 [50:30<29:08,  1.39it/s]
2|4032|Loss: 0.8973363041877747:  62%|██████▏   | 4031/6470 [50:31<29:08,  1.39it/s]
2|4032|Loss: 0.8973363041877747:  62%|██████▏   | 4032/6470 [50:31<28:13,  1.44it/s]
2|4033|Loss: 0.8551339507102966:  62%|██████▏   | 4032/6470 [50:31<28:13,  1.44it/s]
2|4033|Loss: 0.8551339507102966:  62%|██████▏   | 4033/6470 [50:32<28:54,  1.40it/s]
2|4034|Loss: 0.8208683729171753:  62%|██████▏   | 4033/6470 [50:32<28:54,  1.40it/s]
2|4034|Loss: 0.8208683729171753:  62%|██████▏   | 4034/6470 [50:33<31:51,  1.27it/s]
2|4035|Loss: 0.8931152820587158:  62%|██████▏   | 4034/6470 [50:33<31:51,  1.27it/s]
2|4035|Loss: 0.8931152820587158:  62%|██████▏   | 4035/6470 [50:33<33:21,  1.22it/s]
2|4036|Loss: 0.8354969620704651:  62%|██████▏   | 4035/6470 [50:34<33:21,  1.22it/s]
2|4036|Loss: 0.8354969620704651:  62%|██████▏   | 4036/6470 [50:34<30:04,  1.35it/s]
2|4037|Loss: 0.9031304121017456:  62%|██████▏   | 4036/6470 [50:34<30:04,  1.35it/s]
2|4037|Loss: 0.9031304121017456:  62%|██████▏   | 4037/6470 [50:35<30:37,  1.32it/s]
2|4038|Loss: 0.880585789680481:  62%|██████▏   | 4037/6470 [50:35<30:37,  1.32it/s] 
2|4038|Loss: 0.880585789680481:  62%|██████▏   | 4038/6470 [50:36<30:52,  1.31it/s]
2|4039|Loss: 0.8710222840309143:  62%|██████▏   | 4038/6470 [50:36<30:52,  1.31it/s]
2|4039|Loss: 0.8710222840309143:  62%|██████▏   | 4039/6470 [50:36<30:58,  1.31it/s]
2|4040|Loss: 0.831009566783905:  62%|██████▏   | 4039/6470 [50:37<30:58,  1.31it/s] 
2|4040|Loss: 0.831009566783905:  62%|██████▏   | 4040/6470 [50:37<31:09,  1.30it/s]
2|4041|Loss: 0.8154096603393555:  62%|██████▏   | 4040/6470 [50:38<31:09,  1.30it/s]
2|4041|Loss: 0.8154096603393555:  62%|██████▏   | 4041/6470 [50:38<31:15,  1.30it/s]
2|4042|Loss: 0.9367416501045227:  62%|██████▏   | 4041/6470 [50:38<31:15,  1.30it/s]
2|4042|Loss: 0.9367416501045227:  62%|██████▏   | 4042/6470 [50:39<29:48,  1.36it/s]
2|4043|Loss: 0.8799276351928711:  62%|██████▏   | 4042/6470 [50:39<29:48,  1.36it/s]
2|4043|Loss: 0.8799276351928711:  62%|██████▏   | 4043/6470 [50:39<27:31,  1.47it/s]
2|4044|Loss: 0.7901871204376221:  62%|██████▏   | 4043/6470 [50:39<27:31,  1.47it/s]
2|4044|Loss: 0.7901871204376221:  63%|██████▎   | 4044/6470 [50:40<28:06,  1.44it/s]
2|4045|Loss: 0.8539573550224304:  63%|██████▎   | 4044/6470 [50:40<28:06,  1.44it/s]
2|4045|Loss: 0.8539573550224304:  63%|██████▎   | 4045/6470 [50:41<28:23,  1.42it/s]
2|4046|Loss: 0.878216028213501:  63%|██████▎   | 4045/6470 [50:41<28:23,  1.42it/s] 
2|4046|Loss: 0.878216028213501:  63%|██████▎   | 4046/6470 [50:41<27:45,  1.46it/s]
2|4047|Loss: 0.8933774828910828:  63%|██████▎   | 4046/6470 [50:42<27:45,  1.46it/s]
2|4047|Loss: 0.8933774828910828:  63%|██████▎   | 4047/6470 [50:42<27:17,  1.48it/s]
2|4048|Loss: 0.9018127918243408:  63%|██████▎   | 4047/6470 [50:42<27:17,  1.48it/s]
2|4048|Loss: 0.9018127918243408:  63%|██████▎   | 4048/6470 [50:43<27:10,  1.49it/s]
2|4049|Loss: 0.7898250818252563:  63%|██████▎   | 4048/6470 [50:43<27:10,  1.49it/s]
2|4049|Loss: 0.7898250818252563:  63%|██████▎   | 4049/6470 [50:43<27:12,  1.48it/s]
2|4050|Loss: 0.8796117901802063:  63%|██████▎   | 4049/6470 [50:44<27:12,  1.48it/s]
2|4050|Loss: 0.8796117901802063:  63%|██████▎   | 4050/6470 [50:44<29:10,  1.38it/s]
2|4051|Loss: 0.9636426568031311:  63%|██████▎   | 4050/6470 [50:45<29:10,  1.38it/s]
2|4051|Loss: 0.9636426568031311:  63%|██████▎   | 4051/6470 [50:45<31:43,  1.27it/s]
2|4052|Loss: 1.0028282403945923:  63%|██████▎   | 4051/6470 [50:45<31:43,  1.27it/s]
2|4052|Loss: 1.0028282403945923:  63%|██████▎   | 4052/6470 [50:46<33:48,  1.19it/s]
2|4053|Loss: 0.7352921366691589:  63%|██████▎   | 4052/6470 [50:46<33:48,  1.19it/s]
2|4053|Loss: 0.7352921366691589:  63%|██████▎   | 4053/6470 [50:47<30:45,  1.31it/s]
2|4054|Loss: 0.8719968199729919:  63%|██████▎   | 4053/6470 [50:47<30:45,  1.31it/s]
2|4054|Loss: 0.8719968199729919:  63%|██████▎   | 4054/6470 [50:47<30:26,  1.32it/s]
2|4055|Loss: 0.8095523715019226:  63%|██████▎   | 4054/6470 [50:48<30:26,  1.32it/s]
2|4055|Loss: 0.8095523715019226:  63%|██████▎   | 4055/6470 [50:48<31:37,  1.27it/s]
2|4056|Loss: 0.7464132308959961:  63%|██████▎   | 4055/6470 [50:49<31:37,  1.27it/s]
2|4056|Loss: 0.7464132308959961:  63%|██████▎   | 4056/6470 [50:49<32:11,  1.25it/s]
2|4057|Loss: 0.6999524831771851:  63%|██████▎   | 4056/6470 [50:49<32:11,  1.25it/s]
2|4057|Loss: 0.6999524831771851:  63%|██████▎   | 4057/6470 [50:50<32:29,  1.24it/s]
2|4058|Loss: 0.8663806319236755:  63%|██████▎   | 4057/6470 [50:50<32:29,  1.24it/s]
2|4058|Loss: 0.8663806319236755:  63%|██████▎   | 4058/6470 [50:50<31:28,  1.28it/s]
2|4059|Loss: 0.8097546100616455:  63%|██████▎   | 4058/6470 [50:51<31:28,  1.28it/s]
2|4059|Loss: 0.8097546100616455:  63%|██████▎   | 4059/6470 [50:51<29:22,  1.37it/s]
2|4060|Loss: 0.8286711573600769:  63%|██████▎   | 4059/6470 [50:52<29:22,  1.37it/s]
2|4060|Loss: 0.8286711573600769:  63%|██████▎   | 4060/6470 [50:52<31:20,  1.28it/s]
2|4061|Loss: 0.7903475761413574:  63%|██████▎   | 4060/6470 [50:52<31:20,  1.28it/s]
2|4061|Loss: 0.7903475761413574:  63%|██████▎   | 4061/6470 [50:53<28:06,  1.43it/s]
2|4062|Loss: 0.7710920572280884:  63%|██████▎   | 4061/6470 [50:53<28:06,  1.43it/s]
2|4062|Loss: 0.7710920572280884:  63%|██████▎   | 4062/6470 [50:53<25:15,  1.59it/s]
2|4063|Loss: 0.9010587334632874:  63%|██████▎   | 4062/6470 [50:53<25:15,  1.59it/s]
2|4063|Loss: 0.9010587334632874:  63%|██████▎   | 4063/6470 [50:54<26:29,  1.51it/s]
2|4064|Loss: 0.8372594714164734:  63%|██████▎   | 4063/6470 [50:54<26:29,  1.51it/s]
2|4064|Loss: 0.8372594714164734:  63%|██████▎   | 4064/6470 [50:54<26:56,  1.49it/s]
2|4065|Loss: 0.9101786613464355:  63%|██████▎   | 4064/6470 [50:55<26:56,  1.49it/s]
2|4065|Loss: 0.9101786613464355:  63%|██████▎   | 4065/6470 [50:55<28:13,  1.42it/s]
2|4066|Loss: 0.8442885279655457:  63%|██████▎   | 4065/6470 [50:56<28:13,  1.42it/s]
2|4066|Loss: 0.8442885279655457:  63%|██████▎   | 4066/6470 [50:56<28:25,  1.41it/s]
2|4067|Loss: 0.872789204120636:  63%|██████▎   | 4066/6470 [50:56<28:25,  1.41it/s] 
2|4067|Loss: 0.872789204120636:  63%|██████▎   | 4067/6470 [50:57<30:49,  1.30it/s]
2|4068|Loss: 0.969157874584198:  63%|██████▎   | 4067/6470 [50:57<30:49,  1.30it/s]
2|4068|Loss: 0.969157874584198:  63%|██████▎   | 4068/6470 [50:58<34:09,  1.17it/s]
2|4069|Loss: 0.9472158551216125:  63%|██████▎   | 4068/6470 [50:58<34:09,  1.17it/s]
2|4069|Loss: 0.9472158551216125:  63%|██████▎   | 4069/6470 [50:59<31:36,  1.27it/s]
2|4070|Loss: 0.7433138489723206:  63%|██████▎   | 4069/6470 [50:59<31:36,  1.27it/s]
2|4070|Loss: 0.7433138489723206:  63%|██████▎   | 4070/6470 [50:59<30:36,  1.31it/s]
2|4071|Loss: 0.6984356045722961:  63%|██████▎   | 4070/6470 [51:00<30:36,  1.31it/s]
2|4071|Loss: 0.6984356045722961:  63%|██████▎   | 4071/6470 [51:00<30:58,  1.29it/s]
2|4072|Loss: 0.8027738332748413:  63%|██████▎   | 4071/6470 [51:00<30:58,  1.29it/s]
2|4072|Loss: 0.8027738332748413:  63%|██████▎   | 4072/6470 [51:01<28:58,  1.38it/s]
2|4073|Loss: 0.740791380405426:  63%|██████▎   | 4072/6470 [51:01<28:58,  1.38it/s] 
2|4073|Loss: 0.740791380405426:  63%|██████▎   | 4073/6470 [51:01<28:10,  1.42it/s]
2|4074|Loss: 0.8203428387641907:  63%|██████▎   | 4073/6470 [51:02<28:10,  1.42it/s]
2|4074|Loss: 0.8203428387641907:  63%|██████▎   | 4074/6470 [51:02<30:17,  1.32it/s]
2|4075|Loss: 0.7283900380134583:  63%|██████▎   | 4074/6470 [51:03<30:17,  1.32it/s]
2|4075|Loss: 0.7283900380134583:  63%|██████▎   | 4075/6470 [51:03<32:42,  1.22it/s]
2|4076|Loss: 0.9043858647346497:  63%|██████▎   | 4075/6470 [51:04<32:42,  1.22it/s]
2|4076|Loss: 0.9043858647346497:  63%|██████▎   | 4076/6470 [51:04<33:07,  1.20it/s]
2|4077|Loss: 0.8142050504684448:  63%|██████▎   | 4076/6470 [51:04<33:07,  1.20it/s]
2|4077|Loss: 0.8142050504684448:  63%|██████▎   | 4077/6470 [51:05<32:37,  1.22it/s]
2|4078|Loss: 0.783888578414917:  63%|██████▎   | 4077/6470 [51:05<32:37,  1.22it/s] 
2|4078|Loss: 0.783888578414917:  63%|██████▎   | 4078/6470 [51:06<31:51,  1.25it/s]
2|4079|Loss: 0.8425008058547974:  63%|██████▎   | 4078/6470 [51:06<31:51,  1.25it/s]
2|4079|Loss: 0.8425008058547974:  63%|██████▎   | 4079/6470 [51:06<30:42,  1.30it/s]
2|4080|Loss: 0.7141624093055725:  63%|██████▎   | 4079/6470 [51:07<30:42,  1.30it/s]
2|4080|Loss: 0.7141624093055725:  63%|██████▎   | 4080/6470 [51:07<30:48,  1.29it/s]
2|4081|Loss: 0.9014761447906494:  63%|██████▎   | 4080/6470 [51:07<30:48,  1.29it/s]
2|4081|Loss: 0.9014761447906494:  63%|██████▎   | 4081/6470 [51:08<28:41,  1.39it/s]
2|4082|Loss: 0.8989958763122559:  63%|██████▎   | 4081/6470 [51:08<28:41,  1.39it/s]
2|4082|Loss: 0.8989958763122559:  63%|██████▎   | 4082/6470 [51:08<26:28,  1.50it/s]
2|4083|Loss: 0.9613386988639832:  63%|██████▎   | 4082/6470 [51:09<26:28,  1.50it/s]
2|4083|Loss: 0.9613386988639832:  63%|██████▎   | 4083/6470 [51:09<27:21,  1.45it/s]
2|4084|Loss: 0.7188745737075806:  63%|██████▎   | 4083/6470 [51:09<27:21,  1.45it/s]
2|4084|Loss: 0.7188745737075806:  63%|██████▎   | 4084/6470 [51:10<28:09,  1.41it/s]
2|4085|Loss: 0.9431654214859009:  63%|██████▎   | 4084/6470 [51:10<28:09,  1.41it/s]
2|4085|Loss: 0.9431654214859009:  63%|██████▎   | 4085/6470 [51:11<30:07,  1.32it/s]
2|4086|Loss: 0.7539451122283936:  63%|██████▎   | 4085/6470 [51:11<30:07,  1.32it/s]
2|4086|Loss: 0.7539451122283936:  63%|██████▎   | 4086/6470 [51:11<30:56,  1.28it/s]
2|4087|Loss: 0.8318042159080505:  63%|██████▎   | 4086/6470 [51:12<30:56,  1.28it/s]
2|4087|Loss: 0.8318042159080505:  63%|██████▎   | 4087/6470 [51:12<32:41,  1.21it/s]
2|4088|Loss: 0.8714392781257629:  63%|██████▎   | 4087/6470 [51:13<32:41,  1.21it/s]
2|4088|Loss: 0.8714392781257629:  63%|██████▎   | 4088/6470 [51:13<33:23,  1.19it/s]
2|4089|Loss: 0.9391968250274658:  63%|██████▎   | 4088/6470 [51:14<33:23,  1.19it/s]
2|4089|Loss: 0.9391968250274658:  63%|██████▎   | 4089/6470 [51:14<32:58,  1.20it/s]
2|4090|Loss: 0.8866523504257202:  63%|██████▎   | 4089/6470 [51:14<32:58,  1.20it/s]
2|4090|Loss: 0.8866523504257202:  63%|██████▎   | 4090/6470 [51:15<32:28,  1.22it/s]
2|4091|Loss: 0.889561653137207:  63%|██████▎   | 4090/6470 [51:15<32:28,  1.22it/s] 
2|4091|Loss: 0.889561653137207:  63%|██████▎   | 4091/6470 [51:16<33:39,  1.18it/s]
2|4092|Loss: 0.9006634950637817:  63%|██████▎   | 4091/6470 [51:16<33:39,  1.18it/s]
2|4092|Loss: 0.9006634950637817:  63%|██████▎   | 4092/6470 [51:16<31:44,  1.25it/s]
2|4093|Loss: 0.7913945317268372:  63%|██████▎   | 4092/6470 [51:17<31:44,  1.25it/s]
2|4093|Loss: 0.7913945317268372:  63%|██████▎   | 4093/6470 [51:17<31:43,  1.25it/s]
2|4094|Loss: 0.7066242694854736:  63%|██████▎   | 4093/6470 [51:18<31:43,  1.25it/s]
2|4094|Loss: 0.7066242694854736:  63%|██████▎   | 4094/6470 [51:18<30:52,  1.28it/s]
2|4095|Loss: 0.7927622199058533:  63%|██████▎   | 4094/6470 [51:18<30:52,  1.28it/s]
2|4095|Loss: 0.7927622199058533:  63%|██████▎   | 4095/6470 [51:18<28:09,  1.41it/s]
2|4096|Loss: 0.7678375840187073:  63%|██████▎   | 4095/6470 [51:19<28:09,  1.41it/s]
2|4096|Loss: 0.7678375840187073:  63%|██████▎   | 4096/6470 [51:19<27:54,  1.42it/s]
2|4097|Loss: 0.7496863007545471:  63%|██████▎   | 4096/6470 [51:20<27:54,  1.42it/s]
2|4097|Loss: 0.7496863007545471:  63%|██████▎   | 4097/6470 [51:20<30:13,  1.31it/s]
2|4098|Loss: 0.8467194437980652:  63%|██████▎   | 4097/6470 [51:21<30:13,  1.31it/s]
2|4098|Loss: 0.8467194437980652:  63%|██████▎   | 4098/6470 [51:21<30:56,  1.28it/s]
2|4099|Loss: 0.9713380932807922:  63%|██████▎   | 4098/6470 [51:21<30:56,  1.28it/s]
2|4099|Loss: 0.9713380932807922:  63%|██████▎   | 4099/6470 [51:22<32:06,  1.23it/s]
2|4100|Loss: 0.8881975412368774:  63%|██████▎   | 4099/6470 [51:22<32:06,  1.23it/s]
2|4100|Loss: 0.8881975412368774:  63%|██████▎   | 4100/6470 [51:22<30:52,  1.28it/s]
2|4101|Loss: 0.6941870450973511:  63%|██████▎   | 4100/6470 [51:23<30:52,  1.28it/s]
2|4101|Loss: 0.6941870450973511:  63%|██████▎   | 4101/6470 [51:23<29:19,  1.35it/s]
2|4102|Loss: 0.873128354549408:  63%|██████▎   | 4101/6470 [51:24<29:19,  1.35it/s] 
2|4102|Loss: 0.873128354549408:  63%|██████▎   | 4102/6470 [51:24<29:41,  1.33it/s]
2|4103|Loss: 0.7352189421653748:  63%|██████▎   | 4102/6470 [51:24<29:41,  1.33it/s]
2|4103|Loss: 0.7352189421653748:  63%|██████▎   | 4103/6470 [51:24<27:53,  1.41it/s]
2|4104|Loss: 0.8972498178482056:  63%|██████▎   | 4103/6470 [51:25<27:53,  1.41it/s]
2|4104|Loss: 0.8972498178482056:  63%|██████▎   | 4104/6470 [51:25<26:58,  1.46it/s]
2|4105|Loss: 0.8900558948516846:  63%|██████▎   | 4104/6470 [51:26<26:58,  1.46it/s]
2|4105|Loss: 0.8900558948516846:  63%|██████▎   | 4105/6470 [51:26<27:09,  1.45it/s]
2|4106|Loss: 1.065287709236145:  63%|██████▎   | 4105/6470 [51:26<27:09,  1.45it/s] 
2|4106|Loss: 1.065287709236145:  63%|██████▎   | 4106/6470 [51:27<27:49,  1.42it/s]
2|4107|Loss: 0.8003206253051758:  63%|██████▎   | 4106/6470 [51:27<27:49,  1.42it/s]
2|4107|Loss: 0.8003206253051758:  63%|██████▎   | 4107/6470 [51:27<28:23,  1.39it/s]
2|4108|Loss: 0.9240427613258362:  63%|██████▎   | 4107/6470 [51:28<28:23,  1.39it/s]
2|4108|Loss: 0.9240427613258362:  63%|██████▎   | 4108/6470 [51:28<28:18,  1.39it/s]
2|4109|Loss: 0.8138962388038635:  63%|██████▎   | 4108/6470 [51:28<28:18,  1.39it/s]
2|4109|Loss: 0.8138962388038635:  64%|██████▎   | 4109/6470 [51:29<27:49,  1.41it/s]
2|4110|Loss: 0.8798375129699707:  64%|██████▎   | 4109/6470 [51:29<27:49,  1.41it/s]
2|4110|Loss: 0.8798375129699707:  64%|██████▎   | 4110/6470 [51:29<28:10,  1.40it/s]
2|4111|Loss: 0.9346928000450134:  64%|██████▎   | 4110/6470 [51:30<28:10,  1.40it/s]
2|4111|Loss: 0.9346928000450134:  64%|██████▎   | 4111/6470 [51:30<30:00,  1.31it/s]
2|4112|Loss: 0.7372446060180664:  64%|██████▎   | 4111/6470 [51:31<30:00,  1.31it/s]
2|4112|Loss: 0.7372446060180664:  64%|██████▎   | 4112/6470 [51:31<28:10,  1.40it/s]
2|4113|Loss: 0.7857977151870728:  64%|██████▎   | 4112/6470 [51:31<28:10,  1.40it/s]
2|4113|Loss: 0.7857977151870728:  64%|██████▎   | 4113/6470 [51:32<26:51,  1.46it/s]
2|4114|Loss: 0.8961867094039917:  64%|██████▎   | 4113/6470 [51:32<26:51,  1.46it/s]
2|4114|Loss: 0.8961867094039917:  64%|██████▎   | 4114/6470 [51:32<28:59,  1.35it/s]
2|4115|Loss: 0.8058054447174072:  64%|██████▎   | 4114/6470 [51:33<28:59,  1.35it/s]
2|4115|Loss: 0.8058054447174072:  64%|██████▎   | 4115/6470 [51:33<29:46,  1.32it/s]
2|4116|Loss: 0.5315187573432922:  64%|██████▎   | 4115/6470 [51:34<29:46,  1.32it/s]
2|4116|Loss: 0.5315187573432922:  64%|██████▎   | 4116/6470 [51:34<32:31,  1.21it/s]
2|4117|Loss: 0.6370983719825745:  64%|██████▎   | 4116/6470 [51:35<32:31,  1.21it/s]
2|4117|Loss: 0.6370983719825745:  64%|██████▎   | 4117/6470 [51:35<28:54,  1.36it/s]
2|4118|Loss: 0.7638438940048218:  64%|██████▎   | 4117/6470 [51:35<28:54,  1.36it/s]
2|4118|Loss: 0.7638438940048218:  64%|██████▎   | 4118/6470 [51:35<27:45,  1.41it/s]
2|4119|Loss: 0.8720449209213257:  64%|██████▎   | 4118/6470 [51:36<27:45,  1.41it/s]
2|4119|Loss: 0.8720449209213257:  64%|██████▎   | 4119/6470 [51:36<27:29,  1.43it/s]
2|4120|Loss: 0.9553437829017639:  64%|██████▎   | 4119/6470 [51:36<27:29,  1.43it/s]
2|4120|Loss: 0.9553437829017639:  64%|██████▎   | 4120/6470 [51:37<29:33,  1.32it/s]
2|4121|Loss: 0.7553159594535828:  64%|██████▎   | 4120/6470 [51:37<29:33,  1.32it/s]
2|4121|Loss: 0.7553159594535828:  64%|██████▎   | 4121/6470 [51:38<29:58,  1.31it/s]
2|4122|Loss: 0.9018579721450806:  64%|██████▎   | 4121/6470 [51:38<29:58,  1.31it/s]
2|4122|Loss: 0.9018579721450806:  64%|██████▎   | 4122/6470 [51:38<28:36,  1.37it/s]
2|4123|Loss: 0.9467509388923645:  64%|██████▎   | 4122/6470 [51:39<28:36,  1.37it/s]
2|4123|Loss: 0.9467509388923645:  64%|██████▎   | 4123/6470 [51:39<26:29,  1.48it/s]
2|4124|Loss: 0.7424687147140503:  64%|██████▎   | 4123/6470 [51:39<26:29,  1.48it/s]
2|4124|Loss: 0.7424687147140503:  64%|██████▎   | 4124/6470 [51:40<26:00,  1.50it/s]
2|4125|Loss: 0.8665237426757812:  64%|██████▎   | 4124/6470 [51:40<26:00,  1.50it/s]
2|4125|Loss: 0.8665237426757812:  64%|██████▍   | 4125/6470 [51:40<28:55,  1.35it/s]
2|4126|Loss: 0.6374229192733765:  64%|██████▍   | 4125/6470 [51:41<28:55,  1.35it/s]
2|4126|Loss: 0.6374229192733765:  64%|██████▍   | 4126/6470 [51:41<27:40,  1.41it/s]
2|4127|Loss: 0.84980708360672:  64%|██████▍   | 4126/6470 [51:41<27:40,  1.41it/s]  
2|4127|Loss: 0.84980708360672:  64%|██████▍   | 4127/6470 [51:42<26:44,  1.46it/s]
2|4128|Loss: 0.7042921781539917:  64%|██████▍   | 4127/6470 [51:42<26:44,  1.46it/s]
2|4128|Loss: 0.7042921781539917:  64%|██████▍   | 4128/6470 [51:42<26:45,  1.46it/s]
2|4129|Loss: 0.8324918150901794:  64%|██████▍   | 4128/6470 [51:43<26:45,  1.46it/s]
2|4129|Loss: 0.8324918150901794:  64%|██████▍   | 4129/6470 [51:43<27:42,  1.41it/s]
2|4130|Loss: 0.7265382409095764:  64%|██████▍   | 4129/6470 [51:44<27:42,  1.41it/s]
2|4130|Loss: 0.7265382409095764:  64%|██████▍   | 4130/6470 [51:44<28:36,  1.36it/s]
2|4131|Loss: 0.7807215452194214:  64%|██████▍   | 4130/6470 [51:44<28:36,  1.36it/s]
2|4131|Loss: 0.7807215452194214:  64%|██████▍   | 4131/6470 [51:45<27:03,  1.44it/s]
2|4132|Loss: 0.8326717615127563:  64%|██████▍   | 4131/6470 [51:45<27:03,  1.44it/s]
2|4132|Loss: 0.8326717615127563:  64%|██████▍   | 4132/6470 [51:45<25:40,  1.52it/s]
2|4133|Loss: 0.7341575026512146:  64%|██████▍   | 4132/6470 [51:46<25:40,  1.52it/s]
2|4133|Loss: 0.7341575026512146:  64%|██████▍   | 4133/6470 [51:46<28:19,  1.38it/s]
2|4134|Loss: 0.9596903324127197:  64%|██████▍   | 4133/6470 [51:46<28:19,  1.38it/s]
2|4134|Loss: 0.9596903324127197:  64%|██████▍   | 4134/6470 [51:47<28:00,  1.39it/s]
2|4135|Loss: 0.7130391001701355:  64%|██████▍   | 4134/6470 [51:47<28:00,  1.39it/s]
2|4135|Loss: 0.7130391001701355:  64%|██████▍   | 4135/6470 [51:48<29:59,  1.30it/s]
2|4136|Loss: 0.7188073992729187:  64%|██████▍   | 4135/6470 [51:48<29:59,  1.30it/s]
2|4136|Loss: 0.7188073992729187:  64%|██████▍   | 4136/6470 [51:48<30:17,  1.28it/s]
2|4137|Loss: 0.9316422939300537:  64%|██████▍   | 4136/6470 [51:49<30:17,  1.28it/s]
2|4137|Loss: 0.9316422939300537:  64%|██████▍   | 4137/6470 [51:49<30:28,  1.28it/s]
2|4138|Loss: 0.9183928370475769:  64%|██████▍   | 4137/6470 [51:50<30:28,  1.28it/s]
2|4138|Loss: 0.9183928370475769:  64%|██████▍   | 4138/6470 [51:50<31:41,  1.23it/s]
2|4139|Loss: 0.7138442993164062:  64%|██████▍   | 4138/6470 [51:51<31:41,  1.23it/s]
2|4139|Loss: 0.7138442993164062:  64%|██████▍   | 4139/6470 [51:51<31:46,  1.22it/s]
2|4140|Loss: 0.9303814768791199:  64%|██████▍   | 4139/6470 [51:51<31:46,  1.22it/s]
2|4140|Loss: 0.9303814768791199:  64%|██████▍   | 4140/6470 [51:52<28:58,  1.34it/s]
2|4141|Loss: 0.86192387342453:  64%|██████▍   | 4140/6470 [51:52<28:58,  1.34it/s]  
2|4141|Loss: 0.86192387342453:  64%|██████▍   | 4141/6470 [51:52<27:09,  1.43it/s]
2|4142|Loss: 0.7669262290000916:  64%|██████▍   | 4141/6470 [51:52<27:09,  1.43it/s]
2|4142|Loss: 0.7669262290000916:  64%|██████▍   | 4142/6470 [51:53<24:54,  1.56it/s]
2|4143|Loss: 0.9150999188423157:  64%|██████▍   | 4142/6470 [51:53<24:54,  1.56it/s]
2|4143|Loss: 0.9150999188423157:  64%|██████▍   | 4143/6470 [51:53<27:12,  1.43it/s]
2|4144|Loss: 0.8987030386924744:  64%|██████▍   | 4143/6470 [51:54<27:12,  1.43it/s]
2|4144|Loss: 0.8987030386924744:  64%|██████▍   | 4144/6470 [51:54<30:43,  1.26it/s]
2|4145|Loss: 0.9426589608192444:  64%|██████▍   | 4144/6470 [51:55<30:43,  1.26it/s]
2|4145|Loss: 0.9426589608192444:  64%|██████▍   | 4145/6470 [51:56<36:22,  1.07it/s]
2|4146|Loss: 0.7933638691902161:  64%|██████▍   | 4145/6470 [51:56<36:22,  1.07it/s]
2|4146|Loss: 0.7933638691902161:  64%|██████▍   | 4146/6470 [51:57<36:36,  1.06it/s]
2|4147|Loss: 0.7220696806907654:  64%|██████▍   | 4146/6470 [51:57<36:36,  1.06it/s]
2|4147|Loss: 0.7220696806907654:  64%|██████▍   | 4147/6470 [51:58<35:35,  1.09it/s]
2|4148|Loss: 0.8078857660293579:  64%|██████▍   | 4147/6470 [51:58<35:35,  1.09it/s]
2|4148|Loss: 0.8078857660293579:  64%|██████▍   | 4148/6470 [51:58<33:11,  1.17it/s]
2|4149|Loss: 0.8106914758682251:  64%|██████▍   | 4148/6470 [51:59<33:11,  1.17it/s]
2|4149|Loss: 0.8106914758682251:  64%|██████▍   | 4149/6470 [51:59<30:57,  1.25it/s]
2|4150|Loss: 0.7327767610549927:  64%|██████▍   | 4149/6470 [51:59<30:57,  1.25it/s]
2|4150|Loss: 0.7327767610549927:  64%|██████▍   | 4150/6470 [52:00<32:37,  1.19it/s]
2|4151|Loss: 0.7283828258514404:  64%|██████▍   | 4150/6470 [52:00<32:37,  1.19it/s]
2|4151|Loss: 0.7283828258514404:  64%|██████▍   | 4151/6470 [52:01<32:35,  1.19it/s]
2|4152|Loss: 0.8417064547538757:  64%|██████▍   | 4151/6470 [52:01<32:35,  1.19it/s]
2|4152|Loss: 0.8417064547538757:  64%|██████▍   | 4152/6470 [52:02<32:10,  1.20it/s]
2|4153|Loss: 0.8739868402481079:  64%|██████▍   | 4152/6470 [52:02<32:10,  1.20it/s]
2|4153|Loss: 0.8739868402481079:  64%|██████▍   | 4153/6470 [52:02<31:48,  1.21it/s]
2|4154|Loss: 0.8791214823722839:  64%|██████▍   | 4153/6470 [52:03<31:48,  1.21it/s]
2|4154|Loss: 0.8791214823722839:  64%|██████▍   | 4154/6470 [52:03<32:23,  1.19it/s]
2|4155|Loss: 0.9918984174728394:  64%|██████▍   | 4154/6470 [52:04<32:23,  1.19it/s]
2|4155|Loss: 0.9918984174728394:  64%|██████▍   | 4155/6470 [52:04<33:16,  1.16it/s]
2|4156|Loss: 0.7000357508659363:  64%|██████▍   | 4155/6470 [52:05<33:16,  1.16it/s]
2|4156|Loss: 0.7000357508659363:  64%|██████▍   | 4156/6470 [52:05<30:55,  1.25it/s]
2|4157|Loss: 0.6743345856666565:  64%|██████▍   | 4156/6470 [52:05<30:55,  1.25it/s]
2|4157|Loss: 0.6743345856666565:  64%|██████▍   | 4157/6470 [52:05<25:58,  1.48it/s]
2|4158|Loss: 0.858476996421814:  64%|██████▍   | 4157/6470 [52:05<25:58,  1.48it/s] 
2|4158|Loss: 0.858476996421814:  64%|██████▍   | 4158/6470 [52:06<24:51,  1.55it/s]
2|4159|Loss: 0.7622056007385254:  64%|██████▍   | 4158/6470 [52:06<24:51,  1.55it/s]
2|4159|Loss: 0.7622056007385254:  64%|██████▍   | 4159/6470 [52:07<26:55,  1.43it/s]
2|4160|Loss: 0.7869152426719666:  64%|██████▍   | 4159/6470 [52:07<26:55,  1.43it/s]
2|4160|Loss: 0.7869152426719666:  64%|██████▍   | 4160/6470 [52:07<28:05,  1.37it/s]
2|4161|Loss: 0.7618150115013123:  64%|██████▍   | 4160/6470 [52:08<28:05,  1.37it/s]
2|4161|Loss: 0.7618150115013123:  64%|██████▍   | 4161/6470 [52:08<29:28,  1.31it/s]
2|4162|Loss: 0.9017333388328552:  64%|██████▍   | 4161/6470 [52:09<29:28,  1.31it/s]
2|4162|Loss: 0.9017333388328552:  64%|██████▍   | 4162/6470 [52:09<30:33,  1.26it/s]
2|4163|Loss: 0.7985390424728394:  64%|██████▍   | 4162/6470 [52:10<30:33,  1.26it/s]
2|4163|Loss: 0.7985390424728394:  64%|██████▍   | 4163/6470 [52:10<31:13,  1.23it/s]
2|4164|Loss: 0.7551938891410828:  64%|██████▍   | 4163/6470 [52:10<31:13,  1.23it/s]
2|4164|Loss: 0.7551938891410828:  64%|██████▍   | 4164/6470 [52:11<30:59,  1.24it/s]
2|4165|Loss: 0.8950671553611755:  64%|██████▍   | 4164/6470 [52:11<30:59,  1.24it/s]
2|4165|Loss: 0.8950671553611755:  64%|██████▍   | 4165/6470 [52:12<31:22,  1.22it/s]
2|4166|Loss: 0.8945786952972412:  64%|██████▍   | 4165/6470 [52:12<31:22,  1.22it/s]
2|4166|Loss: 0.8945786952972412:  64%|██████▍   | 4166/6470 [52:13<33:18,  1.15it/s]
2|4167|Loss: 0.8219611048698425:  64%|██████▍   | 4166/6470 [52:13<33:18,  1.15it/s]
2|4167|Loss: 0.8219611048698425:  64%|██████▍   | 4167/6470 [52:13<32:09,  1.19it/s]
2|4168|Loss: 0.7771803736686707:  64%|██████▍   | 4167/6470 [52:14<32:09,  1.19it/s]
2|4168|Loss: 0.7771803736686707:  64%|██████▍   | 4168/6470 [52:14<31:09,  1.23it/s]
2|4169|Loss: 0.8498362898826599:  64%|██████▍   | 4168/6470 [52:14<31:09,  1.23it/s]
2|4169|Loss: 0.8498362898826599:  64%|██████▍   | 4169/6470 [52:15<29:22,  1.31it/s]
2|4170|Loss: 1.0293023586273193:  64%|██████▍   | 4169/6470 [52:15<29:22,  1.31it/s]
2|4170|Loss: 1.0293023586273193:  64%|██████▍   | 4170/6470 [52:15<28:17,  1.35it/s]
2|4171|Loss: 0.8320675492286682:  64%|██████▍   | 4170/6470 [52:16<28:17,  1.35it/s]
2|4171|Loss: 0.8320675492286682:  64%|██████▍   | 4171/6470 [52:16<28:42,  1.33it/s]
2|4172|Loss: 0.7947306632995605:  64%|██████▍   | 4171/6470 [52:17<28:42,  1.33it/s]
2|4172|Loss: 0.7947306632995605:  64%|██████▍   | 4172/6470 [52:17<29:23,  1.30it/s]
2|4173|Loss: 0.8008053302764893:  64%|██████▍   | 4172/6470 [52:17<29:23,  1.30it/s]
2|4173|Loss: 0.8008053302764893:  64%|██████▍   | 4173/6470 [52:18<29:51,  1.28it/s]
2|4174|Loss: 0.8477839231491089:  64%|██████▍   | 4173/6470 [52:18<29:51,  1.28it/s]
2|4174|Loss: 0.8477839231491089:  65%|██████▍   | 4174/6470 [52:19<29:55,  1.28it/s]
2|4175|Loss: 0.7794864773750305:  65%|██████▍   | 4174/6470 [52:19<29:55,  1.28it/s]
2|4175|Loss: 0.7794864773750305:  65%|██████▍   | 4175/6470 [52:19<31:22,  1.22it/s]
2|4176|Loss: 0.7938026785850525:  65%|██████▍   | 4175/6470 [52:20<31:22,  1.22it/s]
2|4176|Loss: 0.7938026785850525:  65%|██████▍   | 4176/6470 [52:20<30:21,  1.26it/s]
2|4177|Loss: 0.9033476710319519:  65%|██████▍   | 4176/6470 [52:21<30:21,  1.26it/s]
2|4177|Loss: 0.9033476710319519:  65%|██████▍   | 4177/6470 [52:21<30:40,  1.25it/s]
2|4178|Loss: 0.7606216669082642:  65%|██████▍   | 4177/6470 [52:22<30:40,  1.25it/s]
2|4178|Loss: 0.7606216669082642:  65%|██████▍   | 4178/6470 [52:22<30:48,  1.24it/s]
2|4179|Loss: 0.822073757648468:  65%|██████▍   | 4178/6470 [52:22<30:48,  1.24it/s] 
2|4179|Loss: 0.822073757648468:  65%|██████▍   | 4179/6470 [52:23<31:18,  1.22it/s]
2|4180|Loss: 0.739944338798523:  65%|██████▍   | 4179/6470 [52:23<31:18,  1.22it/s]
2|4180|Loss: 0.739944338798523:  65%|██████▍   | 4180/6470 [52:23<30:46,  1.24it/s]
2|4181|Loss: 0.8170815110206604:  65%|██████▍   | 4180/6470 [52:24<30:46,  1.24it/s]
2|4181|Loss: 0.8170815110206604:  65%|██████▍   | 4181/6470 [52:24<29:27,  1.30it/s]
2|4182|Loss: 0.704813539981842:  65%|██████▍   | 4181/6470 [52:25<29:27,  1.30it/s] 
2|4182|Loss: 0.704813539981842:  65%|██████▍   | 4182/6470 [52:25<27:32,  1.38it/s]
2|4183|Loss: 0.7330709099769592:  65%|██████▍   | 4182/6470 [52:25<27:32,  1.38it/s]
2|4183|Loss: 0.7330709099769592:  65%|██████▍   | 4183/6470 [52:26<27:53,  1.37it/s]
2|4184|Loss: 0.9415475130081177:  65%|██████▍   | 4183/6470 [52:26<27:53,  1.37it/s]
2|4184|Loss: 0.9415475130081177:  65%|██████▍   | 4184/6470 [52:26<27:45,  1.37it/s]
2|4185|Loss: 1.0095701217651367:  65%|██████▍   | 4184/6470 [52:27<27:45,  1.37it/s]
2|4185|Loss: 1.0095701217651367:  65%|██████▍   | 4185/6470 [52:27<27:57,  1.36it/s]
2|4186|Loss: 0.7068050503730774:  65%|██████▍   | 4185/6470 [52:27<27:57,  1.36it/s]
2|4186|Loss: 0.7068050503730774:  65%|██████▍   | 4186/6470 [52:28<27:30,  1.38it/s]
2|4187|Loss: 0.9349527955055237:  65%|██████▍   | 4186/6470 [52:28<27:30,  1.38it/s]
2|4187|Loss: 0.9349527955055237:  65%|██████▍   | 4187/6470 [52:28<28:00,  1.36it/s]
2|4188|Loss: 0.8020428419113159:  65%|██████▍   | 4187/6470 [52:29<28:00,  1.36it/s]
2|4188|Loss: 0.8020428419113159:  65%|██████▍   | 4188/6470 [52:29<27:30,  1.38it/s]
2|4189|Loss: 0.6807135343551636:  65%|██████▍   | 4188/6470 [52:30<27:30,  1.38it/s]
2|4189|Loss: 0.6807135343551636:  65%|██████▍   | 4189/6470 [52:30<26:28,  1.44it/s]
2|4190|Loss: 0.8733760118484497:  65%|██████▍   | 4189/6470 [52:30<26:28,  1.44it/s]
2|4190|Loss: 0.8733760118484497:  65%|██████▍   | 4190/6470 [52:30<25:36,  1.48it/s]
2|4191|Loss: 0.8845900297164917:  65%|██████▍   | 4190/6470 [52:31<25:36,  1.48it/s]
2|4191|Loss: 0.8845900297164917:  65%|██████▍   | 4191/6470 [52:31<27:49,  1.36it/s]
2|4192|Loss: 0.8592502474784851:  65%|██████▍   | 4191/6470 [52:32<27:49,  1.36it/s]
2|4192|Loss: 0.8592502474784851:  65%|██████▍   | 4192/6470 [52:32<28:32,  1.33it/s]
2|4193|Loss: 0.7492210268974304:  65%|██████▍   | 4192/6470 [52:33<28:32,  1.33it/s]
2|4193|Loss: 0.7492210268974304:  65%|██████▍   | 4193/6470 [52:33<30:23,  1.25it/s]
2|4194|Loss: 0.9379866719245911:  65%|██████▍   | 4193/6470 [52:33<30:23,  1.25it/s]
2|4194|Loss: 0.9379866719245911:  65%|██████▍   | 4194/6470 [52:34<28:55,  1.31it/s]
2|4195|Loss: 0.8204184770584106:  65%|██████▍   | 4194/6470 [52:34<28:55,  1.31it/s]
2|4195|Loss: 0.8204184770584106:  65%|██████▍   | 4195/6470 [52:34<26:51,  1.41it/s]
2|4196|Loss: 0.8288688659667969:  65%|██████▍   | 4195/6470 [52:35<26:51,  1.41it/s]
2|4196|Loss: 0.8288688659667969:  65%|██████▍   | 4196/6470 [52:35<28:33,  1.33it/s]
2|4197|Loss: 1.1222283840179443:  65%|██████▍   | 4196/6470 [52:36<28:33,  1.33it/s]
2|4197|Loss: 1.1222283840179443:  65%|██████▍   | 4197/6470 [52:36<30:25,  1.24it/s]
2|4198|Loss: 0.9091351628303528:  65%|██████▍   | 4197/6470 [52:37<30:25,  1.24it/s]
2|4198|Loss: 0.9091351628303528:  65%|██████▍   | 4198/6470 [52:37<31:46,  1.19it/s]
2|4199|Loss: 0.8102114796638489:  65%|██████▍   | 4198/6470 [52:37<31:46,  1.19it/s]
2|4199|Loss: 0.8102114796638489:  65%|██████▍   | 4199/6470 [52:38<29:24,  1.29it/s]
2|4200|Loss: 0.8439503908157349:  65%|██████▍   | 4199/6470 [52:38<29:24,  1.29it/s]
2|4200|Loss: 0.8439503908157349:  65%|██████▍   | 4200/6470 [52:38<29:08,  1.30it/s]
2|4201|Loss: 0.7143295407295227:  65%|██████▍   | 4200/6470 [52:39<29:08,  1.30it/s]
2|4201|Loss: 0.7143295407295227:  65%|██████▍   | 4201/6470 [52:39<29:37,  1.28it/s]
2|4202|Loss: 0.8803989887237549:  65%|██████▍   | 4201/6470 [52:40<29:37,  1.28it/s]
2|4202|Loss: 0.8803989887237549:  65%|██████▍   | 4202/6470 [52:40<28:36,  1.32it/s]
2|4203|Loss: 0.7966622710227966:  65%|██████▍   | 4202/6470 [52:40<28:36,  1.32it/s]
2|4203|Loss: 0.7966622710227966:  65%|██████▍   | 4203/6470 [52:41<28:15,  1.34it/s]
2|4204|Loss: 0.870903730392456:  65%|██████▍   | 4203/6470 [52:41<28:15,  1.34it/s] 
2|4204|Loss: 0.870903730392456:  65%|██████▍   | 4204/6470 [52:41<28:30,  1.32it/s]
2|4205|Loss: 0.8659354448318481:  65%|██████▍   | 4204/6470 [52:42<28:30,  1.32it/s]
2|4205|Loss: 0.8659354448318481:  65%|██████▍   | 4205/6470 [52:42<29:05,  1.30it/s]
2|4206|Loss: 0.88884037733078:  65%|██████▍   | 4205/6470 [52:43<29:05,  1.30it/s]  
2|4206|Loss: 0.88884037733078:  65%|██████▌   | 4206/6470 [52:43<27:57,  1.35it/s]
2|4207|Loss: 0.8115349411964417:  65%|██████▌   | 4206/6470 [52:43<27:57,  1.35it/s]
2|4207|Loss: 0.8115349411964417:  65%|██████▌   | 4207/6470 [52:44<27:42,  1.36it/s]
2|4208|Loss: 0.9443254470825195:  65%|██████▌   | 4207/6470 [52:44<27:42,  1.36it/s]
2|4208|Loss: 0.9443254470825195:  65%|██████▌   | 4208/6470 [52:44<27:09,  1.39it/s]
2|4209|Loss: 0.8675545454025269:  65%|██████▌   | 4208/6470 [52:45<27:09,  1.39it/s]
2|4209|Loss: 0.8675545454025269:  65%|██████▌   | 4209/6470 [52:45<29:12,  1.29it/s]
2|4210|Loss: 0.7532650828361511:  65%|██████▌   | 4209/6470 [52:46<29:12,  1.29it/s]
2|4210|Loss: 0.7532650828361511:  65%|██████▌   | 4210/6470 [52:46<27:20,  1.38it/s]
2|4211|Loss: 0.7845813035964966:  65%|██████▌   | 4210/6470 [52:46<27:20,  1.38it/s]
2|4211|Loss: 0.7845813035964966:  65%|██████▌   | 4211/6470 [52:46<24:53,  1.51it/s]
2|4212|Loss: 0.8980851173400879:  65%|██████▌   | 4211/6470 [52:47<24:53,  1.51it/s]
2|4212|Loss: 0.8980851173400879:  65%|██████▌   | 4212/6470 [52:47<25:19,  1.49it/s]
2|4213|Loss: 0.8268478512763977:  65%|██████▌   | 4212/6470 [52:47<25:19,  1.49it/s]
2|4213|Loss: 0.8268478512763977:  65%|██████▌   | 4213/6470 [52:48<24:22,  1.54it/s]
2|4214|Loss: 0.6983696222305298:  65%|██████▌   | 4213/6470 [52:48<24:22,  1.54it/s]
2|4214|Loss: 0.6983696222305298:  65%|██████▌   | 4214/6470 [52:48<23:53,  1.57it/s]
2|4215|Loss: 0.8911823630332947:  65%|██████▌   | 4214/6470 [52:49<23:53,  1.57it/s]
2|4215|Loss: 0.8911823630332947:  65%|██████▌   | 4215/6470 [52:49<24:58,  1.50it/s]
2|4216|Loss: 0.7915849685668945:  65%|██████▌   | 4215/6470 [52:49<24:58,  1.50it/s]
2|4216|Loss: 0.7915849685668945:  65%|██████▌   | 4216/6470 [52:50<25:10,  1.49it/s]
2|4217|Loss: 0.852999746799469:  65%|██████▌   | 4216/6470 [52:50<25:10,  1.49it/s] 
2|4217|Loss: 0.852999746799469:  65%|██████▌   | 4217/6470 [52:50<27:36,  1.36it/s]
2|4218|Loss: 0.7686978578567505:  65%|██████▌   | 4217/6470 [52:51<27:36,  1.36it/s]
2|4218|Loss: 0.7686978578567505:  65%|██████▌   | 4218/6470 [52:51<29:40,  1.26it/s]
2|4219|Loss: 0.8849473595619202:  65%|██████▌   | 4218/6470 [52:52<29:40,  1.26it/s]
2|4219|Loss: 0.8849473595619202:  65%|██████▌   | 4219/6470 [52:52<28:25,  1.32it/s]
2|4220|Loss: 0.7649955153465271:  65%|██████▌   | 4219/6470 [52:52<28:25,  1.32it/s]
2|4220|Loss: 0.7649955153465271:  65%|██████▌   | 4220/6470 [52:53<28:23,  1.32it/s]
2|4221|Loss: 0.8862223029136658:  65%|██████▌   | 4220/6470 [52:53<28:23,  1.32it/s]
2|4221|Loss: 0.8862223029136658:  65%|██████▌   | 4221/6470 [52:54<30:19,  1.24it/s]
2|4222|Loss: 0.8317568898200989:  65%|██████▌   | 4221/6470 [52:54<30:19,  1.24it/s]
2|4222|Loss: 0.8317568898200989:  65%|██████▌   | 4222/6470 [52:55<30:34,  1.23it/s]
2|4223|Loss: 0.8178358674049377:  65%|██████▌   | 4222/6470 [52:55<30:34,  1.23it/s]
2|4223|Loss: 0.8178358674049377:  65%|██████▌   | 4223/6470 [52:55<29:26,  1.27it/s]
2|4224|Loss: 0.8525802493095398:  65%|██████▌   | 4223/6470 [52:56<29:26,  1.27it/s]
2|4224|Loss: 0.8525802493095398:  65%|██████▌   | 4224/6470 [52:56<30:14,  1.24it/s]
2|4225|Loss: 0.861760675907135:  65%|██████▌   | 4224/6470 [52:57<30:14,  1.24it/s] 
2|4225|Loss: 0.861760675907135:  65%|██████▌   | 4225/6470 [52:57<29:26,  1.27it/s]
2|4226|Loss: 0.9566514492034912:  65%|██████▌   | 4225/6470 [52:57<29:26,  1.27it/s]
2|4226|Loss: 0.9566514492034912:  65%|██████▌   | 4226/6470 [52:58<29:09,  1.28it/s]
2|4227|Loss: 0.9803483486175537:  65%|██████▌   | 4226/6470 [52:58<29:09,  1.28it/s]
2|4227|Loss: 0.9803483486175537:  65%|██████▌   | 4227/6470 [52:58<29:09,  1.28it/s]
2|4228|Loss: 0.8578444123268127:  65%|██████▌   | 4227/6470 [52:59<29:09,  1.28it/s]
2|4228|Loss: 0.8578444123268127:  65%|██████▌   | 4228/6470 [52:59<29:56,  1.25it/s]
2|4229|Loss: 0.9177857041358948:  65%|██████▌   | 4228/6470 [53:00<29:56,  1.25it/s]
2|4229|Loss: 0.9177857041358948:  65%|██████▌   | 4229/6470 [53:00<28:53,  1.29it/s]
2|4230|Loss: 0.6846681833267212:  65%|██████▌   | 4229/6470 [53:00<28:53,  1.29it/s]
2|4230|Loss: 0.6846681833267212:  65%|██████▌   | 4230/6470 [53:00<24:45,  1.51it/s]
2|4231|Loss: 0.9800686240196228:  65%|██████▌   | 4230/6470 [53:01<24:45,  1.51it/s]
2|4231|Loss: 0.9800686240196228:  65%|██████▌   | 4231/6470 [53:01<26:06,  1.43it/s]
2|4232|Loss: 0.8938731551170349:  65%|██████▌   | 4231/6470 [53:02<26:06,  1.43it/s]
2|4232|Loss: 0.8938731551170349:  65%|██████▌   | 4232/6470 [53:02<28:59,  1.29it/s]
2|4233|Loss: 0.9539207220077515:  65%|██████▌   | 4232/6470 [53:03<28:59,  1.29it/s]
2|4233|Loss: 0.9539207220077515:  65%|██████▌   | 4233/6470 [53:03<29:43,  1.25it/s]
2|4234|Loss: 0.7618306875228882:  65%|██████▌   | 4233/6470 [53:03<29:43,  1.25it/s]
2|4234|Loss: 0.7618306875228882:  65%|██████▌   | 4234/6470 [53:04<30:52,  1.21it/s]
2|4235|Loss: 1.0538679361343384:  65%|██████▌   | 4234/6470 [53:04<30:52,  1.21it/s]
2|4235|Loss: 1.0538679361343384:  65%|██████▌   | 4235/6470 [53:05<32:03,  1.16it/s]
2|4236|Loss: 0.7870723605155945:  65%|██████▌   | 4235/6470 [53:05<32:03,  1.16it/s]
2|4236|Loss: 0.7870723605155945:  65%|██████▌   | 4236/6470 [53:06<30:51,  1.21it/s]
2|4237|Loss: 0.9404409527778625:  65%|██████▌   | 4236/6470 [53:06<30:51,  1.21it/s]
2|4237|Loss: 0.9404409527778625:  65%|██████▌   | 4237/6470 [53:06<30:55,  1.20it/s]
2|4238|Loss: 0.8272630572319031:  65%|██████▌   | 4237/6470 [53:07<30:55,  1.20it/s]
2|4238|Loss: 0.8272630572319031:  66%|██████▌   | 4238/6470 [53:07<30:17,  1.23it/s]
2|4239|Loss: 0.9163447022438049:  66%|██████▌   | 4238/6470 [53:08<30:17,  1.23it/s]
2|4239|Loss: 0.9163447022438049:  66%|██████▌   | 4239/6470 [53:08<29:04,  1.28it/s]
2|4240|Loss: 0.7631383538246155:  66%|██████▌   | 4239/6470 [53:08<29:04,  1.28it/s]
2|4240|Loss: 0.7631383538246155:  66%|██████▌   | 4240/6470 [53:09<28:16,  1.31it/s]
2|4241|Loss: 0.8521993160247803:  66%|██████▌   | 4240/6470 [53:09<28:16,  1.31it/s]
2|4241|Loss: 0.8521993160247803:  66%|██████▌   | 4241/6470 [53:09<29:38,  1.25it/s]
2|4242|Loss: 0.7566059231758118:  66%|██████▌   | 4241/6470 [53:10<29:38,  1.25it/s]
2|4242|Loss: 0.7566059231758118:  66%|██████▌   | 4242/6470 [53:10<30:06,  1.23it/s]
2|4243|Loss: 0.8352237343788147:  66%|██████▌   | 4242/6470 [53:11<30:06,  1.23it/s]
2|4243|Loss: 0.8352237343788147:  66%|██████▌   | 4243/6470 [53:11<31:11,  1.19it/s]
2|4244|Loss: 0.9745795130729675:  66%|██████▌   | 4243/6470 [53:12<31:11,  1.19it/s]
2|4244|Loss: 0.9745795130729675:  66%|██████▌   | 4244/6470 [53:12<29:56,  1.24it/s]
2|4245|Loss: 0.7975733876228333:  66%|██████▌   | 4244/6470 [53:12<29:56,  1.24it/s]
2|4245|Loss: 0.7975733876228333:  66%|██████▌   | 4245/6470 [53:13<28:21,  1.31it/s]
2|4246|Loss: 0.6836512088775635:  66%|██████▌   | 4245/6470 [53:13<28:21,  1.31it/s]
2|4246|Loss: 0.6836512088775635:  66%|██████▌   | 4246/6470 [53:14<29:40,  1.25it/s]
2|4247|Loss: 0.9512964487075806:  66%|██████▌   | 4246/6470 [53:14<29:40,  1.25it/s]
2|4247|Loss: 0.9512964487075806:  66%|██████▌   | 4247/6470 [53:14<29:11,  1.27it/s]
2|4248|Loss: 0.8045760989189148:  66%|██████▌   | 4247/6470 [53:15<29:11,  1.27it/s]
2|4248|Loss: 0.8045760989189148:  66%|██████▌   | 4248/6470 [53:15<27:53,  1.33it/s]
2|4249|Loss: 0.7961844801902771:  66%|██████▌   | 4248/6470 [53:15<27:53,  1.33it/s]
2|4249|Loss: 0.7961844801902771:  66%|██████▌   | 4249/6470 [53:16<27:35,  1.34it/s]
2|4250|Loss: 0.8571856021881104:  66%|██████▌   | 4249/6470 [53:16<27:35,  1.34it/s]
2|4250|Loss: 0.8571856021881104:  66%|██████▌   | 4250/6470 [53:16<26:17,  1.41it/s]
2|4251|Loss: 0.784304678440094:  66%|██████▌   | 4250/6470 [53:17<26:17,  1.41it/s] 
2|4251|Loss: 0.784304678440094:  66%|██████▌   | 4251/6470 [53:17<26:51,  1.38it/s]
2|4252|Loss: 0.7476929426193237:  66%|██████▌   | 4251/6470 [53:18<26:51,  1.38it/s]
2|4252|Loss: 0.7476929426193237:  66%|██████▌   | 4252/6470 [53:18<28:15,  1.31it/s]
2|4253|Loss: 1.0520858764648438:  66%|██████▌   | 4252/6470 [53:18<28:15,  1.31it/s]
2|4253|Loss: 1.0520858764648438:  66%|██████▌   | 4253/6470 [53:19<28:24,  1.30it/s]
2|4254|Loss: 0.9042604565620422:  66%|██████▌   | 4253/6470 [53:19<28:24,  1.30it/s]
2|4254|Loss: 0.9042604565620422:  66%|██████▌   | 4254/6470 [53:19<28:12,  1.31it/s]
2|4255|Loss: 0.8486232757568359:  66%|██████▌   | 4254/6470 [53:20<28:12,  1.31it/s]
2|4255|Loss: 0.8486232757568359:  66%|██████▌   | 4255/6470 [53:20<28:22,  1.30it/s]
2|4256|Loss: 0.8280245661735535:  66%|██████▌   | 4255/6470 [53:21<28:22,  1.30it/s]
2|4256|Loss: 0.8280245661735535:  66%|██████▌   | 4256/6470 [53:21<27:52,  1.32it/s]
2|4257|Loss: 0.769083559513092:  66%|██████▌   | 4256/6470 [53:21<27:52,  1.32it/s] 
2|4257|Loss: 0.769083559513092:  66%|██████▌   | 4257/6470 [53:22<28:15,  1.31it/s]
2|4258|Loss: 0.7710147500038147:  66%|██████▌   | 4257/6470 [53:22<28:15,  1.31it/s]
2|4258|Loss: 0.7710147500038147:  66%|██████▌   | 4258/6470 [53:22<25:50,  1.43it/s]
2|4259|Loss: 0.7584446668624878:  66%|██████▌   | 4258/6470 [53:23<25:50,  1.43it/s]
2|4259|Loss: 0.7584446668624878:  66%|██████▌   | 4259/6470 [53:23<25:58,  1.42it/s]
2|4260|Loss: 0.9295658469200134:  66%|██████▌   | 4259/6470 [53:23<25:58,  1.42it/s]
2|4260|Loss: 0.9295658469200134:  66%|██████▌   | 4260/6470 [53:24<25:00,  1.47it/s]
2|4261|Loss: 0.7571300864219666:  66%|██████▌   | 4260/6470 [53:24<25:00,  1.47it/s]
2|4261|Loss: 0.7571300864219666:  66%|██████▌   | 4261/6470 [53:24<23:25,  1.57it/s]
2|4262|Loss: 0.7952081561088562:  66%|██████▌   | 4261/6470 [53:25<23:25,  1.57it/s]
2|4262|Loss: 0.7952081561088562:  66%|██████▌   | 4262/6470 [53:25<25:40,  1.43it/s]
2|4263|Loss: 0.8241783380508423:  66%|██████▌   | 4262/6470 [53:25<25:40,  1.43it/s]
2|4263|Loss: 0.8241783380508423:  66%|██████▌   | 4263/6470 [53:26<25:32,  1.44it/s]
2|4264|Loss: 0.8185882568359375:  66%|██████▌   | 4263/6470 [53:26<25:32,  1.44it/s]
2|4264|Loss: 0.8185882568359375:  66%|██████▌   | 4264/6470 [53:26<23:57,  1.53it/s]
2|4265|Loss: 0.9765422940254211:  66%|██████▌   | 4264/6470 [53:27<23:57,  1.53it/s]
2|4265|Loss: 0.9765422940254211:  66%|██████▌   | 4265/6470 [53:27<24:39,  1.49it/s]
2|4266|Loss: 0.7060176134109497:  66%|██████▌   | 4265/6470 [53:27<24:39,  1.49it/s]
2|4266|Loss: 0.7060176134109497:  66%|██████▌   | 4266/6470 [53:28<25:32,  1.44it/s]
2|4267|Loss: 0.8890711665153503:  66%|██████▌   | 4266/6470 [53:28<25:32,  1.44it/s]
2|4267|Loss: 0.8890711665153503:  66%|██████▌   | 4267/6470 [53:28<24:47,  1.48it/s]
2|4268|Loss: 0.9877660274505615:  66%|██████▌   | 4267/6470 [53:29<24:47,  1.48it/s]
2|4268|Loss: 0.9877660274505615:  66%|██████▌   | 4268/6470 [53:29<26:43,  1.37it/s]
2|4269|Loss: 0.9272118806838989:  66%|██████▌   | 4268/6470 [53:30<26:43,  1.37it/s]
2|4269|Loss: 0.9272118806838989:  66%|██████▌   | 4269/6470 [53:30<29:05,  1.26it/s]
2|4270|Loss: 0.8714677095413208:  66%|██████▌   | 4269/6470 [53:31<29:05,  1.26it/s]
2|4270|Loss: 0.8714677095413208:  66%|██████▌   | 4270/6470 [53:31<29:33,  1.24it/s]
2|4271|Loss: 0.788544237613678:  66%|██████▌   | 4270/6470 [53:31<29:33,  1.24it/s] 
2|4271|Loss: 0.788544237613678:  66%|██████▌   | 4271/6470 [53:32<30:28,  1.20it/s]
2|4272|Loss: 0.9154011011123657:  66%|██████▌   | 4271/6470 [53:32<30:28,  1.20it/s]
2|4272|Loss: 0.9154011011123657:  66%|██████▌   | 4272/6470 [53:33<29:12,  1.25it/s]
2|4273|Loss: 0.841432511806488:  66%|██████▌   | 4272/6470 [53:33<29:12,  1.25it/s] 
2|4273|Loss: 0.841432511806488:  66%|██████▌   | 4273/6470 [53:33<29:57,  1.22it/s]
2|4274|Loss: 0.7767716646194458:  66%|██████▌   | 4273/6470 [53:34<29:57,  1.22it/s]
2|4274|Loss: 0.7767716646194458:  66%|██████▌   | 4274/6470 [53:34<30:06,  1.22it/s]
2|4275|Loss: 0.7356796860694885:  66%|██████▌   | 4274/6470 [53:35<30:06,  1.22it/s]
2|4275|Loss: 0.7356796860694885:  66%|██████▌   | 4275/6470 [53:35<27:54,  1.31it/s]
2|4276|Loss: 0.8547070026397705:  66%|██████▌   | 4275/6470 [53:35<27:54,  1.31it/s]
2|4276|Loss: 0.8547070026397705:  66%|██████▌   | 4276/6470 [53:36<27:52,  1.31it/s]
2|4277|Loss: 0.8353639245033264:  66%|██████▌   | 4276/6470 [53:36<27:52,  1.31it/s]
2|4277|Loss: 0.8353639245033264:  66%|██████▌   | 4277/6470 [53:36<27:54,  1.31it/s]
2|4278|Loss: 1.0385686159133911:  66%|██████▌   | 4277/6470 [53:37<27:54,  1.31it/s]
2|4278|Loss: 1.0385686159133911:  66%|██████▌   | 4278/6470 [53:37<28:47,  1.27it/s]
2|4279|Loss: 0.7684879899024963:  66%|██████▌   | 4278/6470 [53:38<28:47,  1.27it/s]
2|4279|Loss: 0.7684879899024963:  66%|██████▌   | 4279/6470 [53:38<25:20,  1.44it/s]
2|4280|Loss: 0.7643405199050903:  66%|██████▌   | 4279/6470 [53:38<25:20,  1.44it/s]
2|4280|Loss: 0.7643405199050903:  66%|██████▌   | 4280/6470 [53:38<23:48,  1.53it/s]
2|4281|Loss: 0.9833634495735168:  66%|██████▌   | 4280/6470 [53:39<23:48,  1.53it/s]
2|4281|Loss: 0.9833634495735168:  66%|██████▌   | 4281/6470 [53:39<23:19,  1.56it/s]
2|4282|Loss: 0.8693806529045105:  66%|██████▌   | 4281/6470 [53:39<23:19,  1.56it/s]
2|4282|Loss: 0.8693806529045105:  66%|██████▌   | 4282/6470 [53:40<25:37,  1.42it/s]
2|4283|Loss: 0.8771202564239502:  66%|██████▌   | 4282/6470 [53:40<25:37,  1.42it/s]
2|4283|Loss: 0.8771202564239502:  66%|██████▌   | 4283/6470 [53:40<25:46,  1.41it/s]
2|4284|Loss: 0.6903531551361084:  66%|██████▌   | 4283/6470 [53:41<25:46,  1.41it/s]
2|4284|Loss: 0.6903531551361084:  66%|██████▌   | 4284/6470 [53:41<27:03,  1.35it/s]
2|4285|Loss: 0.7772752642631531:  66%|██████▌   | 4284/6470 [53:42<27:03,  1.35it/s]
2|4285|Loss: 0.7772752642631531:  66%|██████▌   | 4285/6470 [53:42<26:55,  1.35it/s]
2|4286|Loss: 0.8646539449691772:  66%|██████▌   | 4285/6470 [53:42<26:55,  1.35it/s]
2|4286|Loss: 0.8646539449691772:  66%|██████▌   | 4286/6470 [53:43<25:54,  1.40it/s]
2|4287|Loss: 0.977444589138031:  66%|██████▌   | 4286/6470 [53:43<25:54,  1.40it/s] 
2|4287|Loss: 0.977444589138031:  66%|██████▋   | 4287/6470 [53:44<26:53,  1.35it/s]
2|4288|Loss: 0.8622305989265442:  66%|██████▋   | 4287/6470 [53:44<26:53,  1.35it/s]
2|4288|Loss: 0.8622305989265442:  66%|██████▋   | 4288/6470 [53:44<28:45,  1.26it/s]
2|4289|Loss: 0.6971876621246338:  66%|██████▋   | 4288/6470 [53:45<28:45,  1.26it/s]
2|4289|Loss: 0.6971876621246338:  66%|██████▋   | 4289/6470 [53:45<28:15,  1.29it/s]
2|4290|Loss: 0.7241794466972351:  66%|██████▋   | 4289/6470 [53:46<28:15,  1.29it/s]
2|4290|Loss: 0.7241794466972351:  66%|██████▋   | 4290/6470 [53:46<29:41,  1.22it/s]
2|4291|Loss: 0.8789043426513672:  66%|██████▋   | 4290/6470 [53:47<29:41,  1.22it/s]
2|4291|Loss: 0.8789043426513672:  66%|██████▋   | 4291/6470 [53:47<31:02,  1.17it/s]
2|4292|Loss: 0.842035710811615:  66%|██████▋   | 4291/6470 [53:47<31:02,  1.17it/s] 
2|4292|Loss: 0.842035710811615:  66%|██████▋   | 4292/6470 [53:48<30:03,  1.21it/s]
2|4293|Loss: 0.7547537684440613:  66%|██████▋   | 4292/6470 [53:48<30:03,  1.21it/s]
2|4293|Loss: 0.7547537684440613:  66%|██████▋   | 4293/6470 [53:49<29:36,  1.23it/s]
2|4294|Loss: 0.8356891870498657:  66%|██████▋   | 4293/6470 [53:49<29:36,  1.23it/s]
2|4294|Loss: 0.8356891870498657:  66%|██████▋   | 4294/6470 [53:49<29:18,  1.24it/s]
2|4295|Loss: 0.9407797455787659:  66%|██████▋   | 4294/6470 [53:50<29:18,  1.24it/s]
2|4295|Loss: 0.9407797455787659:  66%|██████▋   | 4295/6470 [53:50<28:00,  1.29it/s]
2|4296|Loss: 0.8673298358917236:  66%|██████▋   | 4295/6470 [53:50<28:00,  1.29it/s]
2|4296|Loss: 0.8673298358917236:  66%|██████▋   | 4296/6470 [53:51<26:38,  1.36it/s]
2|4297|Loss: 0.9196361899375916:  66%|██████▋   | 4296/6470 [53:51<26:38,  1.36it/s]
2|4297|Loss: 0.9196361899375916:  66%|██████▋   | 4297/6470 [53:51<26:47,  1.35it/s]
2|4298|Loss: 0.7928162217140198:  66%|██████▋   | 4297/6470 [53:52<26:47,  1.35it/s]
2|4298|Loss: 0.7928162217140198:  66%|██████▋   | 4298/6470 [53:52<27:19,  1.32it/s]
2|4299|Loss: 0.9778838753700256:  66%|██████▋   | 4298/6470 [53:53<27:19,  1.32it/s]
2|4299|Loss: 0.9778838753700256:  66%|██████▋   | 4299/6470 [53:53<26:55,  1.34it/s]
2|4300|Loss: 0.6259332895278931:  66%|██████▋   | 4299/6470 [53:53<26:55,  1.34it/s]
2|4300|Loss: 0.6259332895278931:  66%|██████▋   | 4300/6470 [53:54<26:28,  1.37it/s]
2|4301|Loss: 0.7613673210144043:  66%|██████▋   | 4300/6470 [53:54<26:28,  1.37it/s]
2|4301|Loss: 0.7613673210144043:  66%|██████▋   | 4301/6470 [53:54<26:04,  1.39it/s]
2|4302|Loss: 0.7592034935951233:  66%|██████▋   | 4301/6470 [53:55<26:04,  1.39it/s]
2|4302|Loss: 0.7592034935951233:  66%|██████▋   | 4302/6470 [53:55<24:36,  1.47it/s]
2|4303|Loss: 0.8350529074668884:  66%|██████▋   | 4302/6470 [53:55<24:36,  1.47it/s]
2|4303|Loss: 0.8350529074668884:  67%|██████▋   | 4303/6470 [53:56<23:52,  1.51it/s]
2|4304|Loss: 0.7908725738525391:  67%|██████▋   | 4303/6470 [53:56<23:52,  1.51it/s]
2|4304|Loss: 0.7908725738525391:  67%|██████▋   | 4304/6470 [53:56<26:02,  1.39it/s]
2|4305|Loss: 0.7482032775878906:  67%|██████▋   | 4304/6470 [53:57<26:02,  1.39it/s]
2|4305|Loss: 0.7482032775878906:  67%|██████▋   | 4305/6470 [53:57<27:06,  1.33it/s]
2|4306|Loss: 0.7741474509239197:  67%|██████▋   | 4305/6470 [53:58<27:06,  1.33it/s]
2|4306|Loss: 0.7741474509239197:  67%|██████▋   | 4306/6470 [53:58<27:08,  1.33it/s]
2|4307|Loss: 0.786818265914917:  67%|██████▋   | 4306/6470 [53:58<27:08,  1.33it/s] 
2|4307|Loss: 0.786818265914917:  67%|██████▋   | 4307/6470 [53:59<28:09,  1.28it/s]
2|4308|Loss: 0.7944901585578918:  67%|██████▋   | 4307/6470 [53:59<28:09,  1.28it/s]
2|4308|Loss: 0.7944901585578918:  67%|██████▋   | 4308/6470 [54:00<26:57,  1.34it/s]
2|4309|Loss: 0.8967334628105164:  67%|██████▋   | 4308/6470 [54:00<26:57,  1.34it/s]
2|4309|Loss: 0.8967334628105164:  67%|██████▋   | 4309/6470 [54:00<26:09,  1.38it/s]
2|4310|Loss: 0.9284990429878235:  67%|██████▋   | 4309/6470 [54:01<26:09,  1.38it/s]
2|4310|Loss: 0.9284990429878235:  67%|██████▋   | 4310/6470 [54:01<26:39,  1.35it/s]
2|4311|Loss: 0.671421229839325:  67%|██████▋   | 4310/6470 [54:01<26:39,  1.35it/s] 
2|4311|Loss: 0.671421229839325:  67%|██████▋   | 4311/6470 [54:02<28:07,  1.28it/s]
2|4312|Loss: 0.8622137308120728:  67%|██████▋   | 4311/6470 [54:02<28:07,  1.28it/s]
2|4312|Loss: 0.8622137308120728:  67%|██████▋   | 4312/6470 [54:03<27:35,  1.30it/s]
2|4313|Loss: 0.8341787457466125:  67%|██████▋   | 4312/6470 [54:03<27:35,  1.30it/s]
2|4313|Loss: 0.8341787457466125:  67%|██████▋   | 4313/6470 [54:03<26:05,  1.38it/s]
2|4314|Loss: 0.8621250987052917:  67%|██████▋   | 4313/6470 [54:04<26:05,  1.38it/s]
2|4314|Loss: 0.8621250987052917:  67%|██████▋   | 4314/6470 [54:04<26:09,  1.37it/s]
2|4315|Loss: 0.8840215802192688:  67%|██████▋   | 4314/6470 [54:04<26:09,  1.37it/s]
2|4315|Loss: 0.8840215802192688:  67%|██████▋   | 4315/6470 [54:05<26:43,  1.34it/s]
2|4316|Loss: 0.8093217611312866:  67%|██████▋   | 4315/6470 [54:05<26:43,  1.34it/s]
2|4316|Loss: 0.8093217611312866:  67%|██████▋   | 4316/6470 [54:06<27:14,  1.32it/s]
2|4317|Loss: 0.6104596853256226:  67%|██████▋   | 4316/6470 [54:06<27:14,  1.32it/s]
2|4317|Loss: 0.6104596853256226:  67%|██████▋   | 4317/6470 [54:06<28:52,  1.24it/s]
2|4318|Loss: 1.0030579566955566:  67%|██████▋   | 4317/6470 [54:07<28:52,  1.24it/s]
2|4318|Loss: 1.0030579566955566:  67%|██████▋   | 4318/6470 [54:07<26:42,  1.34it/s]
2|4319|Loss: 0.8764556646347046:  67%|██████▋   | 4318/6470 [54:07<26:42,  1.34it/s]
2|4319|Loss: 0.8764556646347046:  67%|██████▋   | 4319/6470 [54:08<26:48,  1.34it/s]
2|4320|Loss: 0.8964859247207642:  67%|██████▋   | 4319/6470 [54:08<26:48,  1.34it/s]
2|4320|Loss: 0.8964859247207642:  67%|██████▋   | 4320/6470 [54:09<28:43,  1.25it/s]
2|4321|Loss: 0.9722033739089966:  67%|██████▋   | 4320/6470 [54:09<28:43,  1.25it/s]
2|4321|Loss: 0.9722033739089966:  67%|██████▋   | 4321/6470 [54:10<30:01,  1.19it/s]
2|4322|Loss: 0.8331640958786011:  67%|██████▋   | 4321/6470 [54:10<30:01,  1.19it/s]
2|4322|Loss: 0.8331640958786011:  67%|██████▋   | 4322/6470 [54:11<30:52,  1.16it/s]
2|4323|Loss: 0.8840075731277466:  67%|██████▋   | 4322/6470 [54:11<30:52,  1.16it/s]
2|4323|Loss: 0.8840075731277466:  67%|██████▋   | 4323/6470 [54:11<27:47,  1.29it/s]
2|4324|Loss: 0.9528745412826538:  67%|██████▋   | 4323/6470 [54:11<27:47,  1.29it/s]
2|4324|Loss: 0.9528745412826538:  67%|██████▋   | 4324/6470 [54:12<27:08,  1.32it/s]
2|4325|Loss: 0.9568865895271301:  67%|██████▋   | 4324/6470 [54:12<27:08,  1.32it/s]
2|4325|Loss: 0.9568865895271301:  67%|██████▋   | 4325/6470 [54:13<26:05,  1.37it/s]
2|4326|Loss: 0.805611252784729:  67%|██████▋   | 4325/6470 [54:13<26:05,  1.37it/s] 
2|4326|Loss: 0.805611252784729:  67%|██████▋   | 4326/6470 [54:13<25:16,  1.41it/s]
2|4327|Loss: 0.8563987016677856:  67%|██████▋   | 4326/6470 [54:14<25:16,  1.41it/s]
2|4327|Loss: 0.8563987016677856:  67%|██████▋   | 4327/6470 [54:14<24:11,  1.48it/s]
2|4328|Loss: 0.919366180896759:  67%|██████▋   | 4327/6470 [54:14<24:11,  1.48it/s] 
2|4328|Loss: 0.919366180896759:  67%|██████▋   | 4328/6470 [54:14<23:47,  1.50it/s]
2|4329|Loss: 0.6886278390884399:  67%|██████▋   | 4328/6470 [54:15<23:47,  1.50it/s]
2|4329|Loss: 0.6886278390884399:  67%|██████▋   | 4329/6470 [54:15<23:57,  1.49it/s]
2|4330|Loss: 0.7944210171699524:  67%|██████▋   | 4329/6470 [54:15<23:57,  1.49it/s]
2|4330|Loss: 0.7944210171699524:  67%|██████▋   | 4330/6470 [54:16<24:12,  1.47it/s]
2|4331|Loss: 0.8517142534255981:  67%|██████▋   | 4330/6470 [54:16<24:12,  1.47it/s]
2|4331|Loss: 0.8517142534255981:  67%|██████▋   | 4331/6470 [54:17<25:20,  1.41it/s]
2|4332|Loss: 0.8311296105384827:  67%|██████▋   | 4331/6470 [54:17<25:20,  1.41it/s]
2|4332|Loss: 0.8311296105384827:  67%|██████▋   | 4332/6470 [54:17<26:25,  1.35it/s]
2|4333|Loss: 0.8007616400718689:  67%|██████▋   | 4332/6470 [54:18<26:25,  1.35it/s]
2|4333|Loss: 0.8007616400718689:  67%|██████▋   | 4333/6470 [54:18<26:39,  1.34it/s]
2|4334|Loss: 0.8629932999610901:  67%|██████▋   | 4333/6470 [54:19<26:39,  1.34it/s]
2|4334|Loss: 0.8629932999610901:  67%|██████▋   | 4334/6470 [54:19<28:00,  1.27it/s]
2|4335|Loss: 0.8799431920051575:  67%|██████▋   | 4334/6470 [54:19<28:00,  1.27it/s]
2|4335|Loss: 0.8799431920051575:  67%|██████▋   | 4335/6470 [54:20<25:57,  1.37it/s]
2|4336|Loss: 0.9331422448158264:  67%|██████▋   | 4335/6470 [54:20<25:57,  1.37it/s]
2|4336|Loss: 0.9331422448158264:  67%|██████▋   | 4336/6470 [54:20<26:05,  1.36it/s]
2|4337|Loss: 0.7922164797782898:  67%|██████▋   | 4336/6470 [54:21<26:05,  1.36it/s]
2|4337|Loss: 0.7922164797782898:  67%|██████▋   | 4337/6470 [54:21<25:15,  1.41it/s]
2|4338|Loss: 0.758002758026123:  67%|██████▋   | 4337/6470 [54:21<25:15,  1.41it/s] 
2|4338|Loss: 0.758002758026123:  67%|██████▋   | 4338/6470 [54:22<24:58,  1.42it/s]
2|4339|Loss: 0.8353103399276733:  67%|██████▋   | 4338/6470 [54:22<24:58,  1.42it/s]
2|4339|Loss: 0.8353103399276733:  67%|██████▋   | 4339/6470 [54:22<24:13,  1.47it/s]
2|4340|Loss: 0.948937177658081:  67%|██████▋   | 4339/6470 [54:23<24:13,  1.47it/s] 
2|4340|Loss: 0.948937177658081:  67%|██████▋   | 4340/6470 [54:23<24:42,  1.44it/s]
2|4341|Loss: 0.8212586045265198:  67%|██████▋   | 4340/6470 [54:23<24:42,  1.44it/s]
2|4341|Loss: 0.8212586045265198:  67%|██████▋   | 4341/6470 [54:24<23:00,  1.54it/s]
2|4342|Loss: 0.773871898651123:  67%|██████▋   | 4341/6470 [54:24<23:00,  1.54it/s] 
2|4342|Loss: 0.773871898651123:  67%|██████▋   | 4342/6470 [54:24<23:19,  1.52it/s]
2|4343|Loss: 0.8699942231178284:  67%|██████▋   | 4342/6470 [54:25<23:19,  1.52it/s]
2|4343|Loss: 0.8699942231178284:  67%|██████▋   | 4343/6470 [54:25<24:39,  1.44it/s]
2|4344|Loss: 0.9563385248184204:  67%|██████▋   | 4343/6470 [54:26<24:39,  1.44it/s]
2|4344|Loss: 0.9563385248184204:  67%|██████▋   | 4344/6470 [54:26<25:55,  1.37it/s]
2|4345|Loss: 0.7105329036712646:  67%|██████▋   | 4344/6470 [54:26<25:55,  1.37it/s]
2|4345|Loss: 0.7105329036712646:  67%|██████▋   | 4345/6470 [54:27<27:34,  1.28it/s]
2|4346|Loss: 0.8835863471031189:  67%|██████▋   | 4345/6470 [54:27<27:34,  1.28it/s]
2|4346|Loss: 0.8835863471031189:  67%|██████▋   | 4346/6470 [54:27<26:49,  1.32it/s]
2|4347|Loss: 0.8948466777801514:  67%|██████▋   | 4346/6470 [54:28<26:49,  1.32it/s]
2|4347|Loss: 0.8948466777801514:  67%|██████▋   | 4347/6470 [54:28<26:01,  1.36it/s]
2|4348|Loss: 0.7913926839828491:  67%|██████▋   | 4347/6470 [54:29<26:01,  1.36it/s]
2|4348|Loss: 0.7913926839828491:  67%|██████▋   | 4348/6470 [54:29<25:39,  1.38it/s]
2|4349|Loss: 0.8275049328804016:  67%|██████▋   | 4348/6470 [54:29<25:39,  1.38it/s]
2|4349|Loss: 0.8275049328804016:  67%|██████▋   | 4349/6470 [54:30<25:21,  1.39it/s]
2|4350|Loss: 0.8614190220832825:  67%|██████▋   | 4349/6470 [54:30<25:21,  1.39it/s]
2|4350|Loss: 0.8614190220832825:  67%|██████▋   | 4350/6470 [54:30<24:58,  1.41it/s]
2|4351|Loss: 0.9344537258148193:  67%|██████▋   | 4350/6470 [54:31<24:58,  1.41it/s]
2|4351|Loss: 0.9344537258148193:  67%|██████▋   | 4351/6470 [54:31<24:47,  1.42it/s]
2|4352|Loss: 0.8225810527801514:  67%|██████▋   | 4351/6470 [54:31<24:47,  1.42it/s]
2|4352|Loss: 0.8225810527801514:  67%|██████▋   | 4352/6470 [54:32<24:32,  1.44it/s]
2|4353|Loss: 0.9327624440193176:  67%|██████▋   | 4352/6470 [54:32<24:32,  1.44it/s]
2|4353|Loss: 0.9327624440193176:  67%|██████▋   | 4353/6470 [54:32<25:02,  1.41it/s]
2|4354|Loss: 0.7902476191520691:  67%|██████▋   | 4353/6470 [54:33<25:02,  1.41it/s]
2|4354|Loss: 0.7902476191520691:  67%|██████▋   | 4354/6470 [54:33<22:33,  1.56it/s]
2|4355|Loss: 0.6584933400154114:  67%|██████▋   | 4354/6470 [54:33<22:33,  1.56it/s]
2|4355|Loss: 0.6584933400154114:  67%|██████▋   | 4355/6470 [54:33<20:46,  1.70it/s]
2|4356|Loss: 0.9647684693336487:  67%|██████▋   | 4355/6470 [54:34<20:46,  1.70it/s]
2|4356|Loss: 0.9647684693336487:  67%|██████▋   | 4356/6470 [54:34<23:11,  1.52it/s]
2|4357|Loss: 0.879219114780426:  67%|██████▋   | 4356/6470 [54:35<23:11,  1.52it/s] 
2|4357|Loss: 0.879219114780426:  67%|██████▋   | 4357/6470 [54:35<23:19,  1.51it/s]
2|4358|Loss: 0.937728226184845:  67%|██████▋   | 4357/6470 [54:35<23:19,  1.51it/s]
2|4358|Loss: 0.937728226184845:  67%|██████▋   | 4358/6470 [54:35<22:48,  1.54it/s]
2|4359|Loss: 0.8212520480155945:  67%|██████▋   | 4358/6470 [54:36<22:48,  1.54it/s]
2|4359|Loss: 0.8212520480155945:  67%|██████▋   | 4359/6470 [54:36<24:43,  1.42it/s]
2|4360|Loss: 0.849155604839325:  67%|██████▋   | 4359/6470 [54:37<24:43,  1.42it/s] 
2|4360|Loss: 0.849155604839325:  67%|██████▋   | 4360/6470 [54:37<24:26,  1.44it/s]
2|4361|Loss: 0.8004248142242432:  67%|██████▋   | 4360/6470 [54:37<24:26,  1.44it/s]
2|4361|Loss: 0.8004248142242432:  67%|██████▋   | 4361/6470 [54:38<25:06,  1.40it/s]
2|4362|Loss: 0.6591127514839172:  67%|██████▋   | 4361/6470 [54:38<25:06,  1.40it/s]
2|4362|Loss: 0.6591127514839172:  67%|██████▋   | 4362/6470 [54:38<25:53,  1.36it/s]
2|4363|Loss: 0.8522306680679321:  67%|██████▋   | 4362/6470 [54:39<25:53,  1.36it/s]
2|4363|Loss: 0.8522306680679321:  67%|██████▋   | 4363/6470 [54:39<25:28,  1.38it/s]
2|4364|Loss: 0.8580014705657959:  67%|██████▋   | 4363/6470 [54:40<25:28,  1.38it/s]
2|4364|Loss: 0.8580014705657959:  67%|██████▋   | 4364/6470 [54:40<25:51,  1.36it/s]
2|4365|Loss: 0.7881810069084167:  67%|██████▋   | 4364/6470 [54:40<25:51,  1.36it/s]
2|4365|Loss: 0.7881810069084167:  67%|██████▋   | 4365/6470 [54:41<24:05,  1.46it/s]
2|4366|Loss: 0.8784098029136658:  67%|██████▋   | 4365/6470 [54:41<24:05,  1.46it/s]
2|4366|Loss: 0.8784098029136658:  67%|██████▋   | 4366/6470 [54:41<23:41,  1.48it/s]
2|4367|Loss: 0.9517614841461182:  67%|██████▋   | 4366/6470 [54:42<23:41,  1.48it/s]
2|4367|Loss: 0.9517614841461182:  67%|██████▋   | 4367/6470 [54:42<23:50,  1.47it/s]
2|4368|Loss: 0.868290901184082:  67%|██████▋   | 4367/6470 [54:42<23:50,  1.47it/s] 
2|4368|Loss: 0.868290901184082:  68%|██████▊   | 4368/6470 [54:43<24:53,  1.41it/s]
2|4369|Loss: 0.6798211932182312:  68%|██████▊   | 4368/6470 [54:43<24:53,  1.41it/s]
2|4369|Loss: 0.6798211932182312:  68%|██████▊   | 4369/6470 [54:43<25:32,  1.37it/s]
2|4370|Loss: 0.6867944002151489:  68%|██████▊   | 4369/6470 [54:44<25:32,  1.37it/s]
2|4370|Loss: 0.6867944002151489:  68%|██████▊   | 4370/6470 [54:44<24:57,  1.40it/s]
2|4371|Loss: 0.8976927399635315:  68%|██████▊   | 4370/6470 [54:45<24:57,  1.40it/s]
2|4371|Loss: 0.8976927399635315:  68%|██████▊   | 4371/6470 [54:45<26:43,  1.31it/s]
2|4372|Loss: 0.74934321641922:  68%|██████▊   | 4371/6470 [54:45<26:43,  1.31it/s]  
2|4372|Loss: 0.74934321641922:  68%|██████▊   | 4372/6470 [54:46<26:59,  1.30it/s]
2|4373|Loss: 0.9663508534431458:  68%|██████▊   | 4372/6470 [54:46<26:59,  1.30it/s]
2|4373|Loss: 0.9663508534431458:  68%|██████▊   | 4373/6470 [54:46<25:26,  1.37it/s]
2|4374|Loss: 0.9133802652359009:  68%|██████▊   | 4373/6470 [54:47<25:26,  1.37it/s]
2|4374|Loss: 0.9133802652359009:  68%|██████▊   | 4374/6470 [54:47<25:31,  1.37it/s]
2|4375|Loss: 0.6694638729095459:  68%|██████▊   | 4374/6470 [54:48<25:31,  1.37it/s]
2|4375|Loss: 0.6694638729095459:  68%|██████▊   | 4375/6470 [54:48<26:36,  1.31it/s]
2|4376|Loss: 0.9951881170272827:  68%|██████▊   | 4375/6470 [54:48<26:36,  1.31it/s]
2|4376|Loss: 0.9951881170272827:  68%|██████▊   | 4376/6470 [54:49<26:16,  1.33it/s]
2|4377|Loss: 0.8704098463058472:  68%|██████▊   | 4376/6470 [54:49<26:16,  1.33it/s]
2|4377|Loss: 0.8704098463058472:  68%|██████▊   | 4377/6470 [54:49<25:10,  1.39it/s]
2|4378|Loss: 0.8121044039726257:  68%|██████▊   | 4377/6470 [54:50<25:10,  1.39it/s]
2|4378|Loss: 0.8121044039726257:  68%|██████▊   | 4378/6470 [54:50<23:58,  1.45it/s]
2|4379|Loss: 0.8286194801330566:  68%|██████▊   | 4378/6470 [54:50<23:58,  1.45it/s]
2|4379|Loss: 0.8286194801330566:  68%|██████▊   | 4379/6470 [54:51<23:53,  1.46it/s]
2|4380|Loss: 0.8527547121047974:  68%|██████▊   | 4379/6470 [54:51<23:53,  1.46it/s]
2|4380|Loss: 0.8527547121047974:  68%|██████▊   | 4380/6470 [54:51<24:33,  1.42it/s]
2|4381|Loss: 0.7933775186538696:  68%|██████▊   | 4380/6470 [54:52<24:33,  1.42it/s]
2|4381|Loss: 0.7933775186538696:  68%|██████▊   | 4381/6470 [54:52<24:23,  1.43it/s]
2|4382|Loss: 0.8584408164024353:  68%|██████▊   | 4381/6470 [54:52<24:23,  1.43it/s]
2|4382|Loss: 0.8584408164024353:  68%|██████▊   | 4382/6470 [54:53<25:01,  1.39it/s]
2|4383|Loss: 0.7231085300445557:  68%|██████▊   | 4382/6470 [54:53<25:01,  1.39it/s]
2|4383|Loss: 0.7231085300445557:  68%|██████▊   | 4383/6470 [54:54<25:50,  1.35it/s]
2|4384|Loss: 0.8984994292259216:  68%|██████▊   | 4383/6470 [54:54<25:50,  1.35it/s]
2|4384|Loss: 0.8984994292259216:  68%|██████▊   | 4384/6470 [54:54<25:27,  1.37it/s]
2|4385|Loss: 1.027797818183899:  68%|██████▊   | 4384/6470 [54:55<25:27,  1.37it/s] 
2|4385|Loss: 1.027797818183899:  68%|██████▊   | 4385/6470 [54:55<26:01,  1.34it/s]
2|4386|Loss: 0.8380653262138367:  68%|██████▊   | 4385/6470 [54:56<26:01,  1.34it/s]
2|4386|Loss: 0.8380653262138367:  68%|██████▊   | 4386/6470 [54:56<26:39,  1.30it/s]
2|4387|Loss: 0.9026361107826233:  68%|██████▊   | 4386/6470 [54:56<26:39,  1.30it/s]
2|4387|Loss: 0.9026361107826233:  68%|██████▊   | 4387/6470 [54:57<27:51,  1.25it/s]
2|4388|Loss: 0.7475913166999817:  68%|██████▊   | 4387/6470 [54:57<27:51,  1.25it/s]
2|4388|Loss: 0.7475913166999817:  68%|██████▊   | 4388/6470 [54:58<29:37,  1.17it/s]
2|4389|Loss: 0.7823920845985413:  68%|██████▊   | 4388/6470 [54:58<29:37,  1.17it/s]
2|4389|Loss: 0.7823920845985413:  68%|██████▊   | 4389/6470 [54:59<29:27,  1.18it/s]
2|4390|Loss: 0.7579886317253113:  68%|██████▊   | 4389/6470 [54:59<29:27,  1.18it/s]
2|4390|Loss: 0.7579886317253113:  68%|██████▊   | 4390/6470 [54:59<28:51,  1.20it/s]
2|4391|Loss: 0.8221602439880371:  68%|██████▊   | 4390/6470 [55:00<28:51,  1.20it/s]
2|4391|Loss: 0.8221602439880371:  68%|██████▊   | 4391/6470 [55:00<28:21,  1.22it/s]
2|4392|Loss: 0.7503182291984558:  68%|██████▊   | 4391/6470 [55:01<28:21,  1.22it/s]
2|4392|Loss: 0.7503182291984558:  68%|██████▊   | 4392/6470 [55:01<27:01,  1.28it/s]
2|4393|Loss: 0.9025790691375732:  68%|██████▊   | 4392/6470 [55:01<27:01,  1.28it/s]
2|4393|Loss: 0.9025790691375732:  68%|██████▊   | 4393/6470 [55:02<25:28,  1.36it/s]
2|4394|Loss: 0.7956929206848145:  68%|██████▊   | 4393/6470 [55:02<25:28,  1.36it/s]
2|4394|Loss: 0.7956929206848145:  68%|██████▊   | 4394/6470 [55:02<23:49,  1.45it/s]
2|4395|Loss: 0.9610148668289185:  68%|██████▊   | 4394/6470 [55:02<23:49,  1.45it/s]
2|4395|Loss: 0.9610148668289185:  68%|██████▊   | 4395/6470 [55:03<21:49,  1.59it/s]
2|4396|Loss: 0.7052100300788879:  68%|██████▊   | 4395/6470 [55:03<21:49,  1.59it/s]
2|4396|Loss: 0.7052100300788879:  68%|██████▊   | 4396/6470 [55:03<23:28,  1.47it/s]
2|4397|Loss: 0.9770025014877319:  68%|██████▊   | 4396/6470 [55:04<23:28,  1.47it/s]
2|4397|Loss: 0.9770025014877319:  68%|██████▊   | 4397/6470 [55:04<24:46,  1.39it/s]
2|4398|Loss: 0.720214307308197:  68%|██████▊   | 4397/6470 [55:05<24:46,  1.39it/s] 
2|4398|Loss: 0.720214307308197:  68%|██████▊   | 4398/6470 [55:05<26:09,  1.32it/s]
2|4399|Loss: 0.6146090626716614:  68%|██████▊   | 4398/6470 [55:05<26:09,  1.32it/s]
2|4399|Loss: 0.6146090626716614:  68%|██████▊   | 4399/6470 [55:06<23:26,  1.47it/s]
2|4400|Loss: 0.8191712498664856:  68%|██████▊   | 4399/6470 [55:06<23:26,  1.47it/s]
2|4400|Loss: 0.8191712498664856:  68%|██████▊   | 4400/6470 [55:06<22:49,  1.51it/s]
2|4401|Loss: 0.8745728731155396:  68%|██████▊   | 4400/6470 [55:07<22:49,  1.51it/s]
2|4401|Loss: 0.8745728731155396:  68%|██████▊   | 4401/6470 [55:07<22:24,  1.54it/s]
2|4402|Loss: 0.9304782748222351:  68%|██████▊   | 4401/6470 [55:07<22:24,  1.54it/s]
2|4402|Loss: 0.9304782748222351:  68%|██████▊   | 4402/6470 [55:08<24:32,  1.40it/s]
2|4403|Loss: 0.9560116529464722:  68%|██████▊   | 4402/6470 [55:08<24:32,  1.40it/s]
2|4403|Loss: 0.9560116529464722:  68%|██████▊   | 4403/6470 [55:08<24:52,  1.39it/s]
2|4404|Loss: 0.8663286566734314:  68%|██████▊   | 4403/6470 [55:09<24:52,  1.39it/s]
2|4404|Loss: 0.8663286566734314:  68%|██████▊   | 4404/6470 [55:09<24:24,  1.41it/s]
2|4405|Loss: 0.6948018074035645:  68%|██████▊   | 4404/6470 [55:10<24:24,  1.41it/s]
2|4405|Loss: 0.6948018074035645:  68%|██████▊   | 4405/6470 [55:10<27:08,  1.27it/s]
2|4406|Loss: 0.7922928929328918:  68%|██████▊   | 4405/6470 [55:11<27:08,  1.27it/s]
2|4406|Loss: 0.7922928929328918:  68%|██████▊   | 4406/6470 [55:11<27:26,  1.25it/s]
2|4407|Loss: 0.8586066961288452:  68%|██████▊   | 4406/6470 [55:11<27:26,  1.25it/s]
2|4407|Loss: 0.8586066961288452:  68%|██████▊   | 4407/6470 [55:11<25:16,  1.36it/s]
2|4408|Loss: 1.0189687013626099:  68%|██████▊   | 4407/6470 [55:12<25:16,  1.36it/s]
2|4408|Loss: 1.0189687013626099:  68%|██████▊   | 4408/6470 [55:12<25:17,  1.36it/s]
2|4409|Loss: 0.8755167126655579:  68%|██████▊   | 4408/6470 [55:13<25:17,  1.36it/s]
2|4409|Loss: 0.8755167126655579:  68%|██████▊   | 4409/6470 [55:13<24:34,  1.40it/s]
2|4410|Loss: 0.8771998286247253:  68%|██████▊   | 4409/6470 [55:13<24:34,  1.40it/s]
2|4410|Loss: 0.8771998286247253:  68%|██████▊   | 4410/6470 [55:14<26:17,  1.31it/s]
2|4411|Loss: 0.8821234703063965:  68%|██████▊   | 4410/6470 [55:14<26:17,  1.31it/s]
2|4411|Loss: 0.8821234703063965:  68%|██████▊   | 4411/6470 [55:15<26:56,  1.27it/s]
2|4412|Loss: 0.7094205617904663:  68%|██████▊   | 4411/6470 [55:15<26:56,  1.27it/s]
2|4412|Loss: 0.7094205617904663:  68%|██████▊   | 4412/6470 [55:15<26:44,  1.28it/s]
2|4413|Loss: 0.8933200240135193:  68%|██████▊   | 4412/6470 [55:16<26:44,  1.28it/s]
2|4413|Loss: 0.8933200240135193:  68%|██████▊   | 4413/6470 [55:16<26:50,  1.28it/s]
2|4414|Loss: 1.0074114799499512:  68%|██████▊   | 4413/6470 [55:17<26:50,  1.28it/s]
2|4414|Loss: 1.0074114799499512:  68%|██████▊   | 4414/6470 [55:17<25:53,  1.32it/s]
2|4415|Loss: 0.8785912990570068:  68%|██████▊   | 4414/6470 [55:17<25:53,  1.32it/s]
2|4415|Loss: 0.8785912990570068:  68%|██████▊   | 4415/6470 [55:18<27:07,  1.26it/s]
2|4416|Loss: 0.7527089715003967:  68%|██████▊   | 4415/6470 [55:18<27:07,  1.26it/s]
2|4416|Loss: 0.7527089715003967:  68%|██████▊   | 4416/6470 [55:18<26:04,  1.31it/s]
2|4417|Loss: 0.781762957572937:  68%|██████▊   | 4416/6470 [55:19<26:04,  1.31it/s] 
2|4417|Loss: 0.781762957572937:  68%|██████▊   | 4417/6470 [55:19<25:57,  1.32it/s]
2|4418|Loss: 0.8922030925750732:  68%|██████▊   | 4417/6470 [55:20<25:57,  1.32it/s]
2|4418|Loss: 0.8922030925750732:  68%|██████▊   | 4418/6470 [55:20<24:50,  1.38it/s]
2|4419|Loss: 0.7716426849365234:  68%|██████▊   | 4418/6470 [55:20<24:50,  1.38it/s]
2|4419|Loss: 0.7716426849365234:  68%|██████▊   | 4419/6470 [55:21<25:09,  1.36it/s]
2|4420|Loss: 0.840771496295929:  68%|██████▊   | 4419/6470 [55:21<25:09,  1.36it/s] 
2|4420|Loss: 0.840771496295929:  68%|██████▊   | 4420/6470 [55:21<25:52,  1.32it/s]
2|4421|Loss: 0.9720548391342163:  68%|██████▊   | 4420/6470 [55:22<25:52,  1.32it/s]
2|4421|Loss: 0.9720548391342163:  68%|██████▊   | 4421/6470 [55:22<26:56,  1.27it/s]
2|4422|Loss: 0.8100264072418213:  68%|██████▊   | 4421/6470 [55:23<26:56,  1.27it/s]
2|4422|Loss: 0.8100264072418213:  68%|██████▊   | 4422/6470 [55:23<25:43,  1.33it/s]
2|4423|Loss: 0.8669426441192627:  68%|██████▊   | 4422/6470 [55:23<25:43,  1.33it/s]
2|4423|Loss: 0.8669426441192627:  68%|██████▊   | 4423/6470 [55:24<26:50,  1.27it/s]
2|4424|Loss: 0.9318126440048218:  68%|██████▊   | 4423/6470 [55:24<26:50,  1.27it/s]
2|4424|Loss: 0.9318126440048218:  68%|██████▊   | 4424/6470 [55:24<25:48,  1.32it/s]
2|4425|Loss: 0.7222651243209839:  68%|██████▊   | 4424/6470 [55:25<25:48,  1.32it/s]
2|4425|Loss: 0.7222651243209839:  68%|██████▊   | 4425/6470 [55:25<24:49,  1.37it/s]
2|4426|Loss: 0.8746830224990845:  68%|██████▊   | 4425/6470 [55:25<24:49,  1.37it/s]
2|4426|Loss: 0.8746830224990845:  68%|██████▊   | 4426/6470 [55:26<24:31,  1.39it/s]
2|4427|Loss: 0.8056563138961792:  68%|██████▊   | 4426/6470 [55:26<24:31,  1.39it/s]
2|4427|Loss: 0.8056563138961792:  68%|██████▊   | 4427/6470 [55:26<23:58,  1.42it/s]
2|4428|Loss: 0.904665470123291:  68%|██████▊   | 4427/6470 [55:27<23:58,  1.42it/s] 
2|4428|Loss: 0.904665470123291:  68%|██████▊   | 4428/6470 [55:27<25:59,  1.31it/s]
2|4429|Loss: 0.9035272002220154:  68%|██████▊   | 4428/6470 [55:28<25:59,  1.31it/s]
2|4429|Loss: 0.9035272002220154:  68%|██████▊   | 4429/6470 [55:28<26:43,  1.27it/s]
2|4430|Loss: 0.9612151384353638:  68%|██████▊   | 4429/6470 [55:29<26:43,  1.27it/s]
2|4430|Loss: 0.9612151384353638:  68%|██████▊   | 4430/6470 [55:29<26:51,  1.27it/s]
2|4431|Loss: 0.8684207797050476:  68%|██████▊   | 4430/6470 [55:29<26:51,  1.27it/s]
2|4431|Loss: 0.8684207797050476:  68%|██████▊   | 4431/6470 [55:30<25:44,  1.32it/s]
2|4432|Loss: 0.9593890309333801:  68%|██████▊   | 4431/6470 [55:30<25:44,  1.32it/s]
2|4432|Loss: 0.9593890309333801:  69%|██████▊   | 4432/6470 [55:30<23:55,  1.42it/s]
2|4433|Loss: 0.8899685144424438:  69%|██████▊   | 4432/6470 [55:31<23:55,  1.42it/s]
2|4433|Loss: 0.8899685144424438:  69%|██████▊   | 4433/6470 [55:31<23:13,  1.46it/s]
2|4434|Loss: 0.8279897570610046:  69%|██████▊   | 4433/6470 [55:31<23:13,  1.46it/s]
2|4434|Loss: 0.8279897570610046:  69%|██████▊   | 4434/6470 [55:32<23:28,  1.45it/s]
2|4435|Loss: 0.8403193950653076:  69%|██████▊   | 4434/6470 [55:32<23:28,  1.45it/s]
2|4435|Loss: 0.8403193950653076:  69%|██████▊   | 4435/6470 [55:32<25:09,  1.35it/s]
2|4436|Loss: 0.9142128825187683:  69%|██████▊   | 4435/6470 [55:33<25:09,  1.35it/s]
2|4436|Loss: 0.9142128825187683:  69%|██████▊   | 4436/6470 [55:33<25:31,  1.33it/s]
2|4437|Loss: 0.7517215013504028:  69%|██████▊   | 4436/6470 [55:34<25:31,  1.33it/s]
2|4437|Loss: 0.7517215013504028:  69%|██████▊   | 4437/6470 [55:34<26:46,  1.27it/s]
2|4438|Loss: 0.8040331602096558:  69%|██████▊   | 4437/6470 [55:35<26:46,  1.27it/s]
2|4438|Loss: 0.8040331602096558:  69%|██████▊   | 4438/6470 [55:35<27:08,  1.25it/s]
2|4439|Loss: 0.738461971282959:  69%|██████▊   | 4438/6470 [55:35<27:08,  1.25it/s] 
2|4439|Loss: 0.738461971282959:  69%|██████▊   | 4439/6470 [55:36<26:06,  1.30it/s]
2|4440|Loss: 0.7662114500999451:  69%|██████▊   | 4439/6470 [55:36<26:06,  1.30it/s]
2|4440|Loss: 0.7662114500999451:  69%|██████▊   | 4440/6470 [55:36<24:36,  1.37it/s]
2|4441|Loss: 1.03653883934021:  69%|██████▊   | 4440/6470 [55:37<24:36,  1.37it/s]  
2|4441|Loss: 1.03653883934021:  69%|██████▊   | 4441/6470 [55:37<26:39,  1.27it/s]
2|4442|Loss: 0.8623067140579224:  69%|██████▊   | 4441/6470 [55:38<26:39,  1.27it/s]
2|4442|Loss: 0.8623067140579224:  69%|██████▊   | 4442/6470 [55:38<27:20,  1.24it/s]
2|4443|Loss: 0.8252840638160706:  69%|██████▊   | 4442/6470 [55:38<27:20,  1.24it/s]
2|4443|Loss: 0.8252840638160706:  69%|██████▊   | 4443/6470 [55:39<26:34,  1.27it/s]
2|4444|Loss: 0.8813508749008179:  69%|██████▊   | 4443/6470 [55:39<26:34,  1.27it/s]
2|4444|Loss: 0.8813508749008179:  69%|██████▊   | 4444/6470 [55:40<27:28,  1.23it/s]
2|4445|Loss: 0.7417720556259155:  69%|██████▊   | 4444/6470 [55:40<27:28,  1.23it/s]
2|4445|Loss: 0.7417720556259155:  69%|██████▊   | 4445/6470 [55:40<24:24,  1.38it/s]
2|4446|Loss: 0.8774869441986084:  69%|██████▊   | 4445/6470 [55:41<24:24,  1.38it/s]
2|4446|Loss: 0.8774869441986084:  69%|██████▊   | 4446/6470 [55:41<23:21,  1.44it/s]
2|4447|Loss: 0.904970109462738:  69%|██████▊   | 4446/6470 [55:41<23:21,  1.44it/s] 
2|4447|Loss: 0.904970109462738:  69%|██████▊   | 4447/6470 [55:42<24:13,  1.39it/s]
2|4448|Loss: 0.6809433102607727:  69%|██████▊   | 4447/6470 [55:42<24:13,  1.39it/s]
2|4448|Loss: 0.6809433102607727:  69%|██████▊   | 4448/6470 [55:42<24:46,  1.36it/s]
2|4449|Loss: 0.8512540459632874:  69%|██████▊   | 4448/6470 [55:43<24:46,  1.36it/s]
2|4449|Loss: 0.8512540459632874:  69%|██████▉   | 4449/6470 [55:43<25:21,  1.33it/s]
2|4450|Loss: 0.9370573163032532:  69%|██████▉   | 4449/6470 [55:44<25:21,  1.33it/s]
2|4450|Loss: 0.9370573163032532:  69%|██████▉   | 4450/6470 [55:44<27:00,  1.25it/s]
2|4451|Loss: 0.9081857800483704:  69%|██████▉   | 4450/6470 [55:44<27:00,  1.25it/s]
2|4451|Loss: 0.9081857800483704:  69%|██████▉   | 4451/6470 [55:45<25:30,  1.32it/s]
2|4452|Loss: 0.8811734914779663:  69%|██████▉   | 4451/6470 [55:45<25:30,  1.32it/s]
2|4452|Loss: 0.8811734914779663:  69%|██████▉   | 4452/6470 [55:45<24:20,  1.38it/s]
2|4453|Loss: 0.7719765305519104:  69%|██████▉   | 4452/6470 [55:46<24:20,  1.38it/s]
2|4453|Loss: 0.7719765305519104:  69%|██████▉   | 4453/6470 [55:46<22:15,  1.51it/s]
2|4454|Loss: 0.7997638583183289:  69%|██████▉   | 4453/6470 [55:46<22:15,  1.51it/s]
2|4454|Loss: 0.7997638583183289:  69%|██████▉   | 4454/6470 [55:46<21:32,  1.56it/s]
2|4455|Loss: 0.6399266719818115:  69%|██████▉   | 4454/6470 [55:47<21:32,  1.56it/s]
2|4455|Loss: 0.6399266719818115:  69%|██████▉   | 4455/6470 [55:47<19:04,  1.76it/s]
2|4456|Loss: 0.9141036868095398:  69%|██████▉   | 4455/6470 [55:47<19:04,  1.76it/s]
2|4456|Loss: 0.9141036868095398:  69%|██████▉   | 4456/6470 [55:47<19:17,  1.74it/s]
2|4457|Loss: 0.9982547760009766:  69%|██████▉   | 4456/6470 [55:48<19:17,  1.74it/s]
2|4457|Loss: 0.9982547760009766:  69%|██████▉   | 4457/6470 [55:48<20:18,  1.65it/s]
2|4458|Loss: 0.7629667520523071:  69%|██████▉   | 4457/6470 [55:49<20:18,  1.65it/s]
2|4458|Loss: 0.7629667520523071:  69%|██████▉   | 4458/6470 [55:49<22:50,  1.47it/s]
2|4459|Loss: 0.794282853603363:  69%|██████▉   | 4458/6470 [55:49<22:50,  1.47it/s] 
2|4459|Loss: 0.794282853603363:  69%|██████▉   | 4459/6470 [55:50<24:06,  1.39it/s]
2|4460|Loss: 0.7757760882377625:  69%|██████▉   | 4459/6470 [55:50<24:06,  1.39it/s]
2|4460|Loss: 0.7757760882377625:  69%|██████▉   | 4460/6470 [55:51<26:32,  1.26it/s]
2|4461|Loss: 0.8222112655639648:  69%|██████▉   | 4460/6470 [55:51<26:32,  1.26it/s]
2|4461|Loss: 0.8222112655639648:  69%|██████▉   | 4461/6470 [55:51<25:06,  1.33it/s]
2|4462|Loss: 0.8395722508430481:  69%|██████▉   | 4461/6470 [55:52<25:06,  1.33it/s]
2|4462|Loss: 0.8395722508430481:  69%|██████▉   | 4462/6470 [55:52<23:20,  1.43it/s]
2|4463|Loss: 0.9480735063552856:  69%|██████▉   | 4462/6470 [55:52<23:20,  1.43it/s]
2|4463|Loss: 0.9480735063552856:  69%|██████▉   | 4463/6470 [55:53<25:24,  1.32it/s]
2|4464|Loss: 0.9153956770896912:  69%|██████▉   | 4463/6470 [55:53<25:24,  1.32it/s]
2|4464|Loss: 0.9153956770896912:  69%|██████▉   | 4464/6470 [55:54<24:22,  1.37it/s]
2|4465|Loss: 0.912811815738678:  69%|██████▉   | 4464/6470 [55:54<24:22,  1.37it/s] 
2|4465|Loss: 0.912811815738678:  69%|██████▉   | 4465/6470 [55:54<22:30,  1.49it/s]
2|4466|Loss: 0.7664995789527893:  69%|██████▉   | 4465/6470 [55:54<22:30,  1.49it/s]
2|4466|Loss: 0.7664995789527893:  69%|██████▉   | 4466/6470 [55:55<22:53,  1.46it/s]
2|4467|Loss: 0.9312071204185486:  69%|██████▉   | 4466/6470 [55:55<22:53,  1.46it/s]
2|4467|Loss: 0.9312071204185486:  69%|██████▉   | 4467/6470 [55:56<25:30,  1.31it/s]
2|4468|Loss: 0.8090920448303223:  69%|██████▉   | 4467/6470 [55:56<25:30,  1.31it/s]
2|4468|Loss: 0.8090920448303223:  69%|██████▉   | 4468/6470 [55:57<26:15,  1.27it/s]
2|4469|Loss: 0.7771748304367065:  69%|██████▉   | 4468/6470 [55:57<26:15,  1.27it/s]
2|4469|Loss: 0.7771748304367065:  69%|██████▉   | 4469/6470 [55:57<26:21,  1.27it/s]
2|4470|Loss: 0.7806704640388489:  69%|██████▉   | 4469/6470 [55:58<26:21,  1.27it/s]
2|4470|Loss: 0.7806704640388489:  69%|██████▉   | 4470/6470 [55:58<26:15,  1.27it/s]
2|4471|Loss: 0.8099198937416077:  69%|██████▉   | 4470/6470 [55:59<26:15,  1.27it/s]
2|4471|Loss: 0.8099198937416077:  69%|██████▉   | 4471/6470 [55:59<26:11,  1.27it/s]
2|4472|Loss: 0.6979550123214722:  69%|██████▉   | 4471/6470 [55:59<26:11,  1.27it/s]
2|4472|Loss: 0.6979550123214722:  69%|██████▉   | 4472/6470 [56:00<26:36,  1.25it/s]
2|4473|Loss: 0.7333446741104126:  69%|██████▉   | 4472/6470 [56:00<26:36,  1.25it/s]
2|4473|Loss: 0.7333446741104126:  69%|██████▉   | 4473/6470 [56:01<26:37,  1.25it/s]
2|4474|Loss: 0.7144467830657959:  69%|██████▉   | 4473/6470 [56:01<26:37,  1.25it/s]
2|4474|Loss: 0.7144467830657959:  69%|██████▉   | 4474/6470 [56:01<26:14,  1.27it/s]
2|4475|Loss: 0.7955314517021179:  69%|██████▉   | 4474/6470 [56:02<26:14,  1.27it/s]
2|4475|Loss: 0.7955314517021179:  69%|██████▉   | 4475/6470 [56:02<26:08,  1.27it/s]
2|4476|Loss: 0.808804988861084:  69%|██████▉   | 4475/6470 [56:03<26:08,  1.27it/s] 
2|4476|Loss: 0.808804988861084:  69%|██████▉   | 4476/6470 [56:03<26:08,  1.27it/s]
2|4477|Loss: 0.8527498245239258:  69%|██████▉   | 4476/6470 [56:03<26:08,  1.27it/s]
2|4477|Loss: 0.8527498245239258:  69%|██████▉   | 4477/6470 [56:04<26:56,  1.23it/s]
2|4478|Loss: 0.8139762878417969:  69%|██████▉   | 4477/6470 [56:04<26:56,  1.23it/s]
2|4478|Loss: 0.8139762878417969:  69%|██████▉   | 4478/6470 [56:05<28:12,  1.18it/s]
2|4479|Loss: 0.9903678297996521:  69%|██████▉   | 4478/6470 [56:05<28:12,  1.18it/s]
2|4479|Loss: 0.9903678297996521:  69%|██████▉   | 4479/6470 [56:06<28:05,  1.18it/s]
2|4480|Loss: 0.916916012763977:  69%|██████▉   | 4479/6470 [56:06<28:05,  1.18it/s] 
2|4480|Loss: 0.916916012763977:  69%|██████▉   | 4480/6470 [56:06<26:28,  1.25it/s]
2|4481|Loss: 0.7781218886375427:  69%|██████▉   | 4480/6470 [56:07<26:28,  1.25it/s]
2|4481|Loss: 0.7781218886375427:  69%|██████▉   | 4481/6470 [56:07<26:02,  1.27it/s]
2|4482|Loss: 0.6044955253601074:  69%|██████▉   | 4481/6470 [56:07<26:02,  1.27it/s]
2|4482|Loss: 0.6044955253601074:  69%|██████▉   | 4482/6470 [56:08<24:04,  1.38it/s]
2|4483|Loss: 1.008602499961853:  69%|██████▉   | 4482/6470 [56:08<24:04,  1.38it/s] 
2|4483|Loss: 1.008602499961853:  69%|██████▉   | 4483/6470 [56:08<23:04,  1.44it/s]
2|4484|Loss: 0.8109826445579529:  69%|██████▉   | 4483/6470 [56:09<23:04,  1.44it/s]
2|4484|Loss: 0.8109826445579529:  69%|██████▉   | 4484/6470 [56:09<25:30,  1.30it/s]
2|4485|Loss: 0.8146874904632568:  69%|██████▉   | 4484/6470 [56:10<25:30,  1.30it/s]
2|4485|Loss: 0.8146874904632568:  69%|██████▉   | 4485/6470 [56:10<26:46,  1.24it/s]
2|4486|Loss: 0.8080258965492249:  69%|██████▉   | 4485/6470 [56:11<26:46,  1.24it/s]
2|4486|Loss: 0.8080258965492249:  69%|██████▉   | 4486/6470 [56:11<26:22,  1.25it/s]
2|4487|Loss: 0.7568841576576233:  69%|██████▉   | 4486/6470 [56:11<26:22,  1.25it/s]
2|4487|Loss: 0.7568841576576233:  69%|██████▉   | 4487/6470 [56:11<23:37,  1.40it/s]
2|4488|Loss: 0.8170108199119568:  69%|██████▉   | 4487/6470 [56:12<23:37,  1.40it/s]
2|4488|Loss: 0.8170108199119568:  69%|██████▉   | 4488/6470 [56:12<23:19,  1.42it/s]
2|4489|Loss: 0.8150274157524109:  69%|██████▉   | 4488/6470 [56:13<23:19,  1.42it/s]
2|4489|Loss: 0.8150274157524109:  69%|██████▉   | 4489/6470 [56:13<25:26,  1.30it/s]
2|4490|Loss: 0.8810690641403198:  69%|██████▉   | 4489/6470 [56:13<25:26,  1.30it/s]
2|4490|Loss: 0.8810690641403198:  69%|██████▉   | 4490/6470 [56:14<25:57,  1.27it/s]
2|4491|Loss: 0.855078935623169:  69%|██████▉   | 4490/6470 [56:14<25:57,  1.27it/s] 
2|4491|Loss: 0.855078935623169:  69%|██████▉   | 4491/6470 [56:15<25:58,  1.27it/s]
2|4492|Loss: 0.8188389539718628:  69%|██████▉   | 4491/6470 [56:15<25:58,  1.27it/s]
2|4492|Loss: 0.8188389539718628:  69%|██████▉   | 4492/6470 [56:15<26:25,  1.25it/s]
2|4493|Loss: 0.9012575745582581:  69%|██████▉   | 4492/6470 [56:16<26:25,  1.25it/s]
2|4493|Loss: 0.9012575745582581:  69%|██████▉   | 4493/6470 [56:17<29:23,  1.12it/s]
2|4494|Loss: 0.9483177065849304:  69%|██████▉   | 4493/6470 [56:17<29:23,  1.12it/s]
2|4494|Loss: 0.9483177065849304:  69%|██████▉   | 4494/6470 [56:17<30:02,  1.10it/s]
2|4495|Loss: 0.946268618106842:  69%|██████▉   | 4494/6470 [56:18<30:02,  1.10it/s] 
2|4495|Loss: 0.946268618106842:  69%|██████▉   | 4495/6470 [56:18<29:08,  1.13it/s]
2|4496|Loss: 0.8974254727363586:  69%|██████▉   | 4495/6470 [56:19<29:08,  1.13it/s]
2|4496|Loss: 0.8974254727363586:  69%|██████▉   | 4496/6470 [56:19<29:20,  1.12it/s]
2|4497|Loss: 0.9326567053794861:  69%|██████▉   | 4496/6470 [56:20<29:20,  1.12it/s]
2|4497|Loss: 0.9326567053794861:  70%|██████▉   | 4497/6470 [56:20<27:37,  1.19it/s]
2|4498|Loss: 0.8866377472877502:  70%|██████▉   | 4497/6470 [56:20<27:37,  1.19it/s]
2|4498|Loss: 0.8866377472877502:  70%|██████▉   | 4498/6470 [56:21<24:54,  1.32it/s]
2|4499|Loss: 0.8520104289054871:  70%|██████▉   | 4498/6470 [56:21<24:54,  1.32it/s]
2|4499|Loss: 0.8520104289054871:  70%|██████▉   | 4499/6470 [56:21<24:22,  1.35it/s]
2|4500|Loss: 0.858735203742981:  70%|██████▉   | 4499/6470 [56:22<24:22,  1.35it/s] 
2|4500|Loss: 0.858735203742981:  70%|██████▉   | 4500/6470 [56:22<25:24,  1.29it/s]
2|4501|Loss: 0.8587040901184082:  70%|██████▉   | 4500/6470 [56:23<25:24,  1.29it/s]
2|4501|Loss: 0.8587040901184082:  70%|██████▉   | 4501/6470 [56:23<25:54,  1.27it/s]
2|4502|Loss: 0.8047932386398315:  70%|██████▉   | 4501/6470 [56:23<25:54,  1.27it/s]
2|4502|Loss: 0.8047932386398315:  70%|██████▉   | 4502/6470 [56:24<24:09,  1.36it/s]
2|4503|Loss: 0.7606220841407776:  70%|██████▉   | 4502/6470 [56:24<24:09,  1.36it/s]
2|4503|Loss: 0.7606220841407776:  70%|██████▉   | 4503/6470 [56:24<24:21,  1.35it/s]
2|4504|Loss: 0.7629272937774658:  70%|██████▉   | 4503/6470 [56:25<24:21,  1.35it/s]
2|4504|Loss: 0.7629272937774658:  70%|██████▉   | 4504/6470 [56:25<25:32,  1.28it/s]
2|4505|Loss: 0.813032865524292:  70%|██████▉   | 4504/6470 [56:26<25:32,  1.28it/s] 
2|4505|Loss: 0.813032865524292:  70%|██████▉   | 4505/6470 [56:26<27:04,  1.21it/s]
2|4506|Loss: 0.7709628343582153:  70%|██████▉   | 4505/6470 [56:27<27:04,  1.21it/s]
2|4506|Loss: 0.7709628343582153:  70%|██████▉   | 4506/6470 [56:27<26:55,  1.22it/s]
2|4507|Loss: 0.8781458139419556:  70%|██████▉   | 4506/6470 [56:27<26:55,  1.22it/s]
2|4507|Loss: 0.8781458139419556:  70%|██████▉   | 4507/6470 [56:28<25:52,  1.26it/s]
2|4508|Loss: 0.8788970112800598:  70%|██████▉   | 4507/6470 [56:28<25:52,  1.26it/s]
2|4508|Loss: 0.8788970112800598:  70%|██████▉   | 4508/6470 [56:28<24:42,  1.32it/s]
2|4509|Loss: 0.9331859946250916:  70%|██████▉   | 4508/6470 [56:29<24:42,  1.32it/s]
2|4509|Loss: 0.9331859946250916:  70%|██████▉   | 4509/6470 [56:29<24:38,  1.33it/s]
2|4510|Loss: 0.9037352800369263:  70%|██████▉   | 4509/6470 [56:29<24:38,  1.33it/s]
2|4510|Loss: 0.9037352800369263:  70%|██████▉   | 4510/6470 [56:30<25:05,  1.30it/s]
2|4511|Loss: 0.8422882556915283:  70%|██████▉   | 4510/6470 [56:30<25:05,  1.30it/s]
2|4511|Loss: 0.8422882556915283:  70%|██████▉   | 4511/6470 [56:31<25:24,  1.29it/s]
2|4512|Loss: 0.9302813410758972:  70%|██████▉   | 4511/6470 [56:31<25:24,  1.29it/s]
2|4512|Loss: 0.9302813410758972:  70%|██████▉   | 4512/6470 [56:31<26:22,  1.24it/s]
2|4513|Loss: 0.9472979307174683:  70%|██████▉   | 4512/6470 [56:32<26:22,  1.24it/s]
2|4513|Loss: 0.9472979307174683:  70%|██████▉   | 4513/6470 [56:32<25:38,  1.27it/s]
2|4514|Loss: 0.8678709268569946:  70%|██████▉   | 4513/6470 [56:33<25:38,  1.27it/s]
2|4514|Loss: 0.8678709268569946:  70%|██████▉   | 4514/6470 [56:33<26:21,  1.24it/s]
2|4515|Loss: 0.8850432634353638:  70%|██████▉   | 4514/6470 [56:34<26:21,  1.24it/s]
2|4515|Loss: 0.8850432634353638:  70%|██████▉   | 4515/6470 [56:34<27:21,  1.19it/s]
2|4516|Loss: 0.8991782069206238:  70%|██████▉   | 4515/6470 [56:34<27:21,  1.19it/s]
2|4516|Loss: 0.8991782069206238:  70%|██████▉   | 4516/6470 [56:35<25:51,  1.26it/s]
2|4517|Loss: 0.7553994059562683:  70%|██████▉   | 4516/6470 [56:35<25:51,  1.26it/s]
2|4517|Loss: 0.7553994059562683:  70%|██████▉   | 4517/6470 [56:36<26:04,  1.25it/s]
2|4518|Loss: 0.8629663586616516:  70%|██████▉   | 4517/6470 [56:36<26:04,  1.25it/s]
2|4518|Loss: 0.8629663586616516:  70%|██████▉   | 4518/6470 [56:36<26:06,  1.25it/s]
2|4519|Loss: 0.5720059275627136:  70%|██████▉   | 4518/6470 [56:37<26:06,  1.25it/s]
2|4519|Loss: 0.5720059275627136:  70%|██████▉   | 4519/6470 [56:37<26:49,  1.21it/s]
2|4520|Loss: 0.696652889251709:  70%|██████▉   | 4519/6470 [56:38<26:49,  1.21it/s] 
2|4520|Loss: 0.696652889251709:  70%|██████▉   | 4520/6470 [56:38<24:30,  1.33it/s]
2|4521|Loss: 0.9494783878326416:  70%|██████▉   | 4520/6470 [56:38<24:30,  1.33it/s]
2|4521|Loss: 0.9494783878326416:  70%|██████▉   | 4521/6470 [56:38<23:12,  1.40it/s]
2|4522|Loss: 0.7873578667640686:  70%|██████▉   | 4521/6470 [56:39<23:12,  1.40it/s]
2|4522|Loss: 0.7873578667640686:  70%|██████▉   | 4522/6470 [56:39<23:42,  1.37it/s]
2|4523|Loss: 1.1146845817565918:  70%|██████▉   | 4522/6470 [56:40<23:42,  1.37it/s]
2|4523|Loss: 1.1146845817565918:  70%|██████▉   | 4523/6470 [56:40<24:13,  1.34it/s]
2|4524|Loss: 0.754339337348938:  70%|██████▉   | 4523/6470 [56:40<24:13,  1.34it/s] 
2|4524|Loss: 0.754339337348938:  70%|██████▉   | 4524/6470 [56:41<24:45,  1.31it/s]
2|4525|Loss: 0.8849314451217651:  70%|██████▉   | 4524/6470 [56:41<24:45,  1.31it/s]
2|4525|Loss: 0.8849314451217651:  70%|██████▉   | 4525/6470 [56:42<24:46,  1.31it/s]
2|4526|Loss: 0.84419846534729:  70%|██████▉   | 4525/6470 [56:42<24:46,  1.31it/s]  
2|4526|Loss: 0.84419846534729:  70%|██████▉   | 4526/6470 [56:42<23:50,  1.36it/s]
2|4527|Loss: 0.8061792254447937:  70%|██████▉   | 4526/6470 [56:43<23:50,  1.36it/s]
2|4527|Loss: 0.8061792254447937:  70%|██████▉   | 4527/6470 [56:43<23:40,  1.37it/s]
2|4528|Loss: 0.7400659918785095:  70%|██████▉   | 4527/6470 [56:43<23:40,  1.37it/s]
2|4528|Loss: 0.7400659918785095:  70%|██████▉   | 4528/6470 [56:44<23:05,  1.40it/s]
2|4529|Loss: 0.8025156259536743:  70%|██████▉   | 4528/6470 [56:44<23:05,  1.40it/s]
2|4529|Loss: 0.8025156259536743:  70%|███████   | 4529/6470 [56:44<21:07,  1.53it/s]
2|4530|Loss: 0.9301555156707764:  70%|███████   | 4529/6470 [56:44<21:07,  1.53it/s]
2|4530|Loss: 0.9301555156707764:  70%|███████   | 4530/6470 [56:45<21:03,  1.54it/s]
2|4531|Loss: 0.8677569031715393:  70%|███████   | 4530/6470 [56:45<21:03,  1.54it/s]
2|4531|Loss: 0.8677569031715393:  70%|███████   | 4531/6470 [56:46<23:51,  1.35it/s]
2|4532|Loss: 0.877236008644104:  70%|███████   | 4531/6470 [56:46<23:51,  1.35it/s] 
2|4532|Loss: 0.877236008644104:  70%|███████   | 4532/6470 [56:46<22:24,  1.44it/s]
2|4533|Loss: 0.9894834160804749:  70%|███████   | 4532/6470 [56:47<22:24,  1.44it/s]
2|4533|Loss: 0.9894834160804749:  70%|███████   | 4533/6470 [56:47<22:26,  1.44it/s]
2|4534|Loss: 0.7317203879356384:  70%|███████   | 4533/6470 [56:47<22:26,  1.44it/s]
2|4534|Loss: 0.7317203879356384:  70%|███████   | 4534/6470 [56:48<21:59,  1.47it/s]
2|4535|Loss: 0.8702369332313538:  70%|███████   | 4534/6470 [56:48<21:59,  1.47it/s]
2|4535|Loss: 0.8702369332313538:  70%|███████   | 4535/6470 [56:48<22:45,  1.42it/s]
2|4536|Loss: 0.8423041701316833:  70%|███████   | 4535/6470 [56:49<22:45,  1.42it/s]
2|4536|Loss: 0.8423041701316833:  70%|███████   | 4536/6470 [56:49<23:48,  1.35it/s]
2|4537|Loss: 0.709830105304718:  70%|███████   | 4536/6470 [56:50<23:48,  1.35it/s] 
2|4537|Loss: 0.709830105304718:  70%|███████   | 4537/6470 [56:50<24:06,  1.34it/s]
2|4538|Loss: 0.9004688262939453:  70%|███████   | 4537/6470 [56:50<24:06,  1.34it/s]
2|4538|Loss: 0.9004688262939453:  70%|███████   | 4538/6470 [56:51<22:29,  1.43it/s]
2|4539|Loss: 0.9410241842269897:  70%|███████   | 4538/6470 [56:51<22:29,  1.43it/s]
2|4539|Loss: 0.9410241842269897:  70%|███████   | 4539/6470 [56:51<21:31,  1.50it/s]
2|4540|Loss: 0.8095216155052185:  70%|███████   | 4539/6470 [56:52<21:31,  1.50it/s]
2|4540|Loss: 0.8095216155052185:  70%|███████   | 4540/6470 [56:52<22:21,  1.44it/s]
2|4541|Loss: 0.7829934358596802:  70%|███████   | 4540/6470 [56:52<22:21,  1.44it/s]
2|4541|Loss: 0.7829934358596802:  70%|███████   | 4541/6470 [56:53<22:21,  1.44it/s]
2|4542|Loss: 0.9904674887657166:  70%|███████   | 4541/6470 [56:53<22:21,  1.44it/s]
2|4542|Loss: 0.9904674887657166:  70%|███████   | 4542/6470 [56:53<24:02,  1.34it/s]
2|4543|Loss: 0.832158625125885:  70%|███████   | 4542/6470 [56:54<24:02,  1.34it/s] 
2|4543|Loss: 0.832158625125885:  70%|███████   | 4543/6470 [56:54<22:56,  1.40it/s]
2|4544|Loss: 0.7298862934112549:  70%|███████   | 4543/6470 [56:54<22:56,  1.40it/s]
2|4544|Loss: 0.7298862934112549:  70%|███████   | 4544/6470 [56:55<22:27,  1.43it/s]
2|4545|Loss: 0.8151246905326843:  70%|███████   | 4544/6470 [56:55<22:27,  1.43it/s]
2|4545|Loss: 0.8151246905326843:  70%|███████   | 4545/6470 [56:55<22:28,  1.43it/s]
2|4546|Loss: 0.7660201191902161:  70%|███████   | 4545/6470 [56:56<22:28,  1.43it/s]
2|4546|Loss: 0.7660201191902161:  70%|███████   | 4546/6470 [56:56<23:08,  1.39it/s]
2|4547|Loss: 0.9144119024276733:  70%|███████   | 4546/6470 [56:57<23:08,  1.39it/s]
2|4547|Loss: 0.9144119024276733:  70%|███████   | 4547/6470 [56:57<24:54,  1.29it/s]
2|4548|Loss: 0.8681824207305908:  70%|███████   | 4547/6470 [56:58<24:54,  1.29it/s]
2|4548|Loss: 0.8681824207305908:  70%|███████   | 4548/6470 [56:58<24:42,  1.30it/s]
2|4549|Loss: 0.7235081791877747:  70%|███████   | 4548/6470 [56:58<24:42,  1.30it/s]
2|4549|Loss: 0.7235081791877747:  70%|███████   | 4549/6470 [56:59<23:46,  1.35it/s]
2|4550|Loss: 0.7598022222518921:  70%|███████   | 4549/6470 [56:59<23:46,  1.35it/s]
2|4550|Loss: 0.7598022222518921:  70%|███████   | 4550/6470 [56:59<23:04,  1.39it/s]
2|4551|Loss: 0.74249666929245:  70%|███████   | 4550/6470 [57:00<23:04,  1.39it/s]  
2|4551|Loss: 0.74249666929245:  70%|███████   | 4551/6470 [57:00<23:37,  1.35it/s]
2|4552|Loss: 0.8141281008720398:  70%|███████   | 4551/6470 [57:00<23:37,  1.35it/s]
2|4552|Loss: 0.8141281008720398:  70%|███████   | 4552/6470 [57:01<23:26,  1.36it/s]
2|4553|Loss: 0.8598908185958862:  70%|███████   | 4552/6470 [57:01<23:26,  1.36it/s]
2|4553|Loss: 0.8598908185958862:  70%|███████   | 4553/6470 [57:02<25:20,  1.26it/s]
2|4554|Loss: 0.9361578226089478:  70%|███████   | 4553/6470 [57:02<25:20,  1.26it/s]
2|4554|Loss: 0.9361578226089478:  70%|███████   | 4554/6470 [57:02<25:06,  1.27it/s]
2|4555|Loss: 0.8075898885726929:  70%|███████   | 4554/6470 [57:03<25:06,  1.27it/s]
2|4555|Loss: 0.8075898885726929:  70%|███████   | 4555/6470 [57:03<24:27,  1.31it/s]
2|4556|Loss: 0.7548258304595947:  70%|███████   | 4555/6470 [57:04<24:27,  1.31it/s]
2|4556|Loss: 0.7548258304595947:  70%|███████   | 4556/6470 [57:04<23:41,  1.35it/s]
2|4557|Loss: 0.8451708555221558:  70%|███████   | 4556/6470 [57:04<23:41,  1.35it/s]
2|4557|Loss: 0.8451708555221558:  70%|███████   | 4557/6470 [57:05<22:46,  1.40it/s]
2|4558|Loss: 0.7677916288375854:  70%|███████   | 4557/6470 [57:05<22:46,  1.40it/s]
2|4558|Loss: 0.7677916288375854:  70%|███████   | 4558/6470 [57:05<22:24,  1.42it/s]
2|4559|Loss: 0.855877161026001:  70%|███████   | 4558/6470 [57:06<22:24,  1.42it/s] 
2|4559|Loss: 0.855877161026001:  70%|███████   | 4559/6470 [57:06<22:45,  1.40it/s]
2|4560|Loss: 0.923686146736145:  70%|███████   | 4559/6470 [57:06<22:45,  1.40it/s]
2|4560|Loss: 0.923686146736145:  70%|███████   | 4560/6470 [57:07<22:40,  1.40it/s]
2|4561|Loss: 0.8252060413360596:  70%|███████   | 4560/6470 [57:07<22:40,  1.40it/s]
2|4561|Loss: 0.8252060413360596:  70%|███████   | 4561/6470 [57:07<22:56,  1.39it/s]
2|4562|Loss: 0.7814954519271851:  70%|███████   | 4561/6470 [57:08<22:56,  1.39it/s]
2|4562|Loss: 0.7814954519271851:  71%|███████   | 4562/6470 [57:08<23:16,  1.37it/s]
2|4563|Loss: 0.8914069533348083:  71%|███████   | 4562/6470 [57:09<23:16,  1.37it/s]
2|4563|Loss: 0.8914069533348083:  71%|███████   | 4563/6470 [57:09<23:34,  1.35it/s]
2|4564|Loss: 0.8786138296127319:  71%|███████   | 4563/6470 [57:09<23:34,  1.35it/s]
2|4564|Loss: 0.8786138296127319:  71%|███████   | 4564/6470 [57:10<23:02,  1.38it/s]
2|4565|Loss: 0.9319296479225159:  71%|███████   | 4564/6470 [57:10<23:02,  1.38it/s]
2|4565|Loss: 0.9319296479225159:  71%|███████   | 4565/6470 [57:10<23:07,  1.37it/s]
2|4566|Loss: 0.868911862373352:  71%|███████   | 4565/6470 [57:11<23:07,  1.37it/s] 
2|4566|Loss: 0.868911862373352:  71%|███████   | 4566/6470 [57:11<22:36,  1.40it/s]
2|4567|Loss: 0.924464225769043:  71%|███████   | 4566/6470 [57:11<22:36,  1.40it/s]
2|4567|Loss: 0.924464225769043:  71%|███████   | 4567/6470 [57:12<21:45,  1.46it/s]
2|4568|Loss: 0.9488137364387512:  71%|███████   | 4567/6470 [57:12<21:45,  1.46it/s]
2|4568|Loss: 0.9488137364387512:  71%|███████   | 4568/6470 [57:12<21:01,  1.51it/s]
2|4569|Loss: 0.7555150985717773:  71%|███████   | 4568/6470 [57:13<21:01,  1.51it/s]
2|4569|Loss: 0.7555150985717773:  71%|███████   | 4569/6470 [57:13<21:54,  1.45it/s]
2|4570|Loss: 0.8777827620506287:  71%|███████   | 4569/6470 [57:13<21:54,  1.45it/s]
2|4570|Loss: 0.8777827620506287:  71%|███████   | 4570/6470 [57:14<21:43,  1.46it/s]
2|4571|Loss: 0.8651203513145447:  71%|███████   | 4570/6470 [57:14<21:43,  1.46it/s]
2|4571|Loss: 0.8651203513145447:  71%|███████   | 4571/6470 [57:14<20:37,  1.53it/s]
2|4572|Loss: 0.893817663192749:  71%|███████   | 4571/6470 [57:15<20:37,  1.53it/s] 
2|4572|Loss: 0.893817663192749:  71%|███████   | 4572/6470 [57:15<22:17,  1.42it/s]
2|4573|Loss: 0.7834542393684387:  71%|███████   | 4572/6470 [57:16<22:17,  1.42it/s]
2|4573|Loss: 0.7834542393684387:  71%|███████   | 4573/6470 [57:16<23:12,  1.36it/s]
2|4574|Loss: 0.83843594789505:  71%|███████   | 4573/6470 [57:16<23:12,  1.36it/s]  
2|4574|Loss: 0.83843594789505:  71%|███████   | 4574/6470 [57:17<23:43,  1.33it/s]
2|4575|Loss: 0.8658348321914673:  71%|███████   | 4574/6470 [57:17<23:43,  1.33it/s]
2|4575|Loss: 0.8658348321914673:  71%|███████   | 4575/6470 [57:18<25:05,  1.26it/s]
2|4576|Loss: 0.9087041020393372:  71%|███████   | 4575/6470 [57:18<25:05,  1.26it/s]
2|4576|Loss: 0.9087041020393372:  71%|███████   | 4576/6470 [57:18<25:09,  1.25it/s]
2|4577|Loss: 0.8853784203529358:  71%|███████   | 4576/6470 [57:19<25:09,  1.25it/s]
2|4577|Loss: 0.8853784203529358:  71%|███████   | 4577/6470 [57:19<25:44,  1.23it/s]
2|4578|Loss: 0.9177407622337341:  71%|███████   | 4577/6470 [57:20<25:44,  1.23it/s]
2|4578|Loss: 0.9177407622337341:  71%|███████   | 4578/6470 [57:20<24:55,  1.26it/s]
2|4579|Loss: 0.8021367788314819:  71%|███████   | 4578/6470 [57:20<24:55,  1.26it/s]
2|4579|Loss: 0.8021367788314819:  71%|███████   | 4579/6470 [57:21<23:33,  1.34it/s]
2|4580|Loss: 0.8438327312469482:  71%|███████   | 4579/6470 [57:21<23:33,  1.34it/s]
2|4580|Loss: 0.8438327312469482:  71%|███████   | 4580/6470 [57:21<22:37,  1.39it/s]
2|4581|Loss: 0.7228934168815613:  71%|███████   | 4580/6470 [57:22<22:37,  1.39it/s]
2|4581|Loss: 0.7228934168815613:  71%|███████   | 4581/6470 [57:22<22:16,  1.41it/s]
2|4582|Loss: 0.8820421695709229:  71%|███████   | 4581/6470 [57:22<22:16,  1.41it/s]
2|4582|Loss: 0.8820421695709229:  71%|███████   | 4582/6470 [57:23<21:28,  1.47it/s]
2|4583|Loss: 0.8477448225021362:  71%|███████   | 4582/6470 [57:23<21:28,  1.47it/s]
2|4583|Loss: 0.8477448225021362:  71%|███████   | 4583/6470 [57:23<21:54,  1.44it/s]
2|4584|Loss: 0.8311625719070435:  71%|███████   | 4583/6470 [57:24<21:54,  1.44it/s]
2|4584|Loss: 0.8311625719070435:  71%|███████   | 4584/6470 [57:24<22:49,  1.38it/s]
2|4585|Loss: 0.8438596725463867:  71%|███████   | 4584/6470 [57:25<22:49,  1.38it/s]
2|4585|Loss: 0.8438596725463867:  71%|███████   | 4585/6470 [57:25<23:15,  1.35it/s]
2|4586|Loss: 0.9059240818023682:  71%|███████   | 4585/6470 [57:25<23:15,  1.35it/s]
2|4586|Loss: 0.9059240818023682:  71%|███████   | 4586/6470 [57:26<23:23,  1.34it/s]
2|4587|Loss: 0.7218747138977051:  71%|███████   | 4586/6470 [57:26<23:23,  1.34it/s]
2|4587|Loss: 0.7218747138977051:  71%|███████   | 4587/6470 [57:26<21:50,  1.44it/s]
2|4588|Loss: 0.8388257026672363:  71%|███████   | 4587/6470 [57:27<21:50,  1.44it/s]
2|4588|Loss: 0.8388257026672363:  71%|███████   | 4588/6470 [57:27<22:44,  1.38it/s]
2|4589|Loss: 0.8116349577903748:  71%|███████   | 4588/6470 [57:27<22:44,  1.38it/s]
2|4589|Loss: 0.8116349577903748:  71%|███████   | 4589/6470 [57:28<22:46,  1.38it/s]
2|4590|Loss: 0.8385435938835144:  71%|███████   | 4589/6470 [57:28<22:46,  1.38it/s]
2|4590|Loss: 0.8385435938835144:  71%|███████   | 4590/6470 [57:28<22:00,  1.42it/s]
2|4591|Loss: 0.9445888996124268:  71%|███████   | 4590/6470 [57:29<22:00,  1.42it/s]
2|4591|Loss: 0.9445888996124268:  71%|███████   | 4591/6470 [57:29<20:58,  1.49it/s]
2|4592|Loss: 0.8514184355735779:  71%|███████   | 4591/6470 [57:29<20:58,  1.49it/s]
2|4592|Loss: 0.8514184355735779:  71%|███████   | 4592/6470 [57:30<21:24,  1.46it/s]
2|4593|Loss: 0.9091008901596069:  71%|███████   | 4592/6470 [57:30<21:24,  1.46it/s]
2|4593|Loss: 0.9091008901596069:  71%|███████   | 4593/6470 [57:30<22:08,  1.41it/s]
2|4594|Loss: 0.9752856492996216:  71%|███████   | 4593/6470 [57:31<22:08,  1.41it/s]
2|4594|Loss: 0.9752856492996216:  71%|███████   | 4594/6470 [57:31<22:07,  1.41it/s]
2|4595|Loss: 1.2315946817398071:  71%|███████   | 4594/6470 [57:32<22:07,  1.41it/s]
2|4595|Loss: 1.2315946817398071:  71%|███████   | 4595/6470 [57:32<22:26,  1.39it/s]
2|4596|Loss: 0.8119519352912903:  71%|███████   | 4595/6470 [57:32<22:26,  1.39it/s]
2|4596|Loss: 0.8119519352912903:  71%|███████   | 4596/6470 [57:33<21:32,  1.45it/s]
2|4597|Loss: 0.9798876047134399:  71%|███████   | 4596/6470 [57:33<21:32,  1.45it/s]
2|4597|Loss: 0.9798876047134399:  71%|███████   | 4597/6470 [57:33<22:11,  1.41it/s]
2|4598|Loss: 0.8892257213592529:  71%|███████   | 4597/6470 [57:34<22:11,  1.41it/s]
2|4598|Loss: 0.8892257213592529:  71%|███████   | 4598/6470 [57:34<22:40,  1.38it/s]
2|4599|Loss: 0.8337206244468689:  71%|███████   | 4598/6470 [57:34<22:40,  1.38it/s]
2|4599|Loss: 0.8337206244468689:  71%|███████   | 4599/6470 [57:35<23:14,  1.34it/s]
2|4600|Loss: 0.8142774105072021:  71%|███████   | 4599/6470 [57:35<23:14,  1.34it/s]
2|4600|Loss: 0.8142774105072021:  71%|███████   | 4600/6470 [57:36<23:30,  1.33it/s]
2|4601|Loss: 0.8456553816795349:  71%|███████   | 4600/6470 [57:36<23:30,  1.33it/s]
2|4601|Loss: 0.8456553816795349:  71%|███████   | 4601/6470 [57:36<23:43,  1.31it/s]
2|4602|Loss: 1.0114810466766357:  71%|███████   | 4601/6470 [57:37<23:43,  1.31it/s]
2|4602|Loss: 1.0114810466766357:  71%|███████   | 4602/6470 [57:37<23:08,  1.35it/s]
2|4603|Loss: 0.7616404891014099:  71%|███████   | 4602/6470 [57:37<23:08,  1.35it/s]
2|4603|Loss: 0.7616404891014099:  71%|███████   | 4603/6470 [57:38<23:29,  1.32it/s]
2|4604|Loss: 0.9283842444419861:  71%|███████   | 4603/6470 [57:38<23:29,  1.32it/s]
2|4604|Loss: 0.9283842444419861:  71%|███████   | 4604/6470 [57:39<24:57,  1.25it/s]
2|4605|Loss: 0.8242374658584595:  71%|███████   | 4604/6470 [57:39<24:57,  1.25it/s]
2|4605|Loss: 0.8242374658584595:  71%|███████   | 4605/6470 [57:40<25:11,  1.23it/s]
2|4606|Loss: 0.8059411644935608:  71%|███████   | 4605/6470 [57:40<25:11,  1.23it/s]
2|4606|Loss: 0.8059411644935608:  71%|███████   | 4606/6470 [57:41<26:31,  1.17it/s]
2|4607|Loss: 0.7452306151390076:  71%|███████   | 4606/6470 [57:41<26:31,  1.17it/s]
2|4607|Loss: 0.7452306151390076:  71%|███████   | 4607/6470 [57:41<25:07,  1.24it/s]
2|4608|Loss: 0.7047401666641235:  71%|███████   | 4607/6470 [57:42<25:07,  1.24it/s]
2|4608|Loss: 0.7047401666641235:  71%|███████   | 4608/6470 [57:42<25:04,  1.24it/s]
2|4609|Loss: 0.7848058938980103:  71%|███████   | 4608/6470 [57:42<25:04,  1.24it/s]
2|4609|Loss: 0.7848058938980103:  71%|███████   | 4609/6470 [57:43<23:43,  1.31it/s]
2|4610|Loss: 0.7208954095840454:  71%|███████   | 4609/6470 [57:43<23:43,  1.31it/s]
2|4610|Loss: 0.7208954095840454:  71%|███████▏  | 4610/6470 [57:43<23:35,  1.31it/s]
2|4611|Loss: 0.8545434474945068:  71%|███████▏  | 4610/6470 [57:44<23:35,  1.31it/s]
2|4611|Loss: 0.8545434474945068:  71%|███████▏  | 4611/6470 [57:44<24:19,  1.27it/s]
2|4612|Loss: 0.8257615566253662:  71%|███████▏  | 4611/6470 [57:45<24:19,  1.27it/s]
2|4612|Loss: 0.8257615566253662:  71%|███████▏  | 4612/6470 [57:45<23:37,  1.31it/s]
2|4613|Loss: 0.8723645806312561:  71%|███████▏  | 4612/6470 [57:45<23:37,  1.31it/s]
2|4613|Loss: 0.8723645806312561:  71%|███████▏  | 4613/6470 [57:46<22:20,  1.39it/s]
2|4614|Loss: 1.0541355609893799:  71%|███████▏  | 4613/6470 [57:46<22:20,  1.39it/s]
2|4614|Loss: 1.0541355609893799:  71%|███████▏  | 4614/6470 [57:46<21:56,  1.41it/s]
2|4615|Loss: 0.8758567571640015:  71%|███████▏  | 4614/6470 [57:47<21:56,  1.41it/s]
2|4615|Loss: 0.8758567571640015:  71%|███████▏  | 4615/6470 [57:47<22:16,  1.39it/s]
2|4616|Loss: 0.807346761226654:  71%|███████▏  | 4615/6470 [57:47<22:16,  1.39it/s] 
2|4616|Loss: 0.807346761226654:  71%|███████▏  | 4616/6470 [57:48<20:56,  1.48it/s]
2|4617|Loss: 0.7862012982368469:  71%|███████▏  | 4616/6470 [57:48<20:56,  1.48it/s]
2|4617|Loss: 0.7862012982368469:  71%|███████▏  | 4617/6470 [57:48<21:31,  1.43it/s]
2|4618|Loss: 0.8689098358154297:  71%|███████▏  | 4617/6470 [57:49<21:31,  1.43it/s]
2|4618|Loss: 0.8689098358154297:  71%|███████▏  | 4618/6470 [57:49<22:17,  1.38it/s]
2|4619|Loss: 0.6774993538856506:  71%|███████▏  | 4618/6470 [57:50<22:17,  1.38it/s]
2|4619|Loss: 0.6774993538856506:  71%|███████▏  | 4619/6470 [57:50<23:04,  1.34it/s]
2|4620|Loss: 0.7882345914840698:  71%|███████▏  | 4619/6470 [57:50<23:04,  1.34it/s]
2|4620|Loss: 0.7882345914840698:  71%|███████▏  | 4620/6470 [57:51<22:28,  1.37it/s]
2|4621|Loss: 0.7855320572853088:  71%|███████▏  | 4620/6470 [57:51<22:28,  1.37it/s]
2|4621|Loss: 0.7855320572853088:  71%|███████▏  | 4621/6470 [57:51<23:15,  1.33it/s]
2|4622|Loss: 0.9396449327468872:  71%|███████▏  | 4621/6470 [57:52<23:15,  1.33it/s]
2|4622|Loss: 0.9396449327468872:  71%|███████▏  | 4622/6470 [57:52<23:03,  1.34it/s]
2|4623|Loss: 0.7422736287117004:  71%|███████▏  | 4622/6470 [57:53<23:03,  1.34it/s]
2|4623|Loss: 0.7422736287117004:  71%|███████▏  | 4623/6470 [57:53<21:59,  1.40it/s]
2|4624|Loss: 0.8289291858673096:  71%|███████▏  | 4623/6470 [57:53<21:59,  1.40it/s]
2|4624|Loss: 0.8289291858673096:  71%|███████▏  | 4624/6470 [57:54<23:47,  1.29it/s]
2|4625|Loss: 0.7546935677528381:  71%|███████▏  | 4624/6470 [57:54<23:47,  1.29it/s]
2|4625|Loss: 0.7546935677528381:  71%|███████▏  | 4625/6470 [57:55<23:44,  1.30it/s]
2|4626|Loss: 0.7856511473655701:  71%|███████▏  | 4625/6470 [57:55<23:44,  1.30it/s]
2|4626|Loss: 0.7856511473655701:  71%|███████▏  | 4626/6470 [57:55<24:42,  1.24it/s]
2|4627|Loss: 0.7559957504272461:  71%|███████▏  | 4626/6470 [57:56<24:42,  1.24it/s]
2|4627|Loss: 0.7559957504272461:  72%|███████▏  | 4627/6470 [57:56<24:34,  1.25it/s]
2|4628|Loss: 0.8918830156326294:  72%|███████▏  | 4627/6470 [57:57<24:34,  1.25it/s]
2|4628|Loss: 0.8918830156326294:  72%|███████▏  | 4628/6470 [57:57<24:28,  1.25it/s]
2|4629|Loss: 0.6501937508583069:  72%|███████▏  | 4628/6470 [57:57<24:28,  1.25it/s]
2|4629|Loss: 0.6501937508583069:  72%|███████▏  | 4629/6470 [57:58<25:00,  1.23it/s]
2|4630|Loss: 0.7022215127944946:  72%|███████▏  | 4629/6470 [57:58<25:00,  1.23it/s]
2|4630|Loss: 0.7022215127944946:  72%|███████▏  | 4630/6470 [57:59<25:39,  1.20it/s]
2|4631|Loss: 0.7980429530143738:  72%|███████▏  | 4630/6470 [57:59<25:39,  1.20it/s]
2|4631|Loss: 0.7980429530143738:  72%|███████▏  | 4631/6470 [57:59<23:48,  1.29it/s]
2|4632|Loss: 0.7914935350418091:  72%|███████▏  | 4631/6470 [58:00<23:48,  1.29it/s]
2|4632|Loss: 0.7914935350418091:  72%|███████▏  | 4632/6470 [58:00<23:31,  1.30it/s]
2|4633|Loss: 0.8984466791152954:  72%|███████▏  | 4632/6470 [58:01<23:31,  1.30it/s]
2|4633|Loss: 0.8984466791152954:  72%|███████▏  | 4633/6470 [58:01<22:04,  1.39it/s]
2|4634|Loss: 0.7000189423561096:  72%|███████▏  | 4633/6470 [58:01<22:04,  1.39it/s]
2|4634|Loss: 0.7000189423561096:  72%|███████▏  | 4634/6470 [58:02<23:48,  1.29it/s]
2|4635|Loss: 0.8537746667861938:  72%|███████▏  | 4634/6470 [58:02<23:48,  1.29it/s]
2|4635|Loss: 0.8537746667861938:  72%|███████▏  | 4635/6470 [58:02<24:19,  1.26it/s]
2|4636|Loss: 0.7667680978775024:  72%|███████▏  | 4635/6470 [58:03<24:19,  1.26it/s]
2|4636|Loss: 0.7667680978775024:  72%|███████▏  | 4636/6470 [58:03<25:12,  1.21it/s]
2|4637|Loss: 0.8163766264915466:  72%|███████▏  | 4636/6470 [58:04<25:12,  1.21it/s]
2|4637|Loss: 0.8163766264915466:  72%|███████▏  | 4637/6470 [58:04<24:17,  1.26it/s]
2|4638|Loss: 0.7520034909248352:  72%|███████▏  | 4637/6470 [58:05<24:17,  1.26it/s]
2|4638|Loss: 0.7520034909248352:  72%|███████▏  | 4638/6470 [58:05<24:59,  1.22it/s]
2|4639|Loss: 0.7962377071380615:  72%|███████▏  | 4638/6470 [58:05<24:59,  1.22it/s]
2|4639|Loss: 0.7962377071380615:  72%|███████▏  | 4639/6470 [58:06<23:28,  1.30it/s]
2|4640|Loss: 0.8115026950836182:  72%|███████▏  | 4639/6470 [58:06<23:28,  1.30it/s]
2|4640|Loss: 0.8115026950836182:  72%|███████▏  | 4640/6470 [58:06<20:55,  1.46it/s]
2|4641|Loss: 0.8714441657066345:  72%|███████▏  | 4640/6470 [58:06<20:55,  1.46it/s]
2|4641|Loss: 0.8714441657066345:  72%|███████▏  | 4641/6470 [58:07<20:21,  1.50it/s]
2|4642|Loss: 0.9988948702812195:  72%|███████▏  | 4641/6470 [58:07<20:21,  1.50it/s]
2|4642|Loss: 0.9988948702812195:  72%|███████▏  | 4642/6470 [58:08<21:15,  1.43it/s]
2|4643|Loss: 0.7490199208259583:  72%|███████▏  | 4642/6470 [58:08<21:15,  1.43it/s]
2|4643|Loss: 0.7490199208259583:  72%|███████▏  | 4643/6470 [58:08<21:18,  1.43it/s]
2|4644|Loss: 0.7737751007080078:  72%|███████▏  | 4643/6470 [58:09<21:18,  1.43it/s]
2|4644|Loss: 0.7737751007080078:  72%|███████▏  | 4644/6470 [58:09<21:53,  1.39it/s]
2|4645|Loss: 0.7458081245422363:  72%|███████▏  | 4644/6470 [58:09<21:53,  1.39it/s]
2|4645|Loss: 0.7458081245422363:  72%|███████▏  | 4645/6470 [58:10<22:27,  1.35it/s]
2|4646|Loss: 0.7420865893363953:  72%|███████▏  | 4645/6470 [58:10<22:27,  1.35it/s]
2|4646|Loss: 0.7420865893363953:  72%|███████▏  | 4646/6470 [58:11<22:42,  1.34it/s]
2|4647|Loss: 0.8745032548904419:  72%|███████▏  | 4646/6470 [58:11<22:42,  1.34it/s]
2|4647|Loss: 0.8745032548904419:  72%|███████▏  | 4647/6470 [58:11<23:47,  1.28it/s]
2|4648|Loss: 0.7824497222900391:  72%|███████▏  | 4647/6470 [58:12<23:47,  1.28it/s]
2|4648|Loss: 0.7824497222900391:  72%|███████▏  | 4648/6470 [58:12<23:05,  1.32it/s]
2|4649|Loss: 0.7245511412620544:  72%|███████▏  | 4648/6470 [58:13<23:05,  1.32it/s]
2|4649|Loss: 0.7245511412620544:  72%|███████▏  | 4649/6470 [58:13<22:52,  1.33it/s]
2|4650|Loss: 0.6962526440620422:  72%|███████▏  | 4649/6470 [58:13<22:52,  1.33it/s]
2|4650|Loss: 0.6962526440620422:  72%|███████▏  | 4650/6470 [58:13<20:52,  1.45it/s]
2|4651|Loss: 0.7964857816696167:  72%|███████▏  | 4650/6470 [58:14<20:52,  1.45it/s]
2|4651|Loss: 0.7964857816696167:  72%|███████▏  | 4651/6470 [58:14<20:16,  1.49it/s]
2|4652|Loss: 0.8516348600387573:  72%|███████▏  | 4651/6470 [58:14<20:16,  1.49it/s]
2|4652|Loss: 0.8516348600387573:  72%|███████▏  | 4652/6470 [58:15<19:57,  1.52it/s]
2|4653|Loss: 0.7979533672332764:  72%|███████▏  | 4652/6470 [58:15<19:57,  1.52it/s]
2|4653|Loss: 0.7979533672332764:  72%|███████▏  | 4653/6470 [58:15<21:15,  1.42it/s]
2|4654|Loss: 0.684257447719574:  72%|███████▏  | 4653/6470 [58:16<21:15,  1.42it/s] 
2|4654|Loss: 0.684257447719574:  72%|███████▏  | 4654/6470 [58:16<23:43,  1.28it/s]
2|4655|Loss: 0.9177888631820679:  72%|███████▏  | 4654/6470 [58:17<23:43,  1.28it/s]
2|4655|Loss: 0.9177888631820679:  72%|███████▏  | 4655/6470 [58:17<23:27,  1.29it/s]
2|4656|Loss: 0.8241762518882751:  72%|███████▏  | 4655/6470 [58:18<23:27,  1.29it/s]
2|4656|Loss: 0.8241762518882751:  72%|███████▏  | 4656/6470 [58:18<24:18,  1.24it/s]
2|4657|Loss: 0.8438937067985535:  72%|███████▏  | 4656/6470 [58:18<24:18,  1.24it/s]
2|4657|Loss: 0.8438937067985535:  72%|███████▏  | 4657/6470 [58:19<22:44,  1.33it/s]
2|4658|Loss: 1.002716064453125:  72%|███████▏  | 4657/6470 [58:19<22:44,  1.33it/s] 
2|4658|Loss: 1.002716064453125:  72%|███████▏  | 4658/6470 [58:19<20:53,  1.45it/s]
2|4659|Loss: 0.9401125311851501:  72%|███████▏  | 4658/6470 [58:20<20:53,  1.45it/s]
2|4659|Loss: 0.9401125311851501:  72%|███████▏  | 4659/6470 [58:20<20:38,  1.46it/s]
2|4660|Loss: 1.0146875381469727:  72%|███████▏  | 4659/6470 [58:20<20:38,  1.46it/s]
2|4660|Loss: 1.0146875381469727:  72%|███████▏  | 4660/6470 [58:21<21:31,  1.40it/s]
2|4661|Loss: 0.795096218585968:  72%|███████▏  | 4660/6470 [58:21<21:31,  1.40it/s] 
2|4661|Loss: 0.795096218585968:  72%|███████▏  | 4661/6470 [58:21<21:35,  1.40it/s]
2|4662|Loss: 0.8918359279632568:  72%|███████▏  | 4661/6470 [58:22<21:35,  1.40it/s]
2|4662|Loss: 0.8918359279632568:  72%|███████▏  | 4662/6470 [58:22<21:45,  1.38it/s]
2|4663|Loss: 0.934604823589325:  72%|███████▏  | 4662/6470 [58:23<21:45,  1.38it/s] 
2|4663|Loss: 0.934604823589325:  72%|███████▏  | 4663/6470 [58:23<22:12,  1.36it/s]
2|4664|Loss: 0.8577806949615479:  72%|███████▏  | 4663/6470 [58:23<22:12,  1.36it/s]
2|4664|Loss: 0.8577806949615479:  72%|███████▏  | 4664/6470 [58:24<23:31,  1.28it/s]
2|4665|Loss: 1.0705513954162598:  72%|███████▏  | 4664/6470 [58:24<23:31,  1.28it/s]
2|4665|Loss: 1.0705513954162598:  72%|███████▏  | 4665/6470 [58:25<23:46,  1.27it/s]
2|4666|Loss: 0.8023856282234192:  72%|███████▏  | 4665/6470 [58:25<23:46,  1.27it/s]
2|4666|Loss: 0.8023856282234192:  72%|███████▏  | 4666/6470 [58:25<23:25,  1.28it/s]
2|4667|Loss: 0.9509549140930176:  72%|███████▏  | 4666/6470 [58:26<23:25,  1.28it/s]
2|4667|Loss: 0.9509549140930176:  72%|███████▏  | 4667/6470 [58:26<24:12,  1.24it/s]
2|4668|Loss: 1.0013396739959717:  72%|███████▏  | 4667/6470 [58:27<24:12,  1.24it/s]
2|4668|Loss: 1.0013396739959717:  72%|███████▏  | 4668/6470 [58:27<23:58,  1.25it/s]
2|4669|Loss: 0.7578673362731934:  72%|███████▏  | 4668/6470 [58:27<23:58,  1.25it/s]
2|4669|Loss: 0.7578673362731934:  72%|███████▏  | 4669/6470 [58:28<22:37,  1.33it/s]
2|4670|Loss: 0.7650895714759827:  72%|███████▏  | 4669/6470 [58:28<22:37,  1.33it/s]
2|4670|Loss: 0.7650895714759827:  72%|███████▏  | 4670/6470 [58:28<21:38,  1.39it/s]
2|4671|Loss: 0.76707524061203:  72%|███████▏  | 4670/6470 [58:29<21:38,  1.39it/s]  
2|4671|Loss: 0.76707524061203:  72%|███████▏  | 4671/6470 [58:29<20:50,  1.44it/s]
2|4672|Loss: 0.8581435084342957:  72%|███████▏  | 4671/6470 [58:29<20:50,  1.44it/s]
2|4672|Loss: 0.8581435084342957:  72%|███████▏  | 4672/6470 [58:30<22:40,  1.32it/s]
2|4673|Loss: 0.8317524790763855:  72%|███████▏  | 4672/6470 [58:30<22:40,  1.32it/s]
2|4673|Loss: 0.8317524790763855:  72%|███████▏  | 4673/6470 [58:31<23:16,  1.29it/s]
2|4674|Loss: 0.8887303471565247:  72%|███████▏  | 4673/6470 [58:31<23:16,  1.29it/s]
2|4674|Loss: 0.8887303471565247:  72%|███████▏  | 4674/6470 [58:31<21:41,  1.38it/s]
2|4675|Loss: 0.7994322180747986:  72%|███████▏  | 4674/6470 [58:32<21:41,  1.38it/s]
2|4675|Loss: 0.7994322180747986:  72%|███████▏  | 4675/6470 [58:32<21:49,  1.37it/s]
2|4676|Loss: 0.9829034209251404:  72%|███████▏  | 4675/6470 [58:32<21:49,  1.37it/s]
2|4676|Loss: 0.9829034209251404:  72%|███████▏  | 4676/6470 [58:33<22:19,  1.34it/s]
2|4677|Loss: 0.8593400716781616:  72%|███████▏  | 4676/6470 [58:33<22:19,  1.34it/s]
2|4677|Loss: 0.8593400716781616:  72%|███████▏  | 4677/6470 [58:34<23:40,  1.26it/s]
2|4678|Loss: 0.7623441815376282:  72%|███████▏  | 4677/6470 [58:34<23:40,  1.26it/s]
2|4678|Loss: 0.7623441815376282:  72%|███████▏  | 4678/6470 [58:35<24:06,  1.24it/s]
2|4679|Loss: 0.8988255858421326:  72%|███████▏  | 4678/6470 [58:35<24:06,  1.24it/s]
2|4679|Loss: 0.8988255858421326:  72%|███████▏  | 4679/6470 [58:35<23:05,  1.29it/s]
2|4680|Loss: 0.8028993606567383:  72%|███████▏  | 4679/6470 [58:36<23:05,  1.29it/s]
2|4680|Loss: 0.8028993606567383:  72%|███████▏  | 4680/6470 [58:36<21:19,  1.40it/s]
2|4681|Loss: 0.9575172066688538:  72%|███████▏  | 4680/6470 [58:36<21:19,  1.40it/s]
2|4681|Loss: 0.9575172066688538:  72%|███████▏  | 4681/6470 [58:37<22:22,  1.33it/s]
2|4682|Loss: 0.8186198472976685:  72%|███████▏  | 4681/6470 [58:37<22:22,  1.33it/s]
2|4682|Loss: 0.8186198472976685:  72%|███████▏  | 4682/6470 [58:37<22:51,  1.30it/s]
2|4683|Loss: 1.0171232223510742:  72%|███████▏  | 4682/6470 [58:38<22:51,  1.30it/s]
2|4683|Loss: 1.0171232223510742:  72%|███████▏  | 4683/6470 [58:38<22:11,  1.34it/s]
2|4684|Loss: 0.6629776358604431:  72%|███████▏  | 4683/6470 [58:39<22:11,  1.34it/s]
2|4684|Loss: 0.6629776358604431:  72%|███████▏  | 4684/6470 [58:39<22:28,  1.32it/s]
2|4685|Loss: 0.8086276054382324:  72%|███████▏  | 4684/6470 [58:39<22:28,  1.32it/s]
2|4685|Loss: 0.8086276054382324:  72%|███████▏  | 4685/6470 [58:40<23:01,  1.29it/s]
2|4686|Loss: 0.9180461764335632:  72%|███████▏  | 4685/6470 [58:40<23:01,  1.29it/s]
2|4686|Loss: 0.9180461764335632:  72%|███████▏  | 4686/6470 [58:40<22:35,  1.32it/s]
2|4687|Loss: 0.8177311420440674:  72%|███████▏  | 4686/6470 [58:41<22:35,  1.32it/s]
2|4687|Loss: 0.8177311420440674:  72%|███████▏  | 4687/6470 [58:41<21:42,  1.37it/s]
2|4688|Loss: 0.849961519241333:  72%|███████▏  | 4687/6470 [58:42<21:42,  1.37it/s] 
2|4688|Loss: 0.849961519241333:  72%|███████▏  | 4688/6470 [58:42<22:03,  1.35it/s]
2|4689|Loss: 0.9291176795959473:  72%|███████▏  | 4688/6470 [58:42<22:03,  1.35it/s]
2|4689|Loss: 0.9291176795959473:  72%|███████▏  | 4689/6470 [58:43<21:02,  1.41it/s]
2|4690|Loss: 0.8855298757553101:  72%|███████▏  | 4689/6470 [58:43<21:02,  1.41it/s]
2|4690|Loss: 0.8855298757553101:  72%|███████▏  | 4690/6470 [58:43<21:56,  1.35it/s]
2|4691|Loss: 0.9207510948181152:  72%|███████▏  | 4690/6470 [58:44<21:56,  1.35it/s]
2|4691|Loss: 0.9207510948181152:  73%|███████▎  | 4691/6470 [58:44<21:40,  1.37it/s]
2|4692|Loss: 0.8333706259727478:  73%|███████▎  | 4691/6470 [58:44<21:40,  1.37it/s]
2|4692|Loss: 0.8333706259727478:  73%|███████▎  | 4692/6470 [58:45<22:05,  1.34it/s]
2|4693|Loss: 0.7444689869880676:  73%|███████▎  | 4692/6470 [58:45<22:05,  1.34it/s]
2|4693|Loss: 0.7444689869880676:  73%|███████▎  | 4693/6470 [58:46<21:41,  1.37it/s]
2|4694|Loss: 0.8142093420028687:  73%|███████▎  | 4693/6470 [58:46<21:41,  1.37it/s]
2|4694|Loss: 0.8142093420028687:  73%|███████▎  | 4694/6470 [58:46<22:52,  1.29it/s]
2|4695|Loss: 0.7056311368942261:  73%|███████▎  | 4694/6470 [58:47<22:52,  1.29it/s]
2|4695|Loss: 0.7056311368942261:  73%|███████▎  | 4695/6470 [58:47<24:01,  1.23it/s]
2|4696|Loss: 0.7848105430603027:  73%|███████▎  | 4695/6470 [58:48<24:01,  1.23it/s]
2|4696|Loss: 0.7848105430603027:  73%|███████▎  | 4696/6470 [58:48<24:47,  1.19it/s]
2|4697|Loss: 0.9032876491546631:  73%|███████▎  | 4696/6470 [58:49<24:47,  1.19it/s]
2|4697|Loss: 0.9032876491546631:  73%|███████▎  | 4697/6470 [58:49<23:36,  1.25it/s]
2|4698|Loss: 0.8364141583442688:  73%|███████▎  | 4697/6470 [58:49<23:36,  1.25it/s]
2|4698|Loss: 0.8364141583442688:  73%|███████▎  | 4698/6470 [58:49<21:29,  1.37it/s]
2|4699|Loss: 0.8531405329704285:  73%|███████▎  | 4698/6470 [58:50<21:29,  1.37it/s]
2|4699|Loss: 0.8531405329704285:  73%|███████▎  | 4699/6470 [58:50<20:09,  1.46it/s]
2|4700|Loss: 0.7955493330955505:  73%|███████▎  | 4699/6470 [58:50<20:09,  1.46it/s]
2|4700|Loss: 0.7955493330955505:  73%|███████▎  | 4700/6470 [58:51<19:49,  1.49it/s]
2|4701|Loss: 0.9110324382781982:  73%|███████▎  | 4700/6470 [58:51<19:49,  1.49it/s]
2|4701|Loss: 0.9110324382781982:  73%|███████▎  | 4701/6470 [58:52<21:37,  1.36it/s]
2|4702|Loss: 0.832146406173706:  73%|███████▎  | 4701/6470 [58:52<21:37,  1.36it/s] 
2|4702|Loss: 0.832146406173706:  73%|███████▎  | 4702/6470 [58:53<23:22,  1.26it/s]
2|4703|Loss: 0.7501745223999023:  73%|███████▎  | 4702/6470 [58:53<23:22,  1.26it/s]
2|4703|Loss: 0.7501745223999023:  73%|███████▎  | 4703/6470 [58:53<23:45,  1.24it/s]
2|4704|Loss: 0.8018660545349121:  73%|███████▎  | 4703/6470 [58:54<23:45,  1.24it/s]
2|4704|Loss: 0.8018660545349121:  73%|███████▎  | 4704/6470 [58:54<21:27,  1.37it/s]
2|4705|Loss: 0.767469048500061:  73%|███████▎  | 4704/6470 [58:54<21:27,  1.37it/s] 
2|4705|Loss: 0.767469048500061:  73%|███████▎  | 4705/6470 [58:55<20:24,  1.44it/s]
2|4706|Loss: 0.7347280383110046:  73%|███████▎  | 4705/6470 [58:55<20:24,  1.44it/s]
2|4706|Loss: 0.7347280383110046:  73%|███████▎  | 4706/6470 [58:55<20:20,  1.44it/s]
2|4707|Loss: 0.7670773267745972:  73%|███████▎  | 4706/6470 [58:56<20:20,  1.44it/s]
2|4707|Loss: 0.7670773267745972:  73%|███████▎  | 4707/6470 [58:56<22:04,  1.33it/s]
2|4708|Loss: 0.8082834482192993:  73%|███████▎  | 4707/6470 [58:57<22:04,  1.33it/s]
2|4708|Loss: 0.8082834482192993:  73%|███████▎  | 4708/6470 [58:57<23:18,  1.26it/s]
2|4709|Loss: 0.799307644367218:  73%|███████▎  | 4708/6470 [58:57<23:18,  1.26it/s] 
2|4709|Loss: 0.799307644367218:  73%|███████▎  | 4709/6470 [58:58<22:33,  1.30it/s]
2|4710|Loss: 0.8762421607971191:  73%|███████▎  | 4709/6470 [58:58<22:33,  1.30it/s]
2|4710|Loss: 0.8762421607971191:  73%|███████▎  | 4710/6470 [58:59<23:40,  1.24it/s]
2|4711|Loss: 0.9137079119682312:  73%|███████▎  | 4710/6470 [58:59<23:40,  1.24it/s]
2|4711|Loss: 0.9137079119682312:  73%|███████▎  | 4711/6470 [58:59<24:18,  1.21it/s]
2|4712|Loss: 0.6286629438400269:  73%|███████▎  | 4711/6470 [59:00<24:18,  1.21it/s]
2|4712|Loss: 0.6286629438400269:  73%|███████▎  | 4712/6470 [59:00<23:50,  1.23it/s]
2|4713|Loss: 0.7515720725059509:  73%|███████▎  | 4712/6470 [59:01<23:50,  1.23it/s]
2|4713|Loss: 0.7515720725059509:  73%|███████▎  | 4713/6470 [59:01<23:32,  1.24it/s]
2|4714|Loss: 1.1300830841064453:  73%|███████▎  | 4713/6470 [59:02<23:32,  1.24it/s]
2|4714|Loss: 1.1300830841064453:  73%|███████▎  | 4714/6470 [59:02<25:21,  1.15it/s]
2|4715|Loss: 0.8473153710365295:  73%|███████▎  | 4714/6470 [59:03<25:21,  1.15it/s]
2|4715|Loss: 0.8473153710365295:  73%|███████▎  | 4715/6470 [59:03<25:59,  1.13it/s]
2|4716|Loss: 0.7399383783340454:  73%|███████▎  | 4715/6470 [59:03<25:59,  1.13it/s]
2|4716|Loss: 0.7399383783340454:  73%|███████▎  | 4716/6470 [59:04<24:42,  1.18it/s]
2|4717|Loss: 0.92763352394104:  73%|███████▎  | 4716/6470 [59:04<24:42,  1.18it/s]  
2|4717|Loss: 0.92763352394104:  73%|███████▎  | 4717/6470 [59:04<22:53,  1.28it/s]
2|4718|Loss: 0.8362300992012024:  73%|███████▎  | 4717/6470 [59:05<22:53,  1.28it/s]
2|4718|Loss: 0.8362300992012024:  73%|███████▎  | 4718/6470 [59:05<21:44,  1.34it/s]
2|4719|Loss: 0.8703505396842957:  73%|███████▎  | 4718/6470 [59:05<21:44,  1.34it/s]
2|4719|Loss: 0.8703505396842957:  73%|███████▎  | 4719/6470 [59:06<21:40,  1.35it/s]
2|4720|Loss: 0.9077652096748352:  73%|███████▎  | 4719/6470 [59:06<21:40,  1.35it/s]
2|4720|Loss: 0.9077652096748352:  73%|███████▎  | 4720/6470 [59:06<21:42,  1.34it/s]
2|4721|Loss: 0.7504787445068359:  73%|███████▎  | 4720/6470 [59:07<21:42,  1.34it/s]
2|4721|Loss: 0.7504787445068359:  73%|███████▎  | 4721/6470 [59:07<21:48,  1.34it/s]
2|4722|Loss: 0.9077978730201721:  73%|███████▎  | 4721/6470 [59:08<21:48,  1.34it/s]
2|4722|Loss: 0.9077978730201721:  73%|███████▎  | 4722/6470 [59:08<21:54,  1.33it/s]
2|4723|Loss: 0.8096643090248108:  73%|███████▎  | 4722/6470 [59:08<21:54,  1.33it/s]
2|4723|Loss: 0.8096643090248108:  73%|███████▎  | 4723/6470 [59:09<21:56,  1.33it/s]
2|4724|Loss: 0.7792984843254089:  73%|███████▎  | 4723/6470 [59:09<21:56,  1.33it/s]
2|4724|Loss: 0.7792984843254089:  73%|███████▎  | 4724/6470 [59:10<21:53,  1.33it/s]
2|4725|Loss: 0.8882098197937012:  73%|███████▎  | 4724/6470 [59:10<21:53,  1.33it/s]
2|4725|Loss: 0.8882098197937012:  73%|███████▎  | 4725/6470 [59:10<21:55,  1.33it/s]
2|4726|Loss: 0.8105528950691223:  73%|███████▎  | 4725/6470 [59:11<21:55,  1.33it/s]
2|4726|Loss: 0.8105528950691223:  73%|███████▎  | 4726/6470 [59:11<22:54,  1.27it/s]
2|4727|Loss: 0.759375810623169:  73%|███████▎  | 4726/6470 [59:12<22:54,  1.27it/s] 
2|4727|Loss: 0.759375810623169:  73%|███████▎  | 4727/6470 [59:12<23:01,  1.26it/s]
2|4728|Loss: 0.7061878442764282:  73%|███████▎  | 4727/6470 [59:12<23:01,  1.26it/s]
2|4728|Loss: 0.7061878442764282:  73%|███████▎  | 4728/6470 [59:13<23:32,  1.23it/s]
2|4729|Loss: 0.8548367023468018:  73%|███████▎  | 4728/6470 [59:13<23:32,  1.23it/s]
2|4729|Loss: 0.8548367023468018:  73%|███████▎  | 4729/6470 [59:14<23:24,  1.24it/s]
2|4730|Loss: 0.7830461859703064:  73%|███████▎  | 4729/6470 [59:14<23:24,  1.24it/s]
2|4730|Loss: 0.7830461859703064:  73%|███████▎  | 4730/6470 [59:15<24:18,  1.19it/s]
2|4731|Loss: 0.8807424306869507:  73%|███████▎  | 4730/6470 [59:15<24:18,  1.19it/s]
2|4731|Loss: 0.8807424306869507:  73%|███████▎  | 4731/6470 [59:15<23:00,  1.26it/s]
2|4732|Loss: 0.8537466526031494:  73%|███████▎  | 4731/6470 [59:16<23:00,  1.26it/s]
2|4732|Loss: 0.8537466526031494:  73%|███████▎  | 4732/6470 [59:16<22:19,  1.30it/s]
2|4733|Loss: 0.7685239315032959:  73%|███████▎  | 4732/6470 [59:16<22:19,  1.30it/s]
2|4733|Loss: 0.7685239315032959:  73%|███████▎  | 4733/6470 [59:17<22:18,  1.30it/s]
2|4734|Loss: 0.8524432182312012:  73%|███████▎  | 4733/6470 [59:17<22:18,  1.30it/s]
2|4734|Loss: 0.8524432182312012:  73%|███████▎  | 4734/6470 [59:17<21:36,  1.34it/s]
2|4735|Loss: 0.7168005704879761:  73%|███████▎  | 4734/6470 [59:18<21:36,  1.34it/s]
2|4735|Loss: 0.7168005704879761:  73%|███████▎  | 4735/6470 [59:18<23:42,  1.22it/s]
2|4736|Loss: 0.916260302066803:  73%|███████▎  | 4735/6470 [59:19<23:42,  1.22it/s] 
2|4736|Loss: 0.916260302066803:  73%|███████▎  | 4736/6470 [59:19<22:43,  1.27it/s]
2|4737|Loss: 1.008719563484192:  73%|███████▎  | 4736/6470 [59:19<22:43,  1.27it/s]
2|4737|Loss: 1.008719563484192:  73%|███████▎  | 4737/6470 [59:20<22:24,  1.29it/s]
2|4738|Loss: 0.9665144681930542:  73%|███████▎  | 4737/6470 [59:20<22:24,  1.29it/s]
2|4738|Loss: 0.9665144681930542:  73%|███████▎  | 4738/6470 [59:21<21:50,  1.32it/s]
2|4739|Loss: 0.7701048254966736:  73%|███████▎  | 4738/6470 [59:21<21:50,  1.32it/s]
2|4739|Loss: 0.7701048254966736:  73%|███████▎  | 4739/6470 [59:21<19:18,  1.49it/s]
2|4740|Loss: 0.7827488780021667:  73%|███████▎  | 4739/6470 [59:21<19:18,  1.49it/s]
2|4740|Loss: 0.7827488780021667:  73%|███████▎  | 4740/6470 [59:22<19:50,  1.45it/s]
2|4741|Loss: 0.7448520660400391:  73%|███████▎  | 4740/6470 [59:22<19:50,  1.45it/s]
2|4741|Loss: 0.7448520660400391:  73%|███████▎  | 4741/6470 [59:22<18:40,  1.54it/s]
2|4742|Loss: 0.891083836555481:  73%|███████▎  | 4741/6470 [59:23<18:40,  1.54it/s] 
2|4742|Loss: 0.891083836555481:  73%|███████▎  | 4742/6470 [59:23<20:24,  1.41it/s]
2|4743|Loss: 0.8314915895462036:  73%|███████▎  | 4742/6470 [59:24<20:24,  1.41it/s]
2|4743|Loss: 0.8314915895462036:  73%|███████▎  | 4743/6470 [59:24<20:03,  1.44it/s]
2|4744|Loss: 0.7787936329841614:  73%|███████▎  | 4743/6470 [59:24<20:03,  1.44it/s]
2|4744|Loss: 0.7787936329841614:  73%|███████▎  | 4744/6470 [59:24<19:44,  1.46it/s]
2|4745|Loss: 0.9544517397880554:  73%|███████▎  | 4744/6470 [59:25<19:44,  1.46it/s]
2|4745|Loss: 0.9544517397880554:  73%|███████▎  | 4745/6470 [59:25<21:51,  1.32it/s]
2|4746|Loss: 0.9205240607261658:  73%|███████▎  | 4745/6470 [59:26<21:51,  1.32it/s]
2|4746|Loss: 0.9205240607261658:  73%|███████▎  | 4746/6470 [59:26<21:39,  1.33it/s]
2|4747|Loss: 0.800107479095459:  73%|███████▎  | 4746/6470 [59:27<21:39,  1.33it/s] 
2|4747|Loss: 0.800107479095459:  73%|███████▎  | 4747/6470 [59:27<21:05,  1.36it/s]
2|4748|Loss: 0.7906512022018433:  73%|███████▎  | 4747/6470 [59:27<21:05,  1.36it/s]
2|4748|Loss: 0.7906512022018433:  73%|███████▎  | 4748/6470 [59:28<21:07,  1.36it/s]
2|4749|Loss: 0.7403417825698853:  73%|███████▎  | 4748/6470 [59:28<21:07,  1.36it/s]
2|4749|Loss: 0.7403417825698853:  73%|███████▎  | 4749/6470 [59:28<21:17,  1.35it/s]
2|4750|Loss: 0.8063873052597046:  73%|███████▎  | 4749/6470 [59:29<21:17,  1.35it/s]
2|4750|Loss: 0.8063873052597046:  73%|███████▎  | 4750/6470 [59:29<22:25,  1.28it/s]
2|4751|Loss: 0.8783641457557678:  73%|███████▎  | 4750/6470 [59:30<22:25,  1.28it/s]
2|4751|Loss: 0.8783641457557678:  73%|███████▎  | 4751/6470 [59:30<22:34,  1.27it/s]
2|4752|Loss: 1.1772589683532715:  73%|███████▎  | 4751/6470 [59:30<22:34,  1.27it/s]
2|4752|Loss: 1.1772589683532715:  73%|███████▎  | 4752/6470 [59:31<22:37,  1.27it/s]
2|4753|Loss: 0.7434307932853699:  73%|███████▎  | 4752/6470 [59:31<22:37,  1.27it/s]
2|4753|Loss: 0.7434307932853699:  73%|███████▎  | 4753/6470 [59:31<21:19,  1.34it/s]
2|4754|Loss: 0.8154283761978149:  73%|███████▎  | 4753/6470 [59:32<21:19,  1.34it/s]
2|4754|Loss: 0.8154283761978149:  73%|███████▎  | 4754/6470 [59:32<21:57,  1.30it/s]
2|4755|Loss: 1.1767191886901855:  73%|███████▎  | 4754/6470 [59:33<21:57,  1.30it/s]
2|4755|Loss: 1.1767191886901855:  73%|███████▎  | 4755/6470 [59:33<21:23,  1.34it/s]
2|4756|Loss: 0.9323915839195251:  73%|███████▎  | 4755/6470 [59:33<21:23,  1.34it/s]
2|4756|Loss: 0.9323915839195251:  74%|███████▎  | 4756/6470 [59:34<20:44,  1.38it/s]
2|4757|Loss: 0.8809693455696106:  74%|███████▎  | 4756/6470 [59:34<20:44,  1.38it/s]
2|4757|Loss: 0.8809693455696106:  74%|███████▎  | 4757/6470 [59:34<21:03,  1.36it/s]
2|4758|Loss: 0.9439749717712402:  74%|███████▎  | 4757/6470 [59:35<21:03,  1.36it/s]
2|4758|Loss: 0.9439749717712402:  74%|███████▎  | 4758/6470 [59:35<20:54,  1.36it/s]
2|4759|Loss: 0.79407799243927:  74%|███████▎  | 4758/6470 [59:35<20:54,  1.36it/s]  
2|4759|Loss: 0.79407799243927:  74%|███████▎  | 4759/6470 [59:36<19:39,  1.45it/s]
2|4760|Loss: 0.9306923747062683:  74%|███████▎  | 4759/6470 [59:36<19:39,  1.45it/s]
2|4760|Loss: 0.9306923747062683:  74%|███████▎  | 4760/6470 [59:36<20:10,  1.41it/s]
2|4761|Loss: 0.9614923000335693:  74%|███████▎  | 4760/6470 [59:37<20:10,  1.41it/s]
2|4761|Loss: 0.9614923000335693:  74%|███████▎  | 4761/6470 [59:37<20:41,  1.38it/s]
2|4762|Loss: 0.7798323035240173:  74%|███████▎  | 4761/6470 [59:38<20:41,  1.38it/s]
2|4762|Loss: 0.7798323035240173:  74%|███████▎  | 4762/6470 [59:38<22:42,  1.25it/s]
2|4763|Loss: 0.798220157623291:  74%|███████▎  | 4762/6470 [59:39<22:42,  1.25it/s] 
2|4763|Loss: 0.798220157623291:  74%|███████▎  | 4763/6470 [59:39<23:28,  1.21it/s]
2|4764|Loss: 0.7741240859031677:  74%|███████▎  | 4763/6470 [59:39<23:28,  1.21it/s]
2|4764|Loss: 0.7741240859031677:  74%|███████▎  | 4764/6470 [59:40<22:14,  1.28it/s]
2|4765|Loss: 0.9014304280281067:  74%|███████▎  | 4764/6470 [59:40<22:14,  1.28it/s]
2|4765|Loss: 0.9014304280281067:  74%|███████▎  | 4765/6470 [59:41<21:55,  1.30it/s]
2|4766|Loss: 0.703805148601532:  74%|███████▎  | 4765/6470 [59:41<21:55,  1.30it/s] 
2|4766|Loss: 0.703805148601532:  74%|███████▎  | 4766/6470 [59:41<21:13,  1.34it/s]
2|4767|Loss: 0.7900362610816956:  74%|███████▎  | 4766/6470 [59:42<21:13,  1.34it/s]
2|4767|Loss: 0.7900362610816956:  74%|███████▎  | 4767/6470 [59:42<21:07,  1.34it/s]
2|4768|Loss: 0.8094282150268555:  74%|███████▎  | 4767/6470 [59:42<21:07,  1.34it/s]
2|4768|Loss: 0.8094282150268555:  74%|███████▎  | 4768/6470 [59:43<20:50,  1.36it/s]
2|4769|Loss: 0.8264074921607971:  74%|███████▎  | 4768/6470 [59:43<20:50,  1.36it/s]
2|4769|Loss: 0.8264074921607971:  74%|███████▎  | 4769/6470 [59:44<22:25,  1.26it/s]
2|4770|Loss: 0.6807906031608582:  74%|███████▎  | 4769/6470 [59:44<22:25,  1.26it/s]
2|4770|Loss: 0.6807906031608582:  74%|███████▎  | 4770/6470 [59:45<23:28,  1.21it/s]
2|4771|Loss: 0.8096253871917725:  74%|███████▎  | 4770/6470 [59:45<23:28,  1.21it/s]
2|4771|Loss: 0.8096253871917725:  74%|███████▎  | 4771/6470 [59:45<24:04,  1.18it/s]
2|4772|Loss: 0.7868247032165527:  74%|███████▎  | 4771/6470 [59:46<24:04,  1.18it/s]
2|4772|Loss: 0.7868247032165527:  74%|███████▍  | 4772/6470 [59:46<24:11,  1.17it/s]
2|4773|Loss: 0.8873438835144043:  74%|███████▍  | 4772/6470 [59:47<24:11,  1.17it/s]
2|4773|Loss: 0.8873438835144043:  74%|███████▍  | 4773/6470 [59:47<22:57,  1.23it/s]
2|4774|Loss: 0.7311785221099854:  74%|███████▍  | 4773/6470 [59:47<22:57,  1.23it/s]
2|4774|Loss: 0.7311785221099854:  74%|███████▍  | 4774/6470 [59:48<22:30,  1.26it/s]
2|4775|Loss: 0.7501142024993896:  74%|███████▍  | 4774/6470 [59:48<22:30,  1.26it/s]
2|4775|Loss: 0.7501142024993896:  74%|███████▍  | 4775/6470 [59:48<22:09,  1.27it/s]
2|4776|Loss: 0.9053070545196533:  74%|███████▍  | 4775/6470 [59:49<22:09,  1.27it/s]
2|4776|Loss: 0.9053070545196533:  74%|███████▍  | 4776/6470 [59:49<21:54,  1.29it/s]
2|4777|Loss: 0.7622580528259277:  74%|███████▍  | 4776/6470 [59:50<21:54,  1.29it/s]
2|4777|Loss: 0.7622580528259277:  74%|███████▍  | 4777/6470 [59:50<23:01,  1.23it/s]
2|4778|Loss: 0.8440830707550049:  74%|███████▍  | 4777/6470 [59:51<23:01,  1.23it/s]
2|4778|Loss: 0.8440830707550049:  74%|███████▍  | 4778/6470 [59:51<21:29,  1.31it/s]
2|4779|Loss: 0.8737490177154541:  74%|███████▍  | 4778/6470 [59:51<21:29,  1.31it/s]
2|4779|Loss: 0.8737490177154541:  74%|███████▍  | 4779/6470 [59:52<21:45,  1.30it/s]
2|4780|Loss: 0.8845964074134827:  74%|███████▍  | 4779/6470 [59:52<21:45,  1.30it/s]
2|4780|Loss: 0.8845964074134827:  74%|███████▍  | 4780/6470 [59:52<22:02,  1.28it/s]
2|4781|Loss: 0.7456235885620117:  74%|███████▍  | 4780/6470 [59:53<22:02,  1.28it/s]
2|4781|Loss: 0.7456235885620117:  74%|███████▍  | 4781/6470 [59:53<21:55,  1.28it/s]
2|4782|Loss: 0.855495035648346:  74%|███████▍  | 4781/6470 [59:54<21:55,  1.28it/s] 
2|4782|Loss: 0.855495035648346:  74%|███████▍  | 4782/6470 [59:54<21:06,  1.33it/s]
2|4783|Loss: 0.9950898289680481:  74%|███████▍  | 4782/6470 [59:54<21:06,  1.33it/s]
2|4783|Loss: 0.9950898289680481:  74%|███████▍  | 4783/6470 [59:55<21:09,  1.33it/s]
2|4784|Loss: 0.9776239395141602:  74%|███████▍  | 4783/6470 [59:55<21:09,  1.33it/s]
2|4784|Loss: 0.9776239395141602:  74%|███████▍  | 4784/6470 [59:55<21:28,  1.31it/s]
2|4785|Loss: 0.7710222601890564:  74%|███████▍  | 4784/6470 [59:56<21:28,  1.31it/s]
2|4785|Loss: 0.7710222601890564:  74%|███████▍  | 4785/6470 [59:56<21:39,  1.30it/s]
2|4786|Loss: 0.8647945523262024:  74%|███████▍  | 4785/6470 [59:57<21:39,  1.30it/s]
2|4786|Loss: 0.8647945523262024:  74%|███████▍  | 4786/6470 [59:57<21:47,  1.29it/s]
2|4787|Loss: 0.6186379790306091:  74%|███████▍  | 4786/6470 [59:57<21:47,  1.29it/s]
2|4787|Loss: 0.6186379790306091:  74%|███████▍  | 4787/6470 [59:58<23:31,  1.19it/s]
2|4788|Loss: 0.9298821687698364:  74%|███████▍  | 4787/6470 [59:58<23:31,  1.19it/s]
2|4788|Loss: 0.9298821687698364:  74%|███████▍  | 4788/6470 [59:59<22:43,  1.23it/s]
2|4789|Loss: 0.8012624979019165:  74%|███████▍  | 4788/6470 [59:59<22:43,  1.23it/s]
2|4789|Loss: 0.8012624979019165:  74%|███████▍  | 4789/6470 [1:00:00<23:02,  1.22it/s]
2|4790|Loss: 0.9692909121513367:  74%|███████▍  | 4789/6470 [1:00:00<23:02,  1.22it/s]
2|4790|Loss: 0.9692909121513367:  74%|███████▍  | 4790/6470 [1:00:00<23:16,  1.20it/s]
2|4791|Loss: 0.8933359980583191:  74%|███████▍  | 4790/6470 [1:00:01<23:16,  1.20it/s]
2|4791|Loss: 0.8933359980583191:  74%|███████▍  | 4791/6470 [1:00:01<21:41,  1.29it/s]
2|4792|Loss: 0.861393392086029:  74%|███████▍  | 4791/6470 [1:00:01<21:41,  1.29it/s] 
2|4792|Loss: 0.861393392086029:  74%|███████▍  | 4792/6470 [1:00:02<22:17,  1.25it/s]
2|4793|Loss: 0.7688702344894409:  74%|███████▍  | 4792/6470 [1:00:02<22:17,  1.25it/s]
2|4793|Loss: 0.7688702344894409:  74%|███████▍  | 4793/6470 [1:00:03<22:06,  1.26it/s]
2|4794|Loss: 0.9145870804786682:  74%|███████▍  | 4793/6470 [1:00:03<22:06,  1.26it/s]
2|4794|Loss: 0.9145870804786682:  74%|███████▍  | 4794/6470 [1:00:03<22:06,  1.26it/s]
2|4795|Loss: 0.9221619367599487:  74%|███████▍  | 4794/6470 [1:00:04<22:06,  1.26it/s]
2|4795|Loss: 0.9221619367599487:  74%|███████▍  | 4795/6470 [1:00:04<22:16,  1.25it/s]
2|4796|Loss: 0.8039702773094177:  74%|███████▍  | 4795/6470 [1:00:05<22:16,  1.25it/s]
2|4796|Loss: 0.8039702773094177:  74%|███████▍  | 4796/6470 [1:00:05<22:08,  1.26it/s]
2|4797|Loss: 0.7246541976928711:  74%|███████▍  | 4796/6470 [1:00:06<22:08,  1.26it/s]
2|4797|Loss: 0.7246541976928711:  74%|███████▍  | 4797/6470 [1:00:06<23:09,  1.20it/s]
2|4798|Loss: 0.6730203628540039:  74%|███████▍  | 4797/6470 [1:00:07<23:09,  1.20it/s]
2|4798|Loss: 0.6730203628540039:  74%|███████▍  | 4798/6470 [1:00:07<23:52,  1.17it/s]
2|4799|Loss: 0.7950745820999146:  74%|███████▍  | 4798/6470 [1:00:07<23:52,  1.17it/s]
2|4799|Loss: 0.7950745820999146:  74%|███████▍  | 4799/6470 [1:00:08<23:12,  1.20it/s]
2|4800|Loss: 0.9714024066925049:  74%|███████▍  | 4799/6470 [1:00:08<23:12,  1.20it/s]
2|4800|Loss: 0.9714024066925049:  74%|███████▍  | 4800/6470 [1:00:08<22:25,  1.24it/s]
2|4801|Loss: 0.9176040291786194:  74%|███████▍  | 4800/6470 [1:00:09<22:25,  1.24it/s]
2|4801|Loss: 0.9176040291786194:  74%|███████▍  | 4801/6470 [1:00:09<22:12,  1.25it/s]
2|4802|Loss: 0.7295041680335999:  74%|███████▍  | 4801/6470 [1:00:10<22:12,  1.25it/s]
2|4802|Loss: 0.7295041680335999:  74%|███████▍  | 4802/6470 [1:00:10<22:49,  1.22it/s]
2|4803|Loss: 0.7799603343009949:  74%|███████▍  | 4802/6470 [1:00:11<22:49,  1.22it/s]
2|4803|Loss: 0.7799603343009949:  74%|███████▍  | 4803/6470 [1:00:11<22:42,  1.22it/s]
2|4804|Loss: 0.7477118968963623:  74%|███████▍  | 4803/6470 [1:00:11<22:42,  1.22it/s]
2|4804|Loss: 0.7477118968963623:  74%|███████▍  | 4804/6470 [1:00:12<22:54,  1.21it/s]
2|4805|Loss: 0.9769265055656433:  74%|███████▍  | 4804/6470 [1:00:12<22:54,  1.21it/s]
2|4805|Loss: 0.9769265055656433:  74%|███████▍  | 4805/6470 [1:00:13<23:17,  1.19it/s]
2|4806|Loss: 0.9269195199012756:  74%|███████▍  | 4805/6470 [1:00:13<23:17,  1.19it/s]
2|4806|Loss: 0.9269195199012756:  74%|███████▍  | 4806/6470 [1:00:13<22:46,  1.22it/s]
2|4807|Loss: 0.7544034719467163:  74%|███████▍  | 4806/6470 [1:00:14<22:46,  1.22it/s]
2|4807|Loss: 0.7544034719467163:  74%|███████▍  | 4807/6470 [1:00:14<21:33,  1.29it/s]
2|4808|Loss: 0.8429084420204163:  74%|███████▍  | 4807/6470 [1:00:14<21:33,  1.29it/s]
2|4808|Loss: 0.8429084420204163:  74%|███████▍  | 4808/6470 [1:00:15<21:12,  1.31it/s]
2|4809|Loss: 0.9994418621063232:  74%|███████▍  | 4808/6470 [1:00:15<21:12,  1.31it/s]
2|4809|Loss: 0.9994418621063232:  74%|███████▍  | 4809/6470 [1:00:16<20:49,  1.33it/s]
2|4810|Loss: 0.9866780638694763:  74%|███████▍  | 4809/6470 [1:00:16<20:49,  1.33it/s]
2|4810|Loss: 0.9866780638694763:  74%|███████▍  | 4810/6470 [1:00:16<20:26,  1.35it/s]
2|4811|Loss: 0.8862478137016296:  74%|███████▍  | 4810/6470 [1:00:17<20:26,  1.35it/s]
2|4811|Loss: 0.8862478137016296:  74%|███████▍  | 4811/6470 [1:00:17<20:38,  1.34it/s]
2|4812|Loss: 1.1050894260406494:  74%|███████▍  | 4811/6470 [1:00:17<20:38,  1.34it/s]
2|4812|Loss: 1.1050894260406494:  74%|███████▍  | 4812/6470 [1:00:18<20:23,  1.36it/s]
2|4813|Loss: 0.9172005653381348:  74%|███████▍  | 4812/6470 [1:00:18<20:23,  1.36it/s]
2|4813|Loss: 0.9172005653381348:  74%|███████▍  | 4813/6470 [1:00:18<20:05,  1.37it/s]
2|4814|Loss: 0.8308444023132324:  74%|███████▍  | 4813/6470 [1:00:19<20:05,  1.37it/s]
2|4814|Loss: 0.8308444023132324:  74%|███████▍  | 4814/6470 [1:00:19<21:09,  1.30it/s]
2|4815|Loss: 0.8259444236755371:  74%|███████▍  | 4814/6470 [1:00:20<21:09,  1.30it/s]
2|4815|Loss: 0.8259444236755371:  74%|███████▍  | 4815/6470 [1:00:20<21:11,  1.30it/s]
2|4816|Loss: 0.8059585094451904:  74%|███████▍  | 4815/6470 [1:00:20<21:11,  1.30it/s]
2|4816|Loss: 0.8059585094451904:  74%|███████▍  | 4816/6470 [1:00:21<19:46,  1.39it/s]
2|4817|Loss: 0.8368903994560242:  74%|███████▍  | 4816/6470 [1:00:21<19:46,  1.39it/s]
2|4817|Loss: 0.8368903994560242:  74%|███████▍  | 4817/6470 [1:00:21<19:56,  1.38it/s]
2|4818|Loss: 0.890137791633606:  74%|███████▍  | 4817/6470 [1:00:22<19:56,  1.38it/s] 
2|4818|Loss: 0.890137791633606:  74%|███████▍  | 4818/6470 [1:00:22<19:21,  1.42it/s]
2|4819|Loss: 0.7760803699493408:  74%|███████▍  | 4818/6470 [1:00:22<19:21,  1.42it/s]
2|4819|Loss: 0.7760803699493408:  74%|███████▍  | 4819/6470 [1:00:23<19:38,  1.40it/s]
2|4820|Loss: 0.7810975313186646:  74%|███████▍  | 4819/6470 [1:00:23<19:38,  1.40it/s]
2|4820|Loss: 0.7810975313186646:  74%|███████▍  | 4820/6470 [1:00:23<18:09,  1.51it/s]
2|4821|Loss: 0.8314318656921387:  74%|███████▍  | 4820/6470 [1:00:24<18:09,  1.51it/s]
2|4821|Loss: 0.8314318656921387:  75%|███████▍  | 4821/6470 [1:00:24<18:31,  1.48it/s]
2|4822|Loss: 0.8683333396911621:  75%|███████▍  | 4821/6470 [1:00:24<18:31,  1.48it/s]
2|4822|Loss: 0.8683333396911621:  75%|███████▍  | 4822/6470 [1:00:25<18:39,  1.47it/s]
2|4823|Loss: 0.7501976490020752:  75%|███████▍  | 4822/6470 [1:00:25<18:39,  1.47it/s]
2|4823|Loss: 0.7501976490020752:  75%|███████▍  | 4823/6470 [1:00:25<18:13,  1.51it/s]
2|4824|Loss: 0.7649536728858948:  75%|███████▍  | 4823/6470 [1:00:26<18:13,  1.51it/s]
2|4824|Loss: 0.7649536728858948:  75%|███████▍  | 4824/6470 [1:00:26<18:10,  1.51it/s]
2|4825|Loss: 0.8710306882858276:  75%|███████▍  | 4824/6470 [1:00:26<18:10,  1.51it/s]
2|4825|Loss: 0.8710306882858276:  75%|███████▍  | 4825/6470 [1:00:27<18:15,  1.50it/s]
2|4826|Loss: 0.9195888638496399:  75%|███████▍  | 4825/6470 [1:00:27<18:15,  1.50it/s]
2|4826|Loss: 0.9195888638496399:  75%|███████▍  | 4826/6470 [1:00:27<18:50,  1.45it/s]
2|4827|Loss: 0.8983373641967773:  75%|███████▍  | 4826/6470 [1:00:28<18:50,  1.45it/s]
2|4827|Loss: 0.8983373641967773:  75%|███████▍  | 4827/6470 [1:00:28<19:30,  1.40it/s]
2|4828|Loss: 0.7875546813011169:  75%|███████▍  | 4827/6470 [1:00:29<19:30,  1.40it/s]
2|4828|Loss: 0.7875546813011169:  75%|███████▍  | 4828/6470 [1:00:29<19:17,  1.42it/s]
2|4829|Loss: 0.8275968432426453:  75%|███████▍  | 4828/6470 [1:00:29<19:17,  1.42it/s]
2|4829|Loss: 0.8275968432426453:  75%|███████▍  | 4829/6470 [1:00:30<20:46,  1.32it/s]
2|4830|Loss: 1.038399577140808:  75%|███████▍  | 4829/6470 [1:00:30<20:46,  1.32it/s] 
2|4830|Loss: 1.038399577140808:  75%|███████▍  | 4830/6470 [1:00:30<20:11,  1.35it/s]
2|4831|Loss: 0.7439103126525879:  75%|███████▍  | 4830/6470 [1:00:31<20:11,  1.35it/s]
2|4831|Loss: 0.7439103126525879:  75%|███████▍  | 4831/6470 [1:00:31<20:49,  1.31it/s]
2|4832|Loss: 0.8172452449798584:  75%|███████▍  | 4831/6470 [1:00:32<20:49,  1.31it/s]
2|4832|Loss: 0.8172452449798584:  75%|███████▍  | 4832/6470 [1:00:32<19:18,  1.41it/s]
2|4833|Loss: 0.8488293886184692:  75%|███████▍  | 4832/6470 [1:00:32<19:18,  1.41it/s]
2|4833|Loss: 0.8488293886184692:  75%|███████▍  | 4833/6470 [1:00:32<18:54,  1.44it/s]
2|4834|Loss: 0.8896388411521912:  75%|███████▍  | 4833/6470 [1:00:33<18:54,  1.44it/s]
2|4834|Loss: 0.8896388411521912:  75%|███████▍  | 4834/6470 [1:00:33<19:38,  1.39it/s]
2|4835|Loss: 0.7642090916633606:  75%|███████▍  | 4834/6470 [1:00:34<19:38,  1.39it/s]
2|4835|Loss: 0.7642090916633606:  75%|███████▍  | 4835/6470 [1:00:34<18:55,  1.44it/s]
2|4836|Loss: 0.8659526109695435:  75%|███████▍  | 4835/6470 [1:00:34<18:55,  1.44it/s]
2|4836|Loss: 0.8659526109695435:  75%|███████▍  | 4836/6470 [1:00:35<18:13,  1.49it/s]
2|4837|Loss: 0.7993391752243042:  75%|███████▍  | 4836/6470 [1:00:35<18:13,  1.49it/s]
2|4837|Loss: 0.7993391752243042:  75%|███████▍  | 4837/6470 [1:00:35<17:49,  1.53it/s]
2|4838|Loss: 0.7559617757797241:  75%|███████▍  | 4837/6470 [1:00:36<17:49,  1.53it/s]
2|4838|Loss: 0.7559617757797241:  75%|███████▍  | 4838/6470 [1:00:36<18:36,  1.46it/s]
2|4839|Loss: 0.9866495132446289:  75%|███████▍  | 4838/6470 [1:00:36<18:36,  1.46it/s]
2|4839|Loss: 0.9866495132446289:  75%|███████▍  | 4839/6470 [1:00:37<19:20,  1.41it/s]
2|4840|Loss: 0.8670130372047424:  75%|███████▍  | 4839/6470 [1:00:37<19:20,  1.41it/s]
2|4840|Loss: 0.8670130372047424:  75%|███████▍  | 4840/6470 [1:00:37<19:16,  1.41it/s]
2|4841|Loss: 0.8070564866065979:  75%|███████▍  | 4840/6470 [1:00:38<19:16,  1.41it/s]
2|4841|Loss: 0.8070564866065979:  75%|███████▍  | 4841/6470 [1:00:38<19:37,  1.38it/s]
2|4842|Loss: 0.8310269713401794:  75%|███████▍  | 4841/6470 [1:00:39<19:37,  1.38it/s]
2|4842|Loss: 0.8310269713401794:  75%|███████▍  | 4842/6470 [1:00:39<19:54,  1.36it/s]
2|4843|Loss: 0.7834008932113647:  75%|███████▍  | 4842/6470 [1:00:39<19:54,  1.36it/s]
2|4843|Loss: 0.7834008932113647:  75%|███████▍  | 4843/6470 [1:00:39<18:40,  1.45it/s]
2|4844|Loss: 0.7697194218635559:  75%|███████▍  | 4843/6470 [1:00:40<18:40,  1.45it/s]
2|4844|Loss: 0.7697194218635559:  75%|███████▍  | 4844/6470 [1:00:40<18:57,  1.43it/s]
2|4845|Loss: 0.7386990785598755:  75%|███████▍  | 4844/6470 [1:00:41<18:57,  1.43it/s]
2|4845|Loss: 0.7386990785598755:  75%|███████▍  | 4845/6470 [1:00:41<18:26,  1.47it/s]
2|4846|Loss: 0.6019225716590881:  75%|███████▍  | 4845/6470 [1:00:41<18:26,  1.47it/s]
2|4846|Loss: 0.6019225716590881:  75%|███████▍  | 4846/6470 [1:00:42<20:31,  1.32it/s]
2|4847|Loss: 0.7735478281974792:  75%|███████▍  | 4846/6470 [1:00:42<20:31,  1.32it/s]
2|4847|Loss: 0.7735478281974792:  75%|███████▍  | 4847/6470 [1:00:43<21:15,  1.27it/s]
2|4848|Loss: 0.8738606572151184:  75%|███████▍  | 4847/6470 [1:00:43<21:15,  1.27it/s]
2|4848|Loss: 0.8738606572151184:  75%|███████▍  | 4848/6470 [1:00:43<21:20,  1.27it/s]
2|4849|Loss: 0.910434365272522:  75%|███████▍  | 4848/6470 [1:00:44<21:20,  1.27it/s] 
2|4849|Loss: 0.910434365272522:  75%|███████▍  | 4849/6470 [1:00:44<20:48,  1.30it/s]
2|4850|Loss: 0.7881823778152466:  75%|███████▍  | 4849/6470 [1:00:45<20:48,  1.30it/s]
2|4850|Loss: 0.7881823778152466:  75%|███████▍  | 4850/6470 [1:00:45<20:01,  1.35it/s]
2|4851|Loss: 0.7037371397018433:  75%|███████▍  | 4850/6470 [1:00:45<20:01,  1.35it/s]
2|4851|Loss: 0.7037371397018433:  75%|███████▍  | 4851/6470 [1:00:45<18:44,  1.44it/s]
2|4852|Loss: 0.8287279009819031:  75%|███████▍  | 4851/6470 [1:00:46<18:44,  1.44it/s]
2|4852|Loss: 0.8287279009819031:  75%|███████▍  | 4852/6470 [1:00:46<18:33,  1.45it/s]
2|4853|Loss: 0.9393001198768616:  75%|███████▍  | 4852/6470 [1:00:46<18:33,  1.45it/s]
2|4853|Loss: 0.9393001198768616:  75%|███████▌  | 4853/6470 [1:00:47<18:28,  1.46it/s]
2|4854|Loss: 0.8845406770706177:  75%|███████▌  | 4853/6470 [1:00:47<18:28,  1.46it/s]
2|4854|Loss: 0.8845406770706177:  75%|███████▌  | 4854/6470 [1:00:48<19:07,  1.41it/s]
2|4855|Loss: 0.9840684533119202:  75%|███████▌  | 4854/6470 [1:00:48<19:07,  1.41it/s]
2|4855|Loss: 0.9840684533119202:  75%|███████▌  | 4855/6470 [1:00:48<19:27,  1.38it/s]
2|4856|Loss: 0.7513274550437927:  75%|███████▌  | 4855/6470 [1:00:49<19:27,  1.38it/s]
2|4856|Loss: 0.7513274550437927:  75%|███████▌  | 4856/6470 [1:00:49<18:07,  1.48it/s]
2|4857|Loss: 0.8764381408691406:  75%|███████▌  | 4856/6470 [1:00:49<18:07,  1.48it/s]
2|4857|Loss: 0.8764381408691406:  75%|███████▌  | 4857/6470 [1:00:49<17:43,  1.52it/s]
2|4858|Loss: 0.8425385355949402:  75%|███████▌  | 4857/6470 [1:00:50<17:43,  1.52it/s]
2|4858|Loss: 0.8425385355949402:  75%|███████▌  | 4858/6470 [1:00:50<17:59,  1.49it/s]
2|4859|Loss: 0.8733396530151367:  75%|███████▌  | 4858/6470 [1:00:51<17:59,  1.49it/s]
2|4859|Loss: 0.8733396530151367:  75%|███████▌  | 4859/6470 [1:00:51<19:17,  1.39it/s]
2|4860|Loss: 0.8672956824302673:  75%|███████▌  | 4859/6470 [1:00:51<19:17,  1.39it/s]
2|4860|Loss: 0.8672956824302673:  75%|███████▌  | 4860/6470 [1:00:52<19:42,  1.36it/s]
2|4861|Loss: 1.0451334714889526:  75%|███████▌  | 4860/6470 [1:00:52<19:42,  1.36it/s]
2|4861|Loss: 1.0451334714889526:  75%|███████▌  | 4861/6470 [1:00:53<20:15,  1.32it/s]
2|4862|Loss: 0.8548927307128906:  75%|███████▌  | 4861/6470 [1:00:53<20:15,  1.32it/s]
2|4862|Loss: 0.8548927307128906:  75%|███████▌  | 4862/6470 [1:00:53<20:29,  1.31it/s]
2|4863|Loss: 0.8363586068153381:  75%|███████▌  | 4862/6470 [1:00:54<20:29,  1.31it/s]
2|4863|Loss: 0.8363586068153381:  75%|███████▌  | 4863/6470 [1:00:54<21:22,  1.25it/s]
2|4864|Loss: 0.7869243621826172:  75%|███████▌  | 4863/6470 [1:00:55<21:22,  1.25it/s]
2|4864|Loss: 0.7869243621826172:  75%|███████▌  | 4864/6470 [1:00:55<22:13,  1.20it/s]
2|4865|Loss: 0.8912366032600403:  75%|███████▌  | 4864/6470 [1:00:56<22:13,  1.20it/s]
2|4865|Loss: 0.8912366032600403:  75%|███████▌  | 4865/6470 [1:00:56<22:05,  1.21it/s]
2|4866|Loss: 0.9112944006919861:  75%|███████▌  | 4865/6470 [1:00:56<22:05,  1.21it/s]
2|4866|Loss: 0.9112944006919861:  75%|███████▌  | 4866/6470 [1:00:57<21:41,  1.23it/s]
2|4867|Loss: 0.7338654398918152:  75%|███████▌  | 4866/6470 [1:00:57<21:41,  1.23it/s]
2|4867|Loss: 0.7338654398918152:  75%|███████▌  | 4867/6470 [1:00:58<22:37,  1.18it/s]
2|4868|Loss: 0.9236500859260559:  75%|███████▌  | 4867/6470 [1:00:58<22:37,  1.18it/s]
2|4868|Loss: 0.9236500859260559:  75%|███████▌  | 4868/6470 [1:00:58<22:27,  1.19it/s]
2|4869|Loss: 0.8492087125778198:  75%|███████▌  | 4868/6470 [1:00:59<22:27,  1.19it/s]
2|4869|Loss: 0.8492087125778198:  75%|███████▌  | 4869/6470 [1:00:59<22:04,  1.21it/s]
2|4870|Loss: 0.8132951855659485:  75%|███████▌  | 4869/6470 [1:01:00<22:04,  1.21it/s]
2|4870|Loss: 0.8132951855659485:  75%|███████▌  | 4870/6470 [1:01:00<22:20,  1.19it/s]
2|4871|Loss: 0.9600528478622437:  75%|███████▌  | 4870/6470 [1:01:01<22:20,  1.19it/s]
2|4871|Loss: 0.9600528478622437:  75%|███████▌  | 4871/6470 [1:01:01<21:48,  1.22it/s]
2|4872|Loss: 0.9295914769172668:  75%|███████▌  | 4871/6470 [1:01:01<21:48,  1.22it/s]
2|4872|Loss: 0.9295914769172668:  75%|███████▌  | 4872/6470 [1:01:02<21:36,  1.23it/s]
2|4873|Loss: 0.8066366910934448:  75%|███████▌  | 4872/6470 [1:01:02<21:36,  1.23it/s]
2|4873|Loss: 0.8066366910934448:  75%|███████▌  | 4873/6470 [1:01:02<20:42,  1.28it/s]
2|4874|Loss: 0.7610934376716614:  75%|███████▌  | 4873/6470 [1:01:03<20:42,  1.28it/s]
2|4874|Loss: 0.7610934376716614:  75%|███████▌  | 4874/6470 [1:01:03<18:59,  1.40it/s]
2|4875|Loss: 0.9177802801132202:  75%|███████▌  | 4874/6470 [1:01:03<18:59,  1.40it/s]
2|4875|Loss: 0.9177802801132202:  75%|███████▌  | 4875/6470 [1:01:04<19:17,  1.38it/s]
2|4876|Loss: 0.7898110151290894:  75%|███████▌  | 4875/6470 [1:01:04<19:17,  1.38it/s]
2|4876|Loss: 0.7898110151290894:  75%|███████▌  | 4876/6470 [1:01:05<19:54,  1.33it/s]
2|4877|Loss: 0.7910091280937195:  75%|███████▌  | 4876/6470 [1:01:05<19:54,  1.33it/s]
2|4877|Loss: 0.7910091280937195:  75%|███████▌  | 4877/6470 [1:01:05<18:19,  1.45it/s]
2|4878|Loss: 0.8569199442863464:  75%|███████▌  | 4877/6470 [1:01:05<18:19,  1.45it/s]
2|4878|Loss: 0.8569199442863464:  75%|███████▌  | 4878/6470 [1:01:06<17:27,  1.52it/s]
2|4879|Loss: 0.8170390129089355:  75%|███████▌  | 4878/6470 [1:01:06<17:27,  1.52it/s]
2|4879|Loss: 0.8170390129089355:  75%|███████▌  | 4879/6470 [1:01:06<18:02,  1.47it/s]
2|4880|Loss: 0.9698639512062073:  75%|███████▌  | 4879/6470 [1:01:07<18:02,  1.47it/s]
2|4880|Loss: 0.9698639512062073:  75%|███████▌  | 4880/6470 [1:01:07<17:34,  1.51it/s]
2|4881|Loss: 0.9564835429191589:  75%|███████▌  | 4880/6470 [1:01:07<17:34,  1.51it/s]
2|4881|Loss: 0.9564835429191589:  75%|███████▌  | 4881/6470 [1:01:08<16:40,  1.59it/s]
2|4882|Loss: 0.899470865726471:  75%|███████▌  | 4881/6470 [1:01:08<16:40,  1.59it/s] 
2|4882|Loss: 0.899470865726471:  75%|███████▌  | 4882/6470 [1:01:08<17:32,  1.51it/s]
2|4883|Loss: 0.8851014375686646:  75%|███████▌  | 4882/6470 [1:01:09<17:32,  1.51it/s]
2|4883|Loss: 0.8851014375686646:  75%|███████▌  | 4883/6470 [1:01:09<18:54,  1.40it/s]
2|4884|Loss: 0.8821751475334167:  75%|███████▌  | 4883/6470 [1:01:10<18:54,  1.40it/s]
2|4884|Loss: 0.8821751475334167:  75%|███████▌  | 4884/6470 [1:01:10<18:34,  1.42it/s]
2|4885|Loss: 0.7696369290351868:  75%|███████▌  | 4884/6470 [1:01:10<18:34,  1.42it/s]
2|4885|Loss: 0.7696369290351868:  76%|███████▌  | 4885/6470 [1:01:10<17:49,  1.48it/s]
2|4886|Loss: 0.8060910701751709:  76%|███████▌  | 4885/6470 [1:01:11<17:49,  1.48it/s]
2|4886|Loss: 0.8060910701751709:  76%|███████▌  | 4886/6470 [1:01:11<18:26,  1.43it/s]
2|4887|Loss: 0.8052595257759094:  76%|███████▌  | 4886/6470 [1:01:12<18:26,  1.43it/s]
2|4887|Loss: 0.8052595257759094:  76%|███████▌  | 4887/6470 [1:01:12<19:40,  1.34it/s]
2|4888|Loss: 0.7632978558540344:  76%|███████▌  | 4887/6470 [1:01:12<19:40,  1.34it/s]
2|4888|Loss: 0.7632978558540344:  76%|███████▌  | 4888/6470 [1:01:13<19:01,  1.39it/s]
2|4889|Loss: 0.9054649472236633:  76%|███████▌  | 4888/6470 [1:01:13<19:01,  1.39it/s]
2|4889|Loss: 0.9054649472236633:  76%|███████▌  | 4889/6470 [1:01:13<19:10,  1.37it/s]
2|4890|Loss: 0.7051971554756165:  76%|███████▌  | 4889/6470 [1:01:14<19:10,  1.37it/s]
2|4890|Loss: 0.7051971554756165:  76%|███████▌  | 4890/6470 [1:01:14<19:24,  1.36it/s]
2|4891|Loss: 0.7761937379837036:  76%|███████▌  | 4890/6470 [1:01:15<19:24,  1.36it/s]
2|4891|Loss: 0.7761937379837036:  76%|███████▌  | 4891/6470 [1:01:15<19:49,  1.33it/s]
2|4892|Loss: 0.8825966715812683:  76%|███████▌  | 4891/6470 [1:01:15<19:49,  1.33it/s]
2|4892|Loss: 0.8825966715812683:  76%|███████▌  | 4892/6470 [1:01:16<20:58,  1.25it/s]
2|4893|Loss: 0.9014318585395813:  76%|███████▌  | 4892/6470 [1:01:16<20:58,  1.25it/s]
2|4893|Loss: 0.9014318585395813:  76%|███████▌  | 4893/6470 [1:01:17<20:32,  1.28it/s]
2|4894|Loss: 0.8737438321113586:  76%|███████▌  | 4893/6470 [1:01:17<20:32,  1.28it/s]
2|4894|Loss: 0.8737438321113586:  76%|███████▌  | 4894/6470 [1:01:18<21:22,  1.23it/s]
2|4895|Loss: 1.0197436809539795:  76%|███████▌  | 4894/6470 [1:01:18<21:22,  1.23it/s]
2|4895|Loss: 1.0197436809539795:  76%|███████▌  | 4895/6470 [1:01:18<20:51,  1.26it/s]
2|4896|Loss: 0.6982472538948059:  76%|███████▌  | 4895/6470 [1:01:19<20:51,  1.26it/s]
2|4896|Loss: 0.6982472538948059:  76%|███████▌  | 4896/6470 [1:01:19<20:29,  1.28it/s]
2|4897|Loss: 0.6267945766448975:  76%|███████▌  | 4896/6470 [1:01:19<20:29,  1.28it/s]
2|4897|Loss: 0.6267945766448975:  76%|███████▌  | 4897/6470 [1:01:20<18:27,  1.42it/s]
2|4898|Loss: 0.7707905769348145:  76%|███████▌  | 4897/6470 [1:01:20<18:27,  1.42it/s]
2|4898|Loss: 0.7707905769348145:  76%|███████▌  | 4898/6470 [1:01:20<19:43,  1.33it/s]
2|4899|Loss: 0.8859349489212036:  76%|███████▌  | 4898/6470 [1:01:21<19:43,  1.33it/s]
2|4899|Loss: 0.8859349489212036:  76%|███████▌  | 4899/6470 [1:01:21<19:36,  1.34it/s]
2|4900|Loss: 0.9166949987411499:  76%|███████▌  | 4899/6470 [1:01:22<19:36,  1.34it/s]
2|4900|Loss: 0.9166949987411499:  76%|███████▌  | 4900/6470 [1:01:22<19:54,  1.31it/s]
2|4901|Loss: 1.004662275314331:  76%|███████▌  | 4900/6470 [1:01:22<19:54,  1.31it/s] 
2|4901|Loss: 1.004662275314331:  76%|███████▌  | 4901/6470 [1:01:23<20:03,  1.30it/s]
2|4902|Loss: 0.7203153371810913:  76%|███████▌  | 4901/6470 [1:01:23<20:03,  1.30it/s]
2|4902|Loss: 0.7203153371810913:  76%|███████▌  | 4902/6470 [1:01:23<19:26,  1.34it/s]
2|4903|Loss: 0.9015281796455383:  76%|███████▌  | 4902/6470 [1:01:24<19:26,  1.34it/s]
2|4903|Loss: 0.9015281796455383:  76%|███████▌  | 4903/6470 [1:01:24<18:33,  1.41it/s]
2|4904|Loss: 0.8317356109619141:  76%|███████▌  | 4903/6470 [1:01:24<18:33,  1.41it/s]
2|4904|Loss: 0.8317356109619141:  76%|███████▌  | 4904/6470 [1:01:25<18:12,  1.43it/s]
2|4905|Loss: 0.9512870907783508:  76%|███████▌  | 4904/6470 [1:01:25<18:12,  1.43it/s]
2|4905|Loss: 0.9512870907783508:  76%|███████▌  | 4905/6470 [1:01:26<19:14,  1.36it/s]
2|4906|Loss: 0.8502002358436584:  76%|███████▌  | 4905/6470 [1:01:26<19:14,  1.36it/s]
2|4906|Loss: 0.8502002358436584:  76%|███████▌  | 4906/6470 [1:01:26<17:48,  1.46it/s]
2|4907|Loss: 0.8754130005836487:  76%|███████▌  | 4906/6470 [1:01:26<17:48,  1.46it/s]
2|4907|Loss: 0.8754130005836487:  76%|███████▌  | 4907/6470 [1:01:27<16:11,  1.61it/s]
2|4908|Loss: 0.8231907486915588:  76%|███████▌  | 4907/6470 [1:01:27<16:11,  1.61it/s]
2|4908|Loss: 0.8231907486915588:  76%|███████▌  | 4908/6470 [1:01:27<16:53,  1.54it/s]
2|4909|Loss: 0.8562598824501038:  76%|███████▌  | 4908/6470 [1:01:28<16:53,  1.54it/s]
2|4909|Loss: 0.8562598824501038:  76%|███████▌  | 4909/6470 [1:01:28<18:56,  1.37it/s]
2|4910|Loss: 0.8747846484184265:  76%|███████▌  | 4909/6470 [1:01:29<18:56,  1.37it/s]
2|4910|Loss: 0.8747846484184265:  76%|███████▌  | 4910/6470 [1:01:29<19:45,  1.32it/s]
2|4911|Loss: 0.7637821435928345:  76%|███████▌  | 4910/6470 [1:01:29<19:45,  1.32it/s]
2|4911|Loss: 0.7637821435928345:  76%|███████▌  | 4911/6470 [1:01:30<19:21,  1.34it/s]
2|4912|Loss: 0.9416435956954956:  76%|███████▌  | 4911/6470 [1:01:30<19:21,  1.34it/s]
2|4912|Loss: 0.9416435956954956:  76%|███████▌  | 4912/6470 [1:01:31<20:06,  1.29it/s]
2|4913|Loss: 0.8437186479568481:  76%|███████▌  | 4912/6470 [1:01:31<20:06,  1.29it/s]
2|4913|Loss: 0.8437186479568481:  76%|███████▌  | 4913/6470 [1:01:31<18:54,  1.37it/s]
2|4914|Loss: 0.8087334632873535:  76%|███████▌  | 4913/6470 [1:01:32<18:54,  1.37it/s]
2|4914|Loss: 0.8087334632873535:  76%|███████▌  | 4914/6470 [1:01:32<18:44,  1.38it/s]
2|4915|Loss: 0.8583865165710449:  76%|███████▌  | 4914/6470 [1:01:32<18:44,  1.38it/s]
2|4915|Loss: 0.8583865165710449:  76%|███████▌  | 4915/6470 [1:01:33<18:39,  1.39it/s]
2|4916|Loss: 0.9951568245887756:  76%|███████▌  | 4915/6470 [1:01:33<18:39,  1.39it/s]
2|4916|Loss: 0.9951568245887756:  76%|███████▌  | 4916/6470 [1:01:33<19:02,  1.36it/s]
2|4917|Loss: 0.7912819385528564:  76%|███████▌  | 4916/6470 [1:01:34<19:02,  1.36it/s]
2|4917|Loss: 0.7912819385528564:  76%|███████▌  | 4917/6470 [1:01:34<19:16,  1.34it/s]
2|4918|Loss: 0.7742061018943787:  76%|███████▌  | 4917/6470 [1:01:35<19:16,  1.34it/s]
2|4918|Loss: 0.7742061018943787:  76%|███████▌  | 4918/6470 [1:01:35<18:07,  1.43it/s]
2|4919|Loss: 0.9255543947219849:  76%|███████▌  | 4918/6470 [1:01:35<18:07,  1.43it/s]
2|4919|Loss: 0.9255543947219849:  76%|███████▌  | 4919/6470 [1:01:35<18:06,  1.43it/s]
2|4920|Loss: 0.8981073498725891:  76%|███████▌  | 4919/6470 [1:01:36<18:06,  1.43it/s]
2|4920|Loss: 0.8981073498725891:  76%|███████▌  | 4920/6470 [1:01:36<18:05,  1.43it/s]
2|4921|Loss: 0.857464611530304:  76%|███████▌  | 4920/6470 [1:01:37<18:05,  1.43it/s] 
2|4921|Loss: 0.857464611530304:  76%|███████▌  | 4921/6470 [1:01:37<18:10,  1.42it/s]
2|4922|Loss: 0.9799984693527222:  76%|███████▌  | 4921/6470 [1:01:37<18:10,  1.42it/s]
2|4922|Loss: 0.9799984693527222:  76%|███████▌  | 4922/6470 [1:01:38<18:06,  1.43it/s]
2|4923|Loss: 0.9287985563278198:  76%|███████▌  | 4922/6470 [1:01:38<18:06,  1.43it/s]
2|4923|Loss: 0.9287985563278198:  76%|███████▌  | 4923/6470 [1:01:38<19:01,  1.36it/s]
2|4924|Loss: 0.9669324159622192:  76%|███████▌  | 4923/6470 [1:01:39<19:01,  1.36it/s]
2|4924|Loss: 0.9669324159622192:  76%|███████▌  | 4924/6470 [1:01:39<18:45,  1.37it/s]
2|4925|Loss: 0.8479824662208557:  76%|███████▌  | 4924/6470 [1:01:40<18:45,  1.37it/s]
2|4925|Loss: 0.8479824662208557:  76%|███████▌  | 4925/6470 [1:01:40<19:06,  1.35it/s]
2|4926|Loss: 0.7711178660392761:  76%|███████▌  | 4925/6470 [1:01:40<19:06,  1.35it/s]
2|4926|Loss: 0.7711178660392761:  76%|███████▌  | 4926/6470 [1:01:41<19:17,  1.33it/s]
2|4927|Loss: 0.993720293045044:  76%|███████▌  | 4926/6470 [1:01:41<19:17,  1.33it/s] 
2|4927|Loss: 0.993720293045044:  76%|███████▌  | 4927/6470 [1:01:41<19:45,  1.30it/s]
2|4928|Loss: 0.807990312576294:  76%|███████▌  | 4927/6470 [1:01:42<19:45,  1.30it/s]
2|4928|Loss: 0.807990312576294:  76%|███████▌  | 4928/6470 [1:01:42<20:02,  1.28it/s]
2|4929|Loss: 0.8573111295700073:  76%|███████▌  | 4928/6470 [1:01:43<20:02,  1.28it/s]
2|4929|Loss: 0.8573111295700073:  76%|███████▌  | 4929/6470 [1:01:43<19:50,  1.29it/s]
2|4930|Loss: 0.8021647930145264:  76%|███████▌  | 4929/6470 [1:01:43<19:50,  1.29it/s]
2|4930|Loss: 0.8021647930145264:  76%|███████▌  | 4930/6470 [1:01:44<19:50,  1.29it/s]
2|4931|Loss: 0.8206039071083069:  76%|███████▌  | 4930/6470 [1:01:44<19:50,  1.29it/s]
2|4931|Loss: 0.8206039071083069:  76%|███████▌  | 4931/6470 [1:01:45<20:21,  1.26it/s]
2|4932|Loss: 0.8787869811058044:  76%|███████▌  | 4931/6470 [1:01:45<20:21,  1.26it/s]
2|4932|Loss: 0.8787869811058044:  76%|███████▌  | 4932/6470 [1:01:45<20:30,  1.25it/s]
2|4933|Loss: 0.9423974752426147:  76%|███████▌  | 4932/6470 [1:01:46<20:30,  1.25it/s]
2|4933|Loss: 0.9423974752426147:  76%|███████▌  | 4933/6470 [1:01:46<19:00,  1.35it/s]
2|4934|Loss: 0.956387460231781:  76%|███████▌  | 4933/6470 [1:01:46<19:00,  1.35it/s] 
2|4934|Loss: 0.956387460231781:  76%|███████▋  | 4934/6470 [1:01:47<19:23,  1.32it/s]
2|4935|Loss: 0.8494346737861633:  76%|███████▋  | 4934/6470 [1:01:47<19:23,  1.32it/s]
2|4935|Loss: 0.8494346737861633:  76%|███████▋  | 4935/6470 [1:01:48<19:28,  1.31it/s]
2|4936|Loss: 0.7339327931404114:  76%|███████▋  | 4935/6470 [1:01:48<19:28,  1.31it/s]
2|4936|Loss: 0.7339327931404114:  76%|███████▋  | 4936/6470 [1:01:48<18:27,  1.39it/s]
2|4937|Loss: 0.827908992767334:  76%|███████▋  | 4936/6470 [1:01:49<18:27,  1.39it/s] 
2|4937|Loss: 0.827908992767334:  76%|███████▋  | 4937/6470 [1:01:49<18:03,  1.42it/s]
2|4938|Loss: 0.8712232112884521:  76%|███████▋  | 4937/6470 [1:01:49<18:03,  1.42it/s]
2|4938|Loss: 0.8712232112884521:  76%|███████▋  | 4938/6470 [1:01:50<17:30,  1.46it/s]
2|4939|Loss: 0.9587132334709167:  76%|███████▋  | 4938/6470 [1:01:50<17:30,  1.46it/s]
2|4939|Loss: 0.9587132334709167:  76%|███████▋  | 4939/6470 [1:01:50<18:45,  1.36it/s]
2|4940|Loss: 0.9005603790283203:  76%|███████▋  | 4939/6470 [1:01:51<18:45,  1.36it/s]
2|4940|Loss: 0.9005603790283203:  76%|███████▋  | 4940/6470 [1:01:51<20:00,  1.27it/s]
2|4941|Loss: 0.9899070262908936:  76%|███████▋  | 4940/6470 [1:01:52<20:00,  1.27it/s]
2|4941|Loss: 0.9899070262908936:  76%|███████▋  | 4941/6470 [1:01:52<20:09,  1.26it/s]
2|4942|Loss: 0.818803608417511:  76%|███████▋  | 4941/6470 [1:01:53<20:09,  1.26it/s] 
2|4942|Loss: 0.818803608417511:  76%|███████▋  | 4942/6470 [1:01:53<20:14,  1.26it/s]
2|4943|Loss: 0.8752619624137878:  76%|███████▋  | 4942/6470 [1:01:53<20:14,  1.26it/s]
2|4943|Loss: 0.8752619624137878:  76%|███████▋  | 4943/6470 [1:01:54<19:43,  1.29it/s]
2|4944|Loss: 0.9016711711883545:  76%|███████▋  | 4943/6470 [1:01:54<19:43,  1.29it/s]
2|4944|Loss: 0.9016711711883545:  76%|███████▋  | 4944/6470 [1:01:54<19:08,  1.33it/s]
2|4945|Loss: 0.8369337916374207:  76%|███████▋  | 4944/6470 [1:01:55<19:08,  1.33it/s]
2|4945|Loss: 0.8369337916374207:  76%|███████▋  | 4945/6470 [1:01:55<19:13,  1.32it/s]
2|4946|Loss: 0.8465930223464966:  76%|███████▋  | 4945/6470 [1:01:56<19:13,  1.32it/s]
2|4946|Loss: 0.8465930223464966:  76%|███████▋  | 4946/6470 [1:01:56<19:27,  1.31it/s]
2|4947|Loss: 0.6893565654754639:  76%|███████▋  | 4946/6470 [1:01:56<19:27,  1.31it/s]
2|4947|Loss: 0.6893565654754639:  76%|███████▋  | 4947/6470 [1:01:57<19:23,  1.31it/s]
2|4948|Loss: 0.8907246589660645:  76%|███████▋  | 4947/6470 [1:01:57<19:23,  1.31it/s]
2|4948|Loss: 0.8907246589660645:  76%|███████▋  | 4948/6470 [1:01:57<18:39,  1.36it/s]
2|4949|Loss: 0.8855355381965637:  76%|███████▋  | 4948/6470 [1:01:58<18:39,  1.36it/s]
2|4949|Loss: 0.8855355381965637:  76%|███████▋  | 4949/6470 [1:01:58<17:20,  1.46it/s]
2|4950|Loss: 0.8207606077194214:  76%|███████▋  | 4949/6470 [1:01:58<17:20,  1.46it/s]
2|4950|Loss: 0.8207606077194214:  77%|███████▋  | 4950/6470 [1:01:59<18:16,  1.39it/s]
2|4951|Loss: 0.9239212870597839:  77%|███████▋  | 4950/6470 [1:01:59<18:16,  1.39it/s]
2|4951|Loss: 0.9239212870597839:  77%|███████▋  | 4951/6470 [1:01:59<18:18,  1.38it/s]
2|4952|Loss: 0.8437256217002869:  77%|███████▋  | 4951/6470 [1:02:00<18:18,  1.38it/s]
2|4952|Loss: 0.8437256217002869:  77%|███████▋  | 4952/6470 [1:02:00<19:11,  1.32it/s]
2|4953|Loss: 0.718923807144165:  77%|███████▋  | 4952/6470 [1:02:01<19:11,  1.32it/s] 
2|4953|Loss: 0.718923807144165:  77%|███████▋  | 4953/6470 [1:02:01<20:08,  1.26it/s]
2|4954|Loss: 0.857321560382843:  77%|███████▋  | 4953/6470 [1:02:02<20:08,  1.26it/s]
2|4954|Loss: 0.857321560382843:  77%|███████▋  | 4954/6470 [1:02:02<20:41,  1.22it/s]
2|4955|Loss: 0.8741884231567383:  77%|███████▋  | 4954/6470 [1:02:03<20:41,  1.22it/s]
2|4955|Loss: 0.8741884231567383:  77%|███████▋  | 4955/6470 [1:02:03<20:39,  1.22it/s]
2|4956|Loss: 0.8474920392036438:  77%|███████▋  | 4955/6470 [1:02:03<20:39,  1.22it/s]
2|4956|Loss: 0.8474920392036438:  77%|███████▋  | 4956/6470 [1:02:04<19:37,  1.29it/s]
2|4957|Loss: 0.7874093055725098:  77%|███████▋  | 4956/6470 [1:02:04<19:37,  1.29it/s]
2|4957|Loss: 0.7874093055725098:  77%|███████▋  | 4957/6470 [1:02:04<19:24,  1.30it/s]
2|4958|Loss: 0.7968920469284058:  77%|███████▋  | 4957/6470 [1:02:05<19:24,  1.30it/s]
2|4958|Loss: 0.7968920469284058:  77%|███████▋  | 4958/6470 [1:02:05<18:34,  1.36it/s]
2|4959|Loss: 0.9045554399490356:  77%|███████▋  | 4958/6470 [1:02:05<18:34,  1.36it/s]
2|4959|Loss: 0.9045554399490356:  77%|███████▋  | 4959/6470 [1:02:06<18:41,  1.35it/s]
2|4960|Loss: 0.9043083786964417:  77%|███████▋  | 4959/6470 [1:02:06<18:41,  1.35it/s]
2|4960|Loss: 0.9043083786964417:  77%|███████▋  | 4960/6470 [1:02:07<19:03,  1.32it/s]
2|4961|Loss: 0.7804726958274841:  77%|███████▋  | 4960/6470 [1:02:07<19:03,  1.32it/s]
2|4961|Loss: 0.7804726958274841:  77%|███████▋  | 4961/6470 [1:02:07<19:58,  1.26it/s]
2|4962|Loss: 0.8584126830101013:  77%|███████▋  | 4961/6470 [1:02:08<19:58,  1.26it/s]
2|4962|Loss: 0.8584126830101013:  77%|███████▋  | 4962/6470 [1:02:08<19:59,  1.26it/s]
2|4963|Loss: 0.783423662185669:  77%|███████▋  | 4962/6470 [1:02:09<19:59,  1.26it/s] 
2|4963|Loss: 0.783423662185669:  77%|███████▋  | 4963/6470 [1:02:09<19:40,  1.28it/s]
2|4964|Loss: 0.8013580441474915:  77%|███████▋  | 4963/6470 [1:02:09<19:40,  1.28it/s]
2|4964|Loss: 0.8013580441474915:  77%|███████▋  | 4964/6470 [1:02:09<17:55,  1.40it/s]
2|4965|Loss: 0.9117403030395508:  77%|███████▋  | 4964/6470 [1:02:10<17:55,  1.40it/s]
2|4965|Loss: 0.9117403030395508:  77%|███████▋  | 4965/6470 [1:02:10<18:23,  1.36it/s]
2|4966|Loss: 0.9410051703453064:  77%|███████▋  | 4965/6470 [1:02:11<18:23,  1.36it/s]
2|4966|Loss: 0.9410051703453064:  77%|███████▋  | 4966/6470 [1:02:11<18:07,  1.38it/s]
2|4967|Loss: 0.952094554901123:  77%|███████▋  | 4966/6470 [1:02:11<18:07,  1.38it/s] 
2|4967|Loss: 0.952094554901123:  77%|███████▋  | 4967/6470 [1:02:12<17:22,  1.44it/s]
2|4968|Loss: 0.8248766660690308:  77%|███████▋  | 4967/6470 [1:02:12<17:22,  1.44it/s]
2|4968|Loss: 0.8248766660690308:  77%|███████▋  | 4968/6470 [1:02:12<17:34,  1.42it/s]
2|4969|Loss: 0.9084812998771667:  77%|███████▋  | 4968/6470 [1:02:13<17:34,  1.42it/s]
2|4969|Loss: 0.9084812998771667:  77%|███████▋  | 4969/6470 [1:02:13<18:52,  1.33it/s]
2|4970|Loss: 0.922512948513031:  77%|███████▋  | 4969/6470 [1:02:14<18:52,  1.33it/s] 
2|4970|Loss: 0.922512948513031:  77%|███████▋  | 4970/6470 [1:02:14<19:15,  1.30it/s]
2|4971|Loss: 0.7366715669631958:  77%|███████▋  | 4970/6470 [1:02:14<19:15,  1.30it/s]
2|4971|Loss: 0.7366715669631958:  77%|███████▋  | 4971/6470 [1:02:15<19:19,  1.29it/s]
2|4972|Loss: 0.8359376788139343:  77%|███████▋  | 4971/6470 [1:02:15<19:19,  1.29it/s]
2|4972|Loss: 0.8359376788139343:  77%|███████▋  | 4972/6470 [1:02:15<18:41,  1.34it/s]
2|4973|Loss: 0.873212993144989:  77%|███████▋  | 4972/6470 [1:02:16<18:41,  1.34it/s] 
2|4973|Loss: 0.873212993144989:  77%|███████▋  | 4973/6470 [1:02:16<17:33,  1.42it/s]
2|4974|Loss: 0.8725907802581787:  77%|███████▋  | 4973/6470 [1:02:16<17:33,  1.42it/s]
2|4974|Loss: 0.8725907802581787:  77%|███████▋  | 4974/6470 [1:02:17<17:21,  1.44it/s]
2|4975|Loss: 0.9021410942077637:  77%|███████▋  | 4974/6470 [1:02:17<17:21,  1.44it/s]
2|4975|Loss: 0.9021410942077637:  77%|███████▋  | 4975/6470 [1:02:17<17:28,  1.43it/s]
2|4976|Loss: 0.817116916179657:  77%|███████▋  | 4975/6470 [1:02:18<17:28,  1.43it/s] 
2|4976|Loss: 0.817116916179657:  77%|███████▋  | 4976/6470 [1:02:18<17:13,  1.45it/s]
2|4977|Loss: 0.8500648140907288:  77%|███████▋  | 4976/6470 [1:02:19<17:13,  1.45it/s]
2|4977|Loss: 0.8500648140907288:  77%|███████▋  | 4977/6470 [1:02:19<17:02,  1.46it/s]
2|4978|Loss: 0.743071436882019:  77%|███████▋  | 4977/6470 [1:02:19<17:02,  1.46it/s] 
2|4978|Loss: 0.743071436882019:  77%|███████▋  | 4978/6470 [1:02:20<17:22,  1.43it/s]
2|4979|Loss: 0.8784776329994202:  77%|███████▋  | 4978/6470 [1:02:20<17:22,  1.43it/s]
2|4979|Loss: 0.8784776329994202:  77%|███████▋  | 4979/6470 [1:02:20<17:27,  1.42it/s]
2|4980|Loss: 0.9080807566642761:  77%|███████▋  | 4979/6470 [1:02:21<17:27,  1.42it/s]
2|4980|Loss: 0.9080807566642761:  77%|███████▋  | 4980/6470 [1:02:21<18:07,  1.37it/s]
2|4981|Loss: 0.7671803832054138:  77%|███████▋  | 4980/6470 [1:02:21<18:07,  1.37it/s]
2|4981|Loss: 0.7671803832054138:  77%|███████▋  | 4981/6470 [1:02:22<18:29,  1.34it/s]
2|4982|Loss: 0.7285637259483337:  77%|███████▋  | 4981/6470 [1:02:22<18:29,  1.34it/s]
2|4982|Loss: 0.7285637259483337:  77%|███████▋  | 4982/6470 [1:02:22<17:46,  1.40it/s]
2|4983|Loss: 1.0186147689819336:  77%|███████▋  | 4982/6470 [1:02:23<17:46,  1.40it/s]
2|4983|Loss: 1.0186147689819336:  77%|███████▋  | 4983/6470 [1:02:23<18:54,  1.31it/s]
2|4984|Loss: 0.8157541751861572:  77%|███████▋  | 4983/6470 [1:02:24<18:54,  1.31it/s]
2|4984|Loss: 0.8157541751861572:  77%|███████▋  | 4984/6470 [1:02:24<19:23,  1.28it/s]
2|4985|Loss: 0.9080187678337097:  77%|███████▋  | 4984/6470 [1:02:25<19:23,  1.28it/s]
2|4985|Loss: 0.9080187678337097:  77%|███████▋  | 4985/6470 [1:02:25<18:54,  1.31it/s]
2|4986|Loss: 0.9653364419937134:  77%|███████▋  | 4985/6470 [1:02:25<18:54,  1.31it/s]
2|4986|Loss: 0.9653364419937134:  77%|███████▋  | 4986/6470 [1:02:26<18:59,  1.30it/s]
2|4987|Loss: 0.7690845727920532:  77%|███████▋  | 4986/6470 [1:02:26<18:59,  1.30it/s]
2|4987|Loss: 0.7690845727920532:  77%|███████▋  | 4987/6470 [1:02:26<16:45,  1.48it/s]
2|4988|Loss: 0.8000232577323914:  77%|███████▋  | 4987/6470 [1:02:26<16:45,  1.48it/s]
2|4988|Loss: 0.8000232577323914:  77%|███████▋  | 4988/6470 [1:02:27<16:26,  1.50it/s]
2|4989|Loss: 0.825630247592926:  77%|███████▋  | 4988/6470 [1:02:27<16:26,  1.50it/s] 
2|4989|Loss: 0.825630247592926:  77%|███████▋  | 4989/6470 [1:02:27<16:32,  1.49it/s]
2|4990|Loss: 0.6901627779006958:  77%|███████▋  | 4989/6470 [1:02:28<16:32,  1.49it/s]
2|4990|Loss: 0.6901627779006958:  77%|███████▋  | 4990/6470 [1:02:28<17:35,  1.40it/s]
2|4991|Loss: 0.6883572340011597:  77%|███████▋  | 4990/6470 [1:02:29<17:35,  1.40it/s]
2|4991|Loss: 0.6883572340011597:  77%|███████▋  | 4991/6470 [1:02:29<18:20,  1.34it/s]
2|4992|Loss: 0.8802273869514465:  77%|███████▋  | 4991/6470 [1:02:30<18:20,  1.34it/s]
2|4992|Loss: 0.8802273869514465:  77%|███████▋  | 4992/6470 [1:02:30<18:45,  1.31it/s]
2|4993|Loss: 0.6899948716163635:  77%|███████▋  | 4992/6470 [1:02:30<18:45,  1.31it/s]
2|4993|Loss: 0.6899948716163635:  77%|███████▋  | 4993/6470 [1:02:31<18:58,  1.30it/s]
2|4994|Loss: 0.72445148229599:  77%|███████▋  | 4993/6470 [1:02:31<18:58,  1.30it/s]  
2|4994|Loss: 0.72445148229599:  77%|███████▋  | 4994/6470 [1:02:31<18:25,  1.34it/s]
2|4995|Loss: 0.9016712307929993:  77%|███████▋  | 4994/6470 [1:02:32<18:25,  1.34it/s]
2|4995|Loss: 0.9016712307929993:  77%|███████▋  | 4995/6470 [1:02:32<18:45,  1.31it/s]
2|4996|Loss: 0.7474424839019775:  77%|███████▋  | 4995/6470 [1:02:33<18:45,  1.31it/s]
2|4996|Loss: 0.7474424839019775:  77%|███████▋  | 4996/6470 [1:02:33<20:01,  1.23it/s]
2|4997|Loss: 0.7415731549263:  77%|███████▋  | 4996/6470 [1:02:34<20:01,  1.23it/s]   
2|4997|Loss: 0.7415731549263:  77%|███████▋  | 4997/6470 [1:02:34<19:25,  1.26it/s]
2|4998|Loss: 0.7965747117996216:  77%|███████▋  | 4997/6470 [1:02:34<19:25,  1.26it/s]
2|4998|Loss: 0.7965747117996216:  77%|███████▋  | 4998/6470 [1:02:35<19:01,  1.29it/s]
2|4999|Loss: 0.8805597424507141:  77%|███████▋  | 4998/6470 [1:02:35<19:01,  1.29it/s]
2|4999|Loss: 0.8805597424507141:  77%|███████▋  | 4999/6470 [1:02:35<17:45,  1.38it/s]
2|5000|Loss: 0.862737774848938:  77%|███████▋  | 4999/6470 [1:02:36<17:45,  1.38it/s] 
2|5000|Loss: 0.862737774848938:  77%|███████▋  | 5000/6470 [1:02:36<17:39,  1.39it/s]
2|5001|Loss: 0.7772390246391296:  77%|███████▋  | 5000/6470 [1:02:36<17:39,  1.39it/s]
2|5001|Loss: 0.7772390246391296:  77%|███████▋  | 5001/6470 [1:02:37<18:01,  1.36it/s]
2|5002|Loss: 0.8026897311210632:  77%|███████▋  | 5001/6470 [1:02:37<18:01,  1.36it/s]
2|5002|Loss: 0.8026897311210632:  77%|███████▋  | 5002/6470 [1:02:37<17:50,  1.37it/s]
2|5003|Loss: 0.9246241450309753:  77%|███████▋  | 5002/6470 [1:02:38<17:50,  1.37it/s]
2|5003|Loss: 0.9246241450309753:  77%|███████▋  | 5003/6470 [1:02:38<17:36,  1.39it/s]
2|5004|Loss: 0.8485782742500305:  77%|███████▋  | 5003/6470 [1:02:38<17:36,  1.39it/s]
2|5004|Loss: 0.8485782742500305:  77%|███████▋  | 5004/6470 [1:02:39<17:25,  1.40it/s]
2|5005|Loss: 0.8610608577728271:  77%|███████▋  | 5004/6470 [1:02:39<17:25,  1.40it/s]
2|5005|Loss: 0.8610608577728271:  77%|███████▋  | 5005/6470 [1:02:39<16:13,  1.50it/s]
2|5006|Loss: 0.7001781463623047:  77%|███████▋  | 5005/6470 [1:02:40<16:13,  1.50it/s]
2|5006|Loss: 0.7001781463623047:  77%|███████▋  | 5006/6470 [1:02:40<15:14,  1.60it/s]
2|5007|Loss: 0.8282850384712219:  77%|███████▋  | 5006/6470 [1:02:40<15:14,  1.60it/s]
2|5007|Loss: 0.8282850384712219:  77%|███████▋  | 5007/6470 [1:02:41<16:34,  1.47it/s]
2|5008|Loss: 1.0445737838745117:  77%|███████▋  | 5007/6470 [1:02:41<16:34,  1.47it/s]
2|5008|Loss: 1.0445737838745117:  77%|███████▋  | 5008/6470 [1:02:41<17:21,  1.40it/s]
2|5009|Loss: 0.6972962021827698:  77%|███████▋  | 5008/6470 [1:02:42<17:21,  1.40it/s]
2|5009|Loss: 0.6972962021827698:  77%|███████▋  | 5009/6470 [1:02:42<17:11,  1.42it/s]
2|5010|Loss: 0.830302894115448:  77%|███████▋  | 5009/6470 [1:02:43<17:11,  1.42it/s] 
2|5010|Loss: 0.830302894115448:  77%|███████▋  | 5010/6470 [1:02:43<17:20,  1.40it/s]
2|5011|Loss: 0.7650769948959351:  77%|███████▋  | 5010/6470 [1:02:43<17:20,  1.40it/s]
2|5011|Loss: 0.7650769948959351:  77%|███████▋  | 5011/6470 [1:02:44<18:08,  1.34it/s]
2|5012|Loss: 0.9176012873649597:  77%|███████▋  | 5011/6470 [1:02:44<18:08,  1.34it/s]
2|5012|Loss: 0.9176012873649597:  77%|███████▋  | 5012/6470 [1:02:44<17:36,  1.38it/s]
2|5013|Loss: 0.8550764322280884:  77%|███████▋  | 5012/6470 [1:02:45<17:36,  1.38it/s]
2|5013|Loss: 0.8550764322280884:  77%|███████▋  | 5013/6470 [1:02:45<17:52,  1.36it/s]
2|5014|Loss: 1.0131796598434448:  77%|███████▋  | 5013/6470 [1:02:46<17:52,  1.36it/s]
2|5014|Loss: 1.0131796598434448:  77%|███████▋  | 5014/6470 [1:02:46<18:42,  1.30it/s]
2|5015|Loss: 0.9180414080619812:  77%|███████▋  | 5014/6470 [1:02:46<18:42,  1.30it/s]
2|5015|Loss: 0.9180414080619812:  78%|███████▊  | 5015/6470 [1:02:47<18:30,  1.31it/s]
2|5016|Loss: 0.9097293615341187:  78%|███████▊  | 5015/6470 [1:02:47<18:30,  1.31it/s]
2|5016|Loss: 0.9097293615341187:  78%|███████▊  | 5016/6470 [1:02:47<17:24,  1.39it/s]
2|5017|Loss: 0.7708858847618103:  78%|███████▊  | 5016/6470 [1:02:48<17:24,  1.39it/s]
2|5017|Loss: 0.7708858847618103:  78%|███████▊  | 5017/6470 [1:02:48<17:43,  1.37it/s]
2|5018|Loss: 0.7178987264633179:  78%|███████▊  | 5017/6470 [1:02:49<17:43,  1.37it/s]
2|5018|Loss: 0.7178987264633179:  78%|███████▊  | 5018/6470 [1:02:49<17:56,  1.35it/s]
2|5019|Loss: 0.8961877822875977:  78%|███████▊  | 5018/6470 [1:02:49<17:56,  1.35it/s]
2|5019|Loss: 0.8961877822875977:  78%|███████▊  | 5019/6470 [1:02:50<19:26,  1.24it/s]
2|5020|Loss: 0.8310284614562988:  78%|███████▊  | 5019/6470 [1:02:50<19:26,  1.24it/s]
2|5020|Loss: 0.8310284614562988:  78%|███████▊  | 5020/6470 [1:02:51<18:57,  1.28it/s]
2|5021|Loss: 1.0120680332183838:  78%|███████▊  | 5020/6470 [1:02:51<18:57,  1.28it/s]
2|5021|Loss: 1.0120680332183838:  78%|███████▊  | 5021/6470 [1:02:51<19:28,  1.24it/s]
2|5022|Loss: 0.7617574334144592:  78%|███████▊  | 5021/6470 [1:02:52<19:28,  1.24it/s]
2|5022|Loss: 0.7617574334144592:  78%|███████▊  | 5022/6470 [1:02:52<18:57,  1.27it/s]
2|5023|Loss: 0.8743279576301575:  78%|███████▊  | 5022/6470 [1:02:53<18:57,  1.27it/s]
2|5023|Loss: 0.8743279576301575:  78%|███████▊  | 5023/6470 [1:02:53<19:32,  1.23it/s]
2|5024|Loss: 0.846110463142395:  78%|███████▊  | 5023/6470 [1:02:54<19:32,  1.23it/s] 
2|5024|Loss: 0.846110463142395:  78%|███████▊  | 5024/6470 [1:02:54<20:27,  1.18it/s]
2|5025|Loss: 0.7906792759895325:  78%|███████▊  | 5024/6470 [1:02:54<20:27,  1.18it/s]
2|5025|Loss: 0.7906792759895325:  78%|███████▊  | 5025/6470 [1:02:55<20:20,  1.18it/s]
2|5026|Loss: 0.8312254548072815:  78%|███████▊  | 5025/6470 [1:02:55<20:20,  1.18it/s]
2|5026|Loss: 0.8312254548072815:  78%|███████▊  | 5026/6470 [1:02:55<19:16,  1.25it/s]
2|5027|Loss: 0.7702563405036926:  78%|███████▊  | 5026/6470 [1:02:56<19:16,  1.25it/s]
2|5027|Loss: 0.7702563405036926:  78%|███████▊  | 5027/6470 [1:02:56<18:05,  1.33it/s]
2|5028|Loss: 0.8055622577667236:  78%|███████▊  | 5027/6470 [1:02:57<18:05,  1.33it/s]
2|5028|Loss: 0.8055622577667236:  78%|███████▊  | 5028/6470 [1:02:57<19:44,  1.22it/s]
2|5029|Loss: 0.8145577311515808:  78%|███████▊  | 5028/6470 [1:02:58<19:44,  1.22it/s]
2|5029|Loss: 0.8145577311515808:  78%|███████▊  | 5029/6470 [1:02:58<20:03,  1.20it/s]
2|5030|Loss: 0.7040392160415649:  78%|███████▊  | 5029/6470 [1:02:58<20:03,  1.20it/s]
2|5030|Loss: 0.7040392160415649:  78%|███████▊  | 5030/6470 [1:02:59<19:54,  1.21it/s]
2|5031|Loss: 0.8036149144172668:  78%|███████▊  | 5030/6470 [1:02:59<19:54,  1.21it/s]
2|5031|Loss: 0.8036149144172668:  78%|███████▊  | 5031/6470 [1:03:00<20:40,  1.16it/s]
2|5032|Loss: 0.8494653105735779:  78%|███████▊  | 5031/6470 [1:03:00<20:40,  1.16it/s]
2|5032|Loss: 0.8494653105735779:  78%|███████▊  | 5032/6470 [1:03:01<20:54,  1.15it/s]
2|5033|Loss: 0.784317672252655:  78%|███████▊  | 5032/6470 [1:03:01<20:54,  1.15it/s] 
2|5033|Loss: 0.784317672252655:  78%|███████▊  | 5033/6470 [1:03:01<18:57,  1.26it/s]
2|5034|Loss: 1.0083993673324585:  78%|███████▊  | 5033/6470 [1:03:02<18:57,  1.26it/s]
2|5034|Loss: 1.0083993673324585:  78%|███████▊  | 5034/6470 [1:03:02<18:03,  1.33it/s]
2|5035|Loss: 0.9329510927200317:  78%|███████▊  | 5034/6470 [1:03:02<18:03,  1.33it/s]
2|5035|Loss: 0.9329510927200317:  78%|███████▊  | 5035/6470 [1:03:02<16:49,  1.42it/s]
2|5036|Loss: 0.8222607970237732:  78%|███████▊  | 5035/6470 [1:03:03<16:49,  1.42it/s]
2|5036|Loss: 0.8222607970237732:  78%|███████▊  | 5036/6470 [1:03:03<18:01,  1.33it/s]
2|5037|Loss: 0.7734849452972412:  78%|███████▊  | 5036/6470 [1:03:04<18:01,  1.33it/s]
2|5037|Loss: 0.7734849452972412:  78%|███████▊  | 5037/6470 [1:03:04<18:38,  1.28it/s]
2|5038|Loss: 0.8713219165802002:  78%|███████▊  | 5037/6470 [1:03:05<18:38,  1.28it/s]
2|5038|Loss: 0.8713219165802002:  78%|███████▊  | 5038/6470 [1:03:05<17:39,  1.35it/s]
2|5039|Loss: 0.7127314209938049:  78%|███████▊  | 5038/6470 [1:03:05<17:39,  1.35it/s]
2|5039|Loss: 0.7127314209938049:  78%|███████▊  | 5039/6470 [1:03:06<18:33,  1.29it/s]
2|5040|Loss: 0.7302070260047913:  78%|███████▊  | 5039/6470 [1:03:06<18:33,  1.29it/s]
2|5040|Loss: 0.7302070260047913:  78%|███████▊  | 5040/6470 [1:03:07<19:15,  1.24it/s]
2|5041|Loss: 0.8735337853431702:  78%|███████▊  | 5040/6470 [1:03:07<19:15,  1.24it/s]
2|5041|Loss: 0.8735337853431702:  78%|███████▊  | 5041/6470 [1:03:07<18:39,  1.28it/s]
2|5042|Loss: 0.809927761554718:  78%|███████▊  | 5041/6470 [1:03:08<18:39,  1.28it/s] 
2|5042|Loss: 0.809927761554718:  78%|███████▊  | 5042/6470 [1:03:08<17:46,  1.34it/s]
2|5043|Loss: 1.0633100271224976:  78%|███████▊  | 5042/6470 [1:03:08<17:46,  1.34it/s]
2|5043|Loss: 1.0633100271224976:  78%|███████▊  | 5043/6470 [1:03:09<17:21,  1.37it/s]
2|5044|Loss: 0.7453432083129883:  78%|███████▊  | 5043/6470 [1:03:09<17:21,  1.37it/s]
2|5044|Loss: 0.7453432083129883:  78%|███████▊  | 5044/6470 [1:03:09<17:59,  1.32it/s]
2|5045|Loss: 0.8355294466018677:  78%|███████▊  | 5044/6470 [1:03:10<17:59,  1.32it/s]
2|5045|Loss: 0.8355294466018677:  78%|███████▊  | 5045/6470 [1:03:10<18:15,  1.30it/s]
2|5046|Loss: 0.9559263586997986:  78%|███████▊  | 5045/6470 [1:03:11<18:15,  1.30it/s]
2|5046|Loss: 0.9559263586997986:  78%|███████▊  | 5046/6470 [1:03:11<18:18,  1.30it/s]
2|5047|Loss: 0.925595760345459:  78%|███████▊  | 5046/6470 [1:03:11<18:18,  1.30it/s] 
2|5047|Loss: 0.925595760345459:  78%|███████▊  | 5047/6470 [1:03:12<18:11,  1.30it/s]
2|5048|Loss: 0.8936383724212646:  78%|███████▊  | 5047/6470 [1:03:12<18:11,  1.30it/s]
2|5048|Loss: 0.8936383724212646:  78%|███████▊  | 5048/6470 [1:03:13<18:39,  1.27it/s]
2|5049|Loss: 0.7709762454032898:  78%|███████▊  | 5048/6470 [1:03:13<18:39,  1.27it/s]
2|5049|Loss: 0.7709762454032898:  78%|███████▊  | 5049/6470 [1:03:13<18:07,  1.31it/s]
2|5050|Loss: 0.6555354595184326:  78%|███████▊  | 5049/6470 [1:03:14<18:07,  1.31it/s]
2|5050|Loss: 0.6555354595184326:  78%|███████▊  | 5050/6470 [1:03:14<18:47,  1.26it/s]
2|5051|Loss: 0.7924290299415588:  78%|███████▊  | 5050/6470 [1:03:15<18:47,  1.26it/s]
2|5051|Loss: 0.7924290299415588:  78%|███████▊  | 5051/6470 [1:03:15<17:57,  1.32it/s]
2|5052|Loss: 0.672137439250946:  78%|███████▊  | 5051/6470 [1:03:15<17:57,  1.32it/s] 
2|5052|Loss: 0.672137439250946:  78%|███████▊  | 5052/6470 [1:03:16<17:23,  1.36it/s]
2|5053|Loss: 0.736598551273346:  78%|███████▊  | 5052/6470 [1:03:16<17:23,  1.36it/s]
2|5053|Loss: 0.736598551273346:  78%|███████▊  | 5053/6470 [1:03:16<17:29,  1.35it/s]
2|5054|Loss: 0.822411060333252:  78%|███████▊  | 5053/6470 [1:03:17<17:29,  1.35it/s]
2|5054|Loss: 0.822411060333252:  78%|███████▊  | 5054/6470 [1:03:17<16:51,  1.40it/s]
2|5055|Loss: 0.8796018958091736:  78%|███████▊  | 5054/6470 [1:03:17<16:51,  1.40it/s]
2|5055|Loss: 0.8796018958091736:  78%|███████▊  | 5055/6470 [1:03:18<17:06,  1.38it/s]
2|5056|Loss: 0.9283251762390137:  78%|███████▊  | 5055/6470 [1:03:18<17:06,  1.38it/s]
2|5056|Loss: 0.9283251762390137:  78%|███████▊  | 5056/6470 [1:03:19<17:17,  1.36it/s]
2|5057|Loss: 0.8726768493652344:  78%|███████▊  | 5056/6470 [1:03:19<17:17,  1.36it/s]
2|5057|Loss: 0.8726768493652344:  78%|███████▊  | 5057/6470 [1:03:19<17:01,  1.38it/s]
2|5058|Loss: 0.8896175026893616:  78%|███████▊  | 5057/6470 [1:03:20<17:01,  1.38it/s]
2|5058|Loss: 0.8896175026893616:  78%|███████▊  | 5058/6470 [1:03:20<17:07,  1.37it/s]
2|5059|Loss: 0.7889219522476196:  78%|███████▊  | 5058/6470 [1:03:20<17:07,  1.37it/s]
2|5059|Loss: 0.7889219522476196:  78%|███████▊  | 5059/6470 [1:03:20<15:16,  1.54it/s]
2|5060|Loss: 0.9101374745368958:  78%|███████▊  | 5059/6470 [1:03:21<15:16,  1.54it/s]
2|5060|Loss: 0.9101374745368958:  78%|███████▊  | 5060/6470 [1:03:21<15:12,  1.55it/s]
2|5061|Loss: 0.7713148593902588:  78%|███████▊  | 5060/6470 [1:03:21<15:12,  1.55it/s]
2|5061|Loss: 0.7713148593902588:  78%|███████▊  | 5061/6470 [1:03:22<16:03,  1.46it/s]
2|5062|Loss: 0.8229798674583435:  78%|███████▊  | 5061/6470 [1:03:22<16:03,  1.46it/s]
2|5062|Loss: 0.8229798674583435:  78%|███████▊  | 5062/6470 [1:03:23<17:46,  1.32it/s]
2|5063|Loss: 0.8655268549919128:  78%|███████▊  | 5062/6470 [1:03:23<17:46,  1.32it/s]
2|5063|Loss: 0.8655268549919128:  78%|███████▊  | 5063/6470 [1:03:23<16:59,  1.38it/s]
2|5064|Loss: 0.8295391201972961:  78%|███████▊  | 5063/6470 [1:03:24<16:59,  1.38it/s]
2|5064|Loss: 0.8295391201972961:  78%|███████▊  | 5064/6470 [1:03:24<16:19,  1.44it/s]
2|5065|Loss: 0.8084506392478943:  78%|███████▊  | 5064/6470 [1:03:24<16:19,  1.44it/s]
2|5065|Loss: 0.8084506392478943:  78%|███████▊  | 5065/6470 [1:03:25<16:20,  1.43it/s]
2|5066|Loss: 0.9285132884979248:  78%|███████▊  | 5065/6470 [1:03:25<16:20,  1.43it/s]
2|5066|Loss: 0.9285132884979248:  78%|███████▊  | 5066/6470 [1:03:25<16:11,  1.45it/s]
2|5067|Loss: 0.9059962034225464:  78%|███████▊  | 5066/6470 [1:03:26<16:11,  1.45it/s]
2|5067|Loss: 0.9059962034225464:  78%|███████▊  | 5067/6470 [1:03:26<17:07,  1.36it/s]
2|5068|Loss: 0.7721557021141052:  78%|███████▊  | 5067/6470 [1:03:27<17:07,  1.36it/s]
2|5068|Loss: 0.7721557021141052:  78%|███████▊  | 5068/6470 [1:03:27<16:13,  1.44it/s]
2|5069|Loss: 0.8771466612815857:  78%|███████▊  | 5068/6470 [1:03:27<16:13,  1.44it/s]
2|5069|Loss: 0.8771466612815857:  78%|███████▊  | 5069/6470 [1:03:28<17:18,  1.35it/s]
2|5070|Loss: 0.9614875912666321:  78%|███████▊  | 5069/6470 [1:03:28<17:18,  1.35it/s]
2|5070|Loss: 0.9614875912666321:  78%|███████▊  | 5070/6470 [1:03:28<17:11,  1.36it/s]
2|5071|Loss: 0.8991552591323853:  78%|███████▊  | 5070/6470 [1:03:29<17:11,  1.36it/s]
2|5071|Loss: 0.8991552591323853:  78%|███████▊  | 5071/6470 [1:03:29<17:11,  1.36it/s]
2|5072|Loss: 0.9043854475021362:  78%|███████▊  | 5071/6470 [1:03:30<17:11,  1.36it/s]
2|5072|Loss: 0.9043854475021362:  78%|███████▊  | 5072/6470 [1:03:30<17:25,  1.34it/s]
2|5073|Loss: 0.8856384754180908:  78%|███████▊  | 5072/6470 [1:03:30<17:25,  1.34it/s]
2|5073|Loss: 0.8856384754180908:  78%|███████▊  | 5073/6470 [1:03:31<17:10,  1.36it/s]
2|5074|Loss: 0.7914364337921143:  78%|███████▊  | 5073/6470 [1:03:31<17:10,  1.36it/s]
2|5074|Loss: 0.7914364337921143:  78%|███████▊  | 5074/6470 [1:03:31<15:56,  1.46it/s]
2|5075|Loss: 0.9643917083740234:  78%|███████▊  | 5074/6470 [1:03:32<15:56,  1.46it/s]
2|5075|Loss: 0.9643917083740234:  78%|███████▊  | 5075/6470 [1:03:32<14:54,  1.56it/s]
2|5076|Loss: 0.8459134697914124:  78%|███████▊  | 5075/6470 [1:03:32<14:54,  1.56it/s]
2|5076|Loss: 0.8459134697914124:  78%|███████▊  | 5076/6470 [1:03:32<15:40,  1.48it/s]
2|5077|Loss: 1.0230238437652588:  78%|███████▊  | 5076/6470 [1:03:33<15:40,  1.48it/s]
2|5077|Loss: 1.0230238437652588:  78%|███████▊  | 5077/6470 [1:03:33<15:31,  1.50it/s]
2|5078|Loss: 0.8229261636734009:  78%|███████▊  | 5077/6470 [1:03:34<15:31,  1.50it/s]
2|5078|Loss: 0.8229261636734009:  78%|███████▊  | 5078/6470 [1:03:34<15:25,  1.50it/s]
2|5079|Loss: 0.8545246720314026:  78%|███████▊  | 5078/6470 [1:03:34<15:25,  1.50it/s]
2|5079|Loss: 0.8545246720314026:  79%|███████▊  | 5079/6470 [1:03:35<16:09,  1.44it/s]
2|5080|Loss: 0.787642240524292:  79%|███████▊  | 5079/6470 [1:03:35<16:09,  1.44it/s] 
2|5080|Loss: 0.787642240524292:  79%|███████▊  | 5080/6470 [1:03:35<16:17,  1.42it/s]
2|5081|Loss: 0.9420515298843384:  79%|███████▊  | 5080/6470 [1:03:36<16:17,  1.42it/s]
2|5081|Loss: 0.9420515298843384:  79%|███████▊  | 5081/6470 [1:03:36<17:07,  1.35it/s]
2|5082|Loss: 0.6937680840492249:  79%|███████▊  | 5081/6470 [1:03:37<17:07,  1.35it/s]
2|5082|Loss: 0.6937680840492249:  79%|███████▊  | 5082/6470 [1:03:37<16:18,  1.42it/s]
2|5083|Loss: 0.8898434638977051:  79%|███████▊  | 5082/6470 [1:03:37<16:18,  1.42it/s]
2|5083|Loss: 0.8898434638977051:  79%|███████▊  | 5083/6470 [1:03:38<16:40,  1.39it/s]
2|5084|Loss: 0.829055666923523:  79%|███████▊  | 5083/6470 [1:03:38<16:40,  1.39it/s] 
2|5084|Loss: 0.829055666923523:  79%|███████▊  | 5084/6470 [1:03:38<17:47,  1.30it/s]
2|5085|Loss: 0.8498629927635193:  79%|███████▊  | 5084/6470 [1:03:39<17:47,  1.30it/s]
2|5085|Loss: 0.8498629927635193:  79%|███████▊  | 5085/6470 [1:03:39<17:23,  1.33it/s]
2|5086|Loss: 0.7709912657737732:  79%|███████▊  | 5085/6470 [1:03:39<17:23,  1.33it/s]
2|5086|Loss: 0.7709912657737732:  79%|███████▊  | 5086/6470 [1:03:40<16:59,  1.36it/s]
2|5087|Loss: 0.6926615834236145:  79%|███████▊  | 5086/6470 [1:03:40<16:59,  1.36it/s]
2|5087|Loss: 0.6926615834236145:  79%|███████▊  | 5087/6470 [1:03:41<18:11,  1.27it/s]
2|5088|Loss: 0.8328187465667725:  79%|███████▊  | 5087/6470 [1:03:41<18:11,  1.27it/s]
2|5088|Loss: 0.8328187465667725:  79%|███████▊  | 5088/6470 [1:03:42<19:15,  1.20it/s]
2|5089|Loss: 0.7935718894004822:  79%|███████▊  | 5088/6470 [1:03:42<19:15,  1.20it/s]
2|5089|Loss: 0.7935718894004822:  79%|███████▊  | 5089/6470 [1:03:42<18:42,  1.23it/s]
2|5090|Loss: 0.8550789952278137:  79%|███████▊  | 5089/6470 [1:03:43<18:42,  1.23it/s]
2|5090|Loss: 0.8550789952278137:  79%|███████▊  | 5090/6470 [1:03:43<18:46,  1.22it/s]
2|5091|Loss: 0.8019638061523438:  79%|███████▊  | 5090/6470 [1:03:44<18:46,  1.22it/s]
2|5091|Loss: 0.8019638061523438:  79%|███████▊  | 5091/6470 [1:03:44<18:28,  1.24it/s]
2|5092|Loss: 1.0011675357818604:  79%|███████▊  | 5091/6470 [1:03:44<18:28,  1.24it/s]
2|5092|Loss: 1.0011675357818604:  79%|███████▊  | 5092/6470 [1:03:45<17:39,  1.30it/s]
2|5093|Loss: 0.8023515939712524:  79%|███████▊  | 5092/6470 [1:03:45<17:39,  1.30it/s]
2|5093|Loss: 0.8023515939712524:  79%|███████▊  | 5093/6470 [1:03:46<18:05,  1.27it/s]
2|5094|Loss: 0.8962220549583435:  79%|███████▊  | 5093/6470 [1:03:46<18:05,  1.27it/s]
2|5094|Loss: 0.8962220549583435:  79%|███████▊  | 5094/6470 [1:03:46<17:47,  1.29it/s]
2|5095|Loss: 0.8651940822601318:  79%|███████▊  | 5094/6470 [1:03:47<17:47,  1.29it/s]
2|5095|Loss: 0.8651940822601318:  79%|███████▊  | 5095/6470 [1:03:47<17:52,  1.28it/s]
2|5096|Loss: 0.8880074620246887:  79%|███████▊  | 5095/6470 [1:03:48<17:52,  1.28it/s]
2|5096|Loss: 0.8880074620246887:  79%|███████▉  | 5096/6470 [1:03:48<18:01,  1.27it/s]
2|5097|Loss: 0.8060206174850464:  79%|███████▉  | 5096/6470 [1:03:48<18:01,  1.27it/s]
2|5097|Loss: 0.8060206174850464:  79%|███████▉  | 5097/6470 [1:03:48<16:44,  1.37it/s]
2|5098|Loss: 0.7582518458366394:  79%|███████▉  | 5097/6470 [1:03:49<16:44,  1.37it/s]
2|5098|Loss: 0.7582518458366394:  79%|███████▉  | 5098/6470 [1:03:49<16:42,  1.37it/s]
2|5099|Loss: 0.9946654438972473:  79%|███████▉  | 5098/6470 [1:03:50<16:42,  1.37it/s]
2|5099|Loss: 0.9946654438972473:  79%|███████▉  | 5099/6470 [1:03:50<16:34,  1.38it/s]
2|5100|Loss: 0.667834997177124:  79%|███████▉  | 5099/6470 [1:03:50<16:34,  1.38it/s] 
2|5100|Loss: 0.667834997177124:  79%|███████▉  | 5100/6470 [1:03:50<15:09,  1.51it/s]
2|5101|Loss: 0.9347485303878784:  79%|███████▉  | 5100/6470 [1:03:51<15:09,  1.51it/s]
2|5101|Loss: 0.9347485303878784:  79%|███████▉  | 5101/6470 [1:03:51<15:31,  1.47it/s]
2|5102|Loss: 0.8835614323616028:  79%|███████▉  | 5101/6470 [1:03:52<15:31,  1.47it/s]
2|5102|Loss: 0.8835614323616028:  79%|███████▉  | 5102/6470 [1:03:52<15:36,  1.46it/s]
2|5103|Loss: 0.9307017922401428:  79%|███████▉  | 5102/6470 [1:03:52<15:36,  1.46it/s]
2|5103|Loss: 0.9307017922401428:  79%|███████▉  | 5103/6470 [1:03:53<17:02,  1.34it/s]
2|5104|Loss: 0.7537690997123718:  79%|███████▉  | 5103/6470 [1:03:53<17:02,  1.34it/s]
2|5104|Loss: 0.7537690997123718:  79%|███████▉  | 5104/6470 [1:03:54<17:21,  1.31it/s]
2|5105|Loss: 0.7533126473426819:  79%|███████▉  | 5104/6470 [1:03:54<17:21,  1.31it/s]
2|5105|Loss: 0.7533126473426819:  79%|███████▉  | 5105/6470 [1:03:54<16:59,  1.34it/s]
2|5106|Loss: 0.9512300491333008:  79%|███████▉  | 5105/6470 [1:03:55<16:59,  1.34it/s]
2|5106|Loss: 0.9512300491333008:  79%|███████▉  | 5106/6470 [1:03:55<16:16,  1.40it/s]
2|5107|Loss: 0.935316801071167:  79%|███████▉  | 5106/6470 [1:03:55<16:16,  1.40it/s] 
2|5107|Loss: 0.935316801071167:  79%|███████▉  | 5107/6470 [1:03:56<17:14,  1.32it/s]
2|5108|Loss: 0.9481461048126221:  79%|███████▉  | 5107/6470 [1:03:56<17:14,  1.32it/s]
2|5108|Loss: 0.9481461048126221:  79%|███████▉  | 5108/6470 [1:03:56<15:29,  1.47it/s]
2|5109|Loss: 0.8132150769233704:  79%|███████▉  | 5108/6470 [1:03:57<15:29,  1.47it/s]
2|5109|Loss: 0.8132150769233704:  79%|███████▉  | 5109/6470 [1:03:57<14:20,  1.58it/s]
2|5110|Loss: 0.8946624398231506:  79%|███████▉  | 5109/6470 [1:03:57<14:20,  1.58it/s]
2|5110|Loss: 0.8946624398231506:  79%|███████▉  | 5110/6470 [1:03:57<14:00,  1.62it/s]
2|5111|Loss: 0.8714565634727478:  79%|███████▉  | 5110/6470 [1:03:58<14:00,  1.62it/s]
2|5111|Loss: 0.8714565634727478:  79%|███████▉  | 5111/6470 [1:03:58<14:44,  1.54it/s]
2|5112|Loss: 0.7927839756011963:  79%|███████▉  | 5111/6470 [1:03:58<14:44,  1.54it/s]
2|5112|Loss: 0.7927839756011963:  79%|███████▉  | 5112/6470 [1:03:59<14:07,  1.60it/s]
2|5113|Loss: 0.9638473987579346:  79%|███████▉  | 5112/6470 [1:03:59<14:07,  1.60it/s]
2|5113|Loss: 0.9638473987579346:  79%|███████▉  | 5113/6470 [1:03:59<15:00,  1.51it/s]
2|5114|Loss: 0.848408579826355:  79%|███████▉  | 5113/6470 [1:04:00<15:00,  1.51it/s] 
2|5114|Loss: 0.848408579826355:  79%|███████▉  | 5114/6470 [1:04:00<15:32,  1.45it/s]
2|5115|Loss: 0.8294493556022644:  79%|███████▉  | 5114/6470 [1:04:01<15:32,  1.45it/s]
2|5115|Loss: 0.8294493556022644:  79%|███████▉  | 5115/6470 [1:04:01<16:32,  1.37it/s]
2|5116|Loss: 0.8737596273422241:  79%|███████▉  | 5115/6470 [1:04:01<16:32,  1.37it/s]
2|5116|Loss: 0.8737596273422241:  79%|███████▉  | 5116/6470 [1:04:02<16:30,  1.37it/s]
2|5117|Loss: 0.8031664490699768:  79%|███████▉  | 5116/6470 [1:04:02<16:30,  1.37it/s]
2|5117|Loss: 0.8031664490699768:  79%|███████▉  | 5117/6470 [1:04:02<15:48,  1.43it/s]
2|5118|Loss: 0.7573948502540588:  79%|███████▉  | 5117/6470 [1:04:03<15:48,  1.43it/s]
2|5118|Loss: 0.7573948502540588:  79%|███████▉  | 5118/6470 [1:04:03<15:26,  1.46it/s]
2|5119|Loss: 0.7783001065254211:  79%|███████▉  | 5118/6470 [1:04:03<15:26,  1.46it/s]
2|5119|Loss: 0.7783001065254211:  79%|███████▉  | 5119/6470 [1:04:04<15:57,  1.41it/s]
2|5120|Loss: 0.8579670786857605:  79%|███████▉  | 5119/6470 [1:04:04<15:57,  1.41it/s]
2|5120|Loss: 0.8579670786857605:  79%|███████▉  | 5120/6470 [1:04:05<16:35,  1.36it/s]
2|5121|Loss: 0.8586593866348267:  79%|███████▉  | 5120/6470 [1:04:05<16:35,  1.36it/s]
2|5121|Loss: 0.8586593866348267:  79%|███████▉  | 5121/6470 [1:04:05<16:05,  1.40it/s]
2|5122|Loss: 1.1280407905578613:  79%|███████▉  | 5121/6470 [1:04:06<16:05,  1.40it/s]
2|5122|Loss: 1.1280407905578613:  79%|███████▉  | 5122/6470 [1:04:06<15:51,  1.42it/s]
2|5123|Loss: 0.9654034376144409:  79%|███████▉  | 5122/6470 [1:04:06<15:51,  1.42it/s]
2|5123|Loss: 0.9654034376144409:  79%|███████▉  | 5123/6470 [1:04:07<15:47,  1.42it/s]
2|5124|Loss: 1.006718397140503:  79%|███████▉  | 5123/6470 [1:04:07<15:47,  1.42it/s] 
2|5124|Loss: 1.006718397140503:  79%|███████▉  | 5124/6470 [1:04:07<16:05,  1.39it/s]
2|5125|Loss: 0.8552533984184265:  79%|███████▉  | 5124/6470 [1:04:08<16:05,  1.39it/s]
2|5125|Loss: 0.8552533984184265:  79%|███████▉  | 5125/6470 [1:04:08<16:22,  1.37it/s]
2|5126|Loss: 0.8275477886199951:  79%|███████▉  | 5125/6470 [1:04:09<16:22,  1.37it/s]
2|5126|Loss: 0.8275477886199951:  79%|███████▉  | 5126/6470 [1:04:09<16:34,  1.35it/s]
2|5127|Loss: 0.8129295110702515:  79%|███████▉  | 5126/6470 [1:04:09<16:34,  1.35it/s]
2|5127|Loss: 0.8129295110702515:  79%|███████▉  | 5127/6470 [1:04:10<16:26,  1.36it/s]
2|5128|Loss: 0.7341836094856262:  79%|███████▉  | 5127/6470 [1:04:10<16:26,  1.36it/s]
2|5128|Loss: 0.7341836094856262:  79%|███████▉  | 5128/6470 [1:04:10<16:30,  1.35it/s]
2|5129|Loss: 0.8616673350334167:  79%|███████▉  | 5128/6470 [1:04:11<16:30,  1.35it/s]
2|5129|Loss: 0.8616673350334167:  79%|███████▉  | 5129/6470 [1:04:11<17:32,  1.27it/s]
2|5130|Loss: 0.6743010878562927:  79%|███████▉  | 5129/6470 [1:04:12<17:32,  1.27it/s]
2|5130|Loss: 0.6743010878562927:  79%|███████▉  | 5130/6470 [1:04:12<17:38,  1.27it/s]
2|5131|Loss: 0.7946996092796326:  79%|███████▉  | 5130/6470 [1:04:12<17:38,  1.27it/s]
2|5131|Loss: 0.7946996092796326:  79%|███████▉  | 5131/6470 [1:04:13<18:00,  1.24it/s]
2|5132|Loss: 0.8991618752479553:  79%|███████▉  | 5131/6470 [1:04:13<18:00,  1.24it/s]
2|5132|Loss: 0.8991618752479553:  79%|███████▉  | 5132/6470 [1:04:14<17:52,  1.25it/s]
2|5133|Loss: 0.7148618102073669:  79%|███████▉  | 5132/6470 [1:04:14<17:52,  1.25it/s]
2|5133|Loss: 0.7148618102073669:  79%|███████▉  | 5133/6470 [1:04:14<17:14,  1.29it/s]
2|5134|Loss: 0.8311249613761902:  79%|███████▉  | 5133/6470 [1:04:15<17:14,  1.29it/s]
2|5134|Loss: 0.8311249613761902:  79%|███████▉  | 5134/6470 [1:04:15<17:10,  1.30it/s]
2|5135|Loss: 0.9249366521835327:  79%|███████▉  | 5134/6470 [1:04:16<17:10,  1.30it/s]
2|5135|Loss: 0.9249366521835327:  79%|███████▉  | 5135/6470 [1:04:16<17:46,  1.25it/s]
2|5136|Loss: 0.7375321388244629:  79%|███████▉  | 5135/6470 [1:04:16<17:46,  1.25it/s]
2|5136|Loss: 0.7375321388244629:  79%|███████▉  | 5136/6470 [1:04:17<17:50,  1.25it/s]
2|5137|Loss: 0.9207055568695068:  79%|███████▉  | 5136/6470 [1:04:17<17:50,  1.25it/s]
2|5137|Loss: 0.9207055568695068:  79%|███████▉  | 5137/6470 [1:04:18<17:34,  1.26it/s]
2|5138|Loss: 0.7584275603294373:  79%|███████▉  | 5137/6470 [1:04:18<17:34,  1.26it/s]
2|5138|Loss: 0.7584275603294373:  79%|███████▉  | 5138/6470 [1:04:18<17:15,  1.29it/s]
2|5139|Loss: 0.552818238735199:  79%|███████▉  | 5138/6470 [1:04:19<17:15,  1.29it/s] 
2|5139|Loss: 0.552818238735199:  79%|███████▉  | 5139/6470 [1:04:19<18:11,  1.22it/s]
2|5140|Loss: 0.6887681484222412:  79%|███████▉  | 5139/6470 [1:04:20<18:11,  1.22it/s]
2|5140|Loss: 0.6887681484222412:  79%|███████▉  | 5140/6470 [1:04:20<18:14,  1.21it/s]
2|5141|Loss: 1.067160964012146:  79%|███████▉  | 5140/6470 [1:04:20<18:14,  1.21it/s] 
2|5141|Loss: 1.067160964012146:  79%|███████▉  | 5141/6470 [1:04:21<17:25,  1.27it/s]
2|5142|Loss: 0.7081961035728455:  79%|███████▉  | 5141/6470 [1:04:21<17:25,  1.27it/s]
2|5142|Loss: 0.7081961035728455:  79%|███████▉  | 5142/6470 [1:04:22<17:17,  1.28it/s]
2|5143|Loss: 0.8865053057670593:  79%|███████▉  | 5142/6470 [1:04:22<17:17,  1.28it/s]
2|5143|Loss: 0.8865053057670593:  79%|███████▉  | 5143/6470 [1:04:22<17:55,  1.23it/s]
2|5144|Loss: 0.8119964003562927:  79%|███████▉  | 5143/6470 [1:04:23<17:55,  1.23it/s]
2|5144|Loss: 0.8119964003562927:  80%|███████▉  | 5144/6470 [1:04:23<17:45,  1.24it/s]
2|5145|Loss: 0.6909515857696533:  80%|███████▉  | 5144/6470 [1:04:24<17:45,  1.24it/s]
2|5145|Loss: 0.6909515857696533:  80%|███████▉  | 5145/6470 [1:04:24<17:34,  1.26it/s]
2|5146|Loss: 0.8466561436653137:  80%|███████▉  | 5145/6470 [1:04:24<17:34,  1.26it/s]
2|5146|Loss: 0.8466561436653137:  80%|███████▉  | 5146/6470 [1:04:25<16:28,  1.34it/s]
2|5147|Loss: 0.7586060762405396:  80%|███████▉  | 5146/6470 [1:04:25<16:28,  1.34it/s]
2|5147|Loss: 0.7586060762405396:  80%|███████▉  | 5147/6470 [1:04:26<17:54,  1.23it/s]
2|5148|Loss: 0.6937059164047241:  80%|███████▉  | 5147/6470 [1:04:26<17:54,  1.23it/s]
2|5148|Loss: 0.6937059164047241:  80%|███████▉  | 5148/6470 [1:04:26<17:38,  1.25it/s]
2|5149|Loss: 1.0992610454559326:  80%|███████▉  | 5148/6470 [1:04:27<17:38,  1.25it/s]
2|5149|Loss: 1.0992610454559326:  80%|███████▉  | 5149/6470 [1:04:27<15:44,  1.40it/s]
2|5150|Loss: 0.8839399218559265:  80%|███████▉  | 5149/6470 [1:04:27<15:44,  1.40it/s]
2|5150|Loss: 0.8839399218559265:  80%|███████▉  | 5150/6470 [1:04:28<15:05,  1.46it/s]
2|5151|Loss: 0.9151245951652527:  80%|███████▉  | 5150/6470 [1:04:28<15:05,  1.46it/s]
2|5151|Loss: 0.9151245951652527:  80%|███████▉  | 5151/6470 [1:04:28<15:27,  1.42it/s]
2|5152|Loss: 0.8373406529426575:  80%|███████▉  | 5151/6470 [1:04:29<15:27,  1.42it/s]
2|5152|Loss: 0.8373406529426575:  80%|███████▉  | 5152/6470 [1:04:29<15:59,  1.37it/s]
2|5153|Loss: 0.8535252213478088:  80%|███████▉  | 5152/6470 [1:04:29<15:59,  1.37it/s]
2|5153|Loss: 0.8535252213478088:  80%|███████▉  | 5153/6470 [1:04:30<15:24,  1.42it/s]
2|5154|Loss: 0.9204347729682922:  80%|███████▉  | 5153/6470 [1:04:30<15:24,  1.42it/s]
2|5154|Loss: 0.9204347729682922:  80%|███████▉  | 5154/6470 [1:04:30<14:23,  1.52it/s]
2|5155|Loss: 0.813401460647583:  80%|███████▉  | 5154/6470 [1:04:31<14:23,  1.52it/s] 
2|5155|Loss: 0.813401460647583:  80%|███████▉  | 5155/6470 [1:04:31<13:59,  1.57it/s]
2|5156|Loss: 0.9891424179077148:  80%|███████▉  | 5155/6470 [1:04:31<13:59,  1.57it/s]
2|5156|Loss: 0.9891424179077148:  80%|███████▉  | 5156/6470 [1:04:32<15:23,  1.42it/s]
2|5157|Loss: 0.8251714706420898:  80%|███████▉  | 5156/6470 [1:04:32<15:23,  1.42it/s]
2|5157|Loss: 0.8251714706420898:  80%|███████▉  | 5157/6470 [1:04:32<15:37,  1.40it/s]
2|5158|Loss: 0.9619481563568115:  80%|███████▉  | 5157/6470 [1:04:33<15:37,  1.40it/s]
2|5158|Loss: 0.9619481563568115:  80%|███████▉  | 5158/6470 [1:04:33<15:57,  1.37it/s]
2|5159|Loss: 0.8082267045974731:  80%|███████▉  | 5158/6470 [1:04:34<15:57,  1.37it/s]
2|5159|Loss: 0.8082267045974731:  80%|███████▉  | 5159/6470 [1:04:34<16:24,  1.33it/s]
2|5160|Loss: 0.8552871346473694:  80%|███████▉  | 5159/6470 [1:04:34<16:24,  1.33it/s]
2|5160|Loss: 0.8552871346473694:  80%|███████▉  | 5160/6470 [1:04:35<16:15,  1.34it/s]
2|5161|Loss: 0.8734556436538696:  80%|███████▉  | 5160/6470 [1:04:35<16:15,  1.34it/s]
2|5161|Loss: 0.8734556436538696:  80%|███████▉  | 5161/6470 [1:04:36<17:27,  1.25it/s]
2|5162|Loss: 0.7970523834228516:  80%|███████▉  | 5161/6470 [1:04:36<17:27,  1.25it/s]
2|5162|Loss: 0.7970523834228516:  80%|███████▉  | 5162/6470 [1:04:37<17:50,  1.22it/s]
2|5163|Loss: 0.7497280240058899:  80%|███████▉  | 5162/6470 [1:04:37<17:50,  1.22it/s]
2|5163|Loss: 0.7497280240058899:  80%|███████▉  | 5163/6470 [1:04:37<17:42,  1.23it/s]
2|5164|Loss: 0.9032217860221863:  80%|███████▉  | 5163/6470 [1:04:38<17:42,  1.23it/s]
2|5164|Loss: 0.9032217860221863:  80%|███████▉  | 5164/6470 [1:04:38<17:04,  1.28it/s]
2|5165|Loss: 0.873208224773407:  80%|███████▉  | 5164/6470 [1:04:38<17:04,  1.28it/s] 
2|5165|Loss: 0.873208224773407:  80%|███████▉  | 5165/6470 [1:04:39<17:04,  1.27it/s]
2|5166|Loss: 0.8593947887420654:  80%|███████▉  | 5165/6470 [1:04:39<17:04,  1.27it/s]
2|5166|Loss: 0.8593947887420654:  80%|███████▉  | 5166/6470 [1:04:40<17:07,  1.27it/s]
2|5167|Loss: 1.070428490638733:  80%|███████▉  | 5166/6470 [1:04:40<17:07,  1.27it/s] 
2|5167|Loss: 1.070428490638733:  80%|███████▉  | 5167/6470 [1:04:40<16:30,  1.32it/s]
2|5168|Loss: 0.8286420106887817:  80%|███████▉  | 5167/6470 [1:04:41<16:30,  1.32it/s]
2|5168|Loss: 0.8286420106887817:  80%|███████▉  | 5168/6470 [1:04:41<15:12,  1.43it/s]
2|5169|Loss: 0.9042700529098511:  80%|███████▉  | 5168/6470 [1:04:41<15:12,  1.43it/s]
2|5169|Loss: 0.9042700529098511:  80%|███████▉  | 5169/6470 [1:04:42<15:20,  1.41it/s]
2|5170|Loss: 0.9730119109153748:  80%|███████▉  | 5169/6470 [1:04:42<15:20,  1.41it/s]
2|5170|Loss: 0.9730119109153748:  80%|███████▉  | 5170/6470 [1:04:42<15:49,  1.37it/s]
2|5171|Loss: 0.9610733389854431:  80%|███████▉  | 5170/6470 [1:04:43<15:49,  1.37it/s]
2|5171|Loss: 0.9610733389854431:  80%|███████▉  | 5171/6470 [1:04:43<16:05,  1.35it/s]
2|5172|Loss: 0.8662503957748413:  80%|███████▉  | 5171/6470 [1:04:44<16:05,  1.35it/s]
2|5172|Loss: 0.8662503957748413:  80%|███████▉  | 5172/6470 [1:04:44<16:37,  1.30it/s]
2|5173|Loss: 0.855975329875946:  80%|███████▉  | 5172/6470 [1:04:44<16:37,  1.30it/s] 
2|5173|Loss: 0.855975329875946:  80%|███████▉  | 5173/6470 [1:04:45<17:10,  1.26it/s]
2|5174|Loss: 0.798443615436554:  80%|███████▉  | 5173/6470 [1:04:45<17:10,  1.26it/s]
2|5174|Loss: 0.798443615436554:  80%|███████▉  | 5174/6470 [1:04:46<17:58,  1.20it/s]
2|5175|Loss: 0.8961785435676575:  80%|███████▉  | 5174/6470 [1:04:46<17:58,  1.20it/s]
2|5175|Loss: 0.8961785435676575:  80%|███████▉  | 5175/6470 [1:04:46<17:12,  1.25it/s]
2|5176|Loss: 0.8605950474739075:  80%|███████▉  | 5175/6470 [1:04:47<17:12,  1.25it/s]
2|5176|Loss: 0.8605950474739075:  80%|████████  | 5176/6470 [1:04:47<15:22,  1.40it/s]
2|5177|Loss: 0.9692515730857849:  80%|████████  | 5176/6470 [1:04:47<15:22,  1.40it/s]
2|5177|Loss: 0.9692515730857849:  80%|████████  | 5177/6470 [1:04:48<14:31,  1.48it/s]
2|5178|Loss: 0.7678821086883545:  80%|████████  | 5177/6470 [1:04:48<14:31,  1.48it/s]
2|5178|Loss: 0.7678821086883545:  80%|████████  | 5178/6470 [1:04:48<15:01,  1.43it/s]
2|5179|Loss: 0.8740218281745911:  80%|████████  | 5178/6470 [1:04:49<15:01,  1.43it/s]
2|5179|Loss: 0.8740218281745911:  80%|████████  | 5179/6470 [1:04:49<16:37,  1.29it/s]
2|5180|Loss: 0.9526164531707764:  80%|████████  | 5179/6470 [1:04:50<16:37,  1.29it/s]
2|5180|Loss: 0.9526164531707764:  80%|████████  | 5180/6470 [1:04:50<16:21,  1.31it/s]
2|5181|Loss: 0.6984851360321045:  80%|████████  | 5180/6470 [1:04:50<16:21,  1.31it/s]
2|5181|Loss: 0.6984851360321045:  80%|████████  | 5181/6470 [1:04:51<17:15,  1.24it/s]
2|5182|Loss: 0.8075833916664124:  80%|████████  | 5181/6470 [1:04:51<17:15,  1.24it/s]
2|5182|Loss: 0.8075833916664124:  80%|████████  | 5182/6470 [1:04:52<17:29,  1.23it/s]
2|5183|Loss: 0.7706702351570129:  80%|████████  | 5182/6470 [1:04:52<17:29,  1.23it/s]
2|5183|Loss: 0.7706702351570129:  80%|████████  | 5183/6470 [1:04:53<17:22,  1.23it/s]
2|5184|Loss: 0.7829094529151917:  80%|████████  | 5183/6470 [1:04:53<17:22,  1.23it/s]
2|5184|Loss: 0.7829094529151917:  80%|████████  | 5184/6470 [1:04:53<16:23,  1.31it/s]
2|5185|Loss: 0.7546188235282898:  80%|████████  | 5184/6470 [1:04:54<16:23,  1.31it/s]
2|5185|Loss: 0.7546188235282898:  80%|████████  | 5185/6470 [1:04:54<16:22,  1.31it/s]
2|5186|Loss: 0.8062388300895691:  80%|████████  | 5185/6470 [1:04:54<16:22,  1.31it/s]
2|5186|Loss: 0.8062388300895691:  80%|████████  | 5186/6470 [1:04:55<15:26,  1.39it/s]
2|5187|Loss: 0.7108644843101501:  80%|████████  | 5186/6470 [1:04:55<15:26,  1.39it/s]
2|5187|Loss: 0.7108644843101501:  80%|████████  | 5187/6470 [1:04:55<16:17,  1.31it/s]
2|5188|Loss: 0.8945066332817078:  80%|████████  | 5187/6470 [1:04:56<16:17,  1.31it/s]
2|5188|Loss: 0.8945066332817078:  80%|████████  | 5188/6470 [1:04:56<15:59,  1.34it/s]
2|5189|Loss: 0.8260034322738647:  80%|████████  | 5188/6470 [1:04:57<15:59,  1.34it/s]
2|5189|Loss: 0.8260034322738647:  80%|████████  | 5189/6470 [1:04:57<16:06,  1.33it/s]
2|5190|Loss: 0.8189590573310852:  80%|████████  | 5189/6470 [1:04:57<16:06,  1.33it/s]
2|5190|Loss: 0.8189590573310852:  80%|████████  | 5190/6470 [1:04:58<16:13,  1.31it/s]
2|5191|Loss: 0.8052535057067871:  80%|████████  | 5190/6470 [1:04:58<16:13,  1.31it/s]
2|5191|Loss: 0.8052535057067871:  80%|████████  | 5191/6470 [1:04:58<15:33,  1.37it/s]
2|5192|Loss: 0.7705401182174683:  80%|████████  | 5191/6470 [1:04:59<15:33,  1.37it/s]
2|5192|Loss: 0.7705401182174683:  80%|████████  | 5192/6470 [1:04:59<17:03,  1.25it/s]
2|5193|Loss: 0.9471268057823181:  80%|████████  | 5192/6470 [1:05:00<17:03,  1.25it/s]
2|5193|Loss: 0.9471268057823181:  80%|████████  | 5193/6470 [1:05:00<17:22,  1.22it/s]
2|5194|Loss: 0.6772381067276001:  80%|████████  | 5193/6470 [1:05:01<17:22,  1.22it/s]
2|5194|Loss: 0.6772381067276001:  80%|████████  | 5194/6470 [1:05:01<16:37,  1.28it/s]
2|5195|Loss: 0.7991018295288086:  80%|████████  | 5194/6470 [1:05:01<16:37,  1.28it/s]
2|5195|Loss: 0.7991018295288086:  80%|████████  | 5195/6470 [1:05:02<17:34,  1.21it/s]
2|5196|Loss: 0.9778739809989929:  80%|████████  | 5195/6470 [1:05:02<17:34,  1.21it/s]
2|5196|Loss: 0.9778739809989929:  80%|████████  | 5196/6470 [1:05:03<17:11,  1.24it/s]
2|5197|Loss: 0.8163304328918457:  80%|████████  | 5196/6470 [1:05:03<17:11,  1.24it/s]
2|5197|Loss: 0.8163304328918457:  80%|████████  | 5197/6470 [1:05:03<16:48,  1.26it/s]
2|5198|Loss: 0.8730112314224243:  80%|████████  | 5197/6470 [1:05:04<16:48,  1.26it/s]
2|5198|Loss: 0.8730112314224243:  80%|████████  | 5198/6470 [1:05:04<16:30,  1.28it/s]
2|5199|Loss: 0.8309801816940308:  80%|████████  | 5198/6470 [1:05:04<16:30,  1.28it/s]
2|5199|Loss: 0.8309801816940308:  80%|████████  | 5199/6470 [1:05:05<16:18,  1.30it/s]
2|5200|Loss: 0.6819025278091431:  80%|████████  | 5199/6470 [1:05:05<16:18,  1.30it/s]
2|5200|Loss: 0.6819025278091431:  80%|████████  | 5200/6470 [1:05:06<15:44,  1.34it/s]
2|5201|Loss: 0.8892823457717896:  80%|████████  | 5200/6470 [1:05:06<15:44,  1.34it/s]
2|5201|Loss: 0.8892823457717896:  80%|████████  | 5201/6470 [1:05:06<15:51,  1.33it/s]
2|5202|Loss: 0.8287038803100586:  80%|████████  | 5201/6470 [1:05:07<15:51,  1.33it/s]
2|5202|Loss: 0.8287038803100586:  80%|████████  | 5202/6470 [1:05:07<15:25,  1.37it/s]
2|5203|Loss: 0.8589820861816406:  80%|████████  | 5202/6470 [1:05:07<15:25,  1.37it/s]
2|5203|Loss: 0.8589820861816406:  80%|████████  | 5203/6470 [1:05:08<15:02,  1.40it/s]
2|5204|Loss: 0.8184230923652649:  80%|████████  | 5203/6470 [1:05:08<15:02,  1.40it/s]
2|5204|Loss: 0.8184230923652649:  80%|████████  | 5204/6470 [1:05:08<14:57,  1.41it/s]
2|5205|Loss: 0.8461369872093201:  80%|████████  | 5204/6470 [1:05:09<14:57,  1.41it/s]
2|5205|Loss: 0.8461369872093201:  80%|████████  | 5205/6470 [1:05:09<13:47,  1.53it/s]
2|5206|Loss: 0.8873065710067749:  80%|████████  | 5205/6470 [1:05:09<13:47,  1.53it/s]
2|5206|Loss: 0.8873065710067749:  80%|████████  | 5206/6470 [1:05:09<12:59,  1.62it/s]
2|5207|Loss: 0.7233479022979736:  80%|████████  | 5206/6470 [1:05:10<12:59,  1.62it/s]
2|5207|Loss: 0.7233479022979736:  80%|████████  | 5207/6470 [1:05:10<13:34,  1.55it/s]
2|5208|Loss: 0.7963784337043762:  80%|████████  | 5207/6470 [1:05:11<13:34,  1.55it/s]
2|5208|Loss: 0.7963784337043762:  80%|████████  | 5208/6470 [1:05:11<14:14,  1.48it/s]
2|5209|Loss: 0.8356777429580688:  80%|████████  | 5208/6470 [1:05:11<14:14,  1.48it/s]
2|5209|Loss: 0.8356777429580688:  81%|████████  | 5209/6470 [1:05:12<15:00,  1.40it/s]
2|5210|Loss: 1.0159670114517212:  81%|████████  | 5209/6470 [1:05:12<15:00,  1.40it/s]
2|5210|Loss: 1.0159670114517212:  81%|████████  | 5210/6470 [1:05:12<15:29,  1.36it/s]
2|5211|Loss: 0.9884647130966187:  81%|████████  | 5210/6470 [1:05:13<15:29,  1.36it/s]
2|5211|Loss: 0.9884647130966187:  81%|████████  | 5211/6470 [1:05:13<15:12,  1.38it/s]
2|5212|Loss: 0.7436335682868958:  81%|████████  | 5211/6470 [1:05:14<15:12,  1.38it/s]
2|5212|Loss: 0.7436335682868958:  81%|████████  | 5212/6470 [1:05:14<14:41,  1.43it/s]
2|5213|Loss: 0.8238803148269653:  81%|████████  | 5212/6470 [1:05:14<14:41,  1.43it/s]
2|5213|Loss: 0.8238803148269653:  81%|████████  | 5213/6470 [1:05:15<15:51,  1.32it/s]
2|5214|Loss: 0.769956111907959:  81%|████████  | 5213/6470 [1:05:15<15:51,  1.32it/s] 
2|5214|Loss: 0.769956111907959:  81%|████████  | 5214/6470 [1:05:15<16:10,  1.29it/s]
2|5215|Loss: 0.8892871737480164:  81%|████████  | 5214/6470 [1:05:16<16:10,  1.29it/s]
2|5215|Loss: 0.8892871737480164:  81%|████████  | 5215/6470 [1:05:16<15:06,  1.38it/s]
2|5216|Loss: 0.966524600982666:  81%|████████  | 5215/6470 [1:05:16<15:06,  1.38it/s] 
2|5216|Loss: 0.966524600982666:  81%|████████  | 5216/6470 [1:05:17<14:17,  1.46it/s]
2|5217|Loss: 0.9524168968200684:  81%|████████  | 5216/6470 [1:05:17<14:17,  1.46it/s]
2|5217|Loss: 0.9524168968200684:  81%|████████  | 5217/6470 [1:05:17<14:16,  1.46it/s]
2|5218|Loss: 0.9003546237945557:  81%|████████  | 5217/6470 [1:05:18<14:16,  1.46it/s]
2|5218|Loss: 0.9003546237945557:  81%|████████  | 5218/6470 [1:05:18<14:15,  1.46it/s]
2|5219|Loss: 0.7976425886154175:  81%|████████  | 5218/6470 [1:05:18<14:15,  1.46it/s]
2|5219|Loss: 0.7976425886154175:  81%|████████  | 5219/6470 [1:05:19<13:34,  1.54it/s]
2|5220|Loss: 0.7846764922142029:  81%|████████  | 5219/6470 [1:05:19<13:34,  1.54it/s]
2|5220|Loss: 0.7846764922142029:  81%|████████  | 5220/6470 [1:05:19<14:16,  1.46it/s]
2|5221|Loss: 0.8114621639251709:  81%|████████  | 5220/6470 [1:05:20<14:16,  1.46it/s]
2|5221|Loss: 0.8114621639251709:  81%|████████  | 5221/6470 [1:05:20<14:30,  1.43it/s]
2|5222|Loss: 0.7884664535522461:  81%|████████  | 5221/6470 [1:05:21<14:30,  1.43it/s]
2|5222|Loss: 0.7884664535522461:  81%|████████  | 5222/6470 [1:05:21<15:03,  1.38it/s]
2|5223|Loss: 0.8109438419342041:  81%|████████  | 5222/6470 [1:05:21<15:03,  1.38it/s]
2|5223|Loss: 0.8109438419342041:  81%|████████  | 5223/6470 [1:05:22<14:20,  1.45it/s]
2|5224|Loss: 0.9510906934738159:  81%|████████  | 5223/6470 [1:05:22<14:20,  1.45it/s]
2|5224|Loss: 0.9510906934738159:  81%|████████  | 5224/6470 [1:05:22<14:31,  1.43it/s]
2|5225|Loss: 0.9437605142593384:  81%|████████  | 5224/6470 [1:05:23<14:31,  1.43it/s]
2|5225|Loss: 0.9437605142593384:  81%|████████  | 5225/6470 [1:05:23<13:32,  1.53it/s]
2|5226|Loss: 0.8509423136711121:  81%|████████  | 5225/6470 [1:05:23<13:32,  1.53it/s]
2|5226|Loss: 0.8509423136711121:  81%|████████  | 5226/6470 [1:05:23<13:45,  1.51it/s]
2|5227|Loss: 0.7862217426300049:  81%|████████  | 5226/6470 [1:05:24<13:45,  1.51it/s]
2|5227|Loss: 0.7862217426300049:  81%|████████  | 5227/6470 [1:05:24<12:27,  1.66it/s]
2|5228|Loss: 0.7807736396789551:  81%|████████  | 5227/6470 [1:05:24<12:27,  1.66it/s]
2|5228|Loss: 0.7807736396789551:  81%|████████  | 5228/6470 [1:05:24<11:36,  1.78it/s]
2|5229|Loss: 0.9172313809394836:  81%|████████  | 5228/6470 [1:05:25<11:36,  1.78it/s]
2|5229|Loss: 0.9172313809394836:  81%|████████  | 5229/6470 [1:05:25<12:06,  1.71it/s]
2|5230|Loss: 0.8036972284317017:  81%|████████  | 5229/6470 [1:05:25<12:06,  1.71it/s]
2|5230|Loss: 0.8036972284317017:  81%|████████  | 5230/6470 [1:05:26<12:17,  1.68it/s]
2|5231|Loss: 0.8789218664169312:  81%|████████  | 5230/6470 [1:05:26<12:17,  1.68it/s]
2|5231|Loss: 0.8789218664169312:  81%|████████  | 5231/6470 [1:05:27<14:01,  1.47it/s]
2|5232|Loss: 0.7541428804397583:  81%|████████  | 5231/6470 [1:05:27<14:01,  1.47it/s]
2|5232|Loss: 0.7541428804397583:  81%|████████  | 5232/6470 [1:05:28<15:57,  1.29it/s]
2|5233|Loss: 0.9056393504142761:  81%|████████  | 5232/6470 [1:05:28<15:57,  1.29it/s]
2|5233|Loss: 0.9056393504142761:  81%|████████  | 5233/6470 [1:05:28<16:58,  1.21it/s]
2|5234|Loss: 0.9202154874801636:  81%|████████  | 5233/6470 [1:05:29<16:58,  1.21it/s]
2|5234|Loss: 0.9202154874801636:  81%|████████  | 5234/6470 [1:05:29<17:58,  1.15it/s]
2|5235|Loss: 0.9881986975669861:  81%|████████  | 5234/6470 [1:05:30<17:58,  1.15it/s]
2|5235|Loss: 0.9881986975669861:  81%|████████  | 5235/6470 [1:05:30<17:47,  1.16it/s]
2|5236|Loss: 0.8887509107589722:  81%|████████  | 5235/6470 [1:05:31<17:47,  1.16it/s]
2|5236|Loss: 0.8887509107589722:  81%|████████  | 5236/6470 [1:05:31<18:02,  1.14it/s]
2|5237|Loss: 0.8128678798675537:  81%|████████  | 5236/6470 [1:05:32<18:02,  1.14it/s]
2|5237|Loss: 0.8128678798675537:  81%|████████  | 5237/6470 [1:05:32<17:40,  1.16it/s]
2|5238|Loss: 0.9146866798400879:  81%|████████  | 5237/6470 [1:05:32<17:40,  1.16it/s]
2|5238|Loss: 0.9146866798400879:  81%|████████  | 5238/6470 [1:05:33<16:13,  1.27it/s]
2|5239|Loss: 0.822955310344696:  81%|████████  | 5238/6470 [1:05:33<16:13,  1.27it/s] 
2|5239|Loss: 0.822955310344696:  81%|████████  | 5239/6470 [1:05:33<15:56,  1.29it/s]
2|5240|Loss: 0.8409214019775391:  81%|████████  | 5239/6470 [1:05:34<15:56,  1.29it/s]
2|5240|Loss: 0.8409214019775391:  81%|████████  | 5240/6470 [1:05:34<15:24,  1.33it/s]
2|5241|Loss: 0.86711186170578:  81%|████████  | 5240/6470 [1:05:35<15:24,  1.33it/s]  
2|5241|Loss: 0.86711186170578:  81%|████████  | 5241/6470 [1:05:35<16:03,  1.28it/s]
2|5242|Loss: 0.8002089262008667:  81%|████████  | 5241/6470 [1:05:35<16:03,  1.28it/s]
2|5242|Loss: 0.8002089262008667:  81%|████████  | 5242/6470 [1:05:36<16:07,  1.27it/s]
2|5243|Loss: 0.829325795173645:  81%|████████  | 5242/6470 [1:05:36<16:07,  1.27it/s] 
2|5243|Loss: 0.829325795173645:  81%|████████  | 5243/6470 [1:05:36<14:54,  1.37it/s]
2|5244|Loss: 0.9616187214851379:  81%|████████  | 5243/6470 [1:05:37<14:54,  1.37it/s]
2|5244|Loss: 0.9616187214851379:  81%|████████  | 5244/6470 [1:05:37<15:27,  1.32it/s]
2|5245|Loss: 0.8443201184272766:  81%|████████  | 5244/6470 [1:05:38<15:27,  1.32it/s]
2|5245|Loss: 0.8443201184272766:  81%|████████  | 5245/6470 [1:05:38<15:49,  1.29it/s]
2|5246|Loss: 0.7474735379219055:  81%|████████  | 5245/6470 [1:05:38<15:49,  1.29it/s]
2|5246|Loss: 0.7474735379219055:  81%|████████  | 5246/6470 [1:05:39<15:31,  1.31it/s]
2|5247|Loss: 0.9042264819145203:  81%|████████  | 5246/6470 [1:05:39<15:31,  1.31it/s]
2|5247|Loss: 0.9042264819145203:  81%|████████  | 5247/6470 [1:05:40<15:58,  1.28it/s]
2|5248|Loss: 0.8452815413475037:  81%|████████  | 5247/6470 [1:05:40<15:58,  1.28it/s]
2|5248|Loss: 0.8452815413475037:  81%|████████  | 5248/6470 [1:05:40<16:11,  1.26it/s]
2|5249|Loss: 0.8897551894187927:  81%|████████  | 5248/6470 [1:05:41<16:11,  1.26it/s]
2|5249|Loss: 0.8897551894187927:  81%|████████  | 5249/6470 [1:05:41<16:48,  1.21it/s]
2|5250|Loss: 0.8398958444595337:  81%|████████  | 5249/6470 [1:05:42<16:48,  1.21it/s]
2|5250|Loss: 0.8398958444595337:  81%|████████  | 5250/6470 [1:05:42<16:12,  1.25it/s]
2|5251|Loss: 0.8855580687522888:  81%|████████  | 5250/6470 [1:05:42<16:12,  1.25it/s]
2|5251|Loss: 0.8855580687522888:  81%|████████  | 5251/6470 [1:05:43<15:10,  1.34it/s]
2|5252|Loss: 0.9270008206367493:  81%|████████  | 5251/6470 [1:05:43<15:10,  1.34it/s]
2|5252|Loss: 0.9270008206367493:  81%|████████  | 5252/6470 [1:05:43<15:16,  1.33it/s]
2|5253|Loss: 0.8956064581871033:  81%|████████  | 5252/6470 [1:05:44<15:16,  1.33it/s]
2|5253|Loss: 0.8956064581871033:  81%|████████  | 5253/6470 [1:05:44<15:21,  1.32it/s]
2|5254|Loss: 0.7872071862220764:  81%|████████  | 5253/6470 [1:05:45<15:21,  1.32it/s]
2|5254|Loss: 0.7872071862220764:  81%|████████  | 5254/6470 [1:05:45<15:32,  1.30it/s]
2|5255|Loss: 0.7351635098457336:  81%|████████  | 5254/6470 [1:05:45<15:32,  1.30it/s]
2|5255|Loss: 0.7351635098457336:  81%|████████  | 5255/6470 [1:05:46<15:40,  1.29it/s]
2|5256|Loss: 0.9216927886009216:  81%|████████  | 5255/6470 [1:05:46<15:40,  1.29it/s]
2|5256|Loss: 0.9216927886009216:  81%|████████  | 5256/6470 [1:05:47<15:47,  1.28it/s]
2|5257|Loss: 0.9613566994667053:  81%|████████  | 5256/6470 [1:05:47<15:47,  1.28it/s]
2|5257|Loss: 0.9613566994667053:  81%|████████▏ | 5257/6470 [1:05:47<15:17,  1.32it/s]
2|5258|Loss: 0.8301167488098145:  81%|████████▏ | 5257/6470 [1:05:48<15:17,  1.32it/s]
2|5258|Loss: 0.8301167488098145:  81%|████████▏ | 5258/6470 [1:05:48<15:19,  1.32it/s]
2|5259|Loss: 0.8594930768013:  81%|████████▏ | 5258/6470 [1:05:48<15:19,  1.32it/s]   
2|5259|Loss: 0.8594930768013:  81%|████████▏ | 5259/6470 [1:05:49<14:00,  1.44it/s]
2|5260|Loss: 0.8430820107460022:  81%|████████▏ | 5259/6470 [1:05:49<14:00,  1.44it/s]
2|5260|Loss: 0.8430820107460022:  81%|████████▏ | 5260/6470 [1:05:49<12:51,  1.57it/s]
2|5261|Loss: 0.7520244717597961:  81%|████████▏ | 5260/6470 [1:05:49<12:51,  1.57it/s]
2|5261|Loss: 0.7520244717597961:  81%|████████▏ | 5261/6470 [1:05:50<12:50,  1.57it/s]
2|5262|Loss: 0.8104225993156433:  81%|████████▏ | 5261/6470 [1:05:50<12:50,  1.57it/s]
2|5262|Loss: 0.8104225993156433:  81%|████████▏ | 5262/6470 [1:05:51<14:14,  1.41it/s]
2|5263|Loss: 0.9234397411346436:  81%|████████▏ | 5262/6470 [1:05:51<14:14,  1.41it/s]
2|5263|Loss: 0.9234397411346436:  81%|████████▏ | 5263/6470 [1:05:51<14:01,  1.43it/s]
2|5264|Loss: 0.8670244216918945:  81%|████████▏ | 5263/6470 [1:05:52<14:01,  1.43it/s]
2|5264|Loss: 0.8670244216918945:  81%|████████▏ | 5264/6470 [1:05:52<15:00,  1.34it/s]
2|5265|Loss: 0.869091808795929:  81%|████████▏ | 5264/6470 [1:05:53<15:00,  1.34it/s] 
2|5265|Loss: 0.869091808795929:  81%|████████▏ | 5265/6470 [1:05:53<14:57,  1.34it/s]
2|5266|Loss: 0.700512170791626:  81%|████████▏ | 5265/6470 [1:05:53<14:57,  1.34it/s]
2|5266|Loss: 0.700512170791626:  81%|████████▏ | 5266/6470 [1:05:53<14:27,  1.39it/s]
2|5267|Loss: 0.8649851083755493:  81%|████████▏ | 5266/6470 [1:05:54<14:27,  1.39it/s]
2|5267|Loss: 0.8649851083755493:  81%|████████▏ | 5267/6470 [1:05:54<15:05,  1.33it/s]
2|5268|Loss: 0.763084352016449:  81%|████████▏ | 5267/6470 [1:05:55<15:05,  1.33it/s] 
2|5268|Loss: 0.763084352016449:  81%|████████▏ | 5268/6470 [1:05:55<14:16,  1.40it/s]
2|5269|Loss: 0.7634608745574951:  81%|████████▏ | 5268/6470 [1:05:55<14:16,  1.40it/s]
2|5269|Loss: 0.7634608745574951:  81%|████████▏ | 5269/6470 [1:05:56<13:36,  1.47it/s]
2|5270|Loss: 0.8677804470062256:  81%|████████▏ | 5269/6470 [1:05:56<13:36,  1.47it/s]
2|5270|Loss: 0.8677804470062256:  81%|████████▏ | 5270/6470 [1:05:56<14:03,  1.42it/s]
2|5271|Loss: 0.9149657487869263:  81%|████████▏ | 5270/6470 [1:05:57<14:03,  1.42it/s]
2|5271|Loss: 0.9149657487869263:  81%|████████▏ | 5271/6470 [1:05:57<14:18,  1.40it/s]
2|5272|Loss: 0.902828574180603:  81%|████████▏ | 5271/6470 [1:05:57<14:18,  1.40it/s] 
2|5272|Loss: 0.902828574180603:  81%|████████▏ | 5272/6470 [1:05:58<14:28,  1.38it/s]
2|5273|Loss: 0.7969984412193298:  81%|████████▏ | 5272/6470 [1:05:58<14:28,  1.38it/s]
2|5273|Loss: 0.7969984412193298:  81%|████████▏ | 5273/6470 [1:05:59<15:10,  1.31it/s]
2|5274|Loss: 0.8598455190658569:  81%|████████▏ | 5273/6470 [1:05:59<15:10,  1.31it/s]
2|5274|Loss: 0.8598455190658569:  82%|████████▏ | 5274/6470 [1:05:59<15:46,  1.26it/s]
2|5275|Loss: 0.8740146160125732:  82%|████████▏ | 5274/6470 [1:06:00<15:46,  1.26it/s]
2|5275|Loss: 0.8740146160125732:  82%|████████▏ | 5275/6470 [1:06:00<16:17,  1.22it/s]
2|5276|Loss: 0.6848974823951721:  82%|████████▏ | 5275/6470 [1:06:01<16:17,  1.22it/s]
2|5276|Loss: 0.6848974823951721:  82%|████████▏ | 5276/6470 [1:06:01<14:17,  1.39it/s]
2|5277|Loss: 0.9338324666023254:  82%|████████▏ | 5276/6470 [1:06:01<14:17,  1.39it/s]
2|5277|Loss: 0.9338324666023254:  82%|████████▏ | 5277/6470 [1:06:01<13:01,  1.53it/s]
2|5278|Loss: 0.7012713551521301:  82%|████████▏ | 5277/6470 [1:06:02<13:01,  1.53it/s]
2|5278|Loss: 0.7012713551521301:  82%|████████▏ | 5278/6470 [1:06:02<12:17,  1.62it/s]
2|5279|Loss: 1.0210944414138794:  82%|████████▏ | 5278/6470 [1:06:02<12:17,  1.62it/s]
2|5279|Loss: 1.0210944414138794:  82%|████████▏ | 5279/6470 [1:06:03<12:18,  1.61it/s]
2|5280|Loss: 0.7828029990196228:  82%|████████▏ | 5279/6470 [1:06:03<12:18,  1.61it/s]
2|5280|Loss: 0.7828029990196228:  82%|████████▏ | 5280/6470 [1:06:03<11:12,  1.77it/s]
2|5281|Loss: 0.8991363048553467:  82%|████████▏ | 5280/6470 [1:06:03<11:12,  1.77it/s]
2|5281|Loss: 0.8991363048553467:  82%|████████▏ | 5281/6470 [1:06:03<10:52,  1.82it/s]
2|5282|Loss: 0.7871561050415039:  82%|████████▏ | 5281/6470 [1:06:04<10:52,  1.82it/s]
2|5282|Loss: 0.7871561050415039:  82%|████████▏ | 5282/6470 [1:06:04<12:29,  1.59it/s]
2|5283|Loss: 0.7822420597076416:  82%|████████▏ | 5282/6470 [1:06:05<12:29,  1.59it/s]
2|5283|Loss: 0.7822420597076416:  82%|████████▏ | 5283/6470 [1:06:05<12:53,  1.53it/s]
2|5284|Loss: 0.8137747645378113:  82%|████████▏ | 5283/6470 [1:06:05<12:53,  1.53it/s]
2|5284|Loss: 0.8137747645378113:  82%|████████▏ | 5284/6470 [1:06:06<12:34,  1.57it/s]
2|5285|Loss: 0.7213575839996338:  82%|████████▏ | 5284/6470 [1:06:06<12:34,  1.57it/s]
2|5285|Loss: 0.7213575839996338:  82%|████████▏ | 5285/6470 [1:06:06<11:09,  1.77it/s]
2|5286|Loss: 0.8916996121406555:  82%|████████▏ | 5285/6470 [1:06:06<11:09,  1.77it/s]
2|5286|Loss: 0.8916996121406555:  82%|████████▏ | 5286/6470 [1:06:07<11:58,  1.65it/s]
2|5287|Loss: 0.8128669857978821:  82%|████████▏ | 5286/6470 [1:06:07<11:58,  1.65it/s]
2|5287|Loss: 0.8128669857978821:  82%|████████▏ | 5287/6470 [1:06:07<12:59,  1.52it/s]
2|5288|Loss: 0.9378185868263245:  82%|████████▏ | 5287/6470 [1:06:08<12:59,  1.52it/s]
2|5288|Loss: 0.9378185868263245:  82%|████████▏ | 5288/6470 [1:06:08<13:49,  1.42it/s]
2|5289|Loss: 0.7680873274803162:  82%|████████▏ | 5288/6470 [1:06:09<13:49,  1.42it/s]
2|5289|Loss: 0.7680873274803162:  82%|████████▏ | 5289/6470 [1:06:09<13:34,  1.45it/s]
2|5290|Loss: 0.7079242467880249:  82%|████████▏ | 5289/6470 [1:06:09<13:34,  1.45it/s]
2|5290|Loss: 0.7079242467880249:  82%|████████▏ | 5290/6470 [1:06:10<12:59,  1.51it/s]
2|5291|Loss: 0.7191545367240906:  82%|████████▏ | 5290/6470 [1:06:10<12:59,  1.51it/s]
2|5291|Loss: 0.7191545367240906:  82%|████████▏ | 5291/6470 [1:06:10<12:43,  1.54it/s]
2|5292|Loss: 0.6127286553382874:  82%|████████▏ | 5291/6470 [1:06:11<12:43,  1.54it/s]
2|5292|Loss: 0.6127286553382874:  82%|████████▏ | 5292/6470 [1:06:11<13:57,  1.41it/s]
2|5293|Loss: 0.9038111567497253:  82%|████████▏ | 5292/6470 [1:06:12<13:57,  1.41it/s]
2|5293|Loss: 0.9038111567497253:  82%|████████▏ | 5293/6470 [1:06:12<15:17,  1.28it/s]
2|5294|Loss: 0.8734860420227051:  82%|████████▏ | 5293/6470 [1:06:12<15:17,  1.28it/s]
2|5294|Loss: 0.8734860420227051:  82%|████████▏ | 5294/6470 [1:06:13<14:25,  1.36it/s]
2|5295|Loss: 0.7625575065612793:  82%|████████▏ | 5294/6470 [1:06:13<14:25,  1.36it/s]
2|5295|Loss: 0.7625575065612793:  82%|████████▏ | 5295/6470 [1:06:13<13:34,  1.44it/s]
2|5296|Loss: 0.8610964417457581:  82%|████████▏ | 5295/6470 [1:06:14<13:34,  1.44it/s]
2|5296|Loss: 0.8610964417457581:  82%|████████▏ | 5296/6470 [1:06:14<14:33,  1.34it/s]
2|5297|Loss: 0.835308313369751:  82%|████████▏ | 5296/6470 [1:06:14<14:33,  1.34it/s] 
2|5297|Loss: 0.835308313369751:  82%|████████▏ | 5297/6470 [1:06:15<14:56,  1.31it/s]
2|5298|Loss: 0.7903514504432678:  82%|████████▏ | 5297/6470 [1:06:15<14:56,  1.31it/s]
2|5298|Loss: 0.7903514504432678:  82%|████████▏ | 5298/6470 [1:06:16<15:38,  1.25it/s]
2|5299|Loss: 0.774897038936615:  82%|████████▏ | 5298/6470 [1:06:16<15:38,  1.25it/s] 
2|5299|Loss: 0.774897038936615:  82%|████████▏ | 5299/6470 [1:06:17<15:33,  1.25it/s]
2|5300|Loss: 0.8704765439033508:  82%|████████▏ | 5299/6470 [1:06:17<15:33,  1.25it/s]
2|5300|Loss: 0.8704765439033508:  82%|████████▏ | 5300/6470 [1:06:17<14:39,  1.33it/s]
2|5301|Loss: 0.9212984442710876:  82%|████████▏ | 5300/6470 [1:06:18<14:39,  1.33it/s]
2|5301|Loss: 0.9212984442710876:  82%|████████▏ | 5301/6470 [1:06:18<14:43,  1.32it/s]
2|5302|Loss: 0.8298510909080505:  82%|████████▏ | 5301/6470 [1:06:18<14:43,  1.32it/s]
2|5302|Loss: 0.8298510909080505:  82%|████████▏ | 5302/6470 [1:06:19<14:56,  1.30it/s]
2|5303|Loss: 0.9333061575889587:  82%|████████▏ | 5302/6470 [1:06:19<14:56,  1.30it/s]
2|5303|Loss: 0.9333061575889587:  82%|████████▏ | 5303/6470 [1:06:19<14:31,  1.34it/s]
2|5304|Loss: 0.8852679133415222:  82%|████████▏ | 5303/6470 [1:06:20<14:31,  1.34it/s]
2|5304|Loss: 0.8852679133415222:  82%|████████▏ | 5304/6470 [1:06:20<14:42,  1.32it/s]
2|5305|Loss: 0.7253677248954773:  82%|████████▏ | 5304/6470 [1:06:21<14:42,  1.32it/s]
2|5305|Loss: 0.7253677248954773:  82%|████████▏ | 5305/6470 [1:06:21<15:41,  1.24it/s]
2|5306|Loss: 1.0789191722869873:  82%|████████▏ | 5305/6470 [1:06:22<15:41,  1.24it/s]
2|5306|Loss: 1.0789191722869873:  82%|████████▏ | 5306/6470 [1:06:22<15:37,  1.24it/s]
2|5307|Loss: 0.6475316882133484:  82%|████████▏ | 5306/6470 [1:06:23<15:37,  1.24it/s]
2|5307|Loss: 0.6475316882133484:  82%|████████▏ | 5307/6470 [1:06:23<19:06,  1.01it/s]
2|5308|Loss: 0.7465102672576904:  82%|████████▏ | 5307/6470 [1:06:24<19:06,  1.01it/s]
2|5308|Loss: 0.7465102672576904:  82%|████████▏ | 5308/6470 [1:06:24<18:44,  1.03it/s]
2|5309|Loss: 0.8170632719993591:  82%|████████▏ | 5308/6470 [1:06:25<18:44,  1.03it/s]
2|5309|Loss: 0.8170632719993591:  82%|████████▏ | 5309/6470 [1:06:25<17:46,  1.09it/s]
2|5310|Loss: 0.8201439380645752:  82%|████████▏ | 5309/6470 [1:06:25<17:46,  1.09it/s]
2|5310|Loss: 0.8201439380645752:  82%|████████▏ | 5310/6470 [1:06:26<16:53,  1.14it/s]
2|5311|Loss: 0.8279914259910583:  82%|████████▏ | 5310/6470 [1:06:26<16:53,  1.14it/s]
2|5311|Loss: 0.8279914259910583:  82%|████████▏ | 5311/6470 [1:06:27<15:54,  1.21it/s]
2|5312|Loss: 0.6857047080993652:  82%|████████▏ | 5311/6470 [1:06:27<15:54,  1.21it/s]
2|5312|Loss: 0.6857047080993652:  82%|████████▏ | 5312/6470 [1:06:27<14:31,  1.33it/s]
2|5313|Loss: 0.8554295301437378:  82%|████████▏ | 5312/6470 [1:06:27<14:31,  1.33it/s]
2|5313|Loss: 0.8554295301437378:  82%|████████▏ | 5313/6470 [1:06:28<13:57,  1.38it/s]
2|5314|Loss: 0.8778349757194519:  82%|████████▏ | 5313/6470 [1:06:28<13:57,  1.38it/s]
2|5314|Loss: 0.8778349757194519:  82%|████████▏ | 5314/6470 [1:06:28<13:49,  1.39it/s]
2|5315|Loss: 0.7207880616188049:  82%|████████▏ | 5314/6470 [1:06:29<13:49,  1.39it/s]
2|5315|Loss: 0.7207880616188049:  82%|████████▏ | 5315/6470 [1:06:29<15:00,  1.28it/s]
2|5316|Loss: 0.9756454229354858:  82%|████████▏ | 5315/6470 [1:06:30<15:00,  1.28it/s]
2|5316|Loss: 0.9756454229354858:  82%|████████▏ | 5316/6470 [1:06:30<15:14,  1.26it/s]
2|5317|Loss: 0.6731203198432922:  82%|████████▏ | 5316/6470 [1:06:31<15:14,  1.26it/s]
2|5317|Loss: 0.6731203198432922:  82%|████████▏ | 5317/6470 [1:06:31<14:28,  1.33it/s]
2|5318|Loss: 0.7102749943733215:  82%|████████▏ | 5317/6470 [1:06:31<14:28,  1.33it/s]
2|5318|Loss: 0.7102749943733215:  82%|████████▏ | 5318/6470 [1:06:32<14:18,  1.34it/s]
2|5319|Loss: 0.8094395399093628:  82%|████████▏ | 5318/6470 [1:06:32<14:18,  1.34it/s]
2|5319|Loss: 0.8094395399093628:  82%|████████▏ | 5319/6470 [1:06:32<14:29,  1.32it/s]
2|5320|Loss: 0.8839971423149109:  82%|████████▏ | 5319/6470 [1:06:33<14:29,  1.32it/s]
2|5320|Loss: 0.8839971423149109:  82%|████████▏ | 5320/6470 [1:06:33<14:09,  1.35it/s]
2|5321|Loss: 0.6971150636672974:  82%|████████▏ | 5320/6470 [1:06:33<14:09,  1.35it/s]
2|5321|Loss: 0.6971150636672974:  82%|████████▏ | 5321/6470 [1:06:34<13:31,  1.42it/s]
2|5322|Loss: 0.908102810382843:  82%|████████▏ | 5321/6470 [1:06:34<13:31,  1.42it/s] 
2|5322|Loss: 0.908102810382843:  82%|████████▏ | 5322/6470 [1:06:34<13:24,  1.43it/s]
2|5323|Loss: 0.9589822292327881:  82%|████████▏ | 5322/6470 [1:06:35<13:24,  1.43it/s]
2|5323|Loss: 0.9589822292327881:  82%|████████▏ | 5323/6470 [1:06:35<13:49,  1.38it/s]
2|5324|Loss: 0.8626872897148132:  82%|████████▏ | 5323/6470 [1:06:36<13:49,  1.38it/s]
2|5324|Loss: 0.8626872897148132:  82%|████████▏ | 5324/6470 [1:06:36<13:03,  1.46it/s]
2|5325|Loss: 0.7962052822113037:  82%|████████▏ | 5324/6470 [1:06:36<13:03,  1.46it/s]
2|5325|Loss: 0.7962052822113037:  82%|████████▏ | 5325/6470 [1:06:37<13:58,  1.37it/s]
2|5326|Loss: 0.79511958360672:  82%|████████▏ | 5325/6470 [1:06:37<13:58,  1.37it/s]  
2|5326|Loss: 0.79511958360672:  82%|████████▏ | 5326/6470 [1:06:37<13:34,  1.41it/s]
2|5327|Loss: 0.8627727031707764:  82%|████████▏ | 5326/6470 [1:06:38<13:34,  1.41it/s]
2|5327|Loss: 0.8627727031707764:  82%|████████▏ | 5327/6470 [1:06:38<13:54,  1.37it/s]
2|5328|Loss: 0.7701771855354309:  82%|████████▏ | 5327/6470 [1:06:38<13:54,  1.37it/s]
2|5328|Loss: 0.7701771855354309:  82%|████████▏ | 5328/6470 [1:06:39<13:44,  1.38it/s]
2|5329|Loss: 0.8113783597946167:  82%|████████▏ | 5328/6470 [1:06:39<13:44,  1.38it/s]
2|5329|Loss: 0.8113783597946167:  82%|████████▏ | 5329/6470 [1:06:40<13:51,  1.37it/s]
2|5330|Loss: 1.00575852394104:  82%|████████▏ | 5329/6470 [1:06:40<13:51,  1.37it/s]  
2|5330|Loss: 1.00575852394104:  82%|████████▏ | 5330/6470 [1:06:40<13:59,  1.36it/s]
2|5331|Loss: 0.8281915187835693:  82%|████████▏ | 5330/6470 [1:06:41<13:59,  1.36it/s]
2|5331|Loss: 0.8281915187835693:  82%|████████▏ | 5331/6470 [1:06:41<13:41,  1.39it/s]
2|5332|Loss: 0.8546393513679504:  82%|████████▏ | 5331/6470 [1:06:41<13:41,  1.39it/s]
2|5332|Loss: 0.8546393513679504:  82%|████████▏ | 5332/6470 [1:06:42<13:45,  1.38it/s]
2|5333|Loss: 0.8747994899749756:  82%|████████▏ | 5332/6470 [1:06:42<13:45,  1.38it/s]
2|5333|Loss: 0.8747994899749756:  82%|████████▏ | 5333/6470 [1:06:43<14:20,  1.32it/s]
2|5334|Loss: 0.8699605464935303:  82%|████████▏ | 5333/6470 [1:06:43<14:20,  1.32it/s]
2|5334|Loss: 0.8699605464935303:  82%|████████▏ | 5334/6470 [1:06:43<13:36,  1.39it/s]
2|5335|Loss: 0.7881157994270325:  82%|████████▏ | 5334/6470 [1:06:44<13:36,  1.39it/s]
2|5335|Loss: 0.7881157994270325:  82%|████████▏ | 5335/6470 [1:06:44<14:14,  1.33it/s]
2|5336|Loss: 0.7396910786628723:  82%|████████▏ | 5335/6470 [1:06:44<14:14,  1.33it/s]
2|5336|Loss: 0.7396910786628723:  82%|████████▏ | 5336/6470 [1:06:45<14:20,  1.32it/s]
2|5337|Loss: 1.032433032989502:  82%|████████▏ | 5336/6470 [1:06:45<14:20,  1.32it/s] 
2|5337|Loss: 1.032433032989502:  82%|████████▏ | 5337/6470 [1:06:46<14:26,  1.31it/s]
2|5338|Loss: 0.8887943625450134:  82%|████████▏ | 5337/6470 [1:06:46<14:26,  1.31it/s]
2|5338|Loss: 0.8887943625450134:  83%|████████▎ | 5338/6470 [1:06:46<13:57,  1.35it/s]
2|5339|Loss: 0.8667440414428711:  83%|████████▎ | 5338/6470 [1:06:47<13:57,  1.35it/s]
2|5339|Loss: 0.8667440414428711:  83%|████████▎ | 5339/6470 [1:06:47<13:30,  1.40it/s]
2|5340|Loss: 0.9131222367286682:  83%|████████▎ | 5339/6470 [1:06:47<13:30,  1.40it/s]
2|5340|Loss: 0.9131222367286682:  83%|████████▎ | 5340/6470 [1:06:47<12:02,  1.56it/s]
2|5341|Loss: 0.8995389938354492:  83%|████████▎ | 5340/6470 [1:06:48<12:02,  1.56it/s]
2|5341|Loss: 0.8995389938354492:  83%|████████▎ | 5341/6470 [1:06:48<12:03,  1.56it/s]
2|5342|Loss: 0.7993737459182739:  83%|████████▎ | 5341/6470 [1:06:48<12:03,  1.56it/s]
2|5342|Loss: 0.7993737459182739:  83%|████████▎ | 5342/6470 [1:06:49<11:50,  1.59it/s]
2|5343|Loss: 0.9919379353523254:  83%|████████▎ | 5342/6470 [1:06:49<11:50,  1.59it/s]
2|5343|Loss: 0.9919379353523254:  83%|████████▎ | 5343/6470 [1:06:49<12:34,  1.49it/s]
2|5344|Loss: 0.8753024339675903:  83%|████████▎ | 5343/6470 [1:06:50<12:34,  1.49it/s]
2|5344|Loss: 0.8753024339675903:  83%|████████▎ | 5344/6470 [1:06:50<14:00,  1.34it/s]
2|5345|Loss: 0.7376488447189331:  83%|████████▎ | 5344/6470 [1:06:51<14:00,  1.34it/s]
2|5345|Loss: 0.7376488447189331:  83%|████████▎ | 5345/6470 [1:06:51<14:33,  1.29it/s]
2|5346|Loss: 0.7661914825439453:  83%|████████▎ | 5345/6470 [1:06:52<14:33,  1.29it/s]
2|5346|Loss: 0.7661914825439453:  83%|████████▎ | 5346/6470 [1:06:52<14:57,  1.25it/s]
2|5347|Loss: 0.842065155506134:  83%|████████▎ | 5346/6470 [1:06:52<14:57,  1.25it/s] 
2|5347|Loss: 0.842065155506134:  83%|████████▎ | 5347/6470 [1:06:53<15:01,  1.25it/s]
2|5348|Loss: 0.861376941204071:  83%|████████▎ | 5347/6470 [1:06:53<15:01,  1.25it/s]
2|5348|Loss: 0.861376941204071:  83%|████████▎ | 5348/6470 [1:06:53<14:17,  1.31it/s]
2|5349|Loss: 0.8592304587364197:  83%|████████▎ | 5348/6470 [1:06:54<14:17,  1.31it/s]
2|5349|Loss: 0.8592304587364197:  83%|████████▎ | 5349/6470 [1:06:54<14:38,  1.28it/s]
2|5350|Loss: 0.8232595920562744:  83%|████████▎ | 5349/6470 [1:06:55<14:38,  1.28it/s]
2|5350|Loss: 0.8232595920562744:  83%|████████▎ | 5350/6470 [1:06:55<14:09,  1.32it/s]
2|5351|Loss: 0.8952651023864746:  83%|████████▎ | 5350/6470 [1:06:55<14:09,  1.32it/s]
2|5351|Loss: 0.8952651023864746:  83%|████████▎ | 5351/6470 [1:06:56<14:09,  1.32it/s]
2|5352|Loss: 0.7894394397735596:  83%|████████▎ | 5351/6470 [1:06:56<14:09,  1.32it/s]
2|5352|Loss: 0.7894394397735596:  83%|████████▎ | 5352/6470 [1:06:56<13:55,  1.34it/s]
2|5353|Loss: 0.7872800230979919:  83%|████████▎ | 5352/6470 [1:06:57<13:55,  1.34it/s]
2|5353|Loss: 0.7872800230979919:  83%|████████▎ | 5353/6470 [1:06:57<14:45,  1.26it/s]
2|5354|Loss: 0.7931463122367859:  83%|████████▎ | 5353/6470 [1:06:58<14:45,  1.26it/s]
2|5354|Loss: 0.7931463122367859:  83%|████████▎ | 5354/6470 [1:06:58<14:25,  1.29it/s]
2|5355|Loss: 0.7498763799667358:  83%|████████▎ | 5354/6470 [1:06:58<14:25,  1.29it/s]
2|5355|Loss: 0.7498763799667358:  83%|████████▎ | 5355/6470 [1:06:59<13:27,  1.38it/s]
2|5356|Loss: 0.887359619140625:  83%|████████▎ | 5355/6470 [1:06:59<13:27,  1.38it/s] 
2|5356|Loss: 0.887359619140625:  83%|████████▎ | 5356/6470 [1:06:59<13:12,  1.41it/s]
2|5357|Loss: 0.7491898536682129:  83%|████████▎ | 5356/6470 [1:07:00<13:12,  1.41it/s]
2|5357|Loss: 0.7491898536682129:  83%|████████▎ | 5357/6470 [1:07:00<13:22,  1.39it/s]
2|5358|Loss: 0.9456876516342163:  83%|████████▎ | 5357/6470 [1:07:00<13:22,  1.39it/s]
2|5358|Loss: 0.9456876516342163:  83%|████████▎ | 5358/6470 [1:07:01<12:47,  1.45it/s]
2|5359|Loss: 0.8688475489616394:  83%|████████▎ | 5358/6470 [1:07:01<12:47,  1.45it/s]
2|5359|Loss: 0.8688475489616394:  83%|████████▎ | 5359/6470 [1:07:01<12:16,  1.51it/s]
2|5360|Loss: 0.8025749921798706:  83%|████████▎ | 5359/6470 [1:07:02<12:16,  1.51it/s]
2|5360|Loss: 0.8025749921798706:  83%|████████▎ | 5360/6470 [1:07:02<12:00,  1.54it/s]
2|5361|Loss: 0.6915879845619202:  83%|████████▎ | 5360/6470 [1:07:02<12:00,  1.54it/s]
2|5361|Loss: 0.6915879845619202:  83%|████████▎ | 5361/6470 [1:07:03<13:11,  1.40it/s]
2|5362|Loss: 0.752668023109436:  83%|████████▎ | 5361/6470 [1:07:03<13:11,  1.40it/s] 
2|5362|Loss: 0.752668023109436:  83%|████████▎ | 5362/6470 [1:07:04<13:47,  1.34it/s]
2|5363|Loss: 0.8593719005584717:  83%|████████▎ | 5362/6470 [1:07:04<13:47,  1.34it/s]
2|5363|Loss: 0.8593719005584717:  83%|████████▎ | 5363/6470 [1:07:05<14:37,  1.26it/s]
2|5364|Loss: 0.8509702682495117:  83%|████████▎ | 5363/6470 [1:07:05<14:37,  1.26it/s]
2|5364|Loss: 0.8509702682495117:  83%|████████▎ | 5364/6470 [1:07:05<14:43,  1.25it/s]
2|5365|Loss: 0.8041951060295105:  83%|████████▎ | 5364/6470 [1:07:06<14:43,  1.25it/s]
2|5365|Loss: 0.8041951060295105:  83%|████████▎ | 5365/6470 [1:07:06<14:27,  1.27it/s]
2|5366|Loss: 0.8520143032073975:  83%|████████▎ | 5365/6470 [1:07:07<14:27,  1.27it/s]
2|5366|Loss: 0.8520143032073975:  83%|████████▎ | 5366/6470 [1:07:07<14:30,  1.27it/s]
2|5367|Loss: 0.8522896766662598:  83%|████████▎ | 5366/6470 [1:07:07<14:30,  1.27it/s]
2|5367|Loss: 0.8522896766662598:  83%|████████▎ | 5367/6470 [1:07:08<13:57,  1.32it/s]
2|5368|Loss: 0.8838245868682861:  83%|████████▎ | 5367/6470 [1:07:08<13:57,  1.32it/s]
2|5368|Loss: 0.8838245868682861:  83%|████████▎ | 5368/6470 [1:07:08<14:14,  1.29it/s]
2|5369|Loss: 0.8847371935844421:  83%|████████▎ | 5368/6470 [1:07:09<14:14,  1.29it/s]
2|5369|Loss: 0.8847371935844421:  83%|████████▎ | 5369/6470 [1:07:09<14:19,  1.28it/s]
2|5370|Loss: 0.6774276494979858:  83%|████████▎ | 5369/6470 [1:07:10<14:19,  1.28it/s]
2|5370|Loss: 0.6774276494979858:  83%|████████▎ | 5370/6470 [1:07:10<14:20,  1.28it/s]
2|5371|Loss: 0.9300577044487:  83%|████████▎ | 5370/6470 [1:07:10<14:20,  1.28it/s]   
2|5371|Loss: 0.9300577044487:  83%|████████▎ | 5371/6470 [1:07:11<14:01,  1.31it/s]
2|5372|Loss: 0.7910029888153076:  83%|████████▎ | 5371/6470 [1:07:11<14:01,  1.31it/s]
2|5372|Loss: 0.7910029888153076:  83%|████████▎ | 5372/6470 [1:07:11<13:37,  1.34it/s]
2|5373|Loss: 0.8698784112930298:  83%|████████▎ | 5372/6470 [1:07:12<13:37,  1.34it/s]
2|5373|Loss: 0.8698784112930298:  83%|████████▎ | 5373/6470 [1:07:12<13:11,  1.39it/s]
2|5374|Loss: 0.8181570172309875:  83%|████████▎ | 5373/6470 [1:07:12<13:11,  1.39it/s]
2|5374|Loss: 0.8181570172309875:  83%|████████▎ | 5374/6470 [1:07:13<12:55,  1.41it/s]
2|5375|Loss: 0.9111489653587341:  83%|████████▎ | 5374/6470 [1:07:13<12:55,  1.41it/s]
2|5375|Loss: 0.9111489653587341:  83%|████████▎ | 5375/6470 [1:07:13<12:46,  1.43it/s]
2|5376|Loss: 0.784008264541626:  83%|████████▎ | 5375/6470 [1:07:14<12:46,  1.43it/s] 
2|5376|Loss: 0.784008264541626:  83%|████████▎ | 5376/6470 [1:07:14<12:37,  1.44it/s]
2|5377|Loss: 0.8318390846252441:  83%|████████▎ | 5376/6470 [1:07:15<12:37,  1.44it/s]
2|5377|Loss: 0.8318390846252441:  83%|████████▎ | 5377/6470 [1:07:15<13:29,  1.35it/s]
2|5378|Loss: 0.8521970510482788:  83%|████████▎ | 5377/6470 [1:07:15<13:29,  1.35it/s]
2|5378|Loss: 0.8521970510482788:  83%|████████▎ | 5378/6470 [1:07:16<12:59,  1.40it/s]
2|5379|Loss: 0.9185159802436829:  83%|████████▎ | 5378/6470 [1:07:16<12:59,  1.40it/s]
2|5379|Loss: 0.9185159802436829:  83%|████████▎ | 5379/6470 [1:07:16<13:00,  1.40it/s]
2|5380|Loss: 0.829637885093689:  83%|████████▎ | 5379/6470 [1:07:17<13:00,  1.40it/s] 
2|5380|Loss: 0.829637885093689:  83%|████████▎ | 5380/6470 [1:07:17<13:19,  1.36it/s]
2|5381|Loss: 0.9209273457527161:  83%|████████▎ | 5380/6470 [1:07:17<13:19,  1.36it/s]
2|5381|Loss: 0.9209273457527161:  83%|████████▎ | 5381/6470 [1:07:18<12:33,  1.45it/s]
2|5382|Loss: 0.8513105511665344:  83%|████████▎ | 5381/6470 [1:07:18<12:33,  1.45it/s]
2|5382|Loss: 0.8513105511665344:  83%|████████▎ | 5382/6470 [1:07:18<12:34,  1.44it/s]
2|5383|Loss: 0.7614011168479919:  83%|████████▎ | 5382/6470 [1:07:19<12:34,  1.44it/s]
2|5383|Loss: 0.7614011168479919:  83%|████████▎ | 5383/6470 [1:07:19<12:30,  1.45it/s]
2|5384|Loss: 0.6995591521263123:  83%|████████▎ | 5383/6470 [1:07:19<12:30,  1.45it/s]
2|5384|Loss: 0.6995591521263123:  83%|████████▎ | 5384/6470 [1:07:20<12:45,  1.42it/s]
2|5385|Loss: 0.8380751013755798:  83%|████████▎ | 5384/6470 [1:07:20<12:45,  1.42it/s]
2|5385|Loss: 0.8380751013755798:  83%|████████▎ | 5385/6470 [1:07:21<13:08,  1.38it/s]
2|5386|Loss: 0.8467321991920471:  83%|████████▎ | 5385/6470 [1:07:21<13:08,  1.38it/s]
2|5386|Loss: 0.8467321991920471:  83%|████████▎ | 5386/6470 [1:07:21<13:28,  1.34it/s]
2|5387|Loss: 0.8834050297737122:  83%|████████▎ | 5386/6470 [1:07:22<13:28,  1.34it/s]
2|5387|Loss: 0.8834050297737122:  83%|████████▎ | 5387/6470 [1:07:22<13:35,  1.33it/s]
2|5388|Loss: 0.7982826232910156:  83%|████████▎ | 5387/6470 [1:07:23<13:35,  1.33it/s]
2|5388|Loss: 0.7982826232910156:  83%|████████▎ | 5388/6470 [1:07:23<13:20,  1.35it/s]
2|5389|Loss: 0.9693168997764587:  83%|████████▎ | 5388/6470 [1:07:23<13:20,  1.35it/s]
2|5389|Loss: 0.9693168997764587:  83%|████████▎ | 5389/6470 [1:07:24<13:04,  1.38it/s]
2|5390|Loss: 0.8165959715843201:  83%|████████▎ | 5389/6470 [1:07:24<13:04,  1.38it/s]
2|5390|Loss: 0.8165959715843201:  83%|████████▎ | 5390/6470 [1:07:24<13:08,  1.37it/s]
2|5391|Loss: 0.7794067859649658:  83%|████████▎ | 5390/6470 [1:07:25<13:08,  1.37it/s]
2|5391|Loss: 0.7794067859649658:  83%|████████▎ | 5391/6470 [1:07:25<13:43,  1.31it/s]
2|5392|Loss: 0.7119658589363098:  83%|████████▎ | 5391/6470 [1:07:26<13:43,  1.31it/s]
2|5392|Loss: 0.7119658589363098:  83%|████████▎ | 5392/6470 [1:07:26<14:00,  1.28it/s]
2|5393|Loss: 0.8406537771224976:  83%|████████▎ | 5392/6470 [1:07:26<14:00,  1.28it/s]
2|5393|Loss: 0.8406537771224976:  83%|████████▎ | 5393/6470 [1:07:27<13:42,  1.31it/s]
2|5394|Loss: 0.8890640735626221:  83%|████████▎ | 5393/6470 [1:07:27<13:42,  1.31it/s]
2|5394|Loss: 0.8890640735626221:  83%|████████▎ | 5394/6470 [1:07:28<14:22,  1.25it/s]
2|5395|Loss: 0.6968593001365662:  83%|████████▎ | 5394/6470 [1:07:28<14:22,  1.25it/s]
2|5395|Loss: 0.6968593001365662:  83%|████████▎ | 5395/6470 [1:07:28<12:51,  1.39it/s]
2|5396|Loss: 0.7383890748023987:  83%|████████▎ | 5395/6470 [1:07:28<12:51,  1.39it/s]
2|5396|Loss: 0.7383890748023987:  83%|████████▎ | 5396/6470 [1:07:29<11:58,  1.50it/s]
2|5397|Loss: 0.845686674118042:  83%|████████▎ | 5396/6470 [1:07:29<11:58,  1.50it/s] 
2|5397|Loss: 0.845686674118042:  83%|████████▎ | 5397/6470 [1:07:30<13:09,  1.36it/s]
2|5398|Loss: 0.8362778425216675:  83%|████████▎ | 5397/6470 [1:07:30<13:09,  1.36it/s]
2|5398|Loss: 0.8362778425216675:  83%|████████▎ | 5398/6470 [1:07:30<13:44,  1.30it/s]
2|5399|Loss: 0.8290300965309143:  83%|████████▎ | 5398/6470 [1:07:31<13:44,  1.30it/s]
2|5399|Loss: 0.8290300965309143:  83%|████████▎ | 5399/6470 [1:07:31<13:11,  1.35it/s]
2|5400|Loss: 0.7983997464179993:  83%|████████▎ | 5399/6470 [1:07:31<13:11,  1.35it/s]
2|5400|Loss: 0.7983997464179993:  83%|████████▎ | 5400/6470 [1:07:32<12:28,  1.43it/s]
2|5401|Loss: 0.7894431948661804:  83%|████████▎ | 5400/6470 [1:07:32<12:28,  1.43it/s]
2|5401|Loss: 0.7894431948661804:  83%|████████▎ | 5401/6470 [1:07:32<11:26,  1.56it/s]
2|5402|Loss: 0.8550562858581543:  83%|████████▎ | 5401/6470 [1:07:33<11:26,  1.56it/s]
2|5402|Loss: 0.8550562858581543:  83%|████████▎ | 5402/6470 [1:07:33<11:00,  1.62it/s]
2|5403|Loss: 0.8785302042961121:  83%|████████▎ | 5402/6470 [1:07:33<11:00,  1.62it/s]
2|5403|Loss: 0.8785302042961121:  84%|████████▎ | 5403/6470 [1:07:34<11:46,  1.51it/s]
2|5404|Loss: 0.9135828614234924:  84%|████████▎ | 5403/6470 [1:07:34<11:46,  1.51it/s]
2|5404|Loss: 0.9135828614234924:  84%|████████▎ | 5404/6470 [1:07:34<11:49,  1.50it/s]
2|5405|Loss: 0.9451640248298645:  84%|████████▎ | 5404/6470 [1:07:35<11:49,  1.50it/s]
2|5405|Loss: 0.9451640248298645:  84%|████████▎ | 5405/6470 [1:07:35<12:16,  1.45it/s]
2|5406|Loss: 0.9600140452384949:  84%|████████▎ | 5405/6470 [1:07:35<12:16,  1.45it/s]
2|5406|Loss: 0.9600140452384949:  84%|████████▎ | 5406/6470 [1:07:36<12:45,  1.39it/s]
2|5407|Loss: 0.7685481309890747:  84%|████████▎ | 5406/6470 [1:07:36<12:45,  1.39it/s]
2|5407|Loss: 0.7685481309890747:  84%|████████▎ | 5407/6470 [1:07:36<12:44,  1.39it/s]
2|5408|Loss: 0.8879872560501099:  84%|████████▎ | 5407/6470 [1:07:37<12:44,  1.39it/s]
2|5408|Loss: 0.8879872560501099:  84%|████████▎ | 5408/6470 [1:07:37<12:51,  1.38it/s]
2|5409|Loss: 0.8414646983146667:  84%|████████▎ | 5408/6470 [1:07:38<12:51,  1.38it/s]
2|5409|Loss: 0.8414646983146667:  84%|████████▎ | 5409/6470 [1:07:38<12:56,  1.37it/s]
2|5410|Loss: 0.886441171169281:  84%|████████▎ | 5409/6470 [1:07:38<12:56,  1.37it/s] 
2|5410|Loss: 0.886441171169281:  84%|████████▎ | 5410/6470 [1:07:39<12:23,  1.43it/s]
2|5411|Loss: 0.9074534773826599:  84%|████████▎ | 5410/6470 [1:07:39<12:23,  1.43it/s]
2|5411|Loss: 0.9074534773826599:  84%|████████▎ | 5411/6470 [1:07:39<12:36,  1.40it/s]
2|5412|Loss: 0.8494018316268921:  84%|████████▎ | 5411/6470 [1:07:40<12:36,  1.40it/s]
2|5412|Loss: 0.8494018316268921:  84%|████████▎ | 5412/6470 [1:07:40<13:31,  1.30it/s]
2|5413|Loss: 0.7742313742637634:  84%|████████▎ | 5412/6470 [1:07:41<13:31,  1.30it/s]
2|5413|Loss: 0.7742313742637634:  84%|████████▎ | 5413/6470 [1:07:41<12:31,  1.41it/s]
2|5414|Loss: 0.8156050443649292:  84%|████████▎ | 5413/6470 [1:07:41<12:31,  1.41it/s]
2|5414|Loss: 0.8156050443649292:  84%|████████▎ | 5414/6470 [1:07:41<12:02,  1.46it/s]
2|5415|Loss: 0.7434864640235901:  84%|████████▎ | 5414/6470 [1:07:42<12:02,  1.46it/s]
2|5415|Loss: 0.7434864640235901:  84%|████████▎ | 5415/6470 [1:07:42<12:44,  1.38it/s]
2|5416|Loss: 0.8202881217002869:  84%|████████▎ | 5415/6470 [1:07:43<12:44,  1.38it/s]
2|5416|Loss: 0.8202881217002869:  84%|████████▎ | 5416/6470 [1:07:43<13:27,  1.31it/s]
2|5417|Loss: 0.7199475169181824:  84%|████████▎ | 5416/6470 [1:07:44<13:27,  1.31it/s]
2|5417|Loss: 0.7199475169181824:  84%|████████▎ | 5417/6470 [1:07:44<13:25,  1.31it/s]
2|5418|Loss: 0.774049699306488:  84%|████████▎ | 5417/6470 [1:07:44<13:25,  1.31it/s] 
2|5418|Loss: 0.774049699306488:  84%|████████▎ | 5418/6470 [1:07:45<14:37,  1.20it/s]
2|5419|Loss: 0.7942743897438049:  84%|████████▎ | 5418/6470 [1:07:45<14:37,  1.20it/s]
2|5419|Loss: 0.7942743897438049:  84%|████████▍ | 5419/6470 [1:07:46<14:40,  1.19it/s]
2|5420|Loss: 0.8028333187103271:  84%|████████▍ | 5419/6470 [1:07:46<14:40,  1.19it/s]
2|5420|Loss: 0.8028333187103271:  84%|████████▍ | 5420/6470 [1:07:46<14:02,  1.25it/s]
2|5421|Loss: 0.8850446343421936:  84%|████████▍ | 5420/6470 [1:07:47<14:02,  1.25it/s]
2|5421|Loss: 0.8850446343421936:  84%|████████▍ | 5421/6470 [1:07:47<13:56,  1.25it/s]
2|5422|Loss: 0.7574355006217957:  84%|████████▍ | 5421/6470 [1:07:48<13:56,  1.25it/s]
2|5422|Loss: 0.7574355006217957:  84%|████████▍ | 5422/6470 [1:07:48<13:43,  1.27it/s]
2|5423|Loss: 0.8040420413017273:  84%|████████▍ | 5422/6470 [1:07:48<13:43,  1.27it/s]
2|5423|Loss: 0.8040420413017273:  84%|████████▍ | 5423/6470 [1:07:49<14:28,  1.21it/s]
2|5424|Loss: 0.8334973454475403:  84%|████████▍ | 5423/6470 [1:07:49<14:28,  1.21it/s]
2|5424|Loss: 0.8334973454475403:  84%|████████▍ | 5424/6470 [1:07:50<14:04,  1.24it/s]
2|5425|Loss: 0.6815605163574219:  84%|████████▍ | 5424/6470 [1:07:50<14:04,  1.24it/s]
2|5425|Loss: 0.6815605163574219:  84%|████████▍ | 5425/6470 [1:07:50<14:18,  1.22it/s]
2|5426|Loss: 0.7672694325447083:  84%|████████▍ | 5425/6470 [1:07:51<14:18,  1.22it/s]
2|5426|Loss: 0.7672694325447083:  84%|████████▍ | 5426/6470 [1:07:51<14:03,  1.24it/s]
2|5427|Loss: 0.9931806325912476:  84%|████████▍ | 5426/6470 [1:07:52<14:03,  1.24it/s]
2|5427|Loss: 0.9931806325912476:  84%|████████▍ | 5427/6470 [1:07:52<14:02,  1.24it/s]
2|5428|Loss: 0.749597430229187:  84%|████████▍ | 5427/6470 [1:07:53<14:02,  1.24it/s] 
2|5428|Loss: 0.749597430229187:  84%|████████▍ | 5428/6470 [1:07:53<13:54,  1.25it/s]
2|5429|Loss: 0.9097430109977722:  84%|████████▍ | 5428/6470 [1:07:53<13:54,  1.25it/s]
2|5429|Loss: 0.9097430109977722:  84%|████████▍ | 5429/6470 [1:07:54<13:50,  1.25it/s]
2|5430|Loss: 0.7903639674186707:  84%|████████▍ | 5429/6470 [1:07:54<13:50,  1.25it/s]
2|5430|Loss: 0.7903639674186707:  84%|████████▍ | 5430/6470 [1:07:54<13:50,  1.25it/s]
2|5431|Loss: 0.8411535620689392:  84%|████████▍ | 5430/6470 [1:07:55<13:50,  1.25it/s]
2|5431|Loss: 0.8411535620689392:  84%|████████▍ | 5431/6470 [1:07:55<13:16,  1.30it/s]
2|5432|Loss: 0.819817841053009:  84%|████████▍ | 5431/6470 [1:07:56<13:16,  1.30it/s] 
2|5432|Loss: 0.819817841053009:  84%|████████▍ | 5432/6470 [1:07:56<13:08,  1.32it/s]
2|5433|Loss: 0.8303024768829346:  84%|████████▍ | 5432/6470 [1:07:56<13:08,  1.32it/s]
2|5433|Loss: 0.8303024768829346:  84%|████████▍ | 5433/6470 [1:07:56<12:10,  1.42it/s]
2|5434|Loss: 0.7662377953529358:  84%|████████▍ | 5433/6470 [1:07:57<12:10,  1.42it/s]
2|5434|Loss: 0.7662377953529358:  84%|████████▍ | 5434/6470 [1:07:57<11:47,  1.46it/s]
2|5435|Loss: 0.8499220609664917:  84%|████████▍ | 5434/6470 [1:07:57<11:47,  1.46it/s]
2|5435|Loss: 0.8499220609664917:  84%|████████▍ | 5435/6470 [1:07:58<11:31,  1.50it/s]
2|5436|Loss: 0.7178847789764404:  84%|████████▍ | 5435/6470 [1:07:58<11:31,  1.50it/s]
2|5436|Loss: 0.7178847789764404:  84%|████████▍ | 5436/6470 [1:07:58<12:01,  1.43it/s]
2|5437|Loss: 0.8352751135826111:  84%|████████▍ | 5436/6470 [1:07:59<12:01,  1.43it/s]
2|5437|Loss: 0.8352751135826111:  84%|████████▍ | 5437/6470 [1:07:59<12:10,  1.41it/s]
2|5438|Loss: 0.8738273978233337:  84%|████████▍ | 5437/6470 [1:08:00<12:10,  1.41it/s]
2|5438|Loss: 0.8738273978233337:  84%|████████▍ | 5438/6470 [1:08:00<12:11,  1.41it/s]
2|5439|Loss: 1.0174527168273926:  84%|████████▍ | 5438/6470 [1:08:00<12:11,  1.41it/s]
2|5439|Loss: 1.0174527168273926:  84%|████████▍ | 5439/6470 [1:08:01<11:41,  1.47it/s]
2|5440|Loss: 0.7418060898780823:  84%|████████▍ | 5439/6470 [1:08:01<11:41,  1.47it/s]
2|5440|Loss: 0.7418060898780823:  84%|████████▍ | 5440/6470 [1:08:01<10:59,  1.56it/s]
2|5441|Loss: 0.8080772161483765:  84%|████████▍ | 5440/6470 [1:08:01<10:59,  1.56it/s]
2|5441|Loss: 0.8080772161483765:  84%|████████▍ | 5441/6470 [1:08:02<11:27,  1.50it/s]
2|5442|Loss: 0.7593576908111572:  84%|████████▍ | 5441/6470 [1:08:02<11:27,  1.50it/s]
2|5442|Loss: 0.7593576908111572:  84%|████████▍ | 5442/6470 [1:08:03<11:32,  1.49it/s]
2|5443|Loss: 0.89096999168396:  84%|████████▍ | 5442/6470 [1:08:03<11:32,  1.49it/s]  
2|5443|Loss: 0.89096999168396:  84%|████████▍ | 5443/6470 [1:08:03<11:29,  1.49it/s]
2|5444|Loss: 0.7662608623504639:  84%|████████▍ | 5443/6470 [1:08:04<11:29,  1.49it/s]
2|5444|Loss: 0.7662608623504639:  84%|████████▍ | 5444/6470 [1:08:04<12:11,  1.40it/s]
2|5445|Loss: 0.9337718486785889:  84%|████████▍ | 5444/6470 [1:08:04<12:11,  1.40it/s]
2|5445|Loss: 0.9337718486785889:  84%|████████▍ | 5445/6470 [1:08:05<12:15,  1.39it/s]
2|5446|Loss: 0.8652603030204773:  84%|████████▍ | 5445/6470 [1:08:05<12:15,  1.39it/s]
2|5446|Loss: 0.8652603030204773:  84%|████████▍ | 5446/6470 [1:08:06<12:58,  1.32it/s]
2|5447|Loss: 0.9085585474967957:  84%|████████▍ | 5446/6470 [1:08:06<12:58,  1.32it/s]
2|5447|Loss: 0.9085585474967957:  84%|████████▍ | 5447/6470 [1:08:06<12:16,  1.39it/s]
2|5448|Loss: 0.8688650131225586:  84%|████████▍ | 5447/6470 [1:08:07<12:16,  1.39it/s]
2|5448|Loss: 0.8688650131225586:  84%|████████▍ | 5448/6470 [1:08:07<11:42,  1.46it/s]
2|5449|Loss: 0.7190024256706238:  84%|████████▍ | 5448/6470 [1:08:07<11:42,  1.46it/s]
2|5449|Loss: 0.7190024256706238:  84%|████████▍ | 5449/6470 [1:08:08<13:05,  1.30it/s]
2|5450|Loss: 0.7157700061798096:  84%|████████▍ | 5449/6470 [1:08:08<13:05,  1.30it/s]
2|5450|Loss: 0.7157700061798096:  84%|████████▍ | 5450/6470 [1:08:09<13:21,  1.27it/s]
2|5451|Loss: 0.7789970636367798:  84%|████████▍ | 5450/6470 [1:08:09<13:21,  1.27it/s]
2|5451|Loss: 0.7789970636367798:  84%|████████▍ | 5451/6470 [1:08:09<12:31,  1.36it/s]
2|5452|Loss: 0.7879945039749146:  84%|████████▍ | 5451/6470 [1:08:10<12:31,  1.36it/s]
2|5452|Loss: 0.7879945039749146:  84%|████████▍ | 5452/6470 [1:08:10<12:13,  1.39it/s]
2|5453|Loss: 0.847933828830719:  84%|████████▍ | 5452/6470 [1:08:10<12:13,  1.39it/s] 
2|5453|Loss: 0.847933828830719:  84%|████████▍ | 5453/6470 [1:08:11<12:34,  1.35it/s]
2|5454|Loss: 0.7775669693946838:  84%|████████▍ | 5453/6470 [1:08:11<12:34,  1.35it/s]
2|5454|Loss: 0.7775669693946838:  84%|████████▍ | 5454/6470 [1:08:12<13:18,  1.27it/s]
2|5455|Loss: 0.9018341898918152:  84%|████████▍ | 5454/6470 [1:08:12<13:18,  1.27it/s]
2|5455|Loss: 0.9018341898918152:  84%|████████▍ | 5455/6470 [1:08:12<12:35,  1.34it/s]
2|5456|Loss: 0.8781861662864685:  84%|████████▍ | 5455/6470 [1:08:13<12:35,  1.34it/s]
2|5456|Loss: 0.8781861662864685:  84%|████████▍ | 5456/6470 [1:08:13<12:26,  1.36it/s]
2|5457|Loss: 0.9772908687591553:  84%|████████▍ | 5456/6470 [1:08:13<12:26,  1.36it/s]
2|5457|Loss: 0.9772908687591553:  84%|████████▍ | 5457/6470 [1:08:14<13:15,  1.27it/s]
2|5458|Loss: 0.5995514988899231:  84%|████████▍ | 5457/6470 [1:08:14<13:15,  1.27it/s]
2|5458|Loss: 0.5995514988899231:  84%|████████▍ | 5458/6470 [1:08:14<11:50,  1.42it/s]
2|5459|Loss: 0.8881431221961975:  84%|████████▍ | 5458/6470 [1:08:15<11:50,  1.42it/s]
2|5459|Loss: 0.8881431221961975:  84%|████████▍ | 5459/6470 [1:08:15<10:46,  1.56it/s]
2|5460|Loss: 0.6940876841545105:  84%|████████▍ | 5459/6470 [1:08:15<10:46,  1.56it/s]
2|5460|Loss: 0.6940876841545105:  84%|████████▍ | 5460/6470 [1:08:16<11:20,  1.49it/s]
2|5461|Loss: 0.7782812714576721:  84%|████████▍ | 5460/6470 [1:08:16<11:20,  1.49it/s]
2|5461|Loss: 0.7782812714576721:  84%|████████▍ | 5461/6470 [1:08:17<12:26,  1.35it/s]
2|5462|Loss: 0.8028676509857178:  84%|████████▍ | 5461/6470 [1:08:17<12:26,  1.35it/s]
2|5462|Loss: 0.8028676509857178:  84%|████████▍ | 5462/6470 [1:08:17<13:06,  1.28it/s]
2|5463|Loss: 0.7602253556251526:  84%|████████▍ | 5462/6470 [1:08:18<13:06,  1.28it/s]
2|5463|Loss: 0.7602253556251526:  84%|████████▍ | 5463/6470 [1:08:18<11:38,  1.44it/s]
2|5464|Loss: 0.9047713279724121:  84%|████████▍ | 5463/6470 [1:08:18<11:38,  1.44it/s]
2|5464|Loss: 0.9047713279724121:  84%|████████▍ | 5464/6470 [1:08:19<11:39,  1.44it/s]
2|5465|Loss: 0.7433566451072693:  84%|████████▍ | 5464/6470 [1:08:19<11:39,  1.44it/s]
2|5465|Loss: 0.7433566451072693:  84%|████████▍ | 5465/6470 [1:08:19<12:32,  1.34it/s]
2|5466|Loss: 0.8200351595878601:  84%|████████▍ | 5465/6470 [1:08:20<12:32,  1.34it/s]
2|5466|Loss: 0.8200351595878601:  84%|████████▍ | 5466/6470 [1:08:20<12:19,  1.36it/s]
2|5467|Loss: 0.755685567855835:  84%|████████▍ | 5466/6470 [1:08:21<12:19,  1.36it/s] 
2|5467|Loss: 0.755685567855835:  84%|████████▍ | 5467/6470 [1:08:21<13:02,  1.28it/s]
2|5468|Loss: 0.8907291293144226:  84%|████████▍ | 5467/6470 [1:08:21<13:02,  1.28it/s]
2|5468|Loss: 0.8907291293144226:  85%|████████▍ | 5468/6470 [1:08:22<13:12,  1.26it/s]
2|5469|Loss: 0.7870643734931946:  85%|████████▍ | 5468/6470 [1:08:22<13:12,  1.26it/s]
2|5469|Loss: 0.7870643734931946:  85%|████████▍ | 5469/6470 [1:08:23<13:09,  1.27it/s]
2|5470|Loss: 0.7411652207374573:  85%|████████▍ | 5469/6470 [1:08:23<13:09,  1.27it/s]
2|5470|Loss: 0.7411652207374573:  85%|████████▍ | 5470/6470 [1:08:23<12:47,  1.30it/s]
2|5471|Loss: 0.7471778988838196:  85%|████████▍ | 5470/6470 [1:08:24<12:47,  1.30it/s]
2|5471|Loss: 0.7471778988838196:  85%|████████▍ | 5471/6470 [1:08:24<12:24,  1.34it/s]
2|5472|Loss: 0.8292173743247986:  85%|████████▍ | 5471/6470 [1:08:24<12:24,  1.34it/s]
2|5472|Loss: 0.8292173743247986:  85%|████████▍ | 5472/6470 [1:08:25<12:51,  1.29it/s]
2|5473|Loss: 0.7188363075256348:  85%|████████▍ | 5472/6470 [1:08:25<12:51,  1.29it/s]
2|5473|Loss: 0.7188363075256348:  85%|████████▍ | 5473/6470 [1:08:26<14:04,  1.18it/s]
2|5474|Loss: 0.7811477780342102:  85%|████████▍ | 5473/6470 [1:08:26<14:04,  1.18it/s]
2|5474|Loss: 0.7811477780342102:  85%|████████▍ | 5474/6470 [1:08:27<13:45,  1.21it/s]
2|5475|Loss: 0.8838834762573242:  85%|████████▍ | 5474/6470 [1:08:27<13:45,  1.21it/s]
2|5475|Loss: 0.8838834762573242:  85%|████████▍ | 5475/6470 [1:08:27<13:31,  1.23it/s]
2|5476|Loss: 0.8111234903335571:  85%|████████▍ | 5475/6470 [1:08:28<13:31,  1.23it/s]
2|5476|Loss: 0.8111234903335571:  85%|████████▍ | 5476/6470 [1:08:28<13:24,  1.24it/s]
2|5477|Loss: 0.8029388785362244:  85%|████████▍ | 5476/6470 [1:08:29<13:24,  1.24it/s]
2|5477|Loss: 0.8029388785362244:  85%|████████▍ | 5477/6470 [1:08:29<13:18,  1.24it/s]
2|5478|Loss: 0.8326345682144165:  85%|████████▍ | 5477/6470 [1:08:29<13:18,  1.24it/s]
2|5478|Loss: 0.8326345682144165:  85%|████████▍ | 5478/6470 [1:08:30<13:15,  1.25it/s]
2|5479|Loss: 0.8623338341712952:  85%|████████▍ | 5478/6470 [1:08:30<13:15,  1.25it/s]
2|5479|Loss: 0.8623338341712952:  85%|████████▍ | 5479/6470 [1:08:31<12:50,  1.29it/s]
2|5480|Loss: 0.8425498008728027:  85%|████████▍ | 5479/6470 [1:08:31<12:50,  1.29it/s]
2|5480|Loss: 0.8425498008728027:  85%|████████▍ | 5480/6470 [1:08:32<13:37,  1.21it/s]
2|5481|Loss: 0.8824512958526611:  85%|████████▍ | 5480/6470 [1:08:32<13:37,  1.21it/s]
2|5481|Loss: 0.8824512958526611:  85%|████████▍ | 5481/6470 [1:08:32<13:18,  1.24it/s]
2|5482|Loss: 0.9270673394203186:  85%|████████▍ | 5481/6470 [1:08:33<13:18,  1.24it/s]
2|5482|Loss: 0.9270673394203186:  85%|████████▍ | 5482/6470 [1:08:33<13:42,  1.20it/s]
2|5483|Loss: 0.8513230681419373:  85%|████████▍ | 5482/6470 [1:08:34<13:42,  1.20it/s]
2|5483|Loss: 0.8513230681419373:  85%|████████▍ | 5483/6470 [1:08:34<13:43,  1.20it/s]
2|5484|Loss: 0.9047250151634216:  85%|████████▍ | 5483/6470 [1:08:34<13:43,  1.20it/s]
2|5484|Loss: 0.9047250151634216:  85%|████████▍ | 5484/6470 [1:08:35<13:29,  1.22it/s]
2|5485|Loss: 0.920842170715332:  85%|████████▍ | 5484/6470 [1:08:35<13:29,  1.22it/s] 
2|5485|Loss: 0.920842170715332:  85%|████████▍ | 5485/6470 [1:08:36<12:59,  1.26it/s]
2|5486|Loss: 0.8670503497123718:  85%|████████▍ | 5485/6470 [1:08:36<12:59,  1.26it/s]
2|5486|Loss: 0.8670503497123718:  85%|████████▍ | 5486/6470 [1:08:36<12:11,  1.34it/s]
2|5487|Loss: 0.6804648637771606:  85%|████████▍ | 5486/6470 [1:08:37<12:11,  1.34it/s]
2|5487|Loss: 0.6804648637771606:  85%|████████▍ | 5487/6470 [1:08:37<12:41,  1.29it/s]
2|5488|Loss: 0.8401069045066833:  85%|████████▍ | 5487/6470 [1:08:37<12:41,  1.29it/s]
2|5488|Loss: 0.8401069045066833:  85%|████████▍ | 5488/6470 [1:08:38<11:53,  1.38it/s]
2|5489|Loss: 0.7345912456512451:  85%|████████▍ | 5488/6470 [1:08:38<11:53,  1.38it/s]
2|5489|Loss: 0.7345912456512451:  85%|████████▍ | 5489/6470 [1:08:38<10:55,  1.50it/s]
2|5490|Loss: 0.6429639458656311:  85%|████████▍ | 5489/6470 [1:08:39<10:55,  1.50it/s]
2|5490|Loss: 0.6429639458656311:  85%|████████▍ | 5490/6470 [1:08:39<11:30,  1.42it/s]
2|5491|Loss: 0.8339826464653015:  85%|████████▍ | 5490/6470 [1:08:39<11:30,  1.42it/s]
2|5491|Loss: 0.8339826464653015:  85%|████████▍ | 5491/6470 [1:08:40<11:31,  1.41it/s]
2|5492|Loss: 0.7881389260292053:  85%|████████▍ | 5491/6470 [1:08:40<11:31,  1.41it/s]
2|5492|Loss: 0.7881389260292053:  85%|████████▍ | 5492/6470 [1:08:40<10:40,  1.53it/s]
2|5493|Loss: 0.774543285369873:  85%|████████▍ | 5492/6470 [1:08:41<10:40,  1.53it/s] 
2|5493|Loss: 0.774543285369873:  85%|████████▍ | 5493/6470 [1:08:41<10:32,  1.55it/s]
2|5494|Loss: 0.7731711268424988:  85%|████████▍ | 5493/6470 [1:08:41<10:32,  1.55it/s]
2|5494|Loss: 0.7731711268424988:  85%|████████▍ | 5494/6470 [1:08:42<11:02,  1.47it/s]
2|5495|Loss: 0.815981924533844:  85%|████████▍ | 5494/6470 [1:08:42<11:02,  1.47it/s] 
2|5495|Loss: 0.815981924533844:  85%|████████▍ | 5495/6470 [1:08:42<10:22,  1.57it/s]
2|5496|Loss: 0.5582138895988464:  85%|████████▍ | 5495/6470 [1:08:42<10:22,  1.57it/s]
2|5496|Loss: 0.5582138895988464:  85%|████████▍ | 5496/6470 [1:08:43<09:54,  1.64it/s]
2|5497|Loss: 0.8454206585884094:  85%|████████▍ | 5496/6470 [1:08:43<09:54,  1.64it/s]
2|5497|Loss: 0.8454206585884094:  85%|████████▍ | 5497/6470 [1:08:43<10:12,  1.59it/s]
2|5498|Loss: 0.699659526348114:  85%|████████▍ | 5497/6470 [1:08:44<10:12,  1.59it/s] 
2|5498|Loss: 0.699659526348114:  85%|████████▍ | 5498/6470 [1:08:44<10:34,  1.53it/s]
2|5499|Loss: 0.908973217010498:  85%|████████▍ | 5498/6470 [1:08:44<10:34,  1.53it/s]
2|5499|Loss: 0.908973217010498:  85%|████████▍ | 5499/6470 [1:08:45<10:02,  1.61it/s]
2|5500|Loss: 0.7547087073326111:  85%|████████▍ | 5499/6470 [1:08:45<10:02,  1.61it/s]
2|5500|Loss: 0.7547087073326111:  85%|████████▌ | 5500/6470 [1:08:45<10:29,  1.54it/s]
2|5501|Loss: 0.9144772887229919:  85%|████████▌ | 5500/6470 [1:08:46<10:29,  1.54it/s]
2|5501|Loss: 0.9144772887229919:  85%|████████▌ | 5501/6470 [1:08:46<10:29,  1.54it/s]
2|5502|Loss: 0.875632643699646:  85%|████████▌ | 5501/6470 [1:08:46<10:29,  1.54it/s] 
2|5502|Loss: 0.875632643699646:  85%|████████▌ | 5502/6470 [1:08:47<10:57,  1.47it/s]
2|5503|Loss: 0.8494712710380554:  85%|████████▌ | 5502/6470 [1:08:47<10:57,  1.47it/s]
2|5503|Loss: 0.8494712710380554:  85%|████████▌ | 5503/6470 [1:08:47<10:57,  1.47it/s]
2|5504|Loss: 1.023099422454834:  85%|████████▌ | 5503/6470 [1:08:48<10:57,  1.47it/s] 
2|5504|Loss: 1.023099422454834:  85%|████████▌ | 5504/6470 [1:08:48<11:13,  1.43it/s]
2|5505|Loss: 0.881493091583252:  85%|████████▌ | 5504/6470 [1:08:49<11:13,  1.43it/s]
2|5505|Loss: 0.881493091583252:  85%|████████▌ | 5505/6470 [1:08:49<11:09,  1.44it/s]
2|5506|Loss: 0.9648422598838806:  85%|████████▌ | 5505/6470 [1:08:49<11:09,  1.44it/s]
2|5506|Loss: 0.9648422598838806:  85%|████████▌ | 5506/6470 [1:08:50<11:12,  1.43it/s]
2|5507|Loss: 0.9185822606086731:  85%|████████▌ | 5506/6470 [1:08:50<11:12,  1.43it/s]
2|5507|Loss: 0.9185822606086731:  85%|████████▌ | 5507/6470 [1:08:50<11:02,  1.45it/s]
2|5508|Loss: 0.9081087708473206:  85%|████████▌ | 5507/6470 [1:08:51<11:02,  1.45it/s]
2|5508|Loss: 0.9081087708473206:  85%|████████▌ | 5508/6470 [1:08:51<10:41,  1.50it/s]
2|5509|Loss: 0.9270711541175842:  85%|████████▌ | 5508/6470 [1:08:51<10:41,  1.50it/s]
2|5509|Loss: 0.9270711541175842:  85%|████████▌ | 5509/6470 [1:08:51<10:43,  1.49it/s]
2|5510|Loss: 0.9357349276542664:  85%|████████▌ | 5509/6470 [1:08:52<10:43,  1.49it/s]
2|5510|Loss: 0.9357349276542664:  85%|████████▌ | 5510/6470 [1:08:52<11:12,  1.43it/s]
2|5511|Loss: 0.8483002781867981:  85%|████████▌ | 5510/6470 [1:08:53<11:12,  1.43it/s]
2|5511|Loss: 0.8483002781867981:  85%|████████▌ | 5511/6470 [1:08:53<11:13,  1.42it/s]
2|5512|Loss: 0.790725827217102:  85%|████████▌ | 5511/6470 [1:08:53<11:13,  1.42it/s] 
2|5512|Loss: 0.790725827217102:  85%|████████▌ | 5512/6470 [1:08:53<10:27,  1.53it/s]
2|5513|Loss: 0.7977294325828552:  85%|████████▌ | 5512/6470 [1:08:54<10:27,  1.53it/s]
2|5513|Loss: 0.7977294325828552:  85%|████████▌ | 5513/6470 [1:08:54<10:47,  1.48it/s]
2|5514|Loss: 0.8449143767356873:  85%|████████▌ | 5513/6470 [1:08:55<10:47,  1.48it/s]
2|5514|Loss: 0.8449143767356873:  85%|████████▌ | 5514/6470 [1:08:55<10:42,  1.49it/s]
2|5515|Loss: 0.8370025157928467:  85%|████████▌ | 5514/6470 [1:08:55<10:42,  1.49it/s]
2|5515|Loss: 0.8370025157928467:  85%|████████▌ | 5515/6470 [1:08:56<11:30,  1.38it/s]
2|5516|Loss: 0.7740738391876221:  85%|████████▌ | 5515/6470 [1:08:56<11:30,  1.38it/s]
2|5516|Loss: 0.7740738391876221:  85%|████████▌ | 5516/6470 [1:08:56<11:26,  1.39it/s]
2|5517|Loss: 0.7444893717765808:  85%|████████▌ | 5516/6470 [1:08:57<11:26,  1.39it/s]
2|5517|Loss: 0.7444893717765808:  85%|████████▌ | 5517/6470 [1:08:57<11:10,  1.42it/s]
2|5518|Loss: 0.7965993881225586:  85%|████████▌ | 5517/6470 [1:08:58<11:10,  1.42it/s]
2|5518|Loss: 0.7965993881225586:  85%|████████▌ | 5518/6470 [1:08:58<11:52,  1.34it/s]
2|5519|Loss: 0.8137640357017517:  85%|████████▌ | 5518/6470 [1:08:58<11:52,  1.34it/s]
2|5519|Loss: 0.8137640357017517:  85%|████████▌ | 5519/6470 [1:08:59<11:58,  1.32it/s]
2|5520|Loss: 0.8072313070297241:  85%|████████▌ | 5519/6470 [1:08:59<11:58,  1.32it/s]
2|5520|Loss: 0.8072313070297241:  85%|████████▌ | 5520/6470 [1:08:59<11:44,  1.35it/s]
2|5521|Loss: 0.9210585951805115:  85%|████████▌ | 5520/6470 [1:09:00<11:44,  1.35it/s]
2|5521|Loss: 0.9210585951805115:  85%|████████▌ | 5521/6470 [1:09:00<12:30,  1.26it/s]
2|5522|Loss: 0.8096837401390076:  85%|████████▌ | 5521/6470 [1:09:01<12:30,  1.26it/s]
2|5522|Loss: 0.8096837401390076:  85%|████████▌ | 5522/6470 [1:09:01<11:52,  1.33it/s]
2|5523|Loss: 0.796441376209259:  85%|████████▌ | 5522/6470 [1:09:01<11:52,  1.33it/s] 
2|5523|Loss: 0.796441376209259:  85%|████████▌ | 5523/6470 [1:09:02<11:53,  1.33it/s]
2|5524|Loss: 0.8796703219413757:  85%|████████▌ | 5523/6470 [1:09:02<11:53,  1.33it/s]
2|5524|Loss: 0.8796703219413757:  85%|████████▌ | 5524/6470 [1:09:03<12:41,  1.24it/s]
2|5525|Loss: 0.7417211532592773:  85%|████████▌ | 5524/6470 [1:09:03<12:41,  1.24it/s]
2|5525|Loss: 0.7417211532592773:  85%|████████▌ | 5525/6470 [1:09:04<13:48,  1.14it/s]
2|5526|Loss: 0.7684080004692078:  85%|████████▌ | 5525/6470 [1:09:04<13:48,  1.14it/s]
2|5526|Loss: 0.7684080004692078:  85%|████████▌ | 5526/6470 [1:09:05<14:09,  1.11it/s]
2|5527|Loss: 0.7697374224662781:  85%|████████▌ | 5526/6470 [1:09:05<14:09,  1.11it/s]
2|5527|Loss: 0.7697374224662781:  85%|████████▌ | 5527/6470 [1:09:05<13:18,  1.18it/s]
2|5528|Loss: 1.1808792352676392:  85%|████████▌ | 5527/6470 [1:09:06<13:18,  1.18it/s]
2|5528|Loss: 1.1808792352676392:  85%|████████▌ | 5528/6470 [1:09:06<13:22,  1.17it/s]
2|5529|Loss: 0.7899727821350098:  85%|████████▌ | 5528/6470 [1:09:07<13:22,  1.17it/s]
2|5529|Loss: 0.7899727821350098:  85%|████████▌ | 5529/6470 [1:09:07<13:04,  1.20it/s]
2|5530|Loss: 1.0152004957199097:  85%|████████▌ | 5529/6470 [1:09:07<13:04,  1.20it/s]
2|5530|Loss: 1.0152004957199097:  85%|████████▌ | 5530/6470 [1:09:08<12:22,  1.27it/s]
2|5531|Loss: 0.8141369223594666:  85%|████████▌ | 5530/6470 [1:09:08<12:22,  1.27it/s]
2|5531|Loss: 0.8141369223594666:  85%|████████▌ | 5531/6470 [1:09:09<12:15,  1.28it/s]
2|5532|Loss: 0.9811554551124573:  85%|████████▌ | 5531/6470 [1:09:09<12:15,  1.28it/s]
2|5532|Loss: 0.9811554551124573:  86%|████████▌ | 5532/6470 [1:09:09<12:55,  1.21it/s]
2|5533|Loss: 0.8318796753883362:  86%|████████▌ | 5532/6470 [1:09:10<12:55,  1.21it/s]
2|5533|Loss: 0.8318796753883362:  86%|████████▌ | 5533/6470 [1:09:10<13:15,  1.18it/s]
2|5534|Loss: 0.7363290190696716:  86%|████████▌ | 5533/6470 [1:09:11<13:15,  1.18it/s]
2|5534|Loss: 0.7363290190696716:  86%|████████▌ | 5534/6470 [1:09:11<12:37,  1.24it/s]
2|5535|Loss: 0.8774397373199463:  86%|████████▌ | 5534/6470 [1:09:11<12:37,  1.24it/s]
2|5535|Loss: 0.8774397373199463:  86%|████████▌ | 5535/6470 [1:09:12<12:30,  1.25it/s]
2|5536|Loss: 1.0744929313659668:  86%|████████▌ | 5535/6470 [1:09:12<12:30,  1.25it/s]
2|5536|Loss: 1.0744929313659668:  86%|████████▌ | 5536/6470 [1:09:13<12:58,  1.20it/s]
2|5537|Loss: 0.8480343818664551:  86%|████████▌ | 5536/6470 [1:09:13<12:58,  1.20it/s]
2|5537|Loss: 0.8480343818664551:  86%|████████▌ | 5537/6470 [1:09:14<12:57,  1.20it/s]
2|5538|Loss: 0.7261496186256409:  86%|████████▌ | 5537/6470 [1:09:14<12:57,  1.20it/s]
2|5538|Loss: 0.7261496186256409:  86%|████████▌ | 5538/6470 [1:09:14<12:19,  1.26it/s]
2|5539|Loss: 0.8063386082649231:  86%|████████▌ | 5538/6470 [1:09:15<12:19,  1.26it/s]
2|5539|Loss: 0.8063386082649231:  86%|████████▌ | 5539/6470 [1:09:15<11:39,  1.33it/s]
2|5540|Loss: 0.7594806551933289:  86%|████████▌ | 5539/6470 [1:09:15<11:39,  1.33it/s]
2|5540|Loss: 0.7594806551933289:  86%|████████▌ | 5540/6470 [1:09:16<12:18,  1.26it/s]
2|5541|Loss: 0.7019181251525879:  86%|████████▌ | 5540/6470 [1:09:16<12:18,  1.26it/s]
2|5541|Loss: 0.7019181251525879:  86%|████████▌ | 5541/6470 [1:09:17<11:59,  1.29it/s]
2|5542|Loss: 0.7561360597610474:  86%|████████▌ | 5541/6470 [1:09:17<11:59,  1.29it/s]
2|5542|Loss: 0.7561360597610474:  86%|████████▌ | 5542/6470 [1:09:17<11:30,  1.34it/s]
2|5543|Loss: 0.873384416103363:  86%|████████▌ | 5542/6470 [1:09:18<11:30,  1.34it/s] 
2|5543|Loss: 0.873384416103363:  86%|████████▌ | 5543/6470 [1:09:18<11:38,  1.33it/s]
2|5544|Loss: 0.8517115712165833:  86%|████████▌ | 5543/6470 [1:09:18<11:38,  1.33it/s]
2|5544|Loss: 0.8517115712165833:  86%|████████▌ | 5544/6470 [1:09:19<12:14,  1.26it/s]
2|5545|Loss: 0.7267792224884033:  86%|████████▌ | 5544/6470 [1:09:19<12:14,  1.26it/s]
2|5545|Loss: 0.7267792224884033:  86%|████████▌ | 5545/6470 [1:09:19<11:12,  1.38it/s]
2|5546|Loss: 0.85329270362854:  86%|████████▌ | 5545/6470 [1:09:20<11:12,  1.38it/s]  
2|5546|Loss: 0.85329270362854:  86%|████████▌ | 5546/6470 [1:09:20<11:10,  1.38it/s]
2|5547|Loss: 0.8544111847877502:  86%|████████▌ | 5546/6470 [1:09:21<11:10,  1.38it/s]
2|5547|Loss: 0.8544111847877502:  86%|████████▌ | 5547/6470 [1:09:21<11:29,  1.34it/s]
2|5548|Loss: 0.8208560347557068:  86%|████████▌ | 5547/6470 [1:09:21<11:29,  1.34it/s]
2|5548|Loss: 0.8208560347557068:  86%|████████▌ | 5548/6470 [1:09:22<11:25,  1.35it/s]
2|5549|Loss: 0.9511077404022217:  86%|████████▌ | 5548/6470 [1:09:22<11:25,  1.35it/s]
2|5549|Loss: 0.9511077404022217:  86%|████████▌ | 5549/6470 [1:09:22<11:12,  1.37it/s]
2|5550|Loss: 0.7183553576469421:  86%|████████▌ | 5549/6470 [1:09:23<11:12,  1.37it/s]
2|5550|Loss: 0.7183553576469421:  86%|████████▌ | 5550/6470 [1:09:23<11:16,  1.36it/s]
2|5551|Loss: 0.847632884979248:  86%|████████▌ | 5550/6470 [1:09:24<11:16,  1.36it/s] 
2|5551|Loss: 0.847632884979248:  86%|████████▌ | 5551/6470 [1:09:24<11:34,  1.32it/s]
2|5552|Loss: 0.8112191557884216:  86%|████████▌ | 5551/6470 [1:09:24<11:34,  1.32it/s]
2|5552|Loss: 0.8112191557884216:  86%|████████▌ | 5552/6470 [1:09:25<12:11,  1.26it/s]
2|5553|Loss: 0.8304809331893921:  86%|████████▌ | 5552/6470 [1:09:25<12:11,  1.26it/s]
2|5553|Loss: 0.8304809331893921:  86%|████████▌ | 5553/6470 [1:09:26<12:08,  1.26it/s]
2|5554|Loss: 1.045562744140625:  86%|████████▌ | 5553/6470 [1:09:26<12:08,  1.26it/s] 
2|5554|Loss: 1.045562744140625:  86%|████████▌ | 5554/6470 [1:09:26<11:55,  1.28it/s]
2|5555|Loss: 0.7530083060264587:  86%|████████▌ | 5554/6470 [1:09:27<11:55,  1.28it/s]
2|5555|Loss: 0.7530083060264587:  86%|████████▌ | 5555/6470 [1:09:27<11:11,  1.36it/s]
2|5556|Loss: 0.8573670387268066:  86%|████████▌ | 5555/6470 [1:09:27<11:11,  1.36it/s]
2|5556|Loss: 0.8573670387268066:  86%|████████▌ | 5556/6470 [1:09:28<11:14,  1.36it/s]
2|5557|Loss: 0.8570492267608643:  86%|████████▌ | 5556/6470 [1:09:28<11:14,  1.36it/s]
2|5557|Loss: 0.8570492267608643:  86%|████████▌ | 5557/6470 [1:09:29<11:53,  1.28it/s]
2|5558|Loss: 0.9367845058441162:  86%|████████▌ | 5557/6470 [1:09:29<11:53,  1.28it/s]
2|5558|Loss: 0.9367845058441162:  86%|████████▌ | 5558/6470 [1:09:29<11:42,  1.30it/s]
2|5559|Loss: 0.9257990121841431:  86%|████████▌ | 5558/6470 [1:09:30<11:42,  1.30it/s]
2|5559|Loss: 0.9257990121841431:  86%|████████▌ | 5559/6470 [1:09:30<12:18,  1.23it/s]
2|5560|Loss: 0.7720393538475037:  86%|████████▌ | 5559/6470 [1:09:31<12:18,  1.23it/s]
2|5560|Loss: 0.7720393538475037:  86%|████████▌ | 5560/6470 [1:09:31<12:05,  1.25it/s]
2|5561|Loss: 0.9038455486297607:  86%|████████▌ | 5560/6470 [1:09:32<12:05,  1.25it/s]
2|5561|Loss: 0.9038455486297607:  86%|████████▌ | 5561/6470 [1:09:32<12:04,  1.25it/s]
2|5562|Loss: 0.7549942135810852:  86%|████████▌ | 5561/6470 [1:09:32<12:04,  1.25it/s]
2|5562|Loss: 0.7549942135810852:  86%|████████▌ | 5562/6470 [1:09:33<11:37,  1.30it/s]
2|5563|Loss: 0.8543814420700073:  86%|████████▌ | 5562/6470 [1:09:33<11:37,  1.30it/s]
2|5563|Loss: 0.8543814420700073:  86%|████████▌ | 5563/6470 [1:09:33<11:31,  1.31it/s]
2|5564|Loss: 0.8387945890426636:  86%|████████▌ | 5563/6470 [1:09:34<11:31,  1.31it/s]
2|5564|Loss: 0.8387945890426636:  86%|████████▌ | 5564/6470 [1:09:34<11:38,  1.30it/s]
2|5565|Loss: 0.7100698947906494:  86%|████████▌ | 5564/6470 [1:09:35<11:38,  1.30it/s]
2|5565|Loss: 0.7100698947906494:  86%|████████▌ | 5565/6470 [1:09:35<11:44,  1.28it/s]
2|5566|Loss: 0.7754209041595459:  86%|████████▌ | 5565/6470 [1:09:35<11:44,  1.28it/s]
2|5566|Loss: 0.7754209041595459:  86%|████████▌ | 5566/6470 [1:09:36<12:02,  1.25it/s]
2|5567|Loss: 0.838437557220459:  86%|████████▌ | 5566/6470 [1:09:36<12:02,  1.25it/s] 
2|5567|Loss: 0.838437557220459:  86%|████████▌ | 5567/6470 [1:09:36<11:03,  1.36it/s]
2|5568|Loss: 0.8501957654953003:  86%|████████▌ | 5567/6470 [1:09:37<11:03,  1.36it/s]
2|5568|Loss: 0.8501957654953003:  86%|████████▌ | 5568/6470 [1:09:37<10:24,  1.44it/s]
2|5569|Loss: 0.8773344159126282:  86%|████████▌ | 5568/6470 [1:09:37<10:24,  1.44it/s]
2|5569|Loss: 0.8773344159126282:  86%|████████▌ | 5569/6470 [1:09:38<10:13,  1.47it/s]
2|5570|Loss: 0.75095534324646:  86%|████████▌ | 5569/6470 [1:09:38<10:13,  1.47it/s]  
2|5570|Loss: 0.75095534324646:  86%|████████▌ | 5570/6470 [1:09:38<10:28,  1.43it/s]
2|5571|Loss: 0.7633473873138428:  86%|████████▌ | 5570/6470 [1:09:39<10:28,  1.43it/s]
2|5571|Loss: 0.7633473873138428:  86%|████████▌ | 5571/6470 [1:09:39<11:13,  1.34it/s]
2|5572|Loss: 0.914324939250946:  86%|████████▌ | 5571/6470 [1:09:40<11:13,  1.34it/s] 
2|5572|Loss: 0.914324939250946:  86%|████████▌ | 5572/6470 [1:09:40<12:00,  1.25it/s]
2|5573|Loss: 0.8390137553215027:  86%|████████▌ | 5572/6470 [1:09:41<12:00,  1.25it/s]
2|5573|Loss: 0.8390137553215027:  86%|████████▌ | 5573/6470 [1:09:41<12:23,  1.21it/s]
2|5574|Loss: 0.8075044751167297:  86%|████████▌ | 5573/6470 [1:09:41<12:23,  1.21it/s]
2|5574|Loss: 0.8075044751167297:  86%|████████▌ | 5574/6470 [1:09:42<11:50,  1.26it/s]
2|5575|Loss: 0.7924119234085083:  86%|████████▌ | 5574/6470 [1:09:42<11:50,  1.26it/s]
2|5575|Loss: 0.7924119234085083:  86%|████████▌ | 5575/6470 [1:09:42<11:18,  1.32it/s]
2|5576|Loss: 0.885818600654602:  86%|████████▌ | 5575/6470 [1:09:43<11:18,  1.32it/s] 
2|5576|Loss: 0.885818600654602:  86%|████████▌ | 5576/6470 [1:09:43<10:14,  1.46it/s]
2|5577|Loss: 0.670026957988739:  86%|████████▌ | 5576/6470 [1:09:43<10:14,  1.46it/s]
2|5577|Loss: 0.670026957988739:  86%|████████▌ | 5577/6470 [1:09:43<09:29,  1.57it/s]
2|5578|Loss: 0.8778905868530273:  86%|████████▌ | 5577/6470 [1:09:44<09:29,  1.57it/s]
2|5578|Loss: 0.8778905868530273:  86%|████████▌ | 5578/6470 [1:09:44<09:50,  1.51it/s]
2|5579|Loss: 0.9184648394584656:  86%|████████▌ | 5578/6470 [1:09:45<09:50,  1.51it/s]
2|5579|Loss: 0.9184648394584656:  86%|████████▌ | 5579/6470 [1:09:45<09:55,  1.50it/s]
2|5580|Loss: 0.8429145812988281:  86%|████████▌ | 5579/6470 [1:09:45<09:55,  1.50it/s]
2|5580|Loss: 0.8429145812988281:  86%|████████▌ | 5580/6470 [1:09:46<10:21,  1.43it/s]
2|5581|Loss: 0.8460938334465027:  86%|████████▌ | 5580/6470 [1:09:46<10:21,  1.43it/s]
2|5581|Loss: 0.8460938334465027:  86%|████████▋ | 5581/6470 [1:09:46<10:28,  1.42it/s]
2|5582|Loss: 1.1063830852508545:  86%|████████▋ | 5581/6470 [1:09:47<10:28,  1.42it/s]
2|5582|Loss: 1.1063830852508545:  86%|████████▋ | 5582/6470 [1:09:47<10:16,  1.44it/s]
2|5583|Loss: 0.7251459360122681:  86%|████████▋ | 5582/6470 [1:09:48<10:16,  1.44it/s]
2|5583|Loss: 0.7251459360122681:  86%|████████▋ | 5583/6470 [1:09:48<11:31,  1.28it/s]
2|5584|Loss: 0.5923893451690674:  86%|████████▋ | 5583/6470 [1:09:49<11:31,  1.28it/s]
2|5584|Loss: 0.5923893451690674:  86%|████████▋ | 5584/6470 [1:09:49<12:02,  1.23it/s]
2|5585|Loss: 0.8807830214500427:  86%|████████▋ | 5584/6470 [1:09:49<12:02,  1.23it/s]
2|5585|Loss: 0.8807830214500427:  86%|████████▋ | 5585/6470 [1:09:50<11:42,  1.26it/s]
2|5586|Loss: 0.8237156867980957:  86%|████████▋ | 5585/6470 [1:09:50<11:42,  1.26it/s]
2|5586|Loss: 0.8237156867980957:  86%|████████▋ | 5586/6470 [1:09:50<11:32,  1.28it/s]
2|5587|Loss: 0.877234160900116:  86%|████████▋ | 5586/6470 [1:09:51<11:32,  1.28it/s] 
2|5587|Loss: 0.877234160900116:  86%|████████▋ | 5587/6470 [1:09:51<10:25,  1.41it/s]
2|5588|Loss: 0.7366266250610352:  86%|████████▋ | 5587/6470 [1:09:51<10:25,  1.41it/s]
2|5588|Loss: 0.7366266250610352:  86%|████████▋ | 5588/6470 [1:09:52<10:03,  1.46it/s]
2|5589|Loss: 0.75686115026474:  86%|████████▋ | 5588/6470 [1:09:52<10:03,  1.46it/s]  
2|5589|Loss: 0.75686115026474:  86%|████████▋ | 5589/6470 [1:09:52<10:43,  1.37it/s]
2|5590|Loss: 0.6604975461959839:  86%|████████▋ | 5589/6470 [1:09:53<10:43,  1.37it/s]
2|5590|Loss: 0.6604975461959839:  86%|████████▋ | 5590/6470 [1:09:53<11:25,  1.28it/s]
2|5591|Loss: 0.7578103542327881:  86%|████████▋ | 5590/6470 [1:09:54<11:25,  1.28it/s]
2|5591|Loss: 0.7578103542327881:  86%|████████▋ | 5591/6470 [1:09:54<11:15,  1.30it/s]
2|5592|Loss: 0.7040581703186035:  86%|████████▋ | 5591/6470 [1:09:54<11:15,  1.30it/s]
2|5592|Loss: 0.7040581703186035:  86%|████████▋ | 5592/6470 [1:09:55<11:06,  1.32it/s]
2|5593|Loss: 0.7652739882469177:  86%|████████▋ | 5592/6470 [1:09:55<11:06,  1.32it/s]
2|5593|Loss: 0.7652739882469177:  86%|████████▋ | 5593/6470 [1:09:55<10:38,  1.37it/s]
2|5594|Loss: 0.8321520090103149:  86%|████████▋ | 5593/6470 [1:09:56<10:38,  1.37it/s]
2|5594|Loss: 0.8321520090103149:  86%|████████▋ | 5594/6470 [1:09:56<11:10,  1.31it/s]
2|5595|Loss: 1.185705304145813:  86%|████████▋ | 5594/6470 [1:09:57<11:10,  1.31it/s] 
2|5595|Loss: 1.185705304145813:  86%|████████▋ | 5595/6470 [1:09:57<10:54,  1.34it/s]
2|5596|Loss: 0.9304274916648865:  86%|████████▋ | 5595/6470 [1:09:57<10:54,  1.34it/s]
2|5596|Loss: 0.9304274916648865:  86%|████████▋ | 5596/6470 [1:09:58<10:20,  1.41it/s]
2|5597|Loss: 0.8354110717773438:  86%|████████▋ | 5596/6470 [1:09:58<10:20,  1.41it/s]
2|5597|Loss: 0.8354110717773438:  87%|████████▋ | 5597/6470 [1:09:58<09:38,  1.51it/s]
2|5598|Loss: 0.6803589463233948:  87%|████████▋ | 5597/6470 [1:09:59<09:38,  1.51it/s]
2|5598|Loss: 0.6803589463233948:  87%|████████▋ | 5598/6470 [1:09:59<09:57,  1.46it/s]
2|5599|Loss: 0.8558789491653442:  87%|████████▋ | 5598/6470 [1:09:59<09:57,  1.46it/s]
2|5599|Loss: 0.8558789491653442:  87%|████████▋ | 5599/6470 [1:10:00<10:08,  1.43it/s]
2|5600|Loss: 0.8033205270767212:  87%|████████▋ | 5599/6470 [1:10:00<10:08,  1.43it/s]
2|5600|Loss: 0.8033205270767212:  87%|████████▋ | 5600/6470 [1:10:00<10:24,  1.39it/s]
2|5601|Loss: 0.8757449388504028:  87%|████████▋ | 5600/6470 [1:10:01<10:24,  1.39it/s]
2|5601|Loss: 0.8757449388504028:  87%|████████▋ | 5601/6470 [1:10:01<10:32,  1.37it/s]
2|5602|Loss: 0.8217208981513977:  87%|████████▋ | 5601/6470 [1:10:02<10:32,  1.37it/s]
2|5602|Loss: 0.8217208981513977:  87%|████████▋ | 5602/6470 [1:10:02<11:10,  1.29it/s]
2|5603|Loss: 0.9428264498710632:  87%|████████▋ | 5602/6470 [1:10:02<11:10,  1.29it/s]
2|5603|Loss: 0.9428264498710632:  87%|████████▋ | 5603/6470 [1:10:03<11:18,  1.28it/s]
2|5604|Loss: 0.8324927687644958:  87%|████████▋ | 5603/6470 [1:10:03<11:18,  1.28it/s]
2|5604|Loss: 0.8324927687644958:  87%|████████▋ | 5604/6470 [1:10:03<10:48,  1.34it/s]
2|5605|Loss: 0.8399885892868042:  87%|████████▋ | 5604/6470 [1:10:04<10:48,  1.34it/s]
2|5605|Loss: 0.8399885892868042:  87%|████████▋ | 5605/6470 [1:10:04<11:18,  1.28it/s]
2|5606|Loss: 0.7356138229370117:  87%|████████▋ | 5605/6470 [1:10:05<11:18,  1.28it/s]
2|5606|Loss: 0.7356138229370117:  87%|████████▋ | 5606/6470 [1:10:05<11:53,  1.21it/s]
2|5607|Loss: 0.7387843132019043:  87%|████████▋ | 5606/6470 [1:10:06<11:53,  1.21it/s]
2|5607|Loss: 0.7387843132019043:  87%|████████▋ | 5607/6470 [1:10:06<11:30,  1.25it/s]
2|5608|Loss: 0.857373058795929:  87%|████████▋ | 5607/6470 [1:10:06<11:30,  1.25it/s] 
2|5608|Loss: 0.857373058795929:  87%|████████▋ | 5608/6470 [1:10:07<11:04,  1.30it/s]
2|5609|Loss: 0.9652453660964966:  87%|████████▋ | 5608/6470 [1:10:07<11:04,  1.30it/s]
2|5609|Loss: 0.9652453660964966:  87%|████████▋ | 5609/6470 [1:10:07<11:02,  1.30it/s]
2|5610|Loss: 0.8407182097434998:  87%|████████▋ | 5609/6470 [1:10:08<11:02,  1.30it/s]
2|5610|Loss: 0.8407182097434998:  87%|████████▋ | 5610/6470 [1:10:08<10:30,  1.36it/s]
2|5611|Loss: 0.8282994031906128:  87%|████████▋ | 5610/6470 [1:10:08<10:30,  1.36it/s]
2|5611|Loss: 0.8282994031906128:  87%|████████▋ | 5611/6470 [1:10:09<10:15,  1.40it/s]
2|5612|Loss: 0.9020158052444458:  87%|████████▋ | 5611/6470 [1:10:09<10:15,  1.40it/s]
2|5612|Loss: 0.9020158052444458:  87%|████████▋ | 5612/6470 [1:10:10<10:45,  1.33it/s]
2|5613|Loss: 0.9703561663627625:  87%|████████▋ | 5612/6470 [1:10:10<10:45,  1.33it/s]
2|5613|Loss: 0.9703561663627625:  87%|████████▋ | 5613/6470 [1:10:10<10:24,  1.37it/s]
2|5614|Loss: 0.8275713920593262:  87%|████████▋ | 5613/6470 [1:10:11<10:24,  1.37it/s]
2|5614|Loss: 0.8275713920593262:  87%|████████▋ | 5614/6470 [1:10:11<10:11,  1.40it/s]
2|5615|Loss: 0.7319052219390869:  87%|████████▋ | 5614/6470 [1:10:11<10:11,  1.40it/s]
2|5615|Loss: 0.7319052219390869:  87%|████████▋ | 5615/6470 [1:10:12<09:40,  1.47it/s]
2|5616|Loss: 0.765763521194458:  87%|████████▋ | 5615/6470 [1:10:12<09:40,  1.47it/s] 
2|5616|Loss: 0.765763521194458:  87%|████████▋ | 5616/6470 [1:10:12<10:24,  1.37it/s]
2|5617|Loss: 0.6074913740158081:  87%|████████▋ | 5616/6470 [1:10:13<10:24,  1.37it/s]
2|5617|Loss: 0.6074913740158081:  87%|████████▋ | 5617/6470 [1:10:13<09:12,  1.54it/s]
2|5618|Loss: 0.7676618099212646:  87%|████████▋ | 5617/6470 [1:10:13<09:12,  1.54it/s]
2|5618|Loss: 0.7676618099212646:  87%|████████▋ | 5618/6470 [1:10:13<08:24,  1.69it/s]
2|5619|Loss: 0.8948853611946106:  87%|████████▋ | 5618/6470 [1:10:14<08:24,  1.69it/s]
2|5619|Loss: 0.8948853611946106:  87%|████████▋ | 5619/6470 [1:10:14<09:06,  1.56it/s]
2|5620|Loss: 0.7192981839179993:  87%|████████▋ | 5619/6470 [1:10:15<09:06,  1.56it/s]
2|5620|Loss: 0.7192981839179993:  87%|████████▋ | 5620/6470 [1:10:15<10:06,  1.40it/s]
2|5621|Loss: 1.005483627319336:  87%|████████▋ | 5620/6470 [1:10:15<10:06,  1.40it/s] 
2|5621|Loss: 1.005483627319336:  87%|████████▋ | 5621/6470 [1:10:16<10:07,  1.40it/s]
2|5622|Loss: 0.660117506980896:  87%|████████▋ | 5621/6470 [1:10:16<10:07,  1.40it/s]
2|5622|Loss: 0.660117506980896:  87%|████████▋ | 5622/6470 [1:10:16<10:13,  1.38it/s]
2|5623|Loss: 1.0766611099243164:  87%|████████▋ | 5622/6470 [1:10:17<10:13,  1.38it/s]
2|5623|Loss: 1.0766611099243164:  87%|████████▋ | 5623/6470 [1:10:17<10:06,  1.40it/s]
2|5624|Loss: 0.7989787459373474:  87%|████████▋ | 5623/6470 [1:10:18<10:06,  1.40it/s]
2|5624|Loss: 0.7989787459373474:  87%|████████▋ | 5624/6470 [1:10:18<09:38,  1.46it/s]
2|5625|Loss: 1.0274125337600708:  87%|████████▋ | 5624/6470 [1:10:18<09:38,  1.46it/s]
2|5625|Loss: 1.0274125337600708:  87%|████████▋ | 5625/6470 [1:10:19<09:58,  1.41it/s]
2|5626|Loss: 0.9063498973846436:  87%|████████▋ | 5625/6470 [1:10:19<09:58,  1.41it/s]
2|5626|Loss: 0.9063498973846436:  87%|████████▋ | 5626/6470 [1:10:19<10:18,  1.36it/s]
2|5627|Loss: 0.7750566005706787:  87%|████████▋ | 5626/6470 [1:10:20<10:18,  1.36it/s]
2|5627|Loss: 0.7750566005706787:  87%|████████▋ | 5627/6470 [1:10:20<10:32,  1.33it/s]
2|5628|Loss: 0.8725352883338928:  87%|████████▋ | 5627/6470 [1:10:20<10:32,  1.33it/s]
2|5628|Loss: 0.8725352883338928:  87%|████████▋ | 5628/6470 [1:10:21<09:59,  1.40it/s]
2|5629|Loss: 0.8772927522659302:  87%|████████▋ | 5628/6470 [1:10:21<09:59,  1.40it/s]
2|5629|Loss: 0.8772927522659302:  87%|████████▋ | 5629/6470 [1:10:21<10:00,  1.40it/s]
2|5630|Loss: 0.648160994052887:  87%|████████▋ | 5629/6470 [1:10:22<10:00,  1.40it/s] 
2|5630|Loss: 0.648160994052887:  87%|████████▋ | 5630/6470 [1:10:22<09:24,  1.49it/s]
2|5631|Loss: 0.8433037996292114:  87%|████████▋ | 5630/6470 [1:10:22<09:24,  1.49it/s]
2|5631|Loss: 0.8433037996292114:  87%|████████▋ | 5631/6470 [1:10:23<09:16,  1.51it/s]
2|5632|Loss: 0.9482579231262207:  87%|████████▋ | 5631/6470 [1:10:23<09:16,  1.51it/s]
2|5632|Loss: 0.9482579231262207:  87%|████████▋ | 5632/6470 [1:10:23<09:37,  1.45it/s]
2|5633|Loss: 0.888512372970581:  87%|████████▋ | 5632/6470 [1:10:24<09:37,  1.45it/s] 
2|5633|Loss: 0.888512372970581:  87%|████████▋ | 5633/6470 [1:10:24<10:33,  1.32it/s]
2|5634|Loss: 0.9378222227096558:  87%|████████▋ | 5633/6470 [1:10:25<10:33,  1.32it/s]
2|5634|Loss: 0.9378222227096558:  87%|████████▋ | 5634/6470 [1:10:25<10:43,  1.30it/s]
2|5635|Loss: 0.6956645846366882:  87%|████████▋ | 5634/6470 [1:10:26<10:43,  1.30it/s]
2|5635|Loss: 0.6956645846366882:  87%|████████▋ | 5635/6470 [1:10:26<10:38,  1.31it/s]
2|5636|Loss: 0.8147534728050232:  87%|████████▋ | 5635/6470 [1:10:26<10:38,  1.31it/s]
2|5636|Loss: 0.8147534728050232:  87%|████████▋ | 5636/6470 [1:10:27<10:10,  1.37it/s]
2|5637|Loss: 0.8611370325088501:  87%|████████▋ | 5636/6470 [1:10:27<10:10,  1.37it/s]
2|5637|Loss: 0.8611370325088501:  87%|████████▋ | 5637/6470 [1:10:27<10:07,  1.37it/s]
2|5638|Loss: 0.8530808687210083:  87%|████████▋ | 5637/6470 [1:10:28<10:07,  1.37it/s]
2|5638|Loss: 0.8530808687210083:  87%|████████▋ | 5638/6470 [1:10:28<10:12,  1.36it/s]
2|5639|Loss: 0.7481659054756165:  87%|████████▋ | 5638/6470 [1:10:28<10:12,  1.36it/s]
2|5639|Loss: 0.7481659054756165:  87%|████████▋ | 5639/6470 [1:10:29<10:14,  1.35it/s]
2|5640|Loss: 0.840155839920044:  87%|████████▋ | 5639/6470 [1:10:29<10:14,  1.35it/s] 
2|5640|Loss: 0.840155839920044:  87%|████████▋ | 5640/6470 [1:10:29<10:05,  1.37it/s]
2|5641|Loss: 0.8506938219070435:  87%|████████▋ | 5640/6470 [1:10:30<10:05,  1.37it/s]
2|5641|Loss: 0.8506938219070435:  87%|████████▋ | 5641/6470 [1:10:30<10:15,  1.35it/s]
2|5642|Loss: 0.88075190782547:  87%|████████▋ | 5641/6470 [1:10:31<10:15,  1.35it/s]  
2|5642|Loss: 0.88075190782547:  87%|████████▋ | 5642/6470 [1:10:31<11:12,  1.23it/s]
2|5643|Loss: 0.799974262714386:  87%|████████▋ | 5642/6470 [1:10:32<11:12,  1.23it/s]
2|5643|Loss: 0.799974262714386:  87%|████████▋ | 5643/6470 [1:10:32<11:12,  1.23it/s]
2|5644|Loss: 0.9372985363006592:  87%|████████▋ | 5643/6470 [1:10:32<11:12,  1.23it/s]
2|5644|Loss: 0.9372985363006592:  87%|████████▋ | 5644/6470 [1:10:33<10:42,  1.29it/s]
2|5645|Loss: 0.8193601369857788:  87%|████████▋ | 5644/6470 [1:10:33<10:42,  1.29it/s]
2|5645|Loss: 0.8193601369857788:  87%|████████▋ | 5645/6470 [1:10:34<11:00,  1.25it/s]
2|5646|Loss: 0.7448094487190247:  87%|████████▋ | 5645/6470 [1:10:34<11:00,  1.25it/s]
2|5646|Loss: 0.7448094487190247:  87%|████████▋ | 5646/6470 [1:10:34<11:00,  1.25it/s]
2|5647|Loss: 0.789411187171936:  87%|████████▋ | 5646/6470 [1:10:35<11:00,  1.25it/s] 
2|5647|Loss: 0.789411187171936:  87%|████████▋ | 5647/6470 [1:10:35<10:07,  1.35it/s]
2|5648|Loss: 0.7326655387878418:  87%|████████▋ | 5647/6470 [1:10:35<10:07,  1.35it/s]
2|5648|Loss: 0.7326655387878418:  87%|████████▋ | 5648/6470 [1:10:36<09:59,  1.37it/s]
2|5649|Loss: 0.850780189037323:  87%|████████▋ | 5648/6470 [1:10:36<09:59,  1.37it/s] 
2|5649|Loss: 0.850780189037323:  87%|████████▋ | 5649/6470 [1:10:36<09:54,  1.38it/s]
2|5650|Loss: 0.7804889678955078:  87%|████████▋ | 5649/6470 [1:10:37<09:54,  1.38it/s]
2|5650|Loss: 0.7804889678955078:  87%|████████▋ | 5650/6470 [1:10:37<09:41,  1.41it/s]
2|5651|Loss: 0.9634678959846497:  87%|████████▋ | 5650/6470 [1:10:38<09:41,  1.41it/s]
2|5651|Loss: 0.9634678959846497:  87%|████████▋ | 5651/6470 [1:10:38<10:07,  1.35it/s]
2|5652|Loss: 0.8352493643760681:  87%|████████▋ | 5651/6470 [1:10:38<10:07,  1.35it/s]
2|5652|Loss: 0.8352493643760681:  87%|████████▋ | 5652/6470 [1:10:39<10:23,  1.31it/s]
2|5653|Loss: 0.8111411929130554:  87%|████████▋ | 5652/6470 [1:10:39<10:23,  1.31it/s]
2|5653|Loss: 0.8111411929130554:  87%|████████▋ | 5653/6470 [1:10:39<09:55,  1.37it/s]
2|5654|Loss: 0.9023945331573486:  87%|████████▋ | 5653/6470 [1:10:40<09:55,  1.37it/s]
2|5654|Loss: 0.9023945331573486:  87%|████████▋ | 5654/6470 [1:10:40<09:55,  1.37it/s]
2|5655|Loss: 0.8808096647262573:  87%|████████▋ | 5654/6470 [1:10:40<09:55,  1.37it/s]
2|5655|Loss: 0.8808096647262573:  87%|████████▋ | 5655/6470 [1:10:41<09:50,  1.38it/s]
2|5656|Loss: 0.9600558280944824:  87%|████████▋ | 5655/6470 [1:10:41<09:50,  1.38it/s]
2|5656|Loss: 0.9600558280944824:  87%|████████▋ | 5656/6470 [1:10:42<09:54,  1.37it/s]
2|5657|Loss: 0.7244617938995361:  87%|████████▋ | 5656/6470 [1:10:42<09:54,  1.37it/s]
2|5657|Loss: 0.7244617938995361:  87%|████████▋ | 5657/6470 [1:10:42<10:24,  1.30it/s]
2|5658|Loss: 0.8415332436561584:  87%|████████▋ | 5657/6470 [1:10:43<10:24,  1.30it/s]
2|5658|Loss: 0.8415332436561584:  87%|████████▋ | 5658/6470 [1:10:43<10:26,  1.30it/s]
2|5659|Loss: 0.7843852639198303:  87%|████████▋ | 5658/6470 [1:10:44<10:26,  1.30it/s]
2|5659|Loss: 0.7843852639198303:  87%|████████▋ | 5659/6470 [1:10:44<10:04,  1.34it/s]
2|5660|Loss: 0.9093261361122131:  87%|████████▋ | 5659/6470 [1:10:44<10:04,  1.34it/s]
2|5660|Loss: 0.9093261361122131:  87%|████████▋ | 5660/6470 [1:10:45<09:44,  1.39it/s]
2|5661|Loss: 0.7794591188430786:  87%|████████▋ | 5660/6470 [1:10:45<09:44,  1.39it/s]
2|5661|Loss: 0.7794591188430786:  87%|████████▋ | 5661/6470 [1:10:45<10:13,  1.32it/s]
2|5662|Loss: 0.8880435824394226:  87%|████████▋ | 5661/6470 [1:10:46<10:13,  1.32it/s]
2|5662|Loss: 0.8880435824394226:  88%|████████▊ | 5662/6470 [1:10:46<10:50,  1.24it/s]
2|5663|Loss: 0.7041741013526917:  88%|████████▊ | 5662/6470 [1:10:47<10:50,  1.24it/s]
2|5663|Loss: 0.7041741013526917:  88%|████████▊ | 5663/6470 [1:10:47<11:14,  1.20it/s]
2|5664|Loss: 0.9331799745559692:  88%|████████▊ | 5663/6470 [1:10:48<11:14,  1.20it/s]
2|5664|Loss: 0.9331799745559692:  88%|████████▊ | 5664/6470 [1:10:48<11:09,  1.20it/s]
2|5665|Loss: 0.6704756021499634:  88%|████████▊ | 5664/6470 [1:10:48<11:09,  1.20it/s]
2|5665|Loss: 0.6704756021499634:  88%|████████▊ | 5665/6470 [1:10:49<10:57,  1.22it/s]
2|5666|Loss: 0.755623996257782:  88%|████████▊ | 5665/6470 [1:10:49<10:57,  1.22it/s] 
2|5666|Loss: 0.755623996257782:  88%|████████▊ | 5666/6470 [1:10:50<11:43,  1.14it/s]
2|5667|Loss: 0.8775444626808167:  88%|████████▊ | 5666/6470 [1:10:50<11:43,  1.14it/s]
2|5667|Loss: 0.8775444626808167:  88%|████████▊ | 5667/6470 [1:10:51<12:12,  1.10it/s]
2|5668|Loss: 0.9203953742980957:  88%|████████▊ | 5667/6470 [1:10:51<12:12,  1.10it/s]
2|5668|Loss: 0.9203953742980957:  88%|████████▊ | 5668/6470 [1:10:52<12:19,  1.08it/s]
2|5669|Loss: 0.8060234189033508:  88%|████████▊ | 5668/6470 [1:10:52<12:19,  1.08it/s]
2|5669|Loss: 0.8060234189033508:  88%|████████▊ | 5669/6470 [1:10:52<11:32,  1.16it/s]
2|5670|Loss: 1.0097051858901978:  88%|████████▊ | 5669/6470 [1:10:53<11:32,  1.16it/s]
2|5670|Loss: 1.0097051858901978:  88%|████████▊ | 5670/6470 [1:10:53<11:38,  1.15it/s]
2|5671|Loss: 0.7063518762588501:  88%|████████▊ | 5670/6470 [1:10:54<11:38,  1.15it/s]
2|5671|Loss: 0.7063518762588501:  88%|████████▊ | 5671/6470 [1:10:54<10:13,  1.30it/s]
2|5672|Loss: 0.8219242095947266:  88%|████████▊ | 5671/6470 [1:10:54<10:13,  1.30it/s]
2|5672|Loss: 0.8219242095947266:  88%|████████▊ | 5672/6470 [1:10:55<09:51,  1.35it/s]
2|5673|Loss: 0.7914141416549683:  88%|████████▊ | 5672/6470 [1:10:55<09:51,  1.35it/s]
2|5673|Loss: 0.7914141416549683:  88%|████████▊ | 5673/6470 [1:10:55<10:06,  1.31it/s]
2|5674|Loss: 0.6918615698814392:  88%|████████▊ | 5673/6470 [1:10:56<10:06,  1.31it/s]
2|5674|Loss: 0.6918615698814392:  88%|████████▊ | 5674/6470 [1:10:56<10:17,  1.29it/s]
2|5675|Loss: 0.837069034576416:  88%|████████▊ | 5674/6470 [1:10:57<10:17,  1.29it/s] 
2|5675|Loss: 0.837069034576416:  88%|████████▊ | 5675/6470 [1:10:57<10:15,  1.29it/s]
2|5676|Loss: 0.9368229508399963:  88%|████████▊ | 5675/6470 [1:10:57<10:15,  1.29it/s]
2|5676|Loss: 0.9368229508399963:  88%|████████▊ | 5676/6470 [1:10:58<09:57,  1.33it/s]
2|5677|Loss: 0.9024560451507568:  88%|████████▊ | 5676/6470 [1:10:58<09:57,  1.33it/s]
2|5677|Loss: 0.9024560451507568:  88%|████████▊ | 5677/6470 [1:10:58<09:46,  1.35it/s]
2|5678|Loss: 0.9446654915809631:  88%|████████▊ | 5677/6470 [1:10:59<09:46,  1.35it/s]
2|5678|Loss: 0.9446654915809631:  88%|████████▊ | 5678/6470 [1:10:59<09:55,  1.33it/s]
2|5679|Loss: 0.9108494520187378:  88%|████████▊ | 5678/6470 [1:11:00<09:55,  1.33it/s]
2|5679|Loss: 0.9108494520187378:  88%|████████▊ | 5679/6470 [1:11:00<09:23,  1.40it/s]
2|5680|Loss: 0.8642122149467468:  88%|████████▊ | 5679/6470 [1:11:00<09:23,  1.40it/s]
2|5680|Loss: 0.8642122149467468:  88%|████████▊ | 5680/6470 [1:11:01<09:29,  1.39it/s]
2|5681|Loss: 0.7529335618019104:  88%|████████▊ | 5680/6470 [1:11:01<09:29,  1.39it/s]
2|5681|Loss: 0.7529335618019104:  88%|████████▊ | 5681/6470 [1:11:01<09:39,  1.36it/s]
2|5682|Loss: 0.8288185596466064:  88%|████████▊ | 5681/6470 [1:11:02<09:39,  1.36it/s]
2|5682|Loss: 0.8288185596466064:  88%|████████▊ | 5682/6470 [1:11:02<09:51,  1.33it/s]
2|5683|Loss: 0.8192790746688843:  88%|████████▊ | 5682/6470 [1:11:03<09:51,  1.33it/s]
2|5683|Loss: 0.8192790746688843:  88%|████████▊ | 5683/6470 [1:11:03<09:59,  1.31it/s]
2|5684|Loss: 0.8354085683822632:  88%|████████▊ | 5683/6470 [1:11:03<09:59,  1.31it/s]
2|5684|Loss: 0.8354085683822632:  88%|████████▊ | 5684/6470 [1:11:04<10:02,  1.30it/s]
2|5685|Loss: 0.8839696049690247:  88%|████████▊ | 5684/6470 [1:11:04<10:02,  1.30it/s]
2|5685|Loss: 0.8839696049690247:  88%|████████▊ | 5685/6470 [1:11:04<10:08,  1.29it/s]
2|5686|Loss: 0.807827353477478:  88%|████████▊ | 5685/6470 [1:11:05<10:08,  1.29it/s] 
2|5686|Loss: 0.807827353477478:  88%|████████▊ | 5686/6470 [1:11:05<09:55,  1.32it/s]
2|5687|Loss: 0.878500759601593:  88%|████████▊ | 5686/6470 [1:11:06<09:55,  1.32it/s]
2|5687|Loss: 0.878500759601593:  88%|████████▊ | 5687/6470 [1:11:06<10:27,  1.25it/s]
2|5688|Loss: 0.9394183158874512:  88%|████████▊ | 5687/6470 [1:11:07<10:27,  1.25it/s]
2|5688|Loss: 0.9394183158874512:  88%|████████▊ | 5688/6470 [1:11:07<10:11,  1.28it/s]
2|5689|Loss: 0.8113544583320618:  88%|████████▊ | 5688/6470 [1:11:07<10:11,  1.28it/s]
2|5689|Loss: 0.8113544583320618:  88%|████████▊ | 5689/6470 [1:11:07<09:24,  1.38it/s]
2|5690|Loss: 0.7054451107978821:  88%|████████▊ | 5689/6470 [1:11:08<09:24,  1.38it/s]
2|5690|Loss: 0.7054451107978821:  88%|████████▊ | 5690/6470 [1:11:08<09:44,  1.33it/s]
2|5691|Loss: 0.8414504528045654:  88%|████████▊ | 5690/6470 [1:11:09<09:44,  1.33it/s]
2|5691|Loss: 0.8414504528045654:  88%|████████▊ | 5691/6470 [1:11:09<08:58,  1.45it/s]
2|5692|Loss: 0.8281650543212891:  88%|████████▊ | 5691/6470 [1:11:09<08:58,  1.45it/s]
2|5692|Loss: 0.8281650543212891:  88%|████████▊ | 5692/6470 [1:11:09<08:42,  1.49it/s]
2|5693|Loss: 0.8702782988548279:  88%|████████▊ | 5692/6470 [1:11:10<08:42,  1.49it/s]
2|5693|Loss: 0.8702782988548279:  88%|████████▊ | 5693/6470 [1:11:10<08:46,  1.48it/s]
2|5694|Loss: 0.9188030362129211:  88%|████████▊ | 5693/6470 [1:11:10<08:46,  1.48it/s]
2|5694|Loss: 0.9188030362129211:  88%|████████▊ | 5694/6470 [1:11:11<08:50,  1.46it/s]
2|5695|Loss: 0.8998546004295349:  88%|████████▊ | 5694/6470 [1:11:11<08:50,  1.46it/s]
2|5695|Loss: 0.8998546004295349:  88%|████████▊ | 5695/6470 [1:11:11<08:23,  1.54it/s]
2|5696|Loss: 0.8309510946273804:  88%|████████▊ | 5695/6470 [1:11:12<08:23,  1.54it/s]
2|5696|Loss: 0.8309510946273804:  88%|████████▊ | 5696/6470 [1:11:12<08:19,  1.55it/s]
2|5697|Loss: 0.835595965385437:  88%|████████▊ | 5696/6470 [1:11:12<08:19,  1.55it/s] 
2|5697|Loss: 0.835595965385437:  88%|████████▊ | 5697/6470 [1:11:13<08:09,  1.58it/s]
2|5698|Loss: 1.0345627069473267:  88%|████████▊ | 5697/6470 [1:11:13<08:09,  1.58it/s]
2|5698|Loss: 1.0345627069473267:  88%|████████▊ | 5698/6470 [1:11:13<08:29,  1.51it/s]
2|5699|Loss: 0.8825222849845886:  88%|████████▊ | 5698/6470 [1:11:14<08:29,  1.51it/s]
2|5699|Loss: 0.8825222849845886:  88%|████████▊ | 5699/6470 [1:11:14<08:28,  1.52it/s]
2|5700|Loss: 1.0372503995895386:  88%|████████▊ | 5699/6470 [1:11:14<08:28,  1.52it/s]
2|5700|Loss: 1.0372503995895386:  88%|████████▊ | 5700/6470 [1:11:15<08:34,  1.50it/s]
2|5701|Loss: 0.8352526426315308:  88%|████████▊ | 5700/6470 [1:11:15<08:34,  1.50it/s]
2|5701|Loss: 0.8352526426315308:  88%|████████▊ | 5701/6470 [1:11:16<09:26,  1.36it/s]
2|5702|Loss: 0.9717778563499451:  88%|████████▊ | 5701/6470 [1:11:16<09:26,  1.36it/s]
2|5702|Loss: 0.9717778563499451:  88%|████████▊ | 5702/6470 [1:11:16<09:32,  1.34it/s]
2|5703|Loss: 0.8508261442184448:  88%|████████▊ | 5702/6470 [1:11:17<09:32,  1.34it/s]
2|5703|Loss: 0.8508261442184448:  88%|████████▊ | 5703/6470 [1:11:17<09:07,  1.40it/s]
2|5704|Loss: 0.9361444711685181:  88%|████████▊ | 5703/6470 [1:11:17<09:07,  1.40it/s]
2|5704|Loss: 0.9361444711685181:  88%|████████▊ | 5704/6470 [1:11:18<09:37,  1.33it/s]
2|5705|Loss: 0.8631529211997986:  88%|████████▊ | 5704/6470 [1:11:18<09:37,  1.33it/s]
2|5705|Loss: 0.8631529211997986:  88%|████████▊ | 5705/6470 [1:11:19<09:41,  1.32it/s]
2|5706|Loss: 0.8093687891960144:  88%|████████▊ | 5705/6470 [1:11:19<09:41,  1.32it/s]
2|5706|Loss: 0.8093687891960144:  88%|████████▊ | 5706/6470 [1:11:19<09:49,  1.30it/s]
2|5707|Loss: 0.8627539873123169:  88%|████████▊ | 5706/6470 [1:11:20<09:49,  1.30it/s]
2|5707|Loss: 0.8627539873123169:  88%|████████▊ | 5707/6470 [1:11:20<09:37,  1.32it/s]
2|5708|Loss: 0.7422716021537781:  88%|████████▊ | 5707/6470 [1:11:20<09:37,  1.32it/s]
2|5708|Loss: 0.7422716021537781:  88%|████████▊ | 5708/6470 [1:11:21<09:10,  1.38it/s]
2|5709|Loss: 0.9812995195388794:  88%|████████▊ | 5708/6470 [1:11:21<09:10,  1.38it/s]
2|5709|Loss: 0.9812995195388794:  88%|████████▊ | 5709/6470 [1:11:21<08:30,  1.49it/s]
2|5710|Loss: 0.8172367215156555:  88%|████████▊ | 5709/6470 [1:11:22<08:30,  1.49it/s]
2|5710|Loss: 0.8172367215156555:  88%|████████▊ | 5710/6470 [1:11:22<09:07,  1.39it/s]
2|5711|Loss: 0.8216447234153748:  88%|████████▊ | 5710/6470 [1:11:23<09:07,  1.39it/s]
2|5711|Loss: 0.8216447234153748:  88%|████████▊ | 5711/6470 [1:11:23<09:04,  1.39it/s]
2|5712|Loss: 0.825377345085144:  88%|████████▊ | 5711/6470 [1:11:23<09:04,  1.39it/s] 
2|5712|Loss: 0.825377345085144:  88%|████████▊ | 5712/6470 [1:11:24<09:09,  1.38it/s]
2|5713|Loss: 0.7543343305587769:  88%|████████▊ | 5712/6470 [1:11:24<09:09,  1.38it/s]
2|5713|Loss: 0.7543343305587769:  88%|████████▊ | 5713/6470 [1:11:24<09:54,  1.27it/s]
2|5714|Loss: 0.8485513925552368:  88%|████████▊ | 5713/6470 [1:11:25<09:54,  1.27it/s]
2|5714|Loss: 0.8485513925552368:  88%|████████▊ | 5714/6470 [1:11:25<09:33,  1.32it/s]
2|5715|Loss: 0.8670592308044434:  88%|████████▊ | 5714/6470 [1:11:26<09:33,  1.32it/s]
2|5715|Loss: 0.8670592308044434:  88%|████████▊ | 5715/6470 [1:11:26<09:13,  1.36it/s]
2|5716|Loss: 0.9449328184127808:  88%|████████▊ | 5715/6470 [1:11:26<09:13,  1.36it/s]
2|5716|Loss: 0.9449328184127808:  88%|████████▊ | 5716/6470 [1:11:27<09:39,  1.30it/s]
2|5717|Loss: 0.8219903111457825:  88%|████████▊ | 5716/6470 [1:11:27<09:39,  1.30it/s]
2|5717|Loss: 0.8219903111457825:  88%|████████▊ | 5717/6470 [1:11:27<09:21,  1.34it/s]
2|5718|Loss: 0.842182457447052:  88%|████████▊ | 5717/6470 [1:11:28<09:21,  1.34it/s] 
2|5718|Loss: 0.842182457447052:  88%|████████▊ | 5718/6470 [1:11:28<09:26,  1.33it/s]
2|5719|Loss: 0.861697793006897:  88%|████████▊ | 5718/6470 [1:11:29<09:26,  1.33it/s]
2|5719|Loss: 0.861697793006897:  88%|████████▊ | 5719/6470 [1:11:29<08:55,  1.40it/s]
2|5720|Loss: 0.8330411911010742:  88%|████████▊ | 5719/6470 [1:11:29<08:55,  1.40it/s]
2|5720|Loss: 0.8330411911010742:  88%|████████▊ | 5720/6470 [1:11:30<08:59,  1.39it/s]
2|5721|Loss: 0.8733084797859192:  88%|████████▊ | 5720/6470 [1:11:30<08:59,  1.39it/s]
2|5721|Loss: 0.8733084797859192:  88%|████████▊ | 5721/6470 [1:11:30<08:58,  1.39it/s]
2|5722|Loss: 0.81463223695755:  88%|████████▊ | 5721/6470 [1:11:31<08:58,  1.39it/s]  
2|5722|Loss: 0.81463223695755:  88%|████████▊ | 5722/6470 [1:11:31<08:29,  1.47it/s]
2|5723|Loss: 0.6841254234313965:  88%|████████▊ | 5722/6470 [1:11:31<08:29,  1.47it/s]
2|5723|Loss: 0.6841254234313965:  88%|████████▊ | 5723/6470 [1:11:32<08:54,  1.40it/s]
2|5724|Loss: 0.703761637210846:  88%|████████▊ | 5723/6470 [1:11:32<08:54,  1.40it/s] 
2|5724|Loss: 0.703761637210846:  88%|████████▊ | 5724/6470 [1:11:32<08:45,  1.42it/s]
2|5725|Loss: 0.8320738077163696:  88%|████████▊ | 5724/6470 [1:11:33<08:45,  1.42it/s]
2|5725|Loss: 0.8320738077163696:  88%|████████▊ | 5725/6470 [1:11:33<08:47,  1.41it/s]
2|5726|Loss: 0.8663291931152344:  88%|████████▊ | 5725/6470 [1:11:33<08:47,  1.41it/s]
2|5726|Loss: 0.8663291931152344:  89%|████████▊ | 5726/6470 [1:11:34<08:46,  1.41it/s]
2|5727|Loss: 0.841622531414032:  89%|████████▊ | 5726/6470 [1:11:34<08:46,  1.41it/s] 
2|5727|Loss: 0.841622531414032:  89%|████████▊ | 5727/6470 [1:11:35<09:02,  1.37it/s]
2|5728|Loss: 0.8756424188613892:  89%|████████▊ | 5727/6470 [1:11:35<09:02,  1.37it/s]
2|5728|Loss: 0.8756424188613892:  89%|████████▊ | 5728/6470 [1:11:35<09:35,  1.29it/s]
2|5729|Loss: 0.8838218450546265:  89%|████████▊ | 5728/6470 [1:11:36<09:35,  1.29it/s]
2|5729|Loss: 0.8838218450546265:  89%|████████▊ | 5729/6470 [1:11:36<10:00,  1.23it/s]
2|5730|Loss: 0.9275965690612793:  89%|████████▊ | 5729/6470 [1:11:37<10:00,  1.23it/s]
2|5730|Loss: 0.9275965690612793:  89%|████████▊ | 5730/6470 [1:11:37<09:59,  1.23it/s]
2|5731|Loss: 0.7103254199028015:  89%|████████▊ | 5730/6470 [1:11:38<09:59,  1.23it/s]
2|5731|Loss: 0.7103254199028015:  89%|████████▊ | 5731/6470 [1:11:38<09:39,  1.28it/s]
2|5732|Loss: 0.819079577922821:  89%|████████▊ | 5731/6470 [1:11:38<09:39,  1.28it/s] 
2|5732|Loss: 0.819079577922821:  89%|████████▊ | 5732/6470 [1:11:39<09:32,  1.29it/s]
2|5733|Loss: 0.785886287689209:  89%|████████▊ | 5732/6470 [1:11:39<09:32,  1.29it/s]
2|5733|Loss: 0.785886287689209:  89%|████████▊ | 5733/6470 [1:11:39<08:53,  1.38it/s]
2|5734|Loss: 0.9809674620628357:  89%|████████▊ | 5733/6470 [1:11:40<08:53,  1.38it/s]
2|5734|Loss: 0.9809674620628357:  89%|████████▊ | 5734/6470 [1:11:40<08:32,  1.43it/s]
2|5735|Loss: 0.8330026268959045:  89%|████████▊ | 5734/6470 [1:11:40<08:32,  1.43it/s]
2|5735|Loss: 0.8330026268959045:  89%|████████▊ | 5735/6470 [1:11:40<08:06,  1.51it/s]
2|5736|Loss: 0.8990939259529114:  89%|████████▊ | 5735/6470 [1:11:41<08:06,  1.51it/s]
2|5736|Loss: 0.8990939259529114:  89%|████████▊ | 5736/6470 [1:11:41<08:26,  1.45it/s]
2|5737|Loss: 0.869340181350708:  89%|████████▊ | 5736/6470 [1:11:42<08:26,  1.45it/s] 
2|5737|Loss: 0.869340181350708:  89%|████████▊ | 5737/6470 [1:11:42<08:27,  1.44it/s]
2|5738|Loss: 0.8666331768035889:  89%|████████▊ | 5737/6470 [1:11:42<08:27,  1.44it/s]
2|5738|Loss: 0.8666331768035889:  89%|████████▊ | 5738/6470 [1:11:43<08:43,  1.40it/s]
2|5739|Loss: 0.8975820541381836:  89%|████████▊ | 5738/6470 [1:11:43<08:43,  1.40it/s]
2|5739|Loss: 0.8975820541381836:  89%|████████▊ | 5739/6470 [1:11:43<08:42,  1.40it/s]
2|5740|Loss: 0.7830066084861755:  89%|████████▊ | 5739/6470 [1:11:44<08:42,  1.40it/s]
2|5740|Loss: 0.7830066084861755:  89%|████████▊ | 5740/6470 [1:11:44<08:52,  1.37it/s]
2|5741|Loss: 0.9456656575202942:  89%|████████▊ | 5740/6470 [1:11:45<08:52,  1.37it/s]
2|5741|Loss: 0.9456656575202942:  89%|████████▊ | 5741/6470 [1:11:45<09:01,  1.35it/s]
2|5742|Loss: 0.8240006566047668:  89%|████████▊ | 5741/6470 [1:11:45<09:01,  1.35it/s]
2|5742|Loss: 0.8240006566047668:  89%|████████▊ | 5742/6470 [1:11:46<08:40,  1.40it/s]
2|5743|Loss: 0.953517735004425:  89%|████████▊ | 5742/6470 [1:11:46<08:40,  1.40it/s] 
2|5743|Loss: 0.953517735004425:  89%|████████▉ | 5743/6470 [1:11:46<09:06,  1.33it/s]
2|5744|Loss: 0.7230291366577148:  89%|████████▉ | 5743/6470 [1:11:47<09:06,  1.33it/s]
2|5744|Loss: 0.7230291366577148:  89%|████████▉ | 5744/6470 [1:11:47<08:47,  1.38it/s]
2|5745|Loss: 0.7974154353141785:  89%|████████▉ | 5744/6470 [1:11:47<08:47,  1.38it/s]
2|5745|Loss: 0.7974154353141785:  89%|████████▉ | 5745/6470 [1:11:48<08:30,  1.42it/s]
2|5746|Loss: 0.89832603931427:  89%|████████▉ | 5745/6470 [1:11:48<08:30,  1.42it/s]  
2|5746|Loss: 0.89832603931427:  89%|████████▉ | 5746/6470 [1:11:48<08:38,  1.40it/s]
2|5747|Loss: 0.8311665654182434:  89%|████████▉ | 5746/6470 [1:11:49<08:38,  1.40it/s]
2|5747|Loss: 0.8311665654182434:  89%|████████▉ | 5747/6470 [1:11:49<07:57,  1.51it/s]
2|5748|Loss: 0.7716690301895142:  89%|████████▉ | 5747/6470 [1:11:49<07:57,  1.51it/s]
2|5748|Loss: 0.7716690301895142:  89%|████████▉ | 5748/6470 [1:11:50<08:03,  1.49it/s]
2|5749|Loss: 0.9056602120399475:  89%|████████▉ | 5748/6470 [1:11:50<08:03,  1.49it/s]
2|5749|Loss: 0.9056602120399475:  89%|████████▉ | 5749/6470 [1:11:50<08:32,  1.41it/s]
2|5750|Loss: 0.782950222492218:  89%|████████▉ | 5749/6470 [1:11:51<08:32,  1.41it/s] 
2|5750|Loss: 0.782950222492218:  89%|████████▉ | 5750/6470 [1:11:51<08:53,  1.35it/s]
2|5751|Loss: 1.0343103408813477:  89%|████████▉ | 5750/6470 [1:11:52<08:53,  1.35it/s]
2|5751|Loss: 1.0343103408813477:  89%|████████▉ | 5751/6470 [1:11:52<08:25,  1.42it/s]
2|5752|Loss: 0.8166972398757935:  89%|████████▉ | 5751/6470 [1:11:52<08:25,  1.42it/s]
2|5752|Loss: 0.8166972398757935:  89%|████████▉ | 5752/6470 [1:11:53<08:35,  1.39it/s]
2|5753|Loss: 0.8096489310264587:  89%|████████▉ | 5752/6470 [1:11:53<08:35,  1.39it/s]
2|5753|Loss: 0.8096489310264587:  89%|████████▉ | 5753/6470 [1:11:53<08:30,  1.40it/s]
2|5754|Loss: 0.9011834263801575:  89%|████████▉ | 5753/6470 [1:11:54<08:30,  1.40it/s]
2|5754|Loss: 0.9011834263801575:  89%|████████▉ | 5754/6470 [1:11:54<08:51,  1.35it/s]
2|5755|Loss: 0.9825549721717834:  89%|████████▉ | 5754/6470 [1:11:55<08:51,  1.35it/s]
2|5755|Loss: 0.9825549721717834:  89%|████████▉ | 5755/6470 [1:11:55<09:26,  1.26it/s]
2|5756|Loss: 0.7143363356590271:  89%|████████▉ | 5755/6470 [1:11:55<09:26,  1.26it/s]
2|5756|Loss: 0.7143363356590271:  89%|████████▉ | 5756/6470 [1:11:56<08:48,  1.35it/s]
2|5757|Loss: 0.7385076880455017:  89%|████████▉ | 5756/6470 [1:11:56<08:48,  1.35it/s]
2|5757|Loss: 0.7385076880455017:  89%|████████▉ | 5757/6470 [1:11:57<09:08,  1.30it/s]
2|5758|Loss: 0.8163527250289917:  89%|████████▉ | 5757/6470 [1:11:57<09:08,  1.30it/s]
2|5758|Loss: 0.8163527250289917:  89%|████████▉ | 5758/6470 [1:11:57<09:01,  1.31it/s]
2|5759|Loss: 0.7819251418113708:  89%|████████▉ | 5758/6470 [1:11:58<09:01,  1.31it/s]
2|5759|Loss: 0.7819251418113708:  89%|████████▉ | 5759/6470 [1:11:58<08:22,  1.42it/s]
2|5760|Loss: 0.7619499564170837:  89%|████████▉ | 5759/6470 [1:11:58<08:22,  1.42it/s]
2|5760|Loss: 0.7619499564170837:  89%|████████▉ | 5760/6470 [1:11:58<08:06,  1.46it/s]
2|5761|Loss: 0.8428399562835693:  89%|████████▉ | 5760/6470 [1:11:59<08:06,  1.46it/s]
2|5761|Loss: 0.8428399562835693:  89%|████████▉ | 5761/6470 [1:11:59<08:47,  1.34it/s]
2|5762|Loss: 0.7471461892127991:  89%|████████▉ | 5761/6470 [1:12:00<08:47,  1.34it/s]
2|5762|Loss: 0.7471461892127991:  89%|████████▉ | 5762/6470 [1:12:00<09:05,  1.30it/s]
2|5763|Loss: 0.8044480085372925:  89%|████████▉ | 5762/6470 [1:12:01<09:05,  1.30it/s]
2|5763|Loss: 0.8044480085372925:  89%|████████▉ | 5763/6470 [1:12:01<08:49,  1.33it/s]
2|5764|Loss: 0.9466857314109802:  89%|████████▉ | 5763/6470 [1:12:01<08:49,  1.33it/s]
2|5764|Loss: 0.9466857314109802:  89%|████████▉ | 5764/6470 [1:12:02<08:54,  1.32it/s]
2|5765|Loss: 0.8880901336669922:  89%|████████▉ | 5764/6470 [1:12:02<08:54,  1.32it/s]
2|5765|Loss: 0.8880901336669922:  89%|████████▉ | 5765/6470 [1:12:02<08:40,  1.35it/s]
2|5766|Loss: 0.7131021618843079:  89%|████████▉ | 5765/6470 [1:12:03<08:40,  1.35it/s]
2|5766|Loss: 0.7131021618843079:  89%|████████▉ | 5766/6470 [1:12:03<08:46,  1.34it/s]
2|5767|Loss: 0.7195361256599426:  89%|████████▉ | 5766/6470 [1:12:04<08:46,  1.34it/s]
2|5767|Loss: 0.7195361256599426:  89%|████████▉ | 5767/6470 [1:12:04<08:49,  1.33it/s]
2|5768|Loss: 1.0711190700531006:  89%|████████▉ | 5767/6470 [1:12:04<08:49,  1.33it/s]
2|5768|Loss: 1.0711190700531006:  89%|████████▉ | 5768/6470 [1:12:05<09:30,  1.23it/s]
2|5769|Loss: 0.8158490657806396:  89%|████████▉ | 5768/6470 [1:12:05<09:30,  1.23it/s]
2|5769|Loss: 0.8158490657806396:  89%|████████▉ | 5769/6470 [1:12:06<09:34,  1.22it/s]
2|5770|Loss: 0.9614343047142029:  89%|████████▉ | 5769/6470 [1:12:06<09:34,  1.22it/s]
2|5770|Loss: 0.9614343047142029:  89%|████████▉ | 5770/6470 [1:12:06<08:59,  1.30it/s]
2|5771|Loss: 0.9094799757003784:  89%|████████▉ | 5770/6470 [1:12:07<08:59,  1.30it/s]
2|5771|Loss: 0.9094799757003784:  89%|████████▉ | 5771/6470 [1:12:07<09:06,  1.28it/s]
2|5772|Loss: 0.7473775744438171:  89%|████████▉ | 5771/6470 [1:12:08<09:06,  1.28it/s]
2|5772|Loss: 0.7473775744438171:  89%|████████▉ | 5772/6470 [1:12:08<08:20,  1.39it/s]
2|5773|Loss: 0.9410547614097595:  89%|████████▉ | 5772/6470 [1:12:08<08:20,  1.39it/s]
2|5773|Loss: 0.9410547614097595:  89%|████████▉ | 5773/6470 [1:12:08<08:02,  1.45it/s]
2|5774|Loss: 1.06991446018219:  89%|████████▉ | 5773/6470 [1:12:09<08:02,  1.45it/s]  
2|5774|Loss: 1.06991446018219:  89%|████████▉ | 5774/6470 [1:12:09<08:21,  1.39it/s]
2|5775|Loss: 0.7693190574645996:  89%|████████▉ | 5774/6470 [1:12:10<08:21,  1.39it/s]
2|5775|Loss: 0.7693190574645996:  89%|████████▉ | 5775/6470 [1:12:10<08:20,  1.39it/s]
2|5776|Loss: 0.925635814666748:  89%|████████▉ | 5775/6470 [1:12:10<08:20,  1.39it/s] 
2|5776|Loss: 0.925635814666748:  89%|████████▉ | 5776/6470 [1:12:10<08:01,  1.44it/s]
2|5777|Loss: 0.8650813698768616:  89%|████████▉ | 5776/6470 [1:12:11<08:01,  1.44it/s]
2|5777|Loss: 0.8650813698768616:  89%|████████▉ | 5777/6470 [1:12:11<08:12,  1.41it/s]
2|5778|Loss: 0.832329273223877:  89%|████████▉ | 5777/6470 [1:12:12<08:12,  1.41it/s] 
2|5778|Loss: 0.832329273223877:  89%|████████▉ | 5778/6470 [1:12:12<07:45,  1.49it/s]
2|5779|Loss: 0.7480005621910095:  89%|████████▉ | 5778/6470 [1:12:12<07:45,  1.49it/s]
2|5779|Loss: 0.7480005621910095:  89%|████████▉ | 5779/6470 [1:12:12<07:37,  1.51it/s]
2|5780|Loss: 0.7088016867637634:  89%|████████▉ | 5779/6470 [1:12:13<07:37,  1.51it/s]
2|5780|Loss: 0.7088016867637634:  89%|████████▉ | 5780/6470 [1:12:13<07:54,  1.45it/s]
2|5781|Loss: 0.839113712310791:  89%|████████▉ | 5780/6470 [1:12:14<07:54,  1.45it/s] 
2|5781|Loss: 0.839113712310791:  89%|████████▉ | 5781/6470 [1:12:14<08:38,  1.33it/s]
2|5782|Loss: 0.8055709600448608:  89%|████████▉ | 5781/6470 [1:12:15<08:38,  1.33it/s]
2|5782|Loss: 0.8055709600448608:  89%|████████▉ | 5782/6470 [1:12:15<08:11,  1.40it/s]
2|5783|Loss: 0.8804733157157898:  89%|████████▉ | 5782/6470 [1:12:15<08:11,  1.40it/s]
2|5783|Loss: 0.8804733157157898:  89%|████████▉ | 5783/6470 [1:12:15<08:14,  1.39it/s]
2|5784|Loss: 0.8189994692802429:  89%|████████▉ | 5783/6470 [1:12:16<08:14,  1.39it/s]
2|5784|Loss: 0.8189994692802429:  89%|████████▉ | 5784/6470 [1:12:16<08:44,  1.31it/s]
2|5785|Loss: 0.6257154941558838:  89%|████████▉ | 5784/6470 [1:12:17<08:44,  1.31it/s]
2|5785|Loss: 0.6257154941558838:  89%|████████▉ | 5785/6470 [1:12:17<07:49,  1.46it/s]
2|5786|Loss: 1.02141273021698:  89%|████████▉ | 5785/6470 [1:12:17<07:49,  1.46it/s]  
2|5786|Loss: 1.02141273021698:  89%|████████▉ | 5786/6470 [1:12:17<07:39,  1.49it/s]
2|5787|Loss: 0.8854355216026306:  89%|████████▉ | 5786/6470 [1:12:18<07:39,  1.49it/s]
2|5787|Loss: 0.8854355216026306:  89%|████████▉ | 5787/6470 [1:12:18<07:43,  1.47it/s]
2|5788|Loss: 0.7762615084648132:  89%|████████▉ | 5787/6470 [1:12:19<07:43,  1.47it/s]
2|5788|Loss: 0.7762615084648132:  89%|████████▉ | 5788/6470 [1:12:19<08:04,  1.41it/s]
2|5789|Loss: 0.8868153691291809:  89%|████████▉ | 5788/6470 [1:12:19<08:04,  1.41it/s]
2|5789|Loss: 0.8868153691291809:  89%|████████▉ | 5789/6470 [1:12:20<08:15,  1.37it/s]
2|5790|Loss: 0.8927236795425415:  89%|████████▉ | 5789/6470 [1:12:20<08:15,  1.37it/s]
2|5790|Loss: 0.8927236795425415:  89%|████████▉ | 5790/6470 [1:12:20<08:20,  1.36it/s]
2|5791|Loss: 0.7694125175476074:  89%|████████▉ | 5790/6470 [1:12:21<08:20,  1.36it/s]
2|5791|Loss: 0.7694125175476074:  90%|████████▉ | 5791/6470 [1:12:21<08:04,  1.40it/s]
2|5792|Loss: 0.8445048332214355:  90%|████████▉ | 5791/6470 [1:12:22<08:04,  1.40it/s]
2|5792|Loss: 0.8445048332214355:  90%|████████▉ | 5792/6470 [1:12:22<08:11,  1.38it/s]
2|5793|Loss: 0.7165819406509399:  90%|████████▉ | 5792/6470 [1:12:22<08:11,  1.38it/s]
2|5793|Loss: 0.7165819406509399:  90%|████████▉ | 5793/6470 [1:12:23<08:17,  1.36it/s]
2|5794|Loss: 0.8701905608177185:  90%|████████▉ | 5793/6470 [1:12:23<08:17,  1.36it/s]
2|5794|Loss: 0.8701905608177185:  90%|████████▉ | 5794/6470 [1:12:23<07:54,  1.42it/s]
2|5795|Loss: 0.7913581132888794:  90%|████████▉ | 5794/6470 [1:12:24<07:54,  1.42it/s]
2|5795|Loss: 0.7913581132888794:  90%|████████▉ | 5795/6470 [1:12:24<07:22,  1.53it/s]
2|5796|Loss: 0.866540789604187:  90%|████████▉ | 5795/6470 [1:12:24<07:22,  1.53it/s] 
2|5796|Loss: 0.866540789604187:  90%|████████▉ | 5796/6470 [1:12:24<07:16,  1.55it/s]
2|5797|Loss: 0.8450202345848083:  90%|████████▉ | 5796/6470 [1:12:25<07:16,  1.55it/s]
2|5797|Loss: 0.8450202345848083:  90%|████████▉ | 5797/6470 [1:12:25<06:59,  1.60it/s]
2|5798|Loss: 0.9403910040855408:  90%|████████▉ | 5797/6470 [1:12:25<06:59,  1.60it/s]
2|5798|Loss: 0.9403910040855408:  90%|████████▉ | 5798/6470 [1:12:26<07:08,  1.57it/s]
2|5799|Loss: 0.9978735446929932:  90%|████████▉ | 5798/6470 [1:12:26<07:08,  1.57it/s]
2|5799|Loss: 0.9978735446929932:  90%|████████▉ | 5799/6470 [1:12:27<07:46,  1.44it/s]
2|5800|Loss: 0.7750241756439209:  90%|████████▉ | 5799/6470 [1:12:27<07:46,  1.44it/s]
2|5800|Loss: 0.7750241756439209:  90%|████████▉ | 5800/6470 [1:12:27<08:30,  1.31it/s]
2|5801|Loss: 0.7893183827400208:  90%|████████▉ | 5800/6470 [1:12:28<08:30,  1.31it/s]
2|5801|Loss: 0.7893183827400208:  90%|████████▉ | 5801/6470 [1:12:28<08:26,  1.32it/s]
2|5802|Loss: 0.8938098549842834:  90%|████████▉ | 5801/6470 [1:12:29<08:26,  1.32it/s]
2|5802|Loss: 0.8938098549842834:  90%|████████▉ | 5802/6470 [1:12:29<08:30,  1.31it/s]
2|5803|Loss: 0.9434236288070679:  90%|████████▉ | 5802/6470 [1:12:29<08:30,  1.31it/s]
2|5803|Loss: 0.9434236288070679:  90%|████████▉ | 5803/6470 [1:12:30<08:50,  1.26it/s]
2|5804|Loss: 0.9375322461128235:  90%|████████▉ | 5803/6470 [1:12:30<08:50,  1.26it/s]
2|5804|Loss: 0.9375322461128235:  90%|████████▉ | 5804/6470 [1:12:31<08:53,  1.25it/s]
2|5805|Loss: 0.8148946166038513:  90%|████████▉ | 5804/6470 [1:12:31<08:53,  1.25it/s]
2|5805|Loss: 0.8148946166038513:  90%|████████▉ | 5805/6470 [1:12:31<08:32,  1.30it/s]
2|5806|Loss: 0.7358770370483398:  90%|████████▉ | 5805/6470 [1:12:32<08:32,  1.30it/s]
2|5806|Loss: 0.7358770370483398:  90%|████████▉ | 5806/6470 [1:12:32<08:30,  1.30it/s]
2|5807|Loss: 0.8910194635391235:  90%|████████▉ | 5806/6470 [1:12:33<08:30,  1.30it/s]
2|5807|Loss: 0.8910194635391235:  90%|████████▉ | 5807/6470 [1:12:33<09:04,  1.22it/s]
2|5808|Loss: 0.8852323889732361:  90%|████████▉ | 5807/6470 [1:12:33<09:04,  1.22it/s]
2|5808|Loss: 0.8852323889732361:  90%|████████▉ | 5808/6470 [1:12:34<08:46,  1.26it/s]
2|5809|Loss: 0.8714821934700012:  90%|████████▉ | 5808/6470 [1:12:34<08:46,  1.26it/s]
2|5809|Loss: 0.8714821934700012:  90%|████████▉ | 5809/6470 [1:12:35<09:06,  1.21it/s]
2|5810|Loss: 0.8778290152549744:  90%|████████▉ | 5809/6470 [1:12:35<09:06,  1.21it/s]
2|5810|Loss: 0.8778290152549744:  90%|████████▉ | 5810/6470 [1:12:35<08:50,  1.24it/s]
2|5811|Loss: 0.7367058992385864:  90%|████████▉ | 5810/6470 [1:12:36<08:50,  1.24it/s]
2|5811|Loss: 0.7367058992385864:  90%|████████▉ | 5811/6470 [1:12:36<07:54,  1.39it/s]
2|5812|Loss: 0.8227522373199463:  90%|████████▉ | 5811/6470 [1:12:36<07:54,  1.39it/s]
2|5812|Loss: 0.8227522373199463:  90%|████████▉ | 5812/6470 [1:12:37<07:55,  1.39it/s]
2|5813|Loss: 0.9267547726631165:  90%|████████▉ | 5812/6470 [1:12:37<07:55,  1.39it/s]
2|5813|Loss: 0.9267547726631165:  90%|████████▉ | 5813/6470 [1:12:37<08:08,  1.35it/s]
2|5814|Loss: 0.8997811079025269:  90%|████████▉ | 5813/6470 [1:12:38<08:08,  1.35it/s]
2|5814|Loss: 0.8997811079025269:  90%|████████▉ | 5814/6470 [1:12:38<08:19,  1.31it/s]
2|5815|Loss: 0.735761821269989:  90%|████████▉ | 5814/6470 [1:12:39<08:19,  1.31it/s] 
2|5815|Loss: 0.735761821269989:  90%|████████▉ | 5815/6470 [1:12:39<08:20,  1.31it/s]
2|5816|Loss: 0.7222460508346558:  90%|████████▉ | 5815/6470 [1:12:39<08:20,  1.31it/s]
2|5816|Loss: 0.7222460508346558:  90%|████████▉ | 5816/6470 [1:12:40<07:45,  1.41it/s]
2|5817|Loss: 0.8010644912719727:  90%|████████▉ | 5816/6470 [1:12:40<07:45,  1.41it/s]
2|5817|Loss: 0.8010644912719727:  90%|████████▉ | 5817/6470 [1:12:40<07:42,  1.41it/s]
2|5818|Loss: 0.9357583522796631:  90%|████████▉ | 5817/6470 [1:12:41<07:42,  1.41it/s]
2|5818|Loss: 0.9357583522796631:  90%|████████▉ | 5818/6470 [1:12:41<08:13,  1.32it/s]
2|5819|Loss: 0.9325379133224487:  90%|████████▉ | 5818/6470 [1:12:42<08:13,  1.32it/s]
2|5819|Loss: 0.9325379133224487:  90%|████████▉ | 5819/6470 [1:12:42<08:09,  1.33it/s]
2|5820|Loss: 0.8655661344528198:  90%|████████▉ | 5819/6470 [1:12:42<08:09,  1.33it/s]
2|5820|Loss: 0.8655661344528198:  90%|████████▉ | 5820/6470 [1:12:43<08:11,  1.32it/s]
2|5821|Loss: 0.987094521522522:  90%|████████▉ | 5820/6470 [1:12:43<08:11,  1.32it/s] 
2|5821|Loss: 0.987094521522522:  90%|████████▉ | 5821/6470 [1:12:43<08:15,  1.31it/s]
2|5822|Loss: 0.7238804697990417:  90%|████████▉ | 5821/6470 [1:12:44<08:15,  1.31it/s]
2|5822|Loss: 0.7238804697990417:  90%|████████▉ | 5822/6470 [1:12:44<08:14,  1.31it/s]
2|5823|Loss: 0.7267646193504333:  90%|████████▉ | 5822/6470 [1:12:45<08:14,  1.31it/s]
2|5823|Loss: 0.7267646193504333:  90%|█████████ | 5823/6470 [1:12:45<07:58,  1.35it/s]
2|5824|Loss: 0.800544798374176:  90%|█████████ | 5823/6470 [1:12:45<07:58,  1.35it/s] 
2|5824|Loss: 0.800544798374176:  90%|█████████ | 5824/6470 [1:12:46<07:56,  1.36it/s]
2|5825|Loss: 0.8394586443901062:  90%|█████████ | 5824/6470 [1:12:46<07:56,  1.36it/s]
2|5825|Loss: 0.8394586443901062:  90%|█████████ | 5825/6470 [1:12:46<07:44,  1.39it/s]
2|5826|Loss: 0.8863047957420349:  90%|█████████ | 5825/6470 [1:12:47<07:44,  1.39it/s]
2|5826|Loss: 0.8863047957420349:  90%|█████████ | 5826/6470 [1:12:47<07:48,  1.38it/s]
2|5827|Loss: 0.9066498279571533:  90%|█████████ | 5826/6470 [1:12:48<07:48,  1.38it/s]
2|5827|Loss: 0.9066498279571533:  90%|█████████ | 5827/6470 [1:12:48<08:17,  1.29it/s]
2|5828|Loss: 0.8484499454498291:  90%|█████████ | 5827/6470 [1:12:48<08:17,  1.29it/s]
2|5828|Loss: 0.8484499454498291:  90%|█████████ | 5828/6470 [1:12:49<08:23,  1.27it/s]
2|5829|Loss: 0.769317090511322:  90%|█████████ | 5828/6470 [1:12:49<08:23,  1.27it/s] 
2|5829|Loss: 0.769317090511322:  90%|█████████ | 5829/6470 [1:12:49<08:03,  1.33it/s]
2|5830|Loss: 0.854268491268158:  90%|█████████ | 5829/6470 [1:12:50<08:03,  1.33it/s]
2|5830|Loss: 0.854268491268158:  90%|█████████ | 5830/6470 [1:12:50<08:01,  1.33it/s]
2|5831|Loss: 0.8434237241744995:  90%|█████████ | 5830/6470 [1:12:51<08:01,  1.33it/s]
2|5831|Loss: 0.8434237241744995:  90%|█████████ | 5831/6470 [1:12:51<07:46,  1.37it/s]
2|5832|Loss: 0.8934956192970276:  90%|█████████ | 5831/6470 [1:12:51<07:46,  1.37it/s]
2|5832|Loss: 0.8934956192970276:  90%|█████████ | 5832/6470 [1:12:52<07:36,  1.40it/s]
2|5833|Loss: 0.8417982459068298:  90%|█████████ | 5832/6470 [1:12:52<07:36,  1.40it/s]
2|5833|Loss: 0.8417982459068298:  90%|█████████ | 5833/6470 [1:12:52<08:03,  1.32it/s]
2|5834|Loss: 1.1659200191497803:  90%|█████████ | 5833/6470 [1:12:53<08:03,  1.32it/s]
2|5834|Loss: 1.1659200191497803:  90%|█████████ | 5834/6470 [1:12:53<08:34,  1.24it/s]
2|5835|Loss: 0.8948602676391602:  90%|█████████ | 5834/6470 [1:12:54<08:34,  1.24it/s]
2|5835|Loss: 0.8948602676391602:  90%|█████████ | 5835/6470 [1:12:54<08:40,  1.22it/s]
2|5836|Loss: 0.8740904331207275:  90%|█████████ | 5835/6470 [1:12:55<08:40,  1.22it/s]
2|5836|Loss: 0.8740904331207275:  90%|█████████ | 5836/6470 [1:12:55<08:16,  1.28it/s]
2|5837|Loss: 0.8131189942359924:  90%|█████████ | 5836/6470 [1:12:55<08:16,  1.28it/s]
2|5837|Loss: 0.8131189942359924:  90%|█████████ | 5837/6470 [1:12:56<08:14,  1.28it/s]
2|5838|Loss: 0.7719288468360901:  90%|█████████ | 5837/6470 [1:12:56<08:14,  1.28it/s]
2|5838|Loss: 0.7719288468360901:  90%|█████████ | 5838/6470 [1:12:56<07:52,  1.34it/s]
2|5839|Loss: 0.830236554145813:  90%|█████████ | 5838/6470 [1:12:57<07:52,  1.34it/s] 
2|5839|Loss: 0.830236554145813:  90%|█████████ | 5839/6470 [1:12:57<07:27,  1.41it/s]
2|5840|Loss: 0.8780553340911865:  90%|█████████ | 5839/6470 [1:12:57<07:27,  1.41it/s]
2|5840|Loss: 0.8780553340911865:  90%|█████████ | 5840/6470 [1:12:58<07:32,  1.39it/s]
2|5841|Loss: 0.9024234414100647:  90%|█████████ | 5840/6470 [1:12:58<07:32,  1.39it/s]
2|5841|Loss: 0.9024234414100647:  90%|█████████ | 5841/6470 [1:12:59<08:00,  1.31it/s]
2|5842|Loss: 0.7788496017456055:  90%|█████████ | 5841/6470 [1:12:59<08:00,  1.31it/s]
2|5842|Loss: 0.7788496017456055:  90%|█████████ | 5842/6470 [1:12:59<08:02,  1.30it/s]
2|5843|Loss: 0.789997935295105:  90%|█████████ | 5842/6470 [1:13:00<08:02,  1.30it/s] 
2|5843|Loss: 0.789997935295105:  90%|█████████ | 5843/6470 [1:13:00<07:59,  1.31it/s]
2|5844|Loss: 0.7373793721199036:  90%|█████████ | 5843/6470 [1:13:01<07:59,  1.31it/s]
2|5844|Loss: 0.7373793721199036:  90%|█████████ | 5844/6470 [1:13:01<07:45,  1.34it/s]
2|5845|Loss: 0.9587801098823547:  90%|█████████ | 5844/6470 [1:13:01<07:45,  1.34it/s]
2|5845|Loss: 0.9587801098823547:  90%|█████████ | 5845/6470 [1:13:02<08:01,  1.30it/s]
2|5846|Loss: 0.8747075200080872:  90%|█████████ | 5845/6470 [1:13:02<08:01,  1.30it/s]
2|5846|Loss: 0.8747075200080872:  90%|█████████ | 5846/6470 [1:13:03<08:27,  1.23it/s]
2|5847|Loss: 0.7439371943473816:  90%|█████████ | 5846/6470 [1:13:03<08:27,  1.23it/s]
2|5847|Loss: 0.7439371943473816:  90%|█████████ | 5847/6470 [1:13:03<08:22,  1.24it/s]
2|5848|Loss: 0.8443394899368286:  90%|█████████ | 5847/6470 [1:13:04<08:22,  1.24it/s]
2|5848|Loss: 0.8443394899368286:  90%|█████████ | 5848/6470 [1:13:04<08:03,  1.29it/s]
2|5849|Loss: 1.0130459070205688:  90%|█████████ | 5848/6470 [1:13:04<08:03,  1.29it/s]
2|5849|Loss: 1.0130459070205688:  90%|█████████ | 5849/6470 [1:13:05<07:14,  1.43it/s]
2|5850|Loss: 0.7280785441398621:  90%|█████████ | 5849/6470 [1:13:05<07:14,  1.43it/s]
2|5850|Loss: 0.7280785441398621:  90%|█████████ | 5850/6470 [1:13:05<07:40,  1.34it/s]
2|5851|Loss: 0.9152161478996277:  90%|█████████ | 5850/6470 [1:13:06<07:40,  1.34it/s]
2|5851|Loss: 0.9152161478996277:  90%|█████████ | 5851/6470 [1:13:06<08:14,  1.25it/s]
2|5852|Loss: 1.0379440784454346:  90%|█████████ | 5851/6470 [1:13:07<08:14,  1.25it/s]
2|5852|Loss: 1.0379440784454346:  90%|█████████ | 5852/6470 [1:13:07<08:03,  1.28it/s]
2|5853|Loss: 0.807153582572937:  90%|█████████ | 5852/6470 [1:13:08<08:03,  1.28it/s] 
2|5853|Loss: 0.807153582572937:  90%|█████████ | 5853/6470 [1:13:08<08:19,  1.23it/s]
2|5854|Loss: 0.790965735912323:  90%|█████████ | 5853/6470 [1:13:08<08:19,  1.23it/s]
2|5854|Loss: 0.790965735912323:  90%|█████████ | 5854/6470 [1:13:09<08:18,  1.24it/s]
2|5855|Loss: 0.8083379864692688:  90%|█████████ | 5854/6470 [1:13:09<08:18,  1.24it/s]
2|5855|Loss: 0.8083379864692688:  90%|█████████ | 5855/6470 [1:13:10<08:53,  1.15it/s]
2|5856|Loss: 0.7640201449394226:  90%|█████████ | 5855/6470 [1:13:10<08:53,  1.15it/s]
2|5856|Loss: 0.7640201449394226:  91%|█████████ | 5856/6470 [1:13:11<08:32,  1.20it/s]
2|5857|Loss: 0.8089394569396973:  91%|█████████ | 5856/6470 [1:13:11<08:32,  1.20it/s]
2|5857|Loss: 0.8089394569396973:  91%|█████████ | 5857/6470 [1:13:11<07:44,  1.32it/s]
2|5858|Loss: 0.8398190140724182:  91%|█████████ | 5857/6470 [1:13:11<07:44,  1.32it/s]
2|5858|Loss: 0.8398190140724182:  91%|█████████ | 5858/6470 [1:13:12<07:05,  1.44it/s]
2|5859|Loss: 0.8110771179199219:  91%|█████████ | 5858/6470 [1:13:12<07:05,  1.44it/s]
2|5859|Loss: 0.8110771179199219:  91%|█████████ | 5859/6470 [1:13:12<07:06,  1.43it/s]
2|5860|Loss: 0.9044261574745178:  91%|█████████ | 5859/6470 [1:13:13<07:06,  1.43it/s]
2|5860|Loss: 0.9044261574745178:  91%|█████████ | 5860/6470 [1:13:13<07:16,  1.40it/s]
2|5861|Loss: 0.8675516843795776:  91%|█████████ | 5860/6470 [1:13:14<07:16,  1.40it/s]
2|5861|Loss: 0.8675516843795776:  91%|█████████ | 5861/6470 [1:13:14<07:31,  1.35it/s]
2|5862|Loss: 0.7186931371688843:  91%|█████████ | 5861/6470 [1:13:14<07:31,  1.35it/s]
2|5862|Loss: 0.7186931371688843:  91%|█████████ | 5862/6470 [1:13:15<07:55,  1.28it/s]
2|5863|Loss: 0.9090012311935425:  91%|█████████ | 5862/6470 [1:13:15<07:55,  1.28it/s]
2|5863|Loss: 0.9090012311935425:  91%|█████████ | 5863/6470 [1:13:16<08:08,  1.24it/s]
2|5864|Loss: 0.6885666847229004:  91%|█████████ | 5863/6470 [1:13:16<08:08,  1.24it/s]
2|5864|Loss: 0.6885666847229004:  91%|█████████ | 5864/6470 [1:13:16<08:05,  1.25it/s]
2|5865|Loss: 0.9485982060432434:  91%|█████████ | 5864/6470 [1:13:17<08:05,  1.25it/s]
2|5865|Loss: 0.9485982060432434:  91%|█████████ | 5865/6470 [1:13:17<08:21,  1.21it/s]
2|5866|Loss: 0.979099690914154:  91%|█████████ | 5865/6470 [1:13:18<08:21,  1.21it/s] 
2|5866|Loss: 0.979099690914154:  91%|█████████ | 5866/6470 [1:13:18<08:06,  1.24it/s]
2|5867|Loss: 0.8946318626403809:  91%|█████████ | 5866/6470 [1:13:19<08:06,  1.24it/s]
2|5867|Loss: 0.8946318626403809:  91%|█████████ | 5867/6470 [1:13:19<07:55,  1.27it/s]
2|5868|Loss: 0.8260572552680969:  91%|█████████ | 5867/6470 [1:13:19<07:55,  1.27it/s]
2|5868|Loss: 0.8260572552680969:  91%|█████████ | 5868/6470 [1:13:20<07:40,  1.31it/s]
2|5869|Loss: 0.7922128438949585:  91%|█████████ | 5868/6470 [1:13:20<07:40,  1.31it/s]
2|5869|Loss: 0.7922128438949585:  91%|█████████ | 5869/6470 [1:13:20<07:16,  1.38it/s]
2|5870|Loss: 0.811973512172699:  91%|█████████ | 5869/6470 [1:13:21<07:16,  1.38it/s] 
2|5870|Loss: 0.811973512172699:  91%|█████████ | 5870/6470 [1:13:21<07:20,  1.36it/s]
2|5871|Loss: 1.0922296047210693:  91%|█████████ | 5870/6470 [1:13:21<07:20,  1.36it/s]
2|5871|Loss: 1.0922296047210693:  91%|█████████ | 5871/6470 [1:13:22<07:29,  1.33it/s]
2|5872|Loss: 0.8548712134361267:  91%|█████████ | 5871/6470 [1:13:22<07:29,  1.33it/s]
2|5872|Loss: 0.8548712134361267:  91%|█████████ | 5872/6470 [1:13:23<07:35,  1.31it/s]
2|5873|Loss: 0.8872667551040649:  91%|█████████ | 5872/6470 [1:13:23<07:35,  1.31it/s]
2|5873|Loss: 0.8872667551040649:  91%|█████████ | 5873/6470 [1:13:23<07:39,  1.30it/s]
2|5874|Loss: 0.8954113721847534:  91%|█████████ | 5873/6470 [1:13:24<07:39,  1.30it/s]
2|5874|Loss: 0.8954113721847534:  91%|█████████ | 5874/6470 [1:13:24<06:39,  1.49it/s]
2|5875|Loss: 0.993121325969696:  91%|█████████ | 5874/6470 [1:13:24<06:39,  1.49it/s] 
2|5875|Loss: 0.993121325969696:  91%|█████████ | 5875/6470 [1:13:24<06:45,  1.47it/s]
2|5876|Loss: 0.9808202385902405:  91%|█████████ | 5875/6470 [1:13:25<06:45,  1.47it/s]
2|5876|Loss: 0.9808202385902405:  91%|█████████ | 5876/6470 [1:13:25<06:43,  1.47it/s]
2|5877|Loss: 0.6865800023078918:  91%|█████████ | 5876/6470 [1:13:26<06:43,  1.47it/s]
2|5877|Loss: 0.6865800023078918:  91%|█████████ | 5877/6470 [1:13:26<07:12,  1.37it/s]
2|5878|Loss: 0.7962172031402588:  91%|█████████ | 5877/6470 [1:13:26<07:12,  1.37it/s]
2|5878|Loss: 0.7962172031402588:  91%|█████████ | 5878/6470 [1:13:27<07:27,  1.32it/s]
2|5879|Loss: 0.9043998122215271:  91%|█████████ | 5878/6470 [1:13:27<07:27,  1.32it/s]
2|5879|Loss: 0.9043998122215271:  91%|█████████ | 5879/6470 [1:13:28<07:21,  1.34it/s]
2|5880|Loss: 0.9450492858886719:  91%|█████████ | 5879/6470 [1:13:28<07:21,  1.34it/s]
2|5880|Loss: 0.9450492858886719:  91%|█████████ | 5880/6470 [1:13:28<07:35,  1.30it/s]
2|5881|Loss: 0.8236584067344666:  91%|█████████ | 5880/6470 [1:13:29<07:35,  1.30it/s]
2|5881|Loss: 0.8236584067344666:  91%|█████████ | 5881/6470 [1:13:29<07:35,  1.29it/s]
2|5882|Loss: 0.8825292587280273:  91%|█████████ | 5881/6470 [1:13:30<07:35,  1.29it/s]
2|5882|Loss: 0.8825292587280273:  91%|█████████ | 5882/6470 [1:13:30<07:22,  1.33it/s]
2|5883|Loss: 0.7860356569290161:  91%|█████████ | 5882/6470 [1:13:30<07:22,  1.33it/s]
2|5883|Loss: 0.7860356569290161:  91%|█████████ | 5883/6470 [1:13:31<07:28,  1.31it/s]
2|5884|Loss: 0.9654296040534973:  91%|█████████ | 5883/6470 [1:13:31<07:28,  1.31it/s]
2|5884|Loss: 0.9654296040534973:  91%|█████████ | 5884/6470 [1:13:32<07:48,  1.25it/s]
2|5885|Loss: 0.8605259656906128:  91%|█████████ | 5884/6470 [1:13:32<07:48,  1.25it/s]
2|5885|Loss: 0.8605259656906128:  91%|█████████ | 5885/6470 [1:13:32<07:47,  1.25it/s]
2|5886|Loss: 0.8147857189178467:  91%|█████████ | 5885/6470 [1:13:33<07:47,  1.25it/s]
2|5886|Loss: 0.8147857189178467:  91%|█████████ | 5886/6470 [1:13:33<07:20,  1.33it/s]
2|5887|Loss: 0.7026833295822144:  91%|█████████ | 5886/6470 [1:13:33<07:20,  1.33it/s]
2|5887|Loss: 0.7026833295822144:  91%|█████████ | 5887/6470 [1:13:34<07:00,  1.38it/s]
2|5888|Loss: 0.8582634925842285:  91%|█████████ | 5887/6470 [1:13:34<07:00,  1.38it/s]
2|5888|Loss: 0.8582634925842285:  91%|█████████ | 5888/6470 [1:13:34<06:43,  1.44it/s]
2|5889|Loss: 0.903368353843689:  91%|█████████ | 5888/6470 [1:13:35<06:43,  1.44it/s] 
2|5889|Loss: 0.903368353843689:  91%|█████████ | 5889/6470 [1:13:35<06:09,  1.57it/s]
2|5890|Loss: 0.6987091898918152:  91%|█████████ | 5889/6470 [1:13:35<06:09,  1.57it/s]
2|5890|Loss: 0.6987091898918152:  91%|█████████ | 5890/6470 [1:13:35<06:10,  1.57it/s]
2|5891|Loss: 0.8888143301010132:  91%|█████████ | 5890/6470 [1:13:36<06:10,  1.57it/s]
2|5891|Loss: 0.8888143301010132:  91%|█████████ | 5891/6470 [1:13:36<06:27,  1.49it/s]
2|5892|Loss: 0.7714353203773499:  91%|█████████ | 5891/6470 [1:13:37<06:27,  1.49it/s]
2|5892|Loss: 0.7714353203773499:  91%|█████████ | 5892/6470 [1:13:37<06:39,  1.45it/s]
2|5893|Loss: 0.8304247856140137:  91%|█████████ | 5892/6470 [1:13:37<06:39,  1.45it/s]
2|5893|Loss: 0.8304247856140137:  91%|█████████ | 5893/6470 [1:13:38<06:49,  1.41it/s]
2|5894|Loss: 1.0361841917037964:  91%|█████████ | 5893/6470 [1:13:38<06:49,  1.41it/s]
2|5894|Loss: 1.0361841917037964:  91%|█████████ | 5894/6470 [1:13:38<06:55,  1.39it/s]
2|5895|Loss: 0.9412581324577332:  91%|█████████ | 5894/6470 [1:13:39<06:55,  1.39it/s]
2|5895|Loss: 0.9412581324577332:  91%|█████████ | 5895/6470 [1:13:39<07:01,  1.36it/s]
2|5896|Loss: 0.6584684252738953:  91%|█████████ | 5895/6470 [1:13:40<07:01,  1.36it/s]
2|5896|Loss: 0.6584684252738953:  91%|█████████ | 5896/6470 [1:13:40<07:08,  1.34it/s]
2|5897|Loss: 0.8845404982566833:  91%|█████████ | 5896/6470 [1:13:40<07:08,  1.34it/s]
2|5897|Loss: 0.8845404982566833:  91%|█████████ | 5897/6470 [1:13:40<06:37,  1.44it/s]
2|5898|Loss: 0.8847183585166931:  91%|█████████ | 5897/6470 [1:13:41<06:37,  1.44it/s]
2|5898|Loss: 0.8847183585166931:  91%|█████████ | 5898/6470 [1:13:41<06:25,  1.48it/s]
2|5899|Loss: 0.8608505129814148:  91%|█████████ | 5898/6470 [1:13:41<06:25,  1.48it/s]
2|5899|Loss: 0.8608505129814148:  91%|█████████ | 5899/6470 [1:13:42<06:24,  1.48it/s]
2|5900|Loss: 1.0970211029052734:  91%|█████████ | 5899/6470 [1:13:42<06:24,  1.48it/s]
2|5900|Loss: 1.0970211029052734:  91%|█████████ | 5900/6470 [1:13:43<06:34,  1.45it/s]
2|5901|Loss: 0.7463648319244385:  91%|█████████ | 5900/6470 [1:13:43<06:34,  1.45it/s]
2|5901|Loss: 0.7463648319244385:  91%|█████████ | 5901/6470 [1:13:43<06:47,  1.40it/s]
2|5902|Loss: 0.8265477418899536:  91%|█████████ | 5901/6470 [1:13:44<06:47,  1.40it/s]
2|5902|Loss: 0.8265477418899536:  91%|█████████ | 5902/6470 [1:13:44<06:54,  1.37it/s]
2|5903|Loss: 0.6893424987792969:  91%|█████████ | 5902/6470 [1:13:45<06:54,  1.37it/s]
2|5903|Loss: 0.6893424987792969:  91%|█████████ | 5903/6470 [1:13:45<07:26,  1.27it/s]
2|5904|Loss: 0.8279926776885986:  91%|█████████ | 5903/6470 [1:13:45<07:26,  1.27it/s]
2|5904|Loss: 0.8279926776885986:  91%|█████████▏| 5904/6470 [1:13:46<07:37,  1.24it/s]
2|5905|Loss: 0.6455543041229248:  91%|█████████▏| 5904/6470 [1:13:46<07:37,  1.24it/s]
2|5905|Loss: 0.6455543041229248:  91%|█████████▏| 5905/6470 [1:13:47<07:18,  1.29it/s]
2|5906|Loss: 0.8101959824562073:  91%|█████████▏| 5905/6470 [1:13:47<07:18,  1.29it/s]
2|5906|Loss: 0.8101959824562073:  91%|█████████▏| 5906/6470 [1:13:47<07:23,  1.27it/s]
2|5907|Loss: 0.7907177805900574:  91%|█████████▏| 5906/6470 [1:13:48<07:23,  1.27it/s]
2|5907|Loss: 0.7907177805900574:  91%|█████████▏| 5907/6470 [1:13:48<07:03,  1.33it/s]
2|5908|Loss: 1.0360417366027832:  91%|█████████▏| 5907/6470 [1:13:48<07:03,  1.33it/s]
2|5908|Loss: 1.0360417366027832:  91%|█████████▏| 5908/6470 [1:13:49<07:03,  1.33it/s]
2|5909|Loss: 0.7252005934715271:  91%|█████████▏| 5908/6470 [1:13:49<07:03,  1.33it/s]
2|5909|Loss: 0.7252005934715271:  91%|█████████▏| 5909/6470 [1:13:49<06:55,  1.35it/s]
2|5910|Loss: 0.9600440263748169:  91%|█████████▏| 5909/6470 [1:13:50<06:55,  1.35it/s]
2|5910|Loss: 0.9600440263748169:  91%|█████████▏| 5910/6470 [1:13:50<06:27,  1.44it/s]
2|5911|Loss: 0.7993836998939514:  91%|█████████▏| 5910/6470 [1:13:50<06:27,  1.44it/s]
2|5911|Loss: 0.7993836998939514:  91%|█████████▏| 5911/6470 [1:13:51<06:34,  1.42it/s]
2|5912|Loss: 0.737082839012146:  91%|█████████▏| 5911/6470 [1:13:51<06:34,  1.42it/s] 
2|5912|Loss: 0.737082839012146:  91%|█████████▏| 5912/6470 [1:13:52<07:10,  1.30it/s]
2|5913|Loss: 0.8449538946151733:  91%|█████████▏| 5912/6470 [1:13:52<07:10,  1.30it/s]
2|5913|Loss: 0.8449538946151733:  91%|█████████▏| 5913/6470 [1:13:53<07:48,  1.19it/s]
2|5914|Loss: 0.89240562915802:  91%|█████████▏| 5913/6470 [1:13:53<07:48,  1.19it/s]  
2|5914|Loss: 0.89240562915802:  91%|█████████▏| 5914/6470 [1:13:53<07:25,  1.25it/s]
2|5915|Loss: 0.8081111311912537:  91%|█████████▏| 5914/6470 [1:13:54<07:25,  1.25it/s]
2|5915|Loss: 0.8081111311912537:  91%|█████████▏| 5915/6470 [1:13:54<06:58,  1.33it/s]
2|5916|Loss: 0.8175677061080933:  91%|█████████▏| 5915/6470 [1:13:54<06:58,  1.33it/s]
2|5916|Loss: 0.8175677061080933:  91%|█████████▏| 5916/6470 [1:13:55<06:54,  1.34it/s]
2|5917|Loss: 0.7854844331741333:  91%|█████████▏| 5916/6470 [1:13:55<06:54,  1.34it/s]
2|5917|Loss: 0.7854844331741333:  91%|█████████▏| 5917/6470 [1:13:55<06:39,  1.38it/s]
2|5918|Loss: 0.7919373512268066:  91%|█████████▏| 5917/6470 [1:13:56<06:39,  1.38it/s]
2|5918|Loss: 0.7919373512268066:  91%|█████████▏| 5918/6470 [1:13:56<06:31,  1.41it/s]
2|5919|Loss: 0.9594741463661194:  91%|█████████▏| 5918/6470 [1:13:57<06:31,  1.41it/s]
2|5919|Loss: 0.9594741463661194:  91%|█████████▏| 5919/6470 [1:13:57<06:29,  1.42it/s]
2|5920|Loss: 0.8779516220092773:  91%|█████████▏| 5919/6470 [1:13:57<06:29,  1.42it/s]
2|5920|Loss: 0.8779516220092773:  91%|█████████▏| 5920/6470 [1:13:58<06:38,  1.38it/s]
2|5921|Loss: 0.7616404891014099:  91%|█████████▏| 5920/6470 [1:13:58<06:38,  1.38it/s]
2|5921|Loss: 0.7616404891014099:  92%|█████████▏| 5921/6470 [1:13:58<06:44,  1.36it/s]
2|5922|Loss: 0.6919481754302979:  92%|█████████▏| 5921/6470 [1:13:59<06:44,  1.36it/s]
2|5922|Loss: 0.6919481754302979:  92%|█████████▏| 5922/6470 [1:13:59<06:59,  1.31it/s]
2|5923|Loss: 0.8633951544761658:  92%|█████████▏| 5922/6470 [1:14:00<06:59,  1.31it/s]
2|5923|Loss: 0.8633951544761658:  92%|█████████▏| 5923/6470 [1:14:00<06:44,  1.35it/s]
2|5924|Loss: 0.8471625447273254:  92%|█████████▏| 5923/6470 [1:14:00<06:44,  1.35it/s]
2|5924|Loss: 0.8471625447273254:  92%|█████████▏| 5924/6470 [1:14:01<06:36,  1.38it/s]
2|5925|Loss: 0.9996675252914429:  92%|█████████▏| 5924/6470 [1:14:01<06:36,  1.38it/s]
2|5925|Loss: 0.9996675252914429:  92%|█████████▏| 5925/6470 [1:14:01<07:02,  1.29it/s]
2|5926|Loss: 0.8350475430488586:  92%|█████████▏| 5925/6470 [1:14:02<07:02,  1.29it/s]
2|5926|Loss: 0.8350475430488586:  92%|█████████▏| 5926/6470 [1:14:02<07:05,  1.28it/s]
2|5927|Loss: 0.7600578665733337:  92%|█████████▏| 5926/6470 [1:14:03<07:05,  1.28it/s]
2|5927|Loss: 0.7600578665733337:  92%|█████████▏| 5927/6470 [1:14:03<07:09,  1.26it/s]
2|5928|Loss: 0.7112141847610474:  92%|█████████▏| 5927/6470 [1:14:04<07:09,  1.26it/s]
2|5928|Loss: 0.7112141847610474:  92%|█████████▏| 5928/6470 [1:14:04<07:20,  1.23it/s]
2|5929|Loss: 0.735153079032898:  92%|█████████▏| 5928/6470 [1:14:04<07:20,  1.23it/s] 
2|5929|Loss: 0.735153079032898:  92%|█████████▏| 5929/6470 [1:14:05<07:17,  1.24it/s]
2|5930|Loss: 0.7704819440841675:  92%|█████████▏| 5929/6470 [1:14:05<07:17,  1.24it/s]
2|5930|Loss: 0.7704819440841675:  92%|█████████▏| 5930/6470 [1:14:06<07:24,  1.21it/s]
2|5931|Loss: 0.768465518951416:  92%|█████████▏| 5930/6470 [1:14:06<07:24,  1.21it/s] 
2|5931|Loss: 0.768465518951416:  92%|█████████▏| 5931/6470 [1:14:06<06:59,  1.29it/s]
2|5932|Loss: 0.8689702153205872:  92%|█████████▏| 5931/6470 [1:14:07<06:59,  1.29it/s]
2|5932|Loss: 0.8689702153205872:  92%|█████████▏| 5932/6470 [1:14:07<06:50,  1.31it/s]
2|5933|Loss: 0.8072001934051514:  92%|█████████▏| 5932/6470 [1:14:07<06:50,  1.31it/s]
2|5933|Loss: 0.8072001934051514:  92%|█████████▏| 5933/6470 [1:14:08<07:05,  1.26it/s]
2|5934|Loss: 1.067335844039917:  92%|█████████▏| 5933/6470 [1:14:08<07:05,  1.26it/s] 
2|5934|Loss: 1.067335844039917:  92%|█████████▏| 5934/6470 [1:14:09<07:08,  1.25it/s]
2|5935|Loss: 0.8385209441184998:  92%|█████████▏| 5934/6470 [1:14:09<07:08,  1.25it/s]
2|5935|Loss: 0.8385209441184998:  92%|█████████▏| 5935/6470 [1:14:09<07:06,  1.26it/s]
2|5936|Loss: 0.8554277420043945:  92%|█████████▏| 5935/6470 [1:14:10<07:06,  1.26it/s]
2|5936|Loss: 0.8554277420043945:  92%|█████████▏| 5936/6470 [1:14:10<07:06,  1.25it/s]
2|5937|Loss: 0.9745858311653137:  92%|█████████▏| 5936/6470 [1:14:11<07:06,  1.25it/s]
2|5937|Loss: 0.9745858311653137:  92%|█████████▏| 5937/6470 [1:14:11<06:39,  1.33it/s]
2|5938|Loss: 0.8633285760879517:  92%|█████████▏| 5937/6470 [1:14:11<06:39,  1.33it/s]
2|5938|Loss: 0.8633285760879517:  92%|█████████▏| 5938/6470 [1:14:12<06:22,  1.39it/s]
2|5939|Loss: 0.7045115828514099:  92%|█████████▏| 5938/6470 [1:14:12<06:22,  1.39it/s]
2|5939|Loss: 0.7045115828514099:  92%|█████████▏| 5939/6470 [1:14:12<06:25,  1.38it/s]
2|5940|Loss: 0.7759952545166016:  92%|█████████▏| 5939/6470 [1:14:13<06:25,  1.38it/s]
2|5940|Loss: 0.7759952545166016:  92%|█████████▏| 5940/6470 [1:14:13<06:17,  1.40it/s]
2|5941|Loss: 0.9722893238067627:  92%|█████████▏| 5940/6470 [1:14:13<06:17,  1.40it/s]
2|5941|Loss: 0.9722893238067627:  92%|█████████▏| 5941/6470 [1:14:14<06:00,  1.47it/s]
2|5942|Loss: 0.8825609683990479:  92%|█████████▏| 5941/6470 [1:14:14<06:00,  1.47it/s]
2|5942|Loss: 0.8825609683990479:  92%|█████████▏| 5942/6470 [1:14:14<05:56,  1.48it/s]
2|5943|Loss: 0.7881106734275818:  92%|█████████▏| 5942/6470 [1:14:15<05:56,  1.48it/s]
2|5943|Loss: 0.7881106734275818:  92%|█████████▏| 5943/6470 [1:14:15<06:06,  1.44it/s]
2|5944|Loss: 0.9844400882720947:  92%|█████████▏| 5943/6470 [1:14:15<06:06,  1.44it/s]
2|5944|Loss: 0.9844400882720947:  92%|█████████▏| 5944/6470 [1:14:16<06:19,  1.39it/s]
2|5945|Loss: 0.7904850244522095:  92%|█████████▏| 5944/6470 [1:14:16<06:19,  1.39it/s]
2|5945|Loss: 0.7904850244522095:  92%|█████████▏| 5945/6470 [1:14:17<06:37,  1.32it/s]
2|5946|Loss: 0.7924805283546448:  92%|█████████▏| 5945/6470 [1:14:17<06:37,  1.32it/s]
2|5946|Loss: 0.7924805283546448:  92%|█████████▏| 5946/6470 [1:14:17<06:33,  1.33it/s]
2|5947|Loss: 0.8501774668693542:  92%|█████████▏| 5946/6470 [1:14:18<06:33,  1.33it/s]
2|5947|Loss: 0.8501774668693542:  92%|█████████▏| 5947/6470 [1:14:18<06:37,  1.31it/s]
2|5948|Loss: 0.7624102234840393:  92%|█████████▏| 5947/6470 [1:14:19<06:37,  1.31it/s]
2|5948|Loss: 0.7624102234840393:  92%|█████████▏| 5948/6470 [1:14:19<06:26,  1.35it/s]
2|5949|Loss: 0.7754110097885132:  92%|█████████▏| 5948/6470 [1:14:19<06:26,  1.35it/s]
2|5949|Loss: 0.7754110097885132:  92%|█████████▏| 5949/6470 [1:14:20<06:57,  1.25it/s]
2|5950|Loss: 0.8447824120521545:  92%|█████████▏| 5949/6470 [1:14:20<06:57,  1.25it/s]
2|5950|Loss: 0.8447824120521545:  92%|█████████▏| 5950/6470 [1:14:21<07:42,  1.12it/s]
2|5951|Loss: 0.8571181297302246:  92%|█████████▏| 5950/6470 [1:14:21<07:42,  1.12it/s]
2|5951|Loss: 0.8571181297302246:  92%|█████████▏| 5951/6470 [1:14:22<07:23,  1.17it/s]
2|5952|Loss: 0.7158694863319397:  92%|█████████▏| 5951/6470 [1:14:22<07:23,  1.17it/s]
2|5952|Loss: 0.7158694863319397:  92%|█████████▏| 5952/6470 [1:14:22<06:39,  1.30it/s]
2|5953|Loss: 0.9188745021820068:  92%|█████████▏| 5952/6470 [1:14:23<06:39,  1.30it/s]
2|5953|Loss: 0.9188745021820068:  92%|█████████▏| 5953/6470 [1:14:23<06:18,  1.36it/s]
2|5954|Loss: 0.8457555174827576:  92%|█████████▏| 5953/6470 [1:14:23<06:18,  1.36it/s]
2|5954|Loss: 0.8457555174827576:  92%|█████████▏| 5954/6470 [1:14:24<06:34,  1.31it/s]
2|5955|Loss: 0.6938083171844482:  92%|█████████▏| 5954/6470 [1:14:24<06:34,  1.31it/s]
2|5955|Loss: 0.6938083171844482:  92%|█████████▏| 5955/6470 [1:14:24<06:00,  1.43it/s]
2|5956|Loss: 1.0306166410446167:  92%|█████████▏| 5955/6470 [1:14:25<06:00,  1.43it/s]
2|5956|Loss: 1.0306166410446167:  92%|█████████▏| 5956/6470 [1:14:25<05:48,  1.48it/s]
2|5957|Loss: 0.6723607182502747:  92%|█████████▏| 5956/6470 [1:14:25<05:48,  1.48it/s]
2|5957|Loss: 0.6723607182502747:  92%|█████████▏| 5957/6470 [1:14:25<05:28,  1.56it/s]
2|5958|Loss: 0.8825379610061646:  92%|█████████▏| 5957/6470 [1:14:26<05:28,  1.56it/s]
2|5958|Loss: 0.8825379610061646:  92%|█████████▏| 5958/6470 [1:14:26<05:36,  1.52it/s]
2|5959|Loss: 0.749733567237854:  92%|█████████▏| 5958/6470 [1:14:27<05:36,  1.52it/s] 
2|5959|Loss: 0.749733567237854:  92%|█████████▏| 5959/6470 [1:14:27<06:10,  1.38it/s]
2|5960|Loss: 0.8604150414466858:  92%|█████████▏| 5959/6470 [1:14:28<06:10,  1.38it/s]
2|5960|Loss: 0.8604150414466858:  92%|█████████▏| 5960/6470 [1:14:28<06:39,  1.28it/s]
2|5961|Loss: 0.7842488288879395:  92%|█████████▏| 5960/6470 [1:14:28<06:39,  1.28it/s]
2|5961|Loss: 0.7842488288879395:  92%|█████████▏| 5961/6470 [1:14:29<06:38,  1.28it/s]
2|5962|Loss: 0.8962883353233337:  92%|█████████▏| 5961/6470 [1:14:29<06:38,  1.28it/s]
2|5962|Loss: 0.8962883353233337:  92%|█████████▏| 5962/6470 [1:14:29<06:37,  1.28it/s]
2|5963|Loss: 0.7899142503738403:  92%|█████████▏| 5962/6470 [1:14:30<06:37,  1.28it/s]
2|5963|Loss: 0.7899142503738403:  92%|█████████▏| 5963/6470 [1:14:30<06:36,  1.28it/s]
2|5964|Loss: 0.8428714871406555:  92%|█████████▏| 5963/6470 [1:14:31<06:36,  1.28it/s]
2|5964|Loss: 0.8428714871406555:  92%|█████████▏| 5964/6470 [1:14:31<06:26,  1.31it/s]
2|5965|Loss: 0.8702861070632935:  92%|█████████▏| 5964/6470 [1:14:31<06:26,  1.31it/s]
2|5965|Loss: 0.8702861070632935:  92%|█████████▏| 5965/6470 [1:14:32<06:15,  1.34it/s]
2|5966|Loss: 0.8975467681884766:  92%|█████████▏| 5965/6470 [1:14:32<06:15,  1.34it/s]
2|5966|Loss: 0.8975467681884766:  92%|█████████▏| 5966/6470 [1:14:32<06:09,  1.37it/s]
2|5967|Loss: 0.7577584981918335:  92%|█████████▏| 5966/6470 [1:14:33<06:09,  1.37it/s]
2|5967|Loss: 0.7577584981918335:  92%|█████████▏| 5967/6470 [1:14:33<05:50,  1.44it/s]
2|5968|Loss: 0.8175417184829712:  92%|█████████▏| 5967/6470 [1:14:33<05:50,  1.44it/s]
2|5968|Loss: 0.8175417184829712:  92%|█████████▏| 5968/6470 [1:14:34<05:54,  1.42it/s]
2|5969|Loss: 0.6409152150154114:  92%|█████████▏| 5968/6470 [1:14:34<05:54,  1.42it/s]
2|5969|Loss: 0.6409152150154114:  92%|█████████▏| 5969/6470 [1:14:34<05:59,  1.39it/s]
2|5970|Loss: 0.8395239114761353:  92%|█████████▏| 5969/6470 [1:14:35<05:59,  1.39it/s]
2|5970|Loss: 0.8395239114761353:  92%|█████████▏| 5970/6470 [1:14:35<06:09,  1.35it/s]
2|5971|Loss: 0.8398700952529907:  92%|█████████▏| 5970/6470 [1:14:36<06:09,  1.35it/s]
2|5971|Loss: 0.8398700952529907:  92%|█████████▏| 5971/6470 [1:14:36<05:48,  1.43it/s]
2|5972|Loss: 0.8159167170524597:  92%|█████████▏| 5971/6470 [1:14:37<05:48,  1.43it/s]
2|5972|Loss: 0.8159167170524597:  92%|█████████▏| 5972/6470 [1:14:37<07:40,  1.08it/s]
2|5973|Loss: 1.0467435121536255:  92%|█████████▏| 5972/6470 [1:14:38<07:40,  1.08it/s]
2|5973|Loss: 1.0467435121536255:  92%|█████████▏| 5973/6470 [1:14:38<07:07,  1.16it/s]
2|5974|Loss: 0.9350579380989075:  92%|█████████▏| 5973/6470 [1:14:38<07:07,  1.16it/s]
2|5974|Loss: 0.9350579380989075:  92%|█████████▏| 5974/6470 [1:14:39<06:21,  1.30it/s]
2|5975|Loss: 0.5933998227119446:  92%|█████████▏| 5974/6470 [1:14:39<06:21,  1.30it/s]
2|5975|Loss: 0.5933998227119446:  92%|█████████▏| 5975/6470 [1:14:40<06:45,  1.22it/s]
2|5976|Loss: 0.8118012547492981:  92%|█████████▏| 5975/6470 [1:14:40<06:45,  1.22it/s]
2|5976|Loss: 0.8118012547492981:  92%|█████████▏| 5976/6470 [1:14:41<07:09,  1.15it/s]
2|5977|Loss: 0.8477550148963928:  92%|█████████▏| 5976/6470 [1:14:41<07:09,  1.15it/s]
2|5977|Loss: 0.8477550148963928:  92%|█████████▏| 5977/6470 [1:14:41<06:35,  1.25it/s]
2|5978|Loss: 0.7814431190490723:  92%|█████████▏| 5977/6470 [1:14:42<06:35,  1.25it/s]
2|5978|Loss: 0.7814431190490723:  92%|█████████▏| 5978/6470 [1:14:42<06:48,  1.20it/s]
2|5979|Loss: 0.9077422618865967:  92%|█████████▏| 5978/6470 [1:14:43<06:48,  1.20it/s]
2|5979|Loss: 0.9077422618865967:  92%|█████████▏| 5979/6470 [1:14:43<06:44,  1.21it/s]
2|5980|Loss: 0.7411748766899109:  92%|█████████▏| 5979/6470 [1:14:43<06:44,  1.21it/s]
2|5980|Loss: 0.7411748766899109:  92%|█████████▏| 5980/6470 [1:14:44<06:49,  1.20it/s]
2|5981|Loss: 0.6769146919250488:  92%|█████████▏| 5980/6470 [1:14:44<06:49,  1.20it/s]
2|5981|Loss: 0.6769146919250488:  92%|█████████▏| 5981/6470 [1:14:44<06:27,  1.26it/s]
2|5982|Loss: 0.6739605665206909:  92%|█████████▏| 5981/6470 [1:14:45<06:27,  1.26it/s]
2|5982|Loss: 0.6739605665206909:  92%|█████████▏| 5982/6470 [1:14:45<06:08,  1.32it/s]
2|5983|Loss: 0.7181880474090576:  92%|█████████▏| 5982/6470 [1:14:46<06:08,  1.32it/s]
2|5983|Loss: 0.7181880474090576:  92%|█████████▏| 5983/6470 [1:14:46<06:10,  1.32it/s]
2|5984|Loss: 0.7122706174850464:  92%|█████████▏| 5983/6470 [1:14:46<06:10,  1.32it/s]
2|5984|Loss: 0.7122706174850464:  92%|█████████▏| 5984/6470 [1:14:46<05:44,  1.41it/s]
2|5985|Loss: 0.8135740756988525:  92%|█████████▏| 5984/6470 [1:14:47<05:44,  1.41it/s]
2|5985|Loss: 0.8135740756988525:  93%|█████████▎| 5985/6470 [1:14:47<05:28,  1.48it/s]
2|5986|Loss: 0.8035706877708435:  93%|█████████▎| 5985/6470 [1:14:47<05:28,  1.48it/s]
2|5986|Loss: 0.8035706877708435:  93%|█████████▎| 5986/6470 [1:14:48<05:39,  1.43it/s]
2|5987|Loss: 0.8810561895370483:  93%|█████████▎| 5986/6470 [1:14:48<05:39,  1.43it/s]
2|5987|Loss: 0.8810561895370483:  93%|█████████▎| 5987/6470 [1:14:49<06:00,  1.34it/s]
2|5988|Loss: 0.7790767550468445:  93%|█████████▎| 5987/6470 [1:14:49<06:00,  1.34it/s]
2|5988|Loss: 0.7790767550468445:  93%|█████████▎| 5988/6470 [1:14:49<05:50,  1.37it/s]
2|5989|Loss: 0.9081323146820068:  93%|█████████▎| 5988/6470 [1:14:50<05:50,  1.37it/s]
2|5989|Loss: 0.9081323146820068:  93%|█████████▎| 5989/6470 [1:14:50<05:29,  1.46it/s]
2|5990|Loss: 0.765211820602417:  93%|█████████▎| 5989/6470 [1:14:50<05:29,  1.46it/s] 
2|5990|Loss: 0.765211820602417:  93%|█████████▎| 5990/6470 [1:14:51<05:28,  1.46it/s]
2|5991|Loss: 0.695305585861206:  93%|█████████▎| 5990/6470 [1:14:51<05:28,  1.46it/s]
2|5991|Loss: 0.695305585861206:  93%|█████████▎| 5991/6470 [1:14:51<05:40,  1.41it/s]
2|5992|Loss: 0.9148334860801697:  93%|█████████▎| 5991/6470 [1:14:52<05:40,  1.41it/s]
2|5992|Loss: 0.9148334860801697:  93%|█████████▎| 5992/6470 [1:14:52<06:04,  1.31it/s]
2|5993|Loss: 0.8606126308441162:  93%|█████████▎| 5992/6470 [1:14:53<06:04,  1.31it/s]
2|5993|Loss: 0.8606126308441162:  93%|█████████▎| 5993/6470 [1:14:53<06:00,  1.32it/s]
2|5994|Loss: 0.9960627555847168:  93%|█████████▎| 5993/6470 [1:14:53<06:00,  1.32it/s]
2|5994|Loss: 0.9960627555847168:  93%|█████████▎| 5994/6470 [1:14:54<06:20,  1.25it/s]
2|5995|Loss: 0.849730908870697:  93%|█████████▎| 5994/6470 [1:14:54<06:20,  1.25it/s] 
2|5995|Loss: 0.849730908870697:  93%|█████████▎| 5995/6470 [1:14:54<05:48,  1.36it/s]
2|5996|Loss: 0.9603014588356018:  93%|█████████▎| 5995/6470 [1:14:55<05:48,  1.36it/s]
2|5996|Loss: 0.9603014588356018:  93%|█████████▎| 5996/6470 [1:14:55<05:46,  1.37it/s]
2|5997|Loss: 0.804564893245697:  93%|█████████▎| 5996/6470 [1:14:56<05:46,  1.37it/s] 
2|5997|Loss: 0.804564893245697:  93%|█████████▎| 5997/6470 [1:14:56<05:58,  1.32it/s]
2|5998|Loss: 0.7919102311134338:  93%|█████████▎| 5997/6470 [1:14:56<05:58,  1.32it/s]
2|5998|Loss: 0.7919102311134338:  93%|█████████▎| 5998/6470 [1:14:57<05:46,  1.36it/s]
2|5999|Loss: 0.8191350698471069:  93%|█████████▎| 5998/6470 [1:14:57<05:46,  1.36it/s]
2|5999|Loss: 0.8191350698471069:  93%|█████████▎| 5999/6470 [1:14:57<05:37,  1.39it/s]
2|6000|Loss: 0.7861459255218506:  93%|█████████▎| 5999/6470 [1:14:58<05:37,  1.39it/s]
2|6000|Loss: 0.7861459255218506:  93%|█████████▎| 6000/6470 [1:14:58<05:30,  1.42it/s]
2|6001|Loss: 0.8892770409584045:  93%|█████████▎| 6000/6470 [1:14:58<05:30,  1.42it/s]
2|6001|Loss: 0.8892770409584045:  93%|█████████▎| 6001/6470 [1:14:59<05:36,  1.39it/s]
2|6002|Loss: 1.1933211088180542:  93%|█████████▎| 6001/6470 [1:14:59<05:36,  1.39it/s]
2|6002|Loss: 1.1933211088180542:  93%|█████████▎| 6002/6470 [1:14:59<05:27,  1.43it/s]
2|6003|Loss: 0.8887711763381958:  93%|█████████▎| 6002/6470 [1:15:00<05:27,  1.43it/s]
2|6003|Loss: 0.8887711763381958:  93%|█████████▎| 6003/6470 [1:15:00<05:48,  1.34it/s]
2|6004|Loss: 0.7767038345336914:  93%|█████████▎| 6003/6470 [1:15:01<05:48,  1.34it/s]
2|6004|Loss: 0.7767038345336914:  93%|█████████▎| 6004/6470 [1:15:01<06:11,  1.25it/s]
2|6005|Loss: 0.8703454732894897:  93%|█████████▎| 6004/6470 [1:15:02<06:11,  1.25it/s]
2|6005|Loss: 0.8703454732894897:  93%|█████████▎| 6005/6470 [1:15:02<06:01,  1.29it/s]
2|6006|Loss: 0.9209319949150085:  93%|█████████▎| 6005/6470 [1:15:02<06:01,  1.29it/s]
2|6006|Loss: 0.9209319949150085:  93%|█████████▎| 6006/6470 [1:15:03<05:45,  1.34it/s]
2|6007|Loss: 0.8350116014480591:  93%|█████████▎| 6006/6470 [1:15:03<05:45,  1.34it/s]
2|6007|Loss: 0.8350116014480591:  93%|█████████▎| 6007/6470 [1:15:03<05:22,  1.44it/s]
2|6008|Loss: 0.7208164930343628:  93%|█████████▎| 6007/6470 [1:15:04<05:22,  1.44it/s]
2|6008|Loss: 0.7208164930343628:  93%|█████████▎| 6008/6470 [1:15:04<04:52,  1.58it/s]
2|6009|Loss: 0.8422849774360657:  93%|█████████▎| 6008/6470 [1:15:04<04:52,  1.58it/s]
2|6009|Loss: 0.8422849774360657:  93%|█████████▎| 6009/6470 [1:15:05<05:15,  1.46it/s]
2|6010|Loss: 0.8620752096176147:  93%|█████████▎| 6009/6470 [1:15:05<05:15,  1.46it/s]
2|6010|Loss: 0.8620752096176147:  93%|█████████▎| 6010/6470 [1:15:05<05:13,  1.47it/s]
2|6011|Loss: 0.692152738571167:  93%|█████████▎| 6010/6470 [1:15:06<05:13,  1.47it/s] 
2|6011|Loss: 0.692152738571167:  93%|█████████▎| 6011/6470 [1:15:06<05:35,  1.37it/s]
2|6012|Loss: 0.8756542801856995:  93%|█████████▎| 6011/6470 [1:15:07<05:35,  1.37it/s]
2|6012|Loss: 0.8756542801856995:  93%|█████████▎| 6012/6470 [1:15:07<05:58,  1.28it/s]
2|6013|Loss: 0.843619704246521:  93%|█████████▎| 6012/6470 [1:15:07<05:58,  1.28it/s] 
2|6013|Loss: 0.843619704246521:  93%|█████████▎| 6013/6470 [1:15:08<06:09,  1.24it/s]
2|6014|Loss: 0.8193337321281433:  93%|█████████▎| 6013/6470 [1:15:08<06:09,  1.24it/s]
2|6014|Loss: 0.8193337321281433:  93%|█████████▎| 6014/6470 [1:15:09<05:55,  1.28it/s]
2|6015|Loss: 0.8976714611053467:  93%|█████████▎| 6014/6470 [1:15:09<05:55,  1.28it/s]
2|6015|Loss: 0.8976714611053467:  93%|█████████▎| 6015/6470 [1:15:09<05:49,  1.30it/s]
2|6016|Loss: 0.7718082070350647:  93%|█████████▎| 6015/6470 [1:15:10<05:49,  1.30it/s]
2|6016|Loss: 0.7718082070350647:  93%|█████████▎| 6016/6470 [1:15:10<05:46,  1.31it/s]
2|6017|Loss: 0.7472041249275208:  93%|█████████▎| 6016/6470 [1:15:10<05:46,  1.31it/s]
2|6017|Loss: 0.7472041249275208:  93%|█████████▎| 6017/6470 [1:15:11<05:21,  1.41it/s]
2|6018|Loss: 0.9046602845191956:  93%|█████████▎| 6017/6470 [1:15:11<05:21,  1.41it/s]
2|6018|Loss: 0.9046602845191956:  93%|█████████▎| 6018/6470 [1:15:11<05:24,  1.39it/s]
2|6019|Loss: 0.6978873014450073:  93%|█████████▎| 6018/6470 [1:15:12<05:24,  1.39it/s]
2|6019|Loss: 0.6978873014450073:  93%|█████████▎| 6019/6470 [1:15:12<05:50,  1.29it/s]
2|6020|Loss: 0.8292160034179688:  93%|█████████▎| 6019/6470 [1:15:13<05:50,  1.29it/s]
2|6020|Loss: 0.8292160034179688:  93%|█████████▎| 6020/6470 [1:15:13<05:46,  1.30it/s]
2|6021|Loss: 0.6388031840324402:  93%|█████████▎| 6020/6470 [1:15:13<05:46,  1.30it/s]
2|6021|Loss: 0.6388031840324402:  93%|█████████▎| 6021/6470 [1:15:14<05:45,  1.30it/s]
2|6022|Loss: 0.8937633633613586:  93%|█████████▎| 6021/6470 [1:15:14<05:45,  1.30it/s]
2|6022|Loss: 0.8937633633613586:  93%|█████████▎| 6022/6470 [1:15:15<05:59,  1.25it/s]
2|6023|Loss: 0.8449427485466003:  93%|█████████▎| 6022/6470 [1:15:15<05:59,  1.25it/s]
2|6023|Loss: 0.8449427485466003:  93%|█████████▎| 6023/6470 [1:15:15<05:40,  1.31it/s]
2|6024|Loss: 0.8121040463447571:  93%|█████████▎| 6023/6470 [1:15:16<05:40,  1.31it/s]
2|6024|Loss: 0.8121040463447571:  93%|█████████▎| 6024/6470 [1:15:16<05:38,  1.32it/s]
2|6025|Loss: 0.9745876789093018:  93%|█████████▎| 6024/6470 [1:15:16<05:38,  1.32it/s]
2|6025|Loss: 0.9745876789093018:  93%|█████████▎| 6025/6470 [1:15:17<05:20,  1.39it/s]
2|6026|Loss: 1.0049262046813965:  93%|█████████▎| 6025/6470 [1:15:17<05:20,  1.39it/s]
2|6026|Loss: 1.0049262046813965:  93%|█████████▎| 6026/6470 [1:15:17<05:18,  1.39it/s]
2|6027|Loss: 0.9567031264305115:  93%|█████████▎| 6026/6470 [1:15:18<05:18,  1.39it/s]
2|6027|Loss: 0.9567031264305115:  93%|█████████▎| 6027/6470 [1:15:18<05:09,  1.43it/s]
2|6028|Loss: 0.8767266869544983:  93%|█████████▎| 6027/6470 [1:15:18<05:09,  1.43it/s]
2|6028|Loss: 0.8767266869544983:  93%|█████████▎| 6028/6470 [1:15:19<05:01,  1.47it/s]
2|6029|Loss: 0.9054641127586365:  93%|█████████▎| 6028/6470 [1:15:19<05:01,  1.47it/s]
2|6029|Loss: 0.9054641127586365:  93%|█████████▎| 6029/6470 [1:15:19<04:53,  1.51it/s]
2|6030|Loss: 0.8836559653282166:  93%|█████████▎| 6029/6470 [1:15:20<04:53,  1.51it/s]
2|6030|Loss: 0.8836559653282166:  93%|█████████▎| 6030/6470 [1:15:20<04:51,  1.51it/s]
2|6031|Loss: 0.8973126411437988:  93%|█████████▎| 6030/6470 [1:15:20<04:51,  1.51it/s]
2|6031|Loss: 0.8973126411437988:  93%|█████████▎| 6031/6470 [1:15:21<05:16,  1.39it/s]
2|6032|Loss: 0.965685248374939:  93%|█████████▎| 6031/6470 [1:15:21<05:16,  1.39it/s] 
2|6032|Loss: 0.965685248374939:  93%|█████████▎| 6032/6470 [1:15:22<05:44,  1.27it/s]
2|6033|Loss: 0.9103654026985168:  93%|█████████▎| 6032/6470 [1:15:22<05:44,  1.27it/s]
2|6033|Loss: 0.9103654026985168:  93%|█████████▎| 6033/6470 [1:15:23<05:51,  1.24it/s]
2|6034|Loss: 0.6997016072273254:  93%|█████████▎| 6033/6470 [1:15:23<05:51,  1.24it/s]
2|6034|Loss: 0.6997016072273254:  93%|█████████▎| 6034/6470 [1:15:24<05:59,  1.21it/s]
2|6035|Loss: 0.8074151277542114:  93%|█████████▎| 6034/6470 [1:15:24<05:59,  1.21it/s]
2|6035|Loss: 0.8074151277542114:  93%|█████████▎| 6035/6470 [1:15:24<06:05,  1.19it/s]
2|6036|Loss: 0.8193570375442505:  93%|█████████▎| 6035/6470 [1:15:25<06:05,  1.19it/s]
2|6036|Loss: 0.8193570375442505:  93%|█████████▎| 6036/6470 [1:15:25<05:51,  1.24it/s]
2|6037|Loss: 0.8603977560997009:  93%|█████████▎| 6036/6470 [1:15:26<05:51,  1.24it/s]
2|6037|Loss: 0.8603977560997009:  93%|█████████▎| 6037/6470 [1:15:26<05:37,  1.28it/s]
2|6038|Loss: 0.7509626746177673:  93%|█████████▎| 6037/6470 [1:15:26<05:37,  1.28it/s]
2|6038|Loss: 0.7509626746177673:  93%|█████████▎| 6038/6470 [1:15:27<05:54,  1.22it/s]
2|6039|Loss: 1.0438346862792969:  93%|█████████▎| 6038/6470 [1:15:27<05:54,  1.22it/s]
2|6039|Loss: 1.0438346862792969:  93%|█████████▎| 6039/6470 [1:15:28<05:45,  1.25it/s]
2|6040|Loss: 0.9196652770042419:  93%|█████████▎| 6039/6470 [1:15:28<05:45,  1.25it/s]
2|6040|Loss: 0.9196652770042419:  93%|█████████▎| 6040/6470 [1:15:28<05:55,  1.21it/s]
2|6041|Loss: 0.9132753610610962:  93%|█████████▎| 6040/6470 [1:15:29<05:55,  1.21it/s]
2|6041|Loss: 0.9132753610610962:  93%|█████████▎| 6041/6470 [1:15:29<05:56,  1.20it/s]
2|6042|Loss: 0.914840042591095:  93%|█████████▎| 6041/6470 [1:15:30<05:56,  1.20it/s] 
2|6042|Loss: 0.914840042591095:  93%|█████████▎| 6042/6470 [1:15:30<06:09,  1.16it/s]
2|6043|Loss: 0.8398762345314026:  93%|█████████▎| 6042/6470 [1:15:31<06:09,  1.16it/s]
2|6043|Loss: 0.8398762345314026:  93%|█████████▎| 6043/6470 [1:15:31<06:06,  1.17it/s]
2|6044|Loss: 0.9890137910842896:  93%|█████████▎| 6043/6470 [1:15:31<06:06,  1.17it/s]
2|6044|Loss: 0.9890137910842896:  93%|█████████▎| 6044/6470 [1:15:32<06:00,  1.18it/s]
2|6045|Loss: 0.8395211100578308:  93%|█████████▎| 6044/6470 [1:15:32<06:00,  1.18it/s]
2|6045|Loss: 0.8395211100578308:  93%|█████████▎| 6045/6470 [1:15:33<05:50,  1.21it/s]
2|6046|Loss: 0.8330639600753784:  93%|█████████▎| 6045/6470 [1:15:33<05:50,  1.21it/s]
2|6046|Loss: 0.8330639600753784:  93%|█████████▎| 6046/6470 [1:15:33<05:45,  1.23it/s]
2|6047|Loss: 0.8281650543212891:  93%|█████████▎| 6046/6470 [1:15:34<05:45,  1.23it/s]
2|6047|Loss: 0.8281650543212891:  93%|█████████▎| 6047/6470 [1:15:34<05:51,  1.20it/s]
2|6048|Loss: 0.7694194316864014:  93%|█████████▎| 6047/6470 [1:15:35<05:51,  1.20it/s]
2|6048|Loss: 0.7694194316864014:  93%|█████████▎| 6048/6470 [1:15:35<05:48,  1.21it/s]
2|6049|Loss: 0.7699518203735352:  93%|█████████▎| 6048/6470 [1:15:36<05:48,  1.21it/s]
2|6049|Loss: 0.7699518203735352:  93%|█████████▎| 6049/6470 [1:15:36<05:43,  1.22it/s]
2|6050|Loss: 0.7517205476760864:  93%|█████████▎| 6049/6470 [1:15:36<05:43,  1.22it/s]
2|6050|Loss: 0.7517205476760864:  94%|█████████▎| 6050/6470 [1:15:37<05:38,  1.24it/s]
2|6051|Loss: 0.6589460968971252:  94%|█████████▎| 6050/6470 [1:15:37<05:38,  1.24it/s]
2|6051|Loss: 0.6589460968971252:  94%|█████████▎| 6051/6470 [1:15:37<05:35,  1.25it/s]
2|6052|Loss: 0.9587548971176147:  94%|█████████▎| 6051/6470 [1:15:38<05:35,  1.25it/s]
2|6052|Loss: 0.9587548971176147:  94%|█████████▎| 6052/6470 [1:15:38<05:14,  1.33it/s]
2|6053|Loss: 0.8134194612503052:  94%|█████████▎| 6052/6470 [1:15:38<05:14,  1.33it/s]
2|6053|Loss: 0.8134194612503052:  94%|█████████▎| 6053/6470 [1:15:39<05:14,  1.33it/s]
2|6054|Loss: 0.8208848834037781:  94%|█████████▎| 6053/6470 [1:15:39<05:14,  1.33it/s]
2|6054|Loss: 0.8208848834037781:  94%|█████████▎| 6054/6470 [1:15:40<05:03,  1.37it/s]
2|6055|Loss: 0.9368044137954712:  94%|█████████▎| 6054/6470 [1:15:40<05:03,  1.37it/s]
2|6055|Loss: 0.9368044137954712:  94%|█████████▎| 6055/6470 [1:15:40<04:49,  1.43it/s]
2|6056|Loss: 0.7908058166503906:  94%|█████████▎| 6055/6470 [1:15:41<04:49,  1.43it/s]
2|6056|Loss: 0.7908058166503906:  94%|█████████▎| 6056/6470 [1:15:41<04:52,  1.42it/s]
2|6057|Loss: 0.7438182830810547:  94%|█████████▎| 6056/6470 [1:15:41<04:52,  1.42it/s]
2|6057|Loss: 0.7438182830810547:  94%|█████████▎| 6057/6470 [1:15:42<05:02,  1.36it/s]
2|6058|Loss: 0.8660224676132202:  94%|█████████▎| 6057/6470 [1:15:42<05:02,  1.36it/s]
2|6058|Loss: 0.8660224676132202:  94%|█████████▎| 6058/6470 [1:15:42<05:10,  1.33it/s]
2|6059|Loss: 0.8317694664001465:  94%|█████████▎| 6058/6470 [1:15:43<05:10,  1.33it/s]
2|6059|Loss: 0.8317694664001465:  94%|█████████▎| 6059/6470 [1:15:43<05:04,  1.35it/s]
2|6060|Loss: 0.8785896301269531:  94%|█████████▎| 6059/6470 [1:15:44<05:04,  1.35it/s]
2|6060|Loss: 0.8785896301269531:  94%|█████████▎| 6060/6470 [1:15:44<04:36,  1.48it/s]
2|6061|Loss: 0.8282702565193176:  94%|█████████▎| 6060/6470 [1:15:44<04:36,  1.48it/s]
2|6061|Loss: 0.8282702565193176:  94%|█████████▎| 6061/6470 [1:15:44<04:40,  1.46it/s]
2|6062|Loss: 0.759445071220398:  94%|█████████▎| 6061/6470 [1:15:45<04:40,  1.46it/s] 
2|6062|Loss: 0.759445071220398:  94%|█████████▎| 6062/6470 [1:15:45<04:41,  1.45it/s]
2|6063|Loss: 0.858575701713562:  94%|█████████▎| 6062/6470 [1:15:46<04:41,  1.45it/s]
2|6063|Loss: 0.858575701713562:  94%|█████████▎| 6063/6470 [1:15:46<05:01,  1.35it/s]
2|6064|Loss: 0.7472410798072815:  94%|█████████▎| 6063/6470 [1:15:46<05:01,  1.35it/s]
2|6064|Loss: 0.7472410798072815:  94%|█████████▎| 6064/6470 [1:15:47<05:08,  1.32it/s]
2|6065|Loss: 0.8106821775436401:  94%|█████████▎| 6064/6470 [1:15:47<05:08,  1.32it/s]
2|6065|Loss: 0.8106821775436401:  94%|█████████▎| 6065/6470 [1:15:47<04:59,  1.35it/s]
2|6066|Loss: 0.8745679259300232:  94%|█████████▎| 6065/6470 [1:15:48<04:59,  1.35it/s]
2|6066|Loss: 0.8745679259300232:  94%|█████████▍| 6066/6470 [1:15:48<04:53,  1.38it/s]
2|6067|Loss: 0.7396612763404846:  94%|█████████▍| 6066/6470 [1:15:48<04:53,  1.38it/s]
2|6067|Loss: 0.7396612763404846:  94%|█████████▍| 6067/6470 [1:15:49<04:21,  1.54it/s]
2|6068|Loss: 0.8446371555328369:  94%|█████████▍| 6067/6470 [1:15:49<04:21,  1.54it/s]
2|6068|Loss: 0.8446371555328369:  94%|█████████▍| 6068/6470 [1:15:49<04:17,  1.56it/s]
2|6069|Loss: 0.7828503251075745:  94%|█████████▍| 6068/6470 [1:15:50<04:17,  1.56it/s]
2|6069|Loss: 0.7828503251075745:  94%|█████████▍| 6069/6470 [1:15:50<04:41,  1.42it/s]
2|6070|Loss: 0.9533413052558899:  94%|█████████▍| 6069/6470 [1:15:51<04:41,  1.42it/s]
2|6070|Loss: 0.9533413052558899:  94%|█████████▍| 6070/6470 [1:15:51<05:06,  1.30it/s]
2|6071|Loss: 0.7859795093536377:  94%|█████████▍| 6070/6470 [1:15:51<05:06,  1.30it/s]
2|6071|Loss: 0.7859795093536377:  94%|█████████▍| 6071/6470 [1:15:52<05:14,  1.27it/s]
2|6072|Loss: 0.96531081199646:  94%|█████████▍| 6071/6470 [1:15:52<05:14,  1.27it/s]  
2|6072|Loss: 0.96531081199646:  94%|█████████▍| 6072/6470 [1:15:53<05:16,  1.26it/s]
2|6073|Loss: 0.7943581938743591:  94%|█████████▍| 6072/6470 [1:15:53<05:16,  1.26it/s]
2|6073|Loss: 0.7943581938743591:  94%|█████████▍| 6073/6470 [1:15:54<05:33,  1.19it/s]
2|6074|Loss: 0.7595176696777344:  94%|█████████▍| 6073/6470 [1:15:54<05:33,  1.19it/s]
2|6074|Loss: 0.7595176696777344:  94%|█████████▍| 6074/6470 [1:15:55<05:45,  1.15it/s]
2|6075|Loss: 0.8732337355613708:  94%|█████████▍| 6074/6470 [1:15:55<05:45,  1.15it/s]
2|6075|Loss: 0.8732337355613708:  94%|█████████▍| 6075/6470 [1:15:55<05:49,  1.13it/s]
2|6076|Loss: 0.8788880705833435:  94%|█████████▍| 6075/6470 [1:15:56<05:49,  1.13it/s]
2|6076|Loss: 0.8788880705833435:  94%|█████████▍| 6076/6470 [1:15:56<05:40,  1.16it/s]
2|6077|Loss: 1.0081735849380493:  94%|█████████▍| 6076/6470 [1:15:57<05:40,  1.16it/s]
2|6077|Loss: 1.0081735849380493:  94%|█████████▍| 6077/6470 [1:15:57<05:19,  1.23it/s]
2|6078|Loss: 0.8976346254348755:  94%|█████████▍| 6077/6470 [1:15:57<05:19,  1.23it/s]
2|6078|Loss: 0.8976346254348755:  94%|█████████▍| 6078/6470 [1:15:58<05:01,  1.30it/s]
2|6079|Loss: 0.8352831602096558:  94%|█████████▍| 6078/6470 [1:15:58<05:01,  1.30it/s]
2|6079|Loss: 0.8352831602096558:  94%|█████████▍| 6079/6470 [1:15:58<04:59,  1.31it/s]
2|6080|Loss: 0.6627835631370544:  94%|█████████▍| 6079/6470 [1:15:59<04:59,  1.31it/s]
2|6080|Loss: 0.6627835631370544:  94%|█████████▍| 6080/6470 [1:15:59<04:50,  1.34it/s]
2|6081|Loss: 0.8537116646766663:  94%|█████████▍| 6080/6470 [1:15:59<04:50,  1.34it/s]
2|6081|Loss: 0.8537116646766663:  94%|█████████▍| 6081/6470 [1:16:00<04:44,  1.37it/s]
2|6082|Loss: 0.7696976065635681:  94%|█████████▍| 6081/6470 [1:16:00<04:44,  1.37it/s]
2|6082|Loss: 0.7696976065635681:  94%|█████████▍| 6082/6470 [1:16:00<04:22,  1.48it/s]
2|6083|Loss: 0.6686273217201233:  94%|█████████▍| 6082/6470 [1:16:01<04:22,  1.48it/s]
2|6083|Loss: 0.6686273217201233:  94%|█████████▍| 6083/6470 [1:16:01<03:50,  1.68it/s]
2|6084|Loss: 0.876469612121582:  94%|█████████▍| 6083/6470 [1:16:01<03:50,  1.68it/s] 
2|6084|Loss: 0.876469612121582:  94%|█████████▍| 6084/6470 [1:16:01<03:52,  1.66it/s]
2|6085|Loss: 0.9101213812828064:  94%|█████████▍| 6084/6470 [1:16:02<03:52,  1.66it/s]
2|6085|Loss: 0.9101213812828064:  94%|█████████▍| 6085/6470 [1:16:02<04:09,  1.54it/s]
2|6086|Loss: 0.762830376625061:  94%|█████████▍| 6085/6470 [1:16:03<04:09,  1.54it/s] 
2|6086|Loss: 0.762830376625061:  94%|█████████▍| 6086/6470 [1:16:03<04:10,  1.53it/s]
2|6087|Loss: 0.8591114282608032:  94%|█████████▍| 6086/6470 [1:16:03<04:10,  1.53it/s]
2|6087|Loss: 0.8591114282608032:  94%|█████████▍| 6087/6470 [1:16:03<03:52,  1.65it/s]
2|6088|Loss: 1.006355881690979:  94%|█████████▍| 6087/6470 [1:16:04<03:52,  1.65it/s] 
2|6088|Loss: 1.006355881690979:  94%|█████████▍| 6088/6470 [1:16:04<04:03,  1.57it/s]
2|6089|Loss: 0.8828160166740417:  94%|█████████▍| 6088/6470 [1:16:04<04:03,  1.57it/s]
2|6089|Loss: 0.8828160166740417:  94%|█████████▍| 6089/6470 [1:16:05<04:09,  1.53it/s]
2|6090|Loss: 0.8342831134796143:  94%|█████████▍| 6089/6470 [1:16:05<04:09,  1.53it/s]
2|6090|Loss: 0.8342831134796143:  94%|█████████▍| 6090/6470 [1:16:05<04:17,  1.47it/s]
2|6091|Loss: 0.8576689958572388:  94%|█████████▍| 6090/6470 [1:16:06<04:17,  1.47it/s]
2|6091|Loss: 0.8576689958572388:  94%|█████████▍| 6091/6470 [1:16:06<04:07,  1.53it/s]
2|6092|Loss: 0.7755734920501709:  94%|█████████▍| 6091/6470 [1:16:06<04:07,  1.53it/s]
2|6092|Loss: 0.7755734920501709:  94%|█████████▍| 6092/6470 [1:16:07<04:12,  1.50it/s]
2|6093|Loss: 0.8802267909049988:  94%|█████████▍| 6092/6470 [1:16:07<04:12,  1.50it/s]
2|6093|Loss: 0.8802267909049988:  94%|█████████▍| 6093/6470 [1:16:07<03:56,  1.59it/s]
2|6094|Loss: 1.0160962343215942:  94%|█████████▍| 6093/6470 [1:16:08<03:56,  1.59it/s]
2|6094|Loss: 1.0160962343215942:  94%|█████████▍| 6094/6470 [1:16:08<04:14,  1.48it/s]
2|6095|Loss: 1.0218651294708252:  94%|█████████▍| 6094/6470 [1:16:08<04:14,  1.48it/s]
2|6095|Loss: 1.0218651294708252:  94%|█████████▍| 6095/6470 [1:16:09<04:07,  1.52it/s]
2|6096|Loss: 0.7977163195610046:  94%|█████████▍| 6095/6470 [1:16:09<04:07,  1.52it/s]
2|6096|Loss: 0.7977163195610046:  94%|█████████▍| 6096/6470 [1:16:09<04:12,  1.48it/s]
2|6097|Loss: 0.8116334080696106:  94%|█████████▍| 6096/6470 [1:16:10<04:12,  1.48it/s]
2|6097|Loss: 0.8116334080696106:  94%|█████████▍| 6097/6470 [1:16:10<04:38,  1.34it/s]
2|6098|Loss: 0.9004208445549011:  94%|█████████▍| 6097/6470 [1:16:11<04:38,  1.34it/s]
2|6098|Loss: 0.9004208445549011:  94%|█████████▍| 6098/6470 [1:16:11<04:32,  1.36it/s]
2|6099|Loss: 0.6969773173332214:  94%|█████████▍| 6098/6470 [1:16:11<04:32,  1.36it/s]
2|6099|Loss: 0.6969773173332214:  94%|█████████▍| 6099/6470 [1:16:12<04:31,  1.37it/s]
2|6100|Loss: 0.8223106861114502:  94%|█████████▍| 6099/6470 [1:16:12<04:31,  1.37it/s]
2|6100|Loss: 0.8223106861114502:  94%|█████████▍| 6100/6470 [1:16:12<04:24,  1.40it/s]
2|6101|Loss: 0.7827832698822021:  94%|█████████▍| 6100/6470 [1:16:13<04:24,  1.40it/s]
2|6101|Loss: 0.7827832698822021:  94%|█████████▍| 6101/6470 [1:16:13<04:01,  1.53it/s]
2|6102|Loss: 0.8288995027542114:  94%|█████████▍| 6101/6470 [1:16:13<04:01,  1.53it/s]
2|6102|Loss: 0.8288995027542114:  94%|█████████▍| 6102/6470 [1:16:14<04:03,  1.51it/s]
2|6103|Loss: 0.876775324344635:  94%|█████████▍| 6102/6470 [1:16:14<04:03,  1.51it/s] 
2|6103|Loss: 0.876775324344635:  94%|█████████▍| 6103/6470 [1:16:14<04:17,  1.43it/s]
2|6104|Loss: 0.778960108757019:  94%|█████████▍| 6103/6470 [1:16:15<04:17,  1.43it/s]
2|6104|Loss: 0.778960108757019:  94%|█████████▍| 6104/6470 [1:16:15<04:11,  1.45it/s]
2|6105|Loss: 1.0075677633285522:  94%|█████████▍| 6104/6470 [1:16:16<04:11,  1.45it/s]
2|6105|Loss: 1.0075677633285522:  94%|█████████▍| 6105/6470 [1:16:16<04:38,  1.31it/s]
2|6106|Loss: 0.9254856705665588:  94%|█████████▍| 6105/6470 [1:16:16<04:38,  1.31it/s]
2|6106|Loss: 0.9254856705665588:  94%|█████████▍| 6106/6470 [1:16:17<04:35,  1.32it/s]
2|6107|Loss: 0.8450186848640442:  94%|█████████▍| 6106/6470 [1:16:17<04:35,  1.32it/s]
2|6107|Loss: 0.8450186848640442:  94%|█████████▍| 6107/6470 [1:16:17<04:24,  1.37it/s]
2|6108|Loss: 0.9325916171073914:  94%|█████████▍| 6107/6470 [1:16:18<04:24,  1.37it/s]
2|6108|Loss: 0.9325916171073914:  94%|█████████▍| 6108/6470 [1:16:18<04:27,  1.35it/s]
2|6109|Loss: 0.8846964836120605:  94%|█████████▍| 6108/6470 [1:16:19<04:27,  1.35it/s]
2|6109|Loss: 0.8846964836120605:  94%|█████████▍| 6109/6470 [1:16:19<04:50,  1.24it/s]
2|6110|Loss: 0.8525869250297546:  94%|█████████▍| 6109/6470 [1:16:20<04:50,  1.24it/s]
2|6110|Loss: 0.8525869250297546:  94%|█████████▍| 6110/6470 [1:16:20<05:07,  1.17it/s]
2|6111|Loss: 0.7327501177787781:  94%|█████████▍| 6110/6470 [1:16:21<05:07,  1.17it/s]
2|6111|Loss: 0.7327501177787781:  94%|█████████▍| 6111/6470 [1:16:21<05:18,  1.13it/s]
2|6112|Loss: 0.9747512340545654:  94%|█████████▍| 6111/6470 [1:16:22<05:18,  1.13it/s]
2|6112|Loss: 0.9747512340545654:  94%|█████████▍| 6112/6470 [1:16:22<05:11,  1.15it/s]
2|6113|Loss: 0.7968283891677856:  94%|█████████▍| 6112/6470 [1:16:22<05:11,  1.15it/s]
2|6113|Loss: 0.7968283891677856:  94%|█████████▍| 6113/6470 [1:16:23<05:12,  1.14it/s]
2|6114|Loss: 0.9214198589324951:  94%|█████████▍| 6113/6470 [1:16:23<05:12,  1.14it/s]
2|6114|Loss: 0.9214198589324951:  94%|█████████▍| 6114/6470 [1:16:24<04:58,  1.19it/s]
2|6115|Loss: 0.9077476859092712:  94%|█████████▍| 6114/6470 [1:16:24<04:58,  1.19it/s]
2|6115|Loss: 0.9077476859092712:  95%|█████████▍| 6115/6470 [1:16:24<04:49,  1.23it/s]
2|6116|Loss: 0.8211532235145569:  95%|█████████▍| 6115/6470 [1:16:25<04:49,  1.23it/s]
2|6116|Loss: 0.8211532235145569:  95%|█████████▍| 6116/6470 [1:16:25<04:28,  1.32it/s]
2|6117|Loss: 0.8130179047584534:  95%|█████████▍| 6116/6470 [1:16:25<04:28,  1.32it/s]
2|6117|Loss: 0.8130179047584534:  95%|█████████▍| 6117/6470 [1:16:26<04:27,  1.32it/s]
2|6118|Loss: 1.0164930820465088:  95%|█████████▍| 6117/6470 [1:16:26<04:27,  1.32it/s]
2|6118|Loss: 1.0164930820465088:  95%|█████████▍| 6118/6470 [1:16:27<04:40,  1.26it/s]
2|6119|Loss: 0.8237599730491638:  95%|█████████▍| 6118/6470 [1:16:27<04:40,  1.26it/s]
2|6119|Loss: 0.8237599730491638:  95%|█████████▍| 6119/6470 [1:16:27<04:20,  1.35it/s]
2|6120|Loss: 0.9116886854171753:  95%|█████████▍| 6119/6470 [1:16:27<04:20,  1.35it/s]
2|6120|Loss: 0.9116886854171753:  95%|█████████▍| 6120/6470 [1:16:28<04:00,  1.46it/s]
2|6121|Loss: 0.8301316499710083:  95%|█████████▍| 6120/6470 [1:16:28<04:00,  1.46it/s]
2|6121|Loss: 0.8301316499710083:  95%|█████████▍| 6121/6470 [1:16:28<04:07,  1.41it/s]
2|6122|Loss: 0.9507659077644348:  95%|█████████▍| 6121/6470 [1:16:29<04:07,  1.41it/s]
2|6122|Loss: 0.9507659077644348:  95%|█████████▍| 6122/6470 [1:16:29<04:07,  1.40it/s]
2|6123|Loss: 0.908681333065033:  95%|█████████▍| 6122/6470 [1:16:30<04:07,  1.40it/s] 
2|6123|Loss: 0.908681333065033:  95%|█████████▍| 6123/6470 [1:16:30<04:21,  1.33it/s]
2|6124|Loss: 0.8897711038589478:  95%|█████████▍| 6123/6470 [1:16:30<04:21,  1.33it/s]
2|6124|Loss: 0.8897711038589478:  95%|█████████▍| 6124/6470 [1:16:31<04:24,  1.31it/s]
2|6125|Loss: 0.7916337251663208:  95%|█████████▍| 6124/6470 [1:16:31<04:24,  1.31it/s]
2|6125|Loss: 0.7916337251663208:  95%|█████████▍| 6125/6470 [1:16:31<03:59,  1.44it/s]
2|6126|Loss: 0.9755436778068542:  95%|█████████▍| 6125/6470 [1:16:32<03:59,  1.44it/s]
2|6126|Loss: 0.9755436778068542:  95%|█████████▍| 6126/6470 [1:16:32<03:50,  1.49it/s]
2|6127|Loss: 0.8324939012527466:  95%|█████████▍| 6126/6470 [1:16:32<03:50,  1.49it/s]
2|6127|Loss: 0.8324939012527466:  95%|█████████▍| 6127/6470 [1:16:33<03:51,  1.48it/s]
2|6128|Loss: 0.840308666229248:  95%|█████████▍| 6127/6470 [1:16:33<03:51,  1.48it/s] 
2|6128|Loss: 0.840308666229248:  95%|█████████▍| 6128/6470 [1:16:33<04:06,  1.39it/s]
2|6129|Loss: 0.9374040961265564:  95%|█████████▍| 6128/6470 [1:16:34<04:06,  1.39it/s]
2|6129|Loss: 0.9374040961265564:  95%|█████████▍| 6129/6470 [1:16:34<04:16,  1.33it/s]
2|6130|Loss: 0.8191219568252563:  95%|█████████▍| 6129/6470 [1:16:35<04:16,  1.33it/s]
2|6130|Loss: 0.8191219568252563:  95%|█████████▍| 6130/6470 [1:16:35<04:17,  1.32it/s]
2|6131|Loss: 0.9340541958808899:  95%|█████████▍| 6130/6470 [1:16:36<04:17,  1.32it/s]
2|6131|Loss: 0.9340541958808899:  95%|█████████▍| 6131/6470 [1:16:36<04:25,  1.27it/s]
2|6132|Loss: 0.7695348262786865:  95%|█████████▍| 6131/6470 [1:16:36<04:25,  1.27it/s]
2|6132|Loss: 0.7695348262786865:  95%|█████████▍| 6132/6470 [1:16:37<04:27,  1.26it/s]
2|6133|Loss: 0.8362240195274353:  95%|█████████▍| 6132/6470 [1:16:37<04:27,  1.26it/s]
2|6133|Loss: 0.8362240195274353:  95%|█████████▍| 6133/6470 [1:16:37<04:18,  1.30it/s]
2|6134|Loss: 0.7737860083580017:  95%|█████████▍| 6133/6470 [1:16:38<04:18,  1.30it/s]
2|6134|Loss: 0.7737860083580017:  95%|█████████▍| 6134/6470 [1:16:38<04:17,  1.30it/s]
2|6135|Loss: 0.7079472541809082:  95%|█████████▍| 6134/6470 [1:16:39<04:17,  1.30it/s]
2|6135|Loss: 0.7079472541809082:  95%|█████████▍| 6135/6470 [1:16:39<04:14,  1.32it/s]
2|6136|Loss: 0.7881218194961548:  95%|█████████▍| 6135/6470 [1:16:39<04:14,  1.32it/s]
2|6136|Loss: 0.7881218194961548:  95%|█████████▍| 6136/6470 [1:16:40<04:05,  1.36it/s]
2|6137|Loss: 0.873845100402832:  95%|█████████▍| 6136/6470 [1:16:40<04:05,  1.36it/s] 
2|6137|Loss: 0.873845100402832:  95%|█████████▍| 6137/6470 [1:16:40<03:48,  1.46it/s]
2|6138|Loss: 0.8749209046363831:  95%|█████████▍| 6137/6470 [1:16:41<03:48,  1.46it/s]
2|6138|Loss: 0.8749209046363831:  95%|█████████▍| 6138/6470 [1:16:41<03:33,  1.56it/s]
2|6139|Loss: 0.8027635812759399:  95%|█████████▍| 6138/6470 [1:16:41<03:33,  1.56it/s]
2|6139|Loss: 0.8027635812759399:  95%|█████████▍| 6139/6470 [1:16:42<03:50,  1.44it/s]
2|6140|Loss: 0.8598976731300354:  95%|█████████▍| 6139/6470 [1:16:42<03:50,  1.44it/s]
2|6140|Loss: 0.8598976731300354:  95%|█████████▍| 6140/6470 [1:16:42<03:52,  1.42it/s]
2|6141|Loss: 0.8049206733703613:  95%|█████████▍| 6140/6470 [1:16:43<03:52,  1.42it/s]
2|6141|Loss: 0.8049206733703613:  95%|█████████▍| 6141/6470 [1:16:43<03:55,  1.39it/s]
2|6142|Loss: 0.7436179518699646:  95%|█████████▍| 6141/6470 [1:16:43<03:55,  1.39it/s]
2|6142|Loss: 0.7436179518699646:  95%|█████████▍| 6142/6470 [1:16:44<03:49,  1.43it/s]
2|6143|Loss: 0.9503616690635681:  95%|█████████▍| 6142/6470 [1:16:44<03:49,  1.43it/s]
2|6143|Loss: 0.9503616690635681:  95%|█████████▍| 6143/6470 [1:16:45<04:00,  1.36it/s]
2|6144|Loss: 0.908202052116394:  95%|█████████▍| 6143/6470 [1:16:45<04:00,  1.36it/s] 
2|6144|Loss: 0.908202052116394:  95%|█████████▍| 6144/6470 [1:16:45<04:07,  1.32it/s]
2|6145|Loss: 0.7772902846336365:  95%|█████████▍| 6144/6470 [1:16:46<04:07,  1.32it/s]
2|6145|Loss: 0.7772902846336365:  95%|█████████▍| 6145/6470 [1:16:46<04:17,  1.26it/s]
2|6146|Loss: 0.6956098079681396:  95%|█████████▍| 6145/6470 [1:16:47<04:17,  1.26it/s]
2|6146|Loss: 0.6956098079681396:  95%|█████████▍| 6146/6470 [1:16:47<03:53,  1.39it/s]
2|6147|Loss: 0.923202633857727:  95%|█████████▍| 6146/6470 [1:16:47<03:53,  1.39it/s] 
2|6147|Loss: 0.923202633857727:  95%|█████████▌| 6147/6470 [1:16:48<03:57,  1.36it/s]
2|6148|Loss: 0.8662874698638916:  95%|█████████▌| 6147/6470 [1:16:48<03:57,  1.36it/s]
2|6148|Loss: 0.8662874698638916:  95%|█████████▌| 6148/6470 [1:16:48<04:02,  1.33it/s]
2|6149|Loss: 0.7853885889053345:  95%|█████████▌| 6148/6470 [1:16:49<04:02,  1.33it/s]
2|6149|Loss: 0.7853885889053345:  95%|█████████▌| 6149/6470 [1:16:49<04:05,  1.31it/s]
2|6150|Loss: 0.8431406617164612:  95%|█████████▌| 6149/6470 [1:16:50<04:05,  1.31it/s]
2|6150|Loss: 0.8431406617164612:  95%|█████████▌| 6150/6470 [1:16:50<03:54,  1.36it/s]
2|6151|Loss: 0.8958324790000916:  95%|█████████▌| 6150/6470 [1:16:50<03:54,  1.36it/s]
2|6151|Loss: 0.8958324790000916:  95%|█████████▌| 6151/6470 [1:16:50<03:48,  1.39it/s]
2|6152|Loss: 0.8254902958869934:  95%|█████████▌| 6151/6470 [1:16:51<03:48,  1.39it/s]
2|6152|Loss: 0.8254902958869934:  95%|█████████▌| 6152/6470 [1:16:51<03:32,  1.50it/s]
2|6153|Loss: 0.8854473233222961:  95%|█████████▌| 6152/6470 [1:16:51<03:32,  1.50it/s]
2|6153|Loss: 0.8854473233222961:  95%|█████████▌| 6153/6470 [1:16:52<03:37,  1.46it/s]
2|6154|Loss: 0.9136414527893066:  95%|█████████▌| 6153/6470 [1:16:52<03:37,  1.46it/s]
2|6154|Loss: 0.9136414527893066:  95%|█████████▌| 6154/6470 [1:16:53<04:00,  1.32it/s]
2|6155|Loss: 0.7661598920822144:  95%|█████████▌| 6154/6470 [1:16:53<04:00,  1.32it/s]
2|6155|Loss: 0.7661598920822144:  95%|█████████▌| 6155/6470 [1:16:53<03:57,  1.33it/s]
2|6156|Loss: 0.7648563981056213:  95%|█████████▌| 6155/6470 [1:16:54<03:57,  1.33it/s]
2|6156|Loss: 0.7648563981056213:  95%|█████████▌| 6156/6470 [1:16:54<03:57,  1.32it/s]
2|6157|Loss: 0.8097273707389832:  95%|█████████▌| 6156/6470 [1:16:55<03:57,  1.32it/s]
2|6157|Loss: 0.8097273707389832:  95%|█████████▌| 6157/6470 [1:16:55<03:53,  1.34it/s]
2|6158|Loss: 1.0616176128387451:  95%|█████████▌| 6157/6470 [1:16:55<03:53,  1.34it/s]
2|6158|Loss: 1.0616176128387451:  95%|█████████▌| 6158/6470 [1:16:56<04:02,  1.29it/s]
2|6159|Loss: 0.9278553128242493:  95%|█████████▌| 6158/6470 [1:16:56<04:02,  1.29it/s]
2|6159|Loss: 0.9278553128242493:  95%|█████████▌| 6159/6470 [1:16:57<04:15,  1.22it/s]
2|6160|Loss: 0.8572126626968384:  95%|█████████▌| 6159/6470 [1:16:57<04:15,  1.22it/s]
2|6160|Loss: 0.8572126626968384:  95%|█████████▌| 6160/6470 [1:16:58<04:26,  1.16it/s]
2|6161|Loss: 0.8325616717338562:  95%|█████████▌| 6160/6470 [1:16:58<04:26,  1.16it/s]
2|6161|Loss: 0.8325616717338562:  95%|█████████▌| 6161/6470 [1:16:58<04:23,  1.17it/s]
2|6162|Loss: 0.9585787653923035:  95%|█████████▌| 6161/6470 [1:16:59<04:23,  1.17it/s]
2|6162|Loss: 0.9585787653923035:  95%|█████████▌| 6162/6470 [1:16:59<03:58,  1.29it/s]
2|6163|Loss: 0.6969337463378906:  95%|█████████▌| 6162/6470 [1:16:59<03:58,  1.29it/s]
2|6163|Loss: 0.6969337463378906:  95%|█████████▌| 6163/6470 [1:17:00<03:33,  1.44it/s]
2|6164|Loss: 0.7143028378486633:  95%|█████████▌| 6163/6470 [1:17:00<03:33,  1.44it/s]
2|6164|Loss: 0.7143028378486633:  95%|█████████▌| 6164/6470 [1:17:00<03:43,  1.37it/s]
2|6165|Loss: 0.7393108606338501:  95%|█████████▌| 6164/6470 [1:17:01<03:43,  1.37it/s]
2|6165|Loss: 0.7393108606338501:  95%|█████████▌| 6165/6470 [1:17:01<04:11,  1.21it/s]
2|6166|Loss: 0.9957765340805054:  95%|█████████▌| 6165/6470 [1:17:02<04:11,  1.21it/s]
2|6166|Loss: 0.9957765340805054:  95%|█████████▌| 6166/6470 [1:17:02<04:21,  1.16it/s]
2|6167|Loss: 0.838729739189148:  95%|█████████▌| 6166/6470 [1:17:03<04:21,  1.16it/s] 
2|6167|Loss: 0.838729739189148:  95%|█████████▌| 6167/6470 [1:17:03<03:58,  1.27it/s]
2|6168|Loss: 0.9163236021995544:  95%|█████████▌| 6167/6470 [1:17:03<03:58,  1.27it/s]
2|6168|Loss: 0.9163236021995544:  95%|█████████▌| 6168/6470 [1:17:04<03:52,  1.30it/s]
2|6169|Loss: 0.9659079909324646:  95%|█████████▌| 6168/6470 [1:17:04<03:52,  1.30it/s]
2|6169|Loss: 0.9659079909324646:  95%|█████████▌| 6169/6470 [1:17:04<03:50,  1.31it/s]
2|6170|Loss: 0.7843502759933472:  95%|█████████▌| 6169/6470 [1:17:05<03:50,  1.31it/s]
2|6170|Loss: 0.7843502759933472:  95%|█████████▌| 6170/6470 [1:17:05<03:50,  1.30it/s]
2|6171|Loss: 0.7921850085258484:  95%|█████████▌| 6170/6470 [1:17:06<03:50,  1.30it/s]
2|6171|Loss: 0.7921850085258484:  95%|█████████▌| 6171/6470 [1:17:06<03:51,  1.29it/s]
2|6172|Loss: 0.9483881592750549:  95%|█████████▌| 6171/6470 [1:17:06<03:51,  1.29it/s]
2|6172|Loss: 0.9483881592750549:  95%|█████████▌| 6172/6470 [1:17:07<03:39,  1.36it/s]
2|6173|Loss: 0.9948827028274536:  95%|█████████▌| 6172/6470 [1:17:07<03:39,  1.36it/s]
2|6173|Loss: 0.9948827028274536:  95%|█████████▌| 6173/6470 [1:17:07<03:41,  1.34it/s]
2|6174|Loss: 0.7375777363777161:  95%|█████████▌| 6173/6470 [1:17:08<03:41,  1.34it/s]
2|6174|Loss: 0.7375777363777161:  95%|█████████▌| 6174/6470 [1:17:08<03:44,  1.32it/s]
2|6175|Loss: 0.7519148588180542:  95%|█████████▌| 6174/6470 [1:17:09<03:44,  1.32it/s]
2|6175|Loss: 0.7519148588180542:  95%|█████████▌| 6175/6470 [1:17:09<03:51,  1.27it/s]
2|6176|Loss: 0.891411304473877:  95%|█████████▌| 6175/6470 [1:17:10<03:51,  1.27it/s] 
2|6176|Loss: 0.891411304473877:  95%|█████████▌| 6176/6470 [1:17:10<03:56,  1.24it/s]
2|6177|Loss: 0.7166056632995605:  95%|█████████▌| 6176/6470 [1:17:10<03:56,  1.24it/s]
2|6177|Loss: 0.7166056632995605:  95%|█████████▌| 6177/6470 [1:17:11<03:42,  1.32it/s]
2|6178|Loss: 0.924837052822113:  95%|█████████▌| 6177/6470 [1:17:11<03:42,  1.32it/s] 
2|6178|Loss: 0.924837052822113:  95%|█████████▌| 6178/6470 [1:17:11<03:40,  1.32it/s]
2|6179|Loss: 0.9397518634796143:  95%|█████████▌| 6178/6470 [1:17:12<03:40,  1.32it/s]
2|6179|Loss: 0.9397518634796143:  96%|█████████▌| 6179/6470 [1:17:12<03:40,  1.32it/s]
2|6180|Loss: 0.8497915863990784:  96%|█████████▌| 6179/6470 [1:17:13<03:40,  1.32it/s]
2|6180|Loss: 0.8497915863990784:  96%|█████████▌| 6180/6470 [1:17:13<03:56,  1.23it/s]
2|6181|Loss: 1.0667753219604492:  96%|█████████▌| 6180/6470 [1:17:14<03:56,  1.23it/s]
2|6181|Loss: 1.0667753219604492:  96%|█████████▌| 6181/6470 [1:17:14<04:08,  1.16it/s]
2|6182|Loss: 0.7701152563095093:  96%|█████████▌| 6181/6470 [1:17:14<04:08,  1.16it/s]
2|6182|Loss: 0.7701152563095093:  96%|█████████▌| 6182/6470 [1:17:15<03:58,  1.21it/s]
2|6183|Loss: 0.8979134559631348:  96%|█████████▌| 6182/6470 [1:17:15<03:58,  1.21it/s]
2|6183|Loss: 0.8979134559631348:  96%|█████████▌| 6183/6470 [1:17:15<03:46,  1.27it/s]
2|6184|Loss: 0.7657493948936462:  96%|█████████▌| 6183/6470 [1:17:16<03:46,  1.27it/s]
2|6184|Loss: 0.7657493948936462:  96%|█████████▌| 6184/6470 [1:17:16<03:27,  1.38it/s]
2|6185|Loss: 0.8643670082092285:  96%|█████████▌| 6184/6470 [1:17:16<03:27,  1.38it/s]
2|6185|Loss: 0.8643670082092285:  96%|█████████▌| 6185/6470 [1:17:17<03:32,  1.34it/s]
2|6186|Loss: 0.8048757910728455:  96%|█████████▌| 6185/6470 [1:17:17<03:32,  1.34it/s]
2|6186|Loss: 0.8048757910728455:  96%|█████████▌| 6186/6470 [1:17:18<03:28,  1.36it/s]
2|6187|Loss: 0.9469299912452698:  96%|█████████▌| 6186/6470 [1:17:18<03:28,  1.36it/s]
2|6187|Loss: 0.9469299912452698:  96%|█████████▌| 6187/6470 [1:17:18<03:31,  1.34it/s]
2|6188|Loss: 0.7149237990379333:  96%|█████████▌| 6187/6470 [1:17:19<03:31,  1.34it/s]
2|6188|Loss: 0.7149237990379333:  96%|█████████▌| 6188/6470 [1:17:19<03:13,  1.46it/s]
2|6189|Loss: 0.810684323310852:  96%|█████████▌| 6188/6470 [1:17:19<03:13,  1.46it/s] 
2|6189|Loss: 0.810684323310852:  96%|█████████▌| 6189/6470 [1:17:19<03:09,  1.49it/s]
2|6190|Loss: 0.7229989767074585:  96%|█████████▌| 6189/6470 [1:17:20<03:09,  1.49it/s]
2|6190|Loss: 0.7229989767074585:  96%|█████████▌| 6190/6470 [1:17:20<03:21,  1.39it/s]
2|6191|Loss: 0.8603703379631042:  96%|█████████▌| 6190/6470 [1:17:21<03:21,  1.39it/s]
2|6191|Loss: 0.8603703379631042:  96%|█████████▌| 6191/6470 [1:17:21<03:09,  1.47it/s]
2|6192|Loss: 0.8083338737487793:  96%|█████████▌| 6191/6470 [1:17:21<03:09,  1.47it/s]
2|6192|Loss: 0.8083338737487793:  96%|█████████▌| 6192/6470 [1:17:22<03:23,  1.37it/s]
2|6193|Loss: 0.8131487369537354:  96%|█████████▌| 6192/6470 [1:17:22<03:23,  1.37it/s]
2|6193|Loss: 0.8131487369537354:  96%|█████████▌| 6193/6470 [1:17:23<03:35,  1.29it/s]
2|6194|Loss: 0.865461528301239:  96%|█████████▌| 6193/6470 [1:17:23<03:35,  1.29it/s] 
2|6194|Loss: 0.865461528301239:  96%|█████████▌| 6194/6470 [1:17:24<03:42,  1.24it/s]
2|6195|Loss: 0.8219262957572937:  96%|█████████▌| 6194/6470 [1:17:24<03:42,  1.24it/s]
2|6195|Loss: 0.8219262957572937:  96%|█████████▌| 6195/6470 [1:17:24<03:50,  1.20it/s]
2|6196|Loss: 0.7767893671989441:  96%|█████████▌| 6195/6470 [1:17:25<03:50,  1.20it/s]
2|6196|Loss: 0.7767893671989441:  96%|█████████▌| 6196/6470 [1:17:25<03:45,  1.22it/s]
2|6197|Loss: 0.8596034049987793:  96%|█████████▌| 6196/6470 [1:17:26<03:45,  1.22it/s]
2|6197|Loss: 0.8596034049987793:  96%|█████████▌| 6197/6470 [1:17:26<03:33,  1.28it/s]
2|6198|Loss: 0.934063196182251:  96%|█████████▌| 6197/6470 [1:17:26<03:33,  1.28it/s] 
2|6198|Loss: 0.934063196182251:  96%|█████████▌| 6198/6470 [1:17:27<03:22,  1.35it/s]
2|6199|Loss: 0.9812778234481812:  96%|█████████▌| 6198/6470 [1:17:27<03:22,  1.35it/s]
2|6199|Loss: 0.9812778234481812:  96%|█████████▌| 6199/6470 [1:17:27<03:29,  1.29it/s]
2|6200|Loss: 0.8303003907203674:  96%|█████████▌| 6199/6470 [1:17:28<03:29,  1.29it/s]
2|6200|Loss: 0.8303003907203674:  96%|█████████▌| 6200/6470 [1:17:28<03:29,  1.29it/s]
2|6201|Loss: 0.7826423048973083:  96%|█████████▌| 6200/6470 [1:17:29<03:29,  1.29it/s]
2|6201|Loss: 0.7826423048973083:  96%|█████████▌| 6201/6470 [1:17:29<03:30,  1.28it/s]
2|6202|Loss: 0.8296822309494019:  96%|█████████▌| 6201/6470 [1:17:29<03:30,  1.28it/s]
2|6202|Loss: 0.8296822309494019:  96%|█████████▌| 6202/6470 [1:17:30<03:23,  1.32it/s]
2|6203|Loss: 0.7854554057121277:  96%|█████████▌| 6202/6470 [1:17:30<03:23,  1.32it/s]
2|6203|Loss: 0.7854554057121277:  96%|█████████▌| 6203/6470 [1:17:31<03:30,  1.27it/s]
2|6204|Loss: 0.7827444672584534:  96%|█████████▌| 6203/6470 [1:17:31<03:30,  1.27it/s]
2|6204|Loss: 0.7827444672584534:  96%|█████████▌| 6204/6470 [1:17:31<03:20,  1.33it/s]
2|6205|Loss: 1.0005695819854736:  96%|█████████▌| 6204/6470 [1:17:32<03:20,  1.33it/s]
2|6205|Loss: 1.0005695819854736:  96%|█████████▌| 6205/6470 [1:17:32<03:14,  1.36it/s]
2|6206|Loss: 0.7264226675033569:  96%|█████████▌| 6205/6470 [1:17:32<03:14,  1.36it/s]
2|6206|Loss: 0.7264226675033569:  96%|█████████▌| 6206/6470 [1:17:33<03:07,  1.41it/s]
2|6207|Loss: 0.8510435819625854:  96%|█████████▌| 6206/6470 [1:17:33<03:07,  1.41it/s]
2|6207|Loss: 0.8510435819625854:  96%|█████████▌| 6207/6470 [1:17:33<02:58,  1.47it/s]
2|6208|Loss: 1.0037908554077148:  96%|█████████▌| 6207/6470 [1:17:34<02:58,  1.47it/s]
2|6208|Loss: 1.0037908554077148:  96%|█████████▌| 6208/6470 [1:17:34<03:00,  1.45it/s]
2|6209|Loss: 0.9308256506919861:  96%|█████████▌| 6208/6470 [1:17:34<03:00,  1.45it/s]
2|6209|Loss: 0.9308256506919861:  96%|█████████▌| 6209/6470 [1:17:35<02:59,  1.46it/s]
2|6210|Loss: 0.7957667708396912:  96%|█████████▌| 6209/6470 [1:17:35<02:59,  1.46it/s]
2|6210|Loss: 0.7957667708396912:  96%|█████████▌| 6210/6470 [1:17:35<03:12,  1.35it/s]
2|6211|Loss: 0.8739098310470581:  96%|█████████▌| 6210/6470 [1:17:36<03:12,  1.35it/s]
2|6211|Loss: 0.8739098310470581:  96%|█████████▌| 6211/6470 [1:17:36<02:59,  1.44it/s]
2|6212|Loss: 0.9466948509216309:  96%|█████████▌| 6211/6470 [1:17:36<02:59,  1.44it/s]
2|6212|Loss: 0.9466948509216309:  96%|█████████▌| 6212/6470 [1:17:37<02:59,  1.44it/s]
2|6213|Loss: 1.004991888999939:  96%|█████████▌| 6212/6470 [1:17:37<02:59,  1.44it/s] 
2|6213|Loss: 1.004991888999939:  96%|█████████▌| 6213/6470 [1:17:37<03:05,  1.39it/s]
2|6214|Loss: 0.7337225675582886:  96%|█████████▌| 6213/6470 [1:17:38<03:05,  1.39it/s]
2|6214|Loss: 0.7337225675582886:  96%|█████████▌| 6214/6470 [1:17:38<03:10,  1.34it/s]
2|6215|Loss: 0.8320693969726562:  96%|█████████▌| 6214/6470 [1:17:39<03:10,  1.34it/s]
2|6215|Loss: 0.8320693969726562:  96%|█████████▌| 6215/6470 [1:17:39<03:20,  1.27it/s]
2|6216|Loss: 0.8894945979118347:  96%|█████████▌| 6215/6470 [1:17:40<03:20,  1.27it/s]
2|6216|Loss: 0.8894945979118347:  96%|█████████▌| 6216/6470 [1:17:40<03:22,  1.26it/s]
2|6217|Loss: 0.9818537831306458:  96%|█████████▌| 6216/6470 [1:17:40<03:22,  1.26it/s]
2|6217|Loss: 0.9818537831306458:  96%|█████████▌| 6217/6470 [1:17:41<03:10,  1.33it/s]
2|6218|Loss: 0.8200865387916565:  96%|█████████▌| 6217/6470 [1:17:41<03:10,  1.33it/s]
2|6218|Loss: 0.8200865387916565:  96%|█████████▌| 6218/6470 [1:17:41<03:10,  1.32it/s]
2|6219|Loss: 0.7836799025535583:  96%|█████████▌| 6218/6470 [1:17:42<03:10,  1.32it/s]
2|6219|Loss: 0.7836799025535583:  96%|█████████▌| 6219/6470 [1:17:42<03:13,  1.30it/s]
2|6220|Loss: 0.8975196480751038:  96%|█████████▌| 6219/6470 [1:17:43<03:13,  1.30it/s]
2|6220|Loss: 0.8975196480751038:  96%|█████████▌| 6220/6470 [1:17:43<03:04,  1.35it/s]
2|6221|Loss: 0.7321929335594177:  96%|█████████▌| 6220/6470 [1:17:43<03:04,  1.35it/s]
2|6221|Loss: 0.7321929335594177:  96%|█████████▌| 6221/6470 [1:17:44<02:59,  1.39it/s]
2|6222|Loss: 0.8268173933029175:  96%|█████████▌| 6221/6470 [1:17:44<02:59,  1.39it/s]
2|6222|Loss: 0.8268173933029175:  96%|█████████▌| 6222/6470 [1:17:44<03:00,  1.37it/s]
2|6223|Loss: 0.8623359203338623:  96%|█████████▌| 6222/6470 [1:17:45<03:00,  1.37it/s]
2|6223|Loss: 0.8623359203338623:  96%|█████████▌| 6223/6470 [1:17:45<03:13,  1.28it/s]
2|6224|Loss: 0.9258313775062561:  96%|█████████▌| 6223/6470 [1:17:46<03:13,  1.28it/s]
2|6224|Loss: 0.9258313775062561:  96%|█████████▌| 6224/6470 [1:17:46<03:02,  1.35it/s]
2|6225|Loss: 0.7436182498931885:  96%|█████████▌| 6224/6470 [1:17:46<03:02,  1.35it/s]
2|6225|Loss: 0.7436182498931885:  96%|█████████▌| 6225/6470 [1:17:47<02:59,  1.36it/s]
2|6226|Loss: 0.9120879173278809:  96%|█████████▌| 6225/6470 [1:17:47<02:59,  1.36it/s]
2|6226|Loss: 0.9120879173278809:  96%|█████████▌| 6226/6470 [1:17:47<03:01,  1.34it/s]
2|6227|Loss: 0.884418785572052:  96%|█████████▌| 6226/6470 [1:17:48<03:01,  1.34it/s] 
2|6227|Loss: 0.884418785572052:  96%|█████████▌| 6227/6470 [1:17:48<03:03,  1.32it/s]
2|6228|Loss: 0.8216041326522827:  96%|█████████▌| 6227/6470 [1:17:49<03:03,  1.32it/s]
2|6228|Loss: 0.8216041326522827:  96%|█████████▋| 6228/6470 [1:17:49<03:06,  1.30it/s]
2|6229|Loss: 0.909473717212677:  96%|█████████▋| 6228/6470 [1:17:49<03:06,  1.30it/s] 
2|6229|Loss: 0.909473717212677:  96%|█████████▋| 6229/6470 [1:17:50<03:15,  1.23it/s]
2|6230|Loss: 0.8591854572296143:  96%|█████████▋| 6229/6470 [1:17:50<03:15,  1.23it/s]
2|6230|Loss: 0.8591854572296143:  96%|█████████▋| 6230/6470 [1:17:51<03:08,  1.27it/s]
2|6231|Loss: 0.8996431827545166:  96%|█████████▋| 6230/6470 [1:17:51<03:08,  1.27it/s]
2|6231|Loss: 0.8996431827545166:  96%|█████████▋| 6231/6470 [1:17:51<02:57,  1.35it/s]
2|6232|Loss: 0.7469521760940552:  96%|█████████▋| 6231/6470 [1:17:52<02:57,  1.35it/s]
2|6232|Loss: 0.7469521760940552:  96%|█████████▋| 6232/6470 [1:17:52<02:48,  1.41it/s]
2|6233|Loss: 0.6519054770469666:  96%|█████████▋| 6232/6470 [1:17:52<02:48,  1.41it/s]
2|6233|Loss: 0.6519054770469666:  96%|█████████▋| 6233/6470 [1:17:52<02:33,  1.54it/s]
2|6234|Loss: 0.8660724759101868:  96%|█████████▋| 6233/6470 [1:17:53<02:33,  1.54it/s]
2|6234|Loss: 0.8660724759101868:  96%|█████████▋| 6234/6470 [1:17:53<02:31,  1.56it/s]
2|6235|Loss: 0.7918735146522522:  96%|█████████▋| 6234/6470 [1:17:53<02:31,  1.56it/s]
2|6235|Loss: 0.7918735146522522:  96%|█████████▋| 6235/6470 [1:17:54<02:34,  1.52it/s]
2|6236|Loss: 1.0308246612548828:  96%|█████████▋| 6235/6470 [1:17:54<02:34,  1.52it/s]
2|6236|Loss: 1.0308246612548828:  96%|█████████▋| 6236/6470 [1:17:55<02:50,  1.37it/s]
2|6237|Loss: 0.8097604513168335:  96%|█████████▋| 6236/6470 [1:17:55<02:50,  1.37it/s]
2|6237|Loss: 0.8097604513168335:  96%|█████████▋| 6237/6470 [1:17:55<03:03,  1.27it/s]
2|6238|Loss: 0.8288097381591797:  96%|█████████▋| 6237/6470 [1:17:56<03:03,  1.27it/s]
2|6238|Loss: 0.8288097381591797:  96%|█████████▋| 6238/6470 [1:17:56<03:05,  1.25it/s]
2|6239|Loss: 0.7958048582077026:  96%|█████████▋| 6238/6470 [1:17:57<03:05,  1.25it/s]
2|6239|Loss: 0.7958048582077026:  96%|█████████▋| 6239/6470 [1:17:57<02:53,  1.33it/s]
2|6240|Loss: 0.8365370631217957:  96%|█████████▋| 6239/6470 [1:17:57<02:53,  1.33it/s]
2|6240|Loss: 0.8365370631217957:  96%|█████████▋| 6240/6470 [1:17:58<02:58,  1.29it/s]
2|6241|Loss: 0.8048727512359619:  96%|█████████▋| 6240/6470 [1:17:58<02:58,  1.29it/s]
2|6241|Loss: 0.8048727512359619:  96%|█████████▋| 6241/6470 [1:17:59<02:54,  1.31it/s]
2|6242|Loss: 0.9674486517906189:  96%|█████████▋| 6241/6470 [1:17:59<02:54,  1.31it/s]
2|6242|Loss: 0.9674486517906189:  96%|█████████▋| 6242/6470 [1:17:59<02:53,  1.31it/s]
2|6243|Loss: 0.8612461090087891:  96%|█████████▋| 6242/6470 [1:18:00<02:53,  1.31it/s]
2|6243|Loss: 0.8612461090087891:  96%|█████████▋| 6243/6470 [1:18:00<02:53,  1.31it/s]
2|6244|Loss: 0.7161723375320435:  96%|█████████▋| 6243/6470 [1:18:00<02:53,  1.31it/s]
2|6244|Loss: 0.7161723375320435:  97%|█████████▋| 6244/6470 [1:18:01<02:44,  1.37it/s]
2|6245|Loss: 0.8111488819122314:  97%|█████████▋| 6244/6470 [1:18:01<02:44,  1.37it/s]
2|6245|Loss: 0.8111488819122314:  97%|█████████▋| 6245/6470 [1:18:01<02:45,  1.36it/s]
2|6246|Loss: 0.6608396768569946:  97%|█████████▋| 6245/6470 [1:18:02<02:45,  1.36it/s]
2|6246|Loss: 0.6608396768569946:  97%|█████████▋| 6246/6470 [1:18:02<02:48,  1.33it/s]
2|6247|Loss: 0.9460363388061523:  97%|█████████▋| 6246/6470 [1:18:03<02:48,  1.33it/s]
2|6247|Loss: 0.9460363388061523:  97%|█████████▋| 6247/6470 [1:18:03<02:40,  1.39it/s]
2|6248|Loss: 0.8415329456329346:  97%|█████████▋| 6247/6470 [1:18:03<02:40,  1.39it/s]
2|6248|Loss: 0.8415329456329346:  97%|█████████▋| 6248/6470 [1:18:03<02:29,  1.49it/s]
2|6249|Loss: 0.9606719017028809:  97%|█████████▋| 6248/6470 [1:18:04<02:29,  1.49it/s]
2|6249|Loss: 0.9606719017028809:  97%|█████████▋| 6249/6470 [1:18:04<02:28,  1.49it/s]
2|6250|Loss: 0.7325965762138367:  97%|█████████▋| 6249/6470 [1:18:05<02:28,  1.49it/s]
2|6250|Loss: 0.7325965762138367:  97%|█████████▋| 6250/6470 [1:18:05<02:32,  1.44it/s]
2|6251|Loss: 0.7963562607765198:  97%|█████████▋| 6250/6470 [1:18:05<02:32,  1.44it/s]
2|6251|Loss: 0.7963562607765198:  97%|█████████▋| 6251/6470 [1:18:06<02:32,  1.44it/s]
2|6252|Loss: 0.8152577877044678:  97%|█████████▋| 6251/6470 [1:18:06<02:32,  1.44it/s]
2|6252|Loss: 0.8152577877044678:  97%|█████████▋| 6252/6470 [1:18:06<02:36,  1.39it/s]
2|6253|Loss: 0.8342701196670532:  97%|█████████▋| 6252/6470 [1:18:07<02:36,  1.39it/s]
2|6253|Loss: 0.8342701196670532:  97%|█████████▋| 6253/6470 [1:18:07<02:33,  1.41it/s]
2|6254|Loss: 0.8495669364929199:  97%|█████████▋| 6253/6470 [1:18:07<02:33,  1.41it/s]
2|6254|Loss: 0.8495669364929199:  97%|█████████▋| 6254/6470 [1:18:08<02:41,  1.34it/s]
2|6255|Loss: 0.7089135050773621:  97%|█████████▋| 6254/6470 [1:18:08<02:41,  1.34it/s]
2|6255|Loss: 0.7089135050773621:  97%|█████████▋| 6255/6470 [1:18:09<02:42,  1.32it/s]
2|6256|Loss: 0.8020132184028625:  97%|█████████▋| 6255/6470 [1:18:09<02:42,  1.32it/s]
2|6256|Loss: 0.8020132184028625:  97%|█████████▋| 6256/6470 [1:18:09<02:34,  1.38it/s]
2|6257|Loss: 0.8347591161727905:  97%|█████████▋| 6256/6470 [1:18:10<02:34,  1.38it/s]
2|6257|Loss: 0.8347591161727905:  97%|█████████▋| 6257/6470 [1:18:10<02:40,  1.32it/s]
2|6258|Loss: 0.7655996680259705:  97%|█████████▋| 6257/6470 [1:18:11<02:40,  1.32it/s]
2|6258|Loss: 0.7655996680259705:  97%|█████████▋| 6258/6470 [1:18:11<02:53,  1.22it/s]
2|6259|Loss: 0.8450210690498352:  97%|█████████▋| 6258/6470 [1:18:12<02:53,  1.22it/s]
2|6259|Loss: 0.8450210690498352:  97%|█████████▋| 6259/6470 [1:18:12<02:52,  1.22it/s]
2|6260|Loss: 0.7682378888130188:  97%|█████████▋| 6259/6470 [1:18:12<02:52,  1.22it/s]
2|6260|Loss: 0.7682378888130188:  97%|█████████▋| 6260/6470 [1:18:12<02:33,  1.37it/s]
2|6261|Loss: 0.7874183654785156:  97%|█████████▋| 6260/6470 [1:18:13<02:33,  1.37it/s]
2|6261|Loss: 0.7874183654785156:  97%|█████████▋| 6261/6470 [1:18:13<02:30,  1.39it/s]
2|6262|Loss: 0.6571199893951416:  97%|█████████▋| 6261/6470 [1:18:14<02:30,  1.39it/s]
2|6262|Loss: 0.6571199893951416:  97%|█████████▋| 6262/6470 [1:18:14<02:27,  1.41it/s]
2|6263|Loss: 0.7628835439682007:  97%|█████████▋| 6262/6470 [1:18:14<02:27,  1.41it/s]
2|6263|Loss: 0.7628835439682007:  97%|█████████▋| 6263/6470 [1:18:15<02:37,  1.32it/s]
2|6264|Loss: 1.0131160020828247:  97%|█████████▋| 6263/6470 [1:18:15<02:37,  1.32it/s]
2|6264|Loss: 1.0131160020828247:  97%|█████████▋| 6264/6470 [1:18:16<02:44,  1.25it/s]
2|6265|Loss: 0.7545056343078613:  97%|█████████▋| 6264/6470 [1:18:16<02:44,  1.25it/s]
2|6265|Loss: 0.7545056343078613:  97%|█████████▋| 6265/6470 [1:18:16<02:32,  1.35it/s]
2|6266|Loss: 0.7637191414833069:  97%|█████████▋| 6265/6470 [1:18:17<02:32,  1.35it/s]
2|6266|Loss: 0.7637191414833069:  97%|█████████▋| 6266/6470 [1:18:17<02:33,  1.33it/s]
2|6267|Loss: 0.8113566637039185:  97%|█████████▋| 6266/6470 [1:18:17<02:33,  1.33it/s]
2|6267|Loss: 0.8113566637039185:  97%|█████████▋| 6267/6470 [1:18:18<02:33,  1.32it/s]
2|6268|Loss: 0.9447960257530212:  97%|█████████▋| 6267/6470 [1:18:18<02:33,  1.32it/s]
2|6268|Loss: 0.9447960257530212:  97%|█████████▋| 6268/6470 [1:18:18<02:28,  1.36it/s]
2|6269|Loss: 0.8548653721809387:  97%|█████████▋| 6268/6470 [1:18:19<02:28,  1.36it/s]
2|6269|Loss: 0.8548653721809387:  97%|█████████▋| 6269/6470 [1:18:19<02:15,  1.48it/s]
2|6270|Loss: 0.9064808487892151:  97%|█████████▋| 6269/6470 [1:18:19<02:15,  1.48it/s]
2|6270|Loss: 0.9064808487892151:  97%|█████████▋| 6270/6470 [1:18:20<02:12,  1.50it/s]
2|6271|Loss: 1.0288258790969849:  97%|█████████▋| 6270/6470 [1:18:20<02:12,  1.50it/s]
2|6271|Loss: 1.0288258790969849:  97%|█████████▋| 6271/6470 [1:18:20<02:22,  1.39it/s]
2|6272|Loss: 0.9963918328285217:  97%|█████████▋| 6271/6470 [1:18:21<02:22,  1.39it/s]
2|6272|Loss: 0.9963918328285217:  97%|█████████▋| 6272/6470 [1:18:21<02:27,  1.34it/s]
2|6273|Loss: 0.8539779782295227:  97%|█████████▋| 6272/6470 [1:18:22<02:27,  1.34it/s]
2|6273|Loss: 0.8539779782295227:  97%|█████████▋| 6273/6470 [1:18:22<02:31,  1.30it/s]
2|6274|Loss: 0.7911392450332642:  97%|█████████▋| 6273/6470 [1:18:22<02:31,  1.30it/s]
2|6274|Loss: 0.7911392450332642:  97%|█████████▋| 6274/6470 [1:18:23<02:30,  1.30it/s]
2|6275|Loss: 0.7387600541114807:  97%|█████████▋| 6274/6470 [1:18:23<02:30,  1.30it/s]
2|6275|Loss: 0.7387600541114807:  97%|█████████▋| 6275/6470 [1:18:23<02:17,  1.41it/s]
2|6276|Loss: 0.8937416076660156:  97%|█████████▋| 6275/6470 [1:18:24<02:17,  1.41it/s]
2|6276|Loss: 0.8937416076660156:  97%|█████████▋| 6276/6470 [1:18:24<02:17,  1.42it/s]
2|6277|Loss: 0.7695767879486084:  97%|█████████▋| 6276/6470 [1:18:25<02:17,  1.42it/s]
2|6277|Loss: 0.7695767879486084:  97%|█████████▋| 6277/6470 [1:18:25<02:24,  1.33it/s]
2|6278|Loss: 0.8578978776931763:  97%|█████████▋| 6277/6470 [1:18:25<02:24,  1.33it/s]
2|6278|Loss: 0.8578978776931763:  97%|█████████▋| 6278/6470 [1:18:26<02:21,  1.35it/s]
2|6279|Loss: 0.7982131838798523:  97%|█████████▋| 6278/6470 [1:18:26<02:21,  1.35it/s]
2|6279|Loss: 0.7982131838798523:  97%|█████████▋| 6279/6470 [1:18:26<02:22,  1.34it/s]
2|6280|Loss: 0.8303905129432678:  97%|█████████▋| 6279/6470 [1:18:27<02:22,  1.34it/s]
2|6280|Loss: 0.8303905129432678:  97%|█████████▋| 6280/6470 [1:18:27<02:15,  1.40it/s]
2|6281|Loss: 0.9109166860580444:  97%|█████████▋| 6280/6470 [1:18:27<02:15,  1.40it/s]
2|6281|Loss: 0.9109166860580444:  97%|█████████▋| 6281/6470 [1:18:28<02:05,  1.51it/s]
2|6282|Loss: 0.8818695545196533:  97%|█████████▋| 6281/6470 [1:18:28<02:05,  1.51it/s]
2|6282|Loss: 0.8818695545196533:  97%|█████████▋| 6282/6470 [1:18:28<02:08,  1.46it/s]
2|6283|Loss: 0.7766616940498352:  97%|█████████▋| 6282/6470 [1:18:29<02:08,  1.46it/s]
2|6283|Loss: 0.7766616940498352:  97%|█████████▋| 6283/6470 [1:18:29<01:55,  1.62it/s]
2|6284|Loss: 0.8862898349761963:  97%|█████████▋| 6283/6470 [1:18:29<01:55,  1.62it/s]
2|6284|Loss: 0.8862898349761963:  97%|█████████▋| 6284/6470 [1:18:30<01:59,  1.55it/s]
2|6285|Loss: 0.7476538419723511:  97%|█████████▋| 6284/6470 [1:18:30<01:59,  1.55it/s]
2|6285|Loss: 0.7476538419723511:  97%|█████████▋| 6285/6470 [1:18:30<02:03,  1.50it/s]
2|6286|Loss: 0.9517302513122559:  97%|█████████▋| 6285/6470 [1:18:31<02:03,  1.50it/s]
2|6286|Loss: 0.9517302513122559:  97%|█████████▋| 6286/6470 [1:18:31<02:08,  1.43it/s]
2|6287|Loss: 0.8116680383682251:  97%|█████████▋| 6286/6470 [1:18:31<02:08,  1.43it/s]
2|6287|Loss: 0.8116680383682251:  97%|█████████▋| 6287/6470 [1:18:32<02:17,  1.33it/s]
2|6288|Loss: 0.7634221911430359:  97%|█████████▋| 6287/6470 [1:18:32<02:17,  1.33it/s]
2|6288|Loss: 0.7634221911430359:  97%|█████████▋| 6288/6470 [1:18:33<02:17,  1.32it/s]
2|6289|Loss: 0.9482381939888:  97%|█████████▋| 6288/6470 [1:18:33<02:17,  1.32it/s]   
2|6289|Loss: 0.9482381939888:  97%|█████████▋| 6289/6470 [1:18:33<02:17,  1.32it/s]
2|6290|Loss: 1.010103702545166:  97%|█████████▋| 6289/6470 [1:18:34<02:17,  1.32it/s]
2|6290|Loss: 1.010103702545166:  97%|█████████▋| 6290/6470 [1:18:34<02:21,  1.28it/s]
2|6291|Loss: 0.8089567422866821:  97%|█████████▋| 6290/6470 [1:18:35<02:21,  1.28it/s]
2|6291|Loss: 0.8089567422866821:  97%|█████████▋| 6291/6470 [1:18:35<02:23,  1.25it/s]
2|6292|Loss: 0.8113747835159302:  97%|█████████▋| 6291/6470 [1:18:36<02:23,  1.25it/s]
2|6292|Loss: 0.8113747835159302:  97%|█████████▋| 6292/6470 [1:18:36<02:27,  1.21it/s]
2|6293|Loss: 0.8504626750946045:  97%|█████████▋| 6292/6470 [1:18:36<02:27,  1.21it/s]
2|6293|Loss: 0.8504626750946045:  97%|█████████▋| 6293/6470 [1:18:37<02:22,  1.24it/s]
2|6294|Loss: 0.7837584614753723:  97%|█████████▋| 6293/6470 [1:18:37<02:22,  1.24it/s]
2|6294|Loss: 0.7837584614753723:  97%|█████████▋| 6294/6470 [1:18:37<02:18,  1.27it/s]
2|6295|Loss: 0.9223973751068115:  97%|█████████▋| 6294/6470 [1:18:38<02:18,  1.27it/s]
2|6295|Loss: 0.9223973751068115:  97%|█████████▋| 6295/6470 [1:18:38<02:20,  1.24it/s]
2|6296|Loss: 0.8095914721488953:  97%|█████████▋| 6295/6470 [1:18:39<02:20,  1.24it/s]
2|6296|Loss: 0.8095914721488953:  97%|█████████▋| 6296/6470 [1:18:39<02:18,  1.26it/s]
2|6297|Loss: 0.722511887550354:  97%|█████████▋| 6296/6470 [1:18:40<02:18,  1.26it/s] 
2|6297|Loss: 0.722511887550354:  97%|█████████▋| 6297/6470 [1:18:40<02:28,  1.17it/s]
2|6298|Loss: 0.7950683236122131:  97%|█████████▋| 6297/6470 [1:18:41<02:28,  1.17it/s]
2|6298|Loss: 0.7950683236122131:  97%|█████████▋| 6298/6470 [1:18:41<02:21,  1.21it/s]
2|6299|Loss: 0.6762572526931763:  97%|█████████▋| 6298/6470 [1:18:41<02:21,  1.21it/s]
2|6299|Loss: 0.6762572526931763:  97%|█████████▋| 6299/6470 [1:18:42<02:26,  1.16it/s]
2|6300|Loss: 0.6821261644363403:  97%|█████████▋| 6299/6470 [1:18:42<02:26,  1.16it/s]
2|6300|Loss: 0.6821261644363403:  97%|█████████▋| 6300/6470 [1:18:43<02:35,  1.09it/s]
2|6301|Loss: 0.749615490436554:  97%|█████████▋| 6300/6470 [1:18:43<02:35,  1.09it/s] 
2|6301|Loss: 0.749615490436554:  97%|█████████▋| 6301/6470 [1:18:44<02:27,  1.14it/s]
2|6302|Loss: 0.8114290237426758:  97%|█████████▋| 6301/6470 [1:18:44<02:27,  1.14it/s]
2|6302|Loss: 0.8114290237426758:  97%|█████████▋| 6302/6470 [1:18:44<02:21,  1.19it/s]
2|6303|Loss: 0.8753187656402588:  97%|█████████▋| 6302/6470 [1:18:45<02:21,  1.19it/s]
2|6303|Loss: 0.8753187656402588:  97%|█████████▋| 6303/6470 [1:18:45<02:18,  1.21it/s]
2|6304|Loss: 0.8679508566856384:  97%|█████████▋| 6303/6470 [1:18:46<02:18,  1.21it/s]
2|6304|Loss: 0.8679508566856384:  97%|█████████▋| 6304/6470 [1:18:46<02:10,  1.27it/s]
2|6305|Loss: 0.9295998811721802:  97%|█████████▋| 6304/6470 [1:18:46<02:10,  1.27it/s]
2|6305|Loss: 0.9295998811721802:  97%|█████████▋| 6305/6470 [1:18:46<02:01,  1.36it/s]
2|6306|Loss: 0.7466832995414734:  97%|█████████▋| 6305/6470 [1:18:47<02:01,  1.36it/s]
2|6306|Loss: 0.7466832995414734:  97%|█████████▋| 6306/6470 [1:18:47<02:07,  1.28it/s]
2|6307|Loss: 0.7831464409828186:  97%|█████████▋| 6306/6470 [1:18:48<02:07,  1.28it/s]
2|6307|Loss: 0.7831464409828186:  97%|█████████▋| 6307/6470 [1:18:48<02:00,  1.35it/s]
2|6308|Loss: 0.8535270690917969:  97%|█████████▋| 6307/6470 [1:18:48<02:00,  1.35it/s]
2|6308|Loss: 0.8535270690917969:  97%|█████████▋| 6308/6470 [1:18:49<01:55,  1.40it/s]
2|6309|Loss: 0.7616745829582214:  97%|█████████▋| 6308/6470 [1:18:49<01:55,  1.40it/s]
2|6309|Loss: 0.7616745829582214:  98%|█████████▊| 6309/6470 [1:18:50<02:04,  1.29it/s]
2|6310|Loss: 0.7480959892272949:  98%|█████████▊| 6309/6470 [1:18:50<02:04,  1.29it/s]
2|6310|Loss: 0.7480959892272949:  98%|█████████▊| 6310/6470 [1:18:50<02:03,  1.30it/s]
2|6311|Loss: 0.869184136390686:  98%|█████████▊| 6310/6470 [1:18:51<02:03,  1.30it/s] 
2|6311|Loss: 0.869184136390686:  98%|█████████▊| 6311/6470 [1:18:51<02:10,  1.22it/s]
2|6312|Loss: 0.8020440340042114:  98%|█████████▊| 6311/6470 [1:18:52<02:10,  1.22it/s]
2|6312|Loss: 0.8020440340042114:  98%|█████████▊| 6312/6470 [1:18:52<02:15,  1.17it/s]
2|6313|Loss: 0.8663267493247986:  98%|█████████▊| 6312/6470 [1:18:53<02:15,  1.17it/s]
2|6313|Loss: 0.8663267493247986:  98%|█████████▊| 6313/6470 [1:18:53<02:03,  1.27it/s]
2|6314|Loss: 0.7097833156585693:  98%|█████████▊| 6313/6470 [1:18:53<02:03,  1.27it/s]
2|6314|Loss: 0.7097833156585693:  98%|█████████▊| 6314/6470 [1:18:53<01:56,  1.34it/s]
2|6315|Loss: 0.8378681540489197:  98%|█████████▊| 6314/6470 [1:18:54<01:56,  1.34it/s]
2|6315|Loss: 0.8378681540489197:  98%|█████████▊| 6315/6470 [1:18:54<02:00,  1.29it/s]
2|6316|Loss: 0.7415114045143127:  98%|█████████▊| 6315/6470 [1:18:55<02:00,  1.29it/s]
2|6316|Loss: 0.7415114045143127:  98%|█████████▊| 6316/6470 [1:18:55<02:01,  1.26it/s]
2|6317|Loss: 0.925190269947052:  98%|█████████▊| 6316/6470 [1:18:56<02:01,  1.26it/s] 
2|6317|Loss: 0.925190269947052:  98%|█████████▊| 6317/6470 [1:18:56<02:05,  1.22it/s]
2|6318|Loss: 0.877917468547821:  98%|█████████▊| 6317/6470 [1:18:57<02:05,  1.22it/s]
2|6318|Loss: 0.877917468547821:  98%|█████████▊| 6318/6470 [1:18:57<02:09,  1.17it/s]
2|6319|Loss: 0.8820130228996277:  98%|█████████▊| 6318/6470 [1:18:57<02:09,  1.17it/s]
2|6319|Loss: 0.8820130228996277:  98%|█████████▊| 6319/6470 [1:18:58<02:07,  1.18it/s]
2|6320|Loss: 0.6125455498695374:  98%|█████████▊| 6319/6470 [1:18:58<02:07,  1.18it/s]
2|6320|Loss: 0.6125455498695374:  98%|█████████▊| 6320/6470 [1:18:59<02:14,  1.11it/s]
2|6321|Loss: 0.7875217199325562:  98%|█████████▊| 6320/6470 [1:18:59<02:14,  1.11it/s]
2|6321|Loss: 0.7875217199325562:  98%|█████████▊| 6321/6470 [1:19:00<02:08,  1.16it/s]
2|6322|Loss: 0.8678611516952515:  98%|█████████▊| 6321/6470 [1:19:00<02:08,  1.16it/s]
2|6322|Loss: 0.8678611516952515:  98%|█████████▊| 6322/6470 [1:19:00<02:06,  1.17it/s]
2|6323|Loss: 0.8343746662139893:  98%|█████████▊| 6322/6470 [1:19:01<02:06,  1.17it/s]
2|6323|Loss: 0.8343746662139893:  98%|█████████▊| 6323/6470 [1:19:01<01:52,  1.31it/s]
2|6324|Loss: 0.7584994435310364:  98%|█████████▊| 6323/6470 [1:19:01<01:52,  1.31it/s]
2|6324|Loss: 0.7584994435310364:  98%|█████████▊| 6324/6470 [1:19:02<01:49,  1.33it/s]
2|6325|Loss: 0.8126305937767029:  98%|█████████▊| 6324/6470 [1:19:02<01:49,  1.33it/s]
2|6325|Loss: 0.8126305937767029:  98%|█████████▊| 6325/6470 [1:19:02<01:49,  1.33it/s]
2|6326|Loss: 0.7594899535179138:  98%|█████████▊| 6325/6470 [1:19:03<01:49,  1.33it/s]
2|6326|Loss: 0.7594899535179138:  98%|█████████▊| 6326/6470 [1:19:03<01:45,  1.36it/s]
2|6327|Loss: 0.9292287826538086:  98%|█████████▊| 6326/6470 [1:19:04<01:45,  1.36it/s]
2|6327|Loss: 0.9292287826538086:  98%|█████████▊| 6327/6470 [1:19:04<01:36,  1.48it/s]
2|6328|Loss: 0.7811362147331238:  98%|█████████▊| 6327/6470 [1:19:04<01:36,  1.48it/s]
2|6328|Loss: 0.7811362147331238:  98%|█████████▊| 6328/6470 [1:19:04<01:36,  1.46it/s]
2|6329|Loss: 0.8825928568840027:  98%|█████████▊| 6328/6470 [1:19:05<01:36,  1.46it/s]
2|6329|Loss: 0.8825928568840027:  98%|█████████▊| 6329/6470 [1:19:05<01:43,  1.36it/s]
2|6330|Loss: 0.8976336121559143:  98%|█████████▊| 6329/6470 [1:19:06<01:43,  1.36it/s]
2|6330|Loss: 0.8976336121559143:  98%|█████████▊| 6330/6470 [1:19:06<01:45,  1.32it/s]
2|6331|Loss: 0.8930997848510742:  98%|█████████▊| 6330/6470 [1:19:06<01:45,  1.32it/s]
2|6331|Loss: 0.8930997848510742:  98%|█████████▊| 6331/6470 [1:19:07<01:43,  1.34it/s]
2|6332|Loss: 0.8380829691886902:  98%|█████████▊| 6331/6470 [1:19:07<01:43,  1.34it/s]
2|6332|Loss: 0.8380829691886902:  98%|█████████▊| 6332/6470 [1:19:08<01:44,  1.31it/s]
2|6333|Loss: 0.9080301523208618:  98%|█████████▊| 6332/6470 [1:19:08<01:44,  1.31it/s]
2|6333|Loss: 0.9080301523208618:  98%|█████████▊| 6333/6470 [1:19:08<01:46,  1.28it/s]
2|6334|Loss: 0.8007640242576599:  98%|█████████▊| 6333/6470 [1:19:09<01:46,  1.28it/s]
2|6334|Loss: 0.8007640242576599:  98%|█████████▊| 6334/6470 [1:19:09<01:39,  1.37it/s]
2|6335|Loss: 0.7473123073577881:  98%|█████████▊| 6334/6470 [1:19:09<01:39,  1.37it/s]
2|6335|Loss: 0.7473123073577881:  98%|█████████▊| 6335/6470 [1:19:10<01:45,  1.27it/s]
2|6336|Loss: 0.7516586780548096:  98%|█████████▊| 6335/6470 [1:19:10<01:45,  1.27it/s]
2|6336|Loss: 0.7516586780548096:  98%|█████████▊| 6336/6470 [1:19:11<01:52,  1.19it/s]
2|6337|Loss: 0.8885961174964905:  98%|█████████▊| 6336/6470 [1:19:11<01:52,  1.19it/s]
2|6337|Loss: 0.8885961174964905:  98%|█████████▊| 6337/6470 [1:19:12<01:55,  1.16it/s]
2|6338|Loss: 0.8766183853149414:  98%|█████████▊| 6337/6470 [1:19:12<01:55,  1.16it/s]
2|6338|Loss: 0.8766183853149414:  98%|█████████▊| 6338/6470 [1:19:13<01:52,  1.17it/s]
2|6339|Loss: 0.8230348825454712:  98%|█████████▊| 6338/6470 [1:19:13<01:52,  1.17it/s]
2|6339|Loss: 0.8230348825454712:  98%|█████████▊| 6339/6470 [1:19:13<01:44,  1.26it/s]
2|6340|Loss: 0.623466432094574:  98%|█████████▊| 6339/6470 [1:19:14<01:44,  1.26it/s] 
2|6340|Loss: 0.623466432094574:  98%|█████████▊| 6340/6470 [1:19:14<01:35,  1.37it/s]
2|6341|Loss: 0.9226875305175781:  98%|█████████▊| 6340/6470 [1:19:14<01:35,  1.37it/s]
2|6341|Loss: 0.9226875305175781:  98%|█████████▊| 6341/6470 [1:19:15<01:32,  1.39it/s]
2|6342|Loss: 0.7108427286148071:  98%|█████████▊| 6341/6470 [1:19:15<01:32,  1.39it/s]
2|6342|Loss: 0.7108427286148071:  98%|█████████▊| 6342/6470 [1:19:15<01:26,  1.48it/s]
2|6343|Loss: 0.7384637594223022:  98%|█████████▊| 6342/6470 [1:19:16<01:26,  1.48it/s]
2|6343|Loss: 0.7384637594223022:  98%|█████████▊| 6343/6470 [1:19:16<01:31,  1.38it/s]
2|6344|Loss: 1.065880298614502:  98%|█████████▊| 6343/6470 [1:19:16<01:31,  1.38it/s] 
2|6344|Loss: 1.065880298614502:  98%|█████████▊| 6344/6470 [1:19:17<01:30,  1.39it/s]
2|6345|Loss: 0.8335532546043396:  98%|█████████▊| 6344/6470 [1:19:17<01:30,  1.39it/s]
2|6345|Loss: 0.8335532546043396:  98%|█████████▊| 6345/6470 [1:19:17<01:27,  1.42it/s]
2|6346|Loss: 0.8754823207855225:  98%|█████████▊| 6345/6470 [1:19:18<01:27,  1.42it/s]
2|6346|Loss: 0.8754823207855225:  98%|█████████▊| 6346/6470 [1:19:18<01:24,  1.47it/s]
2|6347|Loss: 0.8369568586349487:  98%|█████████▊| 6346/6470 [1:19:18<01:24,  1.47it/s]
2|6347|Loss: 0.8369568586349487:  98%|█████████▊| 6347/6470 [1:19:19<01:29,  1.37it/s]
2|6348|Loss: 0.8342006206512451:  98%|█████████▊| 6347/6470 [1:19:19<01:29,  1.37it/s]
2|6348|Loss: 0.8342006206512451:  98%|█████████▊| 6348/6470 [1:19:20<01:32,  1.32it/s]
2|6349|Loss: 0.8277976512908936:  98%|█████████▊| 6348/6470 [1:19:20<01:32,  1.32it/s]
2|6349|Loss: 0.8277976512908936:  98%|█████████▊| 6349/6470 [1:19:20<01:32,  1.31it/s]
2|6350|Loss: 0.8806749582290649:  98%|█████████▊| 6349/6470 [1:19:21<01:32,  1.31it/s]
2|6350|Loss: 0.8806749582290649:  98%|█████████▊| 6350/6470 [1:19:21<01:34,  1.27it/s]
2|6351|Loss: 0.6818687319755554:  98%|█████████▊| 6350/6470 [1:19:22<01:34,  1.27it/s]
2|6351|Loss: 0.6818687319755554:  98%|█████████▊| 6351/6470 [1:19:22<01:25,  1.39it/s]
2|6352|Loss: 0.8739500641822815:  98%|█████████▊| 6351/6470 [1:19:22<01:25,  1.39it/s]
2|6352|Loss: 0.8739500641822815:  98%|█████████▊| 6352/6470 [1:19:22<01:22,  1.44it/s]
2|6353|Loss: 0.7862067818641663:  98%|█████████▊| 6352/6470 [1:19:23<01:22,  1.44it/s]
2|6353|Loss: 0.7862067818641663:  98%|█████████▊| 6353/6470 [1:19:23<01:30,  1.29it/s]
2|6354|Loss: 0.7992566227912903:  98%|█████████▊| 6353/6470 [1:19:24<01:30,  1.29it/s]
2|6354|Loss: 0.7992566227912903:  98%|█████████▊| 6354/6470 [1:19:24<01:31,  1.27it/s]
2|6355|Loss: 0.8854686617851257:  98%|█████████▊| 6354/6470 [1:19:25<01:31,  1.27it/s]
2|6355|Loss: 0.8854686617851257:  98%|█████████▊| 6355/6470 [1:19:25<01:29,  1.29it/s]
2|6356|Loss: 0.7422399520874023:  98%|█████████▊| 6355/6470 [1:19:25<01:29,  1.29it/s]
2|6356|Loss: 0.7422399520874023:  98%|█████████▊| 6356/6470 [1:19:26<01:19,  1.43it/s]
2|6357|Loss: 0.6245973110198975:  98%|█████████▊| 6356/6470 [1:19:26<01:19,  1.43it/s]
2|6357|Loss: 0.6245973110198975:  98%|█████████▊| 6357/6470 [1:19:26<01:18,  1.43it/s]
2|6358|Loss: 1.0334956645965576:  98%|█████████▊| 6357/6470 [1:19:27<01:18,  1.43it/s]
2|6358|Loss: 1.0334956645965576:  98%|█████████▊| 6358/6470 [1:19:27<01:21,  1.37it/s]
2|6359|Loss: 0.8612127900123596:  98%|█████████▊| 6358/6470 [1:19:27<01:21,  1.37it/s]
2|6359|Loss: 0.8612127900123596:  98%|█████████▊| 6359/6470 [1:19:28<01:25,  1.29it/s]
2|6360|Loss: 0.8746428489685059:  98%|█████████▊| 6359/6470 [1:19:28<01:25,  1.29it/s]
2|6360|Loss: 0.8746428489685059:  98%|█████████▊| 6360/6470 [1:19:29<01:30,  1.22it/s]
2|6361|Loss: 0.7073760032653809:  98%|█████████▊| 6360/6470 [1:19:29<01:30,  1.22it/s]
2|6361|Loss: 0.7073760032653809:  98%|█████████▊| 6361/6470 [1:19:29<01:19,  1.37it/s]
2|6362|Loss: 0.7620192766189575:  98%|█████████▊| 6361/6470 [1:19:30<01:19,  1.37it/s]
2|6362|Loss: 0.7620192766189575:  98%|█████████▊| 6362/6470 [1:19:30<01:14,  1.44it/s]
2|6363|Loss: 0.7465351223945618:  98%|█████████▊| 6362/6470 [1:19:30<01:14,  1.44it/s]
2|6363|Loss: 0.7465351223945618:  98%|█████████▊| 6363/6470 [1:19:31<01:15,  1.41it/s]
2|6364|Loss: 0.7899169325828552:  98%|█████████▊| 6363/6470 [1:19:31<01:15,  1.41it/s]
2|6364|Loss: 0.7899169325828552:  98%|█████████▊| 6364/6470 [1:19:31<01:16,  1.38it/s]
2|6365|Loss: 0.8945620656013489:  98%|█████████▊| 6364/6470 [1:19:32<01:16,  1.38it/s]
2|6365|Loss: 0.8945620656013489:  98%|█████████▊| 6365/6470 [1:19:32<01:14,  1.40it/s]
2|6366|Loss: 0.8105756640434265:  98%|█████████▊| 6365/6470 [1:19:33<01:14,  1.40it/s]
2|6366|Loss: 0.8105756640434265:  98%|█████████▊| 6366/6470 [1:19:33<01:16,  1.37it/s]
2|6367|Loss: 0.8456617593765259:  98%|█████████▊| 6366/6470 [1:19:33<01:16,  1.37it/s]
2|6367|Loss: 0.8456617593765259:  98%|█████████▊| 6367/6470 [1:19:34<01:13,  1.40it/s]
2|6368|Loss: 0.774441123008728:  98%|█████████▊| 6367/6470 [1:19:34<01:13,  1.40it/s] 
2|6368|Loss: 0.774441123008728:  98%|█████████▊| 6368/6470 [1:19:34<01:09,  1.46it/s]
2|6369|Loss: 0.7355899810791016:  98%|█████████▊| 6368/6470 [1:19:35<01:09,  1.46it/s]
2|6369|Loss: 0.7355899810791016:  98%|█████████▊| 6369/6470 [1:19:35<01:04,  1.57it/s]
2|6370|Loss: 0.8822553753852844:  98%|█████████▊| 6369/6470 [1:19:35<01:04,  1.57it/s]
2|6370|Loss: 0.8822553753852844:  98%|█████████▊| 6370/6470 [1:19:36<01:11,  1.41it/s]
2|6371|Loss: 0.8028952479362488:  98%|█████████▊| 6370/6470 [1:19:36<01:11,  1.41it/s]
2|6371|Loss: 0.8028952479362488:  98%|█████████▊| 6371/6470 [1:19:36<01:10,  1.40it/s]
2|6372|Loss: 0.8465107083320618:  98%|█████████▊| 6371/6470 [1:19:37<01:10,  1.40it/s]
2|6372|Loss: 0.8465107083320618:  98%|█████████▊| 6372/6470 [1:19:37<01:14,  1.32it/s]
2|6373|Loss: 0.7957205176353455:  98%|█████████▊| 6372/6470 [1:19:38<01:14,  1.32it/s]
2|6373|Loss: 0.7957205176353455:  99%|█████████▊| 6373/6470 [1:19:38<01:14,  1.31it/s]
2|6374|Loss: 0.974845290184021:  99%|█████████▊| 6373/6470 [1:19:38<01:14,  1.31it/s] 
2|6374|Loss: 0.974845290184021:  99%|█████████▊| 6374/6470 [1:19:39<01:09,  1.38it/s]
2|6375|Loss: 0.7335935831069946:  99%|█████████▊| 6374/6470 [1:19:39<01:09,  1.38it/s]
2|6375|Loss: 0.7335935831069946:  99%|█████████▊| 6375/6470 [1:19:39<01:07,  1.40it/s]
2|6376|Loss: 0.8049947023391724:  99%|█████████▊| 6375/6470 [1:19:40<01:07,  1.40it/s]
2|6376|Loss: 0.8049947023391724:  99%|█████████▊| 6376/6470 [1:19:40<01:05,  1.44it/s]
2|6377|Loss: 0.8200966119766235:  99%|█████████▊| 6376/6470 [1:19:40<01:05,  1.44it/s]
2|6377|Loss: 0.8200966119766235:  99%|█████████▊| 6377/6470 [1:19:41<01:03,  1.46it/s]
2|6378|Loss: 0.8242006897926331:  99%|█████████▊| 6377/6470 [1:19:41<01:03,  1.46it/s]
2|6378|Loss: 0.8242006897926331:  99%|█████████▊| 6378/6470 [1:19:41<01:05,  1.40it/s]
2|6379|Loss: 0.9865814447402954:  99%|█████████▊| 6378/6470 [1:19:42<01:05,  1.40it/s]
2|6379|Loss: 0.9865814447402954:  99%|█████████▊| 6379/6470 [1:19:42<01:01,  1.47it/s]
2|6380|Loss: 0.8307706713676453:  99%|█████████▊| 6379/6470 [1:19:42<01:01,  1.47it/s]
2|6380|Loss: 0.8307706713676453:  99%|█████████▊| 6380/6470 [1:19:43<01:00,  1.49it/s]
2|6381|Loss: 0.9606716632843018:  99%|█████████▊| 6380/6470 [1:19:43<01:00,  1.49it/s]
2|6381|Loss: 0.9606716632843018:  99%|█████████▊| 6381/6470 [1:19:43<00:59,  1.50it/s]
2|6382|Loss: 0.8796235918998718:  99%|█████████▊| 6381/6470 [1:19:44<00:59,  1.50it/s]
2|6382|Loss: 0.8796235918998718:  99%|█████████▊| 6382/6470 [1:19:44<01:01,  1.44it/s]
2|6383|Loss: 0.9126684665679932:  99%|█████████▊| 6382/6470 [1:19:44<01:01,  1.44it/s]
2|6383|Loss: 0.9126684665679932:  99%|█████████▊| 6383/6470 [1:19:45<01:00,  1.44it/s]
2|6384|Loss: 0.7544575929641724:  99%|█████████▊| 6383/6470 [1:19:45<01:00,  1.44it/s]
2|6384|Loss: 0.7544575929641724:  99%|█████████▊| 6384/6470 [1:19:45<00:56,  1.52it/s]
2|6385|Loss: 0.8022237420082092:  99%|█████████▊| 6384/6470 [1:19:46<00:56,  1.52it/s]
2|6385|Loss: 0.8022237420082092:  99%|█████████▊| 6385/6470 [1:19:46<00:56,  1.51it/s]
2|6386|Loss: 1.0508285760879517:  99%|█████████▊| 6385/6470 [1:19:46<00:56,  1.51it/s]
2|6386|Loss: 1.0508285760879517:  99%|█████████▊| 6386/6470 [1:19:47<01:02,  1.34it/s]
2|6387|Loss: 0.7954398393630981:  99%|█████████▊| 6386/6470 [1:19:47<01:02,  1.34it/s]
2|6387|Loss: 0.7954398393630981:  99%|█████████▊| 6387/6470 [1:19:48<01:02,  1.34it/s]
2|6388|Loss: 0.5734601020812988:  99%|█████████▊| 6387/6470 [1:19:48<01:02,  1.34it/s]
2|6388|Loss: 0.5734601020812988:  99%|█████████▊| 6388/6470 [1:19:48<00:59,  1.38it/s]
2|6389|Loss: 0.8261953592300415:  99%|█████████▊| 6388/6470 [1:19:49<00:59,  1.38it/s]
2|6389|Loss: 0.8261953592300415:  99%|█████████▊| 6389/6470 [1:19:49<00:57,  1.40it/s]
2|6390|Loss: 0.9913449883460999:  99%|█████████▊| 6389/6470 [1:19:50<00:57,  1.40it/s]
2|6390|Loss: 0.9913449883460999:  99%|█████████▉| 6390/6470 [1:19:50<00:59,  1.34it/s]
2|6391|Loss: 0.8237342238426208:  99%|█████████▉| 6390/6470 [1:19:50<00:59,  1.34it/s]
2|6391|Loss: 0.8237342238426208:  99%|█████████▉| 6391/6470 [1:19:51<01:00,  1.31it/s]
2|6392|Loss: 0.7750856876373291:  99%|█████████▉| 6391/6470 [1:19:51<01:00,  1.31it/s]
2|6392|Loss: 0.7750856876373291:  99%|█████████▉| 6392/6470 [1:19:51<00:54,  1.42it/s]
2|6393|Loss: 0.7643737196922302:  99%|█████████▉| 6392/6470 [1:19:52<00:54,  1.42it/s]
2|6393|Loss: 0.7643737196922302:  99%|█████████▉| 6393/6470 [1:19:52<00:53,  1.43it/s]
2|6394|Loss: 0.7379480004310608:  99%|█████████▉| 6393/6470 [1:19:52<00:53,  1.43it/s]
2|6394|Loss: 0.7379480004310608:  99%|█████████▉| 6394/6470 [1:19:53<00:53,  1.42it/s]
2|6395|Loss: 0.9361158013343811:  99%|█████████▉| 6394/6470 [1:19:53<00:53,  1.42it/s]
2|6395|Loss: 0.9361158013343811:  99%|█████████▉| 6395/6470 [1:19:53<00:52,  1.44it/s]
2|6396|Loss: 0.8749435544013977:  99%|█████████▉| 6395/6470 [1:19:54<00:52,  1.44it/s]
2|6396|Loss: 0.8749435544013977:  99%|█████████▉| 6396/6470 [1:19:54<00:50,  1.47it/s]
2|6397|Loss: 0.8115247488021851:  99%|█████████▉| 6396/6470 [1:19:54<00:50,  1.47it/s]
2|6397|Loss: 0.8115247488021851:  99%|█████████▉| 6397/6470 [1:19:55<00:50,  1.44it/s]
2|6398|Loss: 0.7911393642425537:  99%|█████████▉| 6397/6470 [1:19:55<00:50,  1.44it/s]
2|6398|Loss: 0.7911393642425537:  99%|█████████▉| 6398/6470 [1:19:56<00:54,  1.32it/s]
2|6399|Loss: 0.8638908267021179:  99%|█████████▉| 6398/6470 [1:19:56<00:54,  1.32it/s]
2|6399|Loss: 0.8638908267021179:  99%|█████████▉| 6399/6470 [1:19:56<00:51,  1.37it/s]
2|6400|Loss: 0.8232913017272949:  99%|█████████▉| 6399/6470 [1:19:57<00:51,  1.37it/s]
2|6400|Loss: 0.8232913017272949:  99%|█████████▉| 6400/6470 [1:19:57<00:49,  1.40it/s]
2|6401|Loss: 0.7952827215194702:  99%|█████████▉| 6400/6470 [1:19:57<00:49,  1.40it/s]
2|6401|Loss: 0.7952827215194702:  99%|█████████▉| 6401/6470 [1:19:58<00:52,  1.31it/s]
2|6402|Loss: 0.794773519039154:  99%|█████████▉| 6401/6470 [1:19:58<00:52,  1.31it/s] 
2|6402|Loss: 0.794773519039154:  99%|█████████▉| 6402/6470 [1:19:59<00:51,  1.31it/s]
2|6403|Loss: 0.7765761017799377:  99%|█████████▉| 6402/6470 [1:19:59<00:51,  1.31it/s]
2|6403|Loss: 0.7765761017799377:  99%|█████████▉| 6403/6470 [1:19:59<00:46,  1.45it/s]
2|6404|Loss: 1.0521631240844727:  99%|█████████▉| 6403/6470 [1:19:59<00:46,  1.45it/s]
2|6404|Loss: 1.0521631240844727:  99%|█████████▉| 6404/6470 [1:20:00<00:43,  1.52it/s]
2|6405|Loss: 0.8394697308540344:  99%|█████████▉| 6404/6470 [1:20:00<00:43,  1.52it/s]
2|6405|Loss: 0.8394697308540344:  99%|█████████▉| 6405/6470 [1:20:00<00:44,  1.45it/s]
2|6406|Loss: 0.8846989274024963:  99%|█████████▉| 6405/6470 [1:20:01<00:44,  1.45it/s]
2|6406|Loss: 0.8846989274024963:  99%|█████████▉| 6406/6470 [1:20:01<00:43,  1.48it/s]
2|6407|Loss: 0.8513966202735901:  99%|█████████▉| 6406/6470 [1:20:01<00:43,  1.48it/s]
2|6407|Loss: 0.8513966202735901:  99%|█████████▉| 6407/6470 [1:20:02<00:43,  1.44it/s]
2|6408|Loss: 0.8760343194007874:  99%|█████████▉| 6407/6470 [1:20:02<00:43,  1.44it/s]
2|6408|Loss: 0.8760343194007874:  99%|█████████▉| 6408/6470 [1:20:03<00:42,  1.45it/s]
2|6409|Loss: 0.8181357383728027:  99%|█████████▉| 6408/6470 [1:20:03<00:42,  1.45it/s]
2|6409|Loss: 0.8181357383728027:  99%|█████████▉| 6409/6470 [1:20:03<00:40,  1.49it/s]
2|6410|Loss: 0.7523671984672546:  99%|█████████▉| 6409/6470 [1:20:04<00:40,  1.49it/s]
2|6410|Loss: 0.7523671984672546:  99%|█████████▉| 6410/6470 [1:20:04<00:39,  1.53it/s]
2|6411|Loss: 0.8790935277938843:  99%|█████████▉| 6410/6470 [1:20:04<00:39,  1.53it/s]
2|6411|Loss: 0.8790935277938843:  99%|█████████▉| 6411/6470 [1:20:04<00:39,  1.48it/s]
2|6412|Loss: 0.8059013485908508:  99%|█████████▉| 6411/6470 [1:20:05<00:39,  1.48it/s]
2|6412|Loss: 0.8059013485908508:  99%|█████████▉| 6412/6470 [1:20:05<00:41,  1.41it/s]
2|6413|Loss: 0.9571815729141235:  99%|█████████▉| 6412/6470 [1:20:06<00:41,  1.41it/s]
2|6413|Loss: 0.9571815729141235:  99%|█████████▉| 6413/6470 [1:20:06<00:41,  1.37it/s]
2|6414|Loss: 0.8421366810798645:  99%|█████████▉| 6413/6470 [1:20:06<00:41,  1.37it/s]
2|6414|Loss: 0.8421366810798645:  99%|█████████▉| 6414/6470 [1:20:07<00:42,  1.31it/s]
2|6415|Loss: 0.9563822746276855:  99%|█████████▉| 6414/6470 [1:20:07<00:42,  1.31it/s]
2|6415|Loss: 0.9563822746276855:  99%|█████████▉| 6415/6470 [1:20:08<00:39,  1.38it/s]
2|6416|Loss: 0.8796992897987366:  99%|█████████▉| 6415/6470 [1:20:08<00:39,  1.38it/s]
2|6416|Loss: 0.8796992897987366:  99%|█████████▉| 6416/6470 [1:20:08<00:39,  1.38it/s]
2|6417|Loss: 0.8625984787940979:  99%|█████████▉| 6416/6470 [1:20:09<00:39,  1.38it/s]
2|6417|Loss: 0.8625984787940979:  99%|█████████▉| 6417/6470 [1:20:09<00:38,  1.39it/s]
2|6418|Loss: 0.9265706539154053:  99%|█████████▉| 6417/6470 [1:20:09<00:38,  1.39it/s]
2|6418|Loss: 0.9265706539154053:  99%|█████████▉| 6418/6470 [1:20:10<00:39,  1.33it/s]
2|6419|Loss: 0.8186879754066467:  99%|█████████▉| 6418/6470 [1:20:10<00:39,  1.33it/s]
2|6419|Loss: 0.8186879754066467:  99%|█████████▉| 6419/6470 [1:20:11<00:39,  1.29it/s]
2|6420|Loss: 0.8474264740943909:  99%|█████████▉| 6419/6470 [1:20:11<00:39,  1.29it/s]
2|6420|Loss: 0.8474264740943909:  99%|█████████▉| 6420/6470 [1:20:11<00:38,  1.30it/s]
2|6421|Loss: 0.9487100839614868:  99%|█████████▉| 6420/6470 [1:20:12<00:38,  1.30it/s]
2|6421|Loss: 0.9487100839614868:  99%|█████████▉| 6421/6470 [1:20:12<00:37,  1.31it/s]
2|6422|Loss: 0.8266906142234802:  99%|█████████▉| 6421/6470 [1:20:13<00:37,  1.31it/s]
2|6422|Loss: 0.8266906142234802:  99%|█████████▉| 6422/6470 [1:20:13<00:36,  1.31it/s]
2|6423|Loss: 0.8014907240867615:  99%|█████████▉| 6422/6470 [1:20:13<00:36,  1.31it/s]
2|6423|Loss: 0.8014907240867615:  99%|█████████▉| 6423/6470 [1:20:14<00:37,  1.24it/s]
2|6424|Loss: 0.8083846569061279:  99%|█████████▉| 6423/6470 [1:20:14<00:37,  1.24it/s]
2|6424|Loss: 0.8083846569061279:  99%|█████████▉| 6424/6470 [1:20:14<00:35,  1.31it/s]
2|6425|Loss: 0.8161199688911438:  99%|█████████▉| 6424/6470 [1:20:15<00:35,  1.31it/s]
2|6425|Loss: 0.8161199688911438:  99%|█████████▉| 6425/6470 [1:20:15<00:35,  1.27it/s]
2|6426|Loss: 0.8406643271446228:  99%|█████████▉| 6425/6470 [1:20:16<00:35,  1.27it/s]
2|6426|Loss: 0.8406643271446228:  99%|█████████▉| 6426/6470 [1:20:16<00:35,  1.25it/s]
2|6427|Loss: 0.8142321109771729:  99%|█████████▉| 6426/6470 [1:20:17<00:35,  1.25it/s]
2|6427|Loss: 0.8142321109771729:  99%|█████████▉| 6427/6470 [1:20:17<00:33,  1.28it/s]
2|6428|Loss: 0.9624844193458557:  99%|█████████▉| 6427/6470 [1:20:17<00:33,  1.28it/s]
2|6428|Loss: 0.9624844193458557:  99%|█████████▉| 6428/6470 [1:20:18<00:32,  1.29it/s]
2|6429|Loss: 0.9018592238426208:  99%|█████████▉| 6428/6470 [1:20:18<00:32,  1.29it/s]
2|6429|Loss: 0.9018592238426208:  99%|█████████▉| 6429/6470 [1:20:18<00:30,  1.34it/s]
2|6430|Loss: 0.8824682831764221:  99%|█████████▉| 6429/6470 [1:20:19<00:30,  1.34it/s]
2|6430|Loss: 0.8824682831764221:  99%|█████████▉| 6430/6470 [1:20:19<00:28,  1.38it/s]
2|6431|Loss: 0.8444647789001465:  99%|█████████▉| 6430/6470 [1:20:19<00:28,  1.38it/s]
2|6431|Loss: 0.8444647789001465:  99%|█████████▉| 6431/6470 [1:20:20<00:28,  1.37it/s]
2|6432|Loss: 0.8638582229614258:  99%|█████████▉| 6431/6470 [1:20:20<00:28,  1.37it/s]
2|6432|Loss: 0.8638582229614258:  99%|█████████▉| 6432/6470 [1:20:20<00:26,  1.43it/s]
2|6433|Loss: 0.7251523733139038:  99%|█████████▉| 6432/6470 [1:20:21<00:26,  1.43it/s]
2|6433|Loss: 0.7251523733139038:  99%|█████████▉| 6433/6470 [1:20:21<00:25,  1.45it/s]
2|6434|Loss: 0.8778234124183655:  99%|█████████▉| 6433/6470 [1:20:21<00:25,  1.45it/s]
2|6434|Loss: 0.8778234124183655:  99%|█████████▉| 6434/6470 [1:20:22<00:26,  1.36it/s]
2|6435|Loss: 0.8266534209251404:  99%|█████████▉| 6434/6470 [1:20:22<00:26,  1.36it/s]
2|6435|Loss: 0.8266534209251404:  99%|█████████▉| 6435/6470 [1:20:23<00:27,  1.26it/s]
2|6436|Loss: 0.7269219160079956:  99%|█████████▉| 6435/6470 [1:20:23<00:27,  1.26it/s]
2|6436|Loss: 0.7269219160079956:  99%|█████████▉| 6436/6470 [1:20:23<00:25,  1.31it/s]
2|6437|Loss: 0.7561949491500854:  99%|█████████▉| 6436/6470 [1:20:24<00:25,  1.31it/s]
2|6437|Loss: 0.7561949491500854:  99%|█████████▉| 6437/6470 [1:20:24<00:24,  1.37it/s]
2|6438|Loss: 0.9088847637176514:  99%|█████████▉| 6437/6470 [1:20:25<00:24,  1.37it/s]
2|6438|Loss: 0.9088847637176514: 100%|█████████▉| 6438/6470 [1:20:25<00:24,  1.30it/s]
2|6439|Loss: 0.9710361957550049: 100%|█████████▉| 6438/6470 [1:20:25<00:24,  1.30it/s]
2|6439|Loss: 0.9710361957550049: 100%|█████████▉| 6439/6470 [1:20:26<00:25,  1.23it/s]
2|6440|Loss: 0.6852313280105591: 100%|█████████▉| 6439/6470 [1:20:26<00:25,  1.23it/s]
2|6440|Loss: 0.6852313280105591: 100%|█████████▉| 6440/6470 [1:20:27<00:23,  1.26it/s]
2|6441|Loss: 0.9690271615982056: 100%|█████████▉| 6440/6470 [1:20:27<00:23,  1.26it/s]
2|6441|Loss: 0.9690271615982056: 100%|█████████▉| 6441/6470 [1:20:28<00:23,  1.22it/s]
2|6442|Loss: 0.7581127285957336: 100%|█████████▉| 6441/6470 [1:20:28<00:23,  1.22it/s]
2|6442|Loss: 0.7581127285957336: 100%|█████████▉| 6442/6470 [1:20:28<00:22,  1.26it/s]
2|6443|Loss: 0.8575451374053955: 100%|█████████▉| 6442/6470 [1:20:29<00:22,  1.26it/s]
2|6443|Loss: 0.8575451374053955: 100%|█████████▉| 6443/6470 [1:20:29<00:21,  1.28it/s]
2|6444|Loss: 0.8078499436378479: 100%|█████████▉| 6443/6470 [1:20:29<00:21,  1.28it/s]
2|6444|Loss: 0.8078499436378479: 100%|█████████▉| 6444/6470 [1:20:30<00:19,  1.31it/s]
2|6445|Loss: 0.9305698871612549: 100%|█████████▉| 6444/6470 [1:20:30<00:19,  1.31it/s]
2|6445|Loss: 0.9305698871612549: 100%|█████████▉| 6445/6470 [1:20:30<00:18,  1.36it/s]
2|6446|Loss: 0.9613839983940125: 100%|█████████▉| 6445/6470 [1:20:31<00:18,  1.36it/s]
2|6446|Loss: 0.9613839983940125: 100%|█████████▉| 6446/6470 [1:20:31<00:17,  1.34it/s]
2|6447|Loss: 0.7371962666511536: 100%|█████████▉| 6446/6470 [1:20:32<00:17,  1.34it/s]
2|6447|Loss: 0.7371962666511536: 100%|█████████▉| 6447/6470 [1:20:32<00:16,  1.38it/s]
2|6448|Loss: 0.89085453748703: 100%|█████████▉| 6447/6470 [1:20:32<00:16,  1.38it/s]  
2|6448|Loss: 0.89085453748703: 100%|█████████▉| 6448/6470 [1:20:33<00:17,  1.28it/s]
2|6449|Loss: 0.8028310537338257: 100%|█████████▉| 6448/6470 [1:20:33<00:17,  1.28it/s]
2|6449|Loss: 0.8028310537338257: 100%|█████████▉| 6449/6470 [1:20:34<00:16,  1.29it/s]
2|6450|Loss: 0.9249975681304932: 100%|█████████▉| 6449/6470 [1:20:34<00:16,  1.29it/s]
2|6450|Loss: 0.9249975681304932: 100%|█████████▉| 6450/6470 [1:20:34<00:15,  1.28it/s]
2|6451|Loss: 0.7703728675842285: 100%|█████████▉| 6450/6470 [1:20:35<00:15,  1.28it/s]
2|6451|Loss: 0.7703728675842285: 100%|█████████▉| 6451/6470 [1:20:35<00:15,  1.22it/s]
2|6452|Loss: 0.7742754220962524: 100%|█████████▉| 6451/6470 [1:20:36<00:15,  1.22it/s]
2|6452|Loss: 0.7742754220962524: 100%|█████████▉| 6452/6470 [1:20:36<00:14,  1.25it/s]
2|6453|Loss: 0.6458561420440674: 100%|█████████▉| 6452/6470 [1:20:36<00:14,  1.25it/s]
2|6453|Loss: 0.6458561420440674: 100%|█████████▉| 6453/6470 [1:20:37<00:13,  1.31it/s]
2|6454|Loss: 1.0321331024169922: 100%|█████████▉| 6453/6470 [1:20:37<00:13,  1.31it/s]
2|6454|Loss: 1.0321331024169922: 100%|█████████▉| 6454/6470 [1:20:37<00:11,  1.40it/s]
2|6455|Loss: 0.7859109044075012: 100%|█████████▉| 6454/6470 [1:20:38<00:11,  1.40it/s]
2|6455|Loss: 0.7859109044075012: 100%|█████████▉| 6455/6470 [1:20:38<00:11,  1.32it/s]
2|6456|Loss: 0.7937681674957275: 100%|█████████▉| 6455/6470 [1:20:39<00:11,  1.32it/s]
2|6456|Loss: 0.7937681674957275: 100%|█████████▉| 6456/6470 [1:20:39<00:10,  1.29it/s]
2|6457|Loss: 0.7687339782714844: 100%|█████████▉| 6456/6470 [1:20:39<00:10,  1.29it/s]
2|6457|Loss: 0.7687339782714844: 100%|█████████▉| 6457/6470 [1:20:39<00:08,  1.46it/s]
2|6458|Loss: 0.8668889403343201: 100%|█████████▉| 6457/6470 [1:20:40<00:08,  1.46it/s]
2|6458|Loss: 0.8668889403343201: 100%|█████████▉| 6458/6470 [1:20:40<00:07,  1.55it/s]
2|6459|Loss: 0.7741442918777466: 100%|█████████▉| 6458/6470 [1:20:40<00:07,  1.55it/s]
2|6459|Loss: 0.7741442918777466: 100%|█████████▉| 6459/6470 [1:20:41<00:07,  1.44it/s]
2|6460|Loss: 0.9761037826538086: 100%|█████████▉| 6459/6470 [1:20:41<00:07,  1.44it/s]
2|6460|Loss: 0.9761037826538086: 100%|█████████▉| 6460/6470 [1:20:42<00:07,  1.39it/s]
2|6461|Loss: 0.7326341867446899: 100%|█████████▉| 6460/6470 [1:20:42<00:07,  1.39it/s]
2|6461|Loss: 0.7326341867446899: 100%|█████████▉| 6461/6470 [1:20:42<00:06,  1.40it/s]
2|6462|Loss: 0.9262415766716003: 100%|█████████▉| 6461/6470 [1:20:43<00:06,  1.40it/s]
2|6462|Loss: 0.9262415766716003: 100%|█████████▉| 6462/6470 [1:20:43<00:06,  1.32it/s]
2|6463|Loss: 0.8443769216537476: 100%|█████████▉| 6462/6470 [1:20:44<00:06,  1.32it/s]
2|6463|Loss: 0.8443769216537476: 100%|█████████▉| 6463/6470 [1:20:44<00:05,  1.27it/s]
2|6464|Loss: 0.7204803228378296: 100%|█████████▉| 6463/6470 [1:20:44<00:05,  1.27it/s]
2|6464|Loss: 0.7204803228378296: 100%|█████████▉| 6464/6470 [1:20:45<00:04,  1.25it/s]
2|6465|Loss: 0.8616662621498108: 100%|█████████▉| 6464/6470 [1:20:45<00:04,  1.25it/s]
2|6465|Loss: 0.8616662621498108: 100%|█████████▉| 6465/6470 [1:20:46<00:03,  1.29it/s]
2|6466|Loss: 0.9377252459526062: 100%|█████████▉| 6465/6470 [1:20:46<00:03,  1.29it/s]
2|6466|Loss: 0.9377252459526062: 100%|█████████▉| 6466/6470 [1:20:46<00:03,  1.25it/s]
2|6467|Loss: 0.8128237128257751: 100%|█████████▉| 6466/6470 [1:20:47<00:03,  1.25it/s]
2|6467|Loss: 0.8128237128257751: 100%|█████████▉| 6467/6470 [1:20:47<00:02,  1.19it/s]
2|6468|Loss: 0.9469826817512512: 100%|█████████▉| 6467/6470 [1:20:48<00:02,  1.19it/s]
2|6468|Loss: 0.9469826817512512: 100%|█████████▉| 6468/6470 [1:20:48<00:01,  1.21it/s]
2|6469|Loss: 1.047271490097046: 100%|█████████▉| 6468/6470 [1:20:49<00:01,  1.21it/s] 
2|6469|Loss: 1.047271490097046: 100%|█████████▉| 6469/6470 [1:20:49<00:00,  1.18it/s]
2|6470|Loss: 0.8314648270606995: 100%|█████████▉| 6469/6470 [1:20:49<00:00,  1.18it/s]
2|6470|Loss: 0.8314648270606995: 100%|██████████| 6470/6470 [1:20:50<00:00,  1.20it/s]
2|6470|Loss: 0.8314648270606995: 100%|██████████| 6470/6470 [1:20:50<00:00,  1.33it/s]
Model checkpoint of size 25961 MB saved to output/alpaca-colorful-llama2-finetune/model_1_12940.ckpt
Skipping uploading to HuggingFace Hub (no repo id specified)

  0%|          | 0/6470 [00:00<?, ?it/s]
3|1|Loss: 0.7707856297492981:   0%|          | 0/6470 [00:00<?, ?it/s]
3|1|Loss: 0.7707856297492981:   0%|          | 1/6470 [00:00<1:04:20,  1.68it/s]
3|2|Loss: 0.8351895213127136:   0%|          | 1/6470 [00:00<1:04:20,  1.68it/s]
3|2|Loss: 0.8351895213127136:   0%|          | 2/6470 [00:01<1:10:06,  1.54it/s]
3|3|Loss: 0.7861567139625549:   0%|          | 2/6470 [00:01<1:10:06,  1.54it/s]
3|3|Loss: 0.7861567139625549:   0%|          | 3/6470 [00:02<1:27:34,  1.23it/s]
3|4|Loss: 0.7030385136604309:   0%|          | 3/6470 [00:02<1:27:34,  1.23it/s]
3|4|Loss: 0.7030385136604309:   0%|          | 4/6470 [00:02<1:21:35,  1.32it/s]
3|5|Loss: 0.798353374004364:   0%|          | 4/6470 [00:03<1:21:35,  1.32it/s] 
3|5|Loss: 0.798353374004364:   0%|          | 5/6470 [00:03<1:26:06,  1.25it/s]
3|6|Loss: 0.7387989163398743:   0%|          | 5/6470 [00:04<1:26:06,  1.25it/s]
3|6|Loss: 0.7387989163398743:   0%|          | 6/6470 [00:04<1:28:07,  1.22it/s]
3|7|Loss: 0.6715021729469299:   0%|          | 6/6470 [00:05<1:28:07,  1.22it/s]
3|7|Loss: 0.6715021729469299:   0%|          | 7/6470 [00:05<1:19:47,  1.35it/s]
3|8|Loss: 0.8826126456260681:   0%|          | 7/6470 [00:05<1:19:47,  1.35it/s]
3|8|Loss: 0.8826126456260681:   0%|          | 8/6470 [00:05<1:18:31,  1.37it/s]
3|9|Loss: 0.8445627689361572:   0%|          | 8/6470 [00:06<1:18:31,  1.37it/s]
3|9|Loss: 0.8445627689361572:   0%|          | 9/6470 [00:06<1:18:49,  1.37it/s]
3|10|Loss: 0.7749910354614258:   0%|          | 9/6470 [00:07<1:18:49,  1.37it/s]
3|10|Loss: 0.7749910354614258:   0%|          | 10/6470 [00:07<1:19:45,  1.35it/s]
3|11|Loss: 0.7539274096488953:   0%|          | 10/6470 [00:07<1:19:45,  1.35it/s]
3|11|Loss: 0.7539274096488953:   0%|          | 11/6470 [00:08<1:16:32,  1.41it/s]
3|12|Loss: 0.9929946660995483:   0%|          | 11/6470 [00:08<1:16:32,  1.41it/s]
3|12|Loss: 0.9929946660995483:   0%|          | 12/6470 [00:08<1:12:04,  1.49it/s]
3|13|Loss: 0.7407724261283875:   0%|          | 12/6470 [00:08<1:12:04,  1.49it/s]
3|13|Loss: 0.7407724261283875:   0%|          | 13/6470 [00:09<1:05:58,  1.63it/s]
3|14|Loss: 0.6234472393989563:   0%|          | 13/6470 [00:09<1:05:58,  1.63it/s]
3|14|Loss: 0.6234472393989563:   0%|          | 14/6470 [00:09<1:07:34,  1.59it/s]
3|15|Loss: 0.7608880400657654:   0%|          | 14/6470 [00:10<1:07:34,  1.59it/s]
3|15|Loss: 0.7608880400657654:   0%|          | 15/6470 [00:10<1:18:58,  1.36it/s]
3|16|Loss: 0.8399603962898254:   0%|          | 15/6470 [00:11<1:18:58,  1.36it/s]
3|16|Loss: 0.8399603962898254:   0%|          | 16/6470 [00:11<1:21:07,  1.33it/s]
3|17|Loss: 0.8322595357894897:   0%|          | 16/6470 [00:11<1:21:07,  1.33it/s]
3|17|Loss: 0.8322595357894897:   0%|          | 17/6470 [00:12<1:16:58,  1.40it/s]
3|18|Loss: 0.8908623456954956:   0%|          | 17/6470 [00:12<1:16:58,  1.40it/s]
3|18|Loss: 0.8908623456954956:   0%|          | 18/6470 [00:12<1:16:26,  1.41it/s]
3|19|Loss: 0.7193889021873474:   0%|          | 18/6470 [00:13<1:16:26,  1.41it/s]
3|19|Loss: 0.7193889021873474:   0%|          | 19/6470 [00:13<1:17:49,  1.38it/s]
3|20|Loss: 0.8021298050880432:   0%|          | 19/6470 [00:14<1:17:49,  1.38it/s]
3|20|Loss: 0.8021298050880432:   0%|          | 20/6470 [00:14<1:18:50,  1.36it/s]
3|21|Loss: 0.7367125153541565:   0%|          | 20/6470 [00:14<1:18:50,  1.36it/s]
3|21|Loss: 0.7367125153541565:   0%|          | 21/6470 [00:15<1:15:19,  1.43it/s]
3|22|Loss: 0.8030802607536316:   0%|          | 21/6470 [00:15<1:15:19,  1.43it/s]
3|22|Loss: 0.8030802607536316:   0%|          | 22/6470 [00:15<1:16:41,  1.40it/s]
3|23|Loss: 0.840049147605896:   0%|          | 22/6470 [00:16<1:16:41,  1.40it/s] 
3|23|Loss: 0.840049147605896:   0%|          | 23/6470 [00:16<1:15:33,  1.42it/s]
3|24|Loss: 0.8858805298805237:   0%|          | 23/6470 [00:16<1:15:33,  1.42it/s]
3|24|Loss: 0.8858805298805237:   0%|          | 24/6470 [00:17<1:10:26,  1.53it/s]
3|25|Loss: 0.8556336164474487:   0%|          | 24/6470 [00:17<1:10:26,  1.53it/s]
3|25|Loss: 0.8556336164474487:   0%|          | 25/6470 [00:17<1:11:09,  1.51it/s]
3|26|Loss: 0.7638928890228271:   0%|          | 25/6470 [00:18<1:11:09,  1.51it/s]
3|26|Loss: 0.7638928890228271:   0%|          | 26/6470 [00:18<1:18:00,  1.38it/s]
3|27|Loss: 0.7737535834312439:   0%|          | 26/6470 [00:18<1:18:00,  1.38it/s]
3|27|Loss: 0.7737535834312439:   0%|          | 27/6470 [00:19<1:14:26,  1.44it/s]
3|28|Loss: 0.7806698083877563:   0%|          | 27/6470 [00:19<1:14:26,  1.44it/s]
3|28|Loss: 0.7806698083877563:   0%|          | 28/6470 [00:19<1:14:09,  1.45it/s]
3|29|Loss: 0.8993500471115112:   0%|          | 28/6470 [00:20<1:14:09,  1.45it/s]
3|29|Loss: 0.8993500471115112:   0%|          | 29/6470 [00:20<1:18:15,  1.37it/s]
3|30|Loss: 0.9117184281349182:   0%|          | 29/6470 [00:21<1:18:15,  1.37it/s]
3|30|Loss: 0.9117184281349182:   0%|          | 30/6470 [00:21<1:19:09,  1.36it/s]
3|31|Loss: 0.7754558324813843:   0%|          | 30/6470 [00:21<1:19:09,  1.36it/s]
3|31|Loss: 0.7754558324813843:   0%|          | 31/6470 [00:22<1:19:24,  1.35it/s]
3|32|Loss: 1.0841509103775024:   0%|          | 31/6470 [00:22<1:19:24,  1.35it/s]
3|32|Loss: 1.0841509103775024:   0%|          | 32/6470 [00:23<1:21:52,  1.31it/s]
3|33|Loss: 0.8370441794395447:   0%|          | 32/6470 [00:23<1:21:52,  1.31it/s]
3|33|Loss: 0.8370441794395447:   1%|          | 33/6470 [00:23<1:21:27,  1.32it/s]
3|34|Loss: 0.6956090927124023:   1%|          | 33/6470 [00:24<1:21:27,  1.32it/s]
3|34|Loss: 0.6956090927124023:   1%|          | 34/6470 [00:24<1:18:02,  1.37it/s]
3|35|Loss: 0.751445472240448:   1%|          | 34/6470 [00:24<1:18:02,  1.37it/s] 
3|35|Loss: 0.751445472240448:   1%|          | 35/6470 [00:25<1:20:05,  1.34it/s]
3|36|Loss: 0.6808604598045349:   1%|          | 35/6470 [00:25<1:20:05,  1.34it/s]
3|36|Loss: 0.6808604598045349:   1%|          | 36/6470 [00:26<1:23:20,  1.29it/s]
3|37|Loss: 0.8650848865509033:   1%|          | 36/6470 [00:26<1:23:20,  1.29it/s]
3|37|Loss: 0.8650848865509033:   1%|          | 37/6470 [00:26<1:25:49,  1.25it/s]
3|38|Loss: 1.1034321784973145:   1%|          | 37/6470 [00:27<1:25:49,  1.25it/s]
3|38|Loss: 1.1034321784973145:   1%|          | 38/6470 [00:27<1:32:50,  1.15it/s]
3|39|Loss: 0.853100597858429:   1%|          | 38/6470 [00:28<1:32:50,  1.15it/s] 
3|39|Loss: 0.853100597858429:   1%|          | 39/6470 [00:28<1:28:43,  1.21it/s]
3|40|Loss: 0.9392315149307251:   1%|          | 39/6470 [00:29<1:28:43,  1.21it/s]
3|40|Loss: 0.9392315149307251:   1%|          | 40/6470 [00:29<1:22:55,  1.29it/s]
3|41|Loss: 0.9881062507629395:   1%|          | 40/6470 [00:29<1:22:55,  1.29it/s]
3|41|Loss: 0.9881062507629395:   1%|          | 41/6470 [00:30<1:20:01,  1.34it/s]
3|42|Loss: 1.0366742610931396:   1%|          | 41/6470 [00:30<1:20:01,  1.34it/s]
3|42|Loss: 1.0366742610931396:   1%|          | 42/6470 [00:30<1:15:14,  1.42it/s]
3|43|Loss: 0.7979698777198792:   1%|          | 42/6470 [00:30<1:15:14,  1.42it/s]
3|43|Loss: 0.7979698777198792:   1%|          | 43/6470 [00:31<1:11:44,  1.49it/s]
3|44|Loss: 0.9091771841049194:   1%|          | 43/6470 [00:31<1:11:44,  1.49it/s]
3|44|Loss: 0.9091771841049194:   1%|          | 44/6470 [00:32<1:16:43,  1.40it/s]
3|45|Loss: 0.8130508661270142:   1%|          | 44/6470 [00:32<1:16:43,  1.40it/s]
3|45|Loss: 0.8130508661270142:   1%|          | 45/6470 [00:32<1:19:12,  1.35it/s]
3|46|Loss: 1.0034823417663574:   1%|          | 45/6470 [00:33<1:19:12,  1.35it/s]
3|46|Loss: 1.0034823417663574:   1%|          | 46/6470 [00:33<1:17:38,  1.38it/s]
3|47|Loss: 0.8051701188087463:   1%|          | 46/6470 [00:33<1:17:38,  1.38it/s]
3|47|Loss: 0.8051701188087463:   1%|          | 47/6470 [00:34<1:17:28,  1.38it/s]
3|48|Loss: 0.8283231258392334:   1%|          | 47/6470 [00:34<1:17:28,  1.38it/s]
3|48|Loss: 0.8283231258392334:   1%|          | 48/6470 [00:35<1:18:43,  1.36it/s]
3|49|Loss: 0.8863531351089478:   1%|          | 48/6470 [00:35<1:18:43,  1.36it/s]
3|49|Loss: 0.8863531351089478:   1%|          | 49/6470 [00:35<1:14:59,  1.43it/s]
3|50|Loss: 1.0620129108428955:   1%|          | 49/6470 [00:35<1:14:59,  1.43it/s]
3|50|Loss: 1.0620129108428955:   1%|          | 50/6470 [00:36<1:13:53,  1.45it/s]
3|51|Loss: 0.7549059987068176:   1%|          | 50/6470 [00:36<1:13:53,  1.45it/s]
3|51|Loss: 0.7549059987068176:   1%|          | 51/6470 [00:37<1:15:48,  1.41it/s]
3|52|Loss: 0.7696250081062317:   1%|          | 51/6470 [00:37<1:15:48,  1.41it/s]
3|52|Loss: 0.7696250081062317:   1%|          | 52/6470 [00:37<1:14:57,  1.43it/s]
3|53|Loss: 0.885429859161377:   1%|          | 52/6470 [00:38<1:14:57,  1.43it/s] 
3|53|Loss: 0.885429859161377:   1%|          | 53/6470 [00:38<1:13:28,  1.46it/s]
3|54|Loss: 0.8612255454063416:   1%|          | 53/6470 [00:38<1:13:28,  1.46it/s]
3|54|Loss: 0.8612255454063416:   1%|          | 54/6470 [00:39<1:16:52,  1.39it/s]
3|55|Loss: 0.8275097608566284:   1%|          | 54/6470 [00:39<1:16:52,  1.39it/s]
3|55|Loss: 0.8275097608566284:   1%|          | 55/6470 [00:39<1:19:21,  1.35it/s]
3|56|Loss: 0.6295861005783081:   1%|          | 55/6470 [00:40<1:19:21,  1.35it/s]
3|56|Loss: 0.6295861005783081:   1%|          | 56/6470 [00:40<1:27:03,  1.23it/s]
3|57|Loss: 0.9442988634109497:   1%|          | 56/6470 [00:41<1:27:03,  1.23it/s]
3|57|Loss: 0.9442988634109497:   1%|          | 57/6470 [00:41<1:24:12,  1.27it/s]
3|58|Loss: 0.8213768601417542:   1%|          | 57/6470 [00:42<1:24:12,  1.27it/s]
3|58|Loss: 0.8213768601417542:   1%|          | 58/6470 [00:42<1:22:23,  1.30it/s]
3|59|Loss: 0.8150431513786316:   1%|          | 58/6470 [00:42<1:22:23,  1.30it/s]
3|59|Loss: 0.8150431513786316:   1%|          | 59/6470 [00:43<1:24:37,  1.26it/s]
3|60|Loss: 0.7858339548110962:   1%|          | 59/6470 [00:43<1:24:37,  1.26it/s]
3|60|Loss: 0.7858339548110962:   1%|          | 60/6470 [00:43<1:22:10,  1.30it/s]
3|61|Loss: 0.9607700705528259:   1%|          | 60/6470 [00:44<1:22:10,  1.30it/s]
3|61|Loss: 0.9607700705528259:   1%|          | 61/6470 [00:44<1:19:09,  1.35it/s]
3|62|Loss: 0.7171803116798401:   1%|          | 61/6470 [00:45<1:19:09,  1.35it/s]
3|62|Loss: 0.7171803116798401:   1%|          | 62/6470 [00:45<1:14:00,  1.44it/s]
3|63|Loss: 0.8782430291175842:   1%|          | 62/6470 [00:45<1:14:00,  1.44it/s]
3|63|Loss: 0.8782430291175842:   1%|          | 63/6470 [00:45<1:11:56,  1.48it/s]
3|64|Loss: 0.8450504541397095:   1%|          | 63/6470 [00:46<1:11:56,  1.48it/s]
3|64|Loss: 0.8450504541397095:   1%|          | 64/6470 [00:46<1:14:30,  1.43it/s]
3|65|Loss: 0.7458487153053284:   1%|          | 64/6470 [00:47<1:14:30,  1.43it/s]
3|65|Loss: 0.7458487153053284:   1%|          | 65/6470 [00:47<1:20:13,  1.33it/s]
3|66|Loss: 0.8454258441925049:   1%|          | 65/6470 [00:47<1:20:13,  1.33it/s]
3|66|Loss: 0.8454258441925049:   1%|          | 66/6470 [00:48<1:17:52,  1.37it/s]
3|67|Loss: 0.8335288763046265:   1%|          | 66/6470 [00:48<1:17:52,  1.37it/s]
3|67|Loss: 0.8335288763046265:   1%|          | 67/6470 [00:48<1:14:13,  1.44it/s]
3|68|Loss: 0.8413758873939514:   1%|          | 67/6470 [00:49<1:14:13,  1.44it/s]
3|68|Loss: 0.8413758873939514:   1%|          | 68/6470 [00:49<1:15:32,  1.41it/s]
3|69|Loss: 0.8321111798286438:   1%|          | 68/6470 [00:49<1:15:32,  1.41it/s]
3|69|Loss: 0.8321111798286438:   1%|          | 69/6470 [00:50<1:17:52,  1.37it/s]
3|70|Loss: 0.8629730939865112:   1%|          | 69/6470 [00:50<1:17:52,  1.37it/s]
3|70|Loss: 0.8629730939865112:   1%|          | 70/6470 [00:51<1:20:03,  1.33it/s]
3|71|Loss: 0.73684161901474:   1%|          | 70/6470 [00:51<1:20:03,  1.33it/s]  
3|71|Loss: 0.73684161901474:   1%|          | 71/6470 [00:52<1:27:52,  1.21it/s]
3|72|Loss: 0.9878154993057251:   1%|          | 71/6470 [00:52<1:27:52,  1.21it/s]
3|72|Loss: 0.9878154993057251:   1%|          | 72/6470 [00:52<1:28:16,  1.21it/s]
3|73|Loss: 0.7014238834381104:   1%|          | 72/6470 [00:53<1:28:16,  1.21it/s]
3|73|Loss: 0.7014238834381104:   1%|          | 73/6470 [00:53<1:29:14,  1.19it/s]
3|74|Loss: 0.8874207139015198:   1%|          | 73/6470 [00:54<1:29:14,  1.19it/s]
3|74|Loss: 0.8874207139015198:   1%|          | 74/6470 [00:54<1:29:45,  1.19it/s]
3|75|Loss: 0.8019038438796997:   1%|          | 74/6470 [00:55<1:29:45,  1.19it/s]
3|75|Loss: 0.8019038438796997:   1%|          | 75/6470 [00:55<1:22:52,  1.29it/s]
3|76|Loss: 0.805811882019043:   1%|          | 75/6470 [00:55<1:22:52,  1.29it/s] 
3|76|Loss: 0.805811882019043:   1%|          | 76/6470 [00:56<1:22:13,  1.30it/s]
3|77|Loss: 0.7930628061294556:   1%|          | 76/6470 [00:56<1:22:13,  1.30it/s]
3|77|Loss: 0.7930628061294556:   1%|          | 77/6470 [00:56<1:20:34,  1.32it/s]
3|78|Loss: 0.9610128402709961:   1%|          | 77/6470 [00:57<1:20:34,  1.32it/s]
3|78|Loss: 0.9610128402709961:   1%|          | 78/6470 [00:57<1:20:56,  1.32it/s]
3|79|Loss: 0.7311147451400757:   1%|          | 78/6470 [00:57<1:20:56,  1.32it/s]
3|79|Loss: 0.7311147451400757:   1%|          | 79/6470 [00:58<1:13:44,  1.44it/s]
3|80|Loss: 0.780341625213623:   1%|          | 79/6470 [00:58<1:13:44,  1.44it/s] 
3|80|Loss: 0.780341625213623:   1%|          | 80/6470 [00:58<1:13:11,  1.46it/s]
3|81|Loss: 0.8780235052108765:   1%|          | 80/6470 [00:59<1:13:11,  1.46it/s]
3|81|Loss: 0.8780235052108765:   1%|▏         | 81/6470 [00:59<1:13:03,  1.46it/s]
3|82|Loss: 1.009673833847046:   1%|▏         | 81/6470 [00:59<1:13:03,  1.46it/s] 
3|82|Loss: 1.009673833847046:   1%|▏         | 82/6470 [01:00<1:16:59,  1.38it/s]
3|83|Loss: 0.9049285650253296:   1%|▏         | 82/6470 [01:00<1:16:59,  1.38it/s]
3|83|Loss: 0.9049285650253296:   1%|▏         | 83/6470 [01:01<1:22:29,  1.29it/s]
3|84|Loss: 0.8670983910560608:   1%|▏         | 83/6470 [01:01<1:22:29,  1.29it/s]
3|84|Loss: 0.8670983910560608:   1%|▏         | 84/6470 [01:01<1:21:21,  1.31it/s]
3|85|Loss: 0.8314460515975952:   1%|▏         | 84/6470 [01:02<1:21:21,  1.31it/s]
3|85|Loss: 0.8314460515975952:   1%|▏         | 85/6470 [01:02<1:18:16,  1.36it/s]
3|86|Loss: 0.7947663068771362:   1%|▏         | 85/6470 [01:02<1:18:16,  1.36it/s]
3|86|Loss: 0.7947663068771362:   1%|▏         | 86/6470 [01:03<1:21:19,  1.31it/s]
3|87|Loss: 0.803767204284668:   1%|▏         | 86/6470 [01:03<1:21:19,  1.31it/s] 
3|87|Loss: 0.803767204284668:   1%|▏         | 87/6470 [01:04<1:19:04,  1.35it/s]
3|88|Loss: 0.7544414401054382:   1%|▏         | 87/6470 [01:04<1:19:04,  1.35it/s]
3|88|Loss: 0.7544414401054382:   1%|▏         | 88/6470 [01:04<1:16:08,  1.40it/s]
3|89|Loss: 0.7437342405319214:   1%|▏         | 88/6470 [01:05<1:16:08,  1.40it/s]
3|89|Loss: 0.7437342405319214:   1%|▏         | 89/6470 [01:05<1:16:23,  1.39it/s]
3|90|Loss: 0.8156299591064453:   1%|▏         | 89/6470 [01:05<1:16:23,  1.39it/s]
3|90|Loss: 0.8156299591064453:   1%|▏         | 90/6470 [01:06<1:20:13,  1.33it/s]
3|91|Loss: 0.9130588173866272:   1%|▏         | 90/6470 [01:06<1:20:13,  1.33it/s]
3|91|Loss: 0.9130588173866272:   1%|▏         | 91/6470 [01:06<1:13:55,  1.44it/s]
3|92|Loss: 0.9358944296836853:   1%|▏         | 91/6470 [01:07<1:13:55,  1.44it/s]
3|92|Loss: 0.9358944296836853:   1%|▏         | 92/6470 [01:07<1:11:09,  1.49it/s]
3|93|Loss: 0.9679352045059204:   1%|▏         | 92/6470 [01:07<1:11:09,  1.49it/s]
3|93|Loss: 0.9679352045059204:   1%|▏         | 93/6470 [01:08<1:13:13,  1.45it/s]
3|94|Loss: 0.923577606678009:   1%|▏         | 93/6470 [01:08<1:13:13,  1.45it/s] 
3|94|Loss: 0.923577606678009:   1%|▏         | 94/6470 [01:08<1:14:43,  1.42it/s]
3|95|Loss: 0.819911777973175:   1%|▏         | 94/6470 [01:09<1:14:43,  1.42it/s]
3|95|Loss: 0.819911777973175:   1%|▏         | 95/6470 [01:09<1:19:09,  1.34it/s]
3|96|Loss: 0.7859963774681091:   1%|▏         | 95/6470 [01:10<1:19:09,  1.34it/s]
3|96|Loss: 0.7859963774681091:   1%|▏         | 96/6470 [01:10<1:21:07,  1.31it/s]
3|97|Loss: 0.8297663331031799:   1%|▏         | 96/6470 [01:10<1:21:07,  1.31it/s]
3|97|Loss: 0.8297663331031799:   1%|▏         | 97/6470 [01:11<1:18:56,  1.35it/s]
3|98|Loss: 0.9471533298492432:   1%|▏         | 97/6470 [01:11<1:18:56,  1.35it/s]
3|98|Loss: 0.9471533298492432:   2%|▏         | 98/6470 [01:12<1:21:24,  1.30it/s]
3|99|Loss: 0.771578848361969:   2%|▏         | 98/6470 [01:12<1:21:24,  1.30it/s] 
3|99|Loss: 0.771578848361969:   2%|▏         | 99/6470 [01:12<1:24:00,  1.26it/s]
3|100|Loss: 0.8859025835990906:   2%|▏         | 99/6470 [01:13<1:24:00,  1.26it/s]
3|100|Loss: 0.8859025835990906:   2%|▏         | 100/6470 [01:13<1:22:51,  1.28it/s]
3|101|Loss: 0.9282920956611633:   2%|▏         | 100/6470 [01:14<1:22:51,  1.28it/s]
3|101|Loss: 0.9282920956611633:   2%|▏         | 101/6470 [01:14<1:18:18,  1.36it/s]
3|102|Loss: 0.788805365562439:   2%|▏         | 101/6470 [01:14<1:18:18,  1.36it/s] 
3|102|Loss: 0.788805365562439:   2%|▏         | 102/6470 [01:14<1:14:09,  1.43it/s]
3|103|Loss: 0.8296996355056763:   2%|▏         | 102/6470 [01:15<1:14:09,  1.43it/s]
3|103|Loss: 0.8296996355056763:   2%|▏         | 103/6470 [01:15<1:17:27,  1.37it/s]
3|104|Loss: 0.8602967262268066:   2%|▏         | 103/6470 [01:16<1:17:27,  1.37it/s]
3|104|Loss: 0.8602967262268066:   2%|▏         | 104/6470 [01:16<1:19:38,  1.33it/s]
3|105|Loss: 0.795672595500946:   2%|▏         | 104/6470 [01:16<1:19:38,  1.33it/s] 
3|105|Loss: 0.795672595500946:   2%|▏         | 105/6470 [01:17<1:17:12,  1.37it/s]
3|106|Loss: 0.9327504634857178:   2%|▏         | 105/6470 [01:17<1:17:12,  1.37it/s]
3|106|Loss: 0.9327504634857178:   2%|▏         | 106/6470 [01:17<1:11:54,  1.47it/s]
3|107|Loss: 0.9220110774040222:   2%|▏         | 106/6470 [01:18<1:11:54,  1.47it/s]
3|107|Loss: 0.9220110774040222:   2%|▏         | 107/6470 [01:18<1:09:51,  1.52it/s]
3|108|Loss: 0.818427324295044:   2%|▏         | 107/6470 [01:18<1:09:51,  1.52it/s] 
3|108|Loss: 0.818427324295044:   2%|▏         | 108/6470 [01:19<1:12:44,  1.46it/s]
3|109|Loss: 0.9616791605949402:   2%|▏         | 108/6470 [01:19<1:12:44,  1.46it/s]
3|109|Loss: 0.9616791605949402:   2%|▏         | 109/6470 [01:19<1:14:56,  1.41it/s]
3|110|Loss: 0.8606399297714233:   2%|▏         | 109/6470 [01:20<1:14:56,  1.41it/s]
3|110|Loss: 0.8606399297714233:   2%|▏         | 110/6470 [01:20<1:10:37,  1.50it/s]
3|111|Loss: 0.8607084155082703:   2%|▏         | 110/6470 [01:20<1:10:37,  1.50it/s]
3|111|Loss: 0.8607084155082703:   2%|▏         | 111/6470 [01:21<1:09:25,  1.53it/s]
3|112|Loss: 0.8204052448272705:   2%|▏         | 111/6470 [01:21<1:09:25,  1.53it/s]
3|112|Loss: 0.8204052448272705:   2%|▏         | 112/6470 [01:21<1:12:54,  1.45it/s]
3|113|Loss: 0.8365901708602905:   2%|▏         | 112/6470 [01:22<1:12:54,  1.45it/s]
3|113|Loss: 0.8365901708602905:   2%|▏         | 113/6470 [01:22<1:12:59,  1.45it/s]
3|114|Loss: 0.8065200448036194:   2%|▏         | 113/6470 [01:22<1:12:59,  1.45it/s]
3|114|Loss: 0.8065200448036194:   2%|▏         | 114/6470 [01:23<1:15:04,  1.41it/s]
3|115|Loss: 0.6772595047950745:   2%|▏         | 114/6470 [01:23<1:15:04,  1.41it/s]
3|115|Loss: 0.6772595047950745:   2%|▏         | 115/6470 [01:24<1:15:09,  1.41it/s]
3|116|Loss: 0.9323774576187134:   2%|▏         | 115/6470 [01:24<1:15:09,  1.41it/s]
3|116|Loss: 0.9323774576187134:   2%|▏         | 116/6470 [01:24<1:14:46,  1.42it/s]
3|117|Loss: 0.8439406752586365:   2%|▏         | 116/6470 [01:25<1:14:46,  1.42it/s]
3|117|Loss: 0.8439406752586365:   2%|▏         | 117/6470 [01:25<1:12:37,  1.46it/s]
3|118|Loss: 0.8940544724464417:   2%|▏         | 117/6470 [01:25<1:12:37,  1.46it/s]
3|118|Loss: 0.8940544724464417:   2%|▏         | 118/6470 [01:25<1:11:20,  1.48it/s]
3|119|Loss: 0.8869361877441406:   2%|▏         | 118/6470 [01:26<1:11:20,  1.48it/s]
3|119|Loss: 0.8869361877441406:   2%|▏         | 119/6470 [01:26<1:17:28,  1.37it/s]
3|120|Loss: 0.8684127330780029:   2%|▏         | 119/6470 [01:27<1:17:28,  1.37it/s]
3|120|Loss: 0.8684127330780029:   2%|▏         | 120/6470 [01:27<1:20:29,  1.31it/s]
3|121|Loss: 0.7447669506072998:   2%|▏         | 120/6470 [01:28<1:20:29,  1.31it/s]
3|121|Loss: 0.7447669506072998:   2%|▏         | 121/6470 [01:28<1:21:08,  1.30it/s]
3|122|Loss: 0.7914719581604004:   2%|▏         | 121/6470 [01:28<1:21:08,  1.30it/s]
3|122|Loss: 0.7914719581604004:   2%|▏         | 122/6470 [01:29<1:20:20,  1.32it/s]
3|123|Loss: 0.8736634850502014:   2%|▏         | 122/6470 [01:29<1:20:20,  1.32it/s]
3|123|Loss: 0.8736634850502014:   2%|▏         | 123/6470 [01:29<1:18:21,  1.35it/s]
3|124|Loss: 0.824198305606842:   2%|▏         | 123/6470 [01:30<1:18:21,  1.35it/s] 
3|124|Loss: 0.824198305606842:   2%|▏         | 124/6470 [01:30<1:11:10,  1.49it/s]
3|125|Loss: 1.0489528179168701:   2%|▏         | 124/6470 [01:30<1:11:10,  1.49it/s]
3|125|Loss: 1.0489528179168701:   2%|▏         | 125/6470 [01:30<1:07:56,  1.56it/s]
3|126|Loss: 1.138306736946106:   2%|▏         | 125/6470 [01:31<1:07:56,  1.56it/s] 
3|126|Loss: 1.138306736946106:   2%|▏         | 126/6470 [01:31<1:11:52,  1.47it/s]
3|127|Loss: 0.7931352853775024:   2%|▏         | 126/6470 [01:32<1:11:52,  1.47it/s]
3|127|Loss: 0.7931352853775024:   2%|▏         | 127/6470 [01:32<1:09:36,  1.52it/s]
3|128|Loss: 0.8446370959281921:   2%|▏         | 127/6470 [01:32<1:09:36,  1.52it/s]
3|128|Loss: 0.8446370959281921:   2%|▏         | 128/6470 [01:33<1:17:10,  1.37it/s]
3|129|Loss: 0.8857645392417908:   2%|▏         | 128/6470 [01:33<1:17:10,  1.37it/s]
3|129|Loss: 0.8857645392417908:   2%|▏         | 129/6470 [01:34<1:19:14,  1.33it/s]
3|130|Loss: 0.7806317806243896:   2%|▏         | 129/6470 [01:34<1:19:14,  1.33it/s]
3|130|Loss: 0.7806317806243896:   2%|▏         | 130/6470 [01:34<1:24:55,  1.24it/s]
3|131|Loss: 0.913879930973053:   2%|▏         | 130/6470 [01:35<1:24:55,  1.24it/s] 
3|131|Loss: 0.913879930973053:   2%|▏         | 131/6470 [01:35<1:22:20,  1.28it/s]
3|132|Loss: 0.8495270013809204:   2%|▏         | 131/6470 [01:36<1:22:20,  1.28it/s]
3|132|Loss: 0.8495270013809204:   2%|▏         | 132/6470 [01:36<1:18:46,  1.34it/s]
3|133|Loss: 0.900781512260437:   2%|▏         | 132/6470 [01:36<1:18:46,  1.34it/s] 
3|133|Loss: 0.900781512260437:   2%|▏         | 133/6470 [01:36<1:11:32,  1.48it/s]
3|134|Loss: 0.7704594731330872:   2%|▏         | 133/6470 [01:37<1:11:32,  1.48it/s]
3|134|Loss: 0.7704594731330872:   2%|▏         | 134/6470 [01:37<1:06:38,  1.58it/s]
3|135|Loss: 0.8932316303253174:   2%|▏         | 134/6470 [01:37<1:06:38,  1.58it/s]
3|135|Loss: 0.8932316303253174:   2%|▏         | 135/6470 [01:38<1:07:38,  1.56it/s]
3|136|Loss: 0.6919588446617126:   2%|▏         | 135/6470 [01:38<1:07:38,  1.56it/s]
3|136|Loss: 0.6919588446617126:   2%|▏         | 136/6470 [01:38<1:09:54,  1.51it/s]
3|137|Loss: 0.7616137862205505:   2%|▏         | 136/6470 [01:39<1:09:54,  1.51it/s]
3|137|Loss: 0.7616137862205505:   2%|▏         | 137/6470 [01:39<1:16:57,  1.37it/s]
3|138|Loss: 0.7357352375984192:   2%|▏         | 137/6470 [01:40<1:16:57,  1.37it/s]
3|138|Loss: 0.7357352375984192:   2%|▏         | 138/6470 [01:40<1:21:20,  1.30it/s]
3|139|Loss: 0.844969630241394:   2%|▏         | 138/6470 [01:40<1:21:20,  1.30it/s] 
3|139|Loss: 0.844969630241394:   2%|▏         | 139/6470 [01:41<1:20:07,  1.32it/s]
3|140|Loss: 0.8666186928749084:   2%|▏         | 139/6470 [01:41<1:20:07,  1.32it/s]
3|140|Loss: 0.8666186928749084:   2%|▏         | 140/6470 [01:41<1:15:55,  1.39it/s]
3|141|Loss: 0.6959210634231567:   2%|▏         | 140/6470 [01:42<1:15:55,  1.39it/s]
3|141|Loss: 0.6959210634231567:   2%|▏         | 141/6470 [01:42<1:19:31,  1.33it/s]
3|142|Loss: 1.0636425018310547:   2%|▏         | 141/6470 [01:43<1:19:31,  1.33it/s]
3|142|Loss: 1.0636425018310547:   2%|▏         | 142/6470 [01:43<1:20:10,  1.32it/s]
3|143|Loss: 0.7565124034881592:   2%|▏         | 142/6470 [01:43<1:20:10,  1.32it/s]
3|143|Loss: 0.7565124034881592:   2%|▏         | 143/6470 [01:44<1:18:21,  1.35it/s]
3|144|Loss: 0.8265191316604614:   2%|▏         | 143/6470 [01:44<1:18:21,  1.35it/s]
3|144|Loss: 0.8265191316604614:   2%|▏         | 144/6470 [01:44<1:19:04,  1.33it/s]
3|145|Loss: 0.8486203551292419:   2%|▏         | 144/6470 [01:45<1:19:04,  1.33it/s]
3|145|Loss: 0.8486203551292419:   2%|▏         | 145/6470 [01:45<1:15:50,  1.39it/s]
3|146|Loss: 0.8120118975639343:   2%|▏         | 145/6470 [01:45<1:15:50,  1.39it/s]
3|146|Loss: 0.8120118975639343:   2%|▏         | 146/6470 [01:46<1:09:00,  1.53it/s]
3|147|Loss: 0.8038649559020996:   2%|▏         | 146/6470 [01:46<1:09:00,  1.53it/s]
3|147|Loss: 0.8038649559020996:   2%|▏         | 147/6470 [01:46<1:04:41,  1.63it/s]
3|148|Loss: 0.8129817247390747:   2%|▏         | 147/6470 [01:47<1:04:41,  1.63it/s]
3|148|Loss: 0.8129817247390747:   2%|▏         | 148/6470 [01:47<1:09:07,  1.52it/s]
3|149|Loss: 0.6898950934410095:   2%|▏         | 148/6470 [01:47<1:09:07,  1.52it/s]
3|149|Loss: 0.6898950934410095:   2%|▏         | 149/6470 [01:48<1:10:14,  1.50it/s]
3|150|Loss: 0.7196876406669617:   2%|▏         | 149/6470 [01:48<1:10:14,  1.50it/s]
3|150|Loss: 0.7196876406669617:   2%|▏         | 150/6470 [01:48<1:08:39,  1.53it/s]
3|151|Loss: 0.8667508363723755:   2%|▏         | 150/6470 [01:49<1:08:39,  1.53it/s]
3|151|Loss: 0.8667508363723755:   2%|▏         | 151/6470 [01:49<1:11:08,  1.48it/s]
3|152|Loss: 0.8225134015083313:   2%|▏         | 151/6470 [01:49<1:11:08,  1.48it/s]
3|152|Loss: 0.8225134015083313:   2%|▏         | 152/6470 [01:50<1:13:58,  1.42it/s]
3|153|Loss: 0.7995213866233826:   2%|▏         | 152/6470 [01:50<1:13:58,  1.42it/s]
3|153|Loss: 0.7995213866233826:   2%|▏         | 153/6470 [01:50<1:09:21,  1.52it/s]
3|154|Loss: 0.7668821811676025:   2%|▏         | 153/6470 [01:51<1:09:21,  1.52it/s]
3|154|Loss: 0.7668821811676025:   2%|▏         | 154/6470 [01:51<1:11:54,  1.46it/s]
3|155|Loss: 0.733049750328064:   2%|▏         | 154/6470 [01:51<1:11:54,  1.46it/s] 
3|155|Loss: 0.733049750328064:   2%|▏         | 155/6470 [01:52<1:07:46,  1.55it/s]
3|156|Loss: 0.6343668699264526:   2%|▏         | 155/6470 [01:52<1:07:46,  1.55it/s]
3|156|Loss: 0.6343668699264526:   2%|▏         | 156/6470 [01:52<1:13:09,  1.44it/s]
3|157|Loss: 0.7632855772972107:   2%|▏         | 156/6470 [01:53<1:13:09,  1.44it/s]
3|157|Loss: 0.7632855772972107:   2%|▏         | 157/6470 [01:53<1:09:08,  1.52it/s]
3|158|Loss: 0.6781790256500244:   2%|▏         | 157/6470 [01:53<1:09:08,  1.52it/s]
3|158|Loss: 0.6781790256500244:   2%|▏         | 158/6470 [01:54<1:06:42,  1.58it/s]
3|159|Loss: 0.8238884806632996:   2%|▏         | 158/6470 [01:54<1:06:42,  1.58it/s]
3|159|Loss: 0.8238884806632996:   2%|▏         | 159/6470 [01:54<1:10:25,  1.49it/s]
3|160|Loss: 0.8242505192756653:   2%|▏         | 159/6470 [01:55<1:10:25,  1.49it/s]
3|160|Loss: 0.8242505192756653:   2%|▏         | 160/6470 [01:55<1:14:12,  1.42it/s]
3|161|Loss: 0.8995104432106018:   2%|▏         | 160/6470 [01:56<1:14:12,  1.42it/s]
3|161|Loss: 0.8995104432106018:   2%|▏         | 161/6470 [01:56<1:18:53,  1.33it/s]
3|162|Loss: 0.7129035592079163:   2%|▏         | 161/6470 [01:56<1:18:53,  1.33it/s]
3|162|Loss: 0.7129035592079163:   3%|▎         | 162/6470 [01:57<1:26:55,  1.21it/s]
3|163|Loss: 0.9107515811920166:   3%|▎         | 162/6470 [01:57<1:26:55,  1.21it/s]
3|163|Loss: 0.9107515811920166:   3%|▎         | 163/6470 [01:58<1:23:06,  1.26it/s]
3|164|Loss: 0.858642578125:   3%|▎         | 163/6470 [01:58<1:23:06,  1.26it/s]    
3|164|Loss: 0.858642578125:   3%|▎         | 164/6470 [01:58<1:19:04,  1.33it/s]
3|165|Loss: 0.8821266293525696:   3%|▎         | 164/6470 [01:59<1:19:04,  1.33it/s]
3|165|Loss: 0.8821266293525696:   3%|▎         | 165/6470 [01:59<1:15:31,  1.39it/s]
3|166|Loss: 0.9405116438865662:   3%|▎         | 165/6470 [01:59<1:15:31,  1.39it/s]
3|166|Loss: 0.9405116438865662:   3%|▎         | 166/6470 [01:59<1:08:36,  1.53it/s]
3|167|Loss: 0.7374595403671265:   3%|▎         | 166/6470 [02:00<1:08:36,  1.53it/s]
3|167|Loss: 0.7374595403671265:   3%|▎         | 167/6470 [02:00<1:04:14,  1.64it/s]
3|168|Loss: 0.9289453029632568:   3%|▎         | 167/6470 [02:00<1:04:14,  1.64it/s]
3|168|Loss: 0.9289453029632568:   3%|▎         | 168/6470 [02:01<1:08:37,  1.53it/s]
3|169|Loss: 0.9915449023246765:   3%|▎         | 168/6470 [02:01<1:08:37,  1.53it/s]
3|169|Loss: 0.9915449023246765:   3%|▎         | 169/6470 [02:02<1:15:30,  1.39it/s]
3|170|Loss: 0.7422897815704346:   3%|▎         | 169/6470 [02:02<1:15:30,  1.39it/s]
3|170|Loss: 0.7422897815704346:   3%|▎         | 170/6470 [02:02<1:16:56,  1.36it/s]
3|171|Loss: 0.7694201469421387:   3%|▎         | 170/6470 [02:03<1:16:56,  1.36it/s]
3|171|Loss: 0.7694201469421387:   3%|▎         | 171/6470 [02:03<1:17:23,  1.36it/s]
3|172|Loss: 0.8179183006286621:   3%|▎         | 171/6470 [02:03<1:17:23,  1.36it/s]
3|172|Loss: 0.8179183006286621:   3%|▎         | 172/6470 [02:04<1:17:49,  1.35it/s]
3|173|Loss: 0.8862283229827881:   3%|▎         | 172/6470 [02:04<1:17:49,  1.35it/s]
3|173|Loss: 0.8862283229827881:   3%|▎         | 173/6470 [02:05<1:20:51,  1.30it/s]
3|174|Loss: 0.8233888149261475:   3%|▎         | 173/6470 [02:05<1:20:51,  1.30it/s]
3|174|Loss: 0.8233888149261475:   3%|▎         | 174/6470 [02:05<1:20:44,  1.30it/s]
3|175|Loss: 0.7889748215675354:   3%|▎         | 174/6470 [02:06<1:20:44,  1.30it/s]
3|175|Loss: 0.7889748215675354:   3%|▎         | 175/6470 [02:06<1:17:03,  1.36it/s]
3|176|Loss: 0.7432929277420044:   3%|▎         | 175/6470 [02:06<1:17:03,  1.36it/s]
3|176|Loss: 0.7432929277420044:   3%|▎         | 176/6470 [02:07<1:14:25,  1.41it/s]
3|177|Loss: 0.8233965039253235:   3%|▎         | 176/6470 [02:07<1:14:25,  1.41it/s]
3|177|Loss: 0.8233965039253235:   3%|▎         | 177/6470 [02:08<1:20:49,  1.30it/s]
3|178|Loss: 0.8456240296363831:   3%|▎         | 177/6470 [02:08<1:20:49,  1.30it/s]
3|178|Loss: 0.8456240296363831:   3%|▎         | 178/6470 [02:08<1:21:24,  1.29it/s]
3|179|Loss: 0.8699865341186523:   3%|▎         | 178/6470 [02:09<1:21:24,  1.29it/s]
3|179|Loss: 0.8699865341186523:   3%|▎         | 179/6470 [02:09<1:17:15,  1.36it/s]
3|180|Loss: 0.9042797088623047:   3%|▎         | 179/6470 [02:10<1:17:15,  1.36it/s]
3|180|Loss: 0.9042797088623047:   3%|▎         | 180/6470 [02:10<1:20:49,  1.30it/s]
3|181|Loss: 0.7747021913528442:   3%|▎         | 180/6470 [02:10<1:20:49,  1.30it/s]
3|181|Loss: 0.7747021913528442:   3%|▎         | 181/6470 [02:11<1:19:51,  1.31it/s]
3|182|Loss: 0.9989018440246582:   3%|▎         | 181/6470 [02:11<1:19:51,  1.31it/s]
3|182|Loss: 0.9989018440246582:   3%|▎         | 182/6470 [02:12<1:23:06,  1.26it/s]
3|183|Loss: 0.6829013824462891:   3%|▎         | 182/6470 [02:12<1:23:06,  1.26it/s]
3|183|Loss: 0.6829013824462891:   3%|▎         | 183/6470 [02:12<1:18:29,  1.33it/s]
3|184|Loss: 0.8936865329742432:   3%|▎         | 183/6470 [02:13<1:18:29,  1.33it/s]
3|184|Loss: 0.8936865329742432:   3%|▎         | 184/6470 [02:13<1:22:11,  1.27it/s]
3|185|Loss: 0.8723674416542053:   3%|▎         | 184/6470 [02:14<1:22:11,  1.27it/s]
3|185|Loss: 0.8723674416542053:   3%|▎         | 185/6470 [02:14<1:23:21,  1.26it/s]
3|186|Loss: 0.8624440431594849:   3%|▎         | 185/6470 [02:14<1:23:21,  1.26it/s]
3|186|Loss: 0.8624440431594849:   3%|▎         | 186/6470 [02:15<1:23:29,  1.25it/s]
3|187|Loss: 0.8063324689865112:   3%|▎         | 186/6470 [02:15<1:23:29,  1.25it/s]
3|187|Loss: 0.8063324689865112:   3%|▎         | 187/6470 [02:15<1:20:42,  1.30it/s]
3|188|Loss: 0.8364004492759705:   3%|▎         | 187/6470 [02:16<1:20:42,  1.30it/s]
3|188|Loss: 0.8364004492759705:   3%|▎         | 188/6470 [02:16<1:19:46,  1.31it/s]
3|189|Loss: 0.7470242977142334:   3%|▎         | 188/6470 [02:17<1:19:46,  1.31it/s]
3|189|Loss: 0.7470242977142334:   3%|▎         | 189/6470 [02:17<1:21:59,  1.28it/s]
3|190|Loss: 0.8270200490951538:   3%|▎         | 189/6470 [02:17<1:21:59,  1.28it/s]
3|190|Loss: 0.8270200490951538:   3%|▎         | 190/6470 [02:18<1:23:05,  1.26it/s]
3|191|Loss: 0.7582218647003174:   3%|▎         | 190/6470 [02:18<1:23:05,  1.26it/s]
3|191|Loss: 0.7582218647003174:   3%|▎         | 191/6470 [02:19<1:25:43,  1.22it/s]
3|192|Loss: 0.7798344492912292:   3%|▎         | 191/6470 [02:19<1:25:43,  1.22it/s]
3|192|Loss: 0.7798344492912292:   3%|▎         | 192/6470 [02:19<1:20:23,  1.30it/s]
3|193|Loss: 0.8332168459892273:   3%|▎         | 192/6470 [02:20<1:20:23,  1.30it/s]
3|193|Loss: 0.8332168459892273:   3%|▎         | 193/6470 [02:20<1:19:26,  1.32it/s]
3|194|Loss: 0.6033502817153931:   3%|▎         | 193/6470 [02:20<1:19:26,  1.32it/s]
3|194|Loss: 0.6033502817153931:   3%|▎         | 194/6470 [02:20<1:09:23,  1.51it/s]
3|195|Loss: 0.8646745681762695:   3%|▎         | 194/6470 [02:21<1:09:23,  1.51it/s]
3|195|Loss: 0.8646745681762695:   3%|▎         | 195/6470 [02:21<1:12:59,  1.43it/s]
3|196|Loss: 0.8595051169395447:   3%|▎         | 195/6470 [02:22<1:12:59,  1.43it/s]
3|196|Loss: 0.8595051169395447:   3%|▎         | 196/6470 [02:22<1:14:21,  1.41it/s]
3|197|Loss: 0.8725030422210693:   3%|▎         | 196/6470 [02:22<1:14:21,  1.41it/s]
3|197|Loss: 0.8725030422210693:   3%|▎         | 197/6470 [02:23<1:13:28,  1.42it/s]
3|198|Loss: 0.8104104995727539:   3%|▎         | 197/6470 [02:23<1:13:28,  1.42it/s]
3|198|Loss: 0.8104104995727539:   3%|▎         | 198/6470 [02:24<1:19:09,  1.32it/s]
3|199|Loss: 0.9629836082458496:   3%|▎         | 198/6470 [02:24<1:19:09,  1.32it/s]
3|199|Loss: 0.9629836082458496:   3%|▎         | 199/6470 [02:25<1:25:00,  1.23it/s]
3|200|Loss: 0.8754509687423706:   3%|▎         | 199/6470 [02:25<1:25:00,  1.23it/s]
3|200|Loss: 0.8754509687423706:   3%|▎         | 200/6470 [02:25<1:24:35,  1.24it/s]
3|201|Loss: 0.7603990435600281:   3%|▎         | 200/6470 [02:26<1:24:35,  1.24it/s]
3|201|Loss: 0.7603990435600281:   3%|▎         | 201/6470 [02:26<1:21:40,  1.28it/s]
3|202|Loss: 0.9719836711883545:   3%|▎         | 201/6470 [02:26<1:21:40,  1.28it/s]
3|202|Loss: 0.9719836711883545:   3%|▎         | 202/6470 [02:27<1:18:43,  1.33it/s]
3|203|Loss: 0.8444421291351318:   3%|▎         | 202/6470 [02:27<1:18:43,  1.33it/s]
3|203|Loss: 0.8444421291351318:   3%|▎         | 203/6470 [02:27<1:13:01,  1.43it/s]
3|204|Loss: 0.8481795787811279:   3%|▎         | 203/6470 [02:28<1:13:01,  1.43it/s]
3|204|Loss: 0.8481795787811279:   3%|▎         | 204/6470 [02:28<1:14:16,  1.41it/s]
3|205|Loss: 0.9995843768119812:   3%|▎         | 204/6470 [02:28<1:14:16,  1.41it/s]
3|205|Loss: 0.9995843768119812:   3%|▎         | 205/6470 [02:29<1:16:18,  1.37it/s]
3|206|Loss: 0.8373267650604248:   3%|▎         | 205/6470 [02:29<1:16:18,  1.37it/s]
3|206|Loss: 0.8373267650604248:   3%|▎         | 206/6470 [02:30<1:15:01,  1.39it/s]
3|207|Loss: 0.90679532289505:   3%|▎         | 206/6470 [02:30<1:15:01,  1.39it/s]  
3|207|Loss: 0.90679532289505:   3%|▎         | 207/6470 [02:30<1:20:05,  1.30it/s]
3|208|Loss: 0.8787016272544861:   3%|▎         | 207/6470 [02:31<1:20:05,  1.30it/s]
3|208|Loss: 0.8787016272544861:   3%|▎         | 208/6470 [02:31<1:25:09,  1.23it/s]
3|209|Loss: 0.8405009508132935:   3%|▎         | 208/6470 [02:32<1:25:09,  1.23it/s]
3|209|Loss: 0.8405009508132935:   3%|▎         | 209/6470 [02:32<1:22:05,  1.27it/s]
3|210|Loss: 0.8386731743812561:   3%|▎         | 209/6470 [02:32<1:22:05,  1.27it/s]
3|210|Loss: 0.8386731743812561:   3%|▎         | 210/6470 [02:33<1:21:29,  1.28it/s]
3|211|Loss: 0.7163494229316711:   3%|▎         | 210/6470 [02:33<1:21:29,  1.28it/s]
3|211|Loss: 0.7163494229316711:   3%|▎         | 211/6470 [02:33<1:16:10,  1.37it/s]
3|212|Loss: 0.9515489935874939:   3%|▎         | 211/6470 [02:34<1:16:10,  1.37it/s]
3|212|Loss: 0.9515489935874939:   3%|▎         | 212/6470 [02:34<1:14:15,  1.40it/s]
3|213|Loss: 0.953533947467804:   3%|▎         | 212/6470 [02:35<1:14:15,  1.40it/s] 
3|213|Loss: 0.953533947467804:   3%|▎         | 213/6470 [02:35<1:18:20,  1.33it/s]
3|214|Loss: 0.9308658838272095:   3%|▎         | 213/6470 [02:35<1:18:20,  1.33it/s]
3|214|Loss: 0.9308658838272095:   3%|▎         | 214/6470 [02:36<1:19:13,  1.32it/s]
3|215|Loss: 0.886131227016449:   3%|▎         | 214/6470 [02:36<1:19:13,  1.32it/s] 
3|215|Loss: 0.886131227016449:   3%|▎         | 215/6470 [02:36<1:13:42,  1.41it/s]
3|216|Loss: 0.7979469299316406:   3%|▎         | 215/6470 [02:37<1:13:42,  1.41it/s]
3|216|Loss: 0.7979469299316406:   3%|▎         | 216/6470 [02:37<1:11:26,  1.46it/s]
3|217|Loss: 0.7479047775268555:   3%|▎         | 216/6470 [02:37<1:11:26,  1.46it/s]
3|217|Loss: 0.7479047775268555:   3%|▎         | 217/6470 [02:38<1:10:05,  1.49it/s]
3|218|Loss: 0.8351378440856934:   3%|▎         | 217/6470 [02:38<1:10:05,  1.49it/s]
3|218|Loss: 0.8351378440856934:   3%|▎         | 218/6470 [02:38<1:12:27,  1.44it/s]
3|219|Loss: 0.8600651621818542:   3%|▎         | 218/6470 [02:39<1:12:27,  1.44it/s]
3|219|Loss: 0.8600651621818542:   3%|▎         | 219/6470 [02:39<1:17:49,  1.34it/s]
3|220|Loss: 0.8888174891471863:   3%|▎         | 219/6470 [02:40<1:17:49,  1.34it/s]
3|220|Loss: 0.8888174891471863:   3%|▎         | 220/6470 [02:40<1:18:58,  1.32it/s]
3|221|Loss: 0.7390674948692322:   3%|▎         | 220/6470 [02:40<1:18:58,  1.32it/s]
3|221|Loss: 0.7390674948692322:   3%|▎         | 221/6470 [02:41<1:19:43,  1.31it/s]
3|222|Loss: 1.0213840007781982:   3%|▎         | 221/6470 [02:41<1:19:43,  1.31it/s]
3|222|Loss: 1.0213840007781982:   3%|▎         | 222/6470 [02:41<1:17:10,  1.35it/s]
3|223|Loss: 0.8317659497261047:   3%|▎         | 222/6470 [02:42<1:17:10,  1.35it/s]
3|223|Loss: 0.8317659497261047:   3%|▎         | 223/6470 [02:42<1:20:26,  1.29it/s]
3|224|Loss: 0.8404130935668945:   3%|▎         | 223/6470 [02:43<1:20:26,  1.29it/s]
3|224|Loss: 0.8404130935668945:   3%|▎         | 224/6470 [02:43<1:24:51,  1.23it/s]
3|225|Loss: 0.8060556650161743:   3%|▎         | 224/6470 [02:44<1:24:51,  1.23it/s]
3|225|Loss: 0.8060556650161743:   3%|▎         | 225/6470 [02:44<1:22:29,  1.26it/s]
3|226|Loss: 0.7297594547271729:   3%|▎         | 225/6470 [02:44<1:22:29,  1.26it/s]
3|226|Loss: 0.7297594547271729:   3%|▎         | 226/6470 [02:45<1:15:25,  1.38it/s]
3|227|Loss: 0.7962602972984314:   3%|▎         | 226/6470 [02:45<1:15:25,  1.38it/s]
3|227|Loss: 0.7962602972984314:   4%|▎         | 227/6470 [02:45<1:11:34,  1.45it/s]
3|228|Loss: 0.7664112448692322:   4%|▎         | 227/6470 [02:46<1:11:34,  1.45it/s]
3|228|Loss: 0.7664112448692322:   4%|▎         | 228/6470 [02:46<1:17:53,  1.34it/s]
3|229|Loss: 0.8879331350326538:   4%|▎         | 228/6470 [02:46<1:17:53,  1.34it/s]
3|229|Loss: 0.8879331350326538:   4%|▎         | 229/6470 [02:47<1:19:17,  1.31it/s]
3|230|Loss: 0.8893656730651855:   4%|▎         | 229/6470 [02:47<1:19:17,  1.31it/s]
3|230|Loss: 0.8893656730651855:   4%|▎         | 230/6470 [02:48<1:19:03,  1.32it/s]
3|231|Loss: 0.8349530100822449:   4%|▎         | 230/6470 [02:48<1:19:03,  1.32it/s]
3|231|Loss: 0.8349530100822449:   4%|▎         | 231/6470 [02:48<1:13:32,  1.41it/s]
3|232|Loss: 0.7860023975372314:   4%|▎         | 231/6470 [02:49<1:13:32,  1.41it/s]
3|232|Loss: 0.7860023975372314:   4%|▎         | 232/6470 [02:49<1:15:56,  1.37it/s]
3|233|Loss: 0.8534865379333496:   4%|▎         | 232/6470 [02:49<1:15:56,  1.37it/s]
3|233|Loss: 0.8534865379333496:   4%|▎         | 233/6470 [02:50<1:18:03,  1.33it/s]
3|234|Loss: 0.7802731394767761:   4%|▎         | 233/6470 [02:50<1:18:03,  1.33it/s]
3|234|Loss: 0.7802731394767761:   4%|▎         | 234/6470 [02:51<1:19:06,  1.31it/s]
3|235|Loss: 0.788838803768158:   4%|▎         | 234/6470 [02:51<1:19:06,  1.31it/s] 
3|235|Loss: 0.788838803768158:   4%|▎         | 235/6470 [02:51<1:20:04,  1.30it/s]
3|236|Loss: 0.9229735136032104:   4%|▎         | 235/6470 [02:52<1:20:04,  1.30it/s]
3|236|Loss: 0.9229735136032104:   4%|▎         | 236/6470 [02:52<1:11:05,  1.46it/s]
3|237|Loss: 0.7589446306228638:   4%|▎         | 236/6470 [02:52<1:11:05,  1.46it/s]
3|237|Loss: 0.7589446306228638:   4%|▎         | 237/6470 [02:52<1:10:51,  1.47it/s]
3|238|Loss: 0.8717176914215088:   4%|▎         | 237/6470 [02:53<1:10:51,  1.47it/s]
3|238|Loss: 0.8717176914215088:   4%|▎         | 238/6470 [02:53<1:10:57,  1.46it/s]
3|239|Loss: 0.9373576641082764:   4%|▎         | 238/6470 [02:54<1:10:57,  1.46it/s]
3|239|Loss: 0.9373576641082764:   4%|▎         | 239/6470 [02:54<1:14:02,  1.40it/s]
3|240|Loss: 0.740959107875824:   4%|▎         | 239/6470 [02:54<1:14:02,  1.40it/s] 
3|240|Loss: 0.740959107875824:   4%|▎         | 240/6470 [02:55<1:13:45,  1.41it/s]
3|241|Loss: 0.8437426090240479:   4%|▎         | 240/6470 [02:55<1:13:45,  1.41it/s]
3|241|Loss: 0.8437426090240479:   4%|▎         | 241/6470 [02:55<1:15:57,  1.37it/s]
3|242|Loss: 0.8684781193733215:   4%|▎         | 241/6470 [02:56<1:15:57,  1.37it/s]
3|242|Loss: 0.8684781193733215:   4%|▎         | 242/6470 [02:56<1:17:32,  1.34it/s]
3|243|Loss: 0.8945457935333252:   4%|▎         | 242/6470 [02:57<1:17:32,  1.34it/s]
3|243|Loss: 0.8945457935333252:   4%|▍         | 243/6470 [02:57<1:17:42,  1.34it/s]
3|244|Loss: 0.9306726455688477:   4%|▍         | 243/6470 [02:57<1:17:42,  1.34it/s]
3|244|Loss: 0.9306726455688477:   4%|▍         | 244/6470 [02:58<1:17:37,  1.34it/s]
3|245|Loss: 0.6580749750137329:   4%|▍         | 244/6470 [02:58<1:17:37,  1.34it/s]
3|245|Loss: 0.6580749750137329:   4%|▍         | 245/6470 [02:59<1:20:03,  1.30it/s]
3|246|Loss: 0.759425938129425:   4%|▍         | 245/6470 [02:59<1:20:03,  1.30it/s] 
3|246|Loss: 0.759425938129425:   4%|▍         | 246/6470 [02:59<1:22:15,  1.26it/s]
3|247|Loss: 0.7485709190368652:   4%|▍         | 246/6470 [03:00<1:22:15,  1.26it/s]
3|247|Loss: 0.7485709190368652:   4%|▍         | 247/6470 [03:00<1:25:58,  1.21it/s]
3|248|Loss: 0.7886641621589661:   4%|▍         | 247/6470 [03:01<1:25:58,  1.21it/s]
3|248|Loss: 0.7886641621589661:   4%|▍         | 248/6470 [03:01<1:24:55,  1.22it/s]
3|249|Loss: 0.6939244270324707:   4%|▍         | 248/6470 [03:01<1:24:55,  1.22it/s]
3|249|Loss: 0.6939244270324707:   4%|▍         | 249/6470 [03:02<1:20:54,  1.28it/s]
3|250|Loss: 0.9542537331581116:   4%|▍         | 249/6470 [03:02<1:20:54,  1.28it/s]
3|250|Loss: 0.9542537331581116:   4%|▍         | 250/6470 [03:03<1:20:45,  1.28it/s]
3|251|Loss: 0.753970205783844:   4%|▍         | 250/6470 [03:03<1:20:45,  1.28it/s] 
3|251|Loss: 0.753970205783844:   4%|▍         | 251/6470 [03:03<1:21:45,  1.27it/s]
3|252|Loss: 0.9187954664230347:   4%|▍         | 251/6470 [03:04<1:21:45,  1.27it/s]
3|252|Loss: 0.9187954664230347:   4%|▍         | 252/6470 [03:04<1:15:54,  1.37it/s]
3|253|Loss: 0.910575270652771:   4%|▍         | 252/6470 [03:04<1:15:54,  1.37it/s] 
3|253|Loss: 0.910575270652771:   4%|▍         | 253/6470 [03:04<1:09:46,  1.49it/s]
3|254|Loss: 0.8585745096206665:   4%|▍         | 253/6470 [03:05<1:09:46,  1.49it/s]
3|254|Loss: 0.8585745096206665:   4%|▍         | 254/6470 [03:05<1:12:13,  1.43it/s]
3|255|Loss: 0.7343077659606934:   4%|▍         | 254/6470 [03:06<1:12:13,  1.43it/s]
3|255|Loss: 0.7343077659606934:   4%|▍         | 255/6470 [03:06<1:12:51,  1.42it/s]
3|256|Loss: 0.9591657519340515:   4%|▍         | 255/6470 [03:06<1:12:51,  1.42it/s]
3|256|Loss: 0.9591657519340515:   4%|▍         | 256/6470 [03:07<1:14:14,  1.39it/s]
3|257|Loss: 0.8024871349334717:   4%|▍         | 256/6470 [03:07<1:14:14,  1.39it/s]
3|257|Loss: 0.8024871349334717:   4%|▍         | 257/6470 [03:07<1:14:07,  1.40it/s]
3|258|Loss: 0.9697293043136597:   4%|▍         | 257/6470 [03:08<1:14:07,  1.40it/s]
3|258|Loss: 0.9697293043136597:   4%|▍         | 258/6470 [03:08<1:19:08,  1.31it/s]
3|259|Loss: 0.8458955883979797:   4%|▍         | 258/6470 [03:09<1:19:08,  1.31it/s]
3|259|Loss: 0.8458955883979797:   4%|▍         | 259/6470 [03:09<1:16:49,  1.35it/s]
3|260|Loss: 0.793196976184845:   4%|▍         | 259/6470 [03:09<1:16:49,  1.35it/s] 
3|260|Loss: 0.793196976184845:   4%|▍         | 260/6470 [03:10<1:17:35,  1.33it/s]
3|261|Loss: 0.6727228760719299:   4%|▍         | 260/6470 [03:10<1:17:35,  1.33it/s]
3|261|Loss: 0.6727228760719299:   4%|▍         | 261/6470 [03:11<1:21:42,  1.27it/s]
3|262|Loss: 0.8234193921089172:   4%|▍         | 261/6470 [03:11<1:21:42,  1.27it/s]
3|262|Loss: 0.8234193921089172:   4%|▍         | 262/6470 [03:11<1:16:11,  1.36it/s]
3|263|Loss: 0.8679426312446594:   4%|▍         | 262/6470 [03:12<1:16:11,  1.36it/s]
3|263|Loss: 0.8679426312446594:   4%|▍         | 263/6470 [03:12<1:14:56,  1.38it/s]
3|264|Loss: 0.9114137291908264:   4%|▍         | 263/6470 [03:12<1:14:56,  1.38it/s]
3|264|Loss: 0.9114137291908264:   4%|▍         | 264/6470 [03:13<1:16:16,  1.36it/s]
3|265|Loss: 0.9979492425918579:   4%|▍         | 264/6470 [03:13<1:16:16,  1.36it/s]
3|265|Loss: 0.9979492425918579:   4%|▍         | 265/6470 [03:13<1:17:48,  1.33it/s]
3|266|Loss: 0.8087145686149597:   4%|▍         | 265/6470 [03:14<1:17:48,  1.33it/s]
3|266|Loss: 0.8087145686149597:   4%|▍         | 266/6470 [03:14<1:19:11,  1.31it/s]
3|267|Loss: 0.8405166864395142:   4%|▍         | 266/6470 [03:15<1:19:11,  1.31it/s]
3|267|Loss: 0.8405166864395142:   4%|▍         | 267/6470 [03:15<1:16:42,  1.35it/s]
3|268|Loss: 0.8457823991775513:   4%|▍         | 267/6470 [03:15<1:16:42,  1.35it/s]
3|268|Loss: 0.8457823991775513:   4%|▍         | 268/6470 [03:16<1:18:50,  1.31it/s]
3|269|Loss: 0.7755998373031616:   4%|▍         | 268/6470 [03:16<1:18:50,  1.31it/s]
3|269|Loss: 0.7755998373031616:   4%|▍         | 269/6470 [03:16<1:16:51,  1.34it/s]
3|270|Loss: 0.8234983086585999:   4%|▍         | 269/6470 [03:17<1:16:51,  1.34it/s]
3|270|Loss: 0.8234983086585999:   4%|▍         | 270/6470 [03:17<1:14:55,  1.38it/s]
3|271|Loss: 0.7504702210426331:   4%|▍         | 270/6470 [03:18<1:14:55,  1.38it/s]
3|271|Loss: 0.7504702210426331:   4%|▍         | 271/6470 [03:18<1:16:09,  1.36it/s]
3|272|Loss: 0.8618771433830261:   4%|▍         | 271/6470 [03:18<1:16:09,  1.36it/s]
3|272|Loss: 0.8618771433830261:   4%|▍         | 272/6470 [03:19<1:16:41,  1.35it/s]
3|273|Loss: 0.7295030951499939:   4%|▍         | 272/6470 [03:19<1:16:41,  1.35it/s]
3|273|Loss: 0.7295030951499939:   4%|▍         | 273/6470 [03:19<1:15:41,  1.36it/s]
3|274|Loss: 0.9112849235534668:   4%|▍         | 273/6470 [03:20<1:15:41,  1.36it/s]
3|274|Loss: 0.9112849235534668:   4%|▍         | 274/6470 [03:20<1:14:41,  1.38it/s]
3|275|Loss: 0.8440751433372498:   4%|▍         | 274/6470 [03:20<1:14:41,  1.38it/s]
3|275|Loss: 0.8440751433372498:   4%|▍         | 275/6470 [03:21<1:13:50,  1.40it/s]
3|276|Loss: 0.7930155992507935:   4%|▍         | 275/6470 [03:21<1:13:50,  1.40it/s]
3|276|Loss: 0.7930155992507935:   4%|▍         | 276/6470 [03:21<1:12:17,  1.43it/s]
3|277|Loss: 0.9019289016723633:   4%|▍         | 276/6470 [03:22<1:12:17,  1.43it/s]
3|277|Loss: 0.9019289016723633:   4%|▍         | 277/6470 [03:22<1:11:05,  1.45it/s]
3|278|Loss: 0.87810879945755:   4%|▍         | 277/6470 [03:22<1:11:05,  1.45it/s]  
3|278|Loss: 0.87810879945755:   4%|▍         | 278/6470 [03:23<1:10:16,  1.47it/s]
3|279|Loss: 0.7612814903259277:   4%|▍         | 278/6470 [03:23<1:10:16,  1.47it/s]
3|279|Loss: 0.7612814903259277:   4%|▍         | 279/6470 [03:23<1:06:31,  1.55it/s]
3|280|Loss: 0.9442524313926697:   4%|▍         | 279/6470 [03:24<1:06:31,  1.55it/s]
3|280|Loss: 0.9442524313926697:   4%|▍         | 280/6470 [03:24<1:09:29,  1.48it/s]
3|281|Loss: 0.9342855215072632:   4%|▍         | 280/6470 [03:25<1:09:29,  1.48it/s]
3|281|Loss: 0.9342855215072632:   4%|▍         | 281/6470 [03:25<1:12:26,  1.42it/s]
3|282|Loss: 0.7738377451896667:   4%|▍         | 281/6470 [03:25<1:12:26,  1.42it/s]
3|282|Loss: 0.7738377451896667:   4%|▍         | 282/6470 [03:25<1:08:48,  1.50it/s]
3|283|Loss: 0.9407918453216553:   4%|▍         | 282/6470 [03:26<1:08:48,  1.50it/s]
3|283|Loss: 0.9407918453216553:   4%|▍         | 283/6470 [03:26<1:06:38,  1.55it/s]
3|284|Loss: 0.8576889634132385:   4%|▍         | 283/6470 [03:26<1:06:38,  1.55it/s]
3|284|Loss: 0.8576889634132385:   4%|▍         | 284/6470 [03:27<1:07:28,  1.53it/s]
3|285|Loss: 0.8951409459114075:   4%|▍         | 284/6470 [03:27<1:07:28,  1.53it/s]
3|285|Loss: 0.8951409459114075:   4%|▍         | 285/6470 [03:28<1:15:48,  1.36it/s]
3|286|Loss: 0.5462101101875305:   4%|▍         | 285/6470 [03:28<1:15:48,  1.36it/s]
3|286|Loss: 0.5462101101875305:   4%|▍         | 286/6470 [03:29<1:25:49,  1.20it/s]
3|287|Loss: 0.8887419104576111:   4%|▍         | 286/6470 [03:29<1:25:49,  1.20it/s]
3|287|Loss: 0.8887419104576111:   4%|▍         | 287/6470 [03:30<1:26:33,  1.19it/s]
3|288|Loss: 0.7579554915428162:   4%|▍         | 287/6470 [03:30<1:26:33,  1.19it/s]
3|288|Loss: 0.7579554915428162:   4%|▍         | 288/6470 [03:30<1:24:37,  1.22it/s]
3|289|Loss: 0.972100555896759:   4%|▍         | 288/6470 [03:31<1:24:37,  1.22it/s] 
3|289|Loss: 0.972100555896759:   4%|▍         | 289/6470 [03:31<1:20:30,  1.28it/s]
3|290|Loss: 0.8518509268760681:   4%|▍         | 289/6470 [03:31<1:20:30,  1.28it/s]
3|290|Loss: 0.8518509268760681:   4%|▍         | 290/6470 [03:32<1:16:10,  1.35it/s]
3|291|Loss: 0.7160825133323669:   4%|▍         | 290/6470 [03:32<1:16:10,  1.35it/s]
3|291|Loss: 0.7160825133323669:   4%|▍         | 291/6470 [03:32<1:08:56,  1.49it/s]
3|292|Loss: 0.8963402509689331:   4%|▍         | 291/6470 [03:32<1:08:56,  1.49it/s]
3|292|Loss: 0.8963402509689331:   5%|▍         | 292/6470 [03:33<1:05:00,  1.58it/s]
3|293|Loss: 0.8747769594192505:   5%|▍         | 292/6470 [03:33<1:05:00,  1.58it/s]
3|293|Loss: 0.8747769594192505:   5%|▍         | 293/6470 [03:33<1:08:19,  1.51it/s]
3|294|Loss: 0.8552270531654358:   5%|▍         | 293/6470 [03:34<1:08:19,  1.51it/s]
3|294|Loss: 0.8552270531654358:   5%|▍         | 294/6470 [03:34<1:15:15,  1.37it/s]
3|295|Loss: 0.842991054058075:   5%|▍         | 294/6470 [03:35<1:15:15,  1.37it/s] 
3|295|Loss: 0.842991054058075:   5%|▍         | 295/6470 [03:35<1:12:01,  1.43it/s]
3|296|Loss: 0.8779248595237732:   5%|▍         | 295/6470 [03:35<1:12:01,  1.43it/s]
3|296|Loss: 0.8779248595237732:   5%|▍         | 296/6470 [03:36<1:13:06,  1.41it/s]
3|297|Loss: 0.8109303712844849:   5%|▍         | 296/6470 [03:36<1:13:06,  1.41it/s]
3|297|Loss: 0.8109303712844849:   5%|▍         | 297/6470 [03:36<1:12:43,  1.41it/s]
3|298|Loss: 0.8549802303314209:   5%|▍         | 297/6470 [03:37<1:12:43,  1.41it/s]
3|298|Loss: 0.8549802303314209:   5%|▍         | 298/6470 [03:37<1:19:55,  1.29it/s]
3|299|Loss: 0.8728961944580078:   5%|▍         | 298/6470 [03:38<1:19:55,  1.29it/s]
3|299|Loss: 0.8728961944580078:   5%|▍         | 299/6470 [03:38<1:25:13,  1.21it/s]
3|300|Loss: 0.839699387550354:   5%|▍         | 299/6470 [03:39<1:25:13,  1.21it/s] 
3|300|Loss: 0.839699387550354:   5%|▍         | 300/6470 [03:39<1:20:46,  1.27it/s]
3|301|Loss: 0.9821537137031555:   5%|▍         | 300/6470 [03:39<1:20:46,  1.27it/s]
3|301|Loss: 0.9821537137031555:   5%|▍         | 301/6470 [03:40<1:16:14,  1.35it/s]
3|302|Loss: 0.8300721645355225:   5%|▍         | 301/6470 [03:40<1:16:14,  1.35it/s]
3|302|Loss: 0.8300721645355225:   5%|▍         | 302/6470 [03:40<1:17:02,  1.33it/s]
3|303|Loss: 0.9337899088859558:   5%|▍         | 302/6470 [03:41<1:17:02,  1.33it/s]
3|303|Loss: 0.9337899088859558:   5%|▍         | 303/6470 [03:41<1:18:36,  1.31it/s]
3|304|Loss: 0.8370224237442017:   5%|▍         | 303/6470 [03:42<1:18:36,  1.31it/s]
3|304|Loss: 0.8370224237442017:   5%|▍         | 304/6470 [03:42<1:16:24,  1.35it/s]
3|305|Loss: 0.7722485065460205:   5%|▍         | 304/6470 [03:42<1:16:24,  1.35it/s]
3|305|Loss: 0.7722485065460205:   5%|▍         | 305/6470 [03:42<1:10:39,  1.45it/s]
3|306|Loss: 0.7287952899932861:   5%|▍         | 305/6470 [03:43<1:10:39,  1.45it/s]
3|306|Loss: 0.7287952899932861:   5%|▍         | 306/6470 [03:43<1:09:38,  1.48it/s]
3|307|Loss: 0.7473569512367249:   5%|▍         | 306/6470 [03:44<1:09:38,  1.48it/s]
3|307|Loss: 0.7473569512367249:   5%|▍         | 307/6470 [03:44<1:13:10,  1.40it/s]
3|308|Loss: 0.9753380417823792:   5%|▍         | 307/6470 [03:44<1:13:10,  1.40it/s]
3|308|Loss: 0.9753380417823792:   5%|▍         | 308/6470 [03:45<1:11:54,  1.43it/s]
3|309|Loss: 0.745917797088623:   5%|▍         | 308/6470 [03:45<1:11:54,  1.43it/s] 
3|309|Loss: 0.745917797088623:   5%|▍         | 309/6470 [03:45<1:12:29,  1.42it/s]
3|310|Loss: 0.9762683510780334:   5%|▍         | 309/6470 [03:46<1:12:29,  1.42it/s]
3|310|Loss: 0.9762683510780334:   5%|▍         | 310/6470 [03:46<1:10:23,  1.46it/s]
3|311|Loss: 0.6314970850944519:   5%|▍         | 310/6470 [03:46<1:10:23,  1.46it/s]
3|311|Loss: 0.6314970850944519:   5%|▍         | 311/6470 [03:47<1:18:41,  1.30it/s]
3|312|Loss: 0.852005660533905:   5%|▍         | 311/6470 [03:47<1:18:41,  1.30it/s] 
3|312|Loss: 0.852005660533905:   5%|▍         | 312/6470 [03:48<1:19:51,  1.29it/s]
3|313|Loss: 0.9637571573257446:   5%|▍         | 312/6470 [03:48<1:19:51,  1.29it/s]
3|313|Loss: 0.9637571573257446:   5%|▍         | 313/6470 [03:48<1:17:00,  1.33it/s]
3|314|Loss: 0.7973609566688538:   5%|▍         | 313/6470 [03:49<1:17:00,  1.33it/s]
3|314|Loss: 0.7973609566688538:   5%|▍         | 314/6470 [03:49<1:13:03,  1.40it/s]
3|315|Loss: 0.8237162828445435:   5%|▍         | 314/6470 [03:49<1:13:03,  1.40it/s]
3|315|Loss: 0.8237162828445435:   5%|▍         | 315/6470 [03:50<1:16:54,  1.33it/s]
3|316|Loss: 0.761114776134491:   5%|▍         | 315/6470 [03:50<1:16:54,  1.33it/s] 
3|316|Loss: 0.761114776134491:   5%|▍         | 316/6470 [03:51<1:18:34,  1.31it/s]
3|317|Loss: 0.9173088669776917:   5%|▍         | 316/6470 [03:51<1:18:34,  1.31it/s]
3|317|Loss: 0.9173088669776917:   5%|▍         | 317/6470 [03:52<1:23:13,  1.23it/s]
3|318|Loss: 0.7990463972091675:   5%|▍         | 317/6470 [03:52<1:23:13,  1.23it/s]
3|318|Loss: 0.7990463972091675:   5%|▍         | 318/6470 [03:52<1:24:16,  1.22it/s]
3|319|Loss: 0.9268264174461365:   5%|▍         | 318/6470 [03:53<1:24:16,  1.22it/s]
3|319|Loss: 0.9268264174461365:   5%|▍         | 319/6470 [03:53<1:19:31,  1.29it/s]
3|320|Loss: 0.8506888747215271:   5%|▍         | 319/6470 [03:53<1:19:31,  1.29it/s]
3|320|Loss: 0.8506888747215271:   5%|▍         | 320/6470 [03:54<1:14:18,  1.38it/s]
3|321|Loss: 0.7321110963821411:   5%|▍         | 320/6470 [03:54<1:14:18,  1.38it/s]
3|321|Loss: 0.7321110963821411:   5%|▍         | 321/6470 [03:54<1:17:32,  1.32it/s]
3|322|Loss: 0.797717273235321:   5%|▍         | 321/6470 [03:55<1:17:32,  1.32it/s] 
3|322|Loss: 0.797717273235321:   5%|▍         | 322/6470 [03:55<1:18:47,  1.30it/s]
3|323|Loss: 0.8312309980392456:   5%|▍         | 322/6470 [03:56<1:18:47,  1.30it/s]
3|323|Loss: 0.8312309980392456:   5%|▍         | 323/6470 [03:56<1:16:23,  1.34it/s]
3|324|Loss: 0.7106539607048035:   5%|▍         | 323/6470 [03:56<1:16:23,  1.34it/s]
3|324|Loss: 0.7106539607048035:   5%|▌         | 324/6470 [03:57<1:13:48,  1.39it/s]
3|325|Loss: 0.841194748878479:   5%|▌         | 324/6470 [03:57<1:13:48,  1.39it/s] 
3|325|Loss: 0.841194748878479:   5%|▌         | 325/6470 [03:57<1:17:43,  1.32it/s]
3|326|Loss: 0.7349167466163635:   5%|▌         | 325/6470 [03:58<1:17:43,  1.32it/s]
3|326|Loss: 0.7349167466163635:   5%|▌         | 326/6470 [03:58<1:18:22,  1.31it/s]
3|327|Loss: 0.7286560535430908:   5%|▌         | 326/6470 [03:59<1:18:22,  1.31it/s]
3|327|Loss: 0.7286560535430908:   5%|▌         | 327/6470 [03:59<1:18:31,  1.30it/s]
3|328|Loss: 0.7404947876930237:   5%|▌         | 327/6470 [03:59<1:18:31,  1.30it/s]
3|328|Loss: 0.7404947876930237:   5%|▌         | 328/6470 [04:00<1:16:00,  1.35it/s]
3|329|Loss: 0.9016090035438538:   5%|▌         | 328/6470 [04:00<1:16:00,  1.35it/s]
3|329|Loss: 0.9016090035438538:   5%|▌         | 329/6470 [04:00<1:12:11,  1.42it/s]
3|330|Loss: 0.9735193848609924:   5%|▌         | 329/6470 [04:01<1:12:11,  1.42it/s]
3|330|Loss: 0.9735193848609924:   5%|▌         | 330/6470 [04:01<1:09:49,  1.47it/s]
3|331|Loss: 0.813206672668457:   5%|▌         | 330/6470 [04:01<1:09:49,  1.47it/s] 
3|331|Loss: 0.813206672668457:   5%|▌         | 331/6470 [04:02<1:09:44,  1.47it/s]
3|332|Loss: 0.853506326675415:   5%|▌         | 331/6470 [04:02<1:09:44,  1.47it/s]
3|332|Loss: 0.853506326675415:   5%|▌         | 332/6470 [04:02<1:12:26,  1.41it/s]
3|333|Loss: 0.7634203433990479:   5%|▌         | 332/6470 [04:03<1:12:26,  1.41it/s]
3|333|Loss: 0.7634203433990479:   5%|▌         | 333/6470 [04:03<1:04:49,  1.58it/s]
3|334|Loss: 0.934424877166748:   5%|▌         | 333/6470 [04:03<1:04:49,  1.58it/s] 
3|334|Loss: 0.934424877166748:   5%|▌         | 334/6470 [04:04<1:07:04,  1.52it/s]
3|335|Loss: 0.7514351606369019:   5%|▌         | 334/6470 [04:04<1:07:04,  1.52it/s]
3|335|Loss: 0.7514351606369019:   5%|▌         | 335/6470 [04:04<1:13:09,  1.40it/s]
3|336|Loss: 0.7188762426376343:   5%|▌         | 335/6470 [04:05<1:13:09,  1.40it/s]
3|336|Loss: 0.7188762426376343:   5%|▌         | 336/6470 [04:05<1:13:36,  1.39it/s]
3|337|Loss: 0.7814144492149353:   5%|▌         | 336/6470 [04:06<1:13:36,  1.39it/s]
3|337|Loss: 0.7814144492149353:   5%|▌         | 337/6470 [04:06<1:16:56,  1.33it/s]
3|338|Loss: 0.7569056749343872:   5%|▌         | 337/6470 [04:06<1:16:56,  1.33it/s]
3|338|Loss: 0.7569056749343872:   5%|▌         | 338/6470 [04:07<1:10:46,  1.44it/s]
3|339|Loss: 0.6257364749908447:   5%|▌         | 338/6470 [04:07<1:10:46,  1.44it/s]
3|339|Loss: 0.6257364749908447:   5%|▌         | 339/6470 [04:07<1:02:09,  1.64it/s]
3|340|Loss: 0.9086732268333435:   5%|▌         | 339/6470 [04:07<1:02:09,  1.64it/s]
3|340|Loss: 0.9086732268333435:   5%|▌         | 340/6470 [04:08<1:01:18,  1.67it/s]
3|341|Loss: 0.7385575771331787:   5%|▌         | 340/6470 [04:08<1:01:18,  1.67it/s]
3|341|Loss: 0.7385575771331787:   5%|▌         | 341/6470 [04:08<1:02:18,  1.64it/s]
3|342|Loss: 0.7589011788368225:   5%|▌         | 341/6470 [04:09<1:02:18,  1.64it/s]
3|342|Loss: 0.7589011788368225:   5%|▌         | 342/6470 [04:09<1:10:04,  1.46it/s]
3|343|Loss: 0.8464156985282898:   5%|▌         | 342/6470 [04:10<1:10:04,  1.46it/s]
3|343|Loss: 0.8464156985282898:   5%|▌         | 343/6470 [04:10<1:14:25,  1.37it/s]
3|344|Loss: 0.9277582168579102:   5%|▌         | 343/6470 [04:10<1:14:25,  1.37it/s]
3|344|Loss: 0.9277582168579102:   5%|▌         | 344/6470 [04:11<1:16:28,  1.34it/s]
3|345|Loss: 0.8497545719146729:   5%|▌         | 344/6470 [04:11<1:16:28,  1.34it/s]
3|345|Loss: 0.8497545719146729:   5%|▌         | 345/6470 [04:11<1:16:39,  1.33it/s]
3|346|Loss: 0.8316214084625244:   5%|▌         | 345/6470 [04:12<1:16:39,  1.33it/s]
3|346|Loss: 0.8316214084625244:   5%|▌         | 346/6470 [04:12<1:13:42,  1.38it/s]
3|347|Loss: 0.9027391076087952:   5%|▌         | 346/6470 [04:12<1:13:42,  1.38it/s]
3|347|Loss: 0.9027391076087952:   5%|▌         | 347/6470 [04:13<1:09:03,  1.48it/s]
3|348|Loss: 0.811602771282196:   5%|▌         | 347/6470 [04:13<1:09:03,  1.48it/s] 
3|348|Loss: 0.811602771282196:   5%|▌         | 348/6470 [04:13<1:10:55,  1.44it/s]
3|349|Loss: 0.6738417148590088:   5%|▌         | 348/6470 [04:14<1:10:55,  1.44it/s]
3|349|Loss: 0.6738417148590088:   5%|▌         | 349/6470 [04:14<1:09:45,  1.46it/s]
3|350|Loss: 0.8507183194160461:   5%|▌         | 349/6470 [04:14<1:09:45,  1.46it/s]
3|350|Loss: 0.8507183194160461:   5%|▌         | 350/6470 [04:15<1:11:43,  1.42it/s]
3|351|Loss: 0.7939030528068542:   5%|▌         | 350/6470 [04:15<1:11:43,  1.42it/s]
3|351|Loss: 0.7939030528068542:   5%|▌         | 351/6470 [04:16<1:18:16,  1.30it/s]
3|352|Loss: 0.8304001688957214:   5%|▌         | 351/6470 [04:16<1:18:16,  1.30it/s]
3|352|Loss: 0.8304001688957214:   5%|▌         | 352/6470 [04:17<1:23:14,  1.22it/s]
3|353|Loss: 0.865842342376709:   5%|▌         | 352/6470 [04:17<1:23:14,  1.22it/s] 
3|353|Loss: 0.865842342376709:   5%|▌         | 353/6470 [04:18<1:27:04,  1.17it/s]
3|354|Loss: 1.0879566669464111:   5%|▌         | 353/6470 [04:18<1:27:04,  1.17it/s]
3|354|Loss: 1.0879566669464111:   5%|▌         | 354/6470 [04:18<1:24:12,  1.21it/s]
3|355|Loss: 0.9917398691177368:   5%|▌         | 354/6470 [04:19<1:24:12,  1.21it/s]
3|355|Loss: 0.9917398691177368:   5%|▌         | 355/6470 [04:19<1:26:45,  1.17it/s]
3|356|Loss: 0.8327681422233582:   5%|▌         | 355/6470 [04:20<1:26:45,  1.17it/s]
3|356|Loss: 0.8327681422233582:   6%|▌         | 356/6470 [04:20<1:24:13,  1.21it/s]
3|357|Loss: 0.5834454298019409:   6%|▌         | 356/6470 [04:20<1:24:13,  1.21it/s]
3|357|Loss: 0.5834454298019409:   6%|▌         | 357/6470 [04:21<1:24:16,  1.21it/s]
3|358|Loss: 0.6920942664146423:   6%|▌         | 357/6470 [04:21<1:24:16,  1.21it/s]
3|358|Loss: 0.6920942664146423:   6%|▌         | 358/6470 [04:22<1:22:56,  1.23it/s]
3|359|Loss: 0.7244817614555359:   6%|▌         | 358/6470 [04:22<1:22:56,  1.23it/s]
3|359|Loss: 0.7244817614555359:   6%|▌         | 359/6470 [04:23<1:25:37,  1.19it/s]
3|360|Loss: 0.8195950388908386:   6%|▌         | 359/6470 [04:23<1:25:37,  1.19it/s]
3|360|Loss: 0.8195950388908386:   6%|▌         | 360/6470 [04:23<1:20:59,  1.26it/s]
3|361|Loss: 0.847774088382721:   6%|▌         | 360/6470 [04:24<1:20:59,  1.26it/s] 
3|361|Loss: 0.847774088382721:   6%|▌         | 361/6470 [04:24<1:18:38,  1.29it/s]
3|362|Loss: 1.0425018072128296:   6%|▌         | 361/6470 [04:24<1:18:38,  1.29it/s]
3|362|Loss: 1.0425018072128296:   6%|▌         | 362/6470 [04:25<1:20:48,  1.26it/s]
3|363|Loss: 0.7678365111351013:   6%|▌         | 362/6470 [04:25<1:20:48,  1.26it/s]
3|363|Loss: 0.7678365111351013:   6%|▌         | 363/6470 [04:26<1:22:48,  1.23it/s]
3|364|Loss: 0.8381256461143494:   6%|▌         | 363/6470 [04:26<1:22:48,  1.23it/s]
3|364|Loss: 0.8381256461143494:   6%|▌         | 364/6470 [04:27<1:26:00,  1.18it/s]
3|365|Loss: 0.813713550567627:   6%|▌         | 364/6470 [04:27<1:26:00,  1.18it/s] 
3|365|Loss: 0.813713550567627:   6%|▌         | 365/6470 [04:28<1:30:24,  1.13it/s]
3|366|Loss: 0.8901950716972351:   6%|▌         | 365/6470 [04:28<1:30:24,  1.13it/s]
3|366|Loss: 0.8901950716972351:   6%|▌         | 366/6470 [04:28<1:28:50,  1.15it/s]
3|367|Loss: 0.8038408160209656:   6%|▌         | 366/6470 [04:29<1:28:50,  1.15it/s]
3|367|Loss: 0.8038408160209656:   6%|▌         | 367/6470 [04:29<1:25:31,  1.19it/s]
3|368|Loss: 0.7315999269485474:   6%|▌         | 367/6470 [04:29<1:25:31,  1.19it/s]
3|368|Loss: 0.7315999269485474:   6%|▌         | 368/6470 [04:30<1:15:55,  1.34it/s]
3|369|Loss: 0.8188509345054626:   6%|▌         | 368/6470 [04:30<1:15:55,  1.34it/s]
3|369|Loss: 0.8188509345054626:   6%|▌         | 369/6470 [04:30<1:15:10,  1.35it/s]
3|370|Loss: 0.8639442920684814:   6%|▌         | 369/6470 [04:31<1:15:10,  1.35it/s]
3|370|Loss: 0.8639442920684814:   6%|▌         | 370/6470 [04:31<1:16:14,  1.33it/s]
3|371|Loss: 1.1448204517364502:   6%|▌         | 370/6470 [04:32<1:16:14,  1.33it/s]
3|371|Loss: 1.1448204517364502:   6%|▌         | 371/6470 [04:32<1:13:40,  1.38it/s]
3|372|Loss: 0.7790630459785461:   6%|▌         | 371/6470 [04:32<1:13:40,  1.38it/s]
3|372|Loss: 0.7790630459785461:   6%|▌         | 372/6470 [04:33<1:14:50,  1.36it/s]
3|373|Loss: 0.7988669276237488:   6%|▌         | 372/6470 [04:33<1:14:50,  1.36it/s]
3|373|Loss: 0.7988669276237488:   6%|▌         | 373/6470 [04:33<1:11:57,  1.41it/s]
3|374|Loss: 0.6564452052116394:   6%|▌         | 373/6470 [04:34<1:11:57,  1.41it/s]
3|374|Loss: 0.6564452052116394:   6%|▌         | 374/6470 [04:34<1:13:04,  1.39it/s]
3|375|Loss: 0.6681213974952698:   6%|▌         | 374/6470 [04:34<1:13:04,  1.39it/s]
3|375|Loss: 0.6681213974952698:   6%|▌         | 375/6470 [04:35<1:18:20,  1.30it/s]
3|376|Loss: 0.7633038759231567:   6%|▌         | 375/6470 [04:35<1:18:20,  1.30it/s]
3|376|Loss: 0.7633038759231567:   6%|▌         | 376/6470 [04:35<1:12:37,  1.40it/s]
3|377|Loss: 0.7987873554229736:   6%|▌         | 376/6470 [04:36<1:12:37,  1.40it/s]
3|377|Loss: 0.7987873554229736:   6%|▌         | 377/6470 [04:36<1:11:39,  1.42it/s]
3|378|Loss: 0.7986992597579956:   6%|▌         | 377/6470 [04:37<1:11:39,  1.42it/s]
3|378|Loss: 0.7986992597579956:   6%|▌         | 378/6470 [04:37<1:16:56,  1.32it/s]
3|379|Loss: 0.8864479064941406:   6%|▌         | 378/6470 [04:37<1:16:56,  1.32it/s]
3|379|Loss: 0.8864479064941406:   6%|▌         | 379/6470 [04:38<1:12:32,  1.40it/s]
3|380|Loss: 0.8807242512702942:   6%|▌         | 379/6470 [04:38<1:12:32,  1.40it/s]
3|380|Loss: 0.8807242512702942:   6%|▌         | 380/6470 [04:38<1:09:56,  1.45it/s]
3|381|Loss: 0.9136287569999695:   6%|▌         | 380/6470 [04:39<1:09:56,  1.45it/s]
3|381|Loss: 0.9136287569999695:   6%|▌         | 381/6470 [04:39<1:08:25,  1.48it/s]
3|382|Loss: 0.7586261034011841:   6%|▌         | 381/6470 [04:39<1:08:25,  1.48it/s]
3|382|Loss: 0.7586261034011841:   6%|▌         | 382/6470 [04:39<1:03:04,  1.61it/s]
3|383|Loss: 0.8583381175994873:   6%|▌         | 382/6470 [04:40<1:03:04,  1.61it/s]
3|383|Loss: 0.8583381175994873:   6%|▌         | 383/6470 [04:40<1:06:22,  1.53it/s]
3|384|Loss: 0.8515937924385071:   6%|▌         | 383/6470 [04:41<1:06:22,  1.53it/s]
3|384|Loss: 0.8515937924385071:   6%|▌         | 384/6470 [04:41<1:04:54,  1.56it/s]
3|385|Loss: 0.8484523892402649:   6%|▌         | 384/6470 [04:41<1:04:54,  1.56it/s]
3|385|Loss: 0.8484523892402649:   6%|▌         | 385/6470 [04:41<1:04:30,  1.57it/s]
3|386|Loss: 0.776543915271759:   6%|▌         | 385/6470 [04:42<1:04:30,  1.57it/s] 
3|386|Loss: 0.776543915271759:   6%|▌         | 386/6470 [04:42<1:06:15,  1.53it/s]
3|387|Loss: 0.8244216442108154:   6%|▌         | 386/6470 [04:42<1:06:15,  1.53it/s]
3|387|Loss: 0.8244216442108154:   6%|▌         | 387/6470 [04:43<1:02:14,  1.63it/s]
3|388|Loss: 0.9787189364433289:   6%|▌         | 387/6470 [04:43<1:02:14,  1.63it/s]
3|388|Loss: 0.9787189364433289:   6%|▌         | 388/6470 [04:43<1:01:17,  1.65it/s]
3|389|Loss: 0.8257147669792175:   6%|▌         | 388/6470 [04:44<1:01:17,  1.65it/s]
3|389|Loss: 0.8257147669792175:   6%|▌         | 389/6470 [04:44<1:02:45,  1.61it/s]
3|390|Loss: 0.8076525926589966:   6%|▌         | 389/6470 [04:44<1:02:45,  1.61it/s]
3|390|Loss: 0.8076525926589966:   6%|▌         | 390/6470 [04:45<1:04:25,  1.57it/s]
3|391|Loss: 0.747268557548523:   6%|▌         | 390/6470 [04:45<1:04:25,  1.57it/s] 
3|391|Loss: 0.747268557548523:   6%|▌         | 391/6470 [04:45<1:09:07,  1.47it/s]
3|392|Loss: 0.9121734499931335:   6%|▌         | 391/6470 [04:46<1:09:07,  1.47it/s]
3|392|Loss: 0.9121734499931335:   6%|▌         | 392/6470 [04:46<1:12:42,  1.39it/s]
3|393|Loss: 0.841698944568634:   6%|▌         | 392/6470 [04:47<1:12:42,  1.39it/s] 
3|393|Loss: 0.841698944568634:   6%|▌         | 393/6470 [04:47<1:14:32,  1.36it/s]
3|394|Loss: 0.7884053587913513:   6%|▌         | 393/6470 [04:47<1:14:32,  1.36it/s]
3|394|Loss: 0.7884053587913513:   6%|▌         | 394/6470 [04:48<1:16:15,  1.33it/s]
3|395|Loss: 0.8798416256904602:   6%|▌         | 394/6470 [04:48<1:16:15,  1.33it/s]
3|395|Loss: 0.8798416256904602:   6%|▌         | 395/6470 [04:49<1:20:52,  1.25it/s]
3|396|Loss: 0.7608804106712341:   6%|▌         | 395/6470 [04:49<1:20:52,  1.25it/s]
3|396|Loss: 0.7608804106712341:   6%|▌         | 396/6470 [04:50<1:25:22,  1.19it/s]
3|397|Loss: 0.8792011737823486:   6%|▌         | 396/6470 [04:50<1:25:22,  1.19it/s]
3|397|Loss: 0.8792011737823486:   6%|▌         | 397/6470 [04:50<1:28:20,  1.15it/s]
3|398|Loss: 0.7078946828842163:   6%|▌         | 397/6470 [04:51<1:28:20,  1.15it/s]
3|398|Loss: 0.7078946828842163:   6%|▌         | 398/6470 [04:51<1:29:21,  1.13it/s]
3|399|Loss: 0.7608328461647034:   6%|▌         | 398/6470 [04:52<1:29:21,  1.13it/s]
3|399|Loss: 0.7608328461647034:   6%|▌         | 399/6470 [04:52<1:30:10,  1.12it/s]
3|400|Loss: 0.7234960198402405:   6%|▌         | 399/6470 [04:53<1:30:10,  1.12it/s]
3|400|Loss: 0.7234960198402405:   6%|▌         | 400/6470 [04:53<1:18:00,  1.30it/s]
3|401|Loss: 0.8291320204734802:   6%|▌         | 400/6470 [04:53<1:18:00,  1.30it/s]
3|401|Loss: 0.8291320204734802:   6%|▌         | 401/6470 [04:53<1:15:26,  1.34it/s]
3|402|Loss: 0.7960972189903259:   6%|▌         | 401/6470 [04:54<1:15:26,  1.34it/s]
3|402|Loss: 0.7960972189903259:   6%|▌         | 402/6470 [04:54<1:17:09,  1.31it/s]
3|403|Loss: 0.8163171410560608:   6%|▌         | 402/6470 [04:55<1:17:09,  1.31it/s]
3|403|Loss: 0.8163171410560608:   6%|▌         | 403/6470 [04:55<1:19:02,  1.28it/s]
3|404|Loss: 0.7909966707229614:   6%|▌         | 403/6470 [04:56<1:19:02,  1.28it/s]
3|404|Loss: 0.7909966707229614:   6%|▌         | 404/6470 [04:56<1:21:24,  1.24it/s]
3|405|Loss: 0.741425633430481:   6%|▌         | 404/6470 [04:56<1:21:24,  1.24it/s] 
3|405|Loss: 0.741425633430481:   6%|▋         | 405/6470 [04:57<1:14:31,  1.36it/s]
3|406|Loss: 0.862570583820343:   6%|▋         | 405/6470 [04:57<1:14:31,  1.36it/s]
3|406|Loss: 0.862570583820343:   6%|▋         | 406/6470 [04:57<1:09:15,  1.46it/s]
3|407|Loss: 0.9145786762237549:   6%|▋         | 406/6470 [04:57<1:09:15,  1.46it/s]
3|407|Loss: 0.9145786762237549:   6%|▋         | 407/6470 [04:58<1:04:56,  1.56it/s]
3|408|Loss: 0.6683060526847839:   6%|▋         | 407/6470 [04:58<1:04:56,  1.56it/s]
3|408|Loss: 0.6683060526847839:   6%|▋         | 408/6470 [04:58<1:09:57,  1.44it/s]
3|409|Loss: 0.788362979888916:   6%|▋         | 408/6470 [04:59<1:09:57,  1.44it/s] 
3|409|Loss: 0.788362979888916:   6%|▋         | 409/6470 [04:59<1:11:03,  1.42it/s]
3|410|Loss: 0.8097297549247742:   6%|▋         | 409/6470 [05:00<1:11:03,  1.42it/s]
3|410|Loss: 0.8097297549247742:   6%|▋         | 410/6470 [05:00<1:08:09,  1.48it/s]
3|411|Loss: 0.9075579643249512:   6%|▋         | 410/6470 [05:00<1:08:09,  1.48it/s]
3|411|Loss: 0.9075579643249512:   6%|▋         | 411/6470 [05:01<1:14:07,  1.36it/s]
3|412|Loss: 0.9282532334327698:   6%|▋         | 411/6470 [05:01<1:14:07,  1.36it/s]
3|412|Loss: 0.9282532334327698:   6%|▋         | 412/6470 [05:01<1:17:05,  1.31it/s]
3|413|Loss: 0.7527540922164917:   6%|▋         | 412/6470 [05:02<1:17:05,  1.31it/s]
3|413|Loss: 0.7527540922164917:   6%|▋         | 413/6470 [05:02<1:17:37,  1.30it/s]
3|414|Loss: 0.9678366184234619:   6%|▋         | 413/6470 [05:03<1:17:37,  1.30it/s]
3|414|Loss: 0.9678366184234619:   6%|▋         | 414/6470 [05:03<1:13:05,  1.38it/s]
3|415|Loss: 0.8279063105583191:   6%|▋         | 414/6470 [05:03<1:13:05,  1.38it/s]
3|415|Loss: 0.8279063105583191:   6%|▋         | 415/6470 [05:04<1:12:38,  1.39it/s]
3|416|Loss: 0.8593068718910217:   6%|▋         | 415/6470 [05:04<1:12:38,  1.39it/s]
3|416|Loss: 0.8593068718910217:   6%|▋         | 416/6470 [05:05<1:19:27,  1.27it/s]
3|417|Loss: 0.8422262072563171:   6%|▋         | 416/6470 [05:05<1:19:27,  1.27it/s]
3|417|Loss: 0.8422262072563171:   6%|▋         | 417/6470 [05:05<1:22:24,  1.22it/s]
3|418|Loss: 0.7210178971290588:   6%|▋         | 417/6470 [05:06<1:22:24,  1.22it/s]
3|418|Loss: 0.7210178971290588:   6%|▋         | 418/6470 [05:06<1:20:57,  1.25it/s]
3|419|Loss: 0.5886902213096619:   6%|▋         | 418/6470 [05:07<1:20:57,  1.25it/s]
3|419|Loss: 0.5886902213096619:   6%|▋         | 419/6470 [05:07<1:18:57,  1.28it/s]
3|420|Loss: 0.7953326106071472:   6%|▋         | 419/6470 [05:07<1:18:57,  1.28it/s]
3|420|Loss: 0.7953326106071472:   6%|▋         | 420/6470 [05:07<1:11:11,  1.42it/s]
3|421|Loss: 0.8803618550300598:   6%|▋         | 420/6470 [05:08<1:11:11,  1.42it/s]
3|421|Loss: 0.8803618550300598:   7%|▋         | 421/6470 [05:08<1:08:15,  1.48it/s]
3|422|Loss: 0.9096419215202332:   7%|▋         | 421/6470 [05:08<1:08:15,  1.48it/s]
3|422|Loss: 0.9096419215202332:   7%|▋         | 422/6470 [05:09<1:10:48,  1.42it/s]
3|423|Loss: 0.8633540272712708:   7%|▋         | 422/6470 [05:09<1:10:48,  1.42it/s]
3|423|Loss: 0.8633540272712708:   7%|▋         | 423/6470 [05:10<1:12:22,  1.39it/s]
3|424|Loss: 0.8665518760681152:   7%|▋         | 423/6470 [05:10<1:12:22,  1.39it/s]
3|424|Loss: 0.8665518760681152:   7%|▋         | 424/6470 [05:10<1:08:04,  1.48it/s]
3|425|Loss: 0.7663014531135559:   7%|▋         | 424/6470 [05:11<1:08:04,  1.48it/s]
3|425|Loss: 0.7663014531135559:   7%|▋         | 425/6470 [05:11<1:12:18,  1.39it/s]
3|426|Loss: 0.9241556525230408:   7%|▋         | 425/6470 [05:11<1:12:18,  1.39it/s]
3|426|Loss: 0.9241556525230408:   7%|▋         | 426/6470 [05:12<1:07:31,  1.49it/s]
3|427|Loss: 0.8526321649551392:   7%|▋         | 426/6470 [05:12<1:07:31,  1.49it/s]
3|427|Loss: 0.8526321649551392:   7%|▋         | 427/6470 [05:12<1:04:45,  1.56it/s]
3|428|Loss: 0.9684958457946777:   7%|▋         | 427/6470 [05:12<1:04:45,  1.56it/s]
3|428|Loss: 0.9684958457946777:   7%|▋         | 428/6470 [05:13<1:03:48,  1.58it/s]
3|429|Loss: 0.87212735414505:   7%|▋         | 428/6470 [05:13<1:03:48,  1.58it/s]  
3|429|Loss: 0.87212735414505:   7%|▋         | 429/6470 [05:13<1:05:18,  1.54it/s]
3|430|Loss: 0.7941956520080566:   7%|▋         | 429/6470 [05:14<1:05:18,  1.54it/s]
3|430|Loss: 0.7941956520080566:   7%|▋         | 430/6470 [05:14<1:06:39,  1.51it/s]
3|431|Loss: 0.8976835012435913:   7%|▋         | 430/6470 [05:15<1:06:39,  1.51it/s]
3|431|Loss: 0.8976835012435913:   7%|▋         | 431/6470 [05:15<1:09:59,  1.44it/s]
3|432|Loss: 0.7710120677947998:   7%|▋         | 431/6470 [05:15<1:09:59,  1.44it/s]
3|432|Loss: 0.7710120677947998:   7%|▋         | 432/6470 [05:16<1:11:41,  1.40it/s]
3|433|Loss: 0.8641085624694824:   7%|▋         | 432/6470 [05:16<1:11:41,  1.40it/s]
3|433|Loss: 0.8641085624694824:   7%|▋         | 433/6470 [05:16<1:13:52,  1.36it/s]
3|434|Loss: 0.830157995223999:   7%|▋         | 433/6470 [05:17<1:13:52,  1.36it/s] 
3|434|Loss: 0.830157995223999:   7%|▋         | 434/6470 [05:17<1:16:00,  1.32it/s]
3|435|Loss: 0.8931398987770081:   7%|▋         | 434/6470 [05:18<1:16:00,  1.32it/s]
3|435|Loss: 0.8931398987770081:   7%|▋         | 435/6470 [05:18<1:17:47,  1.29it/s]
3|436|Loss: 0.6821634769439697:   7%|▋         | 435/6470 [05:18<1:17:47,  1.29it/s]
3|436|Loss: 0.6821634769439697:   7%|▋         | 436/6470 [05:19<1:12:36,  1.39it/s]
3|437|Loss: 0.8224488496780396:   7%|▋         | 436/6470 [05:19<1:12:36,  1.39it/s]
3|437|Loss: 0.8224488496780396:   7%|▋         | 437/6470 [05:19<1:15:03,  1.34it/s]
3|438|Loss: 0.7513880729675293:   7%|▋         | 437/6470 [05:20<1:15:03,  1.34it/s]
3|438|Loss: 0.7513880729675293:   7%|▋         | 438/6470 [05:20<1:19:59,  1.26it/s]
3|439|Loss: 0.6879736185073853:   7%|▋         | 438/6470 [05:21<1:19:59,  1.26it/s]
3|439|Loss: 0.6879736185073853:   7%|▋         | 439/6470 [05:21<1:27:58,  1.14it/s]
3|440|Loss: 0.9484456777572632:   7%|▋         | 439/6470 [05:22<1:27:58,  1.14it/s]
3|440|Loss: 0.9484456777572632:   7%|▋         | 440/6470 [05:22<1:27:34,  1.15it/s]
3|441|Loss: 0.8097661733627319:   7%|▋         | 440/6470 [05:23<1:27:34,  1.15it/s]
3|441|Loss: 0.8097661733627319:   7%|▋         | 441/6470 [05:23<1:27:50,  1.14it/s]
3|442|Loss: 0.7315090894699097:   7%|▋         | 441/6470 [05:24<1:27:50,  1.14it/s]
3|442|Loss: 0.7315090894699097:   7%|▋         | 442/6470 [05:24<1:25:39,  1.17it/s]
3|443|Loss: 0.9397522807121277:   7%|▋         | 442/6470 [05:24<1:25:39,  1.17it/s]
3|443|Loss: 0.9397522807121277:   7%|▋         | 443/6470 [05:25<1:20:27,  1.25it/s]
3|444|Loss: 0.7653559446334839:   7%|▋         | 443/6470 [05:25<1:20:27,  1.25it/s]
3|444|Loss: 0.7653559446334839:   7%|▋         | 444/6470 [05:25<1:13:48,  1.36it/s]
3|445|Loss: 0.7731607556343079:   7%|▋         | 444/6470 [05:26<1:13:48,  1.36it/s]
3|445|Loss: 0.7731607556343079:   7%|▋         | 445/6470 [05:26<1:06:30,  1.51it/s]
3|446|Loss: 0.855324923992157:   7%|▋         | 445/6470 [05:26<1:06:30,  1.51it/s] 
3|446|Loss: 0.855324923992157:   7%|▋         | 446/6470 [05:27<1:10:47,  1.42it/s]
3|447|Loss: 0.8479507565498352:   7%|▋         | 446/6470 [05:27<1:10:47,  1.42it/s]
3|447|Loss: 0.8479507565498352:   7%|▋         | 447/6470 [05:27<1:10:52,  1.42it/s]
3|448|Loss: 0.903936505317688:   7%|▋         | 447/6470 [05:28<1:10:52,  1.42it/s] 
3|448|Loss: 0.903936505317688:   7%|▋         | 448/6470 [05:28<1:08:15,  1.47it/s]
3|449|Loss: 0.7928422093391418:   7%|▋         | 448/6470 [05:28<1:08:15,  1.47it/s]
3|449|Loss: 0.7928422093391418:   7%|▋         | 449/6470 [05:28<1:05:50,  1.52it/s]
3|450|Loss: 0.7424767017364502:   7%|▋         | 449/6470 [05:29<1:05:50,  1.52it/s]
3|450|Loss: 0.7424767017364502:   7%|▋         | 450/6470 [05:29<1:11:24,  1.41it/s]
3|451|Loss: 0.7876790761947632:   7%|▋         | 450/6470 [05:30<1:11:24,  1.41it/s]
3|451|Loss: 0.7876790761947632:   7%|▋         | 451/6470 [05:30<1:10:57,  1.41it/s]
3|452|Loss: 0.8605309724807739:   7%|▋         | 451/6470 [05:30<1:10:57,  1.41it/s]
3|452|Loss: 0.8605309724807739:   7%|▋         | 452/6470 [05:31<1:10:35,  1.42it/s]
3|453|Loss: 0.8320848345756531:   7%|▋         | 452/6470 [05:31<1:10:35,  1.42it/s]
3|453|Loss: 0.8320848345756531:   7%|▋         | 453/6470 [05:31<1:08:32,  1.46it/s]
3|454|Loss: 0.7641230821609497:   7%|▋         | 453/6470 [05:32<1:08:32,  1.46it/s]
3|454|Loss: 0.7641230821609497:   7%|▋         | 454/6470 [05:32<1:10:48,  1.42it/s]
3|455|Loss: 0.8404484987258911:   7%|▋         | 454/6470 [05:32<1:10:48,  1.42it/s]
3|455|Loss: 0.8404484987258911:   7%|▋         | 455/6470 [05:33<1:05:51,  1.52it/s]
3|456|Loss: 0.8629892468452454:   7%|▋         | 455/6470 [05:33<1:05:51,  1.52it/s]
3|456|Loss: 0.8629892468452454:   7%|▋         | 456/6470 [05:33<1:01:50,  1.62it/s]
3|457|Loss: 0.8493371605873108:   7%|▋         | 456/6470 [05:33<1:01:50,  1.62it/s]
3|457|Loss: 0.8493371605873108:   7%|▋         | 457/6470 [05:34<55:43,  1.80it/s]  
3|458|Loss: 0.7709940075874329:   7%|▋         | 457/6470 [05:34<55:43,  1.80it/s]
3|458|Loss: 0.7709940075874329:   7%|▋         | 458/6470 [05:34<1:02:40,  1.60it/s]
3|459|Loss: 0.8870729804039001:   7%|▋         | 458/6470 [05:35<1:02:40,  1.60it/s]
3|459|Loss: 0.8870729804039001:   7%|▋         | 459/6470 [05:35<1:08:01,  1.47it/s]
3|460|Loss: 0.8208553194999695:   7%|▋         | 459/6470 [05:36<1:08:01,  1.47it/s]
3|460|Loss: 0.8208553194999695:   7%|▋         | 460/6470 [05:36<1:09:19,  1.44it/s]
3|461|Loss: 0.7514936923980713:   7%|▋         | 460/6470 [05:36<1:09:19,  1.44it/s]
3|461|Loss: 0.7514936923980713:   7%|▋         | 461/6470 [05:36<1:05:44,  1.52it/s]
3|462|Loss: 0.6842330098152161:   7%|▋         | 461/6470 [05:37<1:05:44,  1.52it/s]
3|462|Loss: 0.6842330098152161:   7%|▋         | 462/6470 [05:37<1:05:20,  1.53it/s]
3|463|Loss: 0.821711003780365:   7%|▋         | 462/6470 [05:37<1:05:20,  1.53it/s] 
3|463|Loss: 0.821711003780365:   7%|▋         | 463/6470 [05:38<1:05:59,  1.52it/s]
3|464|Loss: 0.9801231026649475:   7%|▋         | 463/6470 [05:38<1:05:59,  1.52it/s]
3|464|Loss: 0.9801231026649475:   7%|▋         | 464/6470 [05:38<1:06:55,  1.50it/s]
3|465|Loss: 0.8649678826332092:   7%|▋         | 464/6470 [05:39<1:06:55,  1.50it/s]
3|465|Loss: 0.8649678826332092:   7%|▋         | 465/6470 [05:39<1:07:36,  1.48it/s]
3|466|Loss: 0.7181010246276855:   7%|▋         | 465/6470 [05:40<1:07:36,  1.48it/s]
3|466|Loss: 0.7181010246276855:   7%|▋         | 466/6470 [05:40<1:06:43,  1.50it/s]
3|467|Loss: 0.857995867729187:   7%|▋         | 466/6470 [05:40<1:06:43,  1.50it/s] 
3|467|Loss: 0.857995867729187:   7%|▋         | 467/6470 [05:40<1:07:11,  1.49it/s]
3|468|Loss: 0.9965120553970337:   7%|▋         | 467/6470 [05:41<1:07:11,  1.49it/s]
3|468|Loss: 0.9965120553970337:   7%|▋         | 468/6470 [05:41<1:02:36,  1.60it/s]
3|469|Loss: 0.8504140973091125:   7%|▋         | 468/6470 [05:41<1:02:36,  1.60it/s]
3|469|Loss: 0.8504140973091125:   7%|▋         | 469/6470 [05:42<1:08:49,  1.45it/s]
3|470|Loss: 1.0800871849060059:   7%|▋         | 469/6470 [05:42<1:08:49,  1.45it/s]
3|470|Loss: 1.0800871849060059:   7%|▋         | 470/6470 [05:43<1:12:05,  1.39it/s]
3|471|Loss: 0.9214847087860107:   7%|▋         | 470/6470 [05:43<1:12:05,  1.39it/s]
3|471|Loss: 0.9214847087860107:   7%|▋         | 471/6470 [05:44<1:17:29,  1.29it/s]
3|472|Loss: 0.8496058583259583:   7%|▋         | 471/6470 [05:44<1:17:29,  1.29it/s]
3|472|Loss: 0.8496058583259583:   7%|▋         | 472/6470 [05:44<1:16:10,  1.31it/s]
3|473|Loss: 0.7160245180130005:   7%|▋         | 472/6470 [05:45<1:16:10,  1.31it/s]
3|473|Loss: 0.7160245180130005:   7%|▋         | 473/6470 [05:45<1:13:10,  1.37it/s]
3|474|Loss: 1.040345549583435:   7%|▋         | 473/6470 [05:45<1:13:10,  1.37it/s] 
3|474|Loss: 1.040345549583435:   7%|▋         | 474/6470 [05:46<1:13:23,  1.36it/s]
3|475|Loss: 0.7776004076004028:   7%|▋         | 474/6470 [05:46<1:13:23,  1.36it/s]
3|475|Loss: 0.7776004076004028:   7%|▋         | 475/6470 [05:46<1:12:02,  1.39it/s]
3|476|Loss: 0.9500485062599182:   7%|▋         | 475/6470 [05:47<1:12:02,  1.39it/s]
3|476|Loss: 0.9500485062599182:   7%|▋         | 476/6470 [05:47<1:12:32,  1.38it/s]
3|477|Loss: 0.8015305399894714:   7%|▋         | 476/6470 [05:48<1:12:32,  1.38it/s]
3|477|Loss: 0.8015305399894714:   7%|▋         | 477/6470 [05:48<1:14:00,  1.35it/s]
3|478|Loss: 0.8691742420196533:   7%|▋         | 477/6470 [05:48<1:14:00,  1.35it/s]
3|478|Loss: 0.8691742420196533:   7%|▋         | 478/6470 [05:49<1:15:35,  1.32it/s]
3|479|Loss: 0.7623112797737122:   7%|▋         | 478/6470 [05:49<1:15:35,  1.32it/s]
3|479|Loss: 0.7623112797737122:   7%|▋         | 479/6470 [05:49<1:16:44,  1.30it/s]
3|480|Loss: 0.8113433718681335:   7%|▋         | 479/6470 [05:50<1:16:44,  1.30it/s]
3|480|Loss: 0.8113433718681335:   7%|▋         | 480/6470 [05:50<1:13:44,  1.35it/s]
3|481|Loss: 0.7266911864280701:   7%|▋         | 480/6470 [05:51<1:13:44,  1.35it/s]
3|481|Loss: 0.7266911864280701:   7%|▋         | 481/6470 [05:51<1:14:13,  1.34it/s]
3|482|Loss: 0.7579062581062317:   7%|▋         | 481/6470 [05:51<1:14:13,  1.34it/s]
3|482|Loss: 0.7579062581062317:   7%|▋         | 482/6470 [05:52<1:18:53,  1.26it/s]
3|483|Loss: 0.8394218683242798:   7%|▋         | 482/6470 [05:52<1:18:53,  1.26it/s]
3|483|Loss: 0.8394218683242798:   7%|▋         | 483/6470 [05:53<1:19:02,  1.26it/s]
3|484|Loss: 0.828289806842804:   7%|▋         | 483/6470 [05:53<1:19:02,  1.26it/s] 
3|484|Loss: 0.828289806842804:   7%|▋         | 484/6470 [05:53<1:20:53,  1.23it/s]
3|485|Loss: 0.7423638105392456:   7%|▋         | 484/6470 [05:54<1:20:53,  1.23it/s]
3|485|Loss: 0.7423638105392456:   7%|▋         | 485/6470 [05:54<1:17:36,  1.29it/s]
3|486|Loss: 0.8751170039176941:   7%|▋         | 485/6470 [05:55<1:17:36,  1.29it/s]
3|486|Loss: 0.8751170039176941:   8%|▊         | 486/6470 [05:55<1:16:20,  1.31it/s]
3|487|Loss: 0.7176297307014465:   8%|▊         | 486/6470 [05:55<1:16:20,  1.31it/s]
3|487|Loss: 0.7176297307014465:   8%|▊         | 487/6470 [05:56<1:13:59,  1.35it/s]
3|488|Loss: 0.8626059293746948:   8%|▊         | 487/6470 [05:56<1:13:59,  1.35it/s]
3|488|Loss: 0.8626059293746948:   8%|▊         | 488/6470 [05:56<1:15:06,  1.33it/s]
3|489|Loss: 0.9436336159706116:   8%|▊         | 488/6470 [05:57<1:15:06,  1.33it/s]
3|489|Loss: 0.9436336159706116:   8%|▊         | 489/6470 [05:57<1:16:08,  1.31it/s]
3|490|Loss: 0.8694822192192078:   8%|▊         | 489/6470 [05:58<1:16:08,  1.31it/s]
3|490|Loss: 0.8694822192192078:   8%|▊         | 490/6470 [05:58<1:16:39,  1.30it/s]
3|491|Loss: 0.8982933163642883:   8%|▊         | 490/6470 [05:58<1:16:39,  1.30it/s]
3|491|Loss: 0.8982933163642883:   8%|▊         | 491/6470 [05:59<1:13:31,  1.36it/s]
3|492|Loss: 0.8207678198814392:   8%|▊         | 491/6470 [05:59<1:13:31,  1.36it/s]
3|492|Loss: 0.8207678198814392:   8%|▊         | 492/6470 [05:59<1:19:01,  1.26it/s]
3|493|Loss: 0.8486313223838806:   8%|▊         | 492/6470 [06:00<1:19:01,  1.26it/s]
3|493|Loss: 0.8486313223838806:   8%|▊         | 493/6470 [06:00<1:20:25,  1.24it/s]
3|494|Loss: 0.831231415271759:   8%|▊         | 493/6470 [06:01<1:20:25,  1.24it/s] 
3|494|Loss: 0.831231415271759:   8%|▊         | 494/6470 [06:01<1:17:32,  1.28it/s]
3|495|Loss: 0.7687436938285828:   8%|▊         | 494/6470 [06:01<1:17:32,  1.28it/s]
3|495|Loss: 0.7687436938285828:   8%|▊         | 495/6470 [06:01<1:07:19,  1.48it/s]
3|496|Loss: 0.8942099809646606:   8%|▊         | 495/6470 [06:02<1:07:19,  1.48it/s]
3|496|Loss: 0.8942099809646606:   8%|▊         | 496/6470 [06:02<1:05:36,  1.52it/s]
3|497|Loss: 0.8918556571006775:   8%|▊         | 496/6470 [06:03<1:05:36,  1.52it/s]
3|497|Loss: 0.8918556571006775:   8%|▊         | 497/6470 [06:03<1:12:37,  1.37it/s]
3|498|Loss: 0.7895812392234802:   8%|▊         | 497/6470 [06:03<1:12:37,  1.37it/s]
3|498|Loss: 0.7895812392234802:   8%|▊         | 498/6470 [06:04<1:15:34,  1.32it/s]
3|499|Loss: 0.7303307056427002:   8%|▊         | 498/6470 [06:04<1:15:34,  1.32it/s]
3|499|Loss: 0.7303307056427002:   8%|▊         | 499/6470 [06:05<1:19:14,  1.26it/s]
3|500|Loss: 0.8042343854904175:   8%|▊         | 499/6470 [06:05<1:19:14,  1.26it/s]
3|500|Loss: 0.8042343854904175:   8%|▊         | 500/6470 [06:05<1:15:20,  1.32it/s]
3|501|Loss: 0.8186529874801636:   8%|▊         | 500/6470 [06:06<1:15:20,  1.32it/s]
3|501|Loss: 0.8186529874801636:   8%|▊         | 501/6470 [06:06<1:12:47,  1.37it/s]
3|502|Loss: 0.9916617274284363:   8%|▊         | 501/6470 [06:07<1:12:47,  1.37it/s]
3|502|Loss: 0.9916617274284363:   8%|▊         | 502/6470 [06:07<1:17:00,  1.29it/s]
3|503|Loss: 0.8529133200645447:   8%|▊         | 502/6470 [06:07<1:17:00,  1.29it/s]
3|503|Loss: 0.8529133200645447:   8%|▊         | 503/6470 [06:08<1:15:16,  1.32it/s]
3|504|Loss: 0.8079361319541931:   8%|▊         | 503/6470 [06:08<1:15:16,  1.32it/s]
3|504|Loss: 0.8079361319541931:   8%|▊         | 504/6470 [06:08<1:11:07,  1.40it/s]
3|505|Loss: 0.9203813076019287:   8%|▊         | 504/6470 [06:09<1:11:07,  1.40it/s]
3|505|Loss: 0.9203813076019287:   8%|▊         | 505/6470 [06:09<1:15:41,  1.31it/s]
3|506|Loss: 0.7977292537689209:   8%|▊         | 505/6470 [06:10<1:15:41,  1.31it/s]
3|506|Loss: 0.7977292537689209:   8%|▊         | 506/6470 [06:10<1:19:47,  1.25it/s]
3|507|Loss: 0.8795027136802673:   8%|▊         | 506/6470 [06:10<1:19:47,  1.25it/s]
3|507|Loss: 0.8795027136802673:   8%|▊         | 507/6470 [06:11<1:17:42,  1.28it/s]
3|508|Loss: 0.9027456641197205:   8%|▊         | 507/6470 [06:11<1:17:42,  1.28it/s]
3|508|Loss: 0.9027456641197205:   8%|▊         | 508/6470 [06:12<1:19:01,  1.26it/s]
3|509|Loss: 0.9002683162689209:   8%|▊         | 508/6470 [06:12<1:19:01,  1.26it/s]
3|509|Loss: 0.9002683162689209:   8%|▊         | 509/6470 [06:12<1:13:56,  1.34it/s]
3|510|Loss: 0.8285554051399231:   8%|▊         | 509/6470 [06:13<1:13:56,  1.34it/s]
3|510|Loss: 0.8285554051399231:   8%|▊         | 510/6470 [06:13<1:11:39,  1.39it/s]
3|511|Loss: 0.791451632976532:   8%|▊         | 510/6470 [06:13<1:11:39,  1.39it/s] 
3|511|Loss: 0.791451632976532:   8%|▊         | 511/6470 [06:14<1:11:12,  1.39it/s]
3|512|Loss: 0.7436374425888062:   8%|▊         | 511/6470 [06:14<1:11:12,  1.39it/s]
3|512|Loss: 0.7436374425888062:   8%|▊         | 512/6470 [06:14<1:08:40,  1.45it/s]
3|513|Loss: 0.8736768960952759:   8%|▊         | 512/6470 [06:15<1:08:40,  1.45it/s]
3|513|Loss: 0.8736768960952759:   8%|▊         | 513/6470 [06:15<1:10:44,  1.40it/s]
3|514|Loss: 0.9447878003120422:   8%|▊         | 513/6470 [06:15<1:10:44,  1.40it/s]
3|514|Loss: 0.9447878003120422:   8%|▊         | 514/6470 [06:16<1:09:10,  1.44it/s]
3|515|Loss: 0.8654811978340149:   8%|▊         | 514/6470 [06:16<1:09:10,  1.44it/s]
3|515|Loss: 0.8654811978340149:   8%|▊         | 515/6470 [06:16<1:06:37,  1.49it/s]
3|516|Loss: 0.855023980140686:   8%|▊         | 515/6470 [06:17<1:06:37,  1.49it/s] 
3|516|Loss: 0.855023980140686:   8%|▊         | 516/6470 [06:17<1:06:12,  1.50it/s]
3|517|Loss: 0.7713193297386169:   8%|▊         | 516/6470 [06:17<1:06:12,  1.50it/s]
3|517|Loss: 0.7713193297386169:   8%|▊         | 517/6470 [06:18<1:06:55,  1.48it/s]
3|518|Loss: 0.7544591426849365:   8%|▊         | 517/6470 [06:18<1:06:55,  1.48it/s]
3|518|Loss: 0.7544591426849365:   8%|▊         | 518/6470 [06:18<1:11:24,  1.39it/s]
3|519|Loss: 0.8092119693756104:   8%|▊         | 518/6470 [06:19<1:11:24,  1.39it/s]
3|519|Loss: 0.8092119693756104:   8%|▊         | 519/6470 [06:19<1:17:34,  1.28it/s]
3|520|Loss: 0.8317373991012573:   8%|▊         | 519/6470 [06:20<1:17:34,  1.28it/s]
3|520|Loss: 0.8317373991012573:   8%|▊         | 520/6470 [06:20<1:21:47,  1.21it/s]
3|521|Loss: 0.8847742676734924:   8%|▊         | 520/6470 [06:21<1:21:47,  1.21it/s]
3|521|Loss: 0.8847742676734924:   8%|▊         | 521/6470 [06:21<1:16:20,  1.30it/s]
3|522|Loss: 0.8418097496032715:   8%|▊         | 521/6470 [06:21<1:16:20,  1.30it/s]
3|522|Loss: 0.8418097496032715:   8%|▊         | 522/6470 [06:22<1:13:06,  1.36it/s]
3|523|Loss: 0.8762825131416321:   8%|▊         | 522/6470 [06:22<1:13:06,  1.36it/s]
3|523|Loss: 0.8762825131416321:   8%|▊         | 523/6470 [06:22<1:17:47,  1.27it/s]
3|524|Loss: 0.8915716409683228:   8%|▊         | 523/6470 [06:23<1:17:47,  1.27it/s]
3|524|Loss: 0.8915716409683228:   8%|▊         | 524/6470 [06:23<1:21:54,  1.21it/s]
3|525|Loss: 0.9363885521888733:   8%|▊         | 524/6470 [06:24<1:21:54,  1.21it/s]
3|525|Loss: 0.9363885521888733:   8%|▊         | 525/6470 [06:24<1:17:55,  1.27it/s]
3|526|Loss: 0.8923136591911316:   8%|▊         | 525/6470 [06:24<1:17:55,  1.27it/s]
3|526|Loss: 0.8923136591911316:   8%|▊         | 526/6470 [06:25<1:16:41,  1.29it/s]
3|527|Loss: 0.6992822885513306:   8%|▊         | 526/6470 [06:25<1:16:41,  1.29it/s]
3|527|Loss: 0.6992822885513306:   8%|▊         | 527/6470 [06:26<1:18:27,  1.26it/s]
3|528|Loss: 0.7196410894393921:   8%|▊         | 527/6470 [06:26<1:18:27,  1.26it/s]
3|528|Loss: 0.7196410894393921:   8%|▊         | 528/6470 [06:26<1:18:50,  1.26it/s]
3|529|Loss: 0.8050921559333801:   8%|▊         | 528/6470 [06:27<1:18:50,  1.26it/s]
3|529|Loss: 0.8050921559333801:   8%|▊         | 529/6470 [06:27<1:17:44,  1.27it/s]
3|530|Loss: 0.7540622353553772:   8%|▊         | 529/6470 [06:28<1:17:44,  1.27it/s]
3|530|Loss: 0.7540622353553772:   8%|▊         | 530/6470 [06:28<1:16:33,  1.29it/s]
3|531|Loss: 0.9311174154281616:   8%|▊         | 530/6470 [06:28<1:16:33,  1.29it/s]
3|531|Loss: 0.9311174154281616:   8%|▊         | 531/6470 [06:29<1:12:03,  1.37it/s]
3|532|Loss: 0.8459338545799255:   8%|▊         | 531/6470 [06:29<1:12:03,  1.37it/s]
3|532|Loss: 0.8459338545799255:   8%|▊         | 532/6470 [06:29<1:16:22,  1.30it/s]
3|533|Loss: 0.672382116317749:   8%|▊         | 532/6470 [06:30<1:16:22,  1.30it/s] 
3|533|Loss: 0.672382116317749:   8%|▊         | 533/6470 [06:30<1:22:52,  1.19it/s]
3|534|Loss: 0.8051279783248901:   8%|▊         | 533/6470 [06:31<1:22:52,  1.19it/s]
3|534|Loss: 0.8051279783248901:   8%|▊         | 534/6470 [06:31<1:20:47,  1.22it/s]
3|535|Loss: 1.0301097631454468:   8%|▊         | 534/6470 [06:32<1:20:47,  1.22it/s]
3|535|Loss: 1.0301097631454468:   8%|▊         | 535/6470 [06:32<1:19:01,  1.25it/s]
3|536|Loss: 0.8078538179397583:   8%|▊         | 535/6470 [06:32<1:19:01,  1.25it/s]
3|536|Loss: 0.8078538179397583:   8%|▊         | 536/6470 [06:33<1:12:01,  1.37it/s]
3|537|Loss: 0.836234986782074:   8%|▊         | 536/6470 [06:33<1:12:01,  1.37it/s] 
3|537|Loss: 0.836234986782074:   8%|▊         | 537/6470 [06:33<1:16:37,  1.29it/s]
3|538|Loss: 0.792214572429657:   8%|▊         | 537/6470 [06:34<1:16:37,  1.29it/s]
3|538|Loss: 0.792214572429657:   8%|▊         | 538/6470 [06:34<1:18:18,  1.26it/s]
3|539|Loss: 0.8516321182250977:   8%|▊         | 538/6470 [06:35<1:18:18,  1.26it/s]
3|539|Loss: 0.8516321182250977:   8%|▊         | 539/6470 [06:35<1:19:11,  1.25it/s]
3|540|Loss: 0.7599494457244873:   8%|▊         | 539/6470 [06:36<1:19:11,  1.25it/s]
3|540|Loss: 0.7599494457244873:   8%|▊         | 540/6470 [06:36<1:16:44,  1.29it/s]
3|541|Loss: 0.7622167468070984:   8%|▊         | 540/6470 [06:36<1:16:44,  1.29it/s]
3|541|Loss: 0.7622167468070984:   8%|▊         | 541/6470 [06:37<1:16:37,  1.29it/s]
3|542|Loss: 0.8852941393852234:   8%|▊         | 541/6470 [06:37<1:16:37,  1.29it/s]
3|542|Loss: 0.8852941393852234:   8%|▊         | 542/6470 [06:37<1:16:40,  1.29it/s]
3|543|Loss: 0.8371662497520447:   8%|▊         | 542/6470 [06:38<1:16:40,  1.29it/s]
3|543|Loss: 0.8371662497520447:   8%|▊         | 543/6470 [06:38<1:11:12,  1.39it/s]
3|544|Loss: 0.7960728406906128:   8%|▊         | 543/6470 [06:38<1:11:12,  1.39it/s]
3|544|Loss: 0.7960728406906128:   8%|▊         | 544/6470 [06:39<1:12:04,  1.37it/s]
3|545|Loss: 0.843798816204071:   8%|▊         | 544/6470 [06:39<1:12:04,  1.37it/s] 
3|545|Loss: 0.843798816204071:   8%|▊         | 545/6470 [06:39<1:09:16,  1.43it/s]
3|546|Loss: 0.857063353061676:   8%|▊         | 545/6470 [06:40<1:09:16,  1.43it/s]
3|546|Loss: 0.857063353061676:   8%|▊         | 546/6470 [06:40<1:16:32,  1.29it/s]
3|547|Loss: 0.854250431060791:   8%|▊         | 546/6470 [06:41<1:16:32,  1.29it/s]
3|547|Loss: 0.854250431060791:   8%|▊         | 547/6470 [06:41<1:20:07,  1.23it/s]
3|548|Loss: 0.970865786075592:   8%|▊         | 547/6470 [06:42<1:20:07,  1.23it/s]
3|548|Loss: 0.970865786075592:   8%|▊         | 548/6470 [06:42<1:19:21,  1.24it/s]
3|549|Loss: 0.8966195583343506:   8%|▊         | 548/6470 [06:42<1:19:21,  1.24it/s]
3|549|Loss: 0.8966195583343506:   8%|▊         | 549/6470 [06:43<1:18:33,  1.26it/s]
3|550|Loss: 0.8345010280609131:   8%|▊         | 549/6470 [06:43<1:18:33,  1.26it/s]
3|550|Loss: 0.8345010280609131:   9%|▊         | 550/6470 [06:44<1:24:35,  1.17it/s]
3|551|Loss: 0.8130871653556824:   9%|▊         | 550/6470 [06:44<1:24:35,  1.17it/s]
3|551|Loss: 0.8130871653556824:   9%|▊         | 551/6470 [06:45<1:24:55,  1.16it/s]
3|552|Loss: 0.8519920706748962:   9%|▊         | 551/6470 [06:45<1:24:55,  1.16it/s]
3|552|Loss: 0.8519920706748962:   9%|▊         | 552/6470 [06:45<1:20:21,  1.23it/s]
3|553|Loss: 0.6991495490074158:   9%|▊         | 552/6470 [06:46<1:20:21,  1.23it/s]
3|553|Loss: 0.6991495490074158:   9%|▊         | 553/6470 [06:46<1:15:31,  1.31it/s]
3|554|Loss: 0.7988610863685608:   9%|▊         | 553/6470 [06:46<1:15:31,  1.31it/s]
3|554|Loss: 0.7988610863685608:   9%|▊         | 554/6470 [06:47<1:13:12,  1.35it/s]
3|555|Loss: 0.8622074127197266:   9%|▊         | 554/6470 [06:47<1:13:12,  1.35it/s]
3|555|Loss: 0.8622074127197266:   9%|▊         | 555/6470 [06:47<1:11:41,  1.38it/s]
3|556|Loss: 0.7767636179924011:   9%|▊         | 555/6470 [06:48<1:11:41,  1.38it/s]
3|556|Loss: 0.7767636179924011:   9%|▊         | 556/6470 [06:48<1:12:39,  1.36it/s]
3|557|Loss: 0.8125385642051697:   9%|▊         | 556/6470 [06:49<1:12:39,  1.36it/s]
3|557|Loss: 0.8125385642051697:   9%|▊         | 557/6470 [06:49<1:11:01,  1.39it/s]
3|558|Loss: 0.8123297095298767:   9%|▊         | 557/6470 [06:49<1:11:01,  1.39it/s]
3|558|Loss: 0.8123297095298767:   9%|▊         | 558/6470 [06:50<1:18:18,  1.26it/s]
3|559|Loss: 1.084085464477539:   9%|▊         | 558/6470 [06:50<1:18:18,  1.26it/s] 
3|559|Loss: 1.084085464477539:   9%|▊         | 559/6470 [06:50<1:15:36,  1.30it/s]
3|560|Loss: 0.7722013592720032:   9%|▊         | 559/6470 [06:51<1:15:36,  1.30it/s]
3|560|Loss: 0.7722013592720032:   9%|▊         | 560/6470 [06:51<1:10:26,  1.40it/s]
3|561|Loss: 0.8373995423316956:   9%|▊         | 560/6470 [06:51<1:10:26,  1.40it/s]
3|561|Loss: 0.8373995423316956:   9%|▊         | 561/6470 [06:52<1:10:40,  1.39it/s]
3|562|Loss: 0.9241076111793518:   9%|▊         | 561/6470 [06:52<1:10:40,  1.39it/s]
3|562|Loss: 0.9241076111793518:   9%|▊         | 562/6470 [06:52<1:08:42,  1.43it/s]
3|563|Loss: 0.8224796056747437:   9%|▊         | 562/6470 [06:53<1:08:42,  1.43it/s]
3|563|Loss: 0.8224796056747437:   9%|▊         | 563/6470 [06:53<1:10:56,  1.39it/s]
3|564|Loss: 0.8371184468269348:   9%|▊         | 563/6470 [06:54<1:10:56,  1.39it/s]
3|564|Loss: 0.8371184468269348:   9%|▊         | 564/6470 [06:54<1:10:19,  1.40it/s]
3|565|Loss: 0.9182059168815613:   9%|▊         | 564/6470 [06:54<1:10:19,  1.40it/s]
3|565|Loss: 0.9182059168815613:   9%|▊         | 565/6470 [06:55<1:07:50,  1.45it/s]
3|566|Loss: 0.8760290145874023:   9%|▊         | 565/6470 [06:55<1:07:50,  1.45it/s]
3|566|Loss: 0.8760290145874023:   9%|▊         | 566/6470 [06:55<1:06:44,  1.47it/s]
3|567|Loss: 0.7212420701980591:   9%|▊         | 566/6470 [06:56<1:06:44,  1.47it/s]
3|567|Loss: 0.7212420701980591:   9%|▉         | 567/6470 [06:56<1:09:03,  1.42it/s]
3|568|Loss: 0.9094741940498352:   9%|▉         | 567/6470 [06:56<1:09:03,  1.42it/s]
3|568|Loss: 0.9094741940498352:   9%|▉         | 568/6470 [06:57<1:09:14,  1.42it/s]
3|569|Loss: 0.8314800262451172:   9%|▉         | 568/6470 [06:57<1:09:14,  1.42it/s]
3|569|Loss: 0.8314800262451172:   9%|▉         | 569/6470 [06:57<1:11:48,  1.37it/s]
3|570|Loss: 0.7446624636650085:   9%|▉         | 569/6470 [06:58<1:11:48,  1.37it/s]
3|570|Loss: 0.7446624636650085:   9%|▉         | 570/6470 [06:58<1:13:12,  1.34it/s]
3|571|Loss: 0.6733399629592896:   9%|▉         | 570/6470 [06:59<1:13:12,  1.34it/s]
3|571|Loss: 0.6733399629592896:   9%|▉         | 571/6470 [06:59<1:17:52,  1.26it/s]
3|572|Loss: 0.8936796188354492:   9%|▉         | 571/6470 [07:00<1:17:52,  1.26it/s]
3|572|Loss: 0.8936796188354492:   9%|▉         | 572/6470 [07:00<1:20:48,  1.22it/s]
3|573|Loss: 0.747176468372345:   9%|▉         | 572/6470 [07:01<1:20:48,  1.22it/s] 
3|573|Loss: 0.747176468372345:   9%|▉         | 573/6470 [07:01<1:25:18,  1.15it/s]
3|574|Loss: 0.847246527671814:   9%|▉         | 573/6470 [07:01<1:25:18,  1.15it/s]
3|574|Loss: 0.847246527671814:   9%|▉         | 574/6470 [07:02<1:20:14,  1.22it/s]
3|575|Loss: 0.8854516744613647:   9%|▉         | 574/6470 [07:02<1:20:14,  1.22it/s]
3|575|Loss: 0.8854516744613647:   9%|▉         | 575/6470 [07:03<1:22:30,  1.19it/s]
3|576|Loss: 0.80369633436203:   9%|▉         | 575/6470 [07:03<1:22:30,  1.19it/s]  
3|576|Loss: 0.80369633436203:   9%|▉         | 576/6470 [07:03<1:23:39,  1.17it/s]
3|577|Loss: 0.8231057524681091:   9%|▉         | 576/6470 [07:04<1:23:39,  1.17it/s]
3|577|Loss: 0.8231057524681091:   9%|▉         | 577/6470 [07:04<1:21:37,  1.20it/s]
3|578|Loss: 0.8046430349349976:   9%|▉         | 577/6470 [07:05<1:21:37,  1.20it/s]
3|578|Loss: 0.8046430349349976:   9%|▉         | 578/6470 [07:05<1:17:49,  1.26it/s]
3|579|Loss: 0.7279143333435059:   9%|▉         | 578/6470 [07:05<1:17:49,  1.26it/s]
3|579|Loss: 0.7279143333435059:   9%|▉         | 579/6470 [07:06<1:17:46,  1.26it/s]
3|580|Loss: 0.8768903017044067:   9%|▉         | 579/6470 [07:06<1:17:46,  1.26it/s]
3|580|Loss: 0.8768903017044067:   9%|▉         | 580/6470 [07:07<1:19:29,  1.24it/s]
3|581|Loss: 0.8715370297431946:   9%|▉         | 580/6470 [07:07<1:19:29,  1.24it/s]
3|581|Loss: 0.8715370297431946:   9%|▉         | 581/6470 [07:07<1:18:39,  1.25it/s]
3|582|Loss: 0.8667513132095337:   9%|▉         | 581/6470 [07:08<1:18:39,  1.25it/s]
3|582|Loss: 0.8667513132095337:   9%|▉         | 582/6470 [07:08<1:15:09,  1.31it/s]
3|583|Loss: 0.9247465133666992:   9%|▉         | 582/6470 [07:08<1:15:09,  1.31it/s]
3|583|Loss: 0.9247465133666992:   9%|▉         | 583/6470 [07:09<1:15:00,  1.31it/s]
3|584|Loss: 0.7410717010498047:   9%|▉         | 583/6470 [07:09<1:15:00,  1.31it/s]
3|584|Loss: 0.7410717010498047:   9%|▉         | 584/6470 [07:09<1:09:44,  1.41it/s]
3|585|Loss: 0.984697163105011:   9%|▉         | 584/6470 [07:10<1:09:44,  1.41it/s] 
3|585|Loss: 0.984697163105011:   9%|▉         | 585/6470 [07:10<1:09:14,  1.42it/s]
3|586|Loss: 0.8797054886817932:   9%|▉         | 585/6470 [07:11<1:09:14,  1.42it/s]
3|586|Loss: 0.8797054886817932:   9%|▉         | 586/6470 [07:11<1:10:54,  1.38it/s]
3|587|Loss: 0.7812979221343994:   9%|▉         | 586/6470 [07:11<1:10:54,  1.38it/s]
3|587|Loss: 0.7812979221343994:   9%|▉         | 587/6470 [07:12<1:13:11,  1.34it/s]
3|588|Loss: 0.9473111033439636:   9%|▉         | 587/6470 [07:12<1:13:11,  1.34it/s]
3|588|Loss: 0.9473111033439636:   9%|▉         | 588/6470 [07:12<1:15:05,  1.31it/s]
3|589|Loss: 0.905610203742981:   9%|▉         | 588/6470 [07:13<1:15:05,  1.31it/s] 
3|589|Loss: 0.905610203742981:   9%|▉         | 589/6470 [07:13<1:16:25,  1.28it/s]
3|590|Loss: 0.9814960956573486:   9%|▉         | 589/6470 [07:14<1:16:25,  1.28it/s]
3|590|Loss: 0.9814960956573486:   9%|▉         | 590/6470 [07:14<1:17:15,  1.27it/s]
3|591|Loss: 0.8239185810089111:   9%|▉         | 590/6470 [07:15<1:17:15,  1.27it/s]
3|591|Loss: 0.8239185810089111:   9%|▉         | 591/6470 [07:15<1:17:38,  1.26it/s]
3|592|Loss: 0.7560056447982788:   9%|▉         | 591/6470 [07:15<1:17:38,  1.26it/s]
3|592|Loss: 0.7560056447982788:   9%|▉         | 592/6470 [07:16<1:20:50,  1.21it/s]
3|593|Loss: 0.7994361519813538:   9%|▉         | 592/6470 [07:16<1:20:50,  1.21it/s]
3|593|Loss: 0.7994361519813538:   9%|▉         | 593/6470 [07:17<1:18:03,  1.25it/s]
3|594|Loss: 0.9126445651054382:   9%|▉         | 593/6470 [07:17<1:18:03,  1.25it/s]
3|594|Loss: 0.9126445651054382:   9%|▉         | 594/6470 [07:17<1:19:49,  1.23it/s]
3|595|Loss: 0.7884715795516968:   9%|▉         | 594/6470 [07:18<1:19:49,  1.23it/s]
3|595|Loss: 0.7884715795516968:   9%|▉         | 595/6470 [07:18<1:13:45,  1.33it/s]
3|596|Loss: 0.8113141059875488:   9%|▉         | 595/6470 [07:18<1:13:45,  1.33it/s]
3|596|Loss: 0.8113141059875488:   9%|▉         | 596/6470 [07:19<1:08:32,  1.43it/s]
3|597|Loss: 0.7080744504928589:   9%|▉         | 596/6470 [07:19<1:08:32,  1.43it/s]
3|597|Loss: 0.7080744504928589:   9%|▉         | 597/6470 [07:19<1:06:14,  1.48it/s]
3|598|Loss: 1.0162711143493652:   9%|▉         | 597/6470 [07:20<1:06:14,  1.48it/s]
3|598|Loss: 1.0162711143493652:   9%|▉         | 598/6470 [07:20<1:06:20,  1.48it/s]
3|599|Loss: 0.8246127963066101:   9%|▉         | 598/6470 [07:20<1:06:20,  1.48it/s]
3|599|Loss: 0.8246127963066101:   9%|▉         | 599/6470 [07:21<1:09:07,  1.42it/s]
3|600|Loss: 0.8189516067504883:   9%|▉         | 599/6470 [07:21<1:09:07,  1.42it/s]
3|600|Loss: 0.8189516067504883:   9%|▉         | 600/6470 [07:21<1:08:49,  1.42it/s]
3|601|Loss: 0.7608400583267212:   9%|▉         | 600/6470 [07:22<1:08:49,  1.42it/s]
3|601|Loss: 0.7608400583267212:   9%|▉         | 601/6470 [07:22<1:06:28,  1.47it/s]
3|602|Loss: 0.7894588708877563:   9%|▉         | 601/6470 [07:22<1:06:28,  1.47it/s]
3|602|Loss: 0.7894588708877563:   9%|▉         | 602/6470 [07:23<1:10:13,  1.39it/s]
3|603|Loss: 0.9113376140594482:   9%|▉         | 602/6470 [07:23<1:10:13,  1.39it/s]
3|603|Loss: 0.9113376140594482:   9%|▉         | 603/6470 [07:23<1:08:13,  1.43it/s]
3|604|Loss: 0.9261848330497742:   9%|▉         | 603/6470 [07:24<1:08:13,  1.43it/s]
3|604|Loss: 0.9261848330497742:   9%|▉         | 604/6470 [07:24<1:09:52,  1.40it/s]
3|605|Loss: 0.8273986577987671:   9%|▉         | 604/6470 [07:25<1:09:52,  1.40it/s]
3|605|Loss: 0.8273986577987671:   9%|▉         | 605/6470 [07:25<1:12:12,  1.35it/s]
3|606|Loss: 0.7829453945159912:   9%|▉         | 605/6470 [07:25<1:12:12,  1.35it/s]
3|606|Loss: 0.7829453945159912:   9%|▉         | 606/6470 [07:26<1:14:22,  1.31it/s]
3|607|Loss: 0.8529184460639954:   9%|▉         | 606/6470 [07:26<1:14:22,  1.31it/s]
3|607|Loss: 0.8529184460639954:   9%|▉         | 607/6470 [07:27<1:15:28,  1.29it/s]
3|608|Loss: 0.8858090043067932:   9%|▉         | 607/6470 [07:27<1:15:28,  1.29it/s]
3|608|Loss: 0.8858090043067932:   9%|▉         | 608/6470 [07:27<1:16:04,  1.28it/s]
3|609|Loss: 0.7889244556427002:   9%|▉         | 608/6470 [07:28<1:16:04,  1.28it/s]
3|609|Loss: 0.7889244556427002:   9%|▉         | 609/6470 [07:28<1:10:37,  1.38it/s]
3|610|Loss: 0.8105860948562622:   9%|▉         | 609/6470 [07:28<1:10:37,  1.38it/s]
3|610|Loss: 0.8105860948562622:   9%|▉         | 610/6470 [07:29<1:10:29,  1.39it/s]
3|611|Loss: 0.6848050951957703:   9%|▉         | 610/6470 [07:29<1:10:29,  1.39it/s]
3|611|Loss: 0.6848050951957703:   9%|▉         | 611/6470 [07:29<1:12:10,  1.35it/s]
3|612|Loss: 0.9215608835220337:   9%|▉         | 611/6470 [07:30<1:12:10,  1.35it/s]
3|612|Loss: 0.9215608835220337:   9%|▉         | 612/6470 [07:30<1:18:05,  1.25it/s]
3|613|Loss: 0.7436804175376892:   9%|▉         | 612/6470 [07:31<1:18:05,  1.25it/s]
3|613|Loss: 0.7436804175376892:   9%|▉         | 613/6470 [07:31<1:19:31,  1.23it/s]
3|614|Loss: 0.8296946287155151:   9%|▉         | 613/6470 [07:32<1:19:31,  1.23it/s]
3|614|Loss: 0.8296946287155151:   9%|▉         | 614/6470 [07:32<1:16:49,  1.27it/s]
3|615|Loss: 0.7559872269630432:   9%|▉         | 614/6470 [07:32<1:16:49,  1.27it/s]
3|615|Loss: 0.7559872269630432:  10%|▉         | 615/6470 [07:33<1:15:18,  1.30it/s]
3|616|Loss: 0.904551088809967:  10%|▉         | 615/6470 [07:33<1:15:18,  1.30it/s] 
3|616|Loss: 0.904551088809967:  10%|▉         | 616/6470 [07:34<1:18:08,  1.25it/s]
3|617|Loss: 0.8116522431373596:  10%|▉         | 616/6470 [07:34<1:18:08,  1.25it/s]
3|617|Loss: 0.8116522431373596:  10%|▉         | 617/6470 [07:34<1:12:39,  1.34it/s]
3|618|Loss: 0.8744696378707886:  10%|▉         | 617/6470 [07:35<1:12:39,  1.34it/s]
3|618|Loss: 0.8744696378707886:  10%|▉         | 618/6470 [07:35<1:10:22,  1.39it/s]
3|619|Loss: 0.8142802715301514:  10%|▉         | 618/6470 [07:35<1:10:22,  1.39it/s]
3|619|Loss: 0.8142802715301514:  10%|▉         | 619/6470 [07:36<1:22:01,  1.19it/s]
3|620|Loss: 0.8448778390884399:  10%|▉         | 619/6470 [07:36<1:22:01,  1.19it/s]
3|620|Loss: 0.8448778390884399:  10%|▉         | 620/6470 [07:37<1:20:55,  1.20it/s]
3|621|Loss: 0.8918508291244507:  10%|▉         | 620/6470 [07:37<1:20:55,  1.20it/s]
3|621|Loss: 0.8918508291244507:  10%|▉         | 621/6470 [07:38<1:18:23,  1.24it/s]
3|622|Loss: 0.7766572833061218:  10%|▉         | 621/6470 [07:38<1:18:23,  1.24it/s]
3|622|Loss: 0.7766572833061218:  10%|▉         | 622/6470 [07:38<1:22:22,  1.18it/s]
3|623|Loss: 0.8461629152297974:  10%|▉         | 622/6470 [07:39<1:22:22,  1.18it/s]
3|623|Loss: 0.8461629152297974:  10%|▉         | 623/6470 [07:39<1:22:07,  1.19it/s]
3|624|Loss: 0.8681082725524902:  10%|▉         | 623/6470 [07:40<1:22:07,  1.19it/s]
3|624|Loss: 0.8681082725524902:  10%|▉         | 624/6470 [07:40<1:21:04,  1.20it/s]
3|625|Loss: 0.692925214767456:  10%|▉         | 624/6470 [07:41<1:21:04,  1.20it/s] 
3|625|Loss: 0.692925214767456:  10%|▉         | 625/6470 [07:41<1:23:04,  1.17it/s]
3|626|Loss: 0.7050442099571228:  10%|▉         | 625/6470 [07:41<1:23:04,  1.17it/s]
3|626|Loss: 0.7050442099571228:  10%|▉         | 626/6470 [07:42<1:19:56,  1.22it/s]
3|627|Loss: 0.8458251357078552:  10%|▉         | 626/6470 [07:42<1:19:56,  1.22it/s]
3|627|Loss: 0.8458251357078552:  10%|▉         | 627/6470 [07:43<1:17:48,  1.25it/s]
3|628|Loss: 0.6996230483055115:  10%|▉         | 627/6470 [07:43<1:17:48,  1.25it/s]
3|628|Loss: 0.6996230483055115:  10%|▉         | 628/6470 [07:43<1:12:34,  1.34it/s]
3|629|Loss: 0.9051300287246704:  10%|▉         | 628/6470 [07:44<1:12:34,  1.34it/s]
3|629|Loss: 0.9051300287246704:  10%|▉         | 629/6470 [07:44<1:14:02,  1.31it/s]
3|630|Loss: 0.7884871363639832:  10%|▉         | 629/6470 [07:44<1:14:02,  1.31it/s]
3|630|Loss: 0.7884871363639832:  10%|▉         | 630/6470 [07:45<1:14:56,  1.30it/s]
3|631|Loss: 0.8328677415847778:  10%|▉         | 630/6470 [07:45<1:14:56,  1.30it/s]
3|631|Loss: 0.8328677415847778:  10%|▉         | 631/6470 [07:45<1:12:24,  1.34it/s]
3|632|Loss: 0.7497339248657227:  10%|▉         | 631/6470 [07:46<1:12:24,  1.34it/s]
3|632|Loss: 0.7497339248657227:  10%|▉         | 632/6470 [07:46<1:05:33,  1.48it/s]
3|633|Loss: 0.812629222869873:  10%|▉         | 632/6470 [07:46<1:05:33,  1.48it/s] 
3|633|Loss: 0.812629222869873:  10%|▉         | 633/6470 [07:47<1:04:22,  1.51it/s]
3|634|Loss: 0.8221681714057922:  10%|▉         | 633/6470 [07:47<1:04:22,  1.51it/s]
3|634|Loss: 0.8221681714057922:  10%|▉         | 634/6470 [07:47<1:05:27,  1.49it/s]
3|635|Loss: 0.6874627470970154:  10%|▉         | 634/6470 [07:48<1:05:27,  1.49it/s]
3|635|Loss: 0.6874627470970154:  10%|▉         | 635/6470 [07:48<1:05:24,  1.49it/s]
3|636|Loss: 0.6302545666694641:  10%|▉         | 635/6470 [07:48<1:05:24,  1.49it/s]
3|636|Loss: 0.6302545666694641:  10%|▉         | 636/6470 [07:49<1:11:26,  1.36it/s]
3|637|Loss: 0.8323491215705872:  10%|▉         | 636/6470 [07:49<1:11:26,  1.36it/s]
3|637|Loss: 0.8323491215705872:  10%|▉         | 637/6470 [07:50<1:15:48,  1.28it/s]
3|638|Loss: 0.8801706433296204:  10%|▉         | 637/6470 [07:50<1:15:48,  1.28it/s]
3|638|Loss: 0.8801706433296204:  10%|▉         | 638/6470 [07:50<1:14:04,  1.31it/s]
3|639|Loss: 0.8346521258354187:  10%|▉         | 638/6470 [07:51<1:14:04,  1.31it/s]
3|639|Loss: 0.8346521258354187:  10%|▉         | 639/6470 [07:51<1:11:17,  1.36it/s]
3|640|Loss: 0.9341841340065002:  10%|▉         | 639/6470 [07:52<1:11:17,  1.36it/s]
3|640|Loss: 0.9341841340065002:  10%|▉         | 640/6470 [07:52<1:14:11,  1.31it/s]
3|641|Loss: 0.8273423910140991:  10%|▉         | 640/6470 [07:52<1:14:11,  1.31it/s]
3|641|Loss: 0.8273423910140991:  10%|▉         | 641/6470 [07:53<1:10:44,  1.37it/s]
3|642|Loss: 0.8292809724807739:  10%|▉         | 641/6470 [07:53<1:10:44,  1.37it/s]
3|642|Loss: 0.8292809724807739:  10%|▉         | 642/6470 [07:53<1:11:10,  1.36it/s]
3|643|Loss: 0.7327747344970703:  10%|▉         | 642/6470 [07:54<1:11:10,  1.36it/s]
3|643|Loss: 0.7327747344970703:  10%|▉         | 643/6470 [07:54<1:11:52,  1.35it/s]
3|644|Loss: 0.7847638726234436:  10%|▉         | 643/6470 [07:54<1:11:52,  1.35it/s]
3|644|Loss: 0.7847638726234436:  10%|▉         | 644/6470 [07:55<1:12:57,  1.33it/s]
3|645|Loss: 0.8548863530158997:  10%|▉         | 644/6470 [07:55<1:12:57,  1.33it/s]
3|645|Loss: 0.8548863530158997:  10%|▉         | 645/6470 [07:56<1:13:07,  1.33it/s]
3|646|Loss: 0.9158021211624146:  10%|▉         | 645/6470 [07:56<1:13:07,  1.33it/s]
3|646|Loss: 0.9158021211624146:  10%|▉         | 646/6470 [07:56<1:12:11,  1.34it/s]
3|647|Loss: 0.8410780429840088:  10%|▉         | 646/6470 [07:57<1:12:11,  1.34it/s]
3|647|Loss: 0.8410780429840088:  10%|█         | 647/6470 [07:57<1:10:20,  1.38it/s]
3|648|Loss: 0.646163284778595:  10%|█         | 647/6470 [07:57<1:10:20,  1.38it/s] 
3|648|Loss: 0.646163284778595:  10%|█         | 648/6470 [07:58<1:04:04,  1.51it/s]
3|649|Loss: 0.8959662318229675:  10%|█         | 648/6470 [07:58<1:04:04,  1.51it/s]
3|649|Loss: 0.8959662318229675:  10%|█         | 649/6470 [07:58<1:11:42,  1.35it/s]
3|650|Loss: 0.7884665131568909:  10%|█         | 649/6470 [07:59<1:11:42,  1.35it/s]
3|650|Loss: 0.7884665131568909:  10%|█         | 650/6470 [07:59<1:16:28,  1.27it/s]
3|651|Loss: 0.8880051970481873:  10%|█         | 650/6470 [08:00<1:16:28,  1.27it/s]
3|651|Loss: 0.8880051970481873:  10%|█         | 651/6470 [08:00<1:18:49,  1.23it/s]
3|652|Loss: 0.8689976334571838:  10%|█         | 651/6470 [08:01<1:18:49,  1.23it/s]
3|652|Loss: 0.8689976334571838:  10%|█         | 652/6470 [08:01<1:21:15,  1.19it/s]
3|653|Loss: 0.7925991415977478:  10%|█         | 652/6470 [08:02<1:21:15,  1.19it/s]
3|653|Loss: 0.7925991415977478:  10%|█         | 653/6470 [08:02<1:19:25,  1.22it/s]
3|654|Loss: 0.8406803011894226:  10%|█         | 653/6470 [08:02<1:19:25,  1.22it/s]
3|654|Loss: 0.8406803011894226:  10%|█         | 654/6470 [08:03<1:19:41,  1.22it/s]
3|655|Loss: 0.7609453797340393:  10%|█         | 654/6470 [08:03<1:19:41,  1.22it/s]
3|655|Loss: 0.7609453797340393:  10%|█         | 655/6470 [08:04<1:23:54,  1.16it/s]
3|656|Loss: 0.8103477954864502:  10%|█         | 655/6470 [08:04<1:23:54,  1.16it/s]
3|656|Loss: 0.8103477954864502:  10%|█         | 656/6470 [08:04<1:20:48,  1.20it/s]
3|657|Loss: 0.8682118058204651:  10%|█         | 656/6470 [08:05<1:20:48,  1.20it/s]
3|657|Loss: 0.8682118058204651:  10%|█         | 657/6470 [08:05<1:19:30,  1.22it/s]
3|658|Loss: 0.8147430419921875:  10%|█         | 657/6470 [08:06<1:19:30,  1.22it/s]
3|658|Loss: 0.8147430419921875:  10%|█         | 658/6470 [08:06<1:14:49,  1.29it/s]
3|659|Loss: 0.8061031699180603:  10%|█         | 658/6470 [08:06<1:14:49,  1.29it/s]
3|659|Loss: 0.8061031699180603:  10%|█         | 659/6470 [08:07<1:13:15,  1.32it/s]
3|660|Loss: 0.8123133182525635:  10%|█         | 659/6470 [08:07<1:13:15,  1.32it/s]
3|660|Loss: 0.8123133182525635:  10%|█         | 660/6470 [08:07<1:14:14,  1.30it/s]
3|661|Loss: 0.9199118614196777:  10%|█         | 660/6470 [08:08<1:14:14,  1.30it/s]
3|661|Loss: 0.9199118614196777:  10%|█         | 661/6470 [08:08<1:17:33,  1.25it/s]
3|662|Loss: 0.714410126209259:  10%|█         | 661/6470 [08:09<1:17:33,  1.25it/s] 
3|662|Loss: 0.714410126209259:  10%|█         | 662/6470 [08:09<1:11:20,  1.36it/s]
3|663|Loss: 0.8230082988739014:  10%|█         | 662/6470 [08:09<1:11:20,  1.36it/s]
3|663|Loss: 0.8230082988739014:  10%|█         | 663/6470 [08:10<1:15:57,  1.27it/s]
3|664|Loss: 0.8217858672142029:  10%|█         | 663/6470 [08:10<1:15:57,  1.27it/s]
3|664|Loss: 0.8217858672142029:  10%|█         | 664/6470 [08:11<1:19:34,  1.22it/s]
3|665|Loss: 0.860469400882721:  10%|█         | 664/6470 [08:11<1:19:34,  1.22it/s] 
3|665|Loss: 0.860469400882721:  10%|█         | 665/6470 [08:12<1:22:07,  1.18it/s]
3|666|Loss: 0.8131681680679321:  10%|█         | 665/6470 [08:12<1:22:07,  1.18it/s]
3|666|Loss: 0.8131681680679321:  10%|█         | 666/6470 [08:12<1:20:34,  1.20it/s]
3|667|Loss: 0.6465193033218384:  10%|█         | 666/6470 [08:13<1:20:34,  1.20it/s]
3|667|Loss: 0.6465193033218384:  10%|█         | 667/6470 [08:13<1:22:09,  1.18it/s]
3|668|Loss: 0.824824869632721:  10%|█         | 667/6470 [08:14<1:22:09,  1.18it/s] 
3|668|Loss: 0.824824869632721:  10%|█         | 668/6470 [08:14<1:18:33,  1.23it/s]
3|669|Loss: 0.7026334404945374:  10%|█         | 668/6470 [08:14<1:18:33,  1.23it/s]
3|669|Loss: 0.7026334404945374:  10%|█         | 669/6470 [08:14<1:09:39,  1.39it/s]
3|670|Loss: 0.9934000372886658:  10%|█         | 669/6470 [08:15<1:09:39,  1.39it/s]
3|670|Loss: 0.9934000372886658:  10%|█         | 670/6470 [08:15<1:05:29,  1.48it/s]
3|671|Loss: 0.9221761226654053:  10%|█         | 670/6470 [08:15<1:05:29,  1.48it/s]
3|671|Loss: 0.9221761226654053:  10%|█         | 671/6470 [08:16<1:08:20,  1.41it/s]
3|672|Loss: 0.8704593181610107:  10%|█         | 671/6470 [08:16<1:08:20,  1.41it/s]
3|672|Loss: 0.8704593181610107:  10%|█         | 672/6470 [08:16<1:05:19,  1.48it/s]
3|673|Loss: 0.8074909448623657:  10%|█         | 672/6470 [08:17<1:05:19,  1.48it/s]
3|673|Loss: 0.8074909448623657:  10%|█         | 673/6470 [08:17<1:01:39,  1.57it/s]
3|674|Loss: 0.837502658367157:  10%|█         | 673/6470 [08:17<1:01:39,  1.57it/s] 
3|674|Loss: 0.837502658367157:  10%|█         | 674/6470 [08:18<1:02:42,  1.54it/s]
3|675|Loss: 0.8360626697540283:  10%|█         | 674/6470 [08:18<1:02:42,  1.54it/s]
3|675|Loss: 0.8360626697540283:  10%|█         | 675/6470 [08:18<1:06:44,  1.45it/s]
3|676|Loss: 0.7125714421272278:  10%|█         | 675/6470 [08:19<1:06:44,  1.45it/s]
3|676|Loss: 0.7125714421272278:  10%|█         | 676/6470 [08:19<1:09:28,  1.39it/s]
3|677|Loss: 0.8277264833450317:  10%|█         | 676/6470 [08:20<1:09:28,  1.39it/s]
3|677|Loss: 0.8277264833450317:  10%|█         | 677/6470 [08:20<1:08:58,  1.40it/s]
3|678|Loss: 0.9080296754837036:  10%|█         | 677/6470 [08:20<1:08:58,  1.40it/s]
3|678|Loss: 0.9080296754837036:  10%|█         | 678/6470 [08:21<1:09:57,  1.38it/s]
3|679|Loss: 0.8236453533172607:  10%|█         | 678/6470 [08:21<1:09:57,  1.38it/s]
3|679|Loss: 0.8236453533172607:  10%|█         | 679/6470 [08:21<1:08:04,  1.42it/s]
3|680|Loss: 0.8073064684867859:  10%|█         | 679/6470 [08:22<1:08:04,  1.42it/s]
3|680|Loss: 0.8073064684867859:  11%|█         | 680/6470 [08:22<1:12:40,  1.33it/s]
3|681|Loss: 0.7814175486564636:  11%|█         | 680/6470 [08:23<1:12:40,  1.33it/s]
3|681|Loss: 0.7814175486564636:  11%|█         | 681/6470 [08:23<1:15:18,  1.28it/s]
3|682|Loss: 0.8091760277748108:  11%|█         | 681/6470 [08:23<1:15:18,  1.28it/s]
3|682|Loss: 0.8091760277748108:  11%|█         | 682/6470 [08:24<1:12:03,  1.34it/s]
3|683|Loss: 0.7812825441360474:  11%|█         | 682/6470 [08:24<1:12:03,  1.34it/s]
3|683|Loss: 0.7812825441360474:  11%|█         | 683/6470 [08:24<1:10:06,  1.38it/s]
3|684|Loss: 1.1102230548858643:  11%|█         | 683/6470 [08:25<1:10:06,  1.38it/s]
3|684|Loss: 1.1102230548858643:  11%|█         | 684/6470 [08:25<1:07:48,  1.42it/s]
3|685|Loss: 0.8344950079917908:  11%|█         | 684/6470 [08:25<1:07:48,  1.42it/s]
3|685|Loss: 0.8344950079917908:  11%|█         | 685/6470 [08:26<1:09:35,  1.39it/s]
3|686|Loss: 0.6060401797294617:  11%|█         | 685/6470 [08:26<1:09:35,  1.39it/s]
3|686|Loss: 0.6060401797294617:  11%|█         | 686/6470 [08:27<1:16:10,  1.27it/s]
3|687|Loss: 1.0134129524230957:  11%|█         | 686/6470 [08:27<1:16:10,  1.27it/s]
3|687|Loss: 1.0134129524230957:  11%|█         | 687/6470 [08:28<1:17:16,  1.25it/s]
3|688|Loss: 0.8173754811286926:  11%|█         | 687/6470 [08:28<1:17:16,  1.25it/s]
3|688|Loss: 0.8173754811286926:  11%|█         | 688/6470 [08:28<1:17:06,  1.25it/s]
3|689|Loss: 0.7911908626556396:  11%|█         | 688/6470 [08:29<1:17:06,  1.25it/s]
3|689|Loss: 0.7911908626556396:  11%|█         | 689/6470 [08:29<1:07:14,  1.43it/s]
3|690|Loss: 0.9027342200279236:  11%|█         | 689/6470 [08:29<1:07:14,  1.43it/s]
3|690|Loss: 0.9027342200279236:  11%|█         | 690/6470 [08:29<1:03:08,  1.53it/s]
3|691|Loss: 0.8922460079193115:  11%|█         | 690/6470 [08:30<1:03:08,  1.53it/s]
3|691|Loss: 0.8922460079193115:  11%|█         | 691/6470 [08:30<1:10:23,  1.37it/s]
3|692|Loss: 0.8829375505447388:  11%|█         | 691/6470 [08:31<1:10:23,  1.37it/s]
3|692|Loss: 0.8829375505447388:  11%|█         | 692/6470 [08:31<1:07:50,  1.42it/s]
3|693|Loss: 0.8049207329750061:  11%|█         | 692/6470 [08:31<1:07:50,  1.42it/s]
3|693|Loss: 0.8049207329750061:  11%|█         | 693/6470 [08:32<1:10:48,  1.36it/s]
3|694|Loss: 0.9569753408432007:  11%|█         | 693/6470 [08:32<1:10:48,  1.36it/s]
3|694|Loss: 0.9569753408432007:  11%|█         | 694/6470 [08:33<1:10:31,  1.36it/s]
3|695|Loss: 0.9749078750610352:  11%|█         | 694/6470 [08:33<1:10:31,  1.36it/s]
3|695|Loss: 0.9749078750610352:  11%|█         | 695/6470 [08:33<1:10:56,  1.36it/s]
3|696|Loss: 0.8567951321601868:  11%|█         | 695/6470 [08:34<1:10:56,  1.36it/s]
3|696|Loss: 0.8567951321601868:  11%|█         | 696/6470 [08:34<1:08:15,  1.41it/s]
3|697|Loss: 0.9049988985061646:  11%|█         | 696/6470 [08:34<1:08:15,  1.41it/s]
3|697|Loss: 0.9049988985061646:  11%|█         | 697/6470 [08:35<1:13:37,  1.31it/s]
3|698|Loss: 0.9004958868026733:  11%|█         | 697/6470 [08:35<1:13:37,  1.31it/s]
3|698|Loss: 0.9004958868026733:  11%|█         | 698/6470 [08:36<1:15:23,  1.28it/s]
3|699|Loss: 0.8053765296936035:  11%|█         | 698/6470 [08:36<1:15:23,  1.28it/s]
3|699|Loss: 0.8053765296936035:  11%|█         | 699/6470 [08:36<1:12:41,  1.32it/s]
3|700|Loss: 0.7909830212593079:  11%|█         | 699/6470 [08:37<1:12:41,  1.32it/s]
3|700|Loss: 0.7909830212593079:  11%|█         | 700/6470 [08:37<1:10:31,  1.36it/s]
3|701|Loss: 0.871799111366272:  11%|█         | 700/6470 [08:37<1:10:31,  1.36it/s] 
3|701|Loss: 0.871799111366272:  11%|█         | 701/6470 [08:38<1:09:28,  1.38it/s]
3|702|Loss: 1.0673047304153442:  11%|█         | 701/6470 [08:38<1:09:28,  1.38it/s]
3|702|Loss: 1.0673047304153442:  11%|█         | 702/6470 [08:39<1:13:13,  1.31it/s]
3|703|Loss: 0.7933340072631836:  11%|█         | 702/6470 [08:39<1:13:13,  1.31it/s]
3|703|Loss: 0.7933340072631836:  11%|█         | 703/6470 [08:39<1:13:47,  1.30it/s]
3|704|Loss: 0.9367987513542175:  11%|█         | 703/6470 [08:40<1:13:47,  1.30it/s]
3|704|Loss: 0.9367987513542175:  11%|█         | 704/6470 [08:40<1:14:08,  1.30it/s]
3|705|Loss: 0.8791558146476746:  11%|█         | 704/6470 [08:40<1:14:08,  1.30it/s]
3|705|Loss: 0.8791558146476746:  11%|█         | 705/6470 [08:41<1:09:24,  1.38it/s]
3|706|Loss: 0.774127185344696:  11%|█         | 705/6470 [08:41<1:09:24,  1.38it/s] 
3|706|Loss: 0.774127185344696:  11%|█         | 706/6470 [08:41<1:09:24,  1.38it/s]
3|707|Loss: 0.7373512983322144:  11%|█         | 706/6470 [08:42<1:09:24,  1.38it/s]
3|707|Loss: 0.7373512983322144:  11%|█         | 707/6470 [08:42<1:12:52,  1.32it/s]
3|708|Loss: 0.7886600494384766:  11%|█         | 707/6470 [08:43<1:12:52,  1.32it/s]
3|708|Loss: 0.7886600494384766:  11%|█         | 708/6470 [08:43<1:07:52,  1.41it/s]
3|709|Loss: 0.8756403923034668:  11%|█         | 708/6470 [08:43<1:07:52,  1.41it/s]
3|709|Loss: 0.8756403923034668:  11%|█         | 709/6470 [08:44<1:07:58,  1.41it/s]
3|710|Loss: 0.7952157258987427:  11%|█         | 709/6470 [08:44<1:07:58,  1.41it/s]
3|710|Loss: 0.7952157258987427:  11%|█         | 710/6470 [08:44<1:10:28,  1.36it/s]
3|711|Loss: 1.0345311164855957:  11%|█         | 710/6470 [08:45<1:10:28,  1.36it/s]
3|711|Loss: 1.0345311164855957:  11%|█         | 711/6470 [08:45<1:11:11,  1.35it/s]
3|712|Loss: 0.9349760413169861:  11%|█         | 711/6470 [08:46<1:11:11,  1.35it/s]
3|712|Loss: 0.9349760413169861:  11%|█         | 712/6470 [08:46<1:11:33,  1.34it/s]
3|713|Loss: 0.906467080116272:  11%|█         | 712/6470 [08:46<1:11:33,  1.34it/s] 
3|713|Loss: 0.906467080116272:  11%|█         | 713/6470 [08:47<1:09:55,  1.37it/s]
3|714|Loss: 0.8561134338378906:  11%|█         | 713/6470 [08:47<1:09:55,  1.37it/s]
3|714|Loss: 0.8561134338378906:  11%|█         | 714/6470 [08:47<1:12:30,  1.32it/s]
3|715|Loss: 0.7873439192771912:  11%|█         | 714/6470 [08:48<1:12:30,  1.32it/s]
3|715|Loss: 0.7873439192771912:  11%|█         | 715/6470 [08:48<1:16:44,  1.25it/s]
3|716|Loss: 0.8085776567459106:  11%|█         | 715/6470 [08:49<1:16:44,  1.25it/s]
3|716|Loss: 0.8085776567459106:  11%|█         | 716/6470 [08:49<1:14:50,  1.28it/s]
3|717|Loss: 0.9736232757568359:  11%|█         | 716/6470 [08:49<1:14:50,  1.28it/s]
3|717|Loss: 0.9736232757568359:  11%|█         | 717/6470 [08:50<1:10:05,  1.37it/s]
3|718|Loss: 0.877851665019989:  11%|█         | 717/6470 [08:50<1:10:05,  1.37it/s] 
3|718|Loss: 0.877851665019989:  11%|█         | 718/6470 [08:50<1:10:21,  1.36it/s]
3|719|Loss: 0.7691054940223694:  11%|█         | 718/6470 [08:51<1:10:21,  1.36it/s]
3|719|Loss: 0.7691054940223694:  11%|█         | 719/6470 [08:51<1:06:40,  1.44it/s]
3|720|Loss: 0.934837818145752:  11%|█         | 719/6470 [08:51<1:06:40,  1.44it/s] 
3|720|Loss: 0.934837818145752:  11%|█         | 720/6470 [08:52<1:09:33,  1.38it/s]
3|721|Loss: 0.7232637405395508:  11%|█         | 720/6470 [08:52<1:09:33,  1.38it/s]
3|721|Loss: 0.7232637405395508:  11%|█         | 721/6470 [08:53<1:16:25,  1.25it/s]
3|722|Loss: 0.8882258534431458:  11%|█         | 721/6470 [08:53<1:16:25,  1.25it/s]
3|722|Loss: 0.8882258534431458:  11%|█         | 722/6470 [08:54<1:17:31,  1.24it/s]
3|723|Loss: 0.8476079702377319:  11%|█         | 722/6470 [08:54<1:17:31,  1.24it/s]
3|723|Loss: 0.8476079702377319:  11%|█         | 723/6470 [08:54<1:17:29,  1.24it/s]
3|724|Loss: 0.7017729878425598:  11%|█         | 723/6470 [08:55<1:17:29,  1.24it/s]
3|724|Loss: 0.7017729878425598:  11%|█         | 724/6470 [08:55<1:18:32,  1.22it/s]
3|725|Loss: 0.777427613735199:  11%|█         | 724/6470 [08:56<1:18:32,  1.22it/s] 
3|725|Loss: 0.777427613735199:  11%|█         | 725/6470 [08:56<1:17:09,  1.24it/s]
3|726|Loss: 0.8857256770133972:  11%|█         | 725/6470 [08:56<1:17:09,  1.24it/s]
3|726|Loss: 0.8857256770133972:  11%|█         | 726/6470 [08:57<1:16:39,  1.25it/s]
3|727|Loss: 0.8086311221122742:  11%|█         | 726/6470 [08:57<1:16:39,  1.25it/s]
3|727|Loss: 0.8086311221122742:  11%|█         | 727/6470 [08:58<1:19:36,  1.20it/s]
3|728|Loss: 0.7929719090461731:  11%|█         | 727/6470 [08:58<1:19:36,  1.20it/s]
3|728|Loss: 0.7929719090461731:  11%|█▏        | 728/6470 [08:59<1:19:05,  1.21it/s]
3|729|Loss: 0.8074422478675842:  11%|█▏        | 728/6470 [08:59<1:19:05,  1.21it/s]
3|729|Loss: 0.8074422478675842:  11%|█▏        | 729/6470 [08:59<1:10:40,  1.35it/s]
3|730|Loss: 0.7357468605041504:  11%|█▏        | 729/6470 [08:59<1:10:40,  1.35it/s]
3|730|Loss: 0.7357468605041504:  11%|█▏        | 730/6470 [09:00<1:09:58,  1.37it/s]
3|731|Loss: 0.7916797399520874:  11%|█▏        | 730/6470 [09:00<1:09:58,  1.37it/s]
3|731|Loss: 0.7916797399520874:  11%|█▏        | 731/6470 [09:00<1:07:27,  1.42it/s]
3|732|Loss: 0.8320395946502686:  11%|█▏        | 731/6470 [09:01<1:07:27,  1.42it/s]
3|732|Loss: 0.8320395946502686:  11%|█▏        | 732/6470 [09:01<1:02:48,  1.52it/s]
3|733|Loss: 0.8870147466659546:  11%|█▏        | 732/6470 [09:01<1:02:48,  1.52it/s]
3|733|Loss: 0.8870147466659546:  11%|█▏        | 733/6470 [09:02<1:05:47,  1.45it/s]
3|734|Loss: 0.8543685674667358:  11%|█▏        | 733/6470 [09:02<1:05:47,  1.45it/s]
3|734|Loss: 0.8543685674667358:  11%|█▏        | 734/6470 [09:03<1:09:21,  1.38it/s]
3|735|Loss: 0.9517018795013428:  11%|█▏        | 734/6470 [09:03<1:09:21,  1.38it/s]
3|735|Loss: 0.9517018795013428:  11%|█▏        | 735/6470 [09:03<1:13:44,  1.30it/s]
3|736|Loss: 0.7232338786125183:  11%|█▏        | 735/6470 [09:04<1:13:44,  1.30it/s]
3|736|Loss: 0.7232338786125183:  11%|█▏        | 736/6470 [09:04<1:11:45,  1.33it/s]
3|737|Loss: 0.9131636619567871:  11%|█▏        | 736/6470 [09:05<1:11:45,  1.33it/s]
3|737|Loss: 0.9131636619567871:  11%|█▏        | 737/6470 [09:05<1:11:16,  1.34it/s]
3|738|Loss: 0.9051269292831421:  11%|█▏        | 737/6470 [09:05<1:11:16,  1.34it/s]
3|738|Loss: 0.9051269292831421:  11%|█▏        | 738/6470 [09:06<1:12:10,  1.32it/s]
3|739|Loss: 0.7690137028694153:  11%|█▏        | 738/6470 [09:06<1:12:10,  1.32it/s]
3|739|Loss: 0.7690137028694153:  11%|█▏        | 739/6470 [09:07<1:19:06,  1.21it/s]
3|740|Loss: 0.8171038627624512:  11%|█▏        | 739/6470 [09:07<1:19:06,  1.21it/s]
3|740|Loss: 0.8171038627624512:  11%|█▏        | 740/6470 [09:07<1:19:46,  1.20it/s]
3|741|Loss: 0.6311847567558289:  11%|█▏        | 740/6470 [09:08<1:19:46,  1.20it/s]
3|741|Loss: 0.6311847567558289:  11%|█▏        | 741/6470 [09:08<1:15:39,  1.26it/s]
3|742|Loss: 0.887397825717926:  11%|█▏        | 741/6470 [09:09<1:15:39,  1.26it/s] 
3|742|Loss: 0.887397825717926:  11%|█▏        | 742/6470 [09:09<1:12:33,  1.32it/s]
3|743|Loss: 0.8543328642845154:  11%|█▏        | 742/6470 [09:09<1:12:33,  1.32it/s]
3|743|Loss: 0.8543328642845154:  11%|█▏        | 743/6470 [09:10<1:12:04,  1.32it/s]
3|744|Loss: 0.8733553290367126:  11%|█▏        | 743/6470 [09:10<1:12:04,  1.32it/s]
3|744|Loss: 0.8733553290367126:  11%|█▏        | 744/6470 [09:10<1:12:56,  1.31it/s]
3|745|Loss: 0.743216335773468:  11%|█▏        | 744/6470 [09:11<1:12:56,  1.31it/s] 
3|745|Loss: 0.743216335773468:  12%|█▏        | 745/6470 [09:11<1:16:58,  1.24it/s]
3|746|Loss: 0.7775492668151855:  12%|█▏        | 745/6470 [09:12<1:16:58,  1.24it/s]
3|746|Loss: 0.7775492668151855:  12%|█▏        | 746/6470 [09:12<1:14:39,  1.28it/s]
3|747|Loss: 0.7065208554267883:  12%|█▏        | 746/6470 [09:12<1:14:39,  1.28it/s]
3|747|Loss: 0.7065208554267883:  12%|█▏        | 747/6470 [09:13<1:14:12,  1.29it/s]
3|748|Loss: 0.804554283618927:  12%|█▏        | 747/6470 [09:13<1:14:12,  1.29it/s] 
3|748|Loss: 0.804554283618927:  12%|█▏        | 748/6470 [09:13<1:11:46,  1.33it/s]
3|749|Loss: 0.823920488357544:  12%|█▏        | 748/6470 [09:14<1:11:46,  1.33it/s]
3|749|Loss: 0.823920488357544:  12%|█▏        | 749/6470 [09:14<1:06:27,  1.43it/s]
3|750|Loss: 0.6477453112602234:  12%|█▏        | 749/6470 [09:14<1:06:27,  1.43it/s]
3|750|Loss: 0.6477453112602234:  12%|█▏        | 750/6470 [09:15<1:03:40,  1.50it/s]
3|751|Loss: 0.9213494658470154:  12%|█▏        | 750/6470 [09:15<1:03:40,  1.50it/s]
3|751|Loss: 0.9213494658470154:  12%|█▏        | 751/6470 [09:15<1:03:07,  1.51it/s]
3|752|Loss: 0.9229379296302795:  12%|█▏        | 751/6470 [09:16<1:03:07,  1.51it/s]
3|752|Loss: 0.9229379296302795:  12%|█▏        | 752/6470 [09:16<1:09:52,  1.36it/s]
3|753|Loss: 0.9482645988464355:  12%|█▏        | 752/6470 [09:17<1:09:52,  1.36it/s]
3|753|Loss: 0.9482645988464355:  12%|█▏        | 753/6470 [09:17<1:09:09,  1.38it/s]
3|754|Loss: 0.7450806498527527:  12%|█▏        | 753/6470 [09:17<1:09:09,  1.38it/s]
3|754|Loss: 0.7450806498527527:  12%|█▏        | 754/6470 [09:18<1:07:54,  1.40it/s]
3|755|Loss: 0.8146923184394836:  12%|█▏        | 754/6470 [09:18<1:07:54,  1.40it/s]
3|755|Loss: 0.8146923184394836:  12%|█▏        | 755/6470 [09:18<1:05:31,  1.45it/s]
3|756|Loss: 0.8254727721214294:  12%|█▏        | 755/6470 [09:19<1:05:31,  1.45it/s]
3|756|Loss: 0.8254727721214294:  12%|█▏        | 756/6470 [09:19<1:04:33,  1.48it/s]
3|757|Loss: 0.8786744475364685:  12%|█▏        | 756/6470 [09:19<1:04:33,  1.48it/s]
3|757|Loss: 0.8786744475364685:  12%|█▏        | 757/6470 [09:20<1:08:45,  1.38it/s]
3|758|Loss: 0.7996851801872253:  12%|█▏        | 757/6470 [09:20<1:08:45,  1.38it/s]
3|758|Loss: 0.7996851801872253:  12%|█▏        | 758/6470 [09:20<1:09:08,  1.38it/s]
3|759|Loss: 0.7772798538208008:  12%|█▏        | 758/6470 [09:21<1:09:08,  1.38it/s]
3|759|Loss: 0.7772798538208008:  12%|█▏        | 759/6470 [09:21<1:02:07,  1.53it/s]
3|760|Loss: 0.825231671333313:  12%|█▏        | 759/6470 [09:21<1:02:07,  1.53it/s] 
3|760|Loss: 0.825231671333313:  12%|█▏        | 760/6470 [09:21<1:00:02,  1.59it/s]
3|761|Loss: 0.677850604057312:  12%|█▏        | 760/6470 [09:22<1:00:02,  1.59it/s]
3|761|Loss: 0.677850604057312:  12%|█▏        | 761/6470 [09:22<1:02:06,  1.53it/s]
3|762|Loss: 0.738516092300415:  12%|█▏        | 761/6470 [09:23<1:02:06,  1.53it/s]
3|762|Loss: 0.738516092300415:  12%|█▏        | 762/6470 [09:23<1:08:06,  1.40it/s]
3|763|Loss: 0.8008406758308411:  12%|█▏        | 762/6470 [09:23<1:08:06,  1.40it/s]
3|763|Loss: 0.8008406758308411:  12%|█▏        | 763/6470 [09:24<1:07:47,  1.40it/s]
3|764|Loss: 0.89216548204422:  12%|█▏        | 763/6470 [09:24<1:07:47,  1.40it/s]  
3|764|Loss: 0.89216548204422:  12%|█▏        | 764/6470 [09:24<1:07:17,  1.41it/s]
3|765|Loss: 0.8624740242958069:  12%|█▏        | 764/6470 [09:25<1:07:17,  1.41it/s]
3|765|Loss: 0.8624740242958069:  12%|█▏        | 765/6470 [09:25<1:10:50,  1.34it/s]
3|766|Loss: 0.8143682479858398:  12%|█▏        | 765/6470 [09:26<1:10:50,  1.34it/s]
3|766|Loss: 0.8143682479858398:  12%|█▏        | 766/6470 [09:26<1:12:12,  1.32it/s]
3|767|Loss: 0.8296344876289368:  12%|█▏        | 766/6470 [09:27<1:12:12,  1.32it/s]
3|767|Loss: 0.8296344876289368:  12%|█▏        | 767/6470 [09:27<1:12:27,  1.31it/s]
3|768|Loss: 0.7755967378616333:  12%|█▏        | 767/6470 [09:27<1:12:27,  1.31it/s]
3|768|Loss: 0.7755967378616333:  12%|█▏        | 768/6470 [09:28<1:12:15,  1.32it/s]
3|769|Loss: 0.7071751952171326:  12%|█▏        | 768/6470 [09:28<1:12:15,  1.32it/s]
3|769|Loss: 0.7071751952171326:  12%|█▏        | 769/6470 [09:29<1:16:16,  1.25it/s]
3|770|Loss: 0.8881866931915283:  12%|█▏        | 769/6470 [09:29<1:16:16,  1.25it/s]
3|770|Loss: 0.8881866931915283:  12%|█▏        | 770/6470 [09:29<1:12:22,  1.31it/s]
3|771|Loss: 0.8059422373771667:  12%|█▏        | 770/6470 [09:30<1:12:22,  1.31it/s]
3|771|Loss: 0.8059422373771667:  12%|█▏        | 771/6470 [09:30<1:11:02,  1.34it/s]
3|772|Loss: 0.808994710445404:  12%|█▏        | 771/6470 [09:30<1:11:02,  1.34it/s] 
3|772|Loss: 0.808994710445404:  12%|█▏        | 772/6470 [09:31<1:13:21,  1.29it/s]
3|773|Loss: 0.8634383082389832:  12%|█▏        | 772/6470 [09:31<1:13:21,  1.29it/s]
3|773|Loss: 0.8634383082389832:  12%|█▏        | 773/6470 [09:31<1:09:46,  1.36it/s]
3|774|Loss: 0.8037370443344116:  12%|█▏        | 773/6470 [09:32<1:09:46,  1.36it/s]
3|774|Loss: 0.8037370443344116:  12%|█▏        | 774/6470 [09:32<1:09:18,  1.37it/s]
3|775|Loss: 0.7627331614494324:  12%|█▏        | 774/6470 [09:33<1:09:18,  1.37it/s]
3|775|Loss: 0.7627331614494324:  12%|█▏        | 775/6470 [09:33<1:09:52,  1.36it/s]
3|776|Loss: 0.853101909160614:  12%|█▏        | 775/6470 [09:33<1:09:52,  1.36it/s] 
3|776|Loss: 0.853101909160614:  12%|█▏        | 776/6470 [09:34<1:13:40,  1.29it/s]
3|777|Loss: 0.8075277209281921:  12%|█▏        | 776/6470 [09:34<1:13:40,  1.29it/s]
3|777|Loss: 0.8075277209281921:  12%|█▏        | 777/6470 [09:34<1:10:00,  1.36it/s]
3|778|Loss: 0.8521918654441833:  12%|█▏        | 777/6470 [09:35<1:10:00,  1.36it/s]
3|778|Loss: 0.8521918654441833:  12%|█▏        | 778/6470 [09:35<1:10:33,  1.34it/s]
3|779|Loss: 0.7256317734718323:  12%|█▏        | 778/6470 [09:36<1:10:33,  1.34it/s]
3|779|Loss: 0.7256317734718323:  12%|█▏        | 779/6470 [09:36<1:14:13,  1.28it/s]
3|780|Loss: 0.8309980630874634:  12%|█▏        | 779/6470 [09:36<1:14:13,  1.28it/s]
3|780|Loss: 0.8309980630874634:  12%|█▏        | 780/6470 [09:37<1:10:26,  1.35it/s]
3|781|Loss: 0.7374178171157837:  12%|█▏        | 780/6470 [09:37<1:10:26,  1.35it/s]
3|781|Loss: 0.7374178171157837:  12%|█▏        | 781/6470 [09:38<1:13:55,  1.28it/s]
3|782|Loss: 0.9120393991470337:  12%|█▏        | 781/6470 [09:38<1:13:55,  1.28it/s]
3|782|Loss: 0.9120393991470337:  12%|█▏        | 782/6470 [09:38<1:11:14,  1.33it/s]
3|783|Loss: 0.7107692360877991:  12%|█▏        | 782/6470 [09:39<1:11:14,  1.33it/s]
3|783|Loss: 0.7107692360877991:  12%|█▏        | 783/6470 [09:39<1:06:46,  1.42it/s]
3|784|Loss: 0.857688844203949:  12%|█▏        | 783/6470 [09:39<1:06:46,  1.42it/s] 
3|784|Loss: 0.857688844203949:  12%|█▏        | 784/6470 [09:39<1:04:25,  1.47it/s]
3|785|Loss: 0.8672266602516174:  12%|█▏        | 784/6470 [09:40<1:04:25,  1.47it/s]
3|785|Loss: 0.8672266602516174:  12%|█▏        | 785/6470 [09:40<1:03:51,  1.48it/s]
3|786|Loss: 0.8995352983474731:  12%|█▏        | 785/6470 [09:40<1:03:51,  1.48it/s]
3|786|Loss: 0.8995352983474731:  12%|█▏        | 786/6470 [09:41<1:03:31,  1.49it/s]
3|787|Loss: 0.8062014579772949:  12%|█▏        | 786/6470 [09:41<1:03:31,  1.49it/s]
3|787|Loss: 0.8062014579772949:  12%|█▏        | 787/6470 [09:41<1:05:56,  1.44it/s]
3|788|Loss: 0.8982668519020081:  12%|█▏        | 787/6470 [09:42<1:05:56,  1.44it/s]
3|788|Loss: 0.8982668519020081:  12%|█▏        | 788/6470 [09:42<1:06:10,  1.43it/s]
3|789|Loss: 0.8733677268028259:  12%|█▏        | 788/6470 [09:43<1:06:10,  1.43it/s]
3|789|Loss: 0.8733677268028259:  12%|█▏        | 789/6470 [09:43<1:03:40,  1.49it/s]
3|790|Loss: 0.8075879216194153:  12%|█▏        | 789/6470 [09:43<1:03:40,  1.49it/s]
3|790|Loss: 0.8075879216194153:  12%|█▏        | 790/6470 [09:44<1:05:22,  1.45it/s]
3|791|Loss: 0.9249999523162842:  12%|█▏        | 790/6470 [09:44<1:05:22,  1.45it/s]
3|791|Loss: 0.9249999523162842:  12%|█▏        | 791/6470 [09:44<1:02:47,  1.51it/s]
3|792|Loss: 0.7566359639167786:  12%|█▏        | 791/6470 [09:45<1:02:47,  1.51it/s]
3|792|Loss: 0.7566359639167786:  12%|█▏        | 792/6470 [09:45<1:05:23,  1.45it/s]
3|793|Loss: 0.8709359169006348:  12%|█▏        | 792/6470 [09:45<1:05:23,  1.45it/s]
3|793|Loss: 0.8709359169006348:  12%|█▏        | 793/6470 [09:45<1:02:43,  1.51it/s]
3|794|Loss: 0.8508551120758057:  12%|█▏        | 793/6470 [09:46<1:02:43,  1.51it/s]
3|794|Loss: 0.8508551120758057:  12%|█▏        | 794/6470 [09:46<1:02:29,  1.51it/s]
3|795|Loss: 0.7419256567955017:  12%|█▏        | 794/6470 [09:47<1:02:29,  1.51it/s]
3|795|Loss: 0.7419256567955017:  12%|█▏        | 795/6470 [09:47<1:03:18,  1.49it/s]
3|796|Loss: 0.8657047152519226:  12%|█▏        | 795/6470 [09:47<1:03:18,  1.49it/s]
3|796|Loss: 0.8657047152519226:  12%|█▏        | 796/6470 [09:48<1:06:23,  1.42it/s]
3|797|Loss: 0.7515481114387512:  12%|█▏        | 796/6470 [09:48<1:06:23,  1.42it/s]
3|797|Loss: 0.7515481114387512:  12%|█▏        | 797/6470 [09:48<1:06:21,  1.42it/s]
3|798|Loss: 0.7739730477333069:  12%|█▏        | 797/6470 [09:49<1:06:21,  1.42it/s]
3|798|Loss: 0.7739730477333069:  12%|█▏        | 798/6470 [09:49<1:03:38,  1.49it/s]
3|799|Loss: 0.7055421471595764:  12%|█▏        | 798/6470 [09:49<1:03:38,  1.49it/s]
3|799|Loss: 0.7055421471595764:  12%|█▏        | 799/6470 [09:50<1:01:56,  1.53it/s]
3|800|Loss: 0.8075874447822571:  12%|█▏        | 799/6470 [09:50<1:01:56,  1.53it/s]
3|800|Loss: 0.8075874447822571:  12%|█▏        | 800/6470 [09:50<1:02:37,  1.51it/s]
3|801|Loss: 0.8490256071090698:  12%|█▏        | 800/6470 [09:51<1:02:37,  1.51it/s]
3|801|Loss: 0.8490256071090698:  12%|█▏        | 801/6470 [09:51<1:04:56,  1.45it/s]
3|802|Loss: 0.8345494270324707:  12%|█▏        | 801/6470 [09:51<1:04:56,  1.45it/s]
3|802|Loss: 0.8345494270324707:  12%|█▏        | 802/6470 [09:52<1:05:47,  1.44it/s]
3|803|Loss: 0.8694981932640076:  12%|█▏        | 802/6470 [09:52<1:05:47,  1.44it/s]
3|803|Loss: 0.8694981932640076:  12%|█▏        | 803/6470 [09:52<1:05:02,  1.45it/s]
3|804|Loss: 0.7496647834777832:  12%|█▏        | 803/6470 [09:53<1:05:02,  1.45it/s]
3|804|Loss: 0.7496647834777832:  12%|█▏        | 804/6470 [09:53<1:12:45,  1.30it/s]
3|805|Loss: 0.8655261397361755:  12%|█▏        | 804/6470 [09:54<1:12:45,  1.30it/s]
3|805|Loss: 0.8655261397361755:  12%|█▏        | 805/6470 [09:54<1:11:49,  1.31it/s]
3|806|Loss: 0.8934261202812195:  12%|█▏        | 805/6470 [09:54<1:11:49,  1.31it/s]
3|806|Loss: 0.8934261202812195:  12%|█▏        | 806/6470 [09:55<1:04:42,  1.46it/s]
3|807|Loss: 0.8196040391921997:  12%|█▏        | 806/6470 [09:55<1:04:42,  1.46it/s]
3|807|Loss: 0.8196040391921997:  12%|█▏        | 807/6470 [09:55<1:07:12,  1.40it/s]
3|808|Loss: 0.8131939172744751:  12%|█▏        | 807/6470 [09:56<1:07:12,  1.40it/s]
3|808|Loss: 0.8131939172744751:  12%|█▏        | 808/6470 [09:56<1:06:38,  1.42it/s]
3|809|Loss: 0.7228806018829346:  12%|█▏        | 808/6470 [09:56<1:06:38,  1.42it/s]
3|809|Loss: 0.7228806018829346:  13%|█▎        | 809/6470 [09:56<58:36,  1.61it/s]  
3|810|Loss: 0.7670894861221313:  13%|█▎        | 809/6470 [09:57<58:36,  1.61it/s]
3|810|Loss: 0.7670894861221313:  13%|█▎        | 810/6470 [09:57<1:04:03,  1.47it/s]
3|811|Loss: 0.9402440786361694:  13%|█▎        | 810/6470 [09:58<1:04:03,  1.47it/s]
3|811|Loss: 0.9402440786361694:  13%|█▎        | 811/6470 [09:58<1:08:32,  1.38it/s]
3|812|Loss: 0.7735586166381836:  13%|█▎        | 811/6470 [09:59<1:08:32,  1.38it/s]
3|812|Loss: 0.7735586166381836:  13%|█▎        | 812/6470 [09:59<1:10:03,  1.35it/s]
3|813|Loss: 0.7605005502700806:  13%|█▎        | 812/6470 [09:59<1:10:03,  1.35it/s]
3|813|Loss: 0.7605005502700806:  13%|█▎        | 813/6470 [10:00<1:10:55,  1.33it/s]
3|814|Loss: 0.7224969267845154:  13%|█▎        | 813/6470 [10:00<1:10:55,  1.33it/s]
3|814|Loss: 0.7224969267845154:  13%|█▎        | 814/6470 [10:00<1:04:36,  1.46it/s]
3|815|Loss: 0.7178633213043213:  13%|█▎        | 814/6470 [10:01<1:04:36,  1.46it/s]
3|815|Loss: 0.7178633213043213:  13%|█▎        | 815/6470 [10:01<1:05:38,  1.44it/s]
3|816|Loss: 0.8289086222648621:  13%|█▎        | 815/6470 [10:01<1:05:38,  1.44it/s]
3|816|Loss: 0.8289086222648621:  13%|█▎        | 816/6470 [10:02<1:12:00,  1.31it/s]
3|817|Loss: 0.926861047744751:  13%|█▎        | 816/6470 [10:02<1:12:00,  1.31it/s] 
3|817|Loss: 0.926861047744751:  13%|█▎        | 817/6470 [10:02<1:07:17,  1.40it/s]
3|818|Loss: 0.9695208668708801:  13%|█▎        | 817/6470 [10:03<1:07:17,  1.40it/s]
3|818|Loss: 0.9695208668708801:  13%|█▎        | 818/6470 [10:03<1:07:40,  1.39it/s]
3|819|Loss: 0.9629671573638916:  13%|█▎        | 818/6470 [10:04<1:07:40,  1.39it/s]
3|819|Loss: 0.9629671573638916:  13%|█▎        | 819/6470 [10:04<1:03:16,  1.49it/s]
3|820|Loss: 0.8175579905509949:  13%|█▎        | 819/6470 [10:04<1:03:16,  1.49it/s]
3|820|Loss: 0.8175579905509949:  13%|█▎        | 820/6470 [10:04<1:04:32,  1.46it/s]
3|821|Loss: 0.7053388953208923:  13%|█▎        | 820/6470 [10:05<1:04:32,  1.46it/s]
3|821|Loss: 0.7053388953208923:  13%|█▎        | 821/6470 [10:05<1:05:03,  1.45it/s]
3|822|Loss: 0.8660807013511658:  13%|█▎        | 821/6470 [10:06<1:05:03,  1.45it/s]
3|822|Loss: 0.8660807013511658:  13%|█▎        | 822/6470 [10:06<1:04:50,  1.45it/s]
3|823|Loss: 0.7885540127754211:  13%|█▎        | 822/6470 [10:06<1:04:50,  1.45it/s]
3|823|Loss: 0.7885540127754211:  13%|█▎        | 823/6470 [10:07<1:10:30,  1.33it/s]
3|824|Loss: 0.9192520976066589:  13%|█▎        | 823/6470 [10:07<1:10:30,  1.33it/s]
3|824|Loss: 0.9192520976066589:  13%|█▎        | 824/6470 [10:07<1:09:56,  1.35it/s]
3|825|Loss: 0.8265193104743958:  13%|█▎        | 824/6470 [10:08<1:09:56,  1.35it/s]
3|825|Loss: 0.8265193104743958:  13%|█▎        | 825/6470 [10:08<1:09:39,  1.35it/s]
3|826|Loss: 0.7714508175849915:  13%|█▎        | 825/6470 [10:09<1:09:39,  1.35it/s]
3|826|Loss: 0.7714508175849915:  13%|█▎        | 826/6470 [10:09<1:08:33,  1.37it/s]
3|827|Loss: 0.7022061347961426:  13%|█▎        | 826/6470 [10:09<1:08:33,  1.37it/s]
3|827|Loss: 0.7022061347961426:  13%|█▎        | 827/6470 [10:10<1:09:37,  1.35it/s]
3|828|Loss: 0.831497073173523:  13%|█▎        | 827/6470 [10:10<1:09:37,  1.35it/s] 
3|828|Loss: 0.831497073173523:  13%|█▎        | 828/6470 [10:10<1:07:06,  1.40it/s]
3|829|Loss: 0.7455451488494873:  13%|█▎        | 828/6470 [10:11<1:07:06,  1.40it/s]
3|829|Loss: 0.7455451488494873:  13%|█▎        | 829/6470 [10:11<1:07:49,  1.39it/s]
3|830|Loss: 0.696063756942749:  13%|█▎        | 829/6470 [10:11<1:07:49,  1.39it/s] 
3|830|Loss: 0.696063756942749:  13%|█▎        | 830/6470 [10:12<1:10:55,  1.33it/s]
3|831|Loss: 0.7308896780014038:  13%|█▎        | 830/6470 [10:12<1:10:55,  1.33it/s]
3|831|Loss: 0.7308896780014038:  13%|█▎        | 831/6470 [10:13<1:14:06,  1.27it/s]
3|832|Loss: 0.9051390290260315:  13%|█▎        | 831/6470 [10:13<1:14:06,  1.27it/s]
3|832|Loss: 0.9051390290260315:  13%|█▎        | 832/6470 [10:13<1:10:07,  1.34it/s]
3|833|Loss: 0.8434821367263794:  13%|█▎        | 832/6470 [10:14<1:10:07,  1.34it/s]
3|833|Loss: 0.8434821367263794:  13%|█▎        | 833/6470 [10:14<1:12:48,  1.29it/s]
3|834|Loss: 0.9186241030693054:  13%|█▎        | 833/6470 [10:15<1:12:48,  1.29it/s]
3|834|Loss: 0.9186241030693054:  13%|█▎        | 834/6470 [10:15<1:10:00,  1.34it/s]
3|835|Loss: 0.9527400732040405:  13%|█▎        | 834/6470 [10:15<1:10:00,  1.34it/s]
3|835|Loss: 0.9527400732040405:  13%|█▎        | 835/6470 [10:16<1:10:47,  1.33it/s]
3|836|Loss: 0.6707863807678223:  13%|█▎        | 835/6470 [10:16<1:10:47,  1.33it/s]
3|836|Loss: 0.6707863807678223:  13%|█▎        | 836/6470 [10:16<1:08:07,  1.38it/s]
3|837|Loss: 0.9271274209022522:  13%|█▎        | 836/6470 [10:17<1:08:07,  1.38it/s]
3|837|Loss: 0.9271274209022522:  13%|█▎        | 837/6470 [10:17<1:09:29,  1.35it/s]
3|838|Loss: 0.7762680053710938:  13%|█▎        | 837/6470 [10:18<1:09:29,  1.35it/s]
3|838|Loss: 0.7762680053710938:  13%|█▎        | 838/6470 [10:18<1:11:24,  1.31it/s]
3|839|Loss: 0.8191434741020203:  13%|█▎        | 838/6470 [10:18<1:11:24,  1.31it/s]
3|839|Loss: 0.8191434741020203:  13%|█▎        | 839/6470 [10:19<1:12:37,  1.29it/s]
3|840|Loss: 0.9956299662590027:  13%|█▎        | 839/6470 [10:19<1:12:37,  1.29it/s]
3|840|Loss: 0.9956299662590027:  13%|█▎        | 840/6470 [10:19<1:07:40,  1.39it/s]
3|841|Loss: 0.9006728529930115:  13%|█▎        | 840/6470 [10:20<1:07:40,  1.39it/s]
3|841|Loss: 0.9006728529930115:  13%|█▎        | 841/6470 [10:20<1:08:48,  1.36it/s]
3|842|Loss: 0.8333017230033875:  13%|█▎        | 841/6470 [10:21<1:08:48,  1.36it/s]
3|842|Loss: 0.8333017230033875:  13%|█▎        | 842/6470 [10:21<1:13:55,  1.27it/s]
3|843|Loss: 0.7585598826408386:  13%|█▎        | 842/6470 [10:21<1:13:55,  1.27it/s]
3|843|Loss: 0.7585598826408386:  13%|█▎        | 843/6470 [10:22<1:12:38,  1.29it/s]
3|844|Loss: 0.9443102478981018:  13%|█▎        | 843/6470 [10:22<1:12:38,  1.29it/s]
3|844|Loss: 0.9443102478981018:  13%|█▎        | 844/6470 [10:22<1:11:29,  1.31it/s]
3|845|Loss: 0.8323983550071716:  13%|█▎        | 844/6470 [10:23<1:11:29,  1.31it/s]
3|845|Loss: 0.8323983550071716:  13%|█▎        | 845/6470 [10:23<1:07:32,  1.39it/s]
3|846|Loss: 0.9093376398086548:  13%|█▎        | 845/6470 [10:23<1:07:32,  1.39it/s]
3|846|Loss: 0.9093376398086548:  13%|█▎        | 846/6470 [10:24<1:03:06,  1.49it/s]
3|847|Loss: 0.9026351571083069:  13%|█▎        | 846/6470 [10:24<1:03:06,  1.49it/s]
3|847|Loss: 0.9026351571083069:  13%|█▎        | 847/6470 [10:24<1:02:10,  1.51it/s]
3|848|Loss: 0.8511976599693298:  13%|█▎        | 847/6470 [10:25<1:02:10,  1.51it/s]
3|848|Loss: 0.8511976599693298:  13%|█▎        | 848/6470 [10:25<1:05:27,  1.43it/s]
3|849|Loss: 0.7924515008926392:  13%|█▎        | 848/6470 [10:26<1:05:27,  1.43it/s]
3|849|Loss: 0.7924515008926392:  13%|█▎        | 849/6470 [10:26<1:08:19,  1.37it/s]
3|850|Loss: 0.7678316235542297:  13%|█▎        | 849/6470 [10:26<1:08:19,  1.37it/s]
3|850|Loss: 0.7678316235542297:  13%|█▎        | 850/6470 [10:27<1:11:57,  1.30it/s]
3|851|Loss: 0.7230249047279358:  13%|█▎        | 850/6470 [10:27<1:11:57,  1.30it/s]
3|851|Loss: 0.7230249047279358:  13%|█▎        | 851/6470 [10:28<1:13:19,  1.28it/s]
3|852|Loss: 0.8845997452735901:  13%|█▎        | 851/6470 [10:28<1:13:19,  1.28it/s]
3|852|Loss: 0.8845997452735901:  13%|█▎        | 852/6470 [10:28<1:12:54,  1.28it/s]
3|853|Loss: 0.7305852770805359:  13%|█▎        | 852/6470 [10:29<1:12:54,  1.28it/s]
3|853|Loss: 0.7305852770805359:  13%|█▎        | 853/6470 [10:29<1:09:04,  1.36it/s]
3|854|Loss: 0.8313385248184204:  13%|█▎        | 853/6470 [10:29<1:09:04,  1.36it/s]
3|854|Loss: 0.8313385248184204:  13%|█▎        | 854/6470 [10:30<1:09:17,  1.35it/s]
3|855|Loss: 0.8843611478805542:  13%|█▎        | 854/6470 [10:30<1:09:17,  1.35it/s]
3|855|Loss: 0.8843611478805542:  13%|█▎        | 855/6470 [10:30<1:10:23,  1.33it/s]
3|856|Loss: 0.9355262517929077:  13%|█▎        | 855/6470 [10:31<1:10:23,  1.33it/s]
3|856|Loss: 0.9355262517929077:  13%|█▎        | 856/6470 [10:31<1:11:38,  1.31it/s]
3|857|Loss: 0.9135400652885437:  13%|█▎        | 856/6470 [10:32<1:11:38,  1.31it/s]
3|857|Loss: 0.9135400652885437:  13%|█▎        | 857/6470 [10:32<1:09:46,  1.34it/s]
3|858|Loss: 0.7739282846450806:  13%|█▎        | 857/6470 [10:32<1:09:46,  1.34it/s]
3|858|Loss: 0.7739282846450806:  13%|█▎        | 858/6470 [10:33<1:06:47,  1.40it/s]
3|859|Loss: 0.8134071230888367:  13%|█▎        | 858/6470 [10:33<1:06:47,  1.40it/s]
3|859|Loss: 0.8134071230888367:  13%|█▎        | 859/6470 [10:34<1:11:40,  1.30it/s]
3|860|Loss: 0.7700434327125549:  13%|█▎        | 859/6470 [10:34<1:11:40,  1.30it/s]
3|860|Loss: 0.7700434327125549:  13%|█▎        | 860/6470 [10:34<1:09:41,  1.34it/s]
3|861|Loss: 0.8229838013648987:  13%|█▎        | 860/6470 [10:35<1:09:41,  1.34it/s]
3|861|Loss: 0.8229838013648987:  13%|█▎        | 861/6470 [10:35<1:11:35,  1.31it/s]
3|862|Loss: 0.9376503229141235:  13%|█▎        | 861/6470 [10:35<1:11:35,  1.31it/s]
3|862|Loss: 0.9376503229141235:  13%|█▎        | 862/6470 [10:36<1:10:04,  1.33it/s]
3|863|Loss: 0.6119343638420105:  13%|█▎        | 862/6470 [10:36<1:10:04,  1.33it/s]
3|863|Loss: 0.6119343638420105:  13%|█▎        | 863/6470 [10:36<1:07:35,  1.38it/s]
3|864|Loss: 0.8471022248268127:  13%|█▎        | 863/6470 [10:37<1:07:35,  1.38it/s]
3|864|Loss: 0.8471022248268127:  13%|█▎        | 864/6470 [10:37<1:08:14,  1.37it/s]
3|865|Loss: 0.7992302775382996:  13%|█▎        | 864/6470 [10:38<1:08:14,  1.37it/s]
3|865|Loss: 0.7992302775382996:  13%|█▎        | 865/6470 [10:38<1:06:57,  1.40it/s]
3|866|Loss: 0.872901201248169:  13%|█▎        | 865/6470 [10:38<1:06:57,  1.40it/s] 
3|866|Loss: 0.872901201248169:  13%|█▎        | 866/6470 [10:39<1:12:01,  1.30it/s]
3|867|Loss: 0.7738937735557556:  13%|█▎        | 866/6470 [10:39<1:12:01,  1.30it/s]
3|867|Loss: 0.7738937735557556:  13%|█▎        | 867/6470 [10:40<1:13:20,  1.27it/s]
3|868|Loss: 0.8069807887077332:  13%|█▎        | 867/6470 [10:40<1:13:20,  1.27it/s]
3|868|Loss: 0.8069807887077332:  13%|█▎        | 868/6470 [10:40<1:13:06,  1.28it/s]
3|869|Loss: 0.7961646914482117:  13%|█▎        | 868/6470 [10:41<1:13:06,  1.28it/s]
3|869|Loss: 0.7961646914482117:  13%|█▎        | 869/6470 [10:41<1:10:25,  1.33it/s]
3|870|Loss: 0.7945449948310852:  13%|█▎        | 869/6470 [10:41<1:10:25,  1.33it/s]
3|870|Loss: 0.7945449948310852:  13%|█▎        | 870/6470 [10:42<1:06:32,  1.40it/s]
3|871|Loss: 0.6562173962593079:  13%|█▎        | 870/6470 [10:42<1:06:32,  1.40it/s]
3|871|Loss: 0.6562173962593079:  13%|█▎        | 871/6470 [10:42<1:02:01,  1.50it/s]
3|872|Loss: 0.8363156914710999:  13%|█▎        | 871/6470 [10:43<1:02:01,  1.50it/s]
3|872|Loss: 0.8363156914710999:  13%|█▎        | 872/6470 [10:43<1:03:46,  1.46it/s]
3|873|Loss: 1.0271174907684326:  13%|█▎        | 872/6470 [10:43<1:03:46,  1.46it/s]
3|873|Loss: 1.0271174907684326:  13%|█▎        | 873/6470 [10:44<1:09:11,  1.35it/s]
3|874|Loss: 0.7833602428436279:  13%|█▎        | 873/6470 [10:44<1:09:11,  1.35it/s]
3|874|Loss: 0.7833602428436279:  14%|█▎        | 874/6470 [10:45<1:09:16,  1.35it/s]
3|875|Loss: 0.7790406346321106:  14%|█▎        | 874/6470 [10:45<1:09:16,  1.35it/s]
3|875|Loss: 0.7790406346321106:  14%|█▎        | 875/6470 [10:45<1:08:00,  1.37it/s]
3|876|Loss: 0.7798606157302856:  14%|█▎        | 875/6470 [10:46<1:08:00,  1.37it/s]
3|876|Loss: 0.7798606157302856:  14%|█▎        | 876/6470 [10:46<1:04:00,  1.46it/s]
3|877|Loss: 0.8411744236946106:  14%|█▎        | 876/6470 [10:46<1:04:00,  1.46it/s]
3|877|Loss: 0.8411744236946106:  14%|█▎        | 877/6470 [10:46<1:01:08,  1.52it/s]
3|878|Loss: 0.7916245460510254:  14%|█▎        | 877/6470 [10:47<1:01:08,  1.52it/s]
3|878|Loss: 0.7916245460510254:  14%|█▎        | 878/6470 [10:47<1:07:16,  1.39it/s]
3|879|Loss: 0.8524777293205261:  14%|█▎        | 878/6470 [10:48<1:07:16,  1.39it/s]
3|879|Loss: 0.8524777293205261:  14%|█▎        | 879/6470 [10:48<1:06:38,  1.40it/s]
3|880|Loss: 0.8672101497650146:  14%|█▎        | 879/6470 [10:48<1:06:38,  1.40it/s]
3|880|Loss: 0.8672101497650146:  14%|█▎        | 880/6470 [10:49<1:09:07,  1.35it/s]
3|881|Loss: 0.7349740862846375:  14%|█▎        | 880/6470 [10:49<1:09:07,  1.35it/s]
3|881|Loss: 0.7349740862846375:  14%|█▎        | 881/6470 [10:50<1:10:16,  1.33it/s]
3|882|Loss: 0.9077787399291992:  14%|█▎        | 881/6470 [10:50<1:10:16,  1.33it/s]
3|882|Loss: 0.9077787399291992:  14%|█▎        | 882/6470 [10:50<1:04:06,  1.45it/s]
3|883|Loss: 0.9108861088752747:  14%|█▎        | 882/6470 [10:50<1:04:06,  1.45it/s]
3|883|Loss: 0.9108861088752747:  14%|█▎        | 883/6470 [10:51<1:06:54,  1.39it/s]
3|884|Loss: 0.9228679537773132:  14%|█▎        | 883/6470 [10:51<1:06:54,  1.39it/s]
3|884|Loss: 0.9228679537773132:  14%|█▎        | 884/6470 [10:52<1:05:49,  1.41it/s]
3|885|Loss: 0.8285702466964722:  14%|█▎        | 884/6470 [10:52<1:05:49,  1.41it/s]
3|885|Loss: 0.8285702466964722:  14%|█▎        | 885/6470 [10:52<1:07:09,  1.39it/s]
3|886|Loss: 0.7448585629463196:  14%|█▎        | 885/6470 [10:53<1:07:09,  1.39it/s]
3|886|Loss: 0.7448585629463196:  14%|█▎        | 886/6470 [10:53<1:06:12,  1.41it/s]
3|887|Loss: 0.8353205323219299:  14%|█▎        | 886/6470 [10:53<1:06:12,  1.41it/s]
3|887|Loss: 0.8353205323219299:  14%|█▎        | 887/6470 [10:54<1:06:48,  1.39it/s]
3|888|Loss: 0.8358832001686096:  14%|█▎        | 887/6470 [10:54<1:06:48,  1.39it/s]
3|888|Loss: 0.8358832001686096:  14%|█▎        | 888/6470 [10:55<1:08:47,  1.35it/s]
3|889|Loss: 0.7743116021156311:  14%|█▎        | 888/6470 [10:55<1:08:47,  1.35it/s]
3|889|Loss: 0.7743116021156311:  14%|█▎        | 889/6470 [10:55<1:04:29,  1.44it/s]
3|890|Loss: 0.7667522430419922:  14%|█▎        | 889/6470 [10:55<1:04:29,  1.44it/s]
3|890|Loss: 0.7667522430419922:  14%|█▍        | 890/6470 [10:56<1:03:36,  1.46it/s]
3|891|Loss: 1.1346793174743652:  14%|█▍        | 890/6470 [10:56<1:03:36,  1.46it/s]
3|891|Loss: 1.1346793174743652:  14%|█▍        | 891/6470 [10:57<1:06:43,  1.39it/s]
3|892|Loss: 1.0569771528244019:  14%|█▍        | 891/6470 [10:57<1:06:43,  1.39it/s]
3|892|Loss: 1.0569771528244019:  14%|█▍        | 892/6470 [10:57<1:10:57,  1.31it/s]
3|893|Loss: 0.8556278347969055:  14%|█▍        | 892/6470 [10:58<1:10:57,  1.31it/s]
3|893|Loss: 0.8556278347969055:  14%|█▍        | 893/6470 [10:58<1:09:23,  1.34it/s]
3|894|Loss: 0.9822067022323608:  14%|█▍        | 893/6470 [10:59<1:09:23,  1.34it/s]
3|894|Loss: 0.9822067022323608:  14%|█▍        | 894/6470 [10:59<1:07:05,  1.39it/s]
3|895|Loss: 0.8547760248184204:  14%|█▍        | 894/6470 [10:59<1:07:05,  1.39it/s]
3|895|Loss: 0.8547760248184204:  14%|█▍        | 895/6470 [11:00<1:12:09,  1.29it/s]
3|896|Loss: 0.9719681739807129:  14%|█▍        | 895/6470 [11:00<1:12:09,  1.29it/s]
3|896|Loss: 0.9719681739807129:  14%|█▍        | 896/6470 [11:00<1:10:18,  1.32it/s]
3|897|Loss: 0.7369773983955383:  14%|█▍        | 896/6470 [11:01<1:10:18,  1.32it/s]
3|897|Loss: 0.7369773983955383:  14%|█▍        | 897/6470 [11:01<1:13:18,  1.27it/s]
3|898|Loss: 0.8655345439910889:  14%|█▍        | 897/6470 [11:02<1:13:18,  1.27it/s]
3|898|Loss: 0.8655345439910889:  14%|█▍        | 898/6470 [11:02<1:05:45,  1.41it/s]
3|899|Loss: 0.8310792446136475:  14%|█▍        | 898/6470 [11:02<1:05:45,  1.41it/s]
3|899|Loss: 0.8310792446136475:  14%|█▍        | 899/6470 [11:02<1:04:09,  1.45it/s]
3|900|Loss: 0.894934892654419:  14%|█▍        | 899/6470 [11:03<1:04:09,  1.45it/s] 
3|900|Loss: 0.894934892654419:  14%|█▍        | 900/6470 [11:03<1:05:56,  1.41it/s]
3|901|Loss: 0.8576301336288452:  14%|█▍        | 900/6470 [11:04<1:05:56,  1.41it/s]
3|901|Loss: 0.8576301336288452:  14%|█▍        | 901/6470 [11:04<1:10:31,  1.32it/s]
3|902|Loss: 0.8015553951263428:  14%|█▍        | 901/6470 [11:05<1:10:31,  1.32it/s]
3|902|Loss: 0.8015553951263428:  14%|█▍        | 902/6470 [11:05<1:08:02,  1.36it/s]
3|903|Loss: 0.7693166732788086:  14%|█▍        | 902/6470 [11:05<1:08:02,  1.36it/s]
3|903|Loss: 0.7693166732788086:  14%|█▍        | 903/6470 [11:06<1:10:02,  1.32it/s]
3|904|Loss: 0.7825993299484253:  14%|█▍        | 903/6470 [11:06<1:10:02,  1.32it/s]
3|904|Loss: 0.7825993299484253:  14%|█▍        | 904/6470 [11:06<1:08:00,  1.36it/s]
3|905|Loss: 0.8041040897369385:  14%|█▍        | 904/6470 [11:07<1:08:00,  1.36it/s]
3|905|Loss: 0.8041040897369385:  14%|█▍        | 905/6470 [11:07<1:04:25,  1.44it/s]
3|906|Loss: 0.7521706223487854:  14%|█▍        | 905/6470 [11:07<1:04:25,  1.44it/s]
3|906|Loss: 0.7521706223487854:  14%|█▍        | 906/6470 [11:08<1:09:37,  1.33it/s]
3|907|Loss: 0.811122715473175:  14%|█▍        | 906/6470 [11:08<1:09:37,  1.33it/s] 
3|907|Loss: 0.811122715473175:  14%|█▍        | 907/6470 [11:09<1:10:50,  1.31it/s]
3|908|Loss: 0.8239228129386902:  14%|█▍        | 907/6470 [11:09<1:10:50,  1.31it/s]
3|908|Loss: 0.8239228129386902:  14%|█▍        | 908/6470 [11:09<1:11:34,  1.30it/s]
3|909|Loss: 0.8716530203819275:  14%|█▍        | 908/6470 [11:10<1:11:34,  1.30it/s]
3|909|Loss: 0.8716530203819275:  14%|█▍        | 909/6470 [11:10<1:13:33,  1.26it/s]
3|910|Loss: 0.8845389485359192:  14%|█▍        | 909/6470 [11:11<1:13:33,  1.26it/s]
3|910|Loss: 0.8845389485359192:  14%|█▍        | 910/6470 [11:11<1:10:02,  1.32it/s]
3|911|Loss: 0.919040322303772:  14%|█▍        | 910/6470 [11:11<1:10:02,  1.32it/s] 
3|911|Loss: 0.919040322303772:  14%|█▍        | 911/6470 [11:12<1:12:46,  1.27it/s]
3|912|Loss: 0.8707869648933411:  14%|█▍        | 911/6470 [11:12<1:12:46,  1.27it/s]
3|912|Loss: 0.8707869648933411:  14%|█▍        | 912/6470 [11:12<1:04:36,  1.43it/s]
3|913|Loss: 0.9311990737915039:  14%|█▍        | 912/6470 [11:13<1:04:36,  1.43it/s]
3|913|Loss: 0.9311990737915039:  14%|█▍        | 913/6470 [11:13<1:05:00,  1.42it/s]
3|914|Loss: 0.6660793423652649:  14%|█▍        | 913/6470 [11:13<1:05:00,  1.42it/s]
3|914|Loss: 0.6660793423652649:  14%|█▍        | 914/6470 [11:14<1:06:54,  1.38it/s]
3|915|Loss: 0.8530491590499878:  14%|█▍        | 914/6470 [11:14<1:06:54,  1.38it/s]
3|915|Loss: 0.8530491590499878:  14%|█▍        | 915/6470 [11:14<1:04:52,  1.43it/s]
3|916|Loss: 0.5983569622039795:  14%|█▍        | 915/6470 [11:15<1:04:52,  1.43it/s]
3|916|Loss: 0.5983569622039795:  14%|█▍        | 916/6470 [11:15<1:00:42,  1.52it/s]
3|917|Loss: 0.8827042579650879:  14%|█▍        | 916/6470 [11:15<1:00:42,  1.52it/s]
3|917|Loss: 0.8827042579650879:  14%|█▍        | 917/6470 [11:16<1:00:46,  1.52it/s]
3|918|Loss: 0.8655465841293335:  14%|█▍        | 917/6470 [11:16<1:00:46,  1.52it/s]
3|918|Loss: 0.8655465841293335:  14%|█▍        | 918/6470 [11:16<1:01:55,  1.49it/s]
3|919|Loss: 0.9524168372154236:  14%|█▍        | 918/6470 [11:17<1:01:55,  1.49it/s]
3|919|Loss: 0.9524168372154236:  14%|█▍        | 919/6470 [11:17<1:02:30,  1.48it/s]
3|920|Loss: 0.9808041453361511:  14%|█▍        | 919/6470 [11:17<1:02:30,  1.48it/s]
3|920|Loss: 0.9808041453361511:  14%|█▍        | 920/6470 [11:18<1:06:47,  1.38it/s]
3|921|Loss: 0.6753876209259033:  14%|█▍        | 920/6470 [11:18<1:06:47,  1.38it/s]
3|921|Loss: 0.6753876209259033:  14%|█▍        | 921/6470 [11:19<1:12:05,  1.28it/s]
3|922|Loss: 0.7958739995956421:  14%|█▍        | 921/6470 [11:19<1:12:05,  1.28it/s]
3|922|Loss: 0.7958739995956421:  14%|█▍        | 922/6470 [11:19<1:10:20,  1.31it/s]
3|923|Loss: 0.8383521437644958:  14%|█▍        | 922/6470 [11:20<1:10:20,  1.31it/s]
3|923|Loss: 0.8383521437644958:  14%|█▍        | 923/6470 [11:20<1:05:15,  1.42it/s]
3|924|Loss: 0.8364346027374268:  14%|█▍        | 923/6470 [11:20<1:05:15,  1.42it/s]
3|924|Loss: 0.8364346027374268:  14%|█▍        | 924/6470 [11:21<1:05:46,  1.41it/s]
3|925|Loss: 0.745197057723999:  14%|█▍        | 924/6470 [11:21<1:05:46,  1.41it/s] 
3|925|Loss: 0.745197057723999:  14%|█▍        | 925/6470 [11:21<1:01:04,  1.51it/s]
3|926|Loss: 0.933758556842804:  14%|█▍        | 925/6470 [11:22<1:01:04,  1.51it/s]
3|926|Loss: 0.933758556842804:  14%|█▍        | 926/6470 [11:22<1:07:41,  1.36it/s]
3|927|Loss: 0.8601016402244568:  14%|█▍        | 926/6470 [11:23<1:07:41,  1.36it/s]
3|927|Loss: 0.8601016402244568:  14%|█▍        | 927/6470 [11:23<1:05:40,  1.41it/s]
3|928|Loss: 0.7498082518577576:  14%|█▍        | 927/6470 [11:23<1:05:40,  1.41it/s]
3|928|Loss: 0.7498082518577576:  14%|█▍        | 928/6470 [11:24<1:06:19,  1.39it/s]
3|929|Loss: 0.8966138958930969:  14%|█▍        | 928/6470 [11:24<1:06:19,  1.39it/s]
3|929|Loss: 0.8966138958930969:  14%|█▍        | 929/6470 [11:24<1:07:28,  1.37it/s]
3|930|Loss: 0.820774495601654:  14%|█▍        | 929/6470 [11:25<1:07:28,  1.37it/s] 
3|930|Loss: 0.820774495601654:  14%|█▍        | 930/6470 [11:25<1:08:10,  1.35it/s]
3|931|Loss: 0.7860144972801208:  14%|█▍        | 930/6470 [11:25<1:08:10,  1.35it/s]
3|931|Loss: 0.7860144972801208:  14%|█▍        | 931/6470 [11:26<1:06:25,  1.39it/s]
3|932|Loss: 0.9083311557769775:  14%|█▍        | 931/6470 [11:26<1:06:25,  1.39it/s]
3|932|Loss: 0.9083311557769775:  14%|█▍        | 932/6470 [11:27<1:09:34,  1.33it/s]
3|933|Loss: 0.7751848101615906:  14%|█▍        | 932/6470 [11:27<1:09:34,  1.33it/s]
3|933|Loss: 0.7751848101615906:  14%|█▍        | 933/6470 [11:27<1:10:02,  1.32it/s]
3|934|Loss: 0.8142644166946411:  14%|█▍        | 933/6470 [11:28<1:10:02,  1.32it/s]
3|934|Loss: 0.8142644166946411:  14%|█▍        | 934/6470 [11:28<1:10:39,  1.31it/s]
3|935|Loss: 0.774655818939209:  14%|█▍        | 934/6470 [11:29<1:10:39,  1.31it/s] 
3|935|Loss: 0.774655818939209:  14%|█▍        | 935/6470 [11:29<1:09:07,  1.33it/s]
3|936|Loss: 0.8217602968215942:  14%|█▍        | 935/6470 [11:29<1:09:07,  1.33it/s]
3|936|Loss: 0.8217602968215942:  14%|█▍        | 936/6470 [11:29<1:05:52,  1.40it/s]
3|937|Loss: 0.8601205348968506:  14%|█▍        | 936/6470 [11:30<1:05:52,  1.40it/s]
3|937|Loss: 0.8601205348968506:  14%|█▍        | 937/6470 [11:30<1:02:34,  1.47it/s]
3|938|Loss: 0.8711045384407043:  14%|█▍        | 937/6470 [11:30<1:02:34,  1.47it/s]
3|938|Loss: 0.8711045384407043:  14%|█▍        | 938/6470 [11:31<1:04:32,  1.43it/s]
3|939|Loss: 0.7180088758468628:  14%|█▍        | 938/6470 [11:31<1:04:32,  1.43it/s]
3|939|Loss: 0.7180088758468628:  15%|█▍        | 939/6470 [11:32<1:13:15,  1.26it/s]
3|940|Loss: 0.9830957651138306:  15%|█▍        | 939/6470 [11:32<1:13:15,  1.26it/s]
3|940|Loss: 0.9830957651138306:  15%|█▍        | 940/6470 [11:32<1:09:36,  1.32it/s]
3|941|Loss: 0.7122441530227661:  15%|█▍        | 940/6470 [11:33<1:09:36,  1.32it/s]
3|941|Loss: 0.7122441530227661:  15%|█▍        | 941/6470 [11:33<1:06:01,  1.40it/s]
3|942|Loss: 0.891268789768219:  15%|█▍        | 941/6470 [11:34<1:06:01,  1.40it/s] 
3|942|Loss: 0.891268789768219:  15%|█▍        | 942/6470 [11:34<1:07:55,  1.36it/s]
3|943|Loss: 0.7247785329818726:  15%|█▍        | 942/6470 [11:34<1:07:55,  1.36it/s]
3|943|Loss: 0.7247785329818726:  15%|█▍        | 943/6470 [11:35<1:08:53,  1.34it/s]
3|944|Loss: 0.720165491104126:  15%|█▍        | 943/6470 [11:35<1:08:53,  1.34it/s] 
3|944|Loss: 0.720165491104126:  15%|█▍        | 944/6470 [11:35<1:07:04,  1.37it/s]
3|945|Loss: 0.7821136713027954:  15%|█▍        | 944/6470 [11:36<1:07:04,  1.37it/s]
3|945|Loss: 0.7821136713027954:  15%|█▍        | 945/6470 [11:36<1:02:56,  1.46it/s]
3|946|Loss: 0.8071141242980957:  15%|█▍        | 945/6470 [11:36<1:02:56,  1.46it/s]
3|946|Loss: 0.8071141242980957:  15%|█▍        | 946/6470 [11:37<1:05:03,  1.42it/s]
3|947|Loss: 0.8867409229278564:  15%|█▍        | 946/6470 [11:37<1:05:03,  1.42it/s]
3|947|Loss: 0.8867409229278564:  15%|█▍        | 947/6470 [11:37<1:04:39,  1.42it/s]
3|948|Loss: 0.9581202268600464:  15%|█▍        | 947/6470 [11:38<1:04:39,  1.42it/s]
3|948|Loss: 0.9581202268600464:  15%|█▍        | 948/6470 [11:38<1:09:20,  1.33it/s]
3|949|Loss: 0.875418484210968:  15%|█▍        | 948/6470 [11:39<1:09:20,  1.33it/s] 
3|949|Loss: 0.875418484210968:  15%|█▍        | 949/6470 [11:39<1:08:56,  1.33it/s]
3|950|Loss: 0.7442623376846313:  15%|█▍        | 949/6470 [11:39<1:08:56,  1.33it/s]
3|950|Loss: 0.7442623376846313:  15%|█▍        | 950/6470 [11:40<1:06:49,  1.38it/s]
3|951|Loss: 0.7169306874275208:  15%|█▍        | 950/6470 [11:40<1:06:49,  1.38it/s]
3|951|Loss: 0.7169306874275208:  15%|█▍        | 951/6470 [11:40<1:05:11,  1.41it/s]
3|952|Loss: 0.7774033546447754:  15%|█▍        | 951/6470 [11:41<1:05:11,  1.41it/s]
3|952|Loss: 0.7774033546447754:  15%|█▍        | 952/6470 [11:41<1:04:33,  1.42it/s]
3|953|Loss: 0.8533893823623657:  15%|█▍        | 952/6470 [11:41<1:04:33,  1.42it/s]
3|953|Loss: 0.8533893823623657:  15%|█▍        | 953/6470 [11:42<1:08:13,  1.35it/s]
3|954|Loss: 0.9878837466239929:  15%|█▍        | 953/6470 [11:42<1:08:13,  1.35it/s]
3|954|Loss: 0.9878837466239929:  15%|█▍        | 954/6470 [11:43<1:10:02,  1.31it/s]
3|955|Loss: 0.8191556930541992:  15%|█▍        | 954/6470 [11:43<1:10:02,  1.31it/s]
3|955|Loss: 0.8191556930541992:  15%|█▍        | 955/6470 [11:43<1:08:56,  1.33it/s]
3|956|Loss: 0.7141426801681519:  15%|█▍        | 955/6470 [11:44<1:08:56,  1.33it/s]
3|956|Loss: 0.7141426801681519:  15%|█▍        | 956/6470 [11:44<1:07:46,  1.36it/s]
3|957|Loss: 0.8362944722175598:  15%|█▍        | 956/6470 [11:44<1:07:46,  1.36it/s]
3|957|Loss: 0.8362944722175598:  15%|█▍        | 957/6470 [11:45<1:06:39,  1.38it/s]
3|958|Loss: 0.8284666538238525:  15%|█▍        | 957/6470 [11:45<1:06:39,  1.38it/s]
3|958|Loss: 0.8284666538238525:  15%|█▍        | 958/6470 [11:46<1:09:25,  1.32it/s]
3|959|Loss: 0.8434078097343445:  15%|█▍        | 958/6470 [11:46<1:09:25,  1.32it/s]
3|959|Loss: 0.8434078097343445:  15%|█▍        | 959/6470 [11:46<1:06:16,  1.39it/s]
3|960|Loss: 0.7821511626243591:  15%|█▍        | 959/6470 [11:47<1:06:16,  1.39it/s]
3|960|Loss: 0.7821511626243591:  15%|█▍        | 960/6470 [11:47<1:09:29,  1.32it/s]
3|961|Loss: 0.8410494923591614:  15%|█▍        | 960/6470 [11:48<1:09:29,  1.32it/s]
3|961|Loss: 0.8410494923591614:  15%|█▍        | 961/6470 [11:48<1:10:04,  1.31it/s]
3|962|Loss: 0.9119086861610413:  15%|█▍        | 961/6470 [11:48<1:10:04,  1.31it/s]
3|962|Loss: 0.9119086861610413:  15%|█▍        | 962/6470 [11:49<1:08:46,  1.33it/s]
3|963|Loss: 0.7892500162124634:  15%|█▍        | 962/6470 [11:49<1:08:46,  1.33it/s]
3|963|Loss: 0.7892500162124634:  15%|█▍        | 963/6470 [11:49<1:02:23,  1.47it/s]
3|964|Loss: 0.9717189073562622:  15%|█▍        | 963/6470 [11:49<1:02:23,  1.47it/s]
3|964|Loss: 0.9717189073562622:  15%|█▍        | 964/6470 [11:50<1:01:28,  1.49it/s]
3|965|Loss: 0.8894742131233215:  15%|█▍        | 964/6470 [11:50<1:01:28,  1.49it/s]
3|965|Loss: 0.8894742131233215:  15%|█▍        | 965/6470 [11:51<1:03:51,  1.44it/s]
3|966|Loss: 0.8223300576210022:  15%|█▍        | 965/6470 [11:51<1:03:51,  1.44it/s]
3|966|Loss: 0.8223300576210022:  15%|█▍        | 966/6470 [11:51<1:00:43,  1.51it/s]
3|967|Loss: 0.8138235211372375:  15%|█▍        | 966/6470 [11:51<1:00:43,  1.51it/s]
3|967|Loss: 0.8138235211372375:  15%|█▍        | 967/6470 [11:52<1:00:26,  1.52it/s]
3|968|Loss: 0.6425986886024475:  15%|█▍        | 967/6470 [11:52<1:00:26,  1.52it/s]
3|968|Loss: 0.6425986886024475:  15%|█▍        | 968/6470 [11:52<1:02:49,  1.46it/s]
3|969|Loss: 0.8729103803634644:  15%|█▍        | 968/6470 [11:53<1:02:49,  1.46it/s]
3|969|Loss: 0.8729103803634644:  15%|█▍        | 969/6470 [11:53<1:01:42,  1.49it/s]
3|970|Loss: 0.7405475974082947:  15%|█▍        | 969/6470 [11:53<1:01:42,  1.49it/s]
3|970|Loss: 0.7405475974082947:  15%|█▍        | 970/6470 [11:54<1:00:57,  1.50it/s]
3|971|Loss: 0.8976457715034485:  15%|█▍        | 970/6470 [11:54<1:00:57,  1.50it/s]
3|971|Loss: 0.8976457715034485:  15%|█▌        | 971/6470 [11:55<1:06:39,  1.37it/s]
3|972|Loss: 0.8976074457168579:  15%|█▌        | 971/6470 [11:55<1:06:39,  1.37it/s]
3|972|Loss: 0.8976074457168579:  15%|█▌        | 972/6470 [11:55<1:09:27,  1.32it/s]
3|973|Loss: 0.8492671251296997:  15%|█▌        | 972/6470 [11:56<1:09:27,  1.32it/s]
3|973|Loss: 0.8492671251296997:  15%|█▌        | 973/6470 [11:56<1:01:21,  1.49it/s]
3|974|Loss: 0.814050555229187:  15%|█▌        | 973/6470 [11:56<1:01:21,  1.49it/s] 
3|974|Loss: 0.814050555229187:  15%|█▌        | 974/6470 [11:57<59:53,  1.53it/s]  
3|975|Loss: 0.7877907752990723:  15%|█▌        | 974/6470 [11:57<59:53,  1.53it/s]
3|975|Loss: 0.7877907752990723:  15%|█▌        | 975/6470 [11:57<1:04:52,  1.41it/s]
3|976|Loss: 0.8270033001899719:  15%|█▌        | 975/6470 [11:58<1:04:52,  1.41it/s]
3|976|Loss: 0.8270033001899719:  15%|█▌        | 976/6470 [11:58<1:07:54,  1.35it/s]
3|977|Loss: 0.8689218759536743:  15%|█▌        | 976/6470 [11:59<1:07:54,  1.35it/s]
3|977|Loss: 0.8689218759536743:  15%|█▌        | 977/6470 [11:59<1:13:21,  1.25it/s]
3|978|Loss: 0.8205306529998779:  15%|█▌        | 977/6470 [12:00<1:13:21,  1.25it/s]
3|978|Loss: 0.8205306529998779:  15%|█▌        | 978/6470 [12:00<1:11:02,  1.29it/s]
3|979|Loss: 0.6854826211929321:  15%|█▌        | 978/6470 [12:00<1:11:02,  1.29it/s]
3|979|Loss: 0.6854826211929321:  15%|█▌        | 979/6470 [12:00<1:06:44,  1.37it/s]
3|980|Loss: 0.6791581511497498:  15%|█▌        | 979/6470 [12:01<1:06:44,  1.37it/s]
3|980|Loss: 0.6791581511497498:  15%|█▌        | 980/6470 [12:01<1:10:07,  1.30it/s]
3|981|Loss: 0.8708695769309998:  15%|█▌        | 980/6470 [12:02<1:10:07,  1.30it/s]
3|981|Loss: 0.8708695769309998:  15%|█▌        | 981/6470 [12:02<1:11:19,  1.28it/s]
3|982|Loss: 0.8374985456466675:  15%|█▌        | 981/6470 [12:03<1:11:19,  1.28it/s]
3|982|Loss: 0.8374985456466675:  15%|█▌        | 982/6470 [12:03<1:11:34,  1.28it/s]
3|983|Loss: 0.9084760546684265:  15%|█▌        | 982/6470 [12:03<1:11:34,  1.28it/s]
3|983|Loss: 0.9084760546684265:  15%|█▌        | 983/6470 [12:04<1:11:15,  1.28it/s]
3|984|Loss: 0.8000728487968445:  15%|█▌        | 983/6470 [12:04<1:11:15,  1.28it/s]
3|984|Loss: 0.8000728487968445:  15%|█▌        | 984/6470 [12:04<1:08:42,  1.33it/s]
3|985|Loss: 0.89003586769104:  15%|█▌        | 984/6470 [12:05<1:08:42,  1.33it/s]  
3|985|Loss: 0.89003586769104:  15%|█▌        | 985/6470 [12:05<1:12:28,  1.26it/s]
3|986|Loss: 0.8616487979888916:  15%|█▌        | 985/6470 [12:06<1:12:28,  1.26it/s]
3|986|Loss: 0.8616487979888916:  15%|█▌        | 986/6470 [12:06<1:17:52,  1.17it/s]
3|987|Loss: 0.8121518492698669:  15%|█▌        | 986/6470 [12:07<1:17:52,  1.17it/s]
3|987|Loss: 0.8121518492698669:  15%|█▌        | 987/6470 [12:07<1:17:02,  1.19it/s]
3|988|Loss: 0.9487581253051758:  15%|█▌        | 987/6470 [12:08<1:17:02,  1.19it/s]
3|988|Loss: 0.9487581253051758:  15%|█▌        | 988/6470 [12:08<1:14:48,  1.22it/s]
3|989|Loss: 0.8440635204315186:  15%|█▌        | 988/6470 [12:08<1:14:48,  1.22it/s]
3|989|Loss: 0.8440635204315186:  15%|█▌        | 989/6470 [12:09<1:10:57,  1.29it/s]
3|990|Loss: 0.8906666040420532:  15%|█▌        | 989/6470 [12:09<1:10:57,  1.29it/s]
3|990|Loss: 0.8906666040420532:  15%|█▌        | 990/6470 [12:09<1:08:38,  1.33it/s]
3|991|Loss: 0.8126628994941711:  15%|█▌        | 990/6470 [12:10<1:08:38,  1.33it/s]
3|991|Loss: 0.8126628994941711:  15%|█▌        | 991/6470 [12:10<1:12:28,  1.26it/s]
3|992|Loss: 0.9551056623458862:  15%|█▌        | 991/6470 [12:11<1:12:28,  1.26it/s]
3|992|Loss: 0.9551056623458862:  15%|█▌        | 992/6470 [12:11<1:10:30,  1.29it/s]
3|993|Loss: 0.7921851873397827:  15%|█▌        | 992/6470 [12:11<1:10:30,  1.29it/s]
3|993|Loss: 0.7921851873397827:  15%|█▌        | 993/6470 [12:11<1:06:23,  1.37it/s]
3|994|Loss: 0.9192236065864563:  15%|█▌        | 993/6470 [12:12<1:06:23,  1.37it/s]
3|994|Loss: 0.9192236065864563:  15%|█▌        | 994/6470 [12:12<1:07:22,  1.35it/s]
3|995|Loss: 0.87635737657547:  15%|█▌        | 994/6470 [12:13<1:07:22,  1.35it/s]  
3|995|Loss: 0.87635737657547:  15%|█▌        | 995/6470 [12:13<1:07:59,  1.34it/s]
3|996|Loss: 0.6947802901268005:  15%|█▌        | 995/6470 [12:13<1:07:59,  1.34it/s]
3|996|Loss: 0.6947802901268005:  15%|█▌        | 996/6470 [12:14<1:06:55,  1.36it/s]
3|997|Loss: 0.8516791462898254:  15%|█▌        | 996/6470 [12:14<1:06:55,  1.36it/s]
3|997|Loss: 0.8516791462898254:  15%|█▌        | 997/6470 [12:14<1:07:12,  1.36it/s]
3|998|Loss: 0.8707869648933411:  15%|█▌        | 997/6470 [12:15<1:07:12,  1.36it/s]
3|998|Loss: 0.8707869648933411:  15%|█▌        | 998/6470 [12:15<1:05:45,  1.39it/s]
3|999|Loss: 0.7708646655082703:  15%|█▌        | 998/6470 [12:16<1:05:45,  1.39it/s]
3|999|Loss: 0.7708646655082703:  15%|█▌        | 999/6470 [12:16<1:07:06,  1.36it/s]
3|1000|Loss: 0.8394184708595276:  15%|█▌        | 999/6470 [12:16<1:07:06,  1.36it/s]
3|1000|Loss: 0.8394184708595276:  15%|█▌        | 1000/6470 [12:17<1:07:41,  1.35it/s]
3|1001|Loss: 0.7924618721008301:  15%|█▌        | 1000/6470 [12:17<1:07:41,  1.35it/s]
3|1001|Loss: 0.7924618721008301:  15%|█▌        | 1001/6470 [12:17<1:06:43,  1.37it/s]
3|1002|Loss: 0.704319953918457:  15%|█▌        | 1001/6470 [12:18<1:06:43,  1.37it/s] 
3|1002|Loss: 0.704319953918457:  15%|█▌        | 1002/6470 [12:18<57:42,  1.58it/s]  
3|1003|Loss: 0.8408793210983276:  15%|█▌        | 1002/6470 [12:18<57:42,  1.58it/s]
3|1003|Loss: 0.8408793210983276:  16%|█▌        | 1003/6470 [12:18<58:17,  1.56it/s]
3|1004|Loss: 0.8188270330429077:  16%|█▌        | 1003/6470 [12:19<58:17,  1.56it/s]
3|1004|Loss: 0.8188270330429077:  16%|█▌        | 1004/6470 [12:19<1:05:09,  1.40it/s]
3|1005|Loss: 0.8852212429046631:  16%|█▌        | 1004/6470 [12:20<1:05:09,  1.40it/s]
3|1005|Loss: 0.8852212429046631:  16%|█▌        | 1005/6470 [12:20<1:11:12,  1.28it/s]
3|1006|Loss: 0.7605225443840027:  16%|█▌        | 1005/6470 [12:21<1:11:12,  1.28it/s]
3|1006|Loss: 0.7605225443840027:  16%|█▌        | 1006/6470 [12:21<1:15:38,  1.20it/s]
3|1007|Loss: 0.8252527117729187:  16%|█▌        | 1006/6470 [12:22<1:15:38,  1.20it/s]
3|1007|Loss: 0.8252527117729187:  16%|█▌        | 1007/6470 [12:22<1:14:46,  1.22it/s]
3|1008|Loss: 0.8164684772491455:  16%|█▌        | 1007/6470 [12:22<1:14:46,  1.22it/s]
3|1008|Loss: 0.8164684772491455:  16%|█▌        | 1008/6470 [12:23<1:14:01,  1.23it/s]
3|1009|Loss: 0.9224727153778076:  16%|█▌        | 1008/6470 [12:23<1:14:01,  1.23it/s]
3|1009|Loss: 0.9224727153778076:  16%|█▌        | 1009/6470 [12:24<1:13:20,  1.24it/s]
3|1010|Loss: 0.8320443630218506:  16%|█▌        | 1009/6470 [12:24<1:13:20,  1.24it/s]
3|1010|Loss: 0.8320443630218506:  16%|█▌        | 1010/6470 [12:24<1:12:30,  1.26it/s]
3|1011|Loss: 0.904000997543335:  16%|█▌        | 1010/6470 [12:25<1:12:30,  1.26it/s] 
3|1011|Loss: 0.904000997543335:  16%|█▌        | 1011/6470 [12:25<1:12:10,  1.26it/s]
3|1012|Loss: 0.8036786317825317:  16%|█▌        | 1011/6470 [12:26<1:12:10,  1.26it/s]
3|1012|Loss: 0.8036786317825317:  16%|█▌        | 1012/6470 [12:26<1:10:05,  1.30it/s]
3|1013|Loss: 0.9980100393295288:  16%|█▌        | 1012/6470 [12:26<1:10:05,  1.30it/s]
3|1013|Loss: 0.9980100393295288:  16%|█▌        | 1013/6470 [12:27<1:07:33,  1.35it/s]
3|1014|Loss: 0.7709397673606873:  16%|█▌        | 1013/6470 [12:27<1:07:33,  1.35it/s]
3|1014|Loss: 0.7709397673606873:  16%|█▌        | 1014/6470 [12:27<1:01:23,  1.48it/s]
3|1015|Loss: 0.9191317558288574:  16%|█▌        | 1014/6470 [12:27<1:01:23,  1.48it/s]
3|1015|Loss: 0.9191317558288574:  16%|█▌        | 1015/6470 [12:28<1:02:20,  1.46it/s]
3|1016|Loss: 0.9651939272880554:  16%|█▌        | 1015/6470 [12:28<1:02:20,  1.46it/s]
3|1016|Loss: 0.9651939272880554:  16%|█▌        | 1016/6470 [12:29<1:05:28,  1.39it/s]
3|1017|Loss: 0.7330186367034912:  16%|█▌        | 1016/6470 [12:29<1:05:28,  1.39it/s]
3|1017|Loss: 0.7330186367034912:  16%|█▌        | 1017/6470 [12:29<1:05:42,  1.38it/s]
3|1018|Loss: 0.8334991931915283:  16%|█▌        | 1017/6470 [12:30<1:05:42,  1.38it/s]
3|1018|Loss: 0.8334991931915283:  16%|█▌        | 1018/6470 [12:30<1:00:08,  1.51it/s]
3|1019|Loss: 0.711164653301239:  16%|█▌        | 1018/6470 [12:30<1:00:08,  1.51it/s] 
3|1019|Loss: 0.711164653301239:  16%|█▌        | 1019/6470 [12:30<59:53,  1.52it/s]  
3|1020|Loss: 0.8665368556976318:  16%|█▌        | 1019/6470 [12:31<59:53,  1.52it/s]
3|1020|Loss: 0.8665368556976318:  16%|█▌        | 1020/6470 [12:31<1:03:00,  1.44it/s]
3|1021|Loss: 0.7787524461746216:  16%|█▌        | 1020/6470 [12:32<1:03:00,  1.44it/s]
3|1021|Loss: 0.7787524461746216:  16%|█▌        | 1021/6470 [12:32<1:07:54,  1.34it/s]
3|1022|Loss: 0.9274831414222717:  16%|█▌        | 1021/6470 [12:33<1:07:54,  1.34it/s]
3|1022|Loss: 0.9274831414222717:  16%|█▌        | 1022/6470 [12:33<1:08:43,  1.32it/s]
3|1023|Loss: 0.8761230707168579:  16%|█▌        | 1022/6470 [12:33<1:08:43,  1.32it/s]
3|1023|Loss: 0.8761230707168579:  16%|█▌        | 1023/6470 [12:34<1:11:22,  1.27it/s]
3|1024|Loss: 0.7530863881111145:  16%|█▌        | 1023/6470 [12:34<1:11:22,  1.27it/s]
3|1024|Loss: 0.7530863881111145:  16%|█▌        | 1024/6470 [12:34<1:05:02,  1.40it/s]
3|1025|Loss: 0.7444308996200562:  16%|█▌        | 1024/6470 [12:35<1:05:02,  1.40it/s]
3|1025|Loss: 0.7444308996200562:  16%|█▌        | 1025/6470 [12:35<1:03:06,  1.44it/s]
3|1026|Loss: 0.68768310546875:  16%|█▌        | 1025/6470 [12:35<1:03:06,  1.44it/s]  
3|1026|Loss: 0.68768310546875:  16%|█▌        | 1026/6470 [12:36<1:01:50,  1.47it/s]
3|1027|Loss: 0.6643436551094055:  16%|█▌        | 1026/6470 [12:36<1:01:50,  1.47it/s]
3|1027|Loss: 0.6643436551094055:  16%|█▌        | 1027/6470 [12:36<1:07:12,  1.35it/s]
3|1028|Loss: 0.8967558145523071:  16%|█▌        | 1027/6470 [12:37<1:07:12,  1.35it/s]
3|1028|Loss: 0.8967558145523071:  16%|█▌        | 1028/6470 [12:37<1:09:38,  1.30it/s]
3|1029|Loss: 0.854158341884613:  16%|█▌        | 1028/6470 [12:38<1:09:38,  1.30it/s] 
3|1029|Loss: 0.854158341884613:  16%|█▌        | 1029/6470 [12:38<1:11:55,  1.26it/s]
3|1030|Loss: 0.7172867655754089:  16%|█▌        | 1029/6470 [12:39<1:11:55,  1.26it/s]
3|1030|Loss: 0.7172867655754089:  16%|█▌        | 1030/6470 [12:39<1:12:34,  1.25it/s]
3|1031|Loss: 0.8710977435112:  16%|█▌        | 1030/6470 [12:39<1:12:34,  1.25it/s]   
3|1031|Loss: 0.8710977435112:  16%|█▌        | 1031/6470 [12:40<1:06:08,  1.37it/s]
3|1032|Loss: 0.7748169302940369:  16%|█▌        | 1031/6470 [12:40<1:06:08,  1.37it/s]
3|1032|Loss: 0.7748169302940369:  16%|█▌        | 1032/6470 [12:40<1:05:44,  1.38it/s]
3|1033|Loss: 0.9447648525238037:  16%|█▌        | 1032/6470 [12:41<1:05:44,  1.38it/s]
3|1033|Loss: 0.9447648525238037:  16%|█▌        | 1033/6470 [12:41<1:06:45,  1.36it/s]
3|1034|Loss: 0.8591840863227844:  16%|█▌        | 1033/6470 [12:41<1:06:45,  1.36it/s]
3|1034|Loss: 0.8591840863227844:  16%|█▌        | 1034/6470 [12:42<1:06:05,  1.37it/s]
3|1035|Loss: 0.766403079032898:  16%|█▌        | 1034/6470 [12:42<1:06:05,  1.37it/s] 
3|1035|Loss: 0.766403079032898:  16%|█▌        | 1035/6470 [12:42<1:02:05,  1.46it/s]
3|1036|Loss: 0.8318766951560974:  16%|█▌        | 1035/6470 [12:43<1:02:05,  1.46it/s]
3|1036|Loss: 0.8318766951560974:  16%|█▌        | 1036/6470 [12:43<1:05:57,  1.37it/s]
3|1037|Loss: 0.8765830397605896:  16%|█▌        | 1036/6470 [12:44<1:05:57,  1.37it/s]
3|1037|Loss: 0.8765830397605896:  16%|█▌        | 1037/6470 [12:44<1:05:05,  1.39it/s]
3|1038|Loss: 0.7707069516181946:  16%|█▌        | 1037/6470 [12:44<1:05:05,  1.39it/s]
3|1038|Loss: 0.7707069516181946:  16%|█▌        | 1038/6470 [12:45<1:04:05,  1.41it/s]
3|1039|Loss: 0.7771235108375549:  16%|█▌        | 1038/6470 [12:45<1:04:05,  1.41it/s]
3|1039|Loss: 0.7771235108375549:  16%|█▌        | 1039/6470 [12:45<1:02:57,  1.44it/s]
3|1040|Loss: 0.755923867225647:  16%|█▌        | 1039/6470 [12:46<1:02:57,  1.44it/s] 
3|1040|Loss: 0.755923867225647:  16%|█▌        | 1040/6470 [12:46<1:04:59,  1.39it/s]
3|1041|Loss: 0.8432511687278748:  16%|█▌        | 1040/6470 [12:46<1:04:59,  1.39it/s]
3|1041|Loss: 0.8432511687278748:  16%|█▌        | 1041/6470 [12:47<1:05:07,  1.39it/s]
3|1042|Loss: 0.7910493612289429:  16%|█▌        | 1041/6470 [12:47<1:05:07,  1.39it/s]
3|1042|Loss: 0.7910493612289429:  16%|█▌        | 1042/6470 [12:47<1:07:18,  1.34it/s]
3|1043|Loss: 0.8943588137626648:  16%|█▌        | 1042/6470 [12:48<1:07:18,  1.34it/s]
3|1043|Loss: 0.8943588137626648:  16%|█▌        | 1043/6470 [12:48<1:06:57,  1.35it/s]
3|1044|Loss: 0.8640398979187012:  16%|█▌        | 1043/6470 [12:49<1:06:57,  1.35it/s]
3|1044|Loss: 0.8640398979187012:  16%|█▌        | 1044/6470 [12:49<1:10:45,  1.28it/s]
3|1045|Loss: 0.9149184823036194:  16%|█▌        | 1044/6470 [12:50<1:10:45,  1.28it/s]
3|1045|Loss: 0.9149184823036194:  16%|█▌        | 1045/6470 [12:50<1:09:47,  1.30it/s]
3|1046|Loss: 0.9207902550697327:  16%|█▌        | 1045/6470 [12:50<1:09:47,  1.30it/s]
3|1046|Loss: 0.9207902550697327:  16%|█▌        | 1046/6470 [12:51<1:07:55,  1.33it/s]
3|1047|Loss: 0.8553045392036438:  16%|█▌        | 1046/6470 [12:51<1:07:55,  1.33it/s]
3|1047|Loss: 0.8553045392036438:  16%|█▌        | 1047/6470 [12:51<1:08:57,  1.31it/s]
3|1048|Loss: 0.6563251614570618:  16%|█▌        | 1047/6470 [12:52<1:08:57,  1.31it/s]
3|1048|Loss: 0.6563251614570618:  16%|█▌        | 1048/6470 [12:52<1:07:41,  1.33it/s]
3|1049|Loss: 0.8416430354118347:  16%|█▌        | 1048/6470 [12:52<1:07:41,  1.33it/s]
3|1049|Loss: 0.8416430354118347:  16%|█▌        | 1049/6470 [12:53<1:08:20,  1.32it/s]
3|1050|Loss: 0.8592062592506409:  16%|█▌        | 1049/6470 [12:53<1:08:20,  1.32it/s]
3|1050|Loss: 0.8592062592506409:  16%|█▌        | 1050/6470 [12:54<1:09:36,  1.30it/s]
3|1051|Loss: 0.8225414752960205:  16%|█▌        | 1050/6470 [12:54<1:09:36,  1.30it/s]
3|1051|Loss: 0.8225414752960205:  16%|█▌        | 1051/6470 [12:54<1:11:07,  1.27it/s]
3|1052|Loss: 0.7185068726539612:  16%|█▌        | 1051/6470 [12:55<1:11:07,  1.27it/s]
3|1052|Loss: 0.7185068726539612:  16%|█▋        | 1052/6470 [12:55<1:15:51,  1.19it/s]
3|1053|Loss: 0.8543478846549988:  16%|█▋        | 1052/6470 [12:56<1:15:51,  1.19it/s]
3|1053|Loss: 0.8543478846549988:  16%|█▋        | 1053/6470 [12:56<1:18:01,  1.16it/s]
3|1054|Loss: 0.8613688349723816:  16%|█▋        | 1053/6470 [12:57<1:18:01,  1.16it/s]
3|1054|Loss: 0.8613688349723816:  16%|█▋        | 1054/6470 [12:57<1:15:41,  1.19it/s]
3|1055|Loss: 0.8397244215011597:  16%|█▋        | 1054/6470 [12:58<1:15:41,  1.19it/s]
3|1055|Loss: 0.8397244215011597:  16%|█▋        | 1055/6470 [12:58<1:13:18,  1.23it/s]
3|1056|Loss: 0.8881456851959229:  16%|█▋        | 1055/6470 [12:58<1:13:18,  1.23it/s]
3|1056|Loss: 0.8881456851959229:  16%|█▋        | 1056/6470 [12:59<1:12:25,  1.25it/s]
3|1057|Loss: 1.0845352411270142:  16%|█▋        | 1056/6470 [12:59<1:12:25,  1.25it/s]
3|1057|Loss: 1.0845352411270142:  16%|█▋        | 1057/6470 [13:00<1:14:39,  1.21it/s]
3|1058|Loss: 0.8503948450088501:  16%|█▋        | 1057/6470 [13:00<1:14:39,  1.21it/s]
3|1058|Loss: 0.8503948450088501:  16%|█▋        | 1058/6470 [13:00<1:17:53,  1.16it/s]
3|1059|Loss: 0.7984659671783447:  16%|█▋        | 1058/6470 [13:01<1:17:53,  1.16it/s]
3|1059|Loss: 0.7984659671783447:  16%|█▋        | 1059/6470 [13:01<1:17:33,  1.16it/s]
3|1060|Loss: 0.680171012878418:  16%|█▋        | 1059/6470 [13:02<1:17:33,  1.16it/s] 
3|1060|Loss: 0.680171012878418:  16%|█▋        | 1060/6470 [13:02<1:15:00,  1.20it/s]
3|1061|Loss: 0.8010786175727844:  16%|█▋        | 1060/6470 [13:02<1:15:00,  1.20it/s]
3|1061|Loss: 0.8010786175727844:  16%|█▋        | 1061/6470 [13:03<1:10:52,  1.27it/s]
3|1062|Loss: 0.732168436050415:  16%|█▋        | 1061/6470 [13:03<1:10:52,  1.27it/s] 
3|1062|Loss: 0.732168436050415:  16%|█▋        | 1062/6470 [13:04<1:10:09,  1.28it/s]
3|1063|Loss: 0.782136857509613:  16%|█▋        | 1062/6470 [13:04<1:10:09,  1.28it/s]
3|1063|Loss: 0.782136857509613:  16%|█▋        | 1063/6470 [13:04<1:10:52,  1.27it/s]
3|1064|Loss: 0.7608534097671509:  16%|█▋        | 1063/6470 [13:05<1:10:52,  1.27it/s]
3|1064|Loss: 0.7608534097671509:  16%|█▋        | 1064/6470 [13:05<1:09:17,  1.30it/s]
3|1065|Loss: 0.8604312539100647:  16%|█▋        | 1064/6470 [13:05<1:09:17,  1.30it/s]
3|1065|Loss: 0.8604312539100647:  16%|█▋        | 1065/6470 [13:06<1:08:44,  1.31it/s]
3|1066|Loss: 0.8119733333587646:  16%|█▋        | 1065/6470 [13:06<1:08:44,  1.31it/s]
3|1066|Loss: 0.8119733333587646:  16%|█▋        | 1066/6470 [13:07<1:08:27,  1.32it/s]
3|1067|Loss: 0.7480731010437012:  16%|█▋        | 1066/6470 [13:07<1:08:27,  1.32it/s]
3|1067|Loss: 0.7480731010437012:  16%|█▋        | 1067/6470 [13:07<1:07:59,  1.32it/s]
3|1068|Loss: 0.7431545257568359:  16%|█▋        | 1067/6470 [13:08<1:07:59,  1.32it/s]
3|1068|Loss: 0.7431545257568359:  17%|█▋        | 1068/6470 [13:08<1:05:48,  1.37it/s]
3|1069|Loss: 0.9854955673217773:  17%|█▋        | 1068/6470 [13:08<1:05:48,  1.37it/s]
3|1069|Loss: 0.9854955673217773:  17%|█▋        | 1069/6470 [13:09<1:03:14,  1.42it/s]
3|1070|Loss: 0.8440684676170349:  17%|█▋        | 1069/6470 [13:09<1:03:14,  1.42it/s]
3|1070|Loss: 0.8440684676170349:  17%|█▋        | 1070/6470 [13:09<1:01:51,  1.45it/s]
3|1071|Loss: 0.8962905406951904:  17%|█▋        | 1070/6470 [13:10<1:01:51,  1.45it/s]
3|1071|Loss: 0.8962905406951904:  17%|█▋        | 1071/6470 [13:10<1:03:49,  1.41it/s]
3|1072|Loss: 0.7248935103416443:  17%|█▋        | 1071/6470 [13:11<1:03:49,  1.41it/s]
3|1072|Loss: 0.7248935103416443:  17%|█▋        | 1072/6470 [13:11<1:07:42,  1.33it/s]
3|1073|Loss: 0.846433699131012:  17%|█▋        | 1072/6470 [13:11<1:07:42,  1.33it/s] 
3|1073|Loss: 0.846433699131012:  17%|█▋        | 1073/6470 [13:12<1:04:16,  1.40it/s]
3|1074|Loss: 0.9140466451644897:  17%|█▋        | 1073/6470 [13:12<1:04:16,  1.40it/s]
3|1074|Loss: 0.9140466451644897:  17%|█▋        | 1074/6470 [13:12<1:01:08,  1.47it/s]
3|1075|Loss: 0.8147421479225159:  17%|█▋        | 1074/6470 [13:12<1:01:08,  1.47it/s]
3|1075|Loss: 0.8147421479225159:  17%|█▋        | 1075/6470 [13:13<58:19,  1.54it/s]  
3|1076|Loss: 0.8179625272750854:  17%|█▋        | 1075/6470 [13:13<58:19,  1.54it/s]
3|1076|Loss: 0.8179625272750854:  17%|█▋        | 1076/6470 [13:13<1:01:39,  1.46it/s]
3|1077|Loss: 0.8436972498893738:  17%|█▋        | 1076/6470 [13:14<1:01:39,  1.46it/s]
3|1077|Loss: 0.8436972498893738:  17%|█▋        | 1077/6470 [13:14<1:01:18,  1.47it/s]
3|1078|Loss: 0.7479856014251709:  17%|█▋        | 1077/6470 [13:15<1:01:18,  1.47it/s]
3|1078|Loss: 0.7479856014251709:  17%|█▋        | 1078/6470 [13:15<1:03:21,  1.42it/s]
3|1079|Loss: 0.8325858116149902:  17%|█▋        | 1078/6470 [13:15<1:03:21,  1.42it/s]
3|1079|Loss: 0.8325858116149902:  17%|█▋        | 1079/6470 [13:16<1:05:38,  1.37it/s]
3|1080|Loss: 0.8559045791625977:  17%|█▋        | 1079/6470 [13:16<1:05:38,  1.37it/s]
3|1080|Loss: 0.8559045791625977:  17%|█▋        | 1080/6470 [13:16<1:04:51,  1.39it/s]
3|1081|Loss: 0.9145089387893677:  17%|█▋        | 1080/6470 [13:17<1:04:51,  1.39it/s]
3|1081|Loss: 0.9145089387893677:  17%|█▋        | 1081/6470 [13:17<1:02:11,  1.44it/s]
3|1082|Loss: 0.751845121383667:  17%|█▋        | 1081/6470 [13:17<1:02:11,  1.44it/s] 
3|1082|Loss: 0.751845121383667:  17%|█▋        | 1082/6470 [13:18<1:02:53,  1.43it/s]
3|1083|Loss: 0.8089680075645447:  17%|█▋        | 1082/6470 [13:18<1:02:53,  1.43it/s]
3|1083|Loss: 0.8089680075645447:  17%|█▋        | 1083/6470 [13:19<1:06:22,  1.35it/s]
3|1084|Loss: 0.920112669467926:  17%|█▋        | 1083/6470 [13:19<1:06:22,  1.35it/s] 
3|1084|Loss: 0.920112669467926:  17%|█▋        | 1084/6470 [13:19<1:10:07,  1.28it/s]
3|1085|Loss: 0.7492257952690125:  17%|█▋        | 1084/6470 [13:20<1:10:07,  1.28it/s]
3|1085|Loss: 0.7492257952690125:  17%|█▋        | 1085/6470 [13:20<1:11:57,  1.25it/s]
3|1086|Loss: 0.7735306024551392:  17%|█▋        | 1085/6470 [13:21<1:11:57,  1.25it/s]
3|1086|Loss: 0.7735306024551392:  17%|█▋        | 1086/6470 [13:21<1:13:46,  1.22it/s]
3|1087|Loss: 1.0033670663833618:  17%|█▋        | 1086/6470 [13:22<1:13:46,  1.22it/s]
3|1087|Loss: 1.0033670663833618:  17%|█▋        | 1087/6470 [13:22<1:13:04,  1.23it/s]
3|1088|Loss: 0.7886185050010681:  17%|█▋        | 1087/6470 [13:22<1:13:04,  1.23it/s]
3|1088|Loss: 0.7886185050010681:  17%|█▋        | 1088/6470 [13:23<1:12:39,  1.23it/s]
3|1089|Loss: 0.8854203224182129:  17%|█▋        | 1088/6470 [13:23<1:12:39,  1.23it/s]
3|1089|Loss: 0.8854203224182129:  17%|█▋        | 1089/6470 [13:23<1:09:36,  1.29it/s]
3|1090|Loss: 0.7484601140022278:  17%|█▋        | 1089/6470 [13:24<1:09:36,  1.29it/s]
3|1090|Loss: 0.7484601140022278:  17%|█▋        | 1090/6470 [13:24<1:05:19,  1.37it/s]
3|1091|Loss: 0.8380776643753052:  17%|█▋        | 1090/6470 [13:24<1:05:19,  1.37it/s]
3|1091|Loss: 0.8380776643753052:  17%|█▋        | 1091/6470 [13:25<1:05:29,  1.37it/s]
3|1092|Loss: 0.6792633533477783:  17%|█▋        | 1091/6470 [13:25<1:05:29,  1.37it/s]
3|1092|Loss: 0.6792633533477783:  17%|█▋        | 1092/6470 [13:25<1:03:48,  1.40it/s]
3|1093|Loss: 0.6712853312492371:  17%|█▋        | 1092/6470 [13:26<1:03:48,  1.40it/s]
3|1093|Loss: 0.6712853312492371:  17%|█▋        | 1093/6470 [13:26<1:02:41,  1.43it/s]
3|1094|Loss: 0.8926720023155212:  17%|█▋        | 1093/6470 [13:27<1:02:41,  1.43it/s]
3|1094|Loss: 0.8926720023155212:  17%|█▋        | 1094/6470 [13:27<1:04:22,  1.39it/s]
3|1095|Loss: 0.859671950340271:  17%|█▋        | 1094/6470 [13:27<1:04:22,  1.39it/s] 
3|1095|Loss: 0.859671950340271:  17%|█▋        | 1095/6470 [13:28<1:06:13,  1.35it/s]
3|1096|Loss: 0.8453414440155029:  17%|█▋        | 1095/6470 [13:28<1:06:13,  1.35it/s]
3|1096|Loss: 0.8453414440155029:  17%|█▋        | 1096/6470 [13:28<58:56,  1.52it/s]  
3|1097|Loss: 0.6991985440254211:  17%|█▋        | 1096/6470 [13:29<58:56,  1.52it/s]
3|1097|Loss: 0.6991985440254211:  17%|█▋        | 1097/6470 [13:29<58:14,  1.54it/s]
3|1098|Loss: 0.7467369437217712:  17%|█▋        | 1097/6470 [13:29<58:14,  1.54it/s]
3|1098|Loss: 0.7467369437217712:  17%|█▋        | 1098/6470 [13:30<59:41,  1.50it/s]
3|1099|Loss: 0.6930050849914551:  17%|█▋        | 1098/6470 [13:30<59:41,  1.50it/s]
3|1099|Loss: 0.6930050849914551:  17%|█▋        | 1099/6470 [13:30<55:51,  1.60it/s]
3|1100|Loss: 0.9389072060585022:  17%|█▋        | 1099/6470 [13:30<55:51,  1.60it/s]
3|1100|Loss: 0.9389072060585022:  17%|█▋        | 1100/6470 [13:31<55:13,  1.62it/s]
3|1101|Loss: 0.8578773140907288:  17%|█▋        | 1100/6470 [13:31<55:13,  1.62it/s]
3|1101|Loss: 0.8578773140907288:  17%|█▋        | 1101/6470 [13:32<1:03:24,  1.41it/s]
3|1102|Loss: 0.7144231796264648:  17%|█▋        | 1101/6470 [13:32<1:03:24,  1.41it/s]
3|1102|Loss: 0.7144231796264648:  17%|█▋        | 1102/6470 [13:32<1:07:12,  1.33it/s]
3|1103|Loss: 0.8991110920906067:  17%|█▋        | 1102/6470 [13:33<1:07:12,  1.33it/s]
3|1103|Loss: 0.8991110920906067:  17%|█▋        | 1103/6470 [13:33<1:10:48,  1.26it/s]
3|1104|Loss: 0.5978184342384338:  17%|█▋        | 1103/6470 [13:34<1:10:48,  1.26it/s]
3|1104|Loss: 0.5978184342384338:  17%|█▋        | 1104/6470 [13:34<1:08:22,  1.31it/s]
3|1105|Loss: 0.9029942154884338:  17%|█▋        | 1104/6470 [13:34<1:08:22,  1.31it/s]
3|1105|Loss: 0.9029942154884338:  17%|█▋        | 1105/6470 [13:35<1:03:50,  1.40it/s]
3|1106|Loss: 0.7387540936470032:  17%|█▋        | 1105/6470 [13:35<1:03:50,  1.40it/s]
3|1106|Loss: 0.7387540936470032:  17%|█▋        | 1106/6470 [13:36<1:09:38,  1.28it/s]
3|1107|Loss: 0.8383766412734985:  17%|█▋        | 1106/6470 [13:36<1:09:38,  1.28it/s]
3|1107|Loss: 0.8383766412734985:  17%|█▋        | 1107/6470 [13:36<1:12:50,  1.23it/s]
3|1108|Loss: 0.7487480640411377:  17%|█▋        | 1107/6470 [13:37<1:12:50,  1.23it/s]
3|1108|Loss: 0.7487480640411377:  17%|█▋        | 1108/6470 [13:37<1:15:28,  1.18it/s]
3|1109|Loss: 0.7734604477882385:  17%|█▋        | 1108/6470 [13:38<1:15:28,  1.18it/s]
3|1109|Loss: 0.7734604477882385:  17%|█▋        | 1109/6470 [13:38<1:14:50,  1.19it/s]
3|1110|Loss: 0.8718511462211609:  17%|█▋        | 1109/6470 [13:39<1:14:50,  1.19it/s]
3|1110|Loss: 0.8718511462211609:  17%|█▋        | 1110/6470 [13:39<1:18:11,  1.14it/s]
3|1111|Loss: 0.8494997024536133:  17%|█▋        | 1110/6470 [13:40<1:18:11,  1.14it/s]
3|1111|Loss: 0.8494997024536133:  17%|█▋        | 1111/6470 [13:40<1:14:20,  1.20it/s]
3|1112|Loss: 0.8512448072433472:  17%|█▋        | 1111/6470 [13:40<1:14:20,  1.20it/s]
3|1112|Loss: 0.8512448072433472:  17%|█▋        | 1112/6470 [13:41<1:12:23,  1.23it/s]
3|1113|Loss: 0.8100953102111816:  17%|█▋        | 1112/6470 [13:41<1:12:23,  1.23it/s]
3|1113|Loss: 0.8100953102111816:  17%|█▋        | 1113/6470 [13:42<1:14:32,  1.20it/s]
3|1114|Loss: 0.9013880491256714:  17%|█▋        | 1113/6470 [13:42<1:14:32,  1.20it/s]
3|1114|Loss: 0.9013880491256714:  17%|█▋        | 1114/6470 [13:42<1:14:18,  1.20it/s]
3|1115|Loss: 0.8679657578468323:  17%|█▋        | 1114/6470 [13:43<1:14:18,  1.20it/s]
3|1115|Loss: 0.8679657578468323:  17%|█▋        | 1115/6470 [13:43<1:16:23,  1.17it/s]
3|1116|Loss: 0.843626856803894:  17%|█▋        | 1115/6470 [13:44<1:16:23,  1.17it/s] 
3|1116|Loss: 0.843626856803894:  17%|█▋        | 1116/6470 [13:44<1:13:47,  1.21it/s]
3|1117|Loss: 0.8161401748657227:  17%|█▋        | 1116/6470 [13:44<1:13:47,  1.21it/s]
3|1117|Loss: 0.8161401748657227:  17%|█▋        | 1117/6470 [13:45<1:12:49,  1.23it/s]
3|1118|Loss: 0.7831257581710815:  17%|█▋        | 1117/6470 [13:45<1:12:49,  1.23it/s]
3|1118|Loss: 0.7831257581710815:  17%|█▋        | 1118/6470 [13:45<1:09:39,  1.28it/s]
3|1119|Loss: 0.8634788393974304:  17%|█▋        | 1118/6470 [13:46<1:09:39,  1.28it/s]
3|1119|Loss: 0.8634788393974304:  17%|█▋        | 1119/6470 [13:46<1:04:44,  1.38it/s]
3|1120|Loss: 0.5416719317436218:  17%|█▋        | 1119/6470 [13:47<1:04:44,  1.38it/s]
3|1120|Loss: 0.5416719317436218:  17%|█▋        | 1120/6470 [13:47<1:13:14,  1.22it/s]
3|1121|Loss: 0.7921228408813477:  17%|█▋        | 1120/6470 [13:48<1:13:14,  1.22it/s]
3|1121|Loss: 0.7921228408813477:  17%|█▋        | 1121/6470 [13:48<1:10:48,  1.26it/s]
3|1122|Loss: 0.7941617965698242:  17%|█▋        | 1121/6470 [13:48<1:10:48,  1.26it/s]
3|1122|Loss: 0.7941617965698242:  17%|█▋        | 1122/6470 [13:49<1:10:00,  1.27it/s]
3|1123|Loss: 0.8858412504196167:  17%|█▋        | 1122/6470 [13:49<1:10:00,  1.27it/s]
3|1123|Loss: 0.8858412504196167:  17%|█▋        | 1123/6470 [13:49<1:04:49,  1.37it/s]
3|1124|Loss: 0.8005706667900085:  17%|█▋        | 1123/6470 [13:50<1:04:49,  1.37it/s]
3|1124|Loss: 0.8005706667900085:  17%|█▋        | 1124/6470 [13:50<1:00:37,  1.47it/s]
3|1125|Loss: 0.9186875224113464:  17%|█▋        | 1124/6470 [13:50<1:00:37,  1.47it/s]
3|1125|Loss: 0.9186875224113464:  17%|█▋        | 1125/6470 [13:51<1:02:31,  1.42it/s]
3|1126|Loss: 0.7740932703018188:  17%|█▋        | 1125/6470 [13:51<1:02:31,  1.42it/s]
3|1126|Loss: 0.7740932703018188:  17%|█▋        | 1126/6470 [13:51<1:04:26,  1.38it/s]
3|1127|Loss: 0.8672038316726685:  17%|█▋        | 1126/6470 [13:52<1:04:26,  1.38it/s]
3|1127|Loss: 0.8672038316726685:  17%|█▋        | 1127/6470 [13:52<1:04:00,  1.39it/s]
3|1128|Loss: 0.9526311755180359:  17%|█▋        | 1127/6470 [13:52<1:04:00,  1.39it/s]
3|1128|Loss: 0.9526311755180359:  17%|█▋        | 1128/6470 [13:53<1:04:45,  1.37it/s]
3|1129|Loss: 0.6796644926071167:  17%|█▋        | 1128/6470 [13:53<1:04:45,  1.37it/s]
3|1129|Loss: 0.6796644926071167:  17%|█▋        | 1129/6470 [13:53<1:02:36,  1.42it/s]
3|1130|Loss: 0.917094886302948:  17%|█▋        | 1129/6470 [13:54<1:02:36,  1.42it/s] 
3|1130|Loss: 0.917094886302948:  17%|█▋        | 1130/6470 [13:54<58:04,  1.53it/s]  
3|1131|Loss: 0.7190184593200684:  17%|█▋        | 1130/6470 [13:54<58:04,  1.53it/s]
3|1131|Loss: 0.7190184593200684:  17%|█▋        | 1131/6470 [13:55<1:00:00,  1.48it/s]
3|1132|Loss: 0.7095080614089966:  17%|█▋        | 1131/6470 [13:55<1:00:00,  1.48it/s]
3|1132|Loss: 0.7095080614089966:  17%|█▋        | 1132/6470 [13:56<1:06:07,  1.35it/s]
3|1133|Loss: 0.8715060353279114:  17%|█▋        | 1132/6470 [13:56<1:06:07,  1.35it/s]
3|1133|Loss: 0.8715060353279114:  18%|█▊        | 1133/6470 [13:56<1:07:40,  1.31it/s]
3|1134|Loss: 0.768921971321106:  18%|█▊        | 1133/6470 [13:57<1:07:40,  1.31it/s] 
3|1134|Loss: 0.768921971321106:  18%|█▊        | 1134/6470 [13:57<1:08:10,  1.30it/s]
3|1135|Loss: 0.7107815742492676:  18%|█▊        | 1134/6470 [13:58<1:08:10,  1.30it/s]
3|1135|Loss: 0.7107815742492676:  18%|█▊        | 1135/6470 [13:58<1:07:50,  1.31it/s]
3|1136|Loss: 0.7126443982124329:  18%|█▊        | 1135/6470 [13:58<1:07:50,  1.31it/s]
3|1136|Loss: 0.7126443982124329:  18%|█▊        | 1136/6470 [13:59<1:05:26,  1.36it/s]
3|1137|Loss: 0.816023051738739:  18%|█▊        | 1136/6470 [13:59<1:05:26,  1.36it/s] 
3|1137|Loss: 0.816023051738739:  18%|█▊        | 1137/6470 [13:59<1:04:09,  1.39it/s]
3|1138|Loss: 0.8824163675308228:  18%|█▊        | 1137/6470 [14:00<1:04:09,  1.39it/s]
3|1138|Loss: 0.8824163675308228:  18%|█▊        | 1138/6470 [14:00<1:08:49,  1.29it/s]
3|1139|Loss: 0.8873732089996338:  18%|█▊        | 1138/6470 [14:01<1:08:49,  1.29it/s]
3|1139|Loss: 0.8873732089996338:  18%|█▊        | 1139/6470 [14:01<1:08:37,  1.29it/s]
3|1140|Loss: 0.7754082083702087:  18%|█▊        | 1139/6470 [14:01<1:08:37,  1.29it/s]
3|1140|Loss: 0.7754082083702087:  18%|█▊        | 1140/6470 [14:02<1:06:53,  1.33it/s]
3|1141|Loss: 0.9251085519790649:  18%|█▊        | 1140/6470 [14:02<1:06:53,  1.33it/s]
3|1141|Loss: 0.9251085519790649:  18%|█▊        | 1141/6470 [14:02<1:06:42,  1.33it/s]
3|1142|Loss: 0.9068020582199097:  18%|█▊        | 1141/6470 [14:03<1:06:42,  1.33it/s]
3|1142|Loss: 0.9068020582199097:  18%|█▊        | 1142/6470 [14:03<1:07:31,  1.32it/s]
3|1143|Loss: 0.9032257795333862:  18%|█▊        | 1142/6470 [14:04<1:07:31,  1.32it/s]
3|1143|Loss: 0.9032257795333862:  18%|█▊        | 1143/6470 [14:04<1:05:48,  1.35it/s]
3|1144|Loss: 0.8778248429298401:  18%|█▊        | 1143/6470 [14:04<1:05:48,  1.35it/s]
3|1144|Loss: 0.8778248429298401:  18%|█▊        | 1144/6470 [14:05<1:07:43,  1.31it/s]
3|1145|Loss: 0.8062416911125183:  18%|█▊        | 1144/6470 [14:05<1:07:43,  1.31it/s]
3|1145|Loss: 0.8062416911125183:  18%|█▊        | 1145/6470 [14:05<1:05:33,  1.35it/s]
3|1146|Loss: 0.8049640655517578:  18%|█▊        | 1145/6470 [14:06<1:05:33,  1.35it/s]
3|1146|Loss: 0.8049640655517578:  18%|█▊        | 1146/6470 [14:06<1:03:46,  1.39it/s]
3|1147|Loss: 0.796181321144104:  18%|█▊        | 1146/6470 [14:06<1:03:46,  1.39it/s] 
3|1147|Loss: 0.796181321144104:  18%|█▊        | 1147/6470 [14:07<1:00:03,  1.48it/s]
3|1148|Loss: 0.7735137939453125:  18%|█▊        | 1147/6470 [14:07<1:00:03,  1.48it/s]
3|1148|Loss: 0.7735137939453125:  18%|█▊        | 1148/6470 [14:07<59:41,  1.49it/s]  
3|1149|Loss: 0.7736683487892151:  18%|█▊        | 1148/6470 [14:08<59:41,  1.49it/s]
3|1149|Loss: 0.7736683487892151:  18%|█▊        | 1149/6470 [14:08<1:02:39,  1.42it/s]
3|1150|Loss: 0.7782651782035828:  18%|█▊        | 1149/6470 [14:08<1:02:39,  1.42it/s]
3|1150|Loss: 0.7782651782035828:  18%|█▊        | 1150/6470 [14:09<58:36,  1.51it/s]  
3|1151|Loss: 0.9238020181655884:  18%|█▊        | 1150/6470 [14:09<58:36,  1.51it/s]
3|1151|Loss: 0.9238020181655884:  18%|█▊        | 1151/6470 [14:09<1:02:03,  1.43it/s]
3|1152|Loss: 0.7890526056289673:  18%|█▊        | 1151/6470 [14:10<1:02:03,  1.43it/s]
3|1152|Loss: 0.7890526056289673:  18%|█▊        | 1152/6470 [14:10<1:08:27,  1.29it/s]
3|1153|Loss: 0.7923957705497742:  18%|█▊        | 1152/6470 [14:11<1:08:27,  1.29it/s]
3|1153|Loss: 0.7923957705497742:  18%|█▊        | 1153/6470 [14:11<1:09:27,  1.28it/s]
3|1154|Loss: 0.914625346660614:  18%|█▊        | 1153/6470 [14:12<1:09:27,  1.28it/s] 
3|1154|Loss: 0.914625346660614:  18%|█▊        | 1154/6470 [14:12<1:02:41,  1.41it/s]
3|1155|Loss: 0.869299590587616:  18%|█▊        | 1154/6470 [14:12<1:02:41,  1.41it/s]
3|1155|Loss: 0.869299590587616:  18%|█▊        | 1155/6470 [14:12<57:48,  1.53it/s]  
3|1156|Loss: 1.0073513984680176:  18%|█▊        | 1155/6470 [14:13<57:48,  1.53it/s]
3|1156|Loss: 1.0073513984680176:  18%|█▊        | 1156/6470 [14:13<1:00:07,  1.47it/s]
3|1157|Loss: 0.7725317478179932:  18%|█▊        | 1156/6470 [14:13<1:00:07,  1.47it/s]
3|1157|Loss: 0.7725317478179932:  18%|█▊        | 1157/6470 [14:14<1:05:07,  1.36it/s]
3|1158|Loss: 0.7742440104484558:  18%|█▊        | 1157/6470 [14:14<1:05:07,  1.36it/s]
3|1158|Loss: 0.7742440104484558:  18%|█▊        | 1158/6470 [14:15<1:05:23,  1.35it/s]
3|1159|Loss: 0.8631294965744019:  18%|█▊        | 1158/6470 [14:15<1:05:23,  1.35it/s]
3|1159|Loss: 0.8631294965744019:  18%|█▊        | 1159/6470 [14:15<1:04:36,  1.37it/s]
3|1160|Loss: 0.8043578863143921:  18%|█▊        | 1159/6470 [14:16<1:04:36,  1.37it/s]
3|1160|Loss: 0.8043578863143921:  18%|█▊        | 1160/6470 [14:16<1:05:02,  1.36it/s]
3|1161|Loss: 1.0009464025497437:  18%|█▊        | 1160/6470 [14:16<1:05:02,  1.36it/s]
3|1161|Loss: 1.0009464025497437:  18%|█▊        | 1161/6470 [14:17<1:06:23,  1.33it/s]
3|1162|Loss: 0.8941227793693542:  18%|█▊        | 1161/6470 [14:17<1:06:23,  1.33it/s]
3|1162|Loss: 0.8941227793693542:  18%|█▊        | 1162/6470 [14:18<1:07:27,  1.31it/s]
3|1163|Loss: 0.6543981432914734:  18%|█▊        | 1162/6470 [14:18<1:07:27,  1.31it/s]
3|1163|Loss: 0.6543981432914734:  18%|█▊        | 1163/6470 [14:18<1:03:46,  1.39it/s]
3|1164|Loss: 0.8262261748313904:  18%|█▊        | 1163/6470 [14:19<1:03:46,  1.39it/s]
3|1164|Loss: 0.8262261748313904:  18%|█▊        | 1164/6470 [14:19<1:04:46,  1.37it/s]
3|1165|Loss: 0.795647919178009:  18%|█▊        | 1164/6470 [14:20<1:04:46,  1.37it/s] 
3|1165|Loss: 0.795647919178009:  18%|█▊        | 1165/6470 [14:20<1:10:33,  1.25it/s]
3|1166|Loss: 0.9005866050720215:  18%|█▊        | 1165/6470 [14:20<1:10:33,  1.25it/s]
3|1166|Loss: 0.9005866050720215:  18%|█▊        | 1166/6470 [14:21<1:11:40,  1.23it/s]
3|1167|Loss: 0.8139091730117798:  18%|█▊        | 1166/6470 [14:21<1:11:40,  1.23it/s]
3|1167|Loss: 0.8139091730117798:  18%|█▊        | 1167/6470 [14:22<1:13:37,  1.20it/s]
3|1168|Loss: 0.8143054842948914:  18%|█▊        | 1167/6470 [14:22<1:13:37,  1.20it/s]
3|1168|Loss: 0.8143054842948914:  18%|█▊        | 1168/6470 [14:23<1:21:58,  1.08it/s]
3|1169|Loss: 0.6971818804740906:  18%|█▊        | 1168/6470 [14:23<1:21:58,  1.08it/s]
3|1169|Loss: 0.6971818804740906:  18%|█▊        | 1169/6470 [14:24<1:16:44,  1.15it/s]
3|1170|Loss: 0.6942703127861023:  18%|█▊        | 1169/6470 [14:24<1:16:44,  1.15it/s]
3|1170|Loss: 0.6942703127861023:  18%|█▊        | 1170/6470 [14:24<1:10:48,  1.25it/s]
3|1171|Loss: 0.8149818181991577:  18%|█▊        | 1170/6470 [14:25<1:10:48,  1.25it/s]
3|1171|Loss: 0.8149818181991577:  18%|█▊        | 1171/6470 [14:25<1:07:36,  1.31it/s]
3|1172|Loss: 0.6809272170066833:  18%|█▊        | 1171/6470 [14:25<1:07:36,  1.31it/s]
3|1172|Loss: 0.6809272170066833:  18%|█▊        | 1172/6470 [14:26<1:10:52,  1.25it/s]
3|1173|Loss: 0.9239899516105652:  18%|█▊        | 1172/6470 [14:26<1:10:52,  1.25it/s]
3|1173|Loss: 0.9239899516105652:  18%|█▊        | 1173/6470 [14:27<1:13:48,  1.20it/s]
3|1174|Loss: 0.7942618131637573:  18%|█▊        | 1173/6470 [14:27<1:13:48,  1.20it/s]
3|1174|Loss: 0.7942618131637573:  18%|█▊        | 1174/6470 [14:27<1:07:53,  1.30it/s]
3|1175|Loss: 0.8517923951148987:  18%|█▊        | 1174/6470 [14:28<1:07:53,  1.30it/s]
3|1175|Loss: 0.8517923951148987:  18%|█▊        | 1175/6470 [14:28<58:54,  1.50it/s]  
3|1176|Loss: 0.682306170463562:  18%|█▊        | 1175/6470 [14:28<58:54,  1.50it/s] 
3|1176|Loss: 0.682306170463562:  18%|█▊        | 1176/6470 [14:28<51:12,  1.72it/s]
3|1177|Loss: 0.8704210519790649:  18%|█▊        | 1176/6470 [14:28<51:12,  1.72it/s]
3|1177|Loss: 0.8704210519790649:  18%|█▊        | 1177/6470 [14:29<53:49,  1.64it/s]
3|1178|Loss: 0.7263810038566589:  18%|█▊        | 1177/6470 [14:29<53:49,  1.64it/s]
3|1178|Loss: 0.7263810038566589:  18%|█▊        | 1178/6470 [14:30<59:10,  1.49it/s]
3|1179|Loss: 0.8000562191009521:  18%|█▊        | 1178/6470 [14:30<59:10,  1.49it/s]
3|1179|Loss: 0.8000562191009521:  18%|█▊        | 1179/6470 [14:30<1:00:14,  1.46it/s]
3|1180|Loss: 0.9134176969528198:  18%|█▊        | 1179/6470 [14:31<1:00:14,  1.46it/s]
3|1180|Loss: 0.9134176969528198:  18%|█▊        | 1180/6470 [14:31<1:01:54,  1.42it/s]
3|1181|Loss: 0.9450260400772095:  18%|█▊        | 1180/6470 [14:32<1:01:54,  1.42it/s]
3|1181|Loss: 0.9450260400772095:  18%|█▊        | 1181/6470 [14:32<1:06:17,  1.33it/s]
3|1182|Loss: 0.8623372316360474:  18%|█▊        | 1181/6470 [14:32<1:06:17,  1.33it/s]
3|1182|Loss: 0.8623372316360474:  18%|█▊        | 1182/6470 [14:33<1:03:40,  1.38it/s]
3|1183|Loss: 0.8767070174217224:  18%|█▊        | 1182/6470 [14:33<1:03:40,  1.38it/s]
3|1183|Loss: 0.8767070174217224:  18%|█▊        | 1183/6470 [14:33<1:06:47,  1.32it/s]
3|1184|Loss: 0.8378080129623413:  18%|█▊        | 1183/6470 [14:34<1:06:47,  1.32it/s]
3|1184|Loss: 0.8378080129623413:  18%|█▊        | 1184/6470 [14:34<1:01:38,  1.43it/s]
3|1185|Loss: 0.8715730905532837:  18%|█▊        | 1184/6470 [14:34<1:01:38,  1.43it/s]
3|1185|Loss: 0.8715730905532837:  18%|█▊        | 1185/6470 [14:35<1:01:54,  1.42it/s]
3|1186|Loss: 0.969455361366272:  18%|█▊        | 1185/6470 [14:35<1:01:54,  1.42it/s] 
3|1186|Loss: 0.969455361366272:  18%|█▊        | 1186/6470 [14:35<1:01:40,  1.43it/s]
3|1187|Loss: 0.8241551518440247:  18%|█▊        | 1186/6470 [14:36<1:01:40,  1.43it/s]
3|1187|Loss: 0.8241551518440247:  18%|█▊        | 1187/6470 [14:36<57:28,  1.53it/s]  
3|1188|Loss: 0.8998314738273621:  18%|█▊        | 1187/6470 [14:36<57:28,  1.53it/s]
3|1188|Loss: 0.8998314738273621:  18%|█▊        | 1188/6470 [14:37<59:28,  1.48it/s]
3|1189|Loss: 0.9046575427055359:  18%|█▊        | 1188/6470 [14:37<59:28,  1.48it/s]
3|1189|Loss: 0.9046575427055359:  18%|█▊        | 1189/6470 [14:37<57:37,  1.53it/s]
3|1190|Loss: 0.8325437307357788:  18%|█▊        | 1189/6470 [14:38<57:37,  1.53it/s]
3|1190|Loss: 0.8325437307357788:  18%|█▊        | 1190/6470 [14:38<56:22,  1.56it/s]
3|1191|Loss: 0.7278326749801636:  18%|█▊        | 1190/6470 [14:38<56:22,  1.56it/s]
3|1191|Loss: 0.7278326749801636:  18%|█▊        | 1191/6470 [14:39<56:48,  1.55it/s]
3|1192|Loss: 0.8638728857040405:  18%|█▊        | 1191/6470 [14:39<56:48,  1.55it/s]
3|1192|Loss: 0.8638728857040405:  18%|█▊        | 1192/6470 [14:39<1:03:19,  1.39it/s]
3|1193|Loss: 1.0310757160186768:  18%|█▊        | 1192/6470 [14:40<1:03:19,  1.39it/s]
3|1193|Loss: 1.0310757160186768:  18%|█▊        | 1193/6470 [14:40<1:01:47,  1.42it/s]
3|1194|Loss: 0.8004133701324463:  18%|█▊        | 1193/6470 [14:40<1:01:47,  1.42it/s]
3|1194|Loss: 0.8004133701324463:  18%|█▊        | 1194/6470 [14:41<59:13,  1.48it/s]  
3|1195|Loss: 0.7760388851165771:  18%|█▊        | 1194/6470 [14:41<59:13,  1.48it/s]
3|1195|Loss: 0.7760388851165771:  18%|█▊        | 1195/6470 [14:42<1:03:41,  1.38it/s]
3|1196|Loss: 0.8061073422431946:  18%|█▊        | 1195/6470 [14:42<1:03:41,  1.38it/s]
3|1196|Loss: 0.8061073422431946:  18%|█▊        | 1196/6470 [14:42<1:07:29,  1.30it/s]
3|1197|Loss: 0.8922989964485168:  18%|█▊        | 1196/6470 [14:43<1:07:29,  1.30it/s]
3|1197|Loss: 0.8922989964485168:  19%|█▊        | 1197/6470 [14:43<1:07:30,  1.30it/s]
3|1198|Loss: 0.5814612507820129:  19%|█▊        | 1197/6470 [14:44<1:07:30,  1.30it/s]
3|1198|Loss: 0.5814612507820129:  19%|█▊        | 1198/6470 [14:44<1:03:48,  1.38it/s]
3|1199|Loss: 0.7627390623092651:  19%|█▊        | 1198/6470 [14:44<1:03:48,  1.38it/s]
3|1199|Loss: 0.7627390623092651:  19%|█▊        | 1199/6470 [14:45<1:06:06,  1.33it/s]
3|1200|Loss: 0.9755182266235352:  19%|█▊        | 1199/6470 [14:45<1:06:06,  1.33it/s]
3|1200|Loss: 0.9755182266235352:  19%|█▊        | 1200/6470 [14:45<1:07:10,  1.31it/s]
3|1201|Loss: 0.8139179348945618:  19%|█▊        | 1200/6470 [14:46<1:07:10,  1.31it/s]
3|1201|Loss: 0.8139179348945618:  19%|█▊        | 1201/6470 [14:46<1:05:55,  1.33it/s]
3|1202|Loss: 0.8363745808601379:  19%|█▊        | 1201/6470 [14:47<1:05:55,  1.33it/s]
3|1202|Loss: 0.8363745808601379:  19%|█▊        | 1202/6470 [14:47<1:12:35,  1.21it/s]
3|1203|Loss: 0.8721464276313782:  19%|█▊        | 1202/6470 [14:48<1:12:35,  1.21it/s]
3|1203|Loss: 0.8721464276313782:  19%|█▊        | 1203/6470 [14:48<1:11:06,  1.23it/s]
3|1204|Loss: 0.9767608642578125:  19%|█▊        | 1203/6470 [14:48<1:11:06,  1.23it/s]
3|1204|Loss: 0.9767608642578125:  19%|█▊        | 1204/6470 [14:49<1:11:46,  1.22it/s]
3|1205|Loss: 0.8903308510780334:  19%|█▊        | 1204/6470 [14:49<1:11:46,  1.22it/s]
3|1205|Loss: 0.8903308510780334:  19%|█▊        | 1205/6470 [14:49<1:09:01,  1.27it/s]
3|1206|Loss: 0.8157708048820496:  19%|█▊        | 1205/6470 [14:50<1:09:01,  1.27it/s]
3|1206|Loss: 0.8157708048820496:  19%|█▊        | 1206/6470 [14:50<1:07:36,  1.30it/s]
3|1207|Loss: 0.9786526560783386:  19%|█▊        | 1206/6470 [14:51<1:07:36,  1.30it/s]
3|1207|Loss: 0.9786526560783386:  19%|█▊        | 1207/6470 [14:51<1:09:13,  1.27it/s]
3|1208|Loss: 0.8309721350669861:  19%|█▊        | 1207/6470 [14:51<1:09:13,  1.27it/s]
3|1208|Loss: 0.8309721350669861:  19%|█▊        | 1208/6470 [14:52<1:09:27,  1.26it/s]
3|1209|Loss: 0.8093970417976379:  19%|█▊        | 1208/6470 [14:52<1:09:27,  1.26it/s]
3|1209|Loss: 0.8093970417976379:  19%|█▊        | 1209/6470 [14:52<1:04:23,  1.36it/s]
3|1210|Loss: 0.7859766483306885:  19%|█▊        | 1209/6470 [14:53<1:04:23,  1.36it/s]
3|1210|Loss: 0.7859766483306885:  19%|█▊        | 1210/6470 [14:53<1:00:48,  1.44it/s]
3|1211|Loss: 0.9613374471664429:  19%|█▊        | 1210/6470 [14:53<1:00:48,  1.44it/s]
3|1211|Loss: 0.9613374471664429:  19%|█▊        | 1211/6470 [14:54<55:59,  1.57it/s]  
3|1212|Loss: 0.7561783790588379:  19%|█▊        | 1211/6470 [14:54<55:59,  1.57it/s]
3|1212|Loss: 0.7561783790588379:  19%|█▊        | 1212/6470 [14:54<53:24,  1.64it/s]
3|1213|Loss: 0.8416520357131958:  19%|█▊        | 1212/6470 [14:54<53:24,  1.64it/s]
3|1213|Loss: 0.8416520357131958:  19%|█▊        | 1213/6470 [14:55<56:22,  1.55it/s]
3|1214|Loss: 0.8533506989479065:  19%|█▊        | 1213/6470 [14:55<56:22,  1.55it/s]
3|1214|Loss: 0.8533506989479065:  19%|█▉        | 1214/6470 [14:56<1:00:04,  1.46it/s]
3|1215|Loss: 0.8558699488639832:  19%|█▉        | 1214/6470 [14:56<1:00:04,  1.46it/s]
3|1215|Loss: 0.8558699488639832:  19%|█▉        | 1215/6470 [14:56<1:02:09,  1.41it/s]
3|1216|Loss: 0.7022624611854553:  19%|█▉        | 1215/6470 [14:57<1:02:09,  1.41it/s]
3|1216|Loss: 0.7022624611854553:  19%|█▉        | 1216/6470 [14:57<57:32,  1.52it/s]  
3|1217|Loss: 0.912911593914032:  19%|█▉        | 1216/6470 [14:57<57:32,  1.52it/s] 
3|1217|Loss: 0.912911593914032:  19%|█▉        | 1217/6470 [14:58<58:21,  1.50it/s]
3|1218|Loss: 0.8386366367340088:  19%|█▉        | 1217/6470 [14:58<58:21,  1.50it/s]
3|1218|Loss: 0.8386366367340088:  19%|█▉        | 1218/6470 [14:58<59:43,  1.47it/s]
3|1219|Loss: 1.0459251403808594:  19%|█▉        | 1218/6470 [14:59<59:43,  1.47it/s]
3|1219|Loss: 1.0459251403808594:  19%|█▉        | 1219/6470 [14:59<1:03:40,  1.37it/s]
3|1220|Loss: 0.902660608291626:  19%|█▉        | 1219/6470 [15:00<1:03:40,  1.37it/s] 
3|1220|Loss: 0.902660608291626:  19%|█▉        | 1220/6470 [15:00<1:05:44,  1.33it/s]
3|1221|Loss: 0.8703330755233765:  19%|█▉        | 1220/6470 [15:00<1:05:44,  1.33it/s]
3|1221|Loss: 0.8703330755233765:  19%|█▉        | 1221/6470 [15:01<1:03:18,  1.38it/s]
3|1222|Loss: 0.8384213447570801:  19%|█▉        | 1221/6470 [15:01<1:03:18,  1.38it/s]
3|1222|Loss: 0.8384213447570801:  19%|█▉        | 1222/6470 [15:01<1:04:28,  1.36it/s]
3|1223|Loss: 0.8215820789337158:  19%|█▉        | 1222/6470 [15:02<1:04:28,  1.36it/s]
3|1223|Loss: 0.8215820789337158:  19%|█▉        | 1223/6470 [15:02<1:06:23,  1.32it/s]
3|1224|Loss: 0.8025344610214233:  19%|█▉        | 1223/6470 [15:03<1:06:23,  1.32it/s]
3|1224|Loss: 0.8025344610214233:  19%|█▉        | 1224/6470 [15:03<1:04:40,  1.35it/s]
3|1225|Loss: 1.0188277959823608:  19%|█▉        | 1224/6470 [15:03<1:04:40,  1.35it/s]
3|1225|Loss: 1.0188277959823608:  19%|█▉        | 1225/6470 [15:04<1:02:45,  1.39it/s]
3|1226|Loss: 0.8875295519828796:  19%|█▉        | 1225/6470 [15:04<1:02:45,  1.39it/s]
3|1226|Loss: 0.8875295519828796:  19%|█▉        | 1226/6470 [15:04<1:00:36,  1.44it/s]
3|1227|Loss: 0.8673709630966187:  19%|█▉        | 1226/6470 [15:05<1:00:36,  1.44it/s]
3|1227|Loss: 0.8673709630966187:  19%|█▉        | 1227/6470 [15:05<1:03:53,  1.37it/s]
3|1228|Loss: 0.831122100353241:  19%|█▉        | 1227/6470 [15:05<1:03:53,  1.37it/s] 
3|1228|Loss: 0.831122100353241:  19%|█▉        | 1228/6470 [15:06<1:04:01,  1.36it/s]
3|1229|Loss: 0.9436382055282593:  19%|█▉        | 1228/6470 [15:06<1:04:01,  1.36it/s]
3|1229|Loss: 0.9436382055282593:  19%|█▉        | 1229/6470 [15:06<59:43,  1.46it/s]  
3|1230|Loss: 0.804298460483551:  19%|█▉        | 1229/6470 [15:07<59:43,  1.46it/s] 
3|1230|Loss: 0.804298460483551:  19%|█▉        | 1230/6470 [15:07<1:00:27,  1.44it/s]
3|1231|Loss: 0.8098832964897156:  19%|█▉        | 1230/6470 [15:07<1:00:27,  1.44it/s]
3|1231|Loss: 0.8098832964897156:  19%|█▉        | 1231/6470 [15:08<1:01:17,  1.42it/s]
3|1232|Loss: 0.8861834406852722:  19%|█▉        | 1231/6470 [15:08<1:01:17,  1.42it/s]
3|1232|Loss: 0.8861834406852722:  19%|█▉        | 1232/6470 [15:09<1:06:05,  1.32it/s]
3|1233|Loss: 0.8674976825714111:  19%|█▉        | 1232/6470 [15:09<1:06:05,  1.32it/s]
3|1233|Loss: 0.8674976825714111:  19%|█▉        | 1233/6470 [15:09<1:07:59,  1.28it/s]
3|1234|Loss: 0.8788779973983765:  19%|█▉        | 1233/6470 [15:10<1:07:59,  1.28it/s]
3|1234|Loss: 0.8788779973983765:  19%|█▉        | 1234/6470 [15:10<1:08:55,  1.27it/s]
3|1235|Loss: 0.8744837045669556:  19%|█▉        | 1234/6470 [15:11<1:08:55,  1.27it/s]
3|1235|Loss: 0.8744837045669556:  19%|█▉        | 1235/6470 [15:11<1:08:17,  1.28it/s]
3|1236|Loss: 0.8796557784080505:  19%|█▉        | 1235/6470 [15:11<1:08:17,  1.28it/s]
3|1236|Loss: 0.8796557784080505:  19%|█▉        | 1236/6470 [15:12<1:07:27,  1.29it/s]
3|1237|Loss: 0.7216732501983643:  19%|█▉        | 1236/6470 [15:12<1:07:27,  1.29it/s]
3|1237|Loss: 0.7216732501983643:  19%|█▉        | 1237/6470 [15:12<1:02:26,  1.40it/s]
3|1238|Loss: 0.9696887731552124:  19%|█▉        | 1237/6470 [15:13<1:02:26,  1.40it/s]
3|1238|Loss: 0.9696887731552124:  19%|█▉        | 1238/6470 [15:13<1:06:35,  1.31it/s]
3|1239|Loss: 0.803318440914154:  19%|█▉        | 1238/6470 [15:14<1:06:35,  1.31it/s] 
3|1239|Loss: 0.803318440914154:  19%|█▉        | 1239/6470 [15:14<1:09:50,  1.25it/s]
3|1240|Loss: 0.7887257933616638:  19%|█▉        | 1239/6470 [15:15<1:09:50,  1.25it/s]
3|1240|Loss: 0.7887257933616638:  19%|█▉        | 1240/6470 [15:15<1:09:55,  1.25it/s]
3|1241|Loss: 0.9623315334320068:  19%|█▉        | 1240/6470 [15:15<1:09:55,  1.25it/s]
3|1241|Loss: 0.9623315334320068:  19%|█▉        | 1241/6470 [15:16<1:13:57,  1.18it/s]
3|1242|Loss: 0.9805064797401428:  19%|█▉        | 1241/6470 [15:16<1:13:57,  1.18it/s]
3|1242|Loss: 0.9805064797401428:  19%|█▉        | 1242/6470 [15:17<1:13:03,  1.19it/s]
3|1243|Loss: 0.7269586324691772:  19%|█▉        | 1242/6470 [15:17<1:13:03,  1.19it/s]
3|1243|Loss: 0.7269586324691772:  19%|█▉        | 1243/6470 [15:17<1:07:21,  1.29it/s]
3|1244|Loss: 0.9123433828353882:  19%|█▉        | 1243/6470 [15:18<1:07:21,  1.29it/s]
3|1244|Loss: 0.9123433828353882:  19%|█▉        | 1244/6470 [15:18<1:08:27,  1.27it/s]
3|1245|Loss: 0.8454145789146423:  19%|█▉        | 1244/6470 [15:19<1:08:27,  1.27it/s]
3|1245|Loss: 0.8454145789146423:  19%|█▉        | 1245/6470 [15:19<1:09:16,  1.26it/s]
3|1246|Loss: 0.6644831895828247:  19%|█▉        | 1245/6470 [15:19<1:09:16,  1.26it/s]
3|1246|Loss: 0.6644831895828247:  19%|█▉        | 1246/6470 [15:20<1:10:52,  1.23it/s]
3|1247|Loss: 0.7197914123535156:  19%|█▉        | 1246/6470 [15:20<1:10:52,  1.23it/s]
3|1247|Loss: 0.7197914123535156:  19%|█▉        | 1247/6470 [15:20<1:06:55,  1.30it/s]
3|1248|Loss: 0.7499443292617798:  19%|█▉        | 1247/6470 [15:21<1:06:55,  1.30it/s]
3|1248|Loss: 0.7499443292617798:  19%|█▉        | 1248/6470 [15:21<1:06:11,  1.31it/s]
3|1249|Loss: 0.9984337687492371:  19%|█▉        | 1248/6470 [15:22<1:06:11,  1.31it/s]
3|1249|Loss: 0.9984337687492371:  19%|█▉        | 1249/6470 [15:22<1:07:03,  1.30it/s]
3|1250|Loss: 0.6872016191482544:  19%|█▉        | 1249/6470 [15:22<1:07:03,  1.30it/s]
3|1250|Loss: 0.6872016191482544:  19%|█▉        | 1250/6470 [15:22<59:00,  1.47it/s]  
3|1251|Loss: 0.8027195334434509:  19%|█▉        | 1250/6470 [15:23<59:00,  1.47it/s]
3|1251|Loss: 0.8027195334434509:  19%|█▉        | 1251/6470 [15:23<56:04,  1.55it/s]
3|1252|Loss: 0.7984586954116821:  19%|█▉        | 1251/6470 [15:23<56:04,  1.55it/s]
3|1252|Loss: 0.7984586954116821:  19%|█▉        | 1252/6470 [15:24<59:14,  1.47it/s]
3|1253|Loss: 0.9786131978034973:  19%|█▉        | 1252/6470 [15:24<59:14,  1.47it/s]
3|1253|Loss: 0.9786131978034973:  19%|█▉        | 1253/6470 [15:24<57:24,  1.51it/s]
3|1254|Loss: 0.8020504117012024:  19%|█▉        | 1253/6470 [15:25<57:24,  1.51it/s]
3|1254|Loss: 0.8020504117012024:  19%|█▉        | 1254/6470 [15:25<57:37,  1.51it/s]
3|1255|Loss: 0.9447951316833496:  19%|█▉        | 1254/6470 [15:26<57:37,  1.51it/s]
3|1255|Loss: 0.9447951316833496:  19%|█▉        | 1255/6470 [15:26<1:02:02,  1.40it/s]
3|1256|Loss: 0.8122959136962891:  19%|█▉        | 1255/6470 [15:26<1:02:02,  1.40it/s]
3|1256|Loss: 0.8122959136962891:  19%|█▉        | 1256/6470 [15:27<1:03:33,  1.37it/s]
3|1257|Loss: 0.7291423678398132:  19%|█▉        | 1256/6470 [15:27<1:03:33,  1.37it/s]
3|1257|Loss: 0.7291423678398132:  19%|█▉        | 1257/6470 [15:27<1:04:14,  1.35it/s]
3|1258|Loss: 0.7516659498214722:  19%|█▉        | 1257/6470 [15:28<1:04:14,  1.35it/s]
3|1258|Loss: 0.7516659498214722:  19%|█▉        | 1258/6470 [15:28<1:08:01,  1.28it/s]
3|1259|Loss: 0.8381649255752563:  19%|█▉        | 1258/6470 [15:29<1:08:01,  1.28it/s]
3|1259|Loss: 0.8381649255752563:  19%|█▉        | 1259/6470 [15:29<1:08:58,  1.26it/s]
3|1260|Loss: 0.7025561332702637:  19%|█▉        | 1259/6470 [15:30<1:08:58,  1.26it/s]
3|1260|Loss: 0.7025561332702637:  19%|█▉        | 1260/6470 [15:30<1:08:30,  1.27it/s]
3|1261|Loss: 0.9027179479598999:  19%|█▉        | 1260/6470 [15:30<1:08:30,  1.27it/s]
3|1261|Loss: 0.9027179479598999:  19%|█▉        | 1261/6470 [15:31<1:04:39,  1.34it/s]
3|1262|Loss: 0.9339383840560913:  19%|█▉        | 1261/6470 [15:31<1:04:39,  1.34it/s]
3|1262|Loss: 0.9339383840560913:  20%|█▉        | 1262/6470 [15:31<1:00:32,  1.43it/s]
3|1263|Loss: 0.8331325054168701:  20%|█▉        | 1262/6470 [15:32<1:00:32,  1.43it/s]
3|1263|Loss: 0.8331325054168701:  20%|█▉        | 1263/6470 [15:32<1:01:06,  1.42it/s]
3|1264|Loss: 0.8180175423622131:  20%|█▉        | 1263/6470 [15:32<1:01:06,  1.42it/s]
3|1264|Loss: 0.8180175423622131:  20%|█▉        | 1264/6470 [15:33<1:01:10,  1.42it/s]
3|1265|Loss: 0.8620621562004089:  20%|█▉        | 1264/6470 [15:33<1:01:10,  1.42it/s]
3|1265|Loss: 0.8620621562004089:  20%|█▉        | 1265/6470 [15:33<1:04:18,  1.35it/s]
3|1266|Loss: 1.0840635299682617:  20%|█▉        | 1265/6470 [15:34<1:04:18,  1.35it/s]
3|1266|Loss: 1.0840635299682617:  20%|█▉        | 1266/6470 [15:34<1:11:39,  1.21it/s]
3|1267|Loss: 0.8346231579780579:  20%|█▉        | 1266/6470 [15:35<1:11:39,  1.21it/s]
3|1267|Loss: 0.8346231579780579:  20%|█▉        | 1267/6470 [15:35<1:11:43,  1.21it/s]
3|1268|Loss: 0.858000636100769:  20%|█▉        | 1267/6470 [15:36<1:11:43,  1.21it/s] 
3|1268|Loss: 0.858000636100769:  20%|█▉        | 1268/6470 [15:36<1:05:29,  1.32it/s]
3|1269|Loss: 1.082801342010498:  20%|█▉        | 1268/6470 [15:36<1:05:29,  1.32it/s]
3|1269|Loss: 1.082801342010498:  20%|█▉        | 1269/6470 [15:36<1:02:01,  1.40it/s]
3|1270|Loss: 0.847852885723114:  20%|█▉        | 1269/6470 [15:37<1:02:01,  1.40it/s]
3|1270|Loss: 0.847852885723114:  20%|█▉        | 1270/6470 [15:37<1:03:50,  1.36it/s]
3|1271|Loss: 0.948797345161438:  20%|█▉        | 1270/6470 [15:38<1:03:50,  1.36it/s]
3|1271|Loss: 0.948797345161438:  20%|█▉        | 1271/6470 [15:38<1:03:02,  1.37it/s]
3|1272|Loss: 0.845058023929596:  20%|█▉        | 1271/6470 [15:38<1:03:02,  1.37it/s]
3|1272|Loss: 0.845058023929596:  20%|█▉        | 1272/6470 [15:39<1:05:48,  1.32it/s]
3|1273|Loss: 0.75115567445755:  20%|█▉        | 1272/6470 [15:39<1:05:48,  1.32it/s] 
3|1273|Loss: 0.75115567445755:  20%|█▉        | 1273/6470 [15:39<1:00:32,  1.43it/s]
3|1274|Loss: 0.7354809641838074:  20%|█▉        | 1273/6470 [15:40<1:00:32,  1.43it/s]
3|1274|Loss: 0.7354809641838074:  20%|█▉        | 1274/6470 [15:40<59:13,  1.46it/s]  
3|1275|Loss: 0.9665576815605164:  20%|█▉        | 1274/6470 [15:40<59:13,  1.46it/s]
3|1275|Loss: 0.9665576815605164:  20%|█▉        | 1275/6470 [15:41<58:42,  1.47it/s]
3|1276|Loss: 0.8287748694419861:  20%|█▉        | 1275/6470 [15:41<58:42,  1.47it/s]
3|1276|Loss: 0.8287748694419861:  20%|█▉        | 1276/6470 [15:41<1:00:41,  1.43it/s]
3|1277|Loss: 0.8409292101860046:  20%|█▉        | 1276/6470 [15:42<1:00:41,  1.43it/s]
3|1277|Loss: 0.8409292101860046:  20%|█▉        | 1277/6470 [15:42<1:06:24,  1.30it/s]
3|1278|Loss: 1.0686378479003906:  20%|█▉        | 1277/6470 [15:43<1:06:24,  1.30it/s]
3|1278|Loss: 1.0686378479003906:  20%|█▉        | 1278/6470 [15:43<1:10:59,  1.22it/s]
3|1279|Loss: 0.799214780330658:  20%|█▉        | 1278/6470 [15:44<1:10:59,  1.22it/s] 
3|1279|Loss: 0.799214780330658:  20%|█▉        | 1279/6470 [15:44<1:05:54,  1.31it/s]
3|1280|Loss: 0.8920673131942749:  20%|█▉        | 1279/6470 [15:44<1:05:54,  1.31it/s]
3|1280|Loss: 0.8920673131942749:  20%|█▉        | 1280/6470 [15:45<1:01:38,  1.40it/s]
3|1281|Loss: 0.7758079171180725:  20%|█▉        | 1280/6470 [15:45<1:01:38,  1.40it/s]
3|1281|Loss: 0.7758079171180725:  20%|█▉        | 1281/6470 [15:45<1:02:17,  1.39it/s]
3|1282|Loss: 0.8613559603691101:  20%|█▉        | 1281/6470 [15:46<1:02:17,  1.39it/s]
3|1282|Loss: 0.8613559603691101:  20%|█▉        | 1282/6470 [15:46<1:01:16,  1.41it/s]
3|1283|Loss: 0.9178590178489685:  20%|█▉        | 1282/6470 [15:46<1:01:16,  1.41it/s]
3|1283|Loss: 0.9178590178489685:  20%|█▉        | 1283/6470 [15:47<59:55,  1.44it/s]  
3|1284|Loss: 0.8263999223709106:  20%|█▉        | 1283/6470 [15:47<59:55,  1.44it/s]
3|1284|Loss: 0.8263999223709106:  20%|█▉        | 1284/6470 [15:47<1:01:40,  1.40it/s]
3|1285|Loss: 0.9044831991195679:  20%|█▉        | 1284/6470 [15:48<1:01:40,  1.40it/s]
3|1285|Loss: 0.9044831991195679:  20%|█▉        | 1285/6470 [15:48<1:05:26,  1.32it/s]
3|1286|Loss: 0.7922675609588623:  20%|█▉        | 1285/6470 [15:49<1:05:26,  1.32it/s]
3|1286|Loss: 0.7922675609588623:  20%|█▉        | 1286/6470 [15:49<1:04:39,  1.34it/s]
3|1287|Loss: 0.7928929924964905:  20%|█▉        | 1286/6470 [15:49<1:04:39,  1.34it/s]
3|1287|Loss: 0.7928929924964905:  20%|█▉        | 1287/6470 [15:50<1:04:43,  1.33it/s]
3|1288|Loss: 0.7668675184249878:  20%|█▉        | 1287/6470 [15:50<1:04:43,  1.33it/s]
3|1288|Loss: 0.7668675184249878:  20%|█▉        | 1288/6470 [15:51<1:09:19,  1.25it/s]
3|1289|Loss: 0.7552469372749329:  20%|█▉        | 1288/6470 [15:51<1:09:19,  1.25it/s]
3|1289|Loss: 0.7552469372749329:  20%|█▉        | 1289/6470 [15:51<1:07:14,  1.28it/s]
3|1290|Loss: 0.910801112651825:  20%|█▉        | 1289/6470 [15:52<1:07:14,  1.28it/s] 
3|1290|Loss: 0.910801112651825:  20%|█▉        | 1290/6470 [15:52<1:03:40,  1.36it/s]
3|1291|Loss: 0.743188738822937:  20%|█▉        | 1290/6470 [15:52<1:03:40,  1.36it/s]
3|1291|Loss: 0.743188738822937:  20%|█▉        | 1291/6470 [15:53<1:01:27,  1.40it/s]
3|1292|Loss: 0.8131086230278015:  20%|█▉        | 1291/6470 [15:53<1:01:27,  1.40it/s]
3|1292|Loss: 0.8131086230278015:  20%|█▉        | 1292/6470 [15:54<1:06:14,  1.30it/s]
3|1293|Loss: 0.8466991782188416:  20%|█▉        | 1292/6470 [15:54<1:06:14,  1.30it/s]
3|1293|Loss: 0.8466991782188416:  20%|█▉        | 1293/6470 [15:54<1:05:57,  1.31it/s]
3|1294|Loss: 0.9527953863143921:  20%|█▉        | 1293/6470 [15:55<1:05:57,  1.31it/s]
3|1294|Loss: 0.9527953863143921:  20%|██        | 1294/6470 [15:55<1:06:46,  1.29it/s]
3|1295|Loss: 0.697005033493042:  20%|██        | 1294/6470 [15:55<1:06:46,  1.29it/s] 
3|1295|Loss: 0.697005033493042:  20%|██        | 1295/6470 [15:56<1:03:48,  1.35it/s]
3|1296|Loss: 0.8853352665901184:  20%|██        | 1295/6470 [15:56<1:03:48,  1.35it/s]
3|1296|Loss: 0.8853352665901184:  20%|██        | 1296/6470 [15:56<1:04:20,  1.34it/s]
3|1297|Loss: 0.9818562269210815:  20%|██        | 1296/6470 [15:57<1:04:20,  1.34it/s]
3|1297|Loss: 0.9818562269210815:  20%|██        | 1297/6470 [15:57<59:22,  1.45it/s]  
3|1298|Loss: 0.9203722476959229:  20%|██        | 1297/6470 [15:57<59:22,  1.45it/s]
3|1298|Loss: 0.9203722476959229:  20%|██        | 1298/6470 [15:58<59:58,  1.44it/s]
3|1299|Loss: 0.7044078707695007:  20%|██        | 1298/6470 [15:58<59:58,  1.44it/s]
3|1299|Loss: 0.7044078707695007:  20%|██        | 1299/6470 [15:59<1:01:39,  1.40it/s]
3|1300|Loss: 0.8896253108978271:  20%|██        | 1299/6470 [15:59<1:01:39,  1.40it/s]
3|1300|Loss: 0.8896253108978271:  20%|██        | 1300/6470 [15:59<1:07:28,  1.28it/s]
3|1301|Loss: 0.726739227771759:  20%|██        | 1300/6470 [16:00<1:07:28,  1.28it/s] 
3|1301|Loss: 0.726739227771759:  20%|██        | 1301/6470 [16:00<1:02:15,  1.38it/s]
3|1302|Loss: 0.9261668920516968:  20%|██        | 1301/6470 [16:00<1:02:15,  1.38it/s]
3|1302|Loss: 0.9261668920516968:  20%|██        | 1302/6470 [16:01<1:01:18,  1.40it/s]
3|1303|Loss: 0.9443702101707458:  20%|██        | 1302/6470 [16:01<1:01:18,  1.40it/s]
3|1303|Loss: 0.9443702101707458:  20%|██        | 1303/6470 [16:02<1:03:22,  1.36it/s]
3|1304|Loss: 1.0233756303787231:  20%|██        | 1303/6470 [16:02<1:03:22,  1.36it/s]
3|1304|Loss: 1.0233756303787231:  20%|██        | 1304/6470 [16:02<1:03:13,  1.36it/s]
3|1305|Loss: 0.7932897210121155:  20%|██        | 1304/6470 [16:03<1:03:13,  1.36it/s]
3|1305|Loss: 0.7932897210121155:  20%|██        | 1305/6470 [16:03<1:08:27,  1.26it/s]
3|1306|Loss: 0.9109853506088257:  20%|██        | 1305/6470 [16:04<1:08:27,  1.26it/s]
3|1306|Loss: 0.9109853506088257:  20%|██        | 1306/6470 [16:04<1:11:40,  1.20it/s]
3|1307|Loss: 0.7235056161880493:  20%|██        | 1306/6470 [16:05<1:11:40,  1.20it/s]
3|1307|Loss: 0.7235056161880493:  20%|██        | 1307/6470 [16:05<1:08:39,  1.25it/s]
3|1308|Loss: 0.7817742228507996:  20%|██        | 1307/6470 [16:05<1:08:39,  1.25it/s]
3|1308|Loss: 0.7817742228507996:  20%|██        | 1308/6470 [16:05<1:02:00,  1.39it/s]
3|1309|Loss: 0.6064647436141968:  20%|██        | 1308/6470 [16:06<1:02:00,  1.39it/s]
3|1309|Loss: 0.6064647436141968:  20%|██        | 1309/6470 [16:06<1:06:34,  1.29it/s]
3|1310|Loss: 0.8064634203910828:  20%|██        | 1309/6470 [16:07<1:06:34,  1.29it/s]
3|1310|Loss: 0.8064634203910828:  20%|██        | 1310/6470 [16:07<1:05:41,  1.31it/s]
3|1311|Loss: 0.8043884038925171:  20%|██        | 1310/6470 [16:07<1:05:41,  1.31it/s]
3|1311|Loss: 0.8043884038925171:  20%|██        | 1311/6470 [16:08<1:07:31,  1.27it/s]
3|1312|Loss: 0.7027333974838257:  20%|██        | 1311/6470 [16:08<1:07:31,  1.27it/s]
3|1312|Loss: 0.7027333974838257:  20%|██        | 1312/6470 [16:09<1:09:32,  1.24it/s]
3|1313|Loss: 0.7072598338127136:  20%|██        | 1312/6470 [16:09<1:09:32,  1.24it/s]
3|1313|Loss: 0.7072598338127136:  20%|██        | 1313/6470 [16:10<1:12:11,  1.19it/s]
3|1314|Loss: 0.7402815222740173:  20%|██        | 1313/6470 [16:10<1:12:11,  1.19it/s]
3|1314|Loss: 0.7402815222740173:  20%|██        | 1314/6470 [16:10<1:08:01,  1.26it/s]
3|1315|Loss: 0.856018602848053:  20%|██        | 1314/6470 [16:11<1:08:01,  1.26it/s] 
3|1315|Loss: 0.856018602848053:  20%|██        | 1315/6470 [16:11<1:07:11,  1.28it/s]
3|1316|Loss: 0.8471699357032776:  20%|██        | 1315/6470 [16:11<1:07:11,  1.28it/s]
3|1316|Loss: 0.8471699357032776:  20%|██        | 1316/6470 [16:12<1:05:38,  1.31it/s]
3|1317|Loss: 0.7411784529685974:  20%|██        | 1316/6470 [16:12<1:05:38,  1.31it/s]
3|1317|Loss: 0.7411784529685974:  20%|██        | 1317/6470 [16:13<1:07:42,  1.27it/s]
3|1318|Loss: 0.6899991631507874:  20%|██        | 1317/6470 [16:13<1:07:42,  1.27it/s]
3|1318|Loss: 0.6899991631507874:  20%|██        | 1318/6470 [16:13<1:08:00,  1.26it/s]
3|1319|Loss: 0.9588920474052429:  20%|██        | 1318/6470 [16:14<1:08:00,  1.26it/s]
3|1319|Loss: 0.9588920474052429:  20%|██        | 1319/6470 [16:14<1:04:15,  1.34it/s]
3|1320|Loss: 0.7795256972312927:  20%|██        | 1319/6470 [16:14<1:04:15,  1.34it/s]
3|1320|Loss: 0.7795256972312927:  20%|██        | 1320/6470 [16:15<1:02:35,  1.37it/s]
3|1321|Loss: 0.7754414081573486:  20%|██        | 1320/6470 [16:15<1:02:35,  1.37it/s]
3|1321|Loss: 0.7754414081573486:  20%|██        | 1321/6470 [16:15<1:01:30,  1.40it/s]
3|1322|Loss: 0.8321238160133362:  20%|██        | 1321/6470 [16:16<1:01:30,  1.40it/s]
3|1322|Loss: 0.8321238160133362:  20%|██        | 1322/6470 [16:16<1:01:53,  1.39it/s]
3|1323|Loss: 0.7903427481651306:  20%|██        | 1322/6470 [16:17<1:01:53,  1.39it/s]
3|1323|Loss: 0.7903427481651306:  20%|██        | 1323/6470 [16:17<1:00:08,  1.43it/s]
3|1324|Loss: 0.8449937701225281:  20%|██        | 1323/6470 [16:17<1:00:08,  1.43it/s]
3|1324|Loss: 0.8449937701225281:  20%|██        | 1324/6470 [16:18<1:01:35,  1.39it/s]
3|1325|Loss: 0.7581268548965454:  20%|██        | 1324/6470 [16:18<1:01:35,  1.39it/s]
3|1325|Loss: 0.7581268548965454:  20%|██        | 1325/6470 [16:18<1:01:39,  1.39it/s]
3|1326|Loss: 0.7302919626235962:  20%|██        | 1325/6470 [16:19<1:01:39,  1.39it/s]
3|1326|Loss: 0.7302919626235962:  20%|██        | 1326/6470 [16:19<1:03:28,  1.35it/s]
3|1327|Loss: 0.9009231925010681:  20%|██        | 1326/6470 [16:20<1:03:28,  1.35it/s]
3|1327|Loss: 0.9009231925010681:  21%|██        | 1327/6470 [16:20<1:04:20,  1.33it/s]
3|1328|Loss: 0.8662683367729187:  21%|██        | 1327/6470 [16:20<1:04:20,  1.33it/s]
3|1328|Loss: 0.8662683367729187:  21%|██        | 1328/6470 [16:20<58:35,  1.46it/s]  
3|1329|Loss: 0.7974553108215332:  21%|██        | 1328/6470 [16:21<58:35,  1.46it/s]
3|1329|Loss: 0.7974553108215332:  21%|██        | 1329/6470 [16:21<57:23,  1.49it/s]
3|1330|Loss: 1.1147089004516602:  21%|██        | 1329/6470 [16:22<57:23,  1.49it/s]
3|1330|Loss: 1.1147089004516602:  21%|██        | 1330/6470 [16:22<1:04:01,  1.34it/s]
3|1331|Loss: 0.684593677520752:  21%|██        | 1330/6470 [16:22<1:04:01,  1.34it/s] 
3|1331|Loss: 0.684593677520752:  21%|██        | 1331/6470 [16:23<1:06:24,  1.29it/s]
3|1332|Loss: 0.9100086688995361:  21%|██        | 1331/6470 [16:23<1:06:24,  1.29it/s]
3|1332|Loss: 0.9100086688995361:  21%|██        | 1332/6470 [16:24<1:10:18,  1.22it/s]
3|1333|Loss: 0.9347379207611084:  21%|██        | 1332/6470 [16:24<1:10:18,  1.22it/s]
3|1333|Loss: 0.9347379207611084:  21%|██        | 1333/6470 [16:25<1:10:56,  1.21it/s]
3|1334|Loss: 0.8434856534004211:  21%|██        | 1333/6470 [16:25<1:10:56,  1.21it/s]
3|1334|Loss: 0.8434856534004211:  21%|██        | 1334/6470 [16:25<1:05:10,  1.31it/s]
3|1335|Loss: 0.946589469909668:  21%|██        | 1334/6470 [16:25<1:05:10,  1.31it/s] 
3|1335|Loss: 0.946589469909668:  21%|██        | 1335/6470 [16:26<59:18,  1.44it/s]  
3|1336|Loss: 0.765117883682251:  21%|██        | 1335/6470 [16:26<59:18,  1.44it/s]
3|1336|Loss: 0.765117883682251:  21%|██        | 1336/6470 [16:26<58:20,  1.47it/s]
3|1337|Loss: 0.8077713251113892:  21%|██        | 1336/6470 [16:27<58:20,  1.47it/s]
3|1337|Loss: 0.8077713251113892:  21%|██        | 1337/6470 [16:27<56:28,  1.51it/s]
3|1338|Loss: 0.7596263885498047:  21%|██        | 1337/6470 [16:27<56:28,  1.51it/s]
3|1338|Loss: 0.7596263885498047:  21%|██        | 1338/6470 [16:28<1:00:57,  1.40it/s]
3|1339|Loss: 0.9621089100837708:  21%|██        | 1338/6470 [16:28<1:00:57,  1.40it/s]
3|1339|Loss: 0.9621089100837708:  21%|██        | 1339/6470 [16:29<1:00:35,  1.41it/s]
3|1340|Loss: 0.8030332326889038:  21%|██        | 1339/6470 [16:29<1:00:35,  1.41it/s]
3|1340|Loss: 0.8030332326889038:  21%|██        | 1340/6470 [16:29<1:02:10,  1.38it/s]
3|1341|Loss: 0.7435071468353271:  21%|██        | 1340/6470 [16:30<1:02:10,  1.38it/s]
3|1341|Loss: 0.7435071468353271:  21%|██        | 1341/6470 [16:30<1:03:50,  1.34it/s]
3|1342|Loss: 0.8332699537277222:  21%|██        | 1341/6470 [16:30<1:03:50,  1.34it/s]
3|1342|Loss: 0.8332699537277222:  21%|██        | 1342/6470 [16:31<1:03:15,  1.35it/s]
3|1343|Loss: 0.7288163900375366:  21%|██        | 1342/6470 [16:31<1:03:15,  1.35it/s]
3|1343|Loss: 0.7288163900375366:  21%|██        | 1343/6470 [16:32<1:04:14,  1.33it/s]
3|1344|Loss: 1.095259189605713:  21%|██        | 1343/6470 [16:32<1:04:14,  1.33it/s] 
3|1344|Loss: 1.095259189605713:  21%|██        | 1344/6470 [16:33<1:09:40,  1.23it/s]
3|1345|Loss: 0.7576457262039185:  21%|██        | 1344/6470 [16:33<1:09:40,  1.23it/s]
3|1345|Loss: 0.7576457262039185:  21%|██        | 1345/6470 [16:33<1:04:40,  1.32it/s]
3|1346|Loss: 0.8332231640815735:  21%|██        | 1345/6470 [16:34<1:04:40,  1.32it/s]
3|1346|Loss: 0.8332231640815735:  21%|██        | 1346/6470 [16:34<1:02:16,  1.37it/s]
3|1347|Loss: 0.8048937320709229:  21%|██        | 1346/6470 [16:34<1:02:16,  1.37it/s]
3|1347|Loss: 0.8048937320709229:  21%|██        | 1347/6470 [16:35<1:02:03,  1.38it/s]
3|1348|Loss: 0.7210425138473511:  21%|██        | 1347/6470 [16:35<1:02:03,  1.38it/s]
3|1348|Loss: 0.7210425138473511:  21%|██        | 1348/6470 [16:35<1:03:14,  1.35it/s]
3|1349|Loss: 0.7555304765701294:  21%|██        | 1348/6470 [16:36<1:03:14,  1.35it/s]
3|1349|Loss: 0.7555304765701294:  21%|██        | 1349/6470 [16:36<1:03:37,  1.34it/s]
3|1350|Loss: 0.935621976852417:  21%|██        | 1349/6470 [16:36<1:03:37,  1.34it/s] 
3|1350|Loss: 0.935621976852417:  21%|██        | 1350/6470 [16:37<1:02:03,  1.37it/s]
3|1351|Loss: 0.9268231987953186:  21%|██        | 1350/6470 [16:37<1:02:03,  1.37it/s]
3|1351|Loss: 0.9268231987953186:  21%|██        | 1351/6470 [16:37<59:58,  1.42it/s]  
3|1352|Loss: 0.8909660577774048:  21%|██        | 1351/6470 [16:38<59:58,  1.42it/s]
3|1352|Loss: 0.8909660577774048:  21%|██        | 1352/6470 [16:38<1:01:25,  1.39it/s]
3|1353|Loss: 0.866852879524231:  21%|██        | 1352/6470 [16:39<1:01:25,  1.39it/s] 
3|1353|Loss: 0.866852879524231:  21%|██        | 1353/6470 [16:39<1:03:21,  1.35it/s]
3|1354|Loss: 0.8375924825668335:  21%|██        | 1353/6470 [16:39<1:03:21,  1.35it/s]
3|1354|Loss: 0.8375924825668335:  21%|██        | 1354/6470 [16:40<1:07:02,  1.27it/s]
3|1355|Loss: 0.8243813514709473:  21%|██        | 1354/6470 [16:40<1:07:02,  1.27it/s]
3|1355|Loss: 0.8243813514709473:  21%|██        | 1355/6470 [16:41<1:05:00,  1.31it/s]
3|1356|Loss: 0.951560378074646:  21%|██        | 1355/6470 [16:41<1:05:00,  1.31it/s] 
3|1356|Loss: 0.951560378074646:  21%|██        | 1356/6470 [16:41<1:03:04,  1.35it/s]
3|1357|Loss: 0.7740197777748108:  21%|██        | 1356/6470 [16:42<1:03:04,  1.35it/s]
3|1357|Loss: 0.7740197777748108:  21%|██        | 1357/6470 [16:42<1:00:28,  1.41it/s]
3|1358|Loss: 0.6944917440414429:  21%|██        | 1357/6470 [16:42<1:00:28,  1.41it/s]
3|1358|Loss: 0.6944917440414429:  21%|██        | 1358/6470 [16:42<56:26,  1.51it/s]  
3|1359|Loss: 1.1404736042022705:  21%|██        | 1358/6470 [16:43<56:26,  1.51it/s]
3|1359|Loss: 1.1404736042022705:  21%|██        | 1359/6470 [16:44<1:07:03,  1.27it/s]
3|1360|Loss: 0.8875309228897095:  21%|██        | 1359/6470 [16:44<1:07:03,  1.27it/s]
3|1360|Loss: 0.8875309228897095:  21%|██        | 1360/6470 [16:44<1:09:44,  1.22it/s]
3|1361|Loss: 0.7400292158126831:  21%|██        | 1360/6470 [16:45<1:09:44,  1.22it/s]
3|1361|Loss: 0.7400292158126831:  21%|██        | 1361/6470 [16:45<1:08:49,  1.24it/s]
3|1362|Loss: 0.9162576794624329:  21%|██        | 1361/6470 [16:46<1:08:49,  1.24it/s]
3|1362|Loss: 0.9162576794624329:  21%|██        | 1362/6470 [16:46<1:05:28,  1.30it/s]
3|1363|Loss: 0.7814563512802124:  21%|██        | 1362/6470 [16:46<1:05:28,  1.30it/s]
3|1363|Loss: 0.7814563512802124:  21%|██        | 1363/6470 [16:47<1:08:11,  1.25it/s]
3|1364|Loss: 0.8234840631484985:  21%|██        | 1363/6470 [16:47<1:08:11,  1.25it/s]
3|1364|Loss: 0.8234840631484985:  21%|██        | 1364/6470 [16:48<1:09:00,  1.23it/s]
3|1365|Loss: 0.8830053210258484:  21%|██        | 1364/6470 [16:48<1:09:00,  1.23it/s]
3|1365|Loss: 0.8830053210258484:  21%|██        | 1365/6470 [16:49<1:12:22,  1.18it/s]
3|1366|Loss: 0.816498339176178:  21%|██        | 1365/6470 [16:49<1:12:22,  1.18it/s] 
3|1366|Loss: 0.816498339176178:  21%|██        | 1366/6470 [16:49<1:10:10,  1.21it/s]
3|1367|Loss: 0.699194610118866:  21%|██        | 1366/6470 [16:50<1:10:10,  1.21it/s]
3|1367|Loss: 0.699194610118866:  21%|██        | 1367/6470 [16:50<1:14:00,  1.15it/s]
3|1368|Loss: 0.9130507111549377:  21%|██        | 1367/6470 [16:51<1:14:00,  1.15it/s]
3|1368|Loss: 0.9130507111549377:  21%|██        | 1368/6470 [16:51<1:08:16,  1.25it/s]
3|1369|Loss: 0.7714531421661377:  21%|██        | 1368/6470 [16:51<1:08:16,  1.25it/s]
3|1369|Loss: 0.7714531421661377:  21%|██        | 1369/6470 [16:52<1:04:41,  1.31it/s]
3|1370|Loss: 0.8695049285888672:  21%|██        | 1369/6470 [16:52<1:04:41,  1.31it/s]
3|1370|Loss: 0.8695049285888672:  21%|██        | 1370/6470 [16:52<1:03:16,  1.34it/s]
3|1371|Loss: 0.9946566224098206:  21%|██        | 1370/6470 [16:53<1:03:16,  1.34it/s]
3|1371|Loss: 0.9946566224098206:  21%|██        | 1371/6470 [16:53<1:04:08,  1.32it/s]
3|1372|Loss: 0.7553338408470154:  21%|██        | 1371/6470 [16:53<1:04:08,  1.32it/s]
3|1372|Loss: 0.7553338408470154:  21%|██        | 1372/6470 [16:54<1:02:28,  1.36it/s]
3|1373|Loss: 0.8151580691337585:  21%|██        | 1372/6470 [16:54<1:02:28,  1.36it/s]
3|1373|Loss: 0.8151580691337585:  21%|██        | 1373/6470 [16:54<1:00:40,  1.40it/s]
3|1374|Loss: 0.7997802495956421:  21%|██        | 1373/6470 [16:55<1:00:40,  1.40it/s]
3|1374|Loss: 0.7997802495956421:  21%|██        | 1374/6470 [16:55<58:44,  1.45it/s]  
3|1375|Loss: 0.8171342015266418:  21%|██        | 1374/6470 [16:55<58:44,  1.45it/s]
3|1375|Loss: 0.8171342015266418:  21%|██▏       | 1375/6470 [16:56<58:28,  1.45it/s]
3|1376|Loss: 0.7934472560882568:  21%|██▏       | 1375/6470 [16:56<58:28,  1.45it/s]
3|1376|Loss: 0.7934472560882568:  21%|██▏       | 1376/6470 [16:56<58:43,  1.45it/s]
3|1377|Loss: 0.8890476822853088:  21%|██▏       | 1376/6470 [16:57<58:43,  1.45it/s]
3|1377|Loss: 0.8890476822853088:  21%|██▏       | 1377/6470 [16:57<59:50,  1.42it/s]
3|1378|Loss: 0.770745575428009:  21%|██▏       | 1377/6470 [16:58<59:50,  1.42it/s] 
3|1378|Loss: 0.770745575428009:  21%|██▏       | 1378/6470 [16:58<59:37,  1.42it/s]
3|1379|Loss: 0.711031973361969:  21%|██▏       | 1378/6470 [16:58<59:37,  1.42it/s]
3|1379|Loss: 0.711031973361969:  21%|██▏       | 1379/6470 [16:59<1:01:37,  1.38it/s]
3|1380|Loss: 0.8093259930610657:  21%|██▏       | 1379/6470 [16:59<1:01:37,  1.38it/s]
3|1380|Loss: 0.8093259930610657:  21%|██▏       | 1380/6470 [16:59<1:02:40,  1.35it/s]
3|1381|Loss: 0.7369006872177124:  21%|██▏       | 1380/6470 [17:00<1:02:40,  1.35it/s]
3|1381|Loss: 0.7369006872177124:  21%|██▏       | 1381/6470 [17:00<1:05:50,  1.29it/s]
3|1382|Loss: 0.9767119288444519:  21%|██▏       | 1381/6470 [17:01<1:05:50,  1.29it/s]
3|1382|Loss: 0.9767119288444519:  21%|██▏       | 1382/6470 [17:01<1:06:42,  1.27it/s]
3|1383|Loss: 0.8069973587989807:  21%|██▏       | 1382/6470 [17:02<1:06:42,  1.27it/s]
3|1383|Loss: 0.8069973587989807:  21%|██▏       | 1383/6470 [17:02<1:08:58,  1.23it/s]
3|1384|Loss: 0.771434485912323:  21%|██▏       | 1383/6470 [17:02<1:08:58,  1.23it/s] 
3|1384|Loss: 0.771434485912323:  21%|██▏       | 1384/6470 [17:03<1:06:11,  1.28it/s]
3|1385|Loss: 0.7569054961204529:  21%|██▏       | 1384/6470 [17:03<1:06:11,  1.28it/s]
3|1385|Loss: 0.7569054961204529:  21%|██▏       | 1385/6470 [17:03<1:05:13,  1.30it/s]
3|1386|Loss: 0.8851386308670044:  21%|██▏       | 1385/6470 [17:04<1:05:13,  1.30it/s]
3|1386|Loss: 0.8851386308670044:  21%|██▏       | 1386/6470 [17:04<1:05:18,  1.30it/s]
3|1387|Loss: 0.873881459236145:  21%|██▏       | 1386/6470 [17:05<1:05:18,  1.30it/s] 
3|1387|Loss: 0.873881459236145:  21%|██▏       | 1387/6470 [17:05<1:06:05,  1.28it/s]
3|1388|Loss: 0.8167558312416077:  21%|██▏       | 1387/6470 [17:05<1:06:05,  1.28it/s]
3|1388|Loss: 0.8167558312416077:  21%|██▏       | 1388/6470 [17:06<1:06:53,  1.27it/s]
3|1389|Loss: 0.7205169200897217:  21%|██▏       | 1388/6470 [17:06<1:06:53,  1.27it/s]
3|1389|Loss: 0.7205169200897217:  21%|██▏       | 1389/6470 [17:06<1:03:27,  1.33it/s]
3|1390|Loss: 0.8932327032089233:  21%|██▏       | 1389/6470 [17:07<1:03:27,  1.33it/s]
3|1390|Loss: 0.8932327032089233:  21%|██▏       | 1390/6470 [17:07<57:05,  1.48it/s]  
3|1391|Loss: 0.9625572562217712:  21%|██▏       | 1390/6470 [17:07<57:05,  1.48it/s]
3|1391|Loss: 0.9625572562217712:  21%|██▏       | 1391/6470 [17:08<58:32,  1.45it/s]
3|1392|Loss: 0.7301358580589294:  21%|██▏       | 1391/6470 [17:08<58:32,  1.45it/s]
3|1392|Loss: 0.7301358580589294:  22%|██▏       | 1392/6470 [17:08<52:39,  1.61it/s]
3|1393|Loss: 0.8161250948905945:  22%|██▏       | 1392/6470 [17:08<52:39,  1.61it/s]
3|1393|Loss: 0.8161250948905945:  22%|██▏       | 1393/6470 [17:09<54:40,  1.55it/s]
3|1394|Loss: 0.8328444361686707:  22%|██▏       | 1393/6470 [17:09<54:40,  1.55it/s]
3|1394|Loss: 0.8328444361686707:  22%|██▏       | 1394/6470 [17:10<56:38,  1.49it/s]
3|1395|Loss: 0.822170078754425:  22%|██▏       | 1394/6470 [17:10<56:38,  1.49it/s] 
3|1395|Loss: 0.822170078754425:  22%|██▏       | 1395/6470 [17:10<59:48,  1.41it/s]
3|1396|Loss: 0.9070863127708435:  22%|██▏       | 1395/6470 [17:11<59:48,  1.41it/s]
3|1396|Loss: 0.9070863127708435:  22%|██▏       | 1396/6470 [17:11<1:00:16,  1.40it/s]
3|1397|Loss: 0.893151044845581:  22%|██▏       | 1396/6470 [17:12<1:00:16,  1.40it/s] 
3|1397|Loss: 0.893151044845581:  22%|██▏       | 1397/6470 [17:12<1:01:18,  1.38it/s]
3|1398|Loss: 0.776877760887146:  22%|██▏       | 1397/6470 [17:12<1:01:18,  1.38it/s]
3|1398|Loss: 0.776877760887146:  22%|██▏       | 1398/6470 [17:13<59:57,  1.41it/s]  
3|1399|Loss: 0.8547608256340027:  22%|██▏       | 1398/6470 [17:13<59:57,  1.41it/s]
3|1399|Loss: 0.8547608256340027:  22%|██▏       | 1399/6470 [17:13<58:24,  1.45it/s]
3|1400|Loss: 0.7234020233154297:  22%|██▏       | 1399/6470 [17:14<58:24,  1.45it/s]
3|1400|Loss: 0.7234020233154297:  22%|██▏       | 1400/6470 [17:14<56:43,  1.49it/s]
3|1401|Loss: 0.778059720993042:  22%|██▏       | 1400/6470 [17:14<56:43,  1.49it/s] 
3|1401|Loss: 0.778059720993042:  22%|██▏       | 1401/6470 [17:14<56:35,  1.49it/s]
3|1402|Loss: 0.6974148750305176:  22%|██▏       | 1401/6470 [17:15<56:35,  1.49it/s]
3|1402|Loss: 0.6974148750305176:  22%|██▏       | 1402/6470 [17:15<1:00:27,  1.40it/s]
3|1403|Loss: 0.8521373271942139:  22%|██▏       | 1402/6470 [17:16<1:00:27,  1.40it/s]
3|1403|Loss: 0.8521373271942139:  22%|██▏       | 1403/6470 [17:16<1:02:43,  1.35it/s]
3|1404|Loss: 0.8714519143104553:  22%|██▏       | 1403/6470 [17:17<1:02:43,  1.35it/s]
3|1404|Loss: 0.8714519143104553:  22%|██▏       | 1404/6470 [17:17<1:04:00,  1.32it/s]
3|1405|Loss: 0.9084714651107788:  22%|██▏       | 1404/6470 [17:17<1:04:00,  1.32it/s]
3|1405|Loss: 0.9084714651107788:  22%|██▏       | 1405/6470 [17:18<1:04:32,  1.31it/s]
3|1406|Loss: 0.7994625568389893:  22%|██▏       | 1405/6470 [17:18<1:04:32,  1.31it/s]
3|1406|Loss: 0.7994625568389893:  22%|██▏       | 1406/6470 [17:18<1:04:31,  1.31it/s]
3|1407|Loss: 0.8264325857162476:  22%|██▏       | 1406/6470 [17:19<1:04:31,  1.31it/s]
3|1407|Loss: 0.8264325857162476:  22%|██▏       | 1407/6470 [17:19<1:01:39,  1.37it/s]
3|1408|Loss: 0.8255759477615356:  22%|██▏       | 1407/6470 [17:19<1:01:39,  1.37it/s]
3|1408|Loss: 0.8255759477615356:  22%|██▏       | 1408/6470 [17:20<59:53,  1.41it/s]  
3|1409|Loss: 0.8428667187690735:  22%|██▏       | 1408/6470 [17:20<59:53,  1.41it/s]
3|1409|Loss: 0.8428667187690735:  22%|██▏       | 1409/6470 [17:21<1:04:47,  1.30it/s]
3|1410|Loss: 0.8852657079696655:  22%|██▏       | 1409/6470 [17:21<1:04:47,  1.30it/s]
3|1410|Loss: 0.8852657079696655:  22%|██▏       | 1410/6470 [17:21<1:04:45,  1.30it/s]
3|1411|Loss: 0.957144021987915:  22%|██▏       | 1410/6470 [17:22<1:04:45,  1.30it/s] 
3|1411|Loss: 0.957144021987915:  22%|██▏       | 1411/6470 [17:22<1:06:35,  1.27it/s]
3|1412|Loss: 0.7912746667861938:  22%|██▏       | 1411/6470 [17:23<1:06:35,  1.27it/s]
3|1412|Loss: 0.7912746667861938:  22%|██▏       | 1412/6470 [17:23<1:03:25,  1.33it/s]
3|1413|Loss: 0.749487578868866:  22%|██▏       | 1412/6470 [17:23<1:03:25,  1.33it/s] 
3|1413|Loss: 0.749487578868866:  22%|██▏       | 1413/6470 [17:24<1:05:21,  1.29it/s]
3|1414|Loss: 0.8592387437820435:  22%|██▏       | 1413/6470 [17:24<1:05:21,  1.29it/s]
3|1414|Loss: 0.8592387437820435:  22%|██▏       | 1414/6470 [17:24<1:02:54,  1.34it/s]
3|1415|Loss: 0.7477529048919678:  22%|██▏       | 1414/6470 [17:25<1:02:54,  1.34it/s]
3|1415|Loss: 0.7477529048919678:  22%|██▏       | 1415/6470 [17:25<58:00,  1.45it/s]  
3|1416|Loss: 0.7787793874740601:  22%|██▏       | 1415/6470 [17:25<58:00,  1.45it/s]
3|1416|Loss: 0.7787793874740601:  22%|██▏       | 1416/6470 [17:26<59:16,  1.42it/s]
3|1417|Loss: 0.9069790244102478:  22%|██▏       | 1416/6470 [17:26<59:16,  1.42it/s]
3|1417|Loss: 0.9069790244102478:  22%|██▏       | 1417/6470 [17:27<1:01:27,  1.37it/s]
3|1418|Loss: 0.9433507323265076:  22%|██▏       | 1417/6470 [17:27<1:01:27,  1.37it/s]
3|1418|Loss: 0.9433507323265076:  22%|██▏       | 1418/6470 [17:27<1:02:39,  1.34it/s]
3|1419|Loss: 0.7725573182106018:  22%|██▏       | 1418/6470 [17:28<1:02:39,  1.34it/s]
3|1419|Loss: 0.7725573182106018:  22%|██▏       | 1419/6470 [17:28<1:00:03,  1.40it/s]
3|1420|Loss: 0.8237359523773193:  22%|██▏       | 1419/6470 [17:28<1:00:03,  1.40it/s]
3|1420|Loss: 0.8237359523773193:  22%|██▏       | 1420/6470 [17:29<57:16,  1.47it/s]  
3|1421|Loss: 0.7627747058868408:  22%|██▏       | 1420/6470 [17:29<57:16,  1.47it/s]
3|1421|Loss: 0.7627747058868408:  22%|██▏       | 1421/6470 [17:29<59:05,  1.42it/s]
3|1422|Loss: 0.8558021783828735:  22%|██▏       | 1421/6470 [17:30<59:05,  1.42it/s]
3|1422|Loss: 0.8558021783828735:  22%|██▏       | 1422/6470 [17:30<1:02:37,  1.34it/s]
3|1423|Loss: 0.7843100428581238:  22%|██▏       | 1422/6470 [17:31<1:02:37,  1.34it/s]
3|1423|Loss: 0.7843100428581238:  22%|██▏       | 1423/6470 [17:31<59:15,  1.42it/s]  
3|1424|Loss: 0.8878565430641174:  22%|██▏       | 1423/6470 [17:31<59:15,  1.42it/s]
3|1424|Loss: 0.8878565430641174:  22%|██▏       | 1424/6470 [17:31<59:55,  1.40it/s]
3|1425|Loss: 0.9233071208000183:  22%|██▏       | 1424/6470 [17:32<59:55,  1.40it/s]
3|1425|Loss: 0.9233071208000183:  22%|██▏       | 1425/6470 [17:32<1:01:05,  1.38it/s]
3|1426|Loss: 0.7860594391822815:  22%|██▏       | 1425/6470 [17:33<1:01:05,  1.38it/s]
3|1426|Loss: 0.7860594391822815:  22%|██▏       | 1426/6470 [17:33<59:05,  1.42it/s]  
3|1427|Loss: 0.7364319562911987:  22%|██▏       | 1426/6470 [17:33<59:05,  1.42it/s]
3|1427|Loss: 0.7364319562911987:  22%|██▏       | 1427/6470 [17:34<1:00:34,  1.39it/s]
3|1428|Loss: 0.8517162799835205:  22%|██▏       | 1427/6470 [17:34<1:00:34,  1.39it/s]
3|1428|Loss: 0.8517162799835205:  22%|██▏       | 1428/6470 [17:35<1:05:04,  1.29it/s]
3|1429|Loss: 0.8095433712005615:  22%|██▏       | 1428/6470 [17:35<1:05:04,  1.29it/s]
3|1429|Loss: 0.8095433712005615:  22%|██▏       | 1429/6470 [17:35<1:01:51,  1.36it/s]
3|1430|Loss: 0.7982557415962219:  22%|██▏       | 1429/6470 [17:36<1:01:51,  1.36it/s]
3|1430|Loss: 0.7982557415962219:  22%|██▏       | 1430/6470 [17:36<59:44,  1.41it/s]  
3|1431|Loss: 0.8106598854064941:  22%|██▏       | 1430/6470 [17:36<59:44,  1.41it/s]
3|1431|Loss: 0.8106598854064941:  22%|██▏       | 1431/6470 [17:37<1:03:10,  1.33it/s]
3|1432|Loss: 0.9685789942741394:  22%|██▏       | 1431/6470 [17:37<1:03:10,  1.33it/s]
3|1432|Loss: 0.9685789942741394:  22%|██▏       | 1432/6470 [17:38<1:04:36,  1.30it/s]
3|1433|Loss: 0.8541744351387024:  22%|██▏       | 1432/6470 [17:38<1:04:36,  1.30it/s]
3|1433|Loss: 0.8541744351387024:  22%|██▏       | 1433/6470 [17:38<1:02:56,  1.33it/s]
3|1434|Loss: 0.8045814633369446:  22%|██▏       | 1433/6470 [17:39<1:02:56,  1.33it/s]
3|1434|Loss: 0.8045814633369446:  22%|██▏       | 1434/6470 [17:39<1:02:40,  1.34it/s]
3|1435|Loss: 0.7934737801551819:  22%|██▏       | 1434/6470 [17:39<1:02:40,  1.34it/s]
3|1435|Loss: 0.7934737801551819:  22%|██▏       | 1435/6470 [17:40<59:23,  1.41it/s]  
3|1436|Loss: 0.7073252201080322:  22%|██▏       | 1435/6470 [17:40<59:23,  1.41it/s]
3|1436|Loss: 0.7073252201080322:  22%|██▏       | 1436/6470 [17:40<57:43,  1.45it/s]
3|1437|Loss: 0.8314931392669678:  22%|██▏       | 1436/6470 [17:41<57:43,  1.45it/s]
3|1437|Loss: 0.8314931392669678:  22%|██▏       | 1437/6470 [17:41<57:59,  1.45it/s]
3|1438|Loss: 0.8212559223175049:  22%|██▏       | 1437/6470 [17:41<57:59,  1.45it/s]
3|1438|Loss: 0.8212559223175049:  22%|██▏       | 1438/6470 [17:42<1:03:37,  1.32it/s]
3|1439|Loss: 0.8471891283988953:  22%|██▏       | 1438/6470 [17:42<1:03:37,  1.32it/s]
3|1439|Loss: 0.8471891283988953:  22%|██▏       | 1439/6470 [17:43<1:05:24,  1.28it/s]
3|1440|Loss: 0.8460685610771179:  22%|██▏       | 1439/6470 [17:43<1:05:24,  1.28it/s]
3|1440|Loss: 0.8460685610771179:  22%|██▏       | 1440/6470 [17:44<1:08:15,  1.23it/s]
3|1441|Loss: 0.8450555801391602:  22%|██▏       | 1440/6470 [17:44<1:08:15,  1.23it/s]
3|1441|Loss: 0.8450555801391602:  22%|██▏       | 1441/6470 [17:44<1:10:54,  1.18it/s]
3|1442|Loss: 0.7846924066543579:  22%|██▏       | 1441/6470 [17:45<1:10:54,  1.18it/s]
3|1442|Loss: 0.7846924066543579:  22%|██▏       | 1442/6470 [17:45<1:13:06,  1.15it/s]
3|1443|Loss: 0.9908050298690796:  22%|██▏       | 1442/6470 [17:46<1:13:06,  1.15it/s]
3|1443|Loss: 0.9908050298690796:  22%|██▏       | 1443/6470 [17:46<1:10:15,  1.19it/s]
3|1444|Loss: 0.8908211588859558:  22%|██▏       | 1443/6470 [17:46<1:10:15,  1.19it/s]
3|1444|Loss: 0.8908211588859558:  22%|██▏       | 1444/6470 [17:47<1:02:04,  1.35it/s]
3|1445|Loss: 0.8328878879547119:  22%|██▏       | 1444/6470 [17:47<1:02:04,  1.35it/s]
3|1445|Loss: 0.8328878879547119:  22%|██▏       | 1445/6470 [17:47<1:00:39,  1.38it/s]
3|1446|Loss: 0.8965248465538025:  22%|██▏       | 1445/6470 [17:48<1:00:39,  1.38it/s]
3|1446|Loss: 0.8965248465538025:  22%|██▏       | 1446/6470 [17:48<1:03:24,  1.32it/s]
3|1447|Loss: 0.7870728969573975:  22%|██▏       | 1446/6470 [17:49<1:03:24,  1.32it/s]
3|1447|Loss: 0.7870728969573975:  22%|██▏       | 1447/6470 [17:49<1:06:50,  1.25it/s]
3|1448|Loss: 0.890081524848938:  22%|██▏       | 1447/6470 [17:49<1:06:50,  1.25it/s] 
3|1448|Loss: 0.890081524848938:  22%|██▏       | 1448/6470 [17:50<1:03:55,  1.31it/s]
3|1449|Loss: 0.8814614415168762:  22%|██▏       | 1448/6470 [17:50<1:03:55,  1.31it/s]
3|1449|Loss: 0.8814614415168762:  22%|██▏       | 1449/6470 [17:51<1:05:04,  1.29it/s]
3|1450|Loss: 0.6821195483207703:  22%|██▏       | 1449/6470 [17:51<1:05:04,  1.29it/s]
3|1450|Loss: 0.6821195483207703:  22%|██▏       | 1450/6470 [17:51<1:00:58,  1.37it/s]
3|1451|Loss: 0.8324007391929626:  22%|██▏       | 1450/6470 [17:52<1:00:58,  1.37it/s]
3|1451|Loss: 0.8324007391929626:  22%|██▏       | 1451/6470 [17:52<1:00:32,  1.38it/s]
3|1452|Loss: 0.8779280185699463:  22%|██▏       | 1451/6470 [17:52<1:00:32,  1.38it/s]
3|1452|Loss: 0.8779280185699463:  22%|██▏       | 1452/6470 [17:53<1:03:15,  1.32it/s]
3|1453|Loss: 0.7769365906715393:  22%|██▏       | 1452/6470 [17:53<1:03:15,  1.32it/s]
3|1453|Loss: 0.7769365906715393:  22%|██▏       | 1453/6470 [17:54<1:04:40,  1.29it/s]
3|1454|Loss: 0.6975796818733215:  22%|██▏       | 1453/6470 [17:54<1:04:40,  1.29it/s]
3|1454|Loss: 0.6975796818733215:  22%|██▏       | 1454/6470 [17:54<1:04:21,  1.30it/s]
3|1455|Loss: 0.8641907572746277:  22%|██▏       | 1454/6470 [17:55<1:04:21,  1.30it/s]
3|1455|Loss: 0.8641907572746277:  22%|██▏       | 1455/6470 [17:55<1:05:07,  1.28it/s]
3|1456|Loss: 0.9162520170211792:  22%|██▏       | 1455/6470 [17:56<1:05:07,  1.28it/s]
3|1456|Loss: 0.9162520170211792:  23%|██▎       | 1456/6470 [17:56<1:01:36,  1.36it/s]
3|1457|Loss: 0.737302303314209:  23%|██▎       | 1456/6470 [17:56<1:01:36,  1.36it/s] 
3|1457|Loss: 0.737302303314209:  23%|██▎       | 1457/6470 [17:56<55:30,  1.51it/s]  
3|1458|Loss: 0.8329403400421143:  23%|██▎       | 1457/6470 [17:57<55:30,  1.51it/s]
3|1458|Loss: 0.8329403400421143:  23%|██▎       | 1458/6470 [17:57<56:44,  1.47it/s]
3|1459|Loss: 0.7606548070907593:  23%|██▎       | 1458/6470 [17:57<56:44,  1.47it/s]
3|1459|Loss: 0.7606548070907593:  23%|██▎       | 1459/6470 [17:58<54:47,  1.52it/s]
3|1460|Loss: 0.8363592028617859:  23%|██▎       | 1459/6470 [17:58<54:47,  1.52it/s]
3|1460|Loss: 0.8363592028617859:  23%|██▎       | 1460/6470 [17:58<53:58,  1.55it/s]
3|1461|Loss: 0.8219018578529358:  23%|██▎       | 1460/6470 [17:59<53:58,  1.55it/s]
3|1461|Loss: 0.8219018578529358:  23%|██▎       | 1461/6470 [17:59<57:10,  1.46it/s]
3|1462|Loss: 0.7722248435020447:  23%|██▎       | 1461/6470 [17:59<57:10,  1.46it/s]
3|1462|Loss: 0.7722248435020447:  23%|██▎       | 1462/6470 [18:00<57:15,  1.46it/s]
3|1463|Loss: 0.8522555828094482:  23%|██▎       | 1462/6470 [18:00<57:15,  1.46it/s]
3|1463|Loss: 0.8522555828094482:  23%|██▎       | 1463/6470 [18:00<57:09,  1.46it/s]
3|1464|Loss: 0.8749656081199646:  23%|██▎       | 1463/6470 [18:01<57:09,  1.46it/s]
3|1464|Loss: 0.8749656081199646:  23%|██▎       | 1464/6470 [18:01<58:22,  1.43it/s]
3|1465|Loss: 0.7423046827316284:  23%|██▎       | 1464/6470 [18:01<58:22,  1.43it/s]
3|1465|Loss: 0.7423046827316284:  23%|██▎       | 1465/6470 [18:02<59:22,  1.40it/s]
3|1466|Loss: 0.8028008937835693:  23%|██▎       | 1465/6470 [18:02<59:22,  1.40it/s]
3|1466|Loss: 0.8028008937835693:  23%|██▎       | 1466/6470 [18:02<57:07,  1.46it/s]
3|1467|Loss: 0.8600807189941406:  23%|██▎       | 1466/6470 [18:03<57:07,  1.46it/s]
3|1467|Loss: 0.8600807189941406:  23%|██▎       | 1467/6470 [18:03<1:00:54,  1.37it/s]
3|1468|Loss: 0.8772019743919373:  23%|██▎       | 1467/6470 [18:04<1:00:54,  1.37it/s]
3|1468|Loss: 0.8772019743919373:  23%|██▎       | 1468/6470 [18:04<1:01:06,  1.36it/s]
3|1469|Loss: 0.8719434142112732:  23%|██▎       | 1468/6470 [18:04<1:01:06,  1.36it/s]
3|1469|Loss: 0.8719434142112732:  23%|██▎       | 1469/6470 [18:05<59:43,  1.40it/s]  
3|1470|Loss: 0.7273958921432495:  23%|██▎       | 1469/6470 [18:05<59:43,  1.40it/s]
3|1470|Loss: 0.7273958921432495:  23%|██▎       | 1470/6470 [18:05<58:54,  1.41it/s]
3|1471|Loss: 0.9288634657859802:  23%|██▎       | 1470/6470 [18:06<58:54,  1.41it/s]
3|1471|Loss: 0.9288634657859802:  23%|██▎       | 1471/6470 [18:06<58:32,  1.42it/s]
3|1472|Loss: 0.8312522768974304:  23%|██▎       | 1471/6470 [18:07<58:32,  1.42it/s]
3|1472|Loss: 0.8312522768974304:  23%|██▎       | 1472/6470 [18:07<1:01:36,  1.35it/s]
3|1473|Loss: 0.741078794002533:  23%|██▎       | 1472/6470 [18:07<1:01:36,  1.35it/s] 
3|1473|Loss: 0.741078794002533:  23%|██▎       | 1473/6470 [18:08<1:03:37,  1.31it/s]
3|1474|Loss: 0.7750906348228455:  23%|██▎       | 1473/6470 [18:08<1:03:37,  1.31it/s]
3|1474|Loss: 0.7750906348228455:  23%|██▎       | 1474/6470 [18:08<1:00:31,  1.38it/s]
3|1475|Loss: 0.7112258076667786:  23%|██▎       | 1474/6470 [18:09<1:00:31,  1.38it/s]
3|1475|Loss: 0.7112258076667786:  23%|██▎       | 1475/6470 [18:09<58:35,  1.42it/s]  
3|1476|Loss: 0.9085983633995056:  23%|██▎       | 1475/6470 [18:09<58:35,  1.42it/s]
3|1476|Loss: 0.9085983633995056:  23%|██▎       | 1476/6470 [18:10<58:26,  1.42it/s]
3|1477|Loss: 0.7647092938423157:  23%|██▎       | 1476/6470 [18:10<58:26,  1.42it/s]
3|1477|Loss: 0.7647092938423157:  23%|██▎       | 1477/6470 [18:10<53:44,  1.55it/s]
3|1478|Loss: 0.8556398153305054:  23%|██▎       | 1477/6470 [18:11<53:44,  1.55it/s]
3|1478|Loss: 0.8556398153305054:  23%|██▎       | 1478/6470 [18:11<55:33,  1.50it/s]
3|1479|Loss: 0.7685946822166443:  23%|██▎       | 1478/6470 [18:11<55:33,  1.50it/s]
3|1479|Loss: 0.7685946822166443:  23%|██▎       | 1479/6470 [18:12<58:10,  1.43it/s]
3|1480|Loss: 0.8084883093833923:  23%|██▎       | 1479/6470 [18:12<58:10,  1.43it/s]
3|1480|Loss: 0.8084883093833923:  23%|██▎       | 1480/6470 [18:12<59:31,  1.40it/s]
3|1481|Loss: 0.8916311860084534:  23%|██▎       | 1480/6470 [18:13<59:31,  1.40it/s]
3|1481|Loss: 0.8916311860084534:  23%|██▎       | 1481/6470 [18:13<1:01:16,  1.36it/s]
3|1482|Loss: 0.9458895921707153:  23%|██▎       | 1481/6470 [18:14<1:01:16,  1.36it/s]
3|1482|Loss: 0.9458895921707153:  23%|██▎       | 1482/6470 [18:14<1:02:41,  1.33it/s]
3|1483|Loss: 0.913227379322052:  23%|██▎       | 1482/6470 [18:14<1:02:41,  1.33it/s] 
3|1483|Loss: 0.913227379322052:  23%|██▎       | 1483/6470 [18:15<1:03:39,  1.31it/s]
3|1484|Loss: 0.7772084474563599:  23%|██▎       | 1483/6470 [18:15<1:03:39,  1.31it/s]
3|1484|Loss: 0.7772084474563599:  23%|██▎       | 1484/6470 [18:16<1:02:28,  1.33it/s]
3|1485|Loss: 0.8595863580703735:  23%|██▎       | 1484/6470 [18:16<1:02:28,  1.33it/s]
3|1485|Loss: 0.8595863580703735:  23%|██▎       | 1485/6470 [18:16<1:01:15,  1.36it/s]
3|1486|Loss: 0.7556355595588684:  23%|██▎       | 1485/6470 [18:17<1:01:15,  1.36it/s]
3|1486|Loss: 0.7556355595588684:  23%|██▎       | 1486/6470 [18:17<1:01:17,  1.36it/s]
3|1487|Loss: 0.9634928703308105:  23%|██▎       | 1486/6470 [18:17<1:01:17,  1.36it/s]
3|1487|Loss: 0.9634928703308105:  23%|██▎       | 1487/6470 [18:18<1:01:47,  1.34it/s]
3|1488|Loss: 0.9710902571678162:  23%|██▎       | 1487/6470 [18:18<1:01:47,  1.34it/s]
3|1488|Loss: 0.9710902571678162:  23%|██▎       | 1488/6470 [18:19<1:02:15,  1.33it/s]
3|1489|Loss: 0.74100661277771:  23%|██▎       | 1488/6470 [18:19<1:02:15,  1.33it/s]  
3|1489|Loss: 0.74100661277771:  23%|██▎       | 1489/6470 [18:19<1:02:29,  1.33it/s]
3|1490|Loss: 0.6379817724227905:  23%|██▎       | 1489/6470 [18:20<1:02:29,  1.33it/s]
3|1490|Loss: 0.6379817724227905:  23%|██▎       | 1490/6470 [18:20<1:08:36,  1.21it/s]
3|1491|Loss: 0.6972280144691467:  23%|██▎       | 1490/6470 [18:21<1:08:36,  1.21it/s]
3|1491|Loss: 0.6972280144691467:  23%|██▎       | 1491/6470 [18:21<1:06:25,  1.25it/s]
3|1492|Loss: 0.9068336486816406:  23%|██▎       | 1491/6470 [18:21<1:06:25,  1.25it/s]
3|1492|Loss: 0.9068336486816406:  23%|██▎       | 1492/6470 [18:21<57:14,  1.45it/s]  
3|1493|Loss: 0.9219580292701721:  23%|██▎       | 1492/6470 [18:22<57:14,  1.45it/s]
3|1493|Loss: 0.9219580292701721:  23%|██▎       | 1493/6470 [18:22<53:18,  1.56it/s]
3|1494|Loss: 0.8334862589836121:  23%|██▎       | 1493/6470 [18:22<53:18,  1.56it/s]
3|1494|Loss: 0.8334862589836121:  23%|██▎       | 1494/6470 [18:23<59:56,  1.38it/s]
3|1495|Loss: 0.8569884896278381:  23%|██▎       | 1494/6470 [18:23<59:56,  1.38it/s]
3|1495|Loss: 0.8569884896278381:  23%|██▎       | 1495/6470 [18:24<1:04:28,  1.29it/s]
3|1496|Loss: 0.8773919939994812:  23%|██▎       | 1495/6470 [18:24<1:04:28,  1.29it/s]
3|1496|Loss: 0.8773919939994812:  23%|██▎       | 1496/6470 [18:25<1:04:17,  1.29it/s]
3|1497|Loss: 0.9288284182548523:  23%|██▎       | 1496/6470 [18:25<1:04:17,  1.29it/s]
3|1497|Loss: 0.9288284182548523:  23%|██▎       | 1497/6470 [18:25<1:06:23,  1.25it/s]
3|1498|Loss: 0.8374003767967224:  23%|██▎       | 1497/6470 [18:26<1:06:23,  1.25it/s]
3|1498|Loss: 0.8374003767967224:  23%|██▎       | 1498/6470 [18:26<1:00:20,  1.37it/s]
3|1499|Loss: 0.7876762747764587:  23%|██▎       | 1498/6470 [18:26<1:00:20,  1.37it/s]
3|1499|Loss: 0.7876762747764587:  23%|██▎       | 1499/6470 [18:27<57:33,  1.44it/s]  
3|1500|Loss: 0.7156007885932922:  23%|██▎       | 1499/6470 [18:27<57:33,  1.44it/s]
3|1500|Loss: 0.7156007885932922:  23%|██▎       | 1500/6470 [18:27<59:21,  1.40it/s]
3|1501|Loss: 0.8307494521141052:  23%|██▎       | 1500/6470 [18:28<59:21,  1.40it/s]
3|1501|Loss: 0.8307494521141052:  23%|██▎       | 1501/6470 [18:28<59:20,  1.40it/s]
3|1502|Loss: 0.7503796219825745:  23%|██▎       | 1501/6470 [18:29<59:20,  1.40it/s]
3|1502|Loss: 0.7503796219825745:  23%|██▎       | 1502/6470 [18:29<1:01:07,  1.35it/s]
3|1503|Loss: 0.7653486132621765:  23%|██▎       | 1502/6470 [18:29<1:01:07,  1.35it/s]
3|1503|Loss: 0.7653486132621765:  23%|██▎       | 1503/6470 [18:30<1:02:41,  1.32it/s]
3|1504|Loss: 0.7663611173629761:  23%|██▎       | 1503/6470 [18:30<1:02:41,  1.32it/s]
3|1504|Loss: 0.7663611173629761:  23%|██▎       | 1504/6470 [18:30<1:02:44,  1.32it/s]
3|1505|Loss: 0.8726020455360413:  23%|██▎       | 1504/6470 [18:31<1:02:44,  1.32it/s]
3|1505|Loss: 0.8726020455360413:  23%|██▎       | 1505/6470 [18:31<1:01:43,  1.34it/s]
3|1506|Loss: 0.8688069581985474:  23%|██▎       | 1505/6470 [18:31<1:01:43,  1.34it/s]
3|1506|Loss: 0.8688069581985474:  23%|██▎       | 1506/6470 [18:32<56:50,  1.46it/s]  
3|1507|Loss: 0.869245707988739:  23%|██▎       | 1506/6470 [18:32<56:50,  1.46it/s] 
3|1507|Loss: 0.869245707988739:  23%|██▎       | 1507/6470 [18:33<1:01:56,  1.34it/s]
3|1508|Loss: 0.9037553071975708:  23%|██▎       | 1507/6470 [18:33<1:01:56,  1.34it/s]
3|1508|Loss: 0.9037553071975708:  23%|██▎       | 1508/6470 [18:33<1:05:27,  1.26it/s]
3|1509|Loss: 0.8485925793647766:  23%|██▎       | 1508/6470 [18:34<1:05:27,  1.26it/s]
3|1509|Loss: 0.8485925793647766:  23%|██▎       | 1509/6470 [18:34<1:05:27,  1.26it/s]
3|1510|Loss: 0.7832071185112:  23%|██▎       | 1509/6470 [18:35<1:05:27,  1.26it/s]   
3|1510|Loss: 0.7832071185112:  23%|██▎       | 1510/6470 [18:35<1:03:08,  1.31it/s]
3|1511|Loss: 0.8582927584648132:  23%|██▎       | 1510/6470 [18:35<1:03:08,  1.31it/s]
3|1511|Loss: 0.8582927584648132:  23%|██▎       | 1511/6470 [18:36<1:02:33,  1.32it/s]
3|1512|Loss: 0.8916505575180054:  23%|██▎       | 1511/6470 [18:36<1:02:33,  1.32it/s]
3|1512|Loss: 0.8916505575180054:  23%|██▎       | 1512/6470 [18:36<59:47,  1.38it/s]  
3|1513|Loss: 0.7831754088401794:  23%|██▎       | 1512/6470 [18:37<59:47,  1.38it/s]
3|1513|Loss: 0.7831754088401794:  23%|██▎       | 1513/6470 [18:37<54:06,  1.53it/s]
3|1514|Loss: 0.820540726184845:  23%|██▎       | 1513/6470 [18:37<54:06,  1.53it/s] 
3|1514|Loss: 0.820540726184845:  23%|██▎       | 1514/6470 [18:38<55:47,  1.48it/s]
3|1515|Loss: 0.8381527662277222:  23%|██▎       | 1514/6470 [18:38<55:47,  1.48it/s]
3|1515|Loss: 0.8381527662277222:  23%|██▎       | 1515/6470 [18:38<56:22,  1.46it/s]
3|1516|Loss: 0.8094865083694458:  23%|██▎       | 1515/6470 [18:39<56:22,  1.46it/s]
3|1516|Loss: 0.8094865083694458:  23%|██▎       | 1516/6470 [18:39<56:07,  1.47it/s]
3|1517|Loss: 0.8820148706436157:  23%|██▎       | 1516/6470 [18:39<56:07,  1.47it/s]
3|1517|Loss: 0.8820148706436157:  23%|██▎       | 1517/6470 [18:40<58:01,  1.42it/s]
3|1518|Loss: 0.7776703238487244:  23%|██▎       | 1517/6470 [18:40<58:01,  1.42it/s]
3|1518|Loss: 0.7776703238487244:  23%|██▎       | 1518/6470 [18:40<59:34,  1.39it/s]
3|1519|Loss: 0.8105698227882385:  23%|██▎       | 1518/6470 [18:41<59:34,  1.39it/s]
3|1519|Loss: 0.8105698227882385:  23%|██▎       | 1519/6470 [18:41<59:05,  1.40it/s]
3|1520|Loss: 0.8542165756225586:  23%|██▎       | 1519/6470 [18:42<59:05,  1.40it/s]
3|1520|Loss: 0.8542165756225586:  23%|██▎       | 1520/6470 [18:42<1:01:42,  1.34it/s]
3|1521|Loss: 0.6354953050613403:  23%|██▎       | 1520/6470 [18:42<1:01:42,  1.34it/s]
3|1521|Loss: 0.6354953050613403:  24%|██▎       | 1521/6470 [18:43<58:46,  1.40it/s]  
3|1522|Loss: 0.7211107611656189:  24%|██▎       | 1521/6470 [18:43<58:46,  1.40it/s]
3|1522|Loss: 0.7211107611656189:  24%|██▎       | 1522/6470 [18:43<55:24,  1.49it/s]
3|1523|Loss: 0.9499399065971375:  24%|██▎       | 1522/6470 [18:44<55:24,  1.49it/s]
3|1523|Loss: 0.9499399065971375:  24%|██▎       | 1523/6470 [18:44<59:32,  1.38it/s]
3|1524|Loss: 0.9603235125541687:  24%|██▎       | 1523/6470 [18:44<59:32,  1.38it/s]
3|1524|Loss: 0.9603235125541687:  24%|██▎       | 1524/6470 [18:45<59:41,  1.38it/s]
3|1525|Loss: 0.8362002372741699:  24%|██▎       | 1524/6470 [18:45<59:41,  1.38it/s]
3|1525|Loss: 0.8362002372741699:  24%|██▎       | 1525/6470 [18:45<59:04,  1.39it/s]
3|1526|Loss: 0.8610684871673584:  24%|██▎       | 1525/6470 [18:46<59:04,  1.39it/s]
3|1526|Loss: 0.8610684871673584:  24%|██▎       | 1526/6470 [18:46<55:58,  1.47it/s]
3|1527|Loss: 0.7950997948646545:  24%|██▎       | 1526/6470 [18:46<55:58,  1.47it/s]
3|1527|Loss: 0.7950997948646545:  24%|██▎       | 1527/6470 [18:47<56:59,  1.45it/s]
3|1528|Loss: 0.9984576106071472:  24%|██▎       | 1527/6470 [18:47<56:59,  1.45it/s]
3|1528|Loss: 0.9984576106071472:  24%|██▎       | 1528/6470 [18:47<55:13,  1.49it/s]
3|1529|Loss: 0.837902307510376:  24%|██▎       | 1528/6470 [18:48<55:13,  1.49it/s] 
3|1529|Loss: 0.837902307510376:  24%|██▎       | 1529/6470 [18:48<56:21,  1.46it/s]
3|1530|Loss: 0.7977970242500305:  24%|██▎       | 1529/6470 [18:49<56:21,  1.46it/s]
3|1530|Loss: 0.7977970242500305:  24%|██▎       | 1530/6470 [18:49<57:10,  1.44it/s]
3|1531|Loss: 0.8244752287864685:  24%|██▎       | 1530/6470 [18:49<57:10,  1.44it/s]
3|1531|Loss: 0.8244752287864685:  24%|██▎       | 1531/6470 [18:50<57:17,  1.44it/s]
3|1532|Loss: 0.8816195726394653:  24%|██▎       | 1531/6470 [18:50<57:17,  1.44it/s]
3|1532|Loss: 0.8816195726394653:  24%|██▎       | 1532/6470 [18:50<1:02:00,  1.33it/s]
3|1533|Loss: 0.8744577765464783:  24%|██▎       | 1532/6470 [18:51<1:02:00,  1.33it/s]
3|1533|Loss: 0.8744577765464783:  24%|██▎       | 1533/6470 [18:51<1:03:37,  1.29it/s]
3|1534|Loss: 0.8378821611404419:  24%|██▎       | 1533/6470 [18:52<1:03:37,  1.29it/s]
3|1534|Loss: 0.8378821611404419:  24%|██▎       | 1534/6470 [18:52<1:03:58,  1.29it/s]
3|1535|Loss: 0.7855908274650574:  24%|██▎       | 1534/6470 [18:52<1:03:58,  1.29it/s]
3|1535|Loss: 0.7855908274650574:  24%|██▎       | 1535/6470 [18:53<1:02:35,  1.31it/s]
3|1536|Loss: 0.8576439023017883:  24%|██▎       | 1535/6470 [18:53<1:02:35,  1.31it/s]
3|1536|Loss: 0.8576439023017883:  24%|██▎       | 1536/6470 [18:53<1:01:55,  1.33it/s]
3|1537|Loss: 0.8447209000587463:  24%|██▎       | 1536/6470 [18:54<1:01:55,  1.33it/s]
3|1537|Loss: 0.8447209000587463:  24%|██▍       | 1537/6470 [18:54<1:04:02,  1.28it/s]
3|1538|Loss: 0.9495707154273987:  24%|██▍       | 1537/6470 [18:55<1:04:02,  1.28it/s]
3|1538|Loss: 0.9495707154273987:  24%|██▍       | 1538/6470 [18:55<1:04:27,  1.28it/s]
3|1539|Loss: 0.7751230001449585:  24%|██▍       | 1538/6470 [18:56<1:04:27,  1.28it/s]
3|1539|Loss: 0.7751230001449585:  24%|██▍       | 1539/6470 [18:56<1:02:44,  1.31it/s]
3|1540|Loss: 1.0280903577804565:  24%|██▍       | 1539/6470 [18:56<1:02:44,  1.31it/s]
3|1540|Loss: 1.0280903577804565:  24%|██▍       | 1540/6470 [18:57<1:04:28,  1.27it/s]
3|1541|Loss: 0.7980169653892517:  24%|██▍       | 1540/6470 [18:57<1:04:28,  1.27it/s]
3|1541|Loss: 0.7980169653892517:  24%|██▍       | 1541/6470 [18:58<1:05:22,  1.26it/s]
3|1542|Loss: 0.994297444820404:  24%|██▍       | 1541/6470 [18:58<1:05:22,  1.26it/s] 
3|1542|Loss: 0.994297444820404:  24%|██▍       | 1542/6470 [18:58<1:04:45,  1.27it/s]
3|1543|Loss: 0.8308044075965881:  24%|██▍       | 1542/6470 [18:59<1:04:45,  1.27it/s]
3|1543|Loss: 0.8308044075965881:  24%|██▍       | 1543/6470 [18:59<1:03:04,  1.30it/s]
3|1544|Loss: 0.7879595160484314:  24%|██▍       | 1543/6470 [18:59<1:03:04,  1.30it/s]
3|1544|Loss: 0.7879595160484314:  24%|██▍       | 1544/6470 [19:00<1:02:36,  1.31it/s]
3|1545|Loss: 0.7214052677154541:  24%|██▍       | 1544/6470 [19:00<1:02:36,  1.31it/s]
3|1545|Loss: 0.7214052677154541:  24%|██▍       | 1545/6470 [19:01<1:03:23,  1.29it/s]
3|1546|Loss: 0.707041323184967:  24%|██▍       | 1545/6470 [19:01<1:03:23,  1.29it/s] 
3|1546|Loss: 0.707041323184967:  24%|██▍       | 1546/6470 [19:01<57:52,  1.42it/s]  
3|1547|Loss: 1.104148507118225:  24%|██▍       | 1546/6470 [19:01<57:52,  1.42it/s]
3|1547|Loss: 1.104148507118225:  24%|██▍       | 1547/6470 [19:02<1:01:02,  1.34it/s]
3|1548|Loss: 0.7341136336326599:  24%|██▍       | 1547/6470 [19:02<1:01:02,  1.34it/s]
3|1548|Loss: 0.7341136336326599:  24%|██▍       | 1548/6470 [19:03<1:02:57,  1.30it/s]
3|1549|Loss: 0.8661251664161682:  24%|██▍       | 1548/6470 [19:03<1:02:57,  1.30it/s]
3|1549|Loss: 0.8661251664161682:  24%|██▍       | 1549/6470 [19:04<1:04:48,  1.27it/s]
3|1550|Loss: 0.8361639976501465:  24%|██▍       | 1549/6470 [19:04<1:04:48,  1.27it/s]
3|1550|Loss: 0.8361639976501465:  24%|██▍       | 1550/6470 [19:04<1:06:56,  1.22it/s]
3|1551|Loss: 0.9287850260734558:  24%|██▍       | 1550/6470 [19:05<1:06:56,  1.22it/s]
3|1551|Loss: 0.9287850260734558:  24%|██▍       | 1551/6470 [19:05<1:06:41,  1.23it/s]
3|1552|Loss: 0.7861996293067932:  24%|██▍       | 1551/6470 [19:06<1:06:41,  1.23it/s]
3|1552|Loss: 0.7861996293067932:  24%|██▍       | 1552/6470 [19:06<1:03:29,  1.29it/s]
3|1553|Loss: 0.8553085327148438:  24%|██▍       | 1552/6470 [19:06<1:03:29,  1.29it/s]
3|1553|Loss: 0.8553085327148438:  24%|██▍       | 1553/6470 [19:07<1:02:19,  1.31it/s]
3|1554|Loss: 0.8797890543937683:  24%|██▍       | 1553/6470 [19:07<1:02:19,  1.31it/s]
3|1554|Loss: 0.8797890543937683:  24%|██▍       | 1554/6470 [19:07<1:01:52,  1.32it/s]
3|1555|Loss: 0.7917250990867615:  24%|██▍       | 1554/6470 [19:08<1:01:52,  1.32it/s]
3|1555|Loss: 0.7917250990867615:  24%|██▍       | 1555/6470 [19:08<1:04:56,  1.26it/s]
3|1556|Loss: 0.8724794983863831:  24%|██▍       | 1555/6470 [19:09<1:04:56,  1.26it/s]
3|1556|Loss: 0.8724794983863831:  24%|██▍       | 1556/6470 [19:09<1:04:47,  1.26it/s]
3|1557|Loss: 0.937393307685852:  24%|██▍       | 1556/6470 [19:09<1:04:47,  1.26it/s] 
3|1557|Loss: 0.937393307685852:  24%|██▍       | 1557/6470 [19:10<1:02:42,  1.31it/s]
3|1558|Loss: 0.8042606115341187:  24%|██▍       | 1557/6470 [19:10<1:02:42,  1.31it/s]
3|1558|Loss: 0.8042606115341187:  24%|██▍       | 1558/6470 [19:10<57:46,  1.42it/s]  
3|1559|Loss: 0.9157713055610657:  24%|██▍       | 1558/6470 [19:11<57:46,  1.42it/s]
3|1559|Loss: 0.9157713055610657:  24%|██▍       | 1559/6470 [19:11<57:45,  1.42it/s]
3|1560|Loss: 0.7267956733703613:  24%|██▍       | 1559/6470 [19:11<57:45,  1.42it/s]
3|1560|Loss: 0.7267956733703613:  24%|██▍       | 1560/6470 [19:12<59:00,  1.39it/s]
3|1561|Loss: 0.8472699522972107:  24%|██▍       | 1560/6470 [19:12<59:00,  1.39it/s]
3|1561|Loss: 0.8472699522972107:  24%|██▍       | 1561/6470 [19:13<1:00:34,  1.35it/s]
3|1562|Loss: 0.8357716202735901:  24%|██▍       | 1561/6470 [19:13<1:00:34,  1.35it/s]
3|1562|Loss: 0.8357716202735901:  24%|██▍       | 1562/6470 [19:13<55:50,  1.47it/s]  
3|1563|Loss: 0.8252715468406677:  24%|██▍       | 1562/6470 [19:14<55:50,  1.47it/s]
3|1563|Loss: 0.8252715468406677:  24%|██▍       | 1563/6470 [19:14<58:12,  1.41it/s]
3|1564|Loss: 0.8303757309913635:  24%|██▍       | 1563/6470 [19:14<58:12,  1.41it/s]
3|1564|Loss: 0.8303757309913635:  24%|██▍       | 1564/6470 [19:15<58:42,  1.39it/s]
3|1565|Loss: 0.6608918905258179:  24%|██▍       | 1564/6470 [19:15<58:42,  1.39it/s]
3|1565|Loss: 0.6608918905258179:  24%|██▍       | 1565/6470 [19:15<57:43,  1.42it/s]
3|1566|Loss: 0.8012561202049255:  24%|██▍       | 1565/6470 [19:16<57:43,  1.42it/s]
3|1566|Loss: 0.8012561202049255:  24%|██▍       | 1566/6470 [19:16<58:17,  1.40it/s]
3|1567|Loss: 0.7775366306304932:  24%|██▍       | 1566/6470 [19:17<58:17,  1.40it/s]
3|1567|Loss: 0.7775366306304932:  24%|██▍       | 1567/6470 [19:17<59:46,  1.37it/s]
3|1568|Loss: 0.936310350894928:  24%|██▍       | 1567/6470 [19:17<59:46,  1.37it/s] 
3|1568|Loss: 0.936310350894928:  24%|██▍       | 1568/6470 [19:18<1:01:24,  1.33it/s]
3|1569|Loss: 0.878287136554718:  24%|██▍       | 1568/6470 [19:18<1:01:24,  1.33it/s]
3|1569|Loss: 0.878287136554718:  24%|██▍       | 1569/6470 [19:18<1:02:27,  1.31it/s]
3|1570|Loss: 0.8301979303359985:  24%|██▍       | 1569/6470 [19:19<1:02:27,  1.31it/s]
3|1570|Loss: 0.8301979303359985:  24%|██▍       | 1570/6470 [19:19<1:04:35,  1.26it/s]
3|1571|Loss: 0.43778684735298157:  24%|██▍       | 1570/6470 [19:20<1:04:35,  1.26it/s]
3|1571|Loss: 0.43778684735298157:  24%|██▍       | 1571/6470 [19:20<1:04:40,  1.26it/s]
3|1572|Loss: 0.9540342688560486:  24%|██▍       | 1571/6470 [19:20<1:04:40,  1.26it/s] 
3|1572|Loss: 0.9540342688560486:  24%|██▍       | 1572/6470 [19:21<1:00:45,  1.34it/s]
3|1573|Loss: 0.6913788318634033:  24%|██▍       | 1572/6470 [19:21<1:00:45,  1.34it/s]
3|1573|Loss: 0.6913788318634033:  24%|██▍       | 1573/6470 [19:21<1:00:00,  1.36it/s]
3|1574|Loss: 0.9154117107391357:  24%|██▍       | 1573/6470 [19:22<1:00:00,  1.36it/s]
3|1574|Loss: 0.9154117107391357:  24%|██▍       | 1574/6470 [19:22<59:19,  1.38it/s]  
3|1575|Loss: 0.8327964544296265:  24%|██▍       | 1574/6470 [19:23<59:19,  1.38it/s]
3|1575|Loss: 0.8327964544296265:  24%|██▍       | 1575/6470 [19:23<1:00:56,  1.34it/s]
3|1576|Loss: 0.7773796319961548:  24%|██▍       | 1575/6470 [19:23<1:00:56,  1.34it/s]
3|1576|Loss: 0.7773796319961548:  24%|██▍       | 1576/6470 [19:24<1:02:11,  1.31it/s]
3|1577|Loss: 0.937790036201477:  24%|██▍       | 1576/6470 [19:24<1:02:11,  1.31it/s] 
3|1577|Loss: 0.937790036201477:  24%|██▍       | 1577/6470 [19:24<58:26,  1.40it/s]  
3|1578|Loss: 0.974865198135376:  24%|██▍       | 1577/6470 [19:25<58:26,  1.40it/s]
3|1578|Loss: 0.974865198135376:  24%|██▍       | 1578/6470 [19:25<59:10,  1.38it/s]
3|1579|Loss: 0.7093689441680908:  24%|██▍       | 1578/6470 [19:25<59:10,  1.38it/s]
3|1579|Loss: 0.7093689441680908:  24%|██▍       | 1579/6470 [19:26<55:39,  1.46it/s]
3|1580|Loss: 0.753934383392334:  24%|██▍       | 1579/6470 [19:26<55:39,  1.46it/s] 
3|1580|Loss: 0.753934383392334:  24%|██▍       | 1580/6470 [19:26<53:45,  1.52it/s]
3|1581|Loss: 0.7608894109725952:  24%|██▍       | 1580/6470 [19:27<53:45,  1.52it/s]
3|1581|Loss: 0.7608894109725952:  24%|██▍       | 1581/6470 [19:27<56:50,  1.43it/s]
3|1582|Loss: 0.8694273233413696:  24%|██▍       | 1581/6470 [19:27<56:50,  1.43it/s]
3|1582|Loss: 0.8694273233413696:  24%|██▍       | 1582/6470 [19:28<54:49,  1.49it/s]
3|1583|Loss: 0.7448226809501648:  24%|██▍       | 1582/6470 [19:28<54:49,  1.49it/s]
3|1583|Loss: 0.7448226809501648:  24%|██▍       | 1583/6470 [19:28<53:56,  1.51it/s]
3|1584|Loss: 0.676720142364502:  24%|██▍       | 1583/6470 [19:29<53:56,  1.51it/s] 
3|1584|Loss: 0.676720142364502:  24%|██▍       | 1584/6470 [19:29<59:22,  1.37it/s]
3|1585|Loss: 0.819147527217865:  24%|██▍       | 1584/6470 [19:30<59:22,  1.37it/s]
3|1585|Loss: 0.819147527217865:  24%|██▍       | 1585/6470 [19:30<1:02:52,  1.29it/s]
3|1586|Loss: 0.8716408014297485:  24%|██▍       | 1585/6470 [19:31<1:02:52,  1.29it/s]
3|1586|Loss: 0.8716408014297485:  25%|██▍       | 1586/6470 [19:31<1:01:23,  1.33it/s]
3|1587|Loss: 0.8224003314971924:  25%|██▍       | 1586/6470 [19:31<1:01:23,  1.33it/s]
3|1587|Loss: 0.8224003314971924:  25%|██▍       | 1587/6470 [19:31<58:33,  1.39it/s]  
3|1588|Loss: 0.765737771987915:  25%|██▍       | 1587/6470 [19:32<58:33,  1.39it/s] 
3|1588|Loss: 0.765737771987915:  25%|██▍       | 1588/6470 [19:32<57:48,  1.41it/s]
3|1589|Loss: 0.8036944270133972:  25%|██▍       | 1588/6470 [19:33<57:48,  1.41it/s]
3|1589|Loss: 0.8036944270133972:  25%|██▍       | 1589/6470 [19:33<57:38,  1.41it/s]
3|1590|Loss: 0.9002576470375061:  25%|██▍       | 1589/6470 [19:33<57:38,  1.41it/s]
3|1590|Loss: 0.9002576470375061:  25%|██▍       | 1590/6470 [19:34<58:57,  1.38it/s]
3|1591|Loss: 0.8967432975769043:  25%|██▍       | 1590/6470 [19:34<58:57,  1.38it/s]
3|1591|Loss: 0.8967432975769043:  25%|██▍       | 1591/6470 [19:34<57:06,  1.42it/s]
3|1592|Loss: 0.6620479822158813:  25%|██▍       | 1591/6470 [19:35<57:06,  1.42it/s]
3|1592|Loss: 0.6620479822158813:  25%|██▍       | 1592/6470 [19:35<57:28,  1.41it/s]
3|1593|Loss: 0.8759822845458984:  25%|██▍       | 1592/6470 [19:35<57:28,  1.41it/s]
3|1593|Loss: 0.8759822845458984:  25%|██▍       | 1593/6470 [19:36<55:17,  1.47it/s]
3|1594|Loss: 0.5868711471557617:  25%|██▍       | 1593/6470 [19:36<55:17,  1.47it/s]
3|1594|Loss: 0.5868711471557617:  25%|██▍       | 1594/6470 [19:36<54:19,  1.50it/s]
3|1595|Loss: 0.8074535727500916:  25%|██▍       | 1594/6470 [19:37<54:19,  1.50it/s]
3|1595|Loss: 0.8074535727500916:  25%|██▍       | 1595/6470 [19:37<56:56,  1.43it/s]
3|1596|Loss: 0.9618681073188782:  25%|██▍       | 1595/6470 [19:37<56:56,  1.43it/s]
3|1596|Loss: 0.9618681073188782:  25%|██▍       | 1596/6470 [19:38<56:08,  1.45it/s]
3|1597|Loss: 0.8247718214988708:  25%|██▍       | 1596/6470 [19:38<56:08,  1.45it/s]
3|1597|Loss: 0.8247718214988708:  25%|██▍       | 1597/6470 [19:38<57:59,  1.40it/s]
3|1598|Loss: 0.8937693238258362:  25%|██▍       | 1597/6470 [19:39<57:59,  1.40it/s]
3|1598|Loss: 0.8937693238258362:  25%|██▍       | 1598/6470 [19:39<59:34,  1.36it/s]
3|1599|Loss: 0.7645828127861023:  25%|██▍       | 1598/6470 [19:40<59:34,  1.36it/s]
3|1599|Loss: 0.7645828127861023:  25%|██▍       | 1599/6470 [19:40<1:00:56,  1.33it/s]
3|1600|Loss: 0.6779964566230774:  25%|██▍       | 1599/6470 [19:40<1:00:56,  1.33it/s]
3|1600|Loss: 0.6779964566230774:  25%|██▍       | 1600/6470 [19:41<59:29,  1.36it/s]  
3|1601|Loss: 0.9236870408058167:  25%|██▍       | 1600/6470 [19:41<59:29,  1.36it/s]
3|1601|Loss: 0.9236870408058167:  25%|██▍       | 1601/6470 [19:41<1:00:12,  1.35it/s]
3|1602|Loss: 0.869642972946167:  25%|██▍       | 1601/6470 [19:42<1:00:12,  1.35it/s] 
3|1602|Loss: 0.869642972946167:  25%|██▍       | 1602/6470 [19:42<1:00:27,  1.34it/s]
3|1603|Loss: 0.7555830478668213:  25%|██▍       | 1602/6470 [19:43<1:00:27,  1.34it/s]
3|1603|Loss: 0.7555830478668213:  25%|██▍       | 1603/6470 [19:43<1:04:34,  1.26it/s]
3|1604|Loss: 0.8480413556098938:  25%|██▍       | 1603/6470 [19:44<1:04:34,  1.26it/s]
3|1604|Loss: 0.8480413556098938:  25%|██▍       | 1604/6470 [19:44<1:02:34,  1.30it/s]
3|1605|Loss: 0.8032702207565308:  25%|██▍       | 1604/6470 [19:44<1:02:34,  1.30it/s]
3|1605|Loss: 0.8032702207565308:  25%|██▍       | 1605/6470 [19:45<1:02:05,  1.31it/s]
3|1606|Loss: 0.905485987663269:  25%|██▍       | 1605/6470 [19:45<1:02:05,  1.31it/s] 
3|1606|Loss: 0.905485987663269:  25%|██▍       | 1606/6470 [19:45<1:02:01,  1.31it/s]
3|1607|Loss: 0.739129900932312:  25%|██▍       | 1606/6470 [19:46<1:02:01,  1.31it/s]
3|1607|Loss: 0.739129900932312:  25%|██▍       | 1607/6470 [19:46<1:04:39,  1.25it/s]
3|1608|Loss: 0.87786865234375:  25%|██▍       | 1607/6470 [19:47<1:04:39,  1.25it/s] 
3|1608|Loss: 0.87786865234375:  25%|██▍       | 1608/6470 [19:47<1:07:45,  1.20it/s]
3|1609|Loss: 0.7467290163040161:  25%|██▍       | 1608/6470 [19:48<1:07:45,  1.20it/s]
3|1609|Loss: 0.7467290163040161:  25%|██▍       | 1609/6470 [19:48<1:10:26,  1.15it/s]
3|1610|Loss: 0.7361202239990234:  25%|██▍       | 1609/6470 [19:49<1:10:26,  1.15it/s]
3|1610|Loss: 0.7361202239990234:  25%|██▍       | 1610/6470 [19:49<1:09:32,  1.16it/s]
3|1611|Loss: 0.8960348963737488:  25%|██▍       | 1610/6470 [19:49<1:09:32,  1.16it/s]
3|1611|Loss: 0.8960348963737488:  25%|██▍       | 1611/6470 [19:50<1:08:27,  1.18it/s]
3|1612|Loss: 0.7866918444633484:  25%|██▍       | 1611/6470 [19:50<1:08:27,  1.18it/s]
3|1612|Loss: 0.7866918444633484:  25%|██▍       | 1612/6470 [19:50<1:05:44,  1.23it/s]
3|1613|Loss: 0.8622388243675232:  25%|██▍       | 1612/6470 [19:51<1:05:44,  1.23it/s]
3|1613|Loss: 0.8622388243675232:  25%|██▍       | 1613/6470 [19:51<1:07:56,  1.19it/s]
3|1614|Loss: 0.841843843460083:  25%|██▍       | 1613/6470 [19:52<1:07:56,  1.19it/s] 
3|1614|Loss: 0.841843843460083:  25%|██▍       | 1614/6470 [19:52<1:06:02,  1.23it/s]
3|1615|Loss: 0.9238083958625793:  25%|██▍       | 1614/6470 [19:52<1:06:02,  1.23it/s]
3|1615|Loss: 0.9238083958625793:  25%|██▍       | 1615/6470 [19:53<58:47,  1.38it/s]  
3|1616|Loss: 0.8203067779541016:  25%|██▍       | 1615/6470 [19:53<58:47,  1.38it/s]
3|1616|Loss: 0.8203067779541016:  25%|██▍       | 1616/6470 [19:53<58:36,  1.38it/s]
3|1617|Loss: 0.8389419913291931:  25%|██▍       | 1616/6470 [19:54<58:36,  1.38it/s]
3|1617|Loss: 0.8389419913291931:  25%|██▍       | 1617/6470 [19:54<59:40,  1.36it/s]
3|1618|Loss: 0.7397834658622742:  25%|██▍       | 1617/6470 [19:55<59:40,  1.36it/s]
3|1618|Loss: 0.7397834658622742:  25%|██▌       | 1618/6470 [19:55<1:02:51,  1.29it/s]
3|1619|Loss: 0.7797613739967346:  25%|██▌       | 1618/6470 [19:56<1:02:51,  1.29it/s]
3|1619|Loss: 0.7797613739967346:  25%|██▌       | 1619/6470 [19:56<1:05:56,  1.23it/s]
3|1620|Loss: 0.790357768535614:  25%|██▌       | 1619/6470 [19:56<1:05:56,  1.23it/s] 
3|1620|Loss: 0.790357768535614:  25%|██▌       | 1620/6470 [19:57<1:04:01,  1.26it/s]
3|1621|Loss: 0.9055255055427551:  25%|██▌       | 1620/6470 [19:57<1:04:01,  1.26it/s]
3|1621|Loss: 0.9055255055427551:  25%|██▌       | 1621/6470 [19:57<1:03:08,  1.28it/s]
3|1622|Loss: 0.8763360381126404:  25%|██▌       | 1621/6470 [19:58<1:03:08,  1.28it/s]
3|1622|Loss: 0.8763360381126404:  25%|██▌       | 1622/6470 [19:58<1:06:05,  1.22it/s]
3|1623|Loss: 0.8798922300338745:  25%|██▌       | 1622/6470 [19:59<1:06:05,  1.22it/s]
3|1623|Loss: 0.8798922300338745:  25%|██▌       | 1623/6470 [19:59<1:05:37,  1.23it/s]
3|1624|Loss: 0.7857105731964111:  25%|██▌       | 1623/6470 [20:00<1:05:37,  1.23it/s]
3|1624|Loss: 0.7857105731964111:  25%|██▌       | 1624/6470 [20:00<1:09:12,  1.17it/s]
3|1625|Loss: 0.7466729283332825:  25%|██▌       | 1624/6470 [20:01<1:09:12,  1.17it/s]
3|1625|Loss: 0.7466729283332825:  25%|██▌       | 1625/6470 [20:01<1:07:05,  1.20it/s]
3|1626|Loss: 0.8131334781646729:  25%|██▌       | 1625/6470 [20:01<1:07:05,  1.20it/s]
3|1626|Loss: 0.8131334781646729:  25%|██▌       | 1626/6470 [20:02<1:04:05,  1.26it/s]
3|1627|Loss: 0.8972567319869995:  25%|██▌       | 1626/6470 [20:02<1:04:05,  1.26it/s]
3|1627|Loss: 0.8972567319869995:  25%|██▌       | 1627/6470 [20:02<58:06,  1.39it/s]  
3|1628|Loss: 0.8591377139091492:  25%|██▌       | 1627/6470 [20:03<58:06,  1.39it/s]
3|1628|Loss: 0.8591377139091492:  25%|██▌       | 1628/6470 [20:03<59:37,  1.35it/s]
3|1629|Loss: 0.9606297016143799:  25%|██▌       | 1628/6470 [20:03<59:37,  1.35it/s]
3|1629|Loss: 0.9606297016143799:  25%|██▌       | 1629/6470 [20:04<59:33,  1.35it/s]
3|1630|Loss: 0.8466950058937073:  25%|██▌       | 1629/6470 [20:04<59:33,  1.35it/s]
3|1630|Loss: 0.8466950058937073:  25%|██▌       | 1630/6470 [20:04<1:00:32,  1.33it/s]
3|1631|Loss: 0.8728880882263184:  25%|██▌       | 1630/6470 [20:05<1:00:32,  1.33it/s]
3|1631|Loss: 0.8728880882263184:  25%|██▌       | 1631/6470 [20:05<1:00:28,  1.33it/s]
3|1632|Loss: 0.6083188652992249:  25%|██▌       | 1631/6470 [20:06<1:00:28,  1.33it/s]
3|1632|Loss: 0.6083188652992249:  25%|██▌       | 1632/6470 [20:06<1:02:44,  1.29it/s]
3|1633|Loss: 0.8626853823661804:  25%|██▌       | 1632/6470 [20:06<1:02:44,  1.29it/s]
3|1633|Loss: 0.8626853823661804:  25%|██▌       | 1633/6470 [20:07<1:02:33,  1.29it/s]
3|1634|Loss: 0.8297733068466187:  25%|██▌       | 1633/6470 [20:07<1:02:33,  1.29it/s]
3|1634|Loss: 0.8297733068466187:  25%|██▌       | 1634/6470 [20:08<1:02:36,  1.29it/s]
3|1635|Loss: 0.9265272617340088:  25%|██▌       | 1634/6470 [20:08<1:02:36,  1.29it/s]
3|1635|Loss: 0.9265272617340088:  25%|██▌       | 1635/6470 [20:08<1:02:13,  1.30it/s]
3|1636|Loss: 0.8468879461288452:  25%|██▌       | 1635/6470 [20:09<1:02:13,  1.30it/s]
3|1636|Loss: 0.8468879461288452:  25%|██▌       | 1636/6470 [20:09<1:01:44,  1.31it/s]
3|1637|Loss: 0.8048210740089417:  25%|██▌       | 1636/6470 [20:09<1:01:44,  1.31it/s]
3|1637|Loss: 0.8048210740089417:  25%|██▌       | 1637/6470 [20:10<1:00:05,  1.34it/s]
3|1638|Loss: 0.8945274353027344:  25%|██▌       | 1637/6470 [20:10<1:00:05,  1.34it/s]
3|1638|Loss: 0.8945274353027344:  25%|██▌       | 1638/6470 [20:10<59:47,  1.35it/s]  
3|1639|Loss: 0.920628011226654:  25%|██▌       | 1638/6470 [20:11<59:47,  1.35it/s] 
3|1639|Loss: 0.920628011226654:  25%|██▌       | 1639/6470 [20:11<1:00:14,  1.34it/s]
3|1640|Loss: 0.8105762600898743:  25%|██▌       | 1639/6470 [20:12<1:00:14,  1.34it/s]
3|1640|Loss: 0.8105762600898743:  25%|██▌       | 1640/6470 [20:12<1:03:07,  1.28it/s]
3|1641|Loss: 0.7952606081962585:  25%|██▌       | 1640/6470 [20:13<1:03:07,  1.28it/s]
3|1641|Loss: 0.7952606081962585:  25%|██▌       | 1641/6470 [20:13<1:01:25,  1.31it/s]
3|1642|Loss: 0.7514867186546326:  25%|██▌       | 1641/6470 [20:13<1:01:25,  1.31it/s]
3|1642|Loss: 0.7514867186546326:  25%|██▌       | 1642/6470 [20:14<1:01:27,  1.31it/s]
3|1643|Loss: 0.7682965397834778:  25%|██▌       | 1642/6470 [20:14<1:01:27,  1.31it/s]
3|1643|Loss: 0.7682965397834778:  25%|██▌       | 1643/6470 [20:14<1:04:20,  1.25it/s]
3|1644|Loss: 0.8154851198196411:  25%|██▌       | 1643/6470 [20:15<1:04:20,  1.25it/s]
3|1644|Loss: 0.8154851198196411:  25%|██▌       | 1644/6470 [20:15<1:04:57,  1.24it/s]
3|1645|Loss: 0.7567628622055054:  25%|██▌       | 1644/6470 [20:16<1:04:57,  1.24it/s]
3|1645|Loss: 0.7567628622055054:  25%|██▌       | 1645/6470 [20:16<1:05:24,  1.23it/s]
3|1646|Loss: 0.8460280299186707:  25%|██▌       | 1645/6470 [20:17<1:05:24,  1.23it/s]
3|1646|Loss: 0.8460280299186707:  25%|██▌       | 1646/6470 [20:17<1:04:52,  1.24it/s]
3|1647|Loss: 0.9803836345672607:  25%|██▌       | 1646/6470 [20:17<1:04:52,  1.24it/s]
3|1647|Loss: 0.9803836345672607:  25%|██▌       | 1647/6470 [20:18<1:05:01,  1.24it/s]
3|1648|Loss: 0.8515159487724304:  25%|██▌       | 1647/6470 [20:18<1:05:01,  1.24it/s]
3|1648|Loss: 0.8515159487724304:  25%|██▌       | 1648/6470 [20:18<1:02:32,  1.29it/s]
3|1649|Loss: 0.8022530674934387:  25%|██▌       | 1648/6470 [20:19<1:02:32,  1.29it/s]
3|1649|Loss: 0.8022530674934387:  25%|██▌       | 1649/6470 [20:19<1:01:50,  1.30it/s]
3|1650|Loss: 0.7477137446403503:  25%|██▌       | 1649/6470 [20:20<1:01:50,  1.30it/s]
3|1650|Loss: 0.7477137446403503:  26%|██▌       | 1650/6470 [20:20<59:10,  1.36it/s]  
3|1651|Loss: 0.7518181204795837:  26%|██▌       | 1650/6470 [20:20<59:10,  1.36it/s]
3|1651|Loss: 0.7518181204795837:  26%|██▌       | 1651/6470 [20:20<56:28,  1.42it/s]
3|1652|Loss: 0.8006420135498047:  26%|██▌       | 1651/6470 [20:21<56:28,  1.42it/s]
3|1652|Loss: 0.8006420135498047:  26%|██▌       | 1652/6470 [20:21<1:00:03,  1.34it/s]
3|1653|Loss: 0.8104369044303894:  26%|██▌       | 1652/6470 [20:22<1:00:03,  1.34it/s]
3|1653|Loss: 0.8104369044303894:  26%|██▌       | 1653/6470 [20:22<59:39,  1.35it/s]  
3|1654|Loss: 0.7771050930023193:  26%|██▌       | 1653/6470 [20:23<59:39,  1.35it/s]
3|1654|Loss: 0.7771050930023193:  26%|██▌       | 1654/6470 [20:23<1:03:12,  1.27it/s]
3|1655|Loss: 0.8549344539642334:  26%|██▌       | 1654/6470 [20:23<1:03:12,  1.27it/s]
3|1655|Loss: 0.8549344539642334:  26%|██▌       | 1655/6470 [20:24<1:03:29,  1.26it/s]
3|1656|Loss: 0.7827569246292114:  26%|██▌       | 1655/6470 [20:24<1:03:29,  1.26it/s]
3|1656|Loss: 0.7827569246292114:  26%|██▌       | 1656/6470 [20:25<1:06:17,  1.21it/s]
3|1657|Loss: 0.7369673252105713:  26%|██▌       | 1656/6470 [20:25<1:06:17,  1.21it/s]
3|1657|Loss: 0.7369673252105713:  26%|██▌       | 1657/6470 [20:25<1:05:47,  1.22it/s]
3|1658|Loss: 0.9677086472511292:  26%|██▌       | 1657/6470 [20:26<1:05:47,  1.22it/s]
3|1658|Loss: 0.9677086472511292:  26%|██▌       | 1658/6470 [20:26<1:02:58,  1.27it/s]
3|1659|Loss: 0.8567874431610107:  26%|██▌       | 1658/6470 [20:27<1:02:58,  1.27it/s]
3|1659|Loss: 0.8567874431610107:  26%|██▌       | 1659/6470 [20:27<1:03:03,  1.27it/s]
3|1660|Loss: 0.9955441355705261:  26%|██▌       | 1659/6470 [20:27<1:03:03,  1.27it/s]
3|1660|Loss: 0.9955441355705261:  26%|██▌       | 1660/6470 [20:28<58:44,  1.36it/s]  
3|1661|Loss: 1.0907764434814453:  26%|██▌       | 1660/6470 [20:28<58:44,  1.36it/s]
3|1661|Loss: 1.0907764434814453:  26%|██▌       | 1661/6470 [20:28<59:11,  1.35it/s]
3|1662|Loss: 0.7514137625694275:  26%|██▌       | 1661/6470 [20:29<59:11,  1.35it/s]
3|1662|Loss: 0.7514137625694275:  26%|██▌       | 1662/6470 [20:29<55:53,  1.43it/s]
3|1663|Loss: 0.8406668305397034:  26%|██▌       | 1662/6470 [20:29<55:53,  1.43it/s]
3|1663|Loss: 0.8406668305397034:  26%|██▌       | 1663/6470 [20:29<50:45,  1.58it/s]
3|1664|Loss: 0.7497429847717285:  26%|██▌       | 1663/6470 [20:30<50:45,  1.58it/s]
3|1664|Loss: 0.7497429847717285:  26%|██▌       | 1664/6470 [20:30<52:43,  1.52it/s]
3|1665|Loss: 0.8355610370635986:  26%|██▌       | 1664/6470 [20:31<52:43,  1.52it/s]
3|1665|Loss: 0.8355610370635986:  26%|██▌       | 1665/6470 [20:31<51:31,  1.55it/s]
3|1666|Loss: 0.7622712850570679:  26%|██▌       | 1665/6470 [20:31<51:31,  1.55it/s]
3|1666|Loss: 0.7622712850570679:  26%|██▌       | 1666/6470 [20:32<54:24,  1.47it/s]
3|1667|Loss: 0.8537294864654541:  26%|██▌       | 1666/6470 [20:32<54:24,  1.47it/s]
3|1667|Loss: 0.8537294864654541:  26%|██▌       | 1667/6470 [20:32<56:21,  1.42it/s]
3|1668|Loss: 0.8910837173461914:  26%|██▌       | 1667/6470 [20:33<56:21,  1.42it/s]
3|1668|Loss: 0.8910837173461914:  26%|██▌       | 1668/6470 [20:33<56:36,  1.41it/s]
3|1669|Loss: 0.8731842637062073:  26%|██▌       | 1668/6470 [20:33<56:36,  1.41it/s]
3|1669|Loss: 0.8731842637062073:  26%|██▌       | 1669/6470 [20:34<58:22,  1.37it/s]
3|1670|Loss: 1.0788159370422363:  26%|██▌       | 1669/6470 [20:34<58:22,  1.37it/s]
3|1670|Loss: 1.0788159370422363:  26%|██▌       | 1670/6470 [20:34<58:12,  1.37it/s]
3|1671|Loss: 0.9682679772377014:  26%|██▌       | 1670/6470 [20:35<58:12,  1.37it/s]
3|1671|Loss: 0.9682679772377014:  26%|██▌       | 1671/6470 [20:35<58:42,  1.36it/s]
3|1672|Loss: 0.7752364873886108:  26%|██▌       | 1671/6470 [20:36<58:42,  1.36it/s]
3|1672|Loss: 0.7752364873886108:  26%|██▌       | 1672/6470 [20:36<57:17,  1.40it/s]
3|1673|Loss: 0.9137388467788696:  26%|██▌       | 1672/6470 [20:36<57:17,  1.40it/s]
3|1673|Loss: 0.9137388467788696:  26%|██▌       | 1673/6470 [20:36<52:19,  1.53it/s]
3|1674|Loss: 0.8538336157798767:  26%|██▌       | 1673/6470 [20:37<52:19,  1.53it/s]
3|1674|Loss: 0.8538336157798767:  26%|██▌       | 1674/6470 [20:37<52:19,  1.53it/s]
3|1675|Loss: 0.8446621298789978:  26%|██▌       | 1674/6470 [20:37<52:19,  1.53it/s]
3|1675|Loss: 0.8446621298789978:  26%|██▌       | 1675/6470 [20:38<53:41,  1.49it/s]
3|1676|Loss: 0.9253544807434082:  26%|██▌       | 1675/6470 [20:38<53:41,  1.49it/s]
3|1676|Loss: 0.9253544807434082:  26%|██▌       | 1676/6470 [20:39<55:27,  1.44it/s]
3|1677|Loss: 0.8946143388748169:  26%|██▌       | 1676/6470 [20:39<55:27,  1.44it/s]
3|1677|Loss: 0.8946143388748169:  26%|██▌       | 1677/6470 [20:39<1:00:33,  1.32it/s]
3|1678|Loss: 0.8680034279823303:  26%|██▌       | 1677/6470 [20:40<1:00:33,  1.32it/s]
3|1678|Loss: 0.8680034279823303:  26%|██▌       | 1678/6470 [20:40<1:01:58,  1.29it/s]
3|1679|Loss: 0.9074001312255859:  26%|██▌       | 1678/6470 [20:41<1:01:58,  1.29it/s]
3|1679|Loss: 0.9074001312255859:  26%|██▌       | 1679/6470 [20:41<59:14,  1.35it/s]  
3|1680|Loss: 0.9268997311592102:  26%|██▌       | 1679/6470 [20:41<59:14,  1.35it/s]
3|1680|Loss: 0.9268997311592102:  26%|██▌       | 1680/6470 [20:41<55:11,  1.45it/s]
3|1681|Loss: 0.9075280427932739:  26%|██▌       | 1680/6470 [20:42<55:11,  1.45it/s]
3|1681|Loss: 0.9075280427932739:  26%|██▌       | 1681/6470 [20:42<54:33,  1.46it/s]
3|1682|Loss: 0.7987980246543884:  26%|██▌       | 1681/6470 [20:43<54:33,  1.46it/s]
3|1682|Loss: 0.7987980246543884:  26%|██▌       | 1682/6470 [20:43<58:15,  1.37it/s]
3|1683|Loss: 0.868314266204834:  26%|██▌       | 1682/6470 [20:43<58:15,  1.37it/s] 
3|1683|Loss: 0.868314266204834:  26%|██▌       | 1683/6470 [20:44<54:43,  1.46it/s]
3|1684|Loss: 0.7436220645904541:  26%|██▌       | 1683/6470 [20:44<54:43,  1.46it/s]
3|1684|Loss: 0.7436220645904541:  26%|██▌       | 1684/6470 [20:44<53:19,  1.50it/s]
3|1685|Loss: 0.744683563709259:  26%|██▌       | 1684/6470 [20:45<53:19,  1.50it/s] 
3|1685|Loss: 0.744683563709259:  26%|██▌       | 1685/6470 [20:45<50:48,  1.57it/s]
3|1686|Loss: 0.7916414737701416:  26%|██▌       | 1685/6470 [20:45<50:48,  1.57it/s]
3|1686|Loss: 0.7916414737701416:  26%|██▌       | 1686/6470 [20:46<52:58,  1.51it/s]
3|1687|Loss: 0.9692949652671814:  26%|██▌       | 1686/6470 [20:46<52:58,  1.51it/s]
3|1687|Loss: 0.9692949652671814:  26%|██▌       | 1687/6470 [20:46<56:06,  1.42it/s]
3|1688|Loss: 0.7591434121131897:  26%|██▌       | 1687/6470 [20:47<56:06,  1.42it/s]
3|1688|Loss: 0.7591434121131897:  26%|██▌       | 1688/6470 [20:47<50:57,  1.56it/s]
3|1689|Loss: 0.8388312458992004:  26%|██▌       | 1688/6470 [20:47<50:57,  1.56it/s]
3|1689|Loss: 0.8388312458992004:  26%|██▌       | 1689/6470 [20:47<51:59,  1.53it/s]
3|1690|Loss: 0.7442344427108765:  26%|██▌       | 1689/6470 [20:48<51:59,  1.53it/s]
3|1690|Loss: 0.7442344427108765:  26%|██▌       | 1690/6470 [20:48<52:34,  1.52it/s]
3|1691|Loss: 0.5389389395713806:  26%|██▌       | 1690/6470 [20:49<52:34,  1.52it/s]
3|1691|Loss: 0.5389389395713806:  26%|██▌       | 1691/6470 [20:49<57:52,  1.38it/s]
3|1692|Loss: 0.8873768448829651:  26%|██▌       | 1691/6470 [20:49<57:52,  1.38it/s]
3|1692|Loss: 0.8873768448829651:  26%|██▌       | 1692/6470 [20:50<58:36,  1.36it/s]
3|1693|Loss: 0.8588136434555054:  26%|██▌       | 1692/6470 [20:50<58:36,  1.36it/s]
3|1693|Loss: 0.8588136434555054:  26%|██▌       | 1693/6470 [20:51<1:00:03,  1.33it/s]
3|1694|Loss: 0.7916470766067505:  26%|██▌       | 1693/6470 [20:51<1:00:03,  1.33it/s]
3|1694|Loss: 0.7916470766067505:  26%|██▌       | 1694/6470 [20:52<1:04:41,  1.23it/s]
3|1695|Loss: 0.6917443871498108:  26%|██▌       | 1694/6470 [20:52<1:04:41,  1.23it/s]
3|1695|Loss: 0.6917443871498108:  26%|██▌       | 1695/6470 [20:52<1:03:15,  1.26it/s]
3|1696|Loss: 0.8775784969329834:  26%|██▌       | 1695/6470 [20:53<1:03:15,  1.26it/s]
3|1696|Loss: 0.8775784969329834:  26%|██▌       | 1696/6470 [20:53<1:00:28,  1.32it/s]
3|1697|Loss: 0.8532434105873108:  26%|██▌       | 1696/6470 [20:53<1:00:28,  1.32it/s]
3|1697|Loss: 0.8532434105873108:  26%|██▌       | 1697/6470 [20:54<1:00:20,  1.32it/s]
3|1698|Loss: 0.7974914312362671:  26%|██▌       | 1697/6470 [20:54<1:00:20,  1.32it/s]
3|1698|Loss: 0.7974914312362671:  26%|██▌       | 1698/6470 [20:54<55:39,  1.43it/s]  
3|1699|Loss: 0.7600201368331909:  26%|██▌       | 1698/6470 [20:55<55:39,  1.43it/s]
3|1699|Loss: 0.7600201368331909:  26%|██▋       | 1699/6470 [20:55<55:47,  1.43it/s]
3|1700|Loss: 0.9142056107521057:  26%|██▋       | 1699/6470 [20:55<55:47,  1.43it/s]
3|1700|Loss: 0.9142056107521057:  26%|██▋       | 1700/6470 [20:56<58:14,  1.36it/s]
3|1701|Loss: 0.8223481774330139:  26%|██▋       | 1700/6470 [20:56<58:14,  1.36it/s]
3|1701|Loss: 0.8223481774330139:  26%|██▋       | 1701/6470 [20:57<1:01:15,  1.30it/s]
3|1702|Loss: 0.7860413193702698:  26%|██▋       | 1701/6470 [20:57<1:01:15,  1.30it/s]
3|1702|Loss: 0.7860413193702698:  26%|██▋       | 1702/6470 [20:58<1:07:17,  1.18it/s]
3|1703|Loss: 0.9523651599884033:  26%|██▋       | 1702/6470 [20:58<1:07:17,  1.18it/s]
3|1703|Loss: 0.9523651599884033:  26%|██▋       | 1703/6470 [20:59<1:06:42,  1.19it/s]
3|1704|Loss: 0.7098511457443237:  26%|██▋       | 1703/6470 [20:59<1:06:42,  1.19it/s]
3|1704|Loss: 0.7098511457443237:  26%|██▋       | 1704/6470 [20:59<1:04:49,  1.23it/s]
3|1705|Loss: 0.7261756658554077:  26%|██▋       | 1704/6470 [21:00<1:04:49,  1.23it/s]
3|1705|Loss: 0.7261756658554077:  26%|██▋       | 1705/6470 [21:00<1:04:08,  1.24it/s]
3|1706|Loss: 0.8055142164230347:  26%|██▋       | 1705/6470 [21:00<1:04:08,  1.24it/s]
3|1706|Loss: 0.8055142164230347:  26%|██▋       | 1706/6470 [21:01<1:00:07,  1.32it/s]
3|1707|Loss: 0.8431029319763184:  26%|██▋       | 1706/6470 [21:01<1:00:07,  1.32it/s]
3|1707|Loss: 0.8431029319763184:  26%|██▋       | 1707/6470 [21:01<59:17,  1.34it/s]  
3|1708|Loss: 0.8010978102684021:  26%|██▋       | 1707/6470 [21:02<59:17,  1.34it/s]
3|1708|Loss: 0.8010978102684021:  26%|██▋       | 1708/6470 [21:03<1:09:22,  1.14it/s]
3|1709|Loss: 1.0887271165847778:  26%|██▋       | 1708/6470 [21:03<1:09:22,  1.14it/s]
3|1709|Loss: 1.0887271165847778:  26%|██▋       | 1709/6470 [21:03<1:09:43,  1.14it/s]
3|1710|Loss: 0.7405482530593872:  26%|██▋       | 1709/6470 [21:04<1:09:43,  1.14it/s]
3|1710|Loss: 0.7405482530593872:  26%|██▋       | 1710/6470 [21:04<1:10:24,  1.13it/s]
3|1711|Loss: 0.7740232944488525:  26%|██▋       | 1710/6470 [21:05<1:10:24,  1.13it/s]
3|1711|Loss: 0.7740232944488525:  26%|██▋       | 1711/6470 [21:05<1:05:37,  1.21it/s]
3|1712|Loss: 0.830149233341217:  26%|██▋       | 1711/6470 [21:05<1:05:37,  1.21it/s] 
3|1712|Loss: 0.830149233341217:  26%|██▋       | 1712/6470 [21:06<1:03:49,  1.24it/s]
3|1713|Loss: 0.7306143641471863:  26%|██▋       | 1712/6470 [21:06<1:03:49,  1.24it/s]
3|1713|Loss: 0.7306143641471863:  26%|██▋       | 1713/6470 [21:06<1:00:29,  1.31it/s]
3|1714|Loss: 0.8963319063186646:  26%|██▋       | 1713/6470 [21:07<1:00:29,  1.31it/s]
3|1714|Loss: 0.8963319063186646:  26%|██▋       | 1714/6470 [21:07<55:18,  1.43it/s]  
3|1715|Loss: 0.8485136032104492:  26%|██▋       | 1714/6470 [21:08<55:18,  1.43it/s]
3|1715|Loss: 0.8485136032104492:  27%|██▋       | 1715/6470 [21:08<1:01:42,  1.28it/s]
3|1716|Loss: 0.830127477645874:  27%|██▋       | 1715/6470 [21:08<1:01:42,  1.28it/s] 
3|1716|Loss: 0.830127477645874:  27%|██▋       | 1716/6470 [21:09<1:01:29,  1.29it/s]
3|1717|Loss: 0.834640383720398:  27%|██▋       | 1716/6470 [21:09<1:01:29,  1.29it/s]
3|1717|Loss: 0.834640383720398:  27%|██▋       | 1717/6470 [21:10<1:00:28,  1.31it/s]
3|1718|Loss: 0.8145216703414917:  27%|██▋       | 1717/6470 [21:10<1:00:28,  1.31it/s]
3|1718|Loss: 0.8145216703414917:  27%|██▋       | 1718/6470 [21:10<1:03:20,  1.25it/s]
3|1719|Loss: 0.7198277115821838:  27%|██▋       | 1718/6470 [21:11<1:03:20,  1.25it/s]
3|1719|Loss: 0.7198277115821838:  27%|██▋       | 1719/6470 [21:11<1:05:17,  1.21it/s]
3|1720|Loss: 0.7801608443260193:  27%|██▋       | 1719/6470 [21:12<1:05:17,  1.21it/s]
3|1720|Loss: 0.7801608443260193:  27%|██▋       | 1720/6470 [21:12<1:08:53,  1.15it/s]
3|1721|Loss: 0.8957874178886414:  27%|██▋       | 1720/6470 [21:13<1:08:53,  1.15it/s]
3|1721|Loss: 0.8957874178886414:  27%|██▋       | 1721/6470 [21:13<1:10:06,  1.13it/s]
3|1722|Loss: 0.8553065061569214:  27%|██▋       | 1721/6470 [21:14<1:10:06,  1.13it/s]
3|1722|Loss: 0.8553065061569214:  27%|██▋       | 1722/6470 [21:14<1:09:09,  1.14it/s]
3|1723|Loss: 0.8774690628051758:  27%|██▋       | 1722/6470 [21:14<1:09:09,  1.14it/s]
3|1723|Loss: 0.8774690628051758:  27%|██▋       | 1723/6470 [21:15<1:09:17,  1.14it/s]
3|1724|Loss: 0.8301953077316284:  27%|██▋       | 1723/6470 [21:15<1:09:17,  1.14it/s]
3|1724|Loss: 0.8301953077316284:  27%|██▋       | 1724/6470 [21:16<1:05:15,  1.21it/s]
3|1725|Loss: 0.7357697486877441:  27%|██▋       | 1724/6470 [21:16<1:05:15,  1.21it/s]
3|1725|Loss: 0.7357697486877441:  27%|██▋       | 1725/6470 [21:16<1:06:40,  1.19it/s]
3|1726|Loss: 0.8711559772491455:  27%|██▋       | 1725/6470 [21:17<1:06:40,  1.19it/s]
3|1726|Loss: 0.8711559772491455:  27%|██▋       | 1726/6470 [21:17<1:05:37,  1.20it/s]
3|1727|Loss: 0.7149506211280823:  27%|██▋       | 1726/6470 [21:18<1:05:37,  1.20it/s]
3|1727|Loss: 0.7149506211280823:  27%|██▋       | 1727/6470 [21:18<56:49,  1.39it/s]  
3|1728|Loss: 0.635532796382904:  27%|██▋       | 1727/6470 [21:18<56:49,  1.39it/s] 
3|1728|Loss: 0.635532796382904:  27%|██▋       | 1728/6470 [21:19<58:39,  1.35it/s]
3|1729|Loss: 0.8350585103034973:  27%|██▋       | 1728/6470 [21:19<58:39,  1.35it/s]
3|1729|Loss: 0.8350585103034973:  27%|██▋       | 1729/6470 [21:19<59:29,  1.33it/s]
3|1730|Loss: 0.9256172776222229:  27%|██▋       | 1729/6470 [21:20<59:29,  1.33it/s]
3|1730|Loss: 0.9256172776222229:  27%|██▋       | 1730/6470 [21:20<59:43,  1.32it/s]
3|1731|Loss: 0.8955149054527283:  27%|██▋       | 1730/6470 [21:20<59:43,  1.32it/s]
3|1731|Loss: 0.8955149054527283:  27%|██▋       | 1731/6470 [21:21<58:24,  1.35it/s]
3|1732|Loss: 0.8417748808860779:  27%|██▋       | 1731/6470 [21:21<58:24,  1.35it/s]
3|1732|Loss: 0.8417748808860779:  27%|██▋       | 1732/6470 [21:22<58:20,  1.35it/s]
3|1733|Loss: 0.7911168336868286:  27%|██▋       | 1732/6470 [21:22<58:20,  1.35it/s]
3|1733|Loss: 0.7911168336868286:  27%|██▋       | 1733/6470 [21:22<57:08,  1.38it/s]
3|1734|Loss: 0.8785153031349182:  27%|██▋       | 1733/6470 [21:23<57:08,  1.38it/s]
3|1734|Loss: 0.8785153031349182:  27%|██▋       | 1734/6470 [21:23<55:51,  1.41it/s]
3|1735|Loss: 0.7351455092430115:  27%|██▋       | 1734/6470 [21:23<55:51,  1.41it/s]
3|1735|Loss: 0.7351455092430115:  27%|██▋       | 1735/6470 [21:24<53:51,  1.47it/s]
3|1736|Loss: 0.7219908833503723:  27%|██▋       | 1735/6470 [21:24<53:51,  1.47it/s]
3|1736|Loss: 0.7219908833503723:  27%|██▋       | 1736/6470 [21:24<53:52,  1.46it/s]
3|1737|Loss: 0.8021641373634338:  27%|██▋       | 1736/6470 [21:25<53:52,  1.46it/s]
3|1737|Loss: 0.8021641373634338:  27%|██▋       | 1737/6470 [21:25<53:40,  1.47it/s]
3|1738|Loss: 0.72214275598526:  27%|██▋       | 1737/6470 [21:25<53:40,  1.47it/s]  
3|1738|Loss: 0.72214275598526:  27%|██▋       | 1738/6470 [21:26<53:59,  1.46it/s]
3|1739|Loss: 0.9859501719474792:  27%|██▋       | 1738/6470 [21:26<53:59,  1.46it/s]
3|1739|Loss: 0.9859501719474792:  27%|██▋       | 1739/6470 [21:26<55:12,  1.43it/s]
3|1740|Loss: 0.8745901584625244:  27%|██▋       | 1739/6470 [21:27<55:12,  1.43it/s]
3|1740|Loss: 0.8745901584625244:  27%|██▋       | 1740/6470 [21:27<52:47,  1.49it/s]
3|1741|Loss: 0.9631521701812744:  27%|██▋       | 1740/6470 [21:27<52:47,  1.49it/s]
3|1741|Loss: 0.9631521701812744:  27%|██▋       | 1741/6470 [21:27<51:15,  1.54it/s]
3|1742|Loss: 0.9861255884170532:  27%|██▋       | 1741/6470 [21:28<51:15,  1.54it/s]
3|1742|Loss: 0.9861255884170532:  27%|██▋       | 1742/6470 [21:28<52:01,  1.51it/s]
3|1743|Loss: 0.6647511124610901:  27%|██▋       | 1742/6470 [21:29<52:01,  1.51it/s]
3|1743|Loss: 0.6647511124610901:  27%|██▋       | 1743/6470 [21:29<1:00:28,  1.30it/s]
3|1744|Loss: 0.9770631790161133:  27%|██▋       | 1743/6470 [21:30<1:00:28,  1.30it/s]
3|1744|Loss: 0.9770631790161133:  27%|██▋       | 1744/6470 [21:30<1:01:10,  1.29it/s]
3|1745|Loss: 0.7681019902229309:  27%|██▋       | 1744/6470 [21:30<1:01:10,  1.29it/s]
3|1745|Loss: 0.7681019902229309:  27%|██▋       | 1745/6470 [21:31<58:48,  1.34it/s]  
3|1746|Loss: 0.8303637504577637:  27%|██▋       | 1745/6470 [21:31<58:48,  1.34it/s]
3|1746|Loss: 0.8303637504577637:  27%|██▋       | 1746/6470 [21:31<55:23,  1.42it/s]
3|1747|Loss: 0.9824814200401306:  27%|██▋       | 1746/6470 [21:32<55:23,  1.42it/s]
3|1747|Loss: 0.9824814200401306:  27%|██▋       | 1747/6470 [21:32<54:00,  1.46it/s]
3|1748|Loss: 0.8351073265075684:  27%|██▋       | 1747/6470 [21:32<54:00,  1.46it/s]
3|1748|Loss: 0.8351073265075684:  27%|██▋       | 1748/6470 [21:33<54:15,  1.45it/s]
3|1749|Loss: 0.9499897360801697:  27%|██▋       | 1748/6470 [21:33<54:15,  1.45it/s]
3|1749|Loss: 0.9499897360801697:  27%|██▋       | 1749/6470 [21:33<54:38,  1.44it/s]
3|1750|Loss: 0.786970853805542:  27%|██▋       | 1749/6470 [21:34<54:38,  1.44it/s] 
3|1750|Loss: 0.786970853805542:  27%|██▋       | 1750/6470 [21:34<55:50,  1.41it/s]
3|1751|Loss: 0.8718955516815186:  27%|██▋       | 1750/6470 [21:34<55:50,  1.41it/s]
3|1751|Loss: 0.8718955516815186:  27%|██▋       | 1751/6470 [21:35<55:08,  1.43it/s]
3|1752|Loss: 0.9928817749023438:  27%|██▋       | 1751/6470 [21:35<55:08,  1.43it/s]
3|1752|Loss: 0.9928817749023438:  27%|██▋       | 1752/6470 [21:35<55:00,  1.43it/s]
3|1753|Loss: 0.8685693144798279:  27%|██▋       | 1752/6470 [21:36<55:00,  1.43it/s]
3|1753|Loss: 0.8685693144798279:  27%|██▋       | 1753/6470 [21:36<56:31,  1.39it/s]
3|1754|Loss: 0.6570732593536377:  27%|██▋       | 1753/6470 [21:37<56:31,  1.39it/s]
3|1754|Loss: 0.6570732593536377:  27%|██▋       | 1754/6470 [21:37<1:07:54,  1.16it/s]
3|1755|Loss: 0.7541103363037109:  27%|██▋       | 1754/6470 [21:38<1:07:54,  1.16it/s]
3|1755|Loss: 0.7541103363037109:  27%|██▋       | 1755/6470 [21:38<1:07:56,  1.16it/s]
3|1756|Loss: 0.8459101915359497:  27%|██▋       | 1755/6470 [21:39<1:07:56,  1.16it/s]
3|1756|Loss: 0.8459101915359497:  27%|██▋       | 1756/6470 [21:39<59:50,  1.31it/s]  
3|1757|Loss: 0.9280678033828735:  27%|██▋       | 1756/6470 [21:39<59:50,  1.31it/s]
3|1757|Loss: 0.9280678033828735:  27%|██▋       | 1757/6470 [21:39<58:09,  1.35it/s]
3|1758|Loss: 0.7855880260467529:  27%|██▋       | 1757/6470 [21:40<58:09,  1.35it/s]
3|1758|Loss: 0.7855880260467529:  27%|██▋       | 1758/6470 [21:40<1:01:52,  1.27it/s]
3|1759|Loss: 0.7678121328353882:  27%|██▋       | 1758/6470 [21:41<1:01:52,  1.27it/s]
3|1759|Loss: 0.7678121328353882:  27%|██▋       | 1759/6470 [21:41<1:05:36,  1.20it/s]
3|1760|Loss: 0.6792173385620117:  27%|██▋       | 1759/6470 [21:42<1:05:36,  1.20it/s]
3|1760|Loss: 0.6792173385620117:  27%|██▋       | 1760/6470 [21:42<1:05:46,  1.19it/s]
3|1761|Loss: 0.7360350489616394:  27%|██▋       | 1760/6470 [21:43<1:05:46,  1.19it/s]
3|1761|Loss: 0.7360350489616394:  27%|██▋       | 1761/6470 [21:43<1:02:44,  1.25it/s]
3|1762|Loss: 0.8275949358940125:  27%|██▋       | 1761/6470 [21:43<1:02:44,  1.25it/s]
3|1762|Loss: 0.8275949358940125:  27%|██▋       | 1762/6470 [21:44<1:02:06,  1.26it/s]
3|1763|Loss: 0.9178177714347839:  27%|██▋       | 1762/6470 [21:44<1:02:06,  1.26it/s]
3|1763|Loss: 0.9178177714347839:  27%|██▋       | 1763/6470 [21:44<1:01:34,  1.27it/s]
3|1764|Loss: 0.7538606524467468:  27%|██▋       | 1763/6470 [21:45<1:01:34,  1.27it/s]
3|1764|Loss: 0.7538606524467468:  27%|██▋       | 1764/6470 [21:45<1:02:42,  1.25it/s]
3|1765|Loss: 0.7223511338233948:  27%|██▋       | 1764/6470 [21:46<1:02:42,  1.25it/s]
3|1765|Loss: 0.7223511338233948:  27%|██▋       | 1765/6470 [21:46<1:01:14,  1.28it/s]
3|1766|Loss: 0.7991337776184082:  27%|██▋       | 1765/6470 [21:46<1:01:14,  1.28it/s]
3|1766|Loss: 0.7991337776184082:  27%|██▋       | 1766/6470 [21:47<58:39,  1.34it/s]  
3|1767|Loss: 0.8125242590904236:  27%|██▋       | 1766/6470 [21:47<58:39,  1.34it/s]
3|1767|Loss: 0.8125242590904236:  27%|██▋       | 1767/6470 [21:47<1:00:21,  1.30it/s]
3|1768|Loss: 0.9271681904792786:  27%|██▋       | 1767/6470 [21:48<1:00:21,  1.30it/s]
3|1768|Loss: 0.9271681904792786:  27%|██▋       | 1768/6470 [21:48<1:03:00,  1.24it/s]
3|1769|Loss: 0.807135283946991:  27%|██▋       | 1768/6470 [21:49<1:03:00,  1.24it/s] 
3|1769|Loss: 0.807135283946991:  27%|██▋       | 1769/6470 [21:49<59:57,  1.31it/s]  
3|1770|Loss: 0.8640206456184387:  27%|██▋       | 1769/6470 [21:49<59:57,  1.31it/s]
3|1770|Loss: 0.8640206456184387:  27%|██▋       | 1770/6470 [21:50<58:01,  1.35it/s]
3|1771|Loss: 0.6806470155715942:  27%|██▋       | 1770/6470 [21:50<58:01,  1.35it/s]
3|1771|Loss: 0.6806470155715942:  27%|██▋       | 1771/6470 [21:51<58:42,  1.33it/s]
3|1772|Loss: 0.734156608581543:  27%|██▋       | 1771/6470 [21:51<58:42,  1.33it/s] 
3|1772|Loss: 0.734156608581543:  27%|██▋       | 1772/6470 [21:51<1:00:10,  1.30it/s]
3|1773|Loss: 0.8089525103569031:  27%|██▋       | 1772/6470 [21:52<1:00:10,  1.30it/s]
3|1773|Loss: 0.8089525103569031:  27%|██▋       | 1773/6470 [21:52<58:55,  1.33it/s]  
3|1774|Loss: 0.8115978240966797:  27%|██▋       | 1773/6470 [21:52<58:55,  1.33it/s]
3|1774|Loss: 0.8115978240966797:  27%|██▋       | 1774/6470 [21:53<55:56,  1.40it/s]
3|1775|Loss: 0.7504175901412964:  27%|██▋       | 1774/6470 [21:53<55:56,  1.40it/s]
3|1775|Loss: 0.7504175901412964:  27%|██▋       | 1775/6470 [21:54<59:58,  1.30it/s]
3|1776|Loss: 0.9051883220672607:  27%|██▋       | 1775/6470 [21:54<59:58,  1.30it/s]
3|1776|Loss: 0.9051883220672607:  27%|██▋       | 1776/6470 [21:54<1:00:56,  1.28it/s]
3|1777|Loss: 0.6921025514602661:  27%|██▋       | 1776/6470 [21:55<1:00:56,  1.28it/s]
3|1777|Loss: 0.6921025514602661:  27%|██▋       | 1777/6470 [21:55<1:00:50,  1.29it/s]
3|1778|Loss: 0.7614858746528625:  27%|██▋       | 1777/6470 [21:56<1:00:50,  1.29it/s]
3|1778|Loss: 0.7614858746528625:  27%|██▋       | 1778/6470 [21:56<59:22,  1.32it/s]  
3|1779|Loss: 0.8326209783554077:  27%|██▋       | 1778/6470 [21:56<59:22,  1.32it/s]
3|1779|Loss: 0.8326209783554077:  27%|██▋       | 1779/6470 [21:57<1:01:05,  1.28it/s]
3|1780|Loss: 0.8557695150375366:  27%|██▋       | 1779/6470 [21:57<1:01:05,  1.28it/s]
3|1780|Loss: 0.8557695150375366:  28%|██▊       | 1780/6470 [21:57<1:01:19,  1.27it/s]
3|1781|Loss: 0.7261852025985718:  28%|██▊       | 1780/6470 [21:58<1:01:19,  1.27it/s]
3|1781|Loss: 0.7261852025985718:  28%|██▊       | 1781/6470 [21:58<1:00:47,  1.29it/s]
3|1782|Loss: 0.7694878578186035:  28%|██▊       | 1781/6470 [21:59<1:00:47,  1.29it/s]
3|1782|Loss: 0.7694878578186035:  28%|██▊       | 1782/6470 [21:59<59:08,  1.32it/s]  
3|1783|Loss: 0.840612530708313:  28%|██▊       | 1782/6470 [21:59<59:08,  1.32it/s] 
3|1783|Loss: 0.840612530708313:  28%|██▊       | 1783/6470 [22:00<57:21,  1.36it/s]
3|1784|Loss: 0.7755405306816101:  28%|██▊       | 1783/6470 [22:00<57:21,  1.36it/s]
3|1784|Loss: 0.7755405306816101:  28%|██▊       | 1784/6470 [22:00<59:57,  1.30it/s]
3|1785|Loss: 0.8236963152885437:  28%|██▊       | 1784/6470 [22:01<59:57,  1.30it/s]
3|1785|Loss: 0.8236963152885437:  28%|██▊       | 1785/6470 [22:01<1:00:22,  1.29it/s]
3|1786|Loss: 0.7764968872070312:  28%|██▊       | 1785/6470 [22:02<1:00:22,  1.29it/s]
3|1786|Loss: 0.7764968872070312:  28%|██▊       | 1786/6470 [22:02<1:00:28,  1.29it/s]
3|1787|Loss: 0.9957312345504761:  28%|██▊       | 1786/6470 [22:03<1:00:28,  1.29it/s]
3|1787|Loss: 0.9957312345504761:  28%|██▊       | 1787/6470 [22:03<1:02:41,  1.24it/s]
3|1788|Loss: 0.8348820805549622:  28%|██▊       | 1787/6470 [22:03<1:02:41,  1.24it/s]
3|1788|Loss: 0.8348820805549622:  28%|██▊       | 1788/6470 [22:04<1:01:11,  1.28it/s]
3|1789|Loss: 0.8662413954734802:  28%|██▊       | 1788/6470 [22:04<1:01:11,  1.28it/s]
3|1789|Loss: 0.8662413954734802:  28%|██▊       | 1789/6470 [22:04<56:06,  1.39it/s]  
3|1790|Loss: 0.8470401167869568:  28%|██▊       | 1789/6470 [22:05<56:06,  1.39it/s]
3|1790|Loss: 0.8470401167869568:  28%|██▊       | 1790/6470 [22:05<54:10,  1.44it/s]
3|1791|Loss: 0.8406674265861511:  28%|██▊       | 1790/6470 [22:05<54:10,  1.44it/s]
3|1791|Loss: 0.8406674265861511:  28%|██▊       | 1791/6470 [22:05<52:38,  1.48it/s]
3|1792|Loss: 0.8602845072746277:  28%|██▊       | 1791/6470 [22:06<52:38,  1.48it/s]
3|1792|Loss: 0.8602845072746277:  28%|██▊       | 1792/6470 [22:06<48:42,  1.60it/s]
3|1793|Loss: 0.5814164876937866:  28%|██▊       | 1792/6470 [22:06<48:42,  1.60it/s]
3|1793|Loss: 0.5814164876937866:  28%|██▊       | 1793/6470 [22:07<56:13,  1.39it/s]
3|1794|Loss: 0.8722557425498962:  28%|██▊       | 1793/6470 [22:08<56:13,  1.39it/s]
3|1794|Loss: 0.8722557425498962:  28%|██▊       | 1794/6470 [22:08<1:09:07,  1.13it/s]
3|1795|Loss: 0.8081777095794678:  28%|██▊       | 1794/6470 [22:09<1:09:07,  1.13it/s]
3|1795|Loss: 0.8081777095794678:  28%|██▊       | 1795/6470 [22:09<1:06:10,  1.18it/s]
3|1796|Loss: 0.8771472573280334:  28%|██▊       | 1795/6470 [22:09<1:06:10,  1.18it/s]
3|1796|Loss: 0.8771472573280334:  28%|██▊       | 1796/6470 [22:10<1:06:33,  1.17it/s]
3|1797|Loss: 0.7924218773841858:  28%|██▊       | 1796/6470 [22:10<1:06:33,  1.17it/s]
3|1797|Loss: 0.7924218773841858:  28%|██▊       | 1797/6470 [22:11<1:03:34,  1.23it/s]
3|1798|Loss: 0.8778522610664368:  28%|██▊       | 1797/6470 [22:11<1:03:34,  1.23it/s]
3|1798|Loss: 0.8778522610664368:  28%|██▊       | 1798/6470 [22:11<1:05:03,  1.20it/s]
3|1799|Loss: 0.7169763445854187:  28%|██▊       | 1798/6470 [22:12<1:05:03,  1.20it/s]
3|1799|Loss: 0.7169763445854187:  28%|██▊       | 1799/6470 [22:12<1:03:33,  1.22it/s]
3|1800|Loss: 0.8071562647819519:  28%|██▊       | 1799/6470 [22:13<1:03:33,  1.22it/s]
3|1800|Loss: 0.8071562647819519:  28%|██▊       | 1800/6470 [22:13<1:02:32,  1.24it/s]
3|1801|Loss: 0.5643091201782227:  28%|██▊       | 1800/6470 [22:14<1:02:32,  1.24it/s]
3|1801|Loss: 0.5643091201782227:  28%|██▊       | 1801/6470 [22:14<1:07:46,  1.15it/s]
3|1802|Loss: 0.8044613003730774:  28%|██▊       | 1801/6470 [22:14<1:07:46,  1.15it/s]
3|1802|Loss: 0.8044613003730774:  28%|██▊       | 1802/6470 [22:15<1:03:09,  1.23it/s]
3|1803|Loss: 0.7378359436988831:  28%|██▊       | 1802/6470 [22:15<1:03:09,  1.23it/s]
3|1803|Loss: 0.7378359436988831:  28%|██▊       | 1803/6470 [22:15<53:54,  1.44it/s]  
3|1804|Loss: 0.7001817226409912:  28%|██▊       | 1803/6470 [22:15<53:54,  1.44it/s]
3|1804|Loss: 0.7001817226409912:  28%|██▊       | 1804/6470 [22:16<54:13,  1.43it/s]
3|1805|Loss: 0.8472653031349182:  28%|██▊       | 1804/6470 [22:16<54:13,  1.43it/s]
3|1805|Loss: 0.8472653031349182:  28%|██▊       | 1805/6470 [22:17<56:55,  1.37it/s]
3|1806|Loss: 0.9263327717781067:  28%|██▊       | 1805/6470 [22:17<56:55,  1.37it/s]
3|1806|Loss: 0.9263327717781067:  28%|██▊       | 1806/6470 [22:17<56:12,  1.38it/s]
3|1807|Loss: 0.9158501029014587:  28%|██▊       | 1806/6470 [22:18<56:12,  1.38it/s]
3|1807|Loss: 0.9158501029014587:  28%|██▊       | 1807/6470 [22:18<55:28,  1.40it/s]
3|1808|Loss: 0.8408341407775879:  28%|██▊       | 1807/6470 [22:18<55:28,  1.40it/s]
3|1808|Loss: 0.8408341407775879:  28%|██▊       | 1808/6470 [22:19<55:18,  1.40it/s]
3|1809|Loss: 0.7153803110122681:  28%|██▊       | 1808/6470 [22:19<55:18,  1.40it/s]
3|1809|Loss: 0.7153803110122681:  28%|██▊       | 1809/6470 [22:20<59:14,  1.31it/s]
3|1810|Loss: 1.0339186191558838:  28%|██▊       | 1809/6470 [22:20<59:14,  1.31it/s]
3|1810|Loss: 1.0339186191558838:  28%|██▊       | 1810/6470 [22:20<58:42,  1.32it/s]
3|1811|Loss: 0.751472532749176:  28%|██▊       | 1810/6470 [22:21<58:42,  1.32it/s] 
3|1811|Loss: 0.751472532749176:  28%|██▊       | 1811/6470 [22:21<59:09,  1.31it/s]
3|1812|Loss: 0.8818999528884888:  28%|██▊       | 1811/6470 [22:22<59:09,  1.31it/s]
3|1812|Loss: 0.8818999528884888:  28%|██▊       | 1812/6470 [22:22<57:05,  1.36it/s]
3|1813|Loss: 0.8400938510894775:  28%|██▊       | 1812/6470 [22:22<57:05,  1.36it/s]
3|1813|Loss: 0.8400938510894775:  28%|██▊       | 1813/6470 [22:23<56:49,  1.37it/s]
3|1814|Loss: 0.8429545164108276:  28%|██▊       | 1813/6470 [22:23<56:49,  1.37it/s]
3|1814|Loss: 0.8429545164108276:  28%|██▊       | 1814/6470 [22:23<56:09,  1.38it/s]
3|1815|Loss: 0.747206449508667:  28%|██▊       | 1814/6470 [22:24<56:09,  1.38it/s] 
3|1815|Loss: 0.747206449508667:  28%|██▊       | 1815/6470 [22:24<57:39,  1.35it/s]
3|1816|Loss: 0.897144615650177:  28%|██▊       | 1815/6470 [22:24<57:39,  1.35it/s]
3|1816|Loss: 0.897144615650177:  28%|██▊       | 1816/6470 [22:25<1:00:47,  1.28it/s]
3|1817|Loss: 0.8957228064537048:  28%|██▊       | 1816/6470 [22:25<1:00:47,  1.28it/s]
3|1817|Loss: 0.8957228064537048:  28%|██▊       | 1817/6470 [22:26<1:03:18,  1.23it/s]
3|1818|Loss: 0.8943819403648376:  28%|██▊       | 1817/6470 [22:26<1:03:18,  1.23it/s]
3|1818|Loss: 0.8943819403648376:  28%|██▊       | 1818/6470 [22:27<1:03:02,  1.23it/s]
3|1819|Loss: 0.7929927110671997:  28%|██▊       | 1818/6470 [22:27<1:03:02,  1.23it/s]
3|1819|Loss: 0.7929927110671997:  28%|██▊       | 1819/6470 [22:27<58:49,  1.32it/s]  
3|1820|Loss: 0.8262531757354736:  28%|██▊       | 1819/6470 [22:28<58:49,  1.32it/s]
3|1820|Loss: 0.8262531757354736:  28%|██▊       | 1820/6470 [22:28<57:53,  1.34it/s]
3|1821|Loss: 0.8793761730194092:  28%|██▊       | 1820/6470 [22:28<57:53,  1.34it/s]
3|1821|Loss: 0.8793761730194092:  28%|██▊       | 1821/6470 [22:29<57:03,  1.36it/s]
3|1822|Loss: 0.9078941941261292:  28%|██▊       | 1821/6470 [22:29<57:03,  1.36it/s]
3|1822|Loss: 0.9078941941261292:  28%|██▊       | 1822/6470 [22:29<59:18,  1.31it/s]
3|1823|Loss: 0.7664783000946045:  28%|██▊       | 1822/6470 [22:30<59:18,  1.31it/s]
3|1823|Loss: 0.7664783000946045:  28%|██▊       | 1823/6470 [22:30<59:31,  1.30it/s]
3|1824|Loss: 0.8946253657341003:  28%|██▊       | 1823/6470 [22:31<59:31,  1.30it/s]
3|1824|Loss: 0.8946253657341003:  28%|██▊       | 1824/6470 [22:31<1:01:52,  1.25it/s]
3|1825|Loss: 0.7257717251777649:  28%|██▊       | 1824/6470 [22:32<1:01:52,  1.25it/s]
3|1825|Loss: 0.7257717251777649:  28%|██▊       | 1825/6470 [22:32<1:01:32,  1.26it/s]
3|1826|Loss: 0.9882612228393555:  28%|██▊       | 1825/6470 [22:32<1:01:32,  1.26it/s]
3|1826|Loss: 0.9882612228393555:  28%|██▊       | 1826/6470 [22:33<1:01:31,  1.26it/s]
3|1827|Loss: 0.9707065224647522:  28%|██▊       | 1826/6470 [22:33<1:01:31,  1.26it/s]
3|1827|Loss: 0.9707065224647522:  28%|██▊       | 1827/6470 [22:33<55:49,  1.39it/s]  
3|1828|Loss: 0.7592754364013672:  28%|██▊       | 1827/6470 [22:34<55:49,  1.39it/s]
3|1828|Loss: 0.7592754364013672:  28%|██▊       | 1828/6470 [22:34<55:51,  1.39it/s]
3|1829|Loss: 0.7586503624916077:  28%|██▊       | 1828/6470 [22:34<55:51,  1.39it/s]
3|1829|Loss: 0.7586503624916077:  28%|██▊       | 1829/6470 [22:35<57:46,  1.34it/s]
3|1830|Loss: 0.8024256229400635:  28%|██▊       | 1829/6470 [22:35<57:46,  1.34it/s]
3|1830|Loss: 0.8024256229400635:  28%|██▊       | 1830/6470 [22:35<54:41,  1.41it/s]
3|1831|Loss: 0.8085472583770752:  28%|██▊       | 1830/6470 [22:36<54:41,  1.41it/s]
3|1831|Loss: 0.8085472583770752:  28%|██▊       | 1831/6470 [22:36<57:06,  1.35it/s]
3|1832|Loss: 0.9801735877990723:  28%|██▊       | 1831/6470 [22:37<57:06,  1.35it/s]
3|1832|Loss: 0.9801735877990723:  28%|██▊       | 1832/6470 [22:37<56:16,  1.37it/s]
3|1833|Loss: 0.7876318097114563:  28%|██▊       | 1832/6470 [22:37<56:16,  1.37it/s]
3|1833|Loss: 0.7876318097114563:  28%|██▊       | 1833/6470 [22:38<54:07,  1.43it/s]
3|1834|Loss: 0.8684648871421814:  28%|██▊       | 1833/6470 [22:38<54:07,  1.43it/s]
3|1834|Loss: 0.8684648871421814:  28%|██▊       | 1834/6470 [22:38<52:57,  1.46it/s]
3|1835|Loss: 0.9301843643188477:  28%|██▊       | 1834/6470 [22:39<52:57,  1.46it/s]
3|1835|Loss: 0.9301843643188477:  28%|██▊       | 1835/6470 [22:39<52:34,  1.47it/s]
3|1836|Loss: 1.0144627094268799:  28%|██▊       | 1835/6470 [22:39<52:34,  1.47it/s]
3|1836|Loss: 1.0144627094268799:  28%|██▊       | 1836/6470 [22:40<52:50,  1.46it/s]
3|1837|Loss: 0.91490238904953:  28%|██▊       | 1836/6470 [22:40<52:50,  1.46it/s]  
3|1837|Loss: 0.91490238904953:  28%|██▊       | 1837/6470 [22:40<54:06,  1.43it/s]
3|1838|Loss: 0.9101846218109131:  28%|██▊       | 1837/6470 [22:41<54:06,  1.43it/s]
3|1838|Loss: 0.9101846218109131:  28%|██▊       | 1838/6470 [22:41<53:07,  1.45it/s]
3|1839|Loss: 0.6696144938468933:  28%|██▊       | 1838/6470 [22:41<53:07,  1.45it/s]
3|1839|Loss: 0.6696144938468933:  28%|██▊       | 1839/6470 [22:42<50:49,  1.52it/s]
3|1840|Loss: 0.7621327638626099:  28%|██▊       | 1839/6470 [22:42<50:49,  1.52it/s]
3|1840|Loss: 0.7621327638626099:  28%|██▊       | 1840/6470 [22:42<49:34,  1.56it/s]
3|1841|Loss: 0.8473293781280518:  28%|██▊       | 1840/6470 [22:43<49:34,  1.56it/s]
3|1841|Loss: 0.8473293781280518:  28%|██▊       | 1841/6470 [22:43<52:15,  1.48it/s]
3|1842|Loss: 0.7841001749038696:  28%|██▊       | 1841/6470 [22:43<52:15,  1.48it/s]
3|1842|Loss: 0.7841001749038696:  28%|██▊       | 1842/6470 [22:44<54:11,  1.42it/s]
3|1843|Loss: 1.0334420204162598:  28%|██▊       | 1842/6470 [22:44<54:11,  1.42it/s]
3|1843|Loss: 1.0334420204162598:  28%|██▊       | 1843/6470 [22:44<55:33,  1.39it/s]
3|1844|Loss: 0.8789621591567993:  28%|██▊       | 1843/6470 [22:45<55:33,  1.39it/s]
3|1844|Loss: 0.8789621591567993:  29%|██▊       | 1844/6470 [22:45<56:22,  1.37it/s]
3|1845|Loss: 0.9097827076911926:  29%|██▊       | 1844/6470 [22:46<56:22,  1.37it/s]
3|1845|Loss: 0.9097827076911926:  29%|██▊       | 1845/6470 [22:46<56:47,  1.36it/s]
3|1846|Loss: 0.8462790846824646:  29%|██▊       | 1845/6470 [22:46<56:47,  1.36it/s]
3|1846|Loss: 0.8462790846824646:  29%|██▊       | 1846/6470 [22:47<52:43,  1.46it/s]
3|1847|Loss: 0.8657415509223938:  29%|██▊       | 1846/6470 [22:47<52:43,  1.46it/s]
3|1847|Loss: 0.8657415509223938:  29%|██▊       | 1847/6470 [22:47<50:11,  1.53it/s]
3|1848|Loss: 0.8375164866447449:  29%|██▊       | 1847/6470 [22:47<50:11,  1.53it/s]
3|1848|Loss: 0.8375164866447449:  29%|██▊       | 1848/6470 [22:48<52:33,  1.47it/s]
3|1849|Loss: 0.9339773058891296:  29%|██▊       | 1848/6470 [22:48<52:33,  1.47it/s]
3|1849|Loss: 0.9339773058891296:  29%|██▊       | 1849/6470 [22:48<51:22,  1.50it/s]
3|1850|Loss: 0.6852261424064636:  29%|██▊       | 1849/6470 [22:49<51:22,  1.50it/s]
3|1850|Loss: 0.6852261424064636:  29%|██▊       | 1850/6470 [22:49<48:50,  1.58it/s]
3|1851|Loss: 0.8617584109306335:  29%|██▊       | 1850/6470 [22:49<48:50,  1.58it/s]
3|1851|Loss: 0.8617584109306335:  29%|██▊       | 1851/6470 [22:50<51:32,  1.49it/s]
3|1852|Loss: 0.7848628163337708:  29%|██▊       | 1851/6470 [22:50<51:32,  1.49it/s]
3|1852|Loss: 0.7848628163337708:  29%|██▊       | 1852/6470 [22:50<52:20,  1.47it/s]
3|1853|Loss: 1.038869857788086:  29%|██▊       | 1852/6470 [22:51<52:20,  1.47it/s] 
3|1853|Loss: 1.038869857788086:  29%|██▊       | 1853/6470 [22:51<57:03,  1.35it/s]
3|1854|Loss: 0.8429651856422424:  29%|██▊       | 1853/6470 [22:52<57:03,  1.35it/s]
3|1854|Loss: 0.8429651856422424:  29%|██▊       | 1854/6470 [22:52<58:18,  1.32it/s]
3|1855|Loss: 0.7596447467803955:  29%|██▊       | 1854/6470 [22:52<58:18,  1.32it/s]
3|1855|Loss: 0.7596447467803955:  29%|██▊       | 1855/6470 [22:53<51:02,  1.51it/s]
3|1856|Loss: 0.7580046653747559:  29%|██▊       | 1855/6470 [22:53<51:02,  1.51it/s]
3|1856|Loss: 0.7580046653747559:  29%|██▊       | 1856/6470 [22:53<51:29,  1.49it/s]
3|1857|Loss: 0.7290442585945129:  29%|██▊       | 1856/6470 [22:54<51:29,  1.49it/s]
3|1857|Loss: 0.7290442585945129:  29%|██▊       | 1857/6470 [22:54<54:11,  1.42it/s]
3|1858|Loss: 0.8722184896469116:  29%|██▊       | 1857/6470 [22:54<54:11,  1.42it/s]
3|1858|Loss: 0.8722184896469116:  29%|██▊       | 1858/6470 [22:55<53:51,  1.43it/s]
3|1859|Loss: 0.8787286877632141:  29%|██▊       | 1858/6470 [22:55<53:51,  1.43it/s]
3|1859|Loss: 0.8787286877632141:  29%|██▊       | 1859/6470 [22:55<53:40,  1.43it/s]
3|1860|Loss: 0.7373295426368713:  29%|██▊       | 1859/6470 [22:56<53:40,  1.43it/s]
3|1860|Loss: 0.7373295426368713:  29%|██▊       | 1860/6470 [22:56<52:18,  1.47it/s]
3|1861|Loss: 0.6324712038040161:  29%|██▊       | 1860/6470 [22:56<52:18,  1.47it/s]
3|1861|Loss: 0.6324712038040161:  29%|██▉       | 1861/6470 [22:57<50:09,  1.53it/s]
3|1862|Loss: 0.7698320746421814:  29%|██▉       | 1861/6470 [22:57<50:09,  1.53it/s]
3|1862|Loss: 0.7698320746421814:  29%|██▉       | 1862/6470 [22:58<55:43,  1.38it/s]
3|1863|Loss: 1.0089781284332275:  29%|██▉       | 1862/6470 [22:58<55:43,  1.38it/s]
3|1863|Loss: 1.0089781284332275:  29%|██▉       | 1863/6470 [22:58<55:45,  1.38it/s]
3|1864|Loss: 0.7572901844978333:  29%|██▉       | 1863/6470 [22:59<55:45,  1.38it/s]
3|1864|Loss: 0.7572901844978333:  29%|██▉       | 1864/6470 [22:59<59:10,  1.30it/s]
3|1865|Loss: 0.8384988903999329:  29%|██▉       | 1864/6470 [23:00<59:10,  1.30it/s]
3|1865|Loss: 0.8384988903999329:  29%|██▉       | 1865/6470 [23:00<57:22,  1.34it/s]
3|1866|Loss: 0.8250091671943665:  29%|██▉       | 1865/6470 [23:00<57:22,  1.34it/s]
3|1866|Loss: 0.8250091671943665:  29%|██▉       | 1866/6470 [23:01<55:46,  1.38it/s]
3|1867|Loss: 0.8374203443527222:  29%|██▉       | 1866/6470 [23:01<55:46,  1.38it/s]
3|1867|Loss: 0.8374203443527222:  29%|██▉       | 1867/6470 [23:01<55:16,  1.39it/s]
3|1868|Loss: 0.7785190939903259:  29%|██▉       | 1867/6470 [23:02<55:16,  1.39it/s]
3|1868|Loss: 0.7785190939903259:  29%|██▉       | 1868/6470 [23:02<56:17,  1.36it/s]
3|1869|Loss: 0.8331462144851685:  29%|██▉       | 1868/6470 [23:02<56:17,  1.36it/s]
3|1869|Loss: 0.8331462144851685:  29%|██▉       | 1869/6470 [23:03<56:54,  1.35it/s]
3|1870|Loss: 0.7851023077964783:  29%|██▉       | 1869/6470 [23:03<56:54,  1.35it/s]
3|1870|Loss: 0.7851023077964783:  29%|██▉       | 1870/6470 [23:03<52:40,  1.46it/s]
3|1871|Loss: 0.8714955449104309:  29%|██▉       | 1870/6470 [23:04<52:40,  1.46it/s]
3|1871|Loss: 0.8714955449104309:  29%|██▉       | 1871/6470 [23:04<56:22,  1.36it/s]
3|1872|Loss: 0.9760444164276123:  29%|██▉       | 1871/6470 [23:05<56:22,  1.36it/s]
3|1872|Loss: 0.9760444164276123:  29%|██▉       | 1872/6470 [23:05<53:21,  1.44it/s]
3|1873|Loss: 0.7249153256416321:  29%|██▉       | 1872/6470 [23:05<53:21,  1.44it/s]
3|1873|Loss: 0.7249153256416321:  29%|██▉       | 1873/6470 [23:05<51:36,  1.48it/s]
3|1874|Loss: 0.849280834197998:  29%|██▉       | 1873/6470 [23:06<51:36,  1.48it/s] 
3|1874|Loss: 0.849280834197998:  29%|██▉       | 1874/6470 [23:06<54:06,  1.42it/s]
3|1875|Loss: 0.8350027799606323:  29%|██▉       | 1874/6470 [23:07<54:06,  1.42it/s]
3|1875|Loss: 0.8350027799606323:  29%|██▉       | 1875/6470 [23:07<57:53,  1.32it/s]
3|1876|Loss: 0.7773217558860779:  29%|██▉       | 1875/6470 [23:08<57:53,  1.32it/s]
3|1876|Loss: 0.7773217558860779:  29%|██▉       | 1876/6470 [23:08<59:03,  1.30it/s]
3|1877|Loss: 0.8832339644432068:  29%|██▉       | 1876/6470 [23:08<59:03,  1.30it/s]
3|1877|Loss: 0.8832339644432068:  29%|██▉       | 1877/6470 [23:09<59:53,  1.28it/s]
3|1878|Loss: 0.8301780819892883:  29%|██▉       | 1877/6470 [23:09<59:53,  1.28it/s]
3|1878|Loss: 0.8301780819892883:  29%|██▉       | 1878/6470 [23:09<59:32,  1.29it/s]
3|1879|Loss: 0.8734245300292969:  29%|██▉       | 1878/6470 [23:10<59:32,  1.29it/s]
3|1879|Loss: 0.8734245300292969:  29%|██▉       | 1879/6470 [23:10<55:00,  1.39it/s]
3|1880|Loss: 0.722261905670166:  29%|██▉       | 1879/6470 [23:10<55:00,  1.39it/s] 
3|1880|Loss: 0.722261905670166:  29%|██▉       | 1880/6470 [23:11<54:45,  1.40it/s]
3|1881|Loss: 0.7660608887672424:  29%|██▉       | 1880/6470 [23:11<54:45,  1.40it/s]
3|1881|Loss: 0.7660608887672424:  29%|██▉       | 1881/6470 [23:12<58:48,  1.30it/s]
3|1882|Loss: 0.8774916529655457:  29%|██▉       | 1881/6470 [23:12<58:48,  1.30it/s]
3|1882|Loss: 0.8774916529655457:  29%|██▉       | 1882/6470 [23:12<1:00:13,  1.27it/s]
3|1883|Loss: 0.7816857099533081:  29%|██▉       | 1882/6470 [23:13<1:00:13,  1.27it/s]
3|1883|Loss: 0.7816857099533081:  29%|██▉       | 1883/6470 [23:13<57:58,  1.32it/s]  
3|1884|Loss: 0.9319594502449036:  29%|██▉       | 1883/6470 [23:14<57:58,  1.32it/s]
3|1884|Loss: 0.9319594502449036:  29%|██▉       | 1884/6470 [23:14<1:00:42,  1.26it/s]
3|1885|Loss: 0.8936285376548767:  29%|██▉       | 1884/6470 [23:15<1:00:42,  1.26it/s]
3|1885|Loss: 0.8936285376548767:  29%|██▉       | 1885/6470 [23:15<1:00:43,  1.26it/s]
3|1886|Loss: 1.0470155477523804:  29%|██▉       | 1885/6470 [23:15<1:00:43,  1.26it/s]
3|1886|Loss: 1.0470155477523804:  29%|██▉       | 1886/6470 [23:16<1:00:37,  1.26it/s]
3|1887|Loss: 0.8093240261077881:  29%|██▉       | 1886/6470 [23:16<1:00:37,  1.26it/s]
3|1887|Loss: 0.8093240261077881:  29%|██▉       | 1887/6470 [23:17<1:02:35,  1.22it/s]
3|1888|Loss: 0.7530102133750916:  29%|██▉       | 1887/6470 [23:17<1:02:35,  1.22it/s]
3|1888|Loss: 0.7530102133750916:  29%|██▉       | 1888/6470 [23:17<59:25,  1.29it/s]  
3|1889|Loss: 0.6878834962844849:  29%|██▉       | 1888/6470 [23:18<59:25,  1.29it/s]
3|1889|Loss: 0.6878834962844849:  29%|██▉       | 1889/6470 [23:18<1:00:04,  1.27it/s]
3|1890|Loss: 0.8341535925865173:  29%|██▉       | 1889/6470 [23:18<1:00:04,  1.27it/s]
3|1890|Loss: 0.8341535925865173:  29%|██▉       | 1890/6470 [23:19<58:44,  1.30it/s]  
3|1891|Loss: 1.0050294399261475:  29%|██▉       | 1890/6470 [23:19<58:44,  1.30it/s]
3|1891|Loss: 1.0050294399261475:  29%|██▉       | 1891/6470 [23:19<58:16,  1.31it/s]
3|1892|Loss: 0.9246824383735657:  29%|██▉       | 1891/6470 [23:20<58:16,  1.31it/s]
3|1892|Loss: 0.9246824383735657:  29%|██▉       | 1892/6470 [23:20<59:06,  1.29it/s]
3|1893|Loss: 0.7757526636123657:  29%|██▉       | 1892/6470 [23:21<59:06,  1.29it/s]
3|1893|Loss: 0.7757526636123657:  29%|██▉       | 1893/6470 [23:21<59:06,  1.29it/s]
3|1894|Loss: 0.9510088562965393:  29%|██▉       | 1893/6470 [23:21<59:06,  1.29it/s]
3|1894|Loss: 0.9510088562965393:  29%|██▉       | 1894/6470 [23:22<59:07,  1.29it/s]
3|1895|Loss: 0.8455024361610413:  29%|██▉       | 1894/6470 [23:22<59:07,  1.29it/s]
3|1895|Loss: 0.8455024361610413:  29%|██▉       | 1895/6470 [23:23<1:02:03,  1.23it/s]
3|1896|Loss: 0.8293821811676025:  29%|██▉       | 1895/6470 [23:23<1:02:03,  1.23it/s]
3|1896|Loss: 0.8293821811676025:  29%|██▉       | 1896/6470 [23:24<1:05:33,  1.16it/s]
3|1897|Loss: 0.783043622970581:  29%|██▉       | 1896/6470 [23:24<1:05:33,  1.16it/s] 
3|1897|Loss: 0.783043622970581:  29%|██▉       | 1897/6470 [23:25<1:04:20,  1.18it/s]
3|1898|Loss: 0.9318512678146362:  29%|██▉       | 1897/6470 [23:25<1:04:20,  1.18it/s]
3|1898|Loss: 0.9318512678146362:  29%|██▉       | 1898/6470 [23:25<1:00:47,  1.25it/s]
3|1899|Loss: 0.7972192168235779:  29%|██▉       | 1898/6470 [23:26<1:00:47,  1.25it/s]
3|1899|Loss: 0.7972192168235779:  29%|██▉       | 1899/6470 [23:26<59:19,  1.28it/s]  
3|1900|Loss: 0.6941941976547241:  29%|██▉       | 1899/6470 [23:26<59:19,  1.28it/s]
3|1900|Loss: 0.6941941976547241:  29%|██▉       | 1900/6470 [23:27<58:49,  1.29it/s]
3|1901|Loss: 0.8751170039176941:  29%|██▉       | 1900/6470 [23:27<58:49,  1.29it/s]
3|1901|Loss: 0.8751170039176941:  29%|██▉       | 1901/6470 [23:27<57:24,  1.33it/s]
3|1902|Loss: 0.7502437233924866:  29%|██▉       | 1901/6470 [23:28<57:24,  1.33it/s]
3|1902|Loss: 0.7502437233924866:  29%|██▉       | 1902/6470 [23:28<56:13,  1.35it/s]
3|1903|Loss: 0.8107239603996277:  29%|██▉       | 1902/6470 [23:29<56:13,  1.35it/s]
3|1903|Loss: 0.8107239603996277:  29%|██▉       | 1903/6470 [23:29<55:32,  1.37it/s]
3|1904|Loss: 0.923686146736145:  29%|██▉       | 1903/6470 [23:29<55:32,  1.37it/s] 
3|1904|Loss: 0.923686146736145:  29%|██▉       | 1904/6470 [23:30<54:54,  1.39it/s]
3|1905|Loss: 0.9101436138153076:  29%|██▉       | 1904/6470 [23:30<54:54,  1.39it/s]
3|1905|Loss: 0.9101436138153076:  29%|██▉       | 1905/6470 [23:30<52:59,  1.44it/s]
3|1906|Loss: 0.8299019932746887:  29%|██▉       | 1905/6470 [23:31<52:59,  1.44it/s]
3|1906|Loss: 0.8299019932746887:  29%|██▉       | 1906/6470 [23:31<51:13,  1.48it/s]
3|1907|Loss: 0.9143158197402954:  29%|██▉       | 1906/6470 [23:31<51:13,  1.48it/s]
3|1907|Loss: 0.9143158197402954:  29%|██▉       | 1907/6470 [23:32<54:39,  1.39it/s]
3|1908|Loss: 0.7813456058502197:  29%|██▉       | 1907/6470 [23:32<54:39,  1.39it/s]
3|1908|Loss: 0.7813456058502197:  29%|██▉       | 1908/6470 [23:32<52:25,  1.45it/s]
3|1909|Loss: 0.8916017413139343:  29%|██▉       | 1908/6470 [23:33<52:25,  1.45it/s]
3|1909|Loss: 0.8916017413139343:  30%|██▉       | 1909/6470 [23:33<51:06,  1.49it/s]
3|1910|Loss: 0.7963930368423462:  30%|██▉       | 1909/6470 [23:33<51:06,  1.49it/s]
3|1910|Loss: 0.7963930368423462:  30%|██▉       | 1910/6470 [23:33<50:25,  1.51it/s]
3|1911|Loss: 0.8868052363395691:  30%|██▉       | 1910/6470 [23:34<50:25,  1.51it/s]
3|1911|Loss: 0.8868052363395691:  30%|██▉       | 1911/6470 [23:34<53:45,  1.41it/s]
3|1912|Loss: 0.8276245594024658:  30%|██▉       | 1911/6470 [23:35<53:45,  1.41it/s]
3|1912|Loss: 0.8276245594024658:  30%|██▉       | 1912/6470 [23:35<50:29,  1.50it/s]
3|1913|Loss: 0.8042556643486023:  30%|██▉       | 1912/6470 [23:35<50:29,  1.50it/s]
3|1913|Loss: 0.8042556643486023:  30%|██▉       | 1913/6470 [23:36<50:54,  1.49it/s]
3|1914|Loss: 0.8665704727172852:  30%|██▉       | 1913/6470 [23:36<50:54,  1.49it/s]
3|1914|Loss: 0.8665704727172852:  30%|██▉       | 1914/6470 [23:36<49:05,  1.55it/s]
3|1915|Loss: 0.7714234590530396:  30%|██▉       | 1914/6470 [23:37<49:05,  1.55it/s]
3|1915|Loss: 0.7714234590530396:  30%|██▉       | 1915/6470 [23:37<50:34,  1.50it/s]
3|1916|Loss: 0.8353098630905151:  30%|██▉       | 1915/6470 [23:37<50:34,  1.50it/s]
3|1916|Loss: 0.8353098630905151:  30%|██▉       | 1916/6470 [23:38<51:05,  1.49it/s]
3|1917|Loss: 0.7099018692970276:  30%|██▉       | 1916/6470 [23:38<51:05,  1.49it/s]
3|1917|Loss: 0.7099018692970276:  30%|██▉       | 1917/6470 [23:38<56:08,  1.35it/s]
3|1918|Loss: 0.9382601380348206:  30%|██▉       | 1917/6470 [23:39<56:08,  1.35it/s]
3|1918|Loss: 0.9382601380348206:  30%|██▉       | 1918/6470 [23:39<59:49,  1.27it/s]
3|1919|Loss: 0.9425920248031616:  30%|██▉       | 1918/6470 [23:40<59:49,  1.27it/s]
3|1919|Loss: 0.9425920248031616:  30%|██▉       | 1919/6470 [23:40<1:00:21,  1.26it/s]
3|1920|Loss: 0.8648655414581299:  30%|██▉       | 1919/6470 [23:41<1:00:21,  1.26it/s]
3|1920|Loss: 0.8648655414581299:  30%|██▉       | 1920/6470 [23:41<58:04,  1.31it/s]  
3|1921|Loss: 0.8745232224464417:  30%|██▉       | 1920/6470 [23:41<58:04,  1.31it/s]
3|1921|Loss: 0.8745232224464417:  30%|██▉       | 1921/6470 [23:42<58:32,  1.30it/s]
3|1922|Loss: 0.8019346594810486:  30%|██▉       | 1921/6470 [23:42<58:32,  1.30it/s]
3|1922|Loss: 0.8019346594810486:  30%|██▉       | 1922/6470 [23:42<56:05,  1.35it/s]
3|1923|Loss: 0.8301628828048706:  30%|██▉       | 1922/6470 [23:43<56:05,  1.35it/s]
3|1923|Loss: 0.8301628828048706:  30%|██▉       | 1923/6470 [23:43<56:16,  1.35it/s]
3|1924|Loss: 0.7872197031974792:  30%|██▉       | 1923/6470 [23:43<56:16,  1.35it/s]
3|1924|Loss: 0.7872197031974792:  30%|██▉       | 1924/6470 [23:44<57:17,  1.32it/s]
3|1925|Loss: 0.7192631959915161:  30%|██▉       | 1924/6470 [23:44<57:17,  1.32it/s]
3|1925|Loss: 0.7192631959915161:  30%|██▉       | 1925/6470 [23:45<1:03:01,  1.20it/s]
3|1926|Loss: 0.815625011920929:  30%|██▉       | 1925/6470 [23:45<1:03:01,  1.20it/s] 
3|1926|Loss: 0.815625011920929:  30%|██▉       | 1926/6470 [23:46<1:01:11,  1.24it/s]
3|1927|Loss: 0.7290188074111938:  30%|██▉       | 1926/6470 [23:46<1:01:11,  1.24it/s]
3|1927|Loss: 0.7290188074111938:  30%|██▉       | 1927/6470 [23:47<1:03:40,  1.19it/s]
3|1928|Loss: 0.8456613421440125:  30%|██▉       | 1927/6470 [23:47<1:03:40,  1.19it/s]
3|1928|Loss: 0.8456613421440125:  30%|██▉       | 1928/6470 [23:47<59:24,  1.27it/s]  
3|1929|Loss: 0.6717113852500916:  30%|██▉       | 1928/6470 [23:48<59:24,  1.27it/s]
3|1929|Loss: 0.6717113852500916:  30%|██▉       | 1929/6470 [23:48<55:57,  1.35it/s]
3|1930|Loss: 0.6909812688827515:  30%|██▉       | 1929/6470 [23:48<55:57,  1.35it/s]
3|1930|Loss: 0.6909812688827515:  30%|██▉       | 1930/6470 [23:49<1:00:11,  1.26it/s]
3|1931|Loss: 0.954720139503479:  30%|██▉       | 1930/6470 [23:49<1:00:11,  1.26it/s] 
3|1931|Loss: 0.954720139503479:  30%|██▉       | 1931/6470 [23:50<1:01:33,  1.23it/s]
3|1932|Loss: 0.8903159499168396:  30%|██▉       | 1931/6470 [23:50<1:01:33,  1.23it/s]
3|1932|Loss: 0.8903159499168396:  30%|██▉       | 1932/6470 [23:50<1:01:34,  1.23it/s]
3|1933|Loss: 0.6894989013671875:  30%|██▉       | 1932/6470 [23:51<1:01:34,  1.23it/s]
3|1933|Loss: 0.6894989013671875:  30%|██▉       | 1933/6470 [23:51<55:26,  1.36it/s]  
3|1934|Loss: 0.8770224452018738:  30%|██▉       | 1933/6470 [23:51<55:26,  1.36it/s]
3|1934|Loss: 0.8770224452018738:  30%|██▉       | 1934/6470 [23:52<53:28,  1.41it/s]
3|1935|Loss: 0.7703196406364441:  30%|██▉       | 1934/6470 [23:52<53:28,  1.41it/s]
3|1935|Loss: 0.7703196406364441:  30%|██▉       | 1935/6470 [23:52<53:21,  1.42it/s]
3|1936|Loss: 0.8750391006469727:  30%|██▉       | 1935/6470 [23:53<53:21,  1.42it/s]
3|1936|Loss: 0.8750391006469727:  30%|██▉       | 1936/6470 [23:53<52:06,  1.45it/s]
3|1937|Loss: 0.9271093606948853:  30%|██▉       | 1936/6470 [23:53<52:06,  1.45it/s]
3|1937|Loss: 0.9271093606948853:  30%|██▉       | 1937/6470 [23:54<48:59,  1.54it/s]
3|1938|Loss: 0.7911656498908997:  30%|██▉       | 1937/6470 [23:54<48:59,  1.54it/s]
3|1938|Loss: 0.7911656498908997:  30%|██▉       | 1938/6470 [23:54<48:08,  1.57it/s]
3|1939|Loss: 0.7729499936103821:  30%|██▉       | 1938/6470 [23:55<48:08,  1.57it/s]
3|1939|Loss: 0.7729499936103821:  30%|██▉       | 1939/6470 [23:55<51:01,  1.48it/s]
3|1940|Loss: 0.8537531495094299:  30%|██▉       | 1939/6470 [23:55<51:01,  1.48it/s]
3|1940|Loss: 0.8537531495094299:  30%|██▉       | 1940/6470 [23:56<50:28,  1.50it/s]
3|1941|Loss: 0.7308064699172974:  30%|██▉       | 1940/6470 [23:56<50:28,  1.50it/s]
3|1941|Loss: 0.7308064699172974:  30%|███       | 1941/6470 [23:56<52:35,  1.44it/s]
3|1942|Loss: 0.7861639261245728:  30%|███       | 1941/6470 [23:57<52:35,  1.44it/s]
3|1942|Loss: 0.7861639261245728:  30%|███       | 1942/6470 [23:57<52:34,  1.44it/s]
3|1943|Loss: 0.7908653020858765:  30%|███       | 1942/6470 [23:57<52:34,  1.44it/s]
3|1943|Loss: 0.7908653020858765:  30%|███       | 1943/6470 [23:58<54:28,  1.39it/s]
3|1944|Loss: 0.8145081996917725:  30%|███       | 1943/6470 [23:58<54:28,  1.39it/s]
3|1944|Loss: 0.8145081996917725:  30%|███       | 1944/6470 [23:58<51:38,  1.46it/s]
3|1945|Loss: 0.7410380244255066:  30%|███       | 1944/6470 [23:59<51:38,  1.46it/s]
3|1945|Loss: 0.7410380244255066:  30%|███       | 1945/6470 [23:59<50:17,  1.50it/s]
3|1946|Loss: 0.7659977674484253:  30%|███       | 1945/6470 [23:59<50:17,  1.50it/s]
3|1946|Loss: 0.7659977674484253:  30%|███       | 1946/6470 [24:00<51:51,  1.45it/s]
3|1947|Loss: 0.9409675598144531:  30%|███       | 1946/6470 [24:00<51:51,  1.45it/s]
3|1947|Loss: 0.9409675598144531:  30%|███       | 1947/6470 [24:01<54:58,  1.37it/s]
3|1948|Loss: 0.6325768232345581:  30%|███       | 1947/6470 [24:01<54:58,  1.37it/s]
3|1948|Loss: 0.6325768232345581:  30%|███       | 1948/6470 [24:01<56:37,  1.33it/s]
3|1949|Loss: 0.7815325856208801:  30%|███       | 1948/6470 [24:02<56:37,  1.33it/s]
3|1949|Loss: 0.7815325856208801:  30%|███       | 1949/6470 [24:02<1:02:07,  1.21it/s]
3|1950|Loss: 0.7664000988006592:  30%|███       | 1949/6470 [24:03<1:02:07,  1.21it/s]
3|1950|Loss: 0.7664000988006592:  30%|███       | 1950/6470 [24:03<59:51,  1.26it/s]  
3|1951|Loss: 0.9882187843322754:  30%|███       | 1950/6470 [24:03<59:51,  1.26it/s]
3|1951|Loss: 0.9882187843322754:  30%|███       | 1951/6470 [24:04<58:24,  1.29it/s]
3|1952|Loss: 0.9273735880851746:  30%|███       | 1951/6470 [24:04<58:24,  1.29it/s]
3|1952|Loss: 0.9273735880851746:  30%|███       | 1952/6470 [24:04<56:19,  1.34it/s]
3|1953|Loss: 0.7771667242050171:  30%|███       | 1952/6470 [24:05<56:19,  1.34it/s]
3|1953|Loss: 0.7771667242050171:  30%|███       | 1953/6470 [24:05<52:15,  1.44it/s]
3|1954|Loss: 0.9543644189834595:  30%|███       | 1953/6470 [24:06<52:15,  1.44it/s]
3|1954|Loss: 0.9543644189834595:  30%|███       | 1954/6470 [24:06<57:00,  1.32it/s]
3|1955|Loss: 0.7973482012748718:  30%|███       | 1954/6470 [24:06<57:00,  1.32it/s]
3|1955|Loss: 0.7973482012748718:  30%|███       | 1955/6470 [24:07<56:47,  1.33it/s]
3|1956|Loss: 0.8555604219436646:  30%|███       | 1955/6470 [24:07<56:47,  1.33it/s]
3|1956|Loss: 0.8555604219436646:  30%|███       | 1956/6470 [24:07<56:45,  1.33it/s]
3|1957|Loss: 0.7131380438804626:  30%|███       | 1956/6470 [24:08<56:45,  1.33it/s]
3|1957|Loss: 0.7131380438804626:  30%|███       | 1957/6470 [24:08<55:37,  1.35it/s]
3|1958|Loss: 0.8761264681816101:  30%|███       | 1957/6470 [24:08<55:37,  1.35it/s]
3|1958|Loss: 0.8761264681816101:  30%|███       | 1958/6470 [24:09<50:29,  1.49it/s]
3|1959|Loss: 0.8821607232093811:  30%|███       | 1958/6470 [24:09<50:29,  1.49it/s]
3|1959|Loss: 0.8821607232093811:  30%|███       | 1959/6470 [24:09<49:16,  1.53it/s]
3|1960|Loss: 0.873510479927063:  30%|███       | 1959/6470 [24:10<49:16,  1.53it/s] 
3|1960|Loss: 0.873510479927063:  30%|███       | 1960/6470 [24:10<51:45,  1.45it/s]
3|1961|Loss: 0.8313237428665161:  30%|███       | 1960/6470 [24:10<51:45,  1.45it/s]
3|1961|Loss: 0.8313237428665161:  30%|███       | 1961/6470 [24:11<51:58,  1.45it/s]
3|1962|Loss: 0.8900266289710999:  30%|███       | 1961/6470 [24:11<51:58,  1.45it/s]
3|1962|Loss: 0.8900266289710999:  30%|███       | 1962/6470 [24:12<57:22,  1.31it/s]
3|1963|Loss: 0.7823777198791504:  30%|███       | 1962/6470 [24:12<57:22,  1.31it/s]
3|1963|Loss: 0.7823777198791504:  30%|███       | 1963/6470 [24:12<54:30,  1.38it/s]
3|1964|Loss: 0.8300646543502808:  30%|███       | 1963/6470 [24:13<54:30,  1.38it/s]
3|1964|Loss: 0.8300646543502808:  30%|███       | 1964/6470 [24:13<53:14,  1.41it/s]
3|1965|Loss: 0.9783564209938049:  30%|███       | 1964/6470 [24:13<53:14,  1.41it/s]
3|1965|Loss: 0.9783564209938049:  30%|███       | 1965/6470 [24:14<54:54,  1.37it/s]
3|1966|Loss: 0.7827345132827759:  30%|███       | 1965/6470 [24:14<54:54,  1.37it/s]
3|1966|Loss: 0.7827345132827759:  30%|███       | 1966/6470 [24:14<51:49,  1.45it/s]
3|1967|Loss: 0.9601144790649414:  30%|███       | 1966/6470 [24:15<51:49,  1.45it/s]
3|1967|Loss: 0.9601144790649414:  30%|███       | 1967/6470 [24:15<55:14,  1.36it/s]
3|1968|Loss: 0.8830604553222656:  30%|███       | 1967/6470 [24:16<55:14,  1.36it/s]
3|1968|Loss: 0.8830604553222656:  30%|███       | 1968/6470 [24:16<58:51,  1.27it/s]
3|1969|Loss: 0.8038040995597839:  30%|███       | 1968/6470 [24:17<58:51,  1.27it/s]
3|1969|Loss: 0.8038040995597839:  30%|███       | 1969/6470 [24:17<58:41,  1.28it/s]
3|1970|Loss: 0.8693217635154724:  30%|███       | 1969/6470 [24:17<58:41,  1.28it/s]
3|1970|Loss: 0.8693217635154724:  30%|███       | 1970/6470 [24:18<1:01:03,  1.23it/s]
3|1971|Loss: 0.7897164821624756:  30%|███       | 1970/6470 [24:18<1:01:03,  1.23it/s]
3|1971|Loss: 0.7897164821624756:  30%|███       | 1971/6470 [24:19<59:33,  1.26it/s]  
3|1972|Loss: 0.8379162549972534:  30%|███       | 1971/6470 [24:19<59:33,  1.26it/s]
3|1972|Loss: 0.8379162549972534:  30%|███       | 1972/6470 [24:19<57:28,  1.30it/s]
3|1973|Loss: 0.7518174052238464:  30%|███       | 1972/6470 [24:20<57:28,  1.30it/s]
3|1973|Loss: 0.7518174052238464:  30%|███       | 1973/6470 [24:20<1:00:26,  1.24it/s]
3|1974|Loss: 0.7831762433052063:  30%|███       | 1973/6470 [24:21<1:00:26,  1.24it/s]
3|1974|Loss: 0.7831762433052063:  31%|███       | 1974/6470 [24:21<1:01:11,  1.22it/s]
3|1975|Loss: 0.8783348202705383:  31%|███       | 1974/6470 [24:21<1:01:11,  1.22it/s]
3|1975|Loss: 0.8783348202705383:  31%|███       | 1975/6470 [24:22<1:02:25,  1.20it/s]
3|1976|Loss: 0.8427063226699829:  31%|███       | 1975/6470 [24:22<1:02:25,  1.20it/s]
3|1976|Loss: 0.8427063226699829:  31%|███       | 1976/6470 [24:23<1:01:41,  1.21it/s]
3|1977|Loss: 0.9113922119140625:  31%|███       | 1976/6470 [24:23<1:01:41,  1.21it/s]
3|1977|Loss: 0.9113922119140625:  31%|███       | 1977/6470 [24:23<1:00:55,  1.23it/s]
3|1978|Loss: 0.7790010571479797:  31%|███       | 1977/6470 [24:24<1:00:55,  1.23it/s]
3|1978|Loss: 0.7790010571479797:  31%|███       | 1978/6470 [24:24<1:01:48,  1.21it/s]
3|1979|Loss: 0.7911542654037476:  31%|███       | 1978/6470 [24:25<1:01:48,  1.21it/s]
3|1979|Loss: 0.7911542654037476:  31%|███       | 1979/6470 [24:25<57:49,  1.29it/s]  
3|1980|Loss: 0.9098833799362183:  31%|███       | 1979/6470 [24:25<57:49,  1.29it/s]
3|1980|Loss: 0.9098833799362183:  31%|███       | 1980/6470 [24:26<56:36,  1.32it/s]
3|1981|Loss: 0.9098173379898071:  31%|███       | 1980/6470 [24:26<56:36,  1.32it/s]
3|1981|Loss: 0.9098173379898071:  31%|███       | 1981/6470 [24:26<57:18,  1.31it/s]
3|1982|Loss: 0.9313030242919922:  31%|███       | 1981/6470 [24:27<57:18,  1.31it/s]
3|1982|Loss: 0.9313030242919922:  31%|███       | 1982/6470 [24:27<57:31,  1.30it/s]
3|1983|Loss: 0.8499210476875305:  31%|███       | 1982/6470 [24:28<57:31,  1.30it/s]
3|1983|Loss: 0.8499210476875305:  31%|███       | 1983/6470 [24:28<53:20,  1.40it/s]
3|1984|Loss: 0.8222982287406921:  31%|███       | 1983/6470 [24:28<53:20,  1.40it/s]
3|1984|Loss: 0.8222982287406921:  31%|███       | 1984/6470 [24:29<53:16,  1.40it/s]
3|1985|Loss: 0.7453665733337402:  31%|███       | 1984/6470 [24:29<53:16,  1.40it/s]
3|1985|Loss: 0.7453665733337402:  31%|███       | 1985/6470 [24:29<56:49,  1.32it/s]
3|1986|Loss: 0.88311767578125:  31%|███       | 1985/6470 [24:30<56:49,  1.32it/s]  
3|1986|Loss: 0.88311767578125:  31%|███       | 1986/6470 [24:30<58:03,  1.29it/s]
3|1987|Loss: 0.9784756302833557:  31%|███       | 1986/6470 [24:31<58:03,  1.29it/s]
3|1987|Loss: 0.9784756302833557:  31%|███       | 1987/6470 [24:31<56:39,  1.32it/s]
3|1988|Loss: 0.7579469084739685:  31%|███       | 1987/6470 [24:31<56:39,  1.32it/s]
3|1988|Loss: 0.7579469084739685:  31%|███       | 1988/6470 [24:32<52:31,  1.42it/s]
3|1989|Loss: 0.9023782014846802:  31%|███       | 1988/6470 [24:32<52:31,  1.42it/s]
3|1989|Loss: 0.9023782014846802:  31%|███       | 1989/6470 [24:32<53:32,  1.39it/s]
3|1990|Loss: 0.9616974592208862:  31%|███       | 1989/6470 [24:33<53:32,  1.39it/s]
3|1990|Loss: 0.9616974592208862:  31%|███       | 1990/6470 [24:33<55:12,  1.35it/s]
3|1991|Loss: 0.9125076532363892:  31%|███       | 1990/6470 [24:34<55:12,  1.35it/s]
3|1991|Loss: 0.9125076532363892:  31%|███       | 1991/6470 [24:34<58:46,  1.27it/s]
3|1992|Loss: 0.8846928477287292:  31%|███       | 1991/6470 [24:34<58:46,  1.27it/s]
3|1992|Loss: 0.8846928477287292:  31%|███       | 1992/6470 [24:35<1:01:18,  1.22it/s]
3|1993|Loss: 0.6910974979400635:  31%|███       | 1992/6470 [24:35<1:01:18,  1.22it/s]
3|1993|Loss: 0.6910974979400635:  31%|███       | 1993/6470 [24:35<53:34,  1.39it/s]  
3|1994|Loss: 0.7840501666069031:  31%|███       | 1993/6470 [24:36<53:34,  1.39it/s]
3|1994|Loss: 0.7840501666069031:  31%|███       | 1994/6470 [24:36<53:15,  1.40it/s]
3|1995|Loss: 0.9034115076065063:  31%|███       | 1994/6470 [24:36<53:15,  1.40it/s]
3|1995|Loss: 0.9034115076065063:  31%|███       | 1995/6470 [24:37<55:15,  1.35it/s]
3|1996|Loss: 0.9087817072868347:  31%|███       | 1995/6470 [24:37<55:15,  1.35it/s]
3|1996|Loss: 0.9087817072868347:  31%|███       | 1996/6470 [24:38<54:28,  1.37it/s]
3|1997|Loss: 0.8337383270263672:  31%|███       | 1996/6470 [24:38<54:28,  1.37it/s]
3|1997|Loss: 0.8337383270263672:  31%|███       | 1997/6470 [24:38<57:18,  1.30it/s]
3|1998|Loss: 0.7440490126609802:  31%|███       | 1997/6470 [24:39<57:18,  1.30it/s]
3|1998|Loss: 0.7440490126609802:  31%|███       | 1998/6470 [24:39<56:01,  1.33it/s]
3|1999|Loss: 0.9021887183189392:  31%|███       | 1998/6470 [24:40<56:01,  1.33it/s]
3|1999|Loss: 0.9021887183189392:  31%|███       | 1999/6470 [24:40<55:40,  1.34it/s]
3|2000|Loss: 0.9413312077522278:  31%|███       | 1999/6470 [24:40<55:40,  1.34it/s]
3|2000|Loss: 0.9413312077522278:  31%|███       | 2000/6470 [24:40<51:25,  1.45it/s]
3|2001|Loss: 0.8053754568099976:  31%|███       | 2000/6470 [24:41<51:25,  1.45it/s]
3|2001|Loss: 0.8053754568099976:  31%|███       | 2001/6470 [24:41<54:31,  1.37it/s]
3|2002|Loss: 0.8271107077598572:  31%|███       | 2001/6470 [24:42<54:31,  1.37it/s]
3|2002|Loss: 0.8271107077598572:  31%|███       | 2002/6470 [24:42<55:26,  1.34it/s]
3|2003|Loss: 0.731080174446106:  31%|███       | 2002/6470 [24:42<55:26,  1.34it/s] 
3|2003|Loss: 0.731080174446106:  31%|███       | 2003/6470 [24:43<54:52,  1.36it/s]
3|2004|Loss: 0.8248664736747742:  31%|███       | 2003/6470 [24:43<54:52,  1.36it/s]
3|2004|Loss: 0.8248664736747742:  31%|███       | 2004/6470 [24:43<53:35,  1.39it/s]
3|2005|Loss: 0.8174216747283936:  31%|███       | 2004/6470 [24:44<53:35,  1.39it/s]
3|2005|Loss: 0.8174216747283936:  31%|███       | 2005/6470 [24:44<51:42,  1.44it/s]
3|2006|Loss: 0.8278385400772095:  31%|███       | 2005/6470 [24:44<51:42,  1.44it/s]
3|2006|Loss: 0.8278385400772095:  31%|███       | 2006/6470 [24:45<52:27,  1.42it/s]
3|2007|Loss: 0.7404142618179321:  31%|███       | 2006/6470 [24:45<52:27,  1.42it/s]
3|2007|Loss: 0.7404142618179321:  31%|███       | 2007/6470 [24:45<51:01,  1.46it/s]
3|2008|Loss: 0.7900440692901611:  31%|███       | 2007/6470 [24:46<51:01,  1.46it/s]
3|2008|Loss: 0.7900440692901611:  31%|███       | 2008/6470 [24:46<52:23,  1.42it/s]
3|2009|Loss: 0.8761501312255859:  31%|███       | 2008/6470 [24:47<52:23,  1.42it/s]
3|2009|Loss: 0.8761501312255859:  31%|███       | 2009/6470 [24:47<56:16,  1.32it/s]
3|2010|Loss: 0.894256591796875:  31%|███       | 2009/6470 [24:47<56:16,  1.32it/s] 
3|2010|Loss: 0.894256591796875:  31%|███       | 2010/6470 [24:48<55:46,  1.33it/s]
3|2011|Loss: 0.7458800077438354:  31%|███       | 2010/6470 [24:48<55:46,  1.33it/s]
3|2011|Loss: 0.7458800077438354:  31%|███       | 2011/6470 [24:49<55:29,  1.34it/s]
3|2012|Loss: 0.8051849603652954:  31%|███       | 2011/6470 [24:49<55:29,  1.34it/s]
3|2012|Loss: 0.8051849603652954:  31%|███       | 2012/6470 [24:49<57:10,  1.30it/s]
3|2013|Loss: 0.8498620390892029:  31%|███       | 2012/6470 [24:50<57:10,  1.30it/s]
3|2013|Loss: 0.8498620390892029:  31%|███       | 2013/6470 [24:50<58:05,  1.28it/s]
3|2014|Loss: 0.8724744319915771:  31%|███       | 2013/6470 [24:51<58:05,  1.28it/s]
3|2014|Loss: 0.8724744319915771:  31%|███       | 2014/6470 [24:51<57:46,  1.29it/s]
3|2015|Loss: 0.9021303057670593:  31%|███       | 2014/6470 [24:51<57:46,  1.29it/s]
3|2015|Loss: 0.9021303057670593:  31%|███       | 2015/6470 [24:52<54:59,  1.35it/s]
3|2016|Loss: 0.6980032920837402:  31%|███       | 2015/6470 [24:52<54:59,  1.35it/s]
3|2016|Loss: 0.6980032920837402:  31%|███       | 2016/6470 [24:52<51:12,  1.45it/s]
3|2017|Loss: 0.8761292099952698:  31%|███       | 2016/6470 [24:53<51:12,  1.45it/s]
3|2017|Loss: 0.8761292099952698:  31%|███       | 2017/6470 [24:53<50:41,  1.46it/s]
3|2018|Loss: 0.8979747295379639:  31%|███       | 2017/6470 [24:53<50:41,  1.46it/s]
3|2018|Loss: 0.8979747295379639:  31%|███       | 2018/6470 [24:53<50:52,  1.46it/s]
3|2019|Loss: 0.8738804459571838:  31%|███       | 2018/6470 [24:54<50:52,  1.46it/s]
3|2019|Loss: 0.8738804459571838:  31%|███       | 2019/6470 [24:54<53:06,  1.40it/s]
3|2020|Loss: 0.8953713774681091:  31%|███       | 2019/6470 [24:55<53:06,  1.40it/s]
3|2020|Loss: 0.8953713774681091:  31%|███       | 2020/6470 [24:55<55:00,  1.35it/s]
3|2021|Loss: 0.731087863445282:  31%|███       | 2020/6470 [24:55<55:00,  1.35it/s] 
3|2021|Loss: 0.731087863445282:  31%|███       | 2021/6470 [24:56<54:43,  1.35it/s]
3|2022|Loss: 0.9163869023323059:  31%|███       | 2021/6470 [24:56<54:43,  1.35it/s]
3|2022|Loss: 0.9163869023323059:  31%|███▏      | 2022/6470 [24:57<54:03,  1.37it/s]
3|2023|Loss: 0.9585452079772949:  31%|███▏      | 2022/6470 [24:57<54:03,  1.37it/s]
3|2023|Loss: 0.9585452079772949:  31%|███▏      | 2023/6470 [24:57<53:21,  1.39it/s]
3|2024|Loss: 0.9209434390068054:  31%|███▏      | 2023/6470 [24:58<53:21,  1.39it/s]
3|2024|Loss: 0.9209434390068054:  31%|███▏      | 2024/6470 [24:58<53:55,  1.37it/s]
3|2025|Loss: 0.878585934638977:  31%|███▏      | 2024/6470 [24:58<53:55,  1.37it/s] 
3|2025|Loss: 0.878585934638977:  31%|███▏      | 2025/6470 [24:59<53:32,  1.38it/s]
3|2026|Loss: 1.0654075145721436:  31%|███▏      | 2025/6470 [24:59<53:32,  1.38it/s]
3|2026|Loss: 1.0654075145721436:  31%|███▏      | 2026/6470 [24:59<54:50,  1.35it/s]
3|2027|Loss: 0.8694935441017151:  31%|███▏      | 2026/6470 [25:00<54:50,  1.35it/s]
3|2027|Loss: 0.8694935441017151:  31%|███▏      | 2027/6470 [25:00<55:53,  1.33it/s]
3|2028|Loss: 0.7208935618400574:  31%|███▏      | 2027/6470 [25:01<55:53,  1.33it/s]
3|2028|Loss: 0.7208935618400574:  31%|███▏      | 2028/6470 [25:01<53:46,  1.38it/s]
3|2029|Loss: 0.8660606145858765:  31%|███▏      | 2028/6470 [25:01<53:46,  1.38it/s]
3|2029|Loss: 0.8660606145858765:  31%|███▏      | 2029/6470 [25:02<53:34,  1.38it/s]
3|2030|Loss: 0.8287299275398254:  31%|███▏      | 2029/6470 [25:02<53:34,  1.38it/s]
3|2030|Loss: 0.8287299275398254:  31%|███▏      | 2030/6470 [25:02<54:53,  1.35it/s]
3|2031|Loss: 1.057295322418213:  31%|███▏      | 2030/6470 [25:03<54:53,  1.35it/s] 
3|2031|Loss: 1.057295322418213:  31%|███▏      | 2031/6470 [25:03<53:49,  1.37it/s]
3|2032|Loss: 0.7041958570480347:  31%|███▏      | 2031/6470 [25:04<53:49,  1.37it/s]
3|2032|Loss: 0.7041958570480347:  31%|███▏      | 2032/6470 [25:04<57:09,  1.29it/s]
3|2033|Loss: 0.7922916412353516:  31%|███▏      | 2032/6470 [25:04<57:09,  1.29it/s]
3|2033|Loss: 0.7922916412353516:  31%|███▏      | 2033/6470 [25:05<58:33,  1.26it/s]
3|2034|Loss: 0.8760545253753662:  31%|███▏      | 2033/6470 [25:05<58:33,  1.26it/s]
3|2034|Loss: 0.8760545253753662:  31%|███▏      | 2034/6470 [25:06<58:48,  1.26it/s]
3|2035|Loss: 0.7551018595695496:  31%|███▏      | 2034/6470 [25:06<58:48,  1.26it/s]
3|2035|Loss: 0.7551018595695496:  31%|███▏      | 2035/6470 [25:06<59:00,  1.25it/s]
3|2036|Loss: 0.8077462911605835:  31%|███▏      | 2035/6470 [25:07<59:00,  1.25it/s]
3|2036|Loss: 0.8077462911605835:  31%|███▏      | 2036/6470 [25:07<56:48,  1.30it/s]
3|2037|Loss: 0.703740656375885:  31%|███▏      | 2036/6470 [25:08<56:48,  1.30it/s] 
3|2037|Loss: 0.703740656375885:  31%|███▏      | 2037/6470 [25:08<54:28,  1.36it/s]
3|2038|Loss: 0.7192733287811279:  31%|███▏      | 2037/6470 [25:08<54:28,  1.36it/s]
3|2038|Loss: 0.7192733287811279:  31%|███▏      | 2038/6470 [25:08<52:02,  1.42it/s]
3|2039|Loss: 0.8594640493392944:  31%|███▏      | 2038/6470 [25:09<52:02,  1.42it/s]
3|2039|Loss: 0.8594640493392944:  32%|███▏      | 2039/6470 [25:09<52:30,  1.41it/s]
3|2040|Loss: 0.8801586627960205:  32%|███▏      | 2039/6470 [25:10<52:30,  1.41it/s]
3|2040|Loss: 0.8801586627960205:  32%|███▏      | 2040/6470 [25:10<56:55,  1.30it/s]
3|2041|Loss: 0.9068809151649475:  32%|███▏      | 2040/6470 [25:11<56:55,  1.30it/s]
3|2041|Loss: 0.9068809151649475:  32%|███▏      | 2041/6470 [25:11<1:00:45,  1.21it/s]
3|2042|Loss: 0.7353852391242981:  32%|███▏      | 2041/6470 [25:12<1:00:45,  1.21it/s]
3|2042|Loss: 0.7353852391242981:  32%|███▏      | 2042/6470 [25:12<1:02:08,  1.19it/s]
3|2043|Loss: 0.8067739009857178:  32%|███▏      | 2042/6470 [25:12<1:02:08,  1.19it/s]
3|2043|Loss: 0.8067739009857178:  32%|███▏      | 2043/6470 [25:13<1:04:07,  1.15it/s]
3|2044|Loss: 0.9535062909126282:  32%|███▏      | 2043/6470 [25:13<1:04:07,  1.15it/s]
3|2044|Loss: 0.9535062909126282:  32%|███▏      | 2044/6470 [25:14<1:05:32,  1.13it/s]
3|2045|Loss: 0.7031847834587097:  32%|███▏      | 2044/6470 [25:14<1:05:32,  1.13it/s]
3|2045|Loss: 0.7031847834587097:  32%|███▏      | 2045/6470 [25:15<1:06:51,  1.10it/s]
3|2046|Loss: 0.7536454200744629:  32%|███▏      | 2045/6470 [25:15<1:06:51,  1.10it/s]
3|2046|Loss: 0.7536454200744629:  32%|███▏      | 2046/6470 [25:16<1:05:18,  1.13it/s]
3|2047|Loss: 0.9297801852226257:  32%|███▏      | 2046/6470 [25:16<1:05:18,  1.13it/s]
3|2047|Loss: 0.9297801852226257:  32%|███▏      | 2047/6470 [25:16<1:05:09,  1.13it/s]
3|2048|Loss: 0.9224883913993835:  32%|███▏      | 2047/6470 [25:17<1:05:09,  1.13it/s]
3|2048|Loss: 0.9224883913993835:  32%|███▏      | 2048/6470 [25:17<1:02:42,  1.18it/s]
3|2049|Loss: 0.7347843050956726:  32%|███▏      | 2048/6470 [25:18<1:02:42,  1.18it/s]
3|2049|Loss: 0.7347843050956726:  32%|███▏      | 2049/6470 [25:18<58:10,  1.27it/s]  
3|2050|Loss: 0.9902759790420532:  32%|███▏      | 2049/6470 [25:18<58:10,  1.27it/s]
3|2050|Loss: 0.9902759790420532:  32%|███▏      | 2050/6470 [25:19<1:00:45,  1.21it/s]
3|2051|Loss: 0.7114983797073364:  32%|███▏      | 2050/6470 [25:19<1:00:45,  1.21it/s]
3|2051|Loss: 0.7114983797073364:  32%|███▏      | 2051/6470 [25:20<1:00:31,  1.22it/s]
3|2052|Loss: 0.8284447193145752:  32%|███▏      | 2051/6470 [25:20<1:00:31,  1.22it/s]
3|2052|Loss: 0.8284447193145752:  32%|███▏      | 2052/6470 [25:20<1:00:08,  1.22it/s]
3|2053|Loss: 0.7943676114082336:  32%|███▏      | 2052/6470 [25:21<1:00:08,  1.22it/s]
3|2053|Loss: 0.7943676114082336:  32%|███▏      | 2053/6470 [25:21<1:02:48,  1.17it/s]
3|2054|Loss: 1.0897345542907715:  32%|███▏      | 2053/6470 [25:22<1:02:48,  1.17it/s]
3|2054|Loss: 1.0897345542907715:  32%|███▏      | 2054/6470 [25:22<58:21,  1.26it/s]  
3|2055|Loss: 0.7069371342658997:  32%|███▏      | 2054/6470 [25:22<58:21,  1.26it/s]
3|2055|Loss: 0.7069371342658997:  32%|███▏      | 2055/6470 [25:22<51:45,  1.42it/s]
3|2056|Loss: 0.8025071024894714:  32%|███▏      | 2055/6470 [25:23<51:45,  1.42it/s]
3|2056|Loss: 0.8025071024894714:  32%|███▏      | 2056/6470 [25:23<52:14,  1.41it/s]
3|2057|Loss: 0.9073055982589722:  32%|███▏      | 2056/6470 [25:24<52:14,  1.41it/s]
3|2057|Loss: 0.9073055982589722:  32%|███▏      | 2057/6470 [25:24<54:10,  1.36it/s]
3|2058|Loss: 0.6694503426551819:  32%|███▏      | 2057/6470 [25:24<54:10,  1.36it/s]
3|2058|Loss: 0.6694503426551819:  32%|███▏      | 2058/6470 [25:25<55:29,  1.33it/s]
3|2059|Loss: 0.6933584809303284:  32%|███▏      | 2058/6470 [25:25<55:29,  1.33it/s]
3|2059|Loss: 0.6933584809303284:  32%|███▏      | 2059/6470 [25:25<54:39,  1.35it/s]
3|2060|Loss: 0.7399491667747498:  32%|███▏      | 2059/6470 [25:26<54:39,  1.35it/s]
3|2060|Loss: 0.7399491667747498:  32%|███▏      | 2060/6470 [25:26<55:01,  1.34it/s]
3|2061|Loss: 0.683495819568634:  32%|███▏      | 2060/6470 [25:27<55:01,  1.34it/s] 
3|2061|Loss: 0.683495819568634:  32%|███▏      | 2061/6470 [25:27<55:52,  1.32it/s]
3|2062|Loss: 0.9091573357582092:  32%|███▏      | 2061/6470 [25:27<55:52,  1.32it/s]
3|2062|Loss: 0.9091573357582092:  32%|███▏      | 2062/6470 [25:28<56:47,  1.29it/s]
3|2063|Loss: 0.7621369361877441:  32%|███▏      | 2062/6470 [25:28<56:47,  1.29it/s]
3|2063|Loss: 0.7621369361877441:  32%|███▏      | 2063/6470 [25:29<55:01,  1.33it/s]
3|2064|Loss: 0.9962648749351501:  32%|███▏      | 2063/6470 [25:29<55:01,  1.33it/s]
3|2064|Loss: 0.9962648749351501:  32%|███▏      | 2064/6470 [25:29<56:59,  1.29it/s]
3|2065|Loss: 0.868208646774292:  32%|███▏      | 2064/6470 [25:30<56:59,  1.29it/s] 
3|2065|Loss: 0.868208646774292:  32%|███▏      | 2065/6470 [25:30<57:36,  1.27it/s]
3|2066|Loss: 0.8596140742301941:  32%|███▏      | 2065/6470 [25:31<57:36,  1.27it/s]
3|2066|Loss: 0.8596140742301941:  32%|███▏      | 2066/6470 [25:31<59:09,  1.24it/s]
3|2067|Loss: 0.8456813097000122:  32%|███▏      | 2066/6470 [25:31<59:09,  1.24it/s]
3|2067|Loss: 0.8456813097000122:  32%|███▏      | 2067/6470 [25:32<57:08,  1.28it/s]
3|2068|Loss: 0.9000088572502136:  32%|███▏      | 2067/6470 [25:32<57:08,  1.28it/s]
3|2068|Loss: 0.9000088572502136:  32%|███▏      | 2068/6470 [25:33<59:11,  1.24it/s]
3|2069|Loss: 0.8746448159217834:  32%|███▏      | 2068/6470 [25:33<59:11,  1.24it/s]
3|2069|Loss: 0.8746448159217834:  32%|███▏      | 2069/6470 [25:33<57:21,  1.28it/s]
3|2070|Loss: 0.8385216593742371:  32%|███▏      | 2069/6470 [25:34<57:21,  1.28it/s]
3|2070|Loss: 0.8385216593742371:  32%|███▏      | 2070/6470 [25:34<59:47,  1.23it/s]
3|2071|Loss: 0.8536971211433411:  32%|███▏      | 2070/6470 [25:35<59:47,  1.23it/s]
3|2071|Loss: 0.8536971211433411:  32%|███▏      | 2071/6470 [25:35<57:36,  1.27it/s]
3|2072|Loss: 0.8980321884155273:  32%|███▏      | 2071/6470 [25:35<57:36,  1.27it/s]
3|2072|Loss: 0.8980321884155273:  32%|███▏      | 2072/6470 [25:36<59:09,  1.24it/s]
3|2073|Loss: 0.8196277022361755:  32%|███▏      | 2072/6470 [25:36<59:09,  1.24it/s]
3|2073|Loss: 0.8196277022361755:  32%|███▏      | 2073/6470 [25:36<56:18,  1.30it/s]
3|2074|Loss: 0.650221586227417:  32%|███▏      | 2073/6470 [25:37<56:18,  1.30it/s] 
3|2074|Loss: 0.650221586227417:  32%|███▏      | 2074/6470 [25:37<58:41,  1.25it/s]
3|2075|Loss: 0.9818060398101807:  32%|███▏      | 2074/6470 [25:38<58:41,  1.25it/s]
3|2075|Loss: 0.9818060398101807:  32%|███▏      | 2075/6470 [25:38<57:49,  1.27it/s]
3|2076|Loss: 0.8451479077339172:  32%|███▏      | 2075/6470 [25:39<57:49,  1.27it/s]
3|2076|Loss: 0.8451479077339172:  32%|███▏      | 2076/6470 [25:39<55:29,  1.32it/s]
3|2077|Loss: 0.8714609742164612:  32%|███▏      | 2076/6470 [25:39<55:29,  1.32it/s]
3|2077|Loss: 0.8714609742164612:  32%|███▏      | 2077/6470 [25:39<52:07,  1.40it/s]
3|2078|Loss: 0.8316354751586914:  32%|███▏      | 2077/6470 [25:40<52:07,  1.40it/s]
3|2078|Loss: 0.8316354751586914:  32%|███▏      | 2078/6470 [25:40<55:14,  1.33it/s]
3|2079|Loss: 0.8458251357078552:  32%|███▏      | 2078/6470 [25:41<55:14,  1.33it/s]
3|2079|Loss: 0.8458251357078552:  32%|███▏      | 2079/6470 [25:41<55:48,  1.31it/s]
3|2080|Loss: 0.7728128433227539:  32%|███▏      | 2079/6470 [25:41<55:48,  1.31it/s]
3|2080|Loss: 0.7728128433227539:  32%|███▏      | 2080/6470 [25:42<53:32,  1.37it/s]
3|2081|Loss: 0.8769990801811218:  32%|███▏      | 2080/6470 [25:42<53:32,  1.37it/s]
3|2081|Loss: 0.8769990801811218:  32%|███▏      | 2081/6470 [25:42<53:57,  1.36it/s]
3|2082|Loss: 0.9007635116577148:  32%|███▏      | 2081/6470 [25:43<53:57,  1.36it/s]
3|2082|Loss: 0.9007635116577148:  32%|███▏      | 2082/6470 [25:43<57:12,  1.28it/s]
3|2083|Loss: 0.849316418170929:  32%|███▏      | 2082/6470 [25:44<57:12,  1.28it/s] 
3|2083|Loss: 0.849316418170929:  32%|███▏      | 2083/6470 [25:44<58:21,  1.25it/s]
3|2084|Loss: 0.9670456051826477:  32%|███▏      | 2083/6470 [25:45<58:21,  1.25it/s]
3|2084|Loss: 0.9670456051826477:  32%|███▏      | 2084/6470 [25:45<59:27,  1.23it/s]
3|2085|Loss: 0.8950508832931519:  32%|███▏      | 2084/6470 [25:45<59:27,  1.23it/s]
3|2085|Loss: 0.8950508832931519:  32%|███▏      | 2085/6470 [25:46<55:02,  1.33it/s]
3|2086|Loss: 0.9210852384567261:  32%|███▏      | 2085/6470 [25:46<55:02,  1.33it/s]
3|2086|Loss: 0.9210852384567261:  32%|███▏      | 2086/6470 [25:46<54:05,  1.35it/s]
3|2087|Loss: 0.8745378851890564:  32%|███▏      | 2086/6470 [25:47<54:05,  1.35it/s]
3|2087|Loss: 0.8745378851890564:  32%|███▏      | 2087/6470 [25:47<52:01,  1.40it/s]
3|2088|Loss: 0.727332353591919:  32%|███▏      | 2087/6470 [25:47<52:01,  1.40it/s] 
3|2088|Loss: 0.727332353591919:  32%|███▏      | 2088/6470 [25:48<55:19,  1.32it/s]
3|2089|Loss: 0.9350802302360535:  32%|███▏      | 2088/6470 [25:48<55:19,  1.32it/s]
3|2089|Loss: 0.9350802302360535:  32%|███▏      | 2089/6470 [25:49<55:57,  1.31it/s]
3|2090|Loss: 0.7623165249824524:  32%|███▏      | 2089/6470 [25:49<55:57,  1.31it/s]
3|2090|Loss: 0.7623165249824524:  32%|███▏      | 2090/6470 [25:49<56:29,  1.29it/s]
3|2091|Loss: 0.7811734676361084:  32%|███▏      | 2090/6470 [25:50<56:29,  1.29it/s]
3|2091|Loss: 0.7811734676361084:  32%|███▏      | 2091/6470 [25:50<1:01:14,  1.19it/s]
3|2092|Loss: 0.6207832098007202:  32%|███▏      | 2091/6470 [25:51<1:01:14,  1.19it/s]
3|2092|Loss: 0.6207832098007202:  32%|███▏      | 2092/6470 [25:51<59:49,  1.22it/s]  
3|2093|Loss: 0.7819525003433228:  32%|███▏      | 2092/6470 [25:52<59:49,  1.22it/s]
3|2093|Loss: 0.7819525003433228:  32%|███▏      | 2093/6470 [25:52<55:45,  1.31it/s]
3|2094|Loss: 0.8979365229606628:  32%|███▏      | 2093/6470 [25:52<55:45,  1.31it/s]
3|2094|Loss: 0.8979365229606628:  32%|███▏      | 2094/6470 [25:52<52:32,  1.39it/s]
3|2095|Loss: 0.8592483997344971:  32%|███▏      | 2094/6470 [25:53<52:32,  1.39it/s]
3|2095|Loss: 0.8592483997344971:  32%|███▏      | 2095/6470 [25:53<53:28,  1.36it/s]
3|2096|Loss: 0.9379569292068481:  32%|███▏      | 2095/6470 [25:54<53:28,  1.36it/s]
3|2096|Loss: 0.9379569292068481:  32%|███▏      | 2096/6470 [25:54<54:45,  1.33it/s]
3|2097|Loss: 0.757461667060852:  32%|███▏      | 2096/6470 [25:54<54:45,  1.33it/s] 
3|2097|Loss: 0.757461667060852:  32%|███▏      | 2097/6470 [25:55<53:21,  1.37it/s]
3|2098|Loss: 0.972741425037384:  32%|███▏      | 2097/6470 [25:55<53:21,  1.37it/s]
3|2098|Loss: 0.972741425037384:  32%|███▏      | 2098/6470 [25:55<51:46,  1.41it/s]
3|2099|Loss: 0.8040388226509094:  32%|███▏      | 2098/6470 [25:56<51:46,  1.41it/s]
3|2099|Loss: 0.8040388226509094:  32%|███▏      | 2099/6470 [25:56<50:53,  1.43it/s]
3|2100|Loss: 0.8275400400161743:  32%|███▏      | 2099/6470 [25:56<50:53,  1.43it/s]
3|2100|Loss: 0.8275400400161743:  32%|███▏      | 2100/6470 [25:57<46:48,  1.56it/s]
3|2101|Loss: 0.9110286831855774:  32%|███▏      | 2100/6470 [25:57<46:48,  1.56it/s]
3|2101|Loss: 0.9110286831855774:  32%|███▏      | 2101/6470 [25:57<49:49,  1.46it/s]
3|2102|Loss: 0.6631180644035339:  32%|███▏      | 2101/6470 [25:58<49:49,  1.46it/s]
3|2102|Loss: 0.6631180644035339:  32%|███▏      | 2102/6470 [25:58<47:00,  1.55it/s]
3|2103|Loss: 0.8273005485534668:  32%|███▏      | 2102/6470 [25:58<47:00,  1.55it/s]
3|2103|Loss: 0.8273005485534668:  33%|███▎      | 2103/6470 [25:58<45:50,  1.59it/s]
3|2104|Loss: 0.8284863829612732:  33%|███▎      | 2103/6470 [25:59<45:50,  1.59it/s]
3|2104|Loss: 0.8284863829612732:  33%|███▎      | 2104/6470 [25:59<47:01,  1.55it/s]
3|2105|Loss: 0.8035041093826294:  33%|███▎      | 2104/6470 [26:00<47:01,  1.55it/s]
3|2105|Loss: 0.8035041093826294:  33%|███▎      | 2105/6470 [26:00<50:03,  1.45it/s]
3|2106|Loss: 0.8560435771942139:  33%|███▎      | 2105/6470 [26:00<50:03,  1.45it/s]
3|2106|Loss: 0.8560435771942139:  33%|███▎      | 2106/6470 [26:01<52:35,  1.38it/s]
3|2107|Loss: 0.860676109790802:  33%|███▎      | 2106/6470 [26:01<52:35,  1.38it/s] 
3|2107|Loss: 0.860676109790802:  33%|███▎      | 2107/6470 [26:02<53:20,  1.36it/s]
3|2108|Loss: 0.5962044596672058:  33%|███▎      | 2107/6470 [26:02<53:20,  1.36it/s]
3|2108|Loss: 0.5962044596672058:  33%|███▎      | 2108/6470 [26:02<52:03,  1.40it/s]
3|2109|Loss: 0.9236444234848022:  33%|███▎      | 2108/6470 [26:03<52:03,  1.40it/s]
3|2109|Loss: 0.9236444234848022:  33%|███▎      | 2109/6470 [26:03<52:27,  1.39it/s]
3|2110|Loss: 0.9361075162887573:  33%|███▎      | 2109/6470 [26:03<52:27,  1.39it/s]
3|2110|Loss: 0.9361075162887573:  33%|███▎      | 2110/6470 [26:04<52:54,  1.37it/s]
3|2111|Loss: 0.7652831673622131:  33%|███▎      | 2110/6470 [26:04<52:54,  1.37it/s]
3|2111|Loss: 0.7652831673622131:  33%|███▎      | 2111/6470 [26:04<55:03,  1.32it/s]
3|2112|Loss: 0.9089305996894836:  33%|███▎      | 2111/6470 [26:05<55:03,  1.32it/s]
3|2112|Loss: 0.9089305996894836:  33%|███▎      | 2112/6470 [26:05<56:16,  1.29it/s]
3|2113|Loss: 0.8879026770591736:  33%|███▎      | 2112/6470 [26:06<56:16,  1.29it/s]
3|2113|Loss: 0.8879026770591736:  33%|███▎      | 2113/6470 [26:06<55:05,  1.32it/s]
3|2114|Loss: 0.8300676345825195:  33%|███▎      | 2113/6470 [26:06<55:05,  1.32it/s]
3|2114|Loss: 0.8300676345825195:  33%|███▎      | 2114/6470 [26:07<55:26,  1.31it/s]
3|2115|Loss: 0.862614095211029:  33%|███▎      | 2114/6470 [26:07<55:26,  1.31it/s] 
3|2115|Loss: 0.862614095211029:  33%|███▎      | 2115/6470 [26:07<53:24,  1.36it/s]
3|2116|Loss: 0.7522487044334412:  33%|███▎      | 2115/6470 [26:08<53:24,  1.36it/s]
3|2116|Loss: 0.7522487044334412:  33%|███▎      | 2116/6470 [26:08<53:59,  1.34it/s]
3|2117|Loss: 0.9350049495697021:  33%|███▎      | 2116/6470 [26:09<53:59,  1.34it/s]
3|2117|Loss: 0.9350049495697021:  33%|███▎      | 2117/6470 [26:09<54:27,  1.33it/s]
3|2118|Loss: 0.9126595258712769:  33%|███▎      | 2117/6470 [26:09<54:27,  1.33it/s]
3|2118|Loss: 0.9126595258712769:  33%|███▎      | 2118/6470 [26:10<57:15,  1.27it/s]
3|2119|Loss: 0.7244858145713806:  33%|███▎      | 2118/6470 [26:10<57:15,  1.27it/s]
3|2119|Loss: 0.7244858145713806:  33%|███▎      | 2119/6470 [26:11<55:35,  1.30it/s]
3|2120|Loss: 0.7564804553985596:  33%|███▎      | 2119/6470 [26:11<55:35,  1.30it/s]
3|2120|Loss: 0.7564804553985596:  33%|███▎      | 2120/6470 [26:11<53:17,  1.36it/s]
3|2121|Loss: 0.8047930598258972:  33%|███▎      | 2120/6470 [26:12<53:17,  1.36it/s]
3|2121|Loss: 0.8047930598258972:  33%|███▎      | 2121/6470 [26:12<55:54,  1.30it/s]
3|2122|Loss: 0.7737069129943848:  33%|███▎      | 2121/6470 [26:13<55:54,  1.30it/s]
3|2122|Loss: 0.7737069129943848:  33%|███▎      | 2122/6470 [26:13<56:12,  1.29it/s]
3|2123|Loss: 0.7770608067512512:  33%|███▎      | 2122/6470 [26:13<56:12,  1.29it/s]
3|2123|Loss: 0.7770608067512512:  33%|███▎      | 2123/6470 [26:14<52:26,  1.38it/s]
3|2124|Loss: 0.8396881818771362:  33%|███▎      | 2123/6470 [26:14<52:26,  1.38it/s]
3|2124|Loss: 0.8396881818771362:  33%|███▎      | 2124/6470 [26:14<53:08,  1.36it/s]
3|2125|Loss: 0.8547533750534058:  33%|███▎      | 2124/6470 [26:15<53:08,  1.36it/s]
3|2125|Loss: 0.8547533750534058:  33%|███▎      | 2125/6470 [26:15<53:58,  1.34it/s]
3|2126|Loss: 0.7356778383255005:  33%|███▎      | 2125/6470 [26:15<53:58,  1.34it/s]
3|2126|Loss: 0.7356778383255005:  33%|███▎      | 2126/6470 [26:16<54:30,  1.33it/s]
3|2127|Loss: 0.7197163701057434:  33%|███▎      | 2126/6470 [26:16<54:30,  1.33it/s]
3|2127|Loss: 0.7197163701057434:  33%|███▎      | 2127/6470 [26:17<57:37,  1.26it/s]
3|2128|Loss: 0.8007053136825562:  33%|███▎      | 2127/6470 [26:17<57:37,  1.26it/s]
3|2128|Loss: 0.8007053136825562:  33%|███▎      | 2128/6470 [26:18<58:18,  1.24it/s]
3|2129|Loss: 0.8703187108039856:  33%|███▎      | 2128/6470 [26:18<58:18,  1.24it/s]
3|2129|Loss: 0.8703187108039856:  33%|███▎      | 2129/6470 [26:18<54:56,  1.32it/s]
3|2130|Loss: 0.8808292746543884:  33%|███▎      | 2129/6470 [26:19<54:56,  1.32it/s]
3|2130|Loss: 0.8808292746543884:  33%|███▎      | 2130/6470 [26:19<54:10,  1.34it/s]
3|2131|Loss: 0.7926729321479797:  33%|███▎      | 2130/6470 [26:19<54:10,  1.34it/s]
3|2131|Loss: 0.7926729321479797:  33%|███▎      | 2131/6470 [26:20<52:44,  1.37it/s]
3|2132|Loss: 0.7221052646636963:  33%|███▎      | 2131/6470 [26:20<52:44,  1.37it/s]
3|2132|Loss: 0.7221052646636963:  33%|███▎      | 2132/6470 [26:20<54:55,  1.32it/s]
3|2133|Loss: 0.7619435787200928:  33%|███▎      | 2132/6470 [26:21<54:55,  1.32it/s]
3|2133|Loss: 0.7619435787200928:  33%|███▎      | 2133/6470 [26:21<1:00:30,  1.19it/s]
3|2134|Loss: 0.8283147811889648:  33%|███▎      | 2133/6470 [26:22<1:00:30,  1.19it/s]
3|2134|Loss: 0.8283147811889648:  33%|███▎      | 2134/6470 [26:22<58:55,  1.23it/s]  
3|2135|Loss: 0.7292829155921936:  33%|███▎      | 2134/6470 [26:23<58:55,  1.23it/s]
3|2135|Loss: 0.7292829155921936:  33%|███▎      | 2135/6470 [26:23<57:24,  1.26it/s]
3|2136|Loss: 0.918708860874176:  33%|███▎      | 2135/6470 [26:23<57:24,  1.26it/s] 
3|2136|Loss: 0.918708860874176:  33%|███▎      | 2136/6470 [26:24<58:21,  1.24it/s]
3|2137|Loss: 0.8162677884101868:  33%|███▎      | 2136/6470 [26:24<58:21,  1.24it/s]
3|2137|Loss: 0.8162677884101868:  33%|███▎      | 2137/6470 [26:25<56:30,  1.28it/s]
3|2138|Loss: 0.6160129308700562:  33%|███▎      | 2137/6470 [26:25<56:30,  1.28it/s]
3|2138|Loss: 0.6160129308700562:  33%|███▎      | 2138/6470 [26:25<48:47,  1.48it/s]
3|2139|Loss: 0.7051311731338501:  33%|███▎      | 2138/6470 [26:25<48:47,  1.48it/s]
3|2139|Loss: 0.7051311731338501:  33%|███▎      | 2139/6470 [26:26<52:09,  1.38it/s]
3|2140|Loss: 0.888831615447998:  33%|███▎      | 2139/6470 [26:26<52:09,  1.38it/s] 
3|2140|Loss: 0.888831615447998:  33%|███▎      | 2140/6470 [26:27<52:38,  1.37it/s]
3|2141|Loss: 0.8705241680145264:  33%|███▎      | 2140/6470 [26:27<52:38,  1.37it/s]
3|2141|Loss: 0.8705241680145264:  33%|███▎      | 2141/6470 [26:27<53:07,  1.36it/s]
3|2142|Loss: 0.8308371305465698:  33%|███▎      | 2141/6470 [26:28<53:07,  1.36it/s]
3|2142|Loss: 0.8308371305465698:  33%|███▎      | 2142/6470 [26:28<53:25,  1.35it/s]
3|2143|Loss: 0.7981529831886292:  33%|███▎      | 2142/6470 [26:28<53:25,  1.35it/s]
3|2143|Loss: 0.7981529831886292:  33%|███▎      | 2143/6470 [26:29<53:39,  1.34it/s]
3|2144|Loss: 0.7136628031730652:  33%|███▎      | 2143/6470 [26:29<53:39,  1.34it/s]
3|2144|Loss: 0.7136628031730652:  33%|███▎      | 2144/6470 [26:30<53:47,  1.34it/s]
3|2145|Loss: 0.7812775373458862:  33%|███▎      | 2144/6470 [26:30<53:47,  1.34it/s]
3|2145|Loss: 0.7812775373458862:  33%|███▎      | 2145/6470 [26:30<53:57,  1.34it/s]
3|2146|Loss: 0.9488799571990967:  33%|███▎      | 2145/6470 [26:31<53:57,  1.34it/s]
3|2146|Loss: 0.9488799571990967:  33%|███▎      | 2146/6470 [26:31<54:57,  1.31it/s]
3|2147|Loss: 0.7574670910835266:  33%|███▎      | 2146/6470 [26:32<54:57,  1.31it/s]
3|2147|Loss: 0.7574670910835266:  33%|███▎      | 2147/6470 [26:32<55:02,  1.31it/s]
3|2148|Loss: 0.67552649974823:  33%|███▎      | 2147/6470 [26:32<55:02,  1.31it/s]  
3|2148|Loss: 0.67552649974823:  33%|███▎      | 2148/6470 [26:33<59:02,  1.22it/s]
3|2149|Loss: 0.7615328431129456:  33%|███▎      | 2148/6470 [26:33<59:02,  1.22it/s]
3|2149|Loss: 0.7615328431129456:  33%|███▎      | 2149/6470 [26:33<56:31,  1.27it/s]
3|2150|Loss: 0.9348322153091431:  33%|███▎      | 2149/6470 [26:34<56:31,  1.27it/s]
3|2150|Loss: 0.9348322153091431:  33%|███▎      | 2150/6470 [26:34<50:22,  1.43it/s]
3|2151|Loss: 0.8940892815589905:  33%|███▎      | 2150/6470 [26:34<50:22,  1.43it/s]
3|2151|Loss: 0.8940892815589905:  33%|███▎      | 2151/6470 [26:35<50:18,  1.43it/s]
3|2152|Loss: 0.6915642619132996:  33%|███▎      | 2151/6470 [26:35<50:18,  1.43it/s]
3|2152|Loss: 0.6915642619132996:  33%|███▎      | 2152/6470 [26:36<54:25,  1.32it/s]
3|2153|Loss: 0.6918468475341797:  33%|███▎      | 2152/6470 [26:36<54:25,  1.32it/s]
3|2153|Loss: 0.6918468475341797:  33%|███▎      | 2153/6470 [26:36<55:56,  1.29it/s]
3|2154|Loss: 0.8275197744369507:  33%|███▎      | 2153/6470 [26:37<55:56,  1.29it/s]
3|2154|Loss: 0.8275197744369507:  33%|███▎      | 2154/6470 [26:37<54:27,  1.32it/s]
3|2155|Loss: 0.8062805533409119:  33%|███▎      | 2154/6470 [26:38<54:27,  1.32it/s]
3|2155|Loss: 0.8062805533409119:  33%|███▎      | 2155/6470 [26:38<58:26,  1.23it/s]
3|2156|Loss: 0.7721351385116577:  33%|███▎      | 2155/6470 [26:39<58:26,  1.23it/s]
3|2156|Loss: 0.7721351385116577:  33%|███▎      | 2156/6470 [26:39<57:00,  1.26it/s]
3|2157|Loss: 0.8030192255973816:  33%|███▎      | 2156/6470 [26:39<57:00,  1.26it/s]
3|2157|Loss: 0.8030192255973816:  33%|███▎      | 2157/6470 [26:39<54:25,  1.32it/s]
3|2158|Loss: 0.7612015008926392:  33%|███▎      | 2157/6470 [26:40<54:25,  1.32it/s]
3|2158|Loss: 0.7612015008926392:  33%|███▎      | 2158/6470 [26:40<53:57,  1.33it/s]
3|2159|Loss: 0.8212339282035828:  33%|███▎      | 2158/6470 [26:41<53:57,  1.33it/s]
3|2159|Loss: 0.8212339282035828:  33%|███▎      | 2159/6470 [26:41<54:25,  1.32it/s]
3|2160|Loss: 0.7497446537017822:  33%|███▎      | 2159/6470 [26:41<54:25,  1.32it/s]
3|2160|Loss: 0.7497446537017822:  33%|███▎      | 2160/6470 [26:42<55:11,  1.30it/s]
3|2161|Loss: 0.8786660432815552:  33%|███▎      | 2160/6470 [26:42<55:11,  1.30it/s]
3|2161|Loss: 0.8786660432815552:  33%|███▎      | 2161/6470 [26:43<55:49,  1.29it/s]
3|2162|Loss: 0.7587636709213257:  33%|███▎      | 2161/6470 [26:43<55:49,  1.29it/s]
3|2162|Loss: 0.7587636709213257:  33%|███▎      | 2162/6470 [26:43<49:14,  1.46it/s]
3|2163|Loss: 0.8421838879585266:  33%|███▎      | 2162/6470 [26:43<49:14,  1.46it/s]
3|2163|Loss: 0.8421838879585266:  33%|███▎      | 2163/6470 [26:44<46:25,  1.55it/s]
3|2164|Loss: 0.8128967881202698:  33%|███▎      | 2163/6470 [26:44<46:25,  1.55it/s]
3|2164|Loss: 0.8128967881202698:  33%|███▎      | 2164/6470 [26:44<46:33,  1.54it/s]
3|2165|Loss: 0.859247624874115:  33%|███▎      | 2164/6470 [26:45<46:33,  1.54it/s] 
3|2165|Loss: 0.859247624874115:  33%|███▎      | 2165/6470 [26:45<50:17,  1.43it/s]
3|2166|Loss: 0.802361249923706:  33%|███▎      | 2165/6470 [26:45<50:17,  1.43it/s]
3|2166|Loss: 0.802361249923706:  33%|███▎      | 2166/6470 [26:46<50:16,  1.43it/s]
3|2167|Loss: 0.7005712389945984:  33%|███▎      | 2166/6470 [26:46<50:16,  1.43it/s]
3|2167|Loss: 0.7005712389945984:  33%|███▎      | 2167/6470 [26:46<47:34,  1.51it/s]
3|2168|Loss: 0.878584086894989:  33%|███▎      | 2167/6470 [26:47<47:34,  1.51it/s] 
3|2168|Loss: 0.878584086894989:  34%|███▎      | 2168/6470 [26:47<49:06,  1.46it/s]
3|2169|Loss: 0.7616879940032959:  34%|███▎      | 2168/6470 [26:48<49:06,  1.46it/s]
3|2169|Loss: 0.7616879940032959:  34%|███▎      | 2169/6470 [26:48<53:13,  1.35it/s]
3|2170|Loss: 0.7199311852455139:  34%|███▎      | 2169/6470 [26:48<53:13,  1.35it/s]
3|2170|Loss: 0.7199311852455139:  34%|███▎      | 2170/6470 [26:49<56:44,  1.26it/s]
3|2171|Loss: 0.9085114598274231:  34%|███▎      | 2170/6470 [26:49<56:44,  1.26it/s]
3|2171|Loss: 0.9085114598274231:  34%|███▎      | 2171/6470 [26:50<56:53,  1.26it/s]
3|2172|Loss: 0.855185329914093:  34%|███▎      | 2171/6470 [26:50<56:53,  1.26it/s] 
3|2172|Loss: 0.855185329914093:  34%|███▎      | 2172/6470 [26:51<59:15,  1.21it/s]
3|2173|Loss: 0.8352693319320679:  34%|███▎      | 2172/6470 [26:51<59:15,  1.21it/s]
3|2173|Loss: 0.8352693319320679:  34%|███▎      | 2173/6470 [26:51<59:19,  1.21it/s]
3|2174|Loss: 0.7553961873054504:  34%|███▎      | 2173/6470 [26:52<59:19,  1.21it/s]
3|2174|Loss: 0.7553961873054504:  34%|███▎      | 2174/6470 [26:52<57:11,  1.25it/s]
3|2175|Loss: 0.8785935640335083:  34%|███▎      | 2174/6470 [26:53<57:11,  1.25it/s]
3|2175|Loss: 0.8785935640335083:  34%|███▎      | 2175/6470 [26:53<56:47,  1.26it/s]
3|2176|Loss: 0.8709314465522766:  34%|███▎      | 2175/6470 [26:53<56:47,  1.26it/s]
3|2176|Loss: 0.8709314465522766:  34%|███▎      | 2176/6470 [26:54<55:02,  1.30it/s]
3|2177|Loss: 1.1146199703216553:  34%|███▎      | 2176/6470 [26:54<55:02,  1.30it/s]
3|2177|Loss: 1.1146199703216553:  34%|███▎      | 2177/6470 [26:55<57:19,  1.25it/s]
3|2178|Loss: 0.9326444268226624:  34%|███▎      | 2177/6470 [26:55<57:19,  1.25it/s]
3|2178|Loss: 0.9326444268226624:  34%|███▎      | 2178/6470 [26:55<57:10,  1.25it/s]
3|2179|Loss: 0.7856897711753845:  34%|███▎      | 2178/6470 [26:56<57:10,  1.25it/s]
3|2179|Loss: 0.7856897711753845:  34%|███▎      | 2179/6470 [26:56<55:25,  1.29it/s]
3|2180|Loss: 0.8229345679283142:  34%|███▎      | 2179/6470 [26:56<55:25,  1.29it/s]
3|2180|Loss: 0.8229345679283142:  34%|███▎      | 2180/6470 [26:57<53:23,  1.34it/s]
3|2181|Loss: 0.9418923258781433:  34%|███▎      | 2180/6470 [26:57<53:23,  1.34it/s]
3|2181|Loss: 0.9418923258781433:  34%|███▎      | 2181/6470 [26:57<53:39,  1.33it/s]
3|2182|Loss: 0.8893210887908936:  34%|███▎      | 2181/6470 [26:58<53:39,  1.33it/s]
3|2182|Loss: 0.8893210887908936:  34%|███▎      | 2182/6470 [26:58<52:52,  1.35it/s]
3|2183|Loss: 0.8530566096305847:  34%|███▎      | 2182/6470 [26:59<52:52,  1.35it/s]
3|2183|Loss: 0.8530566096305847:  34%|███▎      | 2183/6470 [26:59<55:56,  1.28it/s]
3|2184|Loss: 0.9200983047485352:  34%|███▎      | 2183/6470 [27:00<55:56,  1.28it/s]
3|2184|Loss: 0.9200983047485352:  34%|███▍      | 2184/6470 [27:00<58:54,  1.21it/s]
3|2185|Loss: 0.8591495156288147:  34%|███▍      | 2184/6470 [27:00<58:54,  1.21it/s]
3|2185|Loss: 0.8591495156288147:  34%|███▍      | 2185/6470 [27:01<58:34,  1.22it/s]
3|2186|Loss: 0.6877788305282593:  34%|███▍      | 2185/6470 [27:01<58:34,  1.22it/s]
3|2186|Loss: 0.6877788305282593:  34%|███▍      | 2186/6470 [27:01<53:44,  1.33it/s]
3|2187|Loss: 0.8466196656227112:  34%|███▍      | 2186/6470 [27:02<53:44,  1.33it/s]
3|2187|Loss: 0.8466196656227112:  34%|███▍      | 2187/6470 [27:02<49:03,  1.46it/s]
3|2188|Loss: 0.9324089884757996:  34%|███▍      | 2187/6470 [27:02<49:03,  1.46it/s]
3|2188|Loss: 0.9324089884757996:  34%|███▍      | 2188/6470 [27:03<52:53,  1.35it/s]
3|2189|Loss: 0.9197998642921448:  34%|███▍      | 2188/6470 [27:03<52:53,  1.35it/s]
3|2189|Loss: 0.9197998642921448:  34%|███▍      | 2189/6470 [27:03<50:33,  1.41it/s]
3|2190|Loss: 0.8482123017311096:  34%|███▍      | 2189/6470 [27:04<50:33,  1.41it/s]
3|2190|Loss: 0.8482123017311096:  34%|███▍      | 2190/6470 [27:04<52:36,  1.36it/s]
3|2191|Loss: 0.914879560470581:  34%|███▍      | 2190/6470 [27:05<52:36,  1.36it/s] 
3|2191|Loss: 0.914879560470581:  34%|███▍      | 2191/6470 [27:05<51:43,  1.38it/s]
3|2192|Loss: 0.8080178499221802:  34%|███▍      | 2191/6470 [27:05<51:43,  1.38it/s]
3|2192|Loss: 0.8080178499221802:  34%|███▍      | 2192/6470 [27:06<52:30,  1.36it/s]
3|2193|Loss: 0.9051845073699951:  34%|███▍      | 2192/6470 [27:06<52:30,  1.36it/s]
3|2193|Loss: 0.9051845073699951:  34%|███▍      | 2193/6470 [27:07<54:49,  1.30it/s]
3|2194|Loss: 0.8374043703079224:  34%|███▍      | 2193/6470 [27:07<54:49,  1.30it/s]
3|2194|Loss: 0.8374043703079224:  34%|███▍      | 2194/6470 [27:07<55:49,  1.28it/s]
3|2195|Loss: 0.7265813946723938:  34%|███▍      | 2194/6470 [27:08<55:49,  1.28it/s]
3|2195|Loss: 0.7265813946723938:  34%|███▍      | 2195/6470 [27:08<49:05,  1.45it/s]
3|2196|Loss: 0.8221059441566467:  34%|███▍      | 2195/6470 [27:08<49:05,  1.45it/s]
3|2196|Loss: 0.8221059441566467:  34%|███▍      | 2196/6470 [27:09<49:39,  1.43it/s]
3|2197|Loss: 0.8330665230751038:  34%|███▍      | 2196/6470 [27:09<49:39,  1.43it/s]
3|2197|Loss: 0.8330665230751038:  34%|███▍      | 2197/6470 [27:09<45:05,  1.58it/s]
3|2198|Loss: 0.7933229207992554:  34%|███▍      | 2197/6470 [27:09<45:05,  1.58it/s]
3|2198|Loss: 0.7933229207992554:  34%|███▍      | 2198/6470 [27:10<45:56,  1.55it/s]
3|2199|Loss: 0.8278986811637878:  34%|███▍      | 2198/6470 [27:10<45:56,  1.55it/s]
3|2199|Loss: 0.8278986811637878:  34%|███▍      | 2199/6470 [27:10<49:12,  1.45it/s]
3|2200|Loss: 0.8742862343788147:  34%|███▍      | 2199/6470 [27:11<49:12,  1.45it/s]
3|2200|Loss: 0.8742862343788147:  34%|███▍      | 2200/6470 [27:11<53:27,  1.33it/s]
3|2201|Loss: 0.995746374130249:  34%|███▍      | 2200/6470 [27:12<53:27,  1.33it/s] 
3|2201|Loss: 0.995746374130249:  34%|███▍      | 2201/6470 [27:12<52:43,  1.35it/s]
3|2202|Loss: 0.810813844203949:  34%|███▍      | 2201/6470 [27:12<52:43,  1.35it/s]
3|2202|Loss: 0.810813844203949:  34%|███▍      | 2202/6470 [27:13<52:36,  1.35it/s]
3|2203|Loss: 0.7503976821899414:  34%|███▍      | 2202/6470 [27:13<52:36,  1.35it/s]
3|2203|Loss: 0.7503976821899414:  34%|███▍      | 2203/6470 [27:14<51:58,  1.37it/s]
3|2204|Loss: 0.7558202743530273:  34%|███▍      | 2203/6470 [27:14<51:58,  1.37it/s]
3|2204|Loss: 0.7558202743530273:  34%|███▍      | 2204/6470 [27:14<54:04,  1.31it/s]
3|2205|Loss: 0.8769753575325012:  34%|███▍      | 2204/6470 [27:15<54:04,  1.31it/s]
3|2205|Loss: 0.8769753575325012:  34%|███▍      | 2205/6470 [27:15<58:02,  1.22it/s]
3|2206|Loss: 0.8437825441360474:  34%|███▍      | 2205/6470 [27:16<58:02,  1.22it/s]
3|2206|Loss: 0.8437825441360474:  34%|███▍      | 2206/6470 [27:16<56:22,  1.26it/s]
3|2207|Loss: 0.7485832571983337:  34%|███▍      | 2206/6470 [27:17<56:22,  1.26it/s]
3|2207|Loss: 0.7485832571983337:  34%|███▍      | 2207/6470 [27:17<58:12,  1.22it/s]
3|2208|Loss: 0.7840533256530762:  34%|███▍      | 2207/6470 [27:17<58:12,  1.22it/s]
3|2208|Loss: 0.7840533256530762:  34%|███▍      | 2208/6470 [27:18<58:17,  1.22it/s]
3|2209|Loss: 0.8968746662139893:  34%|███▍      | 2208/6470 [27:18<58:17,  1.22it/s]
3|2209|Loss: 0.8968746662139893:  34%|███▍      | 2209/6470 [27:19<56:46,  1.25it/s]
3|2210|Loss: 0.87748783826828:  34%|███▍      | 2209/6470 [27:19<56:46,  1.25it/s]  
3|2210|Loss: 0.87748783826828:  34%|███▍      | 2210/6470 [27:19<57:57,  1.23it/s]
3|2211|Loss: 0.772606372833252:  34%|███▍      | 2210/6470 [27:20<57:57,  1.23it/s]
3|2211|Loss: 0.772606372833252:  34%|███▍      | 2211/6470 [27:20<57:24,  1.24it/s]
3|2212|Loss: 0.8933608531951904:  34%|███▍      | 2211/6470 [27:21<57:24,  1.24it/s]
3|2212|Loss: 0.8933608531951904:  34%|███▍      | 2212/6470 [27:21<55:19,  1.28it/s]
3|2213|Loss: 0.7851659059524536:  34%|███▍      | 2212/6470 [27:21<55:19,  1.28it/s]
3|2213|Loss: 0.7851659059524536:  34%|███▍      | 2213/6470 [27:22<55:20,  1.28it/s]
3|2214|Loss: 0.8776237368583679:  34%|███▍      | 2213/6470 [27:22<55:20,  1.28it/s]
3|2214|Loss: 0.8776237368583679:  34%|███▍      | 2214/6470 [27:22<53:27,  1.33it/s]
3|2215|Loss: 0.8684638142585754:  34%|███▍      | 2214/6470 [27:23<53:27,  1.33it/s]
3|2215|Loss: 0.8684638142585754:  34%|███▍      | 2215/6470 [27:23<52:57,  1.34it/s]
3|2216|Loss: 0.7652120590209961:  34%|███▍      | 2215/6470 [27:24<52:57,  1.34it/s]
3|2216|Loss: 0.7652120590209961:  34%|███▍      | 2216/6470 [27:24<54:38,  1.30it/s]
3|2217|Loss: 1.0758264064788818:  34%|███▍      | 2216/6470 [27:24<54:38,  1.30it/s]
3|2217|Loss: 1.0758264064788818:  34%|███▍      | 2217/6470 [27:25<54:04,  1.31it/s]
3|2218|Loss: 0.8582453727722168:  34%|███▍      | 2217/6470 [27:25<54:04,  1.31it/s]
3|2218|Loss: 0.8582453727722168:  34%|███▍      | 2218/6470 [27:25<55:40,  1.27it/s]
3|2219|Loss: 0.8452867269515991:  34%|███▍      | 2218/6470 [27:26<55:40,  1.27it/s]
3|2219|Loss: 0.8452867269515991:  34%|███▍      | 2219/6470 [27:26<53:59,  1.31it/s]
3|2220|Loss: 0.7427076697349548:  34%|███▍      | 2219/6470 [27:27<53:59,  1.31it/s]
3|2220|Loss: 0.7427076697349548:  34%|███▍      | 2220/6470 [27:27<49:10,  1.44it/s]
3|2221|Loss: 0.9336898326873779:  34%|███▍      | 2220/6470 [27:27<49:10,  1.44it/s]
3|2221|Loss: 0.9336898326873779:  34%|███▍      | 2221/6470 [27:28<51:08,  1.38it/s]
3|2222|Loss: 0.9036380052566528:  34%|███▍      | 2221/6470 [27:28<51:08,  1.38it/s]
3|2222|Loss: 0.9036380052566528:  34%|███▍      | 2222/6470 [27:28<52:17,  1.35it/s]
3|2223|Loss: 0.9544082880020142:  34%|███▍      | 2222/6470 [27:29<52:17,  1.35it/s]
3|2223|Loss: 0.9544082880020142:  34%|███▍      | 2223/6470 [27:29<52:48,  1.34it/s]
3|2224|Loss: 0.6451661586761475:  34%|███▍      | 2223/6470 [27:30<52:48,  1.34it/s]
3|2224|Loss: 0.6451661586761475:  34%|███▍      | 2224/6470 [27:30<55:15,  1.28it/s]
3|2225|Loss: 0.8807746767997742:  34%|███▍      | 2224/6470 [27:30<55:15,  1.28it/s]
3|2225|Loss: 0.8807746767997742:  34%|███▍      | 2225/6470 [27:31<55:47,  1.27it/s]
3|2226|Loss: 0.8551456332206726:  34%|███▍      | 2225/6470 [27:31<55:47,  1.27it/s]
3|2226|Loss: 0.8551456332206726:  34%|███▍      | 2226/6470 [27:32<56:04,  1.26it/s]
3|2227|Loss: 0.7812124490737915:  34%|███▍      | 2226/6470 [27:32<56:04,  1.26it/s]
3|2227|Loss: 0.7812124490737915:  34%|███▍      | 2227/6470 [27:32<50:53,  1.39it/s]
3|2228|Loss: 0.9382344484329224:  34%|███▍      | 2227/6470 [27:32<50:53,  1.39it/s]
3|2228|Loss: 0.9382344484329224:  34%|███▍      | 2228/6470 [27:33<47:54,  1.48it/s]
3|2229|Loss: 0.9180532097816467:  34%|███▍      | 2228/6470 [27:33<47:54,  1.48it/s]
3|2229|Loss: 0.9180532097816467:  34%|███▍      | 2229/6470 [27:33<46:41,  1.51it/s]
3|2230|Loss: 0.8307856321334839:  34%|███▍      | 2229/6470 [27:34<46:41,  1.51it/s]
3|2230|Loss: 0.8307856321334839:  34%|███▍      | 2230/6470 [27:34<50:02,  1.41it/s]
3|2231|Loss: 0.8716034889221191:  34%|███▍      | 2230/6470 [27:34<50:02,  1.41it/s]
3|2231|Loss: 0.8716034889221191:  34%|███▍      | 2231/6470 [27:35<48:50,  1.45it/s]
3|2232|Loss: 0.7458139657974243:  34%|███▍      | 2231/6470 [27:35<48:50,  1.45it/s]
3|2232|Loss: 0.7458139657974243:  34%|███▍      | 2232/6470 [27:36<50:27,  1.40it/s]
3|2233|Loss: 0.8931981325149536:  34%|███▍      | 2232/6470 [27:36<50:27,  1.40it/s]
3|2233|Loss: 0.8931981325149536:  35%|███▍      | 2233/6470 [27:36<51:58,  1.36it/s]
3|2234|Loss: 0.8709807991981506:  35%|███▍      | 2233/6470 [27:37<51:58,  1.36it/s]
3|2234|Loss: 0.8709807991981506:  35%|███▍      | 2234/6470 [27:37<56:18,  1.25it/s]
3|2235|Loss: 0.9402663707733154:  35%|███▍      | 2234/6470 [27:38<56:18,  1.25it/s]
3|2235|Loss: 0.9402663707733154:  35%|███▍      | 2235/6470 [27:38<55:13,  1.28it/s]
3|2236|Loss: 0.9193999171257019:  35%|███▍      | 2235/6470 [27:38<55:13,  1.28it/s]
3|2236|Loss: 0.9193999171257019:  35%|███▍      | 2236/6470 [27:39<57:28,  1.23it/s]
3|2237|Loss: 0.8231595754623413:  35%|███▍      | 2236/6470 [27:39<57:28,  1.23it/s]
3|2237|Loss: 0.8231595754623413:  35%|███▍      | 2237/6470 [27:40<58:50,  1.20it/s]
3|2238|Loss: 0.917163610458374:  35%|███▍      | 2237/6470 [27:40<58:50,  1.20it/s] 
3|2238|Loss: 0.917163610458374:  35%|███▍      | 2238/6470 [27:40<56:38,  1.25it/s]
3|2239|Loss: 0.7348634600639343:  35%|███▍      | 2238/6470 [27:41<56:38,  1.25it/s]
3|2239|Loss: 0.7348634600639343:  35%|███▍      | 2239/6470 [27:41<53:49,  1.31it/s]
3|2240|Loss: 0.9159537553787231:  35%|███▍      | 2239/6470 [27:42<53:49,  1.31it/s]
3|2240|Loss: 0.9159537553787231:  35%|███▍      | 2240/6470 [27:42<56:17,  1.25it/s]
3|2241|Loss: 0.9962500333786011:  35%|███▍      | 2240/6470 [27:43<56:17,  1.25it/s]
3|2241|Loss: 0.9962500333786011:  35%|███▍      | 2241/6470 [27:43<58:22,  1.21it/s]
3|2242|Loss: 0.9698746204376221:  35%|███▍      | 2241/6470 [27:43<58:22,  1.21it/s]
3|2242|Loss: 0.9698746204376221:  35%|███▍      | 2242/6470 [27:44<52:48,  1.33it/s]
3|2243|Loss: 0.7607284188270569:  35%|███▍      | 2242/6470 [27:44<52:48,  1.33it/s]
3|2243|Loss: 0.7607284188270569:  35%|███▍      | 2243/6470 [27:44<49:29,  1.42it/s]
3|2244|Loss: 0.8635399341583252:  35%|███▍      | 2243/6470 [27:44<49:29,  1.42it/s]
3|2244|Loss: 0.8635399341583252:  35%|███▍      | 2244/6470 [27:45<48:08,  1.46it/s]
3|2245|Loss: 0.9769957661628723:  35%|███▍      | 2244/6470 [27:45<48:08,  1.46it/s]
3|2245|Loss: 0.9769957661628723:  35%|███▍      | 2245/6470 [27:46<49:50,  1.41it/s]
3|2246|Loss: 0.7830132246017456:  35%|███▍      | 2245/6470 [27:46<49:50,  1.41it/s]
3|2246|Loss: 0.7830132246017456:  35%|███▍      | 2246/6470 [27:46<50:03,  1.41it/s]
3|2247|Loss: 0.8389172554016113:  35%|███▍      | 2246/6470 [27:47<50:03,  1.41it/s]
3|2247|Loss: 0.8389172554016113:  35%|███▍      | 2247/6470 [27:47<53:39,  1.31it/s]
3|2248|Loss: 0.7776882648468018:  35%|███▍      | 2247/6470 [27:48<53:39,  1.31it/s]
3|2248|Loss: 0.7776882648468018:  35%|███▍      | 2248/6470 [27:48<57:41,  1.22it/s]
3|2249|Loss: 0.9010214805603027:  35%|███▍      | 2248/6470 [27:49<57:41,  1.22it/s]
3|2249|Loss: 0.9010214805603027:  35%|███▍      | 2249/6470 [27:49<57:19,  1.23it/s]
3|2250|Loss: 0.7713183164596558:  35%|███▍      | 2249/6470 [27:49<57:19,  1.23it/s]
3|2250|Loss: 0.7713183164596558:  35%|███▍      | 2250/6470 [27:50<56:40,  1.24it/s]
3|2251|Loss: 0.7823812365531921:  35%|███▍      | 2250/6470 [27:50<56:40,  1.24it/s]
3|2251|Loss: 0.7823812365531921:  35%|███▍      | 2251/6470 [27:50<54:15,  1.30it/s]
3|2252|Loss: 0.8001854419708252:  35%|███▍      | 2251/6470 [27:51<54:15,  1.30it/s]
3|2252|Loss: 0.8001854419708252:  35%|███▍      | 2252/6470 [27:51<50:15,  1.40it/s]
3|2253|Loss: 0.7494168877601624:  35%|███▍      | 2252/6470 [27:51<50:15,  1.40it/s]
3|2253|Loss: 0.7494168877601624:  35%|███▍      | 2253/6470 [27:52<52:01,  1.35it/s]
3|2254|Loss: 0.6632936596870422:  35%|███▍      | 2253/6470 [27:52<52:01,  1.35it/s]
3|2254|Loss: 0.6632936596870422:  35%|███▍      | 2254/6470 [27:53<1:07:09,  1.05it/s]
3|2255|Loss: 0.7321236729621887:  35%|███▍      | 2254/6470 [27:54<1:07:09,  1.05it/s]
3|2255|Loss: 0.7321236729621887:  35%|███▍      | 2255/6470 [27:54<1:03:07,  1.11it/s]
3|2256|Loss: 0.8786753416061401:  35%|███▍      | 2255/6470 [27:54<1:03:07,  1.11it/s]
3|2256|Loss: 0.8786753416061401:  35%|███▍      | 2256/6470 [27:55<1:01:29,  1.14it/s]
3|2257|Loss: 0.8691895008087158:  35%|███▍      | 2256/6470 [27:55<1:01:29,  1.14it/s]
3|2257|Loss: 0.8691895008087158:  35%|███▍      | 2257/6470 [27:55<57:49,  1.21it/s]  
3|2258|Loss: 0.8573518991470337:  35%|███▍      | 2257/6470 [27:56<57:49,  1.21it/s]
3|2258|Loss: 0.8573518991470337:  35%|███▍      | 2258/6470 [27:56<53:49,  1.30it/s]
3|2259|Loss: 0.8881728649139404:  35%|███▍      | 2258/6470 [27:56<53:49,  1.30it/s]
3|2259|Loss: 0.8881728649139404:  35%|███▍      | 2259/6470 [27:57<51:36,  1.36it/s]
3|2260|Loss: 0.8155078887939453:  35%|███▍      | 2259/6470 [27:57<51:36,  1.36it/s]
3|2260|Loss: 0.8155078887939453:  35%|███▍      | 2260/6470 [27:58<55:06,  1.27it/s]
3|2261|Loss: 0.7239717245101929:  35%|███▍      | 2260/6470 [27:58<55:06,  1.27it/s]
3|2261|Loss: 0.7239717245101929:  35%|███▍      | 2261/6470 [27:59<56:13,  1.25it/s]
3|2262|Loss: 0.7817420363426208:  35%|███▍      | 2261/6470 [27:59<56:13,  1.25it/s]
3|2262|Loss: 0.7817420363426208:  35%|███▍      | 2262/6470 [27:59<54:34,  1.29it/s]
3|2263|Loss: 0.8426647782325745:  35%|███▍      | 2262/6470 [28:00<54:34,  1.29it/s]
3|2263|Loss: 0.8426647782325745:  35%|███▍      | 2263/6470 [28:00<53:52,  1.30it/s]
3|2264|Loss: 0.8020560145378113:  35%|███▍      | 2263/6470 [28:00<53:52,  1.30it/s]
3|2264|Loss: 0.8020560145378113:  35%|███▍      | 2264/6470 [28:01<56:04,  1.25it/s]
3|2265|Loss: 0.871791422367096:  35%|███▍      | 2264/6470 [28:01<56:04,  1.25it/s] 
3|2265|Loss: 0.871791422367096:  35%|███▌      | 2265/6470 [28:02<58:07,  1.21it/s]
3|2266|Loss: 0.7893632054328918:  35%|███▌      | 2265/6470 [28:02<58:07,  1.21it/s]
3|2266|Loss: 0.7893632054328918:  35%|███▌      | 2266/6470 [28:02<56:11,  1.25it/s]
3|2267|Loss: 1.012945294380188:  35%|███▌      | 2266/6470 [28:03<56:11,  1.25it/s] 
3|2267|Loss: 1.012945294380188:  35%|███▌      | 2267/6470 [28:03<55:00,  1.27it/s]
3|2268|Loss: 0.9307869672775269:  35%|███▌      | 2267/6470 [28:04<55:00,  1.27it/s]
3|2268|Loss: 0.9307869672775269:  35%|███▌      | 2268/6470 [28:04<59:12,  1.18it/s]
3|2269|Loss: 1.0291117429733276:  35%|███▌      | 2268/6470 [28:05<59:12,  1.18it/s]
3|2269|Loss: 1.0291117429733276:  35%|███▌      | 2269/6470 [28:05<1:01:33,  1.14it/s]
3|2270|Loss: 1.0164371728897095:  35%|███▌      | 2269/6470 [28:06<1:01:33,  1.14it/s]
3|2270|Loss: 1.0164371728897095:  35%|███▌      | 2270/6470 [28:06<57:46,  1.21it/s]  
3|2271|Loss: 0.8724324703216553:  35%|███▌      | 2270/6470 [28:06<57:46,  1.21it/s]
3|2271|Loss: 0.8724324703216553:  35%|███▌      | 2271/6470 [28:06<52:53,  1.32it/s]
3|2272|Loss: 0.9027122855186462:  35%|███▌      | 2271/6470 [28:07<52:53,  1.32it/s]
3|2272|Loss: 0.9027122855186462:  35%|███▌      | 2272/6470 [28:07<52:39,  1.33it/s]
3|2273|Loss: 0.7514200210571289:  35%|███▌      | 2272/6470 [28:08<52:39,  1.33it/s]
3|2273|Loss: 0.7514200210571289:  35%|███▌      | 2273/6470 [28:08<52:48,  1.32it/s]
3|2274|Loss: 0.8532551527023315:  35%|███▌      | 2273/6470 [28:08<52:48,  1.32it/s]
3|2274|Loss: 0.8532551527023315:  35%|███▌      | 2274/6470 [28:09<51:17,  1.36it/s]
3|2275|Loss: 0.7808645367622375:  35%|███▌      | 2274/6470 [28:09<51:17,  1.36it/s]
3|2275|Loss: 0.7808645367622375:  35%|███▌      | 2275/6470 [28:09<49:56,  1.40it/s]
3|2276|Loss: 0.8314327597618103:  35%|███▌      | 2275/6470 [28:10<49:56,  1.40it/s]
3|2276|Loss: 0.8314327597618103:  35%|███▌      | 2276/6470 [28:10<54:23,  1.29it/s]
3|2277|Loss: 0.8642611503601074:  35%|███▌      | 2276/6470 [28:11<54:23,  1.29it/s]
3|2277|Loss: 0.8642611503601074:  35%|███▌      | 2277/6470 [28:11<57:34,  1.21it/s]
3|2278|Loss: 0.9793772101402283:  35%|███▌      | 2277/6470 [28:12<57:34,  1.21it/s]
3|2278|Loss: 0.9793772101402283:  35%|███▌      | 2278/6470 [28:12<56:36,  1.23it/s]
3|2279|Loss: 0.7263702154159546:  35%|███▌      | 2278/6470 [28:12<56:36,  1.23it/s]
3|2279|Loss: 0.7263702154159546:  35%|███▌      | 2279/6470 [28:13<54:07,  1.29it/s]
3|2280|Loss: 0.9256102442741394:  35%|███▌      | 2279/6470 [28:13<54:07,  1.29it/s]
3|2280|Loss: 0.9256102442741394:  35%|███▌      | 2280/6470 [28:13<53:15,  1.31it/s]
3|2281|Loss: 0.8049449920654297:  35%|███▌      | 2280/6470 [28:14<53:15,  1.31it/s]
3|2281|Loss: 0.8049449920654297:  35%|███▌      | 2281/6470 [28:14<57:02,  1.22it/s]
3|2282|Loss: 0.8754649758338928:  35%|███▌      | 2281/6470 [28:15<57:02,  1.22it/s]
3|2282|Loss: 0.8754649758338928:  35%|███▌      | 2282/6470 [28:15<55:01,  1.27it/s]
3|2283|Loss: 0.7079339027404785:  35%|███▌      | 2282/6470 [28:15<55:01,  1.27it/s]
3|2283|Loss: 0.7079339027404785:  35%|███▌      | 2283/6470 [28:16<54:22,  1.28it/s]
3|2284|Loss: 0.964088499546051:  35%|███▌      | 2283/6470 [28:16<54:22,  1.28it/s] 
3|2284|Loss: 0.964088499546051:  35%|███▌      | 2284/6470 [28:17<56:34,  1.23it/s]
3|2285|Loss: 0.7843441367149353:  35%|███▌      | 2284/6470 [28:17<56:34,  1.23it/s]
3|2285|Loss: 0.7843441367149353:  35%|███▌      | 2285/6470 [28:17<55:22,  1.26it/s]
3|2286|Loss: 0.7649608850479126:  35%|███▌      | 2285/6470 [28:18<55:22,  1.26it/s]
3|2286|Loss: 0.7649608850479126:  35%|███▌      | 2286/6470 [28:18<51:03,  1.37it/s]
3|2287|Loss: 0.8256803750991821:  35%|███▌      | 2286/6470 [28:18<51:03,  1.37it/s]
3|2287|Loss: 0.8256803750991821:  35%|███▌      | 2287/6470 [28:19<48:54,  1.43it/s]
3|2288|Loss: 0.9272458553314209:  35%|███▌      | 2287/6470 [28:19<48:54,  1.43it/s]
3|2288|Loss: 0.9272458553314209:  35%|███▌      | 2288/6470 [28:19<50:15,  1.39it/s]
3|2289|Loss: 0.7717078924179077:  35%|███▌      | 2288/6470 [28:20<50:15,  1.39it/s]
3|2289|Loss: 0.7717078924179077:  35%|███▌      | 2289/6470 [28:20<51:54,  1.34it/s]
3|2290|Loss: 0.808262288570404:  35%|███▌      | 2289/6470 [28:21<51:54,  1.34it/s] 
3|2290|Loss: 0.808262288570404:  35%|███▌      | 2290/6470 [28:21<51:29,  1.35it/s]
3|2291|Loss: 0.7578477263450623:  35%|███▌      | 2290/6470 [28:21<51:29,  1.35it/s]
3|2291|Loss: 0.7578477263450623:  35%|███▌      | 2291/6470 [28:22<48:17,  1.44it/s]
3|2292|Loss: 0.9161171317100525:  35%|███▌      | 2291/6470 [28:22<48:17,  1.44it/s]
3|2292|Loss: 0.9161171317100525:  35%|███▌      | 2292/6470 [28:22<46:18,  1.50it/s]
3|2293|Loss: 0.9659633636474609:  35%|███▌      | 2292/6470 [28:23<46:18,  1.50it/s]
3|2293|Loss: 0.9659633636474609:  35%|███▌      | 2293/6470 [28:23<45:23,  1.53it/s]
3|2294|Loss: 0.9459569454193115:  35%|███▌      | 2293/6470 [28:23<45:23,  1.53it/s]
3|2294|Loss: 0.9459569454193115:  35%|███▌      | 2294/6470 [28:23<44:53,  1.55it/s]
3|2295|Loss: 1.0007894039154053:  35%|███▌      | 2294/6470 [28:24<44:53,  1.55it/s]
3|2295|Loss: 1.0007894039154053:  35%|███▌      | 2295/6470 [28:24<40:32,  1.72it/s]
3|2296|Loss: 0.9479376673698425:  35%|███▌      | 2295/6470 [28:24<40:32,  1.72it/s]
3|2296|Loss: 0.9479376673698425:  35%|███▌      | 2296/6470 [28:25<43:36,  1.60it/s]
3|2297|Loss: 0.710316002368927:  35%|███▌      | 2296/6470 [28:25<43:36,  1.60it/s] 
3|2297|Loss: 0.710316002368927:  36%|███▌      | 2297/6470 [28:25<45:21,  1.53it/s]
3|2298|Loss: 0.7805945873260498:  36%|███▌      | 2297/6470 [28:26<45:21,  1.53it/s]
3|2298|Loss: 0.7805945873260498:  36%|███▌      | 2298/6470 [28:26<47:13,  1.47it/s]
3|2299|Loss: 0.743703305721283:  36%|███▌      | 2298/6470 [28:26<47:13,  1.47it/s] 
3|2299|Loss: 0.743703305721283:  36%|███▌      | 2299/6470 [28:27<46:42,  1.49it/s]
3|2300|Loss: 0.7639896869659424:  36%|███▌      | 2299/6470 [28:27<46:42,  1.49it/s]
3|2300|Loss: 0.7639896869659424:  36%|███▌      | 2300/6470 [28:28<50:50,  1.37it/s]
3|2301|Loss: 0.742596447467804:  36%|███▌      | 2300/6470 [28:28<50:50,  1.37it/s] 
3|2301|Loss: 0.742596447467804:  36%|███▌      | 2301/6470 [28:28<49:20,  1.41it/s]
3|2302|Loss: 0.721764326095581:  36%|███▌      | 2301/6470 [28:29<49:20,  1.41it/s]
3|2302|Loss: 0.721764326095581:  36%|███▌      | 2302/6470 [28:29<49:26,  1.40it/s]
3|2303|Loss: 0.7301559448242188:  36%|███▌      | 2302/6470 [28:29<49:26,  1.40it/s]
3|2303|Loss: 0.7301559448242188:  36%|███▌      | 2303/6470 [28:30<50:04,  1.39it/s]
3|2304|Loss: 0.8230143785476685:  36%|███▌      | 2303/6470 [28:30<50:04,  1.39it/s]
3|2304|Loss: 0.8230143785476685:  36%|███▌      | 2304/6470 [28:30<50:47,  1.37it/s]
3|2305|Loss: 0.8555907011032104:  36%|███▌      | 2304/6470 [28:31<50:47,  1.37it/s]
3|2305|Loss: 0.8555907011032104:  36%|███▌      | 2305/6470 [28:31<49:49,  1.39it/s]
3|2306|Loss: 0.9642447829246521:  36%|███▌      | 2305/6470 [28:31<49:49,  1.39it/s]
3|2306|Loss: 0.9642447829246521:  36%|███▌      | 2306/6470 [28:32<46:41,  1.49it/s]
3|2307|Loss: 0.8313328623771667:  36%|███▌      | 2306/6470 [28:32<46:41,  1.49it/s]
3|2307|Loss: 0.8313328623771667:  36%|███▌      | 2307/6470 [28:32<46:40,  1.49it/s]
3|2308|Loss: 0.8280549645423889:  36%|███▌      | 2307/6470 [28:33<46:40,  1.49it/s]
3|2308|Loss: 0.8280549645423889:  36%|███▌      | 2308/6470 [28:33<48:51,  1.42it/s]
3|2309|Loss: 0.7369673252105713:  36%|███▌      | 2308/6470 [28:34<48:51,  1.42it/s]
3|2309|Loss: 0.7369673252105713:  36%|███▌      | 2309/6470 [28:34<53:54,  1.29it/s]
3|2310|Loss: 0.7570831179618835:  36%|███▌      | 2309/6470 [28:35<53:54,  1.29it/s]
3|2310|Loss: 0.7570831179618835:  36%|███▌      | 2310/6470 [28:35<52:53,  1.31it/s]
3|2311|Loss: 0.7142530679702759:  36%|███▌      | 2310/6470 [28:35<52:53,  1.31it/s]
3|2311|Loss: 0.7142530679702759:  36%|███▌      | 2311/6470 [28:36<52:16,  1.33it/s]
3|2312|Loss: 0.9141724109649658:  36%|███▌      | 2311/6470 [28:36<52:16,  1.33it/s]
3|2312|Loss: 0.9141724109649658:  36%|███▌      | 2312/6470 [28:36<50:49,  1.36it/s]
3|2313|Loss: 0.7261857390403748:  36%|███▌      | 2312/6470 [28:37<50:49,  1.36it/s]
3|2313|Loss: 0.7261857390403748:  36%|███▌      | 2313/6470 [28:37<46:17,  1.50it/s]
3|2314|Loss: 0.8795505166053772:  36%|███▌      | 2313/6470 [28:37<46:17,  1.50it/s]
3|2314|Loss: 0.8795505166053772:  36%|███▌      | 2314/6470 [28:38<48:52,  1.42it/s]
3|2315|Loss: 0.750862717628479:  36%|███▌      | 2314/6470 [28:38<48:52,  1.42it/s] 
3|2315|Loss: 0.750862717628479:  36%|███▌      | 2315/6470 [28:38<47:30,  1.46it/s]
3|2316|Loss: 0.9546594619750977:  36%|███▌      | 2315/6470 [28:39<47:30,  1.46it/s]
3|2316|Loss: 0.9546594619750977:  36%|███▌      | 2316/6470 [28:39<49:53,  1.39it/s]
3|2317|Loss: 0.6994354724884033:  36%|███▌      | 2316/6470 [28:39<49:53,  1.39it/s]
3|2317|Loss: 0.6994354724884033:  36%|███▌      | 2317/6470 [28:40<51:03,  1.36it/s]
3|2318|Loss: 0.7531145215034485:  36%|███▌      | 2317/6470 [28:40<51:03,  1.36it/s]
3|2318|Loss: 0.7531145215034485:  36%|███▌      | 2318/6470 [28:41<52:58,  1.31it/s]
3|2319|Loss: 0.9730325937271118:  36%|███▌      | 2318/6470 [28:41<52:58,  1.31it/s]
3|2319|Loss: 0.9730325937271118:  36%|███▌      | 2319/6470 [28:41<53:51,  1.28it/s]
3|2320|Loss: 0.8121260404586792:  36%|███▌      | 2319/6470 [28:42<53:51,  1.28it/s]
3|2320|Loss: 0.8121260404586792:  36%|███▌      | 2320/6470 [28:42<56:36,  1.22it/s]
3|2321|Loss: 0.7590587735176086:  36%|███▌      | 2320/6470 [28:43<56:36,  1.22it/s]
3|2321|Loss: 0.7590587735176086:  36%|███▌      | 2321/6470 [28:43<54:36,  1.27it/s]
3|2322|Loss: 0.8260878324508667:  36%|███▌      | 2321/6470 [28:43<54:36,  1.27it/s]
3|2322|Loss: 0.8260878324508667:  36%|███▌      | 2322/6470 [28:44<52:28,  1.32it/s]
3|2323|Loss: 0.7790461182594299:  36%|███▌      | 2322/6470 [28:44<52:28,  1.32it/s]
3|2323|Loss: 0.7790461182594299:  36%|███▌      | 2323/6470 [28:44<50:08,  1.38it/s]
3|2324|Loss: 0.7637568116188049:  36%|███▌      | 2323/6470 [28:45<50:08,  1.38it/s]
3|2324|Loss: 0.7637568116188049:  36%|███▌      | 2324/6470 [28:45<49:13,  1.40it/s]
3|2325|Loss: 0.9029099941253662:  36%|███▌      | 2324/6470 [28:45<49:13,  1.40it/s]
3|2325|Loss: 0.9029099941253662:  36%|███▌      | 2325/6470 [28:46<47:57,  1.44it/s]
3|2326|Loss: 0.8550180196762085:  36%|███▌      | 2325/6470 [28:46<47:57,  1.44it/s]
3|2326|Loss: 0.8550180196762085:  36%|███▌      | 2326/6470 [28:46<49:11,  1.40it/s]
3|2327|Loss: 0.7415249347686768:  36%|███▌      | 2326/6470 [28:47<49:11,  1.40it/s]
3|2327|Loss: 0.7415249347686768:  36%|███▌      | 2327/6470 [28:47<47:01,  1.47it/s]
3|2328|Loss: 0.8023843765258789:  36%|███▌      | 2327/6470 [28:47<47:01,  1.47it/s]
3|2328|Loss: 0.8023843765258789:  36%|███▌      | 2328/6470 [28:48<44:09,  1.56it/s]
3|2329|Loss: 0.9052619934082031:  36%|███▌      | 2328/6470 [28:48<44:09,  1.56it/s]
3|2329|Loss: 0.9052619934082031:  36%|███▌      | 2329/6470 [28:48<46:13,  1.49it/s]
3|2330|Loss: 0.8391309380531311:  36%|███▌      | 2329/6470 [28:49<46:13,  1.49it/s]
3|2330|Loss: 0.8391309380531311:  36%|███▌      | 2330/6470 [28:49<47:11,  1.46it/s]
3|2331|Loss: 0.8114356994628906:  36%|███▌      | 2330/6470 [28:49<47:11,  1.46it/s]
3|2331|Loss: 0.8114356994628906:  36%|███▌      | 2331/6470 [28:50<47:37,  1.45it/s]
3|2332|Loss: 0.7483540773391724:  36%|███▌      | 2331/6470 [28:50<47:37,  1.45it/s]
3|2332|Loss: 0.7483540773391724:  36%|███▌      | 2332/6470 [28:51<51:18,  1.34it/s]
3|2333|Loss: 0.8926766514778137:  36%|███▌      | 2332/6470 [28:51<51:18,  1.34it/s]
3|2333|Loss: 0.8926766514778137:  36%|███▌      | 2333/6470 [28:51<51:08,  1.35it/s]
3|2334|Loss: 0.6650999188423157:  36%|███▌      | 2333/6470 [28:52<51:08,  1.35it/s]
3|2334|Loss: 0.6650999188423157:  36%|███▌      | 2334/6470 [28:52<48:49,  1.41it/s]
3|2335|Loss: 0.794750988483429:  36%|███▌      | 2334/6470 [28:52<48:49,  1.41it/s] 
3|2335|Loss: 0.794750988483429:  36%|███▌      | 2335/6470 [28:53<52:04,  1.32it/s]
3|2336|Loss: 0.7957531213760376:  36%|███▌      | 2335/6470 [28:53<52:04,  1.32it/s]
3|2336|Loss: 0.7957531213760376:  36%|███▌      | 2336/6470 [28:54<51:44,  1.33it/s]
3|2337|Loss: 0.8782380223274231:  36%|███▌      | 2336/6470 [28:54<51:44,  1.33it/s]
3|2337|Loss: 0.8782380223274231:  36%|███▌      | 2337/6470 [28:54<51:46,  1.33it/s]
3|2338|Loss: 0.8415871262550354:  36%|███▌      | 2337/6470 [28:55<51:46,  1.33it/s]
3|2338|Loss: 0.8415871262550354:  36%|███▌      | 2338/6470 [28:55<50:11,  1.37it/s]
3|2339|Loss: 0.7568398118019104:  36%|███▌      | 2338/6470 [28:55<50:11,  1.37it/s]
3|2339|Loss: 0.7568398118019104:  36%|███▌      | 2339/6470 [28:56<52:49,  1.30it/s]
3|2340|Loss: 0.9727521538734436:  36%|███▌      | 2339/6470 [28:56<52:49,  1.30it/s]
3|2340|Loss: 0.9727521538734436:  36%|███▌      | 2340/6470 [28:57<52:56,  1.30it/s]
3|2341|Loss: 0.8403488397598267:  36%|███▌      | 2340/6470 [28:57<52:56,  1.30it/s]
3|2341|Loss: 0.8403488397598267:  36%|███▌      | 2341/6470 [28:57<53:15,  1.29it/s]
3|2342|Loss: 0.9192151427268982:  36%|███▌      | 2341/6470 [28:58<53:15,  1.29it/s]
3|2342|Loss: 0.9192151427268982:  36%|███▌      | 2342/6470 [28:58<53:37,  1.28it/s]
3|2343|Loss: 0.8790109753608704:  36%|███▌      | 2342/6470 [28:59<53:37,  1.28it/s]
3|2343|Loss: 0.8790109753608704:  36%|███▌      | 2343/6470 [28:59<55:53,  1.23it/s]
3|2344|Loss: 0.8744298815727234:  36%|███▌      | 2343/6470 [29:00<55:53,  1.23it/s]
3|2344|Loss: 0.8744298815727234:  36%|███▌      | 2344/6470 [29:00<56:12,  1.22it/s]
3|2345|Loss: 0.8742345571517944:  36%|███▌      | 2344/6470 [29:00<56:12,  1.22it/s]
3|2345|Loss: 0.8742345571517944:  36%|███▌      | 2345/6470 [29:01<53:01,  1.30it/s]
3|2346|Loss: 0.8135706782341003:  36%|███▌      | 2345/6470 [29:01<53:01,  1.30it/s]
3|2346|Loss: 0.8135706782341003:  36%|███▋      | 2346/6470 [29:01<52:02,  1.32it/s]
3|2347|Loss: 0.7853864431381226:  36%|███▋      | 2346/6470 [29:02<52:02,  1.32it/s]
3|2347|Loss: 0.7853864431381226:  36%|███▋      | 2347/6470 [29:02<52:15,  1.31it/s]
3|2348|Loss: 0.8560307025909424:  36%|███▋      | 2347/6470 [29:03<52:15,  1.31it/s]
3|2348|Loss: 0.8560307025909424:  36%|███▋      | 2348/6470 [29:03<51:04,  1.35it/s]
3|2349|Loss: 0.7408830523490906:  36%|███▋      | 2348/6470 [29:03<51:04,  1.35it/s]
3|2349|Loss: 0.7408830523490906:  36%|███▋      | 2349/6470 [29:04<50:03,  1.37it/s]
3|2350|Loss: 0.8234781622886658:  36%|███▋      | 2349/6470 [29:04<50:03,  1.37it/s]
3|2350|Loss: 0.8234781622886658:  36%|███▋      | 2350/6470 [29:04<52:07,  1.32it/s]
3|2351|Loss: 0.7650370001792908:  36%|███▋      | 2350/6470 [29:05<52:07,  1.32it/s]
3|2351|Loss: 0.7650370001792908:  36%|███▋      | 2351/6470 [29:05<52:23,  1.31it/s]
3|2352|Loss: 0.6513734459877014:  36%|███▋      | 2351/6470 [29:06<52:23,  1.31it/s]
3|2352|Loss: 0.6513734459877014:  36%|███▋      | 2352/6470 [29:06<52:43,  1.30it/s]
3|2353|Loss: 0.7831893563270569:  36%|███▋      | 2352/6470 [29:06<52:43,  1.30it/s]
3|2353|Loss: 0.7831893563270569:  36%|███▋      | 2353/6470 [29:07<51:06,  1.34it/s]
3|2354|Loss: 0.806452751159668:  36%|███▋      | 2353/6470 [29:07<51:06,  1.34it/s] 
3|2354|Loss: 0.806452751159668:  36%|███▋      | 2354/6470 [29:07<48:07,  1.43it/s]
3|2355|Loss: 0.8510108590126038:  36%|███▋      | 2354/6470 [29:08<48:07,  1.43it/s]
3|2355|Loss: 0.8510108590126038:  36%|███▋      | 2355/6470 [29:08<49:24,  1.39it/s]
3|2356|Loss: 0.9069400429725647:  36%|███▋      | 2355/6470 [29:08<49:24,  1.39it/s]
3|2356|Loss: 0.9069400429725647:  36%|███▋      | 2356/6470 [29:09<52:59,  1.29it/s]
3|2357|Loss: 0.8664589524269104:  36%|███▋      | 2356/6470 [29:09<52:59,  1.29it/s]
3|2357|Loss: 0.8664589524269104:  36%|███▋      | 2357/6470 [29:10<56:08,  1.22it/s]
3|2358|Loss: 0.9139599204063416:  36%|███▋      | 2357/6470 [29:10<56:08,  1.22it/s]
3|2358|Loss: 0.9139599204063416:  36%|███▋      | 2358/6470 [29:10<54:19,  1.26it/s]
3|2359|Loss: 0.825709879398346:  36%|███▋      | 2358/6470 [29:11<54:19,  1.26it/s] 
3|2359|Loss: 0.825709879398346:  36%|███▋      | 2359/6470 [29:11<52:10,  1.31it/s]
3|2360|Loss: 0.9411874413490295:  36%|███▋      | 2359/6470 [29:12<52:10,  1.31it/s]
3|2360|Loss: 0.9411874413490295:  36%|███▋      | 2360/6470 [29:12<50:53,  1.35it/s]
3|2361|Loss: 0.836166501045227:  36%|███▋      | 2360/6470 [29:12<50:53,  1.35it/s] 
3|2361|Loss: 0.836166501045227:  36%|███▋      | 2361/6470 [29:12<48:10,  1.42it/s]
3|2362|Loss: 0.7660216093063354:  36%|███▋      | 2361/6470 [29:13<48:10,  1.42it/s]
3|2362|Loss: 0.7660216093063354:  37%|███▋      | 2362/6470 [29:13<50:49,  1.35it/s]
3|2363|Loss: 0.794038712978363:  37%|███▋      | 2362/6470 [29:14<50:49,  1.35it/s] 
3|2363|Loss: 0.794038712978363:  37%|███▋      | 2363/6470 [29:14<54:04,  1.27it/s]
3|2364|Loss: 0.8337681293487549:  37%|███▋      | 2363/6470 [29:15<54:04,  1.27it/s]
3|2364|Loss: 0.8337681293487549:  37%|███▋      | 2364/6470 [29:15<48:27,  1.41it/s]
3|2365|Loss: 0.7814325094223022:  37%|███▋      | 2364/6470 [29:15<48:27,  1.41it/s]
3|2365|Loss: 0.7814325094223022:  37%|███▋      | 2365/6470 [29:15<48:59,  1.40it/s]
3|2366|Loss: 0.8580952286720276:  37%|███▋      | 2365/6470 [29:16<48:59,  1.40it/s]
3|2366|Loss: 0.8580952286720276:  37%|███▋      | 2366/6470 [29:16<53:09,  1.29it/s]
3|2367|Loss: 0.8535298705101013:  37%|███▋      | 2366/6470 [29:17<53:09,  1.29it/s]
3|2367|Loss: 0.8535298705101013:  37%|███▋      | 2367/6470 [29:17<47:39,  1.43it/s]
3|2368|Loss: 0.923409640789032:  37%|███▋      | 2367/6470 [29:17<47:39,  1.43it/s] 
3|2368|Loss: 0.923409640789032:  37%|███▋      | 2368/6470 [29:18<47:02,  1.45it/s]
3|2369|Loss: 0.7926602959632874:  37%|███▋      | 2368/6470 [29:18<47:02,  1.45it/s]
3|2369|Loss: 0.7926602959632874:  37%|███▋      | 2369/6470 [29:18<47:07,  1.45it/s]
3|2370|Loss: 0.8820161819458008:  37%|███▋      | 2369/6470 [29:19<47:07,  1.45it/s]
3|2370|Loss: 0.8820161819458008:  37%|███▋      | 2370/6470 [29:19<47:20,  1.44it/s]
3|2371|Loss: 0.8125261664390564:  37%|███▋      | 2370/6470 [29:19<47:20,  1.44it/s]
3|2371|Loss: 0.8125261664390564:  37%|███▋      | 2371/6470 [29:20<47:40,  1.43it/s]
3|2372|Loss: 0.7624171376228333:  37%|███▋      | 2371/6470 [29:20<47:40,  1.43it/s]
3|2372|Loss: 0.7624171376228333:  37%|███▋      | 2372/6470 [29:20<46:32,  1.47it/s]
3|2373|Loss: 0.8565515279769897:  37%|███▋      | 2372/6470 [29:21<46:32,  1.47it/s]
3|2373|Loss: 0.8565515279769897:  37%|███▋      | 2373/6470 [29:21<45:37,  1.50it/s]
3|2374|Loss: 0.792055070400238:  37%|███▋      | 2373/6470 [29:21<45:37,  1.50it/s] 
3|2374|Loss: 0.792055070400238:  37%|███▋      | 2374/6470 [29:22<46:45,  1.46it/s]
3|2375|Loss: 0.9688763618469238:  37%|███▋      | 2374/6470 [29:22<46:45,  1.46it/s]
3|2375|Loss: 0.9688763618469238:  37%|███▋      | 2375/6470 [29:23<49:43,  1.37it/s]
3|2376|Loss: 0.7344124913215637:  37%|███▋      | 2375/6470 [29:23<49:43,  1.37it/s]
3|2376|Loss: 0.7344124913215637:  37%|███▋      | 2376/6470 [29:23<49:31,  1.38it/s]
3|2377|Loss: 0.8195046782493591:  37%|███▋      | 2376/6470 [29:24<49:31,  1.38it/s]
3|2377|Loss: 0.8195046782493591:  37%|███▋      | 2377/6470 [29:24<50:17,  1.36it/s]
3|2378|Loss: 0.8888324499130249:  37%|███▋      | 2377/6470 [29:24<50:17,  1.36it/s]
3|2378|Loss: 0.8888324499130249:  37%|███▋      | 2378/6470 [29:25<52:08,  1.31it/s]
3|2379|Loss: 0.8403967618942261:  37%|███▋      | 2378/6470 [29:25<52:08,  1.31it/s]
3|2379|Loss: 0.8403967618942261:  37%|███▋      | 2379/6470 [29:26<55:48,  1.22it/s]
3|2380|Loss: 0.8440653681755066:  37%|███▋      | 2379/6470 [29:26<55:48,  1.22it/s]
3|2380|Loss: 0.8440653681755066:  37%|███▋      | 2380/6470 [29:27<58:59,  1.16it/s]
3|2381|Loss: 0.7913229465484619:  37%|███▋      | 2380/6470 [29:27<58:59,  1.16it/s]
3|2381|Loss: 0.7913229465484619:  37%|███▋      | 2381/6470 [29:28<1:00:48,  1.12it/s]
3|2382|Loss: 0.7409082055091858:  37%|███▋      | 2381/6470 [29:28<1:00:48,  1.12it/s]
3|2382|Loss: 0.7409082055091858:  37%|███▋      | 2382/6470 [29:29<59:20,  1.15it/s]  
3|2383|Loss: 0.7658185958862305:  37%|███▋      | 2382/6470 [29:29<59:20,  1.15it/s]
3|2383|Loss: 0.7658185958862305:  37%|███▋      | 2383/6470 [29:29<59:45,  1.14it/s]
3|2384|Loss: 0.9621506929397583:  37%|███▋      | 2383/6470 [29:30<59:45,  1.14it/s]
3|2384|Loss: 0.9621506929397583:  37%|███▋      | 2384/6470 [29:30<58:24,  1.17it/s]
3|2385|Loss: 0.8648282289505005:  37%|███▋      | 2384/6470 [29:31<58:24,  1.17it/s]
3|2385|Loss: 0.8648282289505005:  37%|███▋      | 2385/6470 [29:31<57:04,  1.19it/s]
3|2386|Loss: 0.8270275592803955:  37%|███▋      | 2385/6470 [29:31<57:04,  1.19it/s]
3|2386|Loss: 0.8270275592803955:  37%|███▋      | 2386/6470 [29:32<52:25,  1.30it/s]
3|2387|Loss: 0.8448732495307922:  37%|███▋      | 2386/6470 [29:32<52:25,  1.30it/s]
3|2387|Loss: 0.8448732495307922:  37%|███▋      | 2387/6470 [29:32<52:09,  1.30it/s]
3|2388|Loss: 0.8317924737930298:  37%|███▋      | 2387/6470 [29:33<52:09,  1.30it/s]
3|2388|Loss: 0.8317924737930298:  37%|███▋      | 2388/6470 [29:33<54:50,  1.24it/s]
3|2389|Loss: 0.9193211197853088:  37%|███▋      | 2388/6470 [29:34<54:50,  1.24it/s]
3|2389|Loss: 0.9193211197853088:  37%|███▋      | 2389/6470 [29:34<55:03,  1.24it/s]
3|2390|Loss: 0.8008098602294922:  37%|███▋      | 2389/6470 [29:35<55:03,  1.24it/s]
3|2390|Loss: 0.8008098602294922:  37%|███▋      | 2390/6470 [29:35<52:48,  1.29it/s]
3|2391|Loss: 0.8554871678352356:  37%|███▋      | 2390/6470 [29:35<52:48,  1.29it/s]
3|2391|Loss: 0.8554871678352356:  37%|███▋      | 2391/6470 [29:36<52:43,  1.29it/s]
3|2392|Loss: 0.7198799252510071:  37%|███▋      | 2391/6470 [29:36<52:43,  1.29it/s]
3|2392|Loss: 0.7198799252510071:  37%|███▋      | 2392/6470 [29:36<51:24,  1.32it/s]
3|2393|Loss: 0.9026800394058228:  37%|███▋      | 2392/6470 [29:37<51:24,  1.32it/s]
3|2393|Loss: 0.9026800394058228:  37%|███▋      | 2393/6470 [29:37<50:08,  1.36it/s]
3|2394|Loss: 0.8603230118751526:  37%|███▋      | 2393/6470 [29:37<50:08,  1.36it/s]
3|2394|Loss: 0.8603230118751526:  37%|███▋      | 2394/6470 [29:38<51:48,  1.31it/s]
3|2395|Loss: 0.8005753755569458:  37%|███▋      | 2394/6470 [29:38<51:48,  1.31it/s]
3|2395|Loss: 0.8005753755569458:  37%|███▋      | 2395/6470 [29:39<53:37,  1.27it/s]
3|2396|Loss: 0.9159541726112366:  37%|███▋      | 2395/6470 [29:39<53:37,  1.27it/s]
3|2396|Loss: 0.9159541726112366:  37%|███▋      | 2396/6470 [29:40<55:57,  1.21it/s]
3|2397|Loss: 0.7317602038383484:  37%|███▋      | 2396/6470 [29:40<55:57,  1.21it/s]
3|2397|Loss: 0.7317602038383484:  37%|███▋      | 2397/6470 [29:40<54:06,  1.25it/s]
3|2398|Loss: 0.754830539226532:  37%|███▋      | 2397/6470 [29:41<54:06,  1.25it/s] 
3|2398|Loss: 0.754830539226532:  37%|███▋      | 2398/6470 [29:41<54:01,  1.26it/s]
3|2399|Loss: 0.9428569078445435:  37%|███▋      | 2398/6470 [29:42<54:01,  1.26it/s]
3|2399|Loss: 0.9428569078445435:  37%|███▋      | 2399/6470 [29:42<53:52,  1.26it/s]
3|2400|Loss: 0.9130761623382568:  37%|███▋      | 2399/6470 [29:42<53:52,  1.26it/s]
3|2400|Loss: 0.9130761623382568:  37%|███▋      | 2400/6470 [29:43<55:05,  1.23it/s]
3|2401|Loss: 0.9478259086608887:  37%|███▋      | 2400/6470 [29:43<55:05,  1.23it/s]
3|2401|Loss: 0.9478259086608887:  37%|███▋      | 2401/6470 [29:44<56:08,  1.21it/s]
3|2402|Loss: 0.7783412337303162:  37%|███▋      | 2401/6470 [29:44<56:08,  1.21it/s]
3|2402|Loss: 0.7783412337303162:  37%|███▋      | 2402/6470 [29:44<56:42,  1.20it/s]
3|2403|Loss: 0.7563742995262146:  37%|███▋      | 2402/6470 [29:45<56:42,  1.20it/s]
3|2403|Loss: 0.7563742995262146:  37%|███▋      | 2403/6470 [29:45<55:34,  1.22it/s]
3|2404|Loss: 0.9277926683425903:  37%|███▋      | 2403/6470 [29:46<55:34,  1.22it/s]
3|2404|Loss: 0.9277926683425903:  37%|███▋      | 2404/6470 [29:46<55:05,  1.23it/s]
3|2405|Loss: 0.9994781613349915:  37%|███▋      | 2404/6470 [29:46<55:05,  1.23it/s]
3|2405|Loss: 0.9994781613349915:  37%|███▋      | 2405/6470 [29:47<53:26,  1.27it/s]
3|2406|Loss: 0.8377798795700073:  37%|███▋      | 2405/6470 [29:47<53:26,  1.27it/s]
3|2406|Loss: 0.8377798795700073:  37%|███▋      | 2406/6470 [29:48<54:52,  1.23it/s]
3|2407|Loss: 0.6772031784057617:  37%|███▋      | 2406/6470 [29:48<54:52,  1.23it/s]
3|2407|Loss: 0.6772031784057617:  37%|███▋      | 2407/6470 [29:48<54:48,  1.24it/s]
3|2408|Loss: 0.9612143039703369:  37%|███▋      | 2407/6470 [29:49<54:48,  1.24it/s]
3|2408|Loss: 0.9612143039703369:  37%|███▋      | 2408/6470 [29:49<54:37,  1.24it/s]
3|2409|Loss: 0.8084549307823181:  37%|███▋      | 2408/6470 [29:50<54:37,  1.24it/s]
3|2409|Loss: 0.8084549307823181:  37%|███▋      | 2409/6470 [29:50<55:15,  1.22it/s]
3|2410|Loss: 0.898705780506134:  37%|███▋      | 2409/6470 [29:51<55:15,  1.22it/s] 
3|2410|Loss: 0.898705780506134:  37%|███▋      | 2410/6470 [29:51<58:50,  1.15it/s]
3|2411|Loss: 0.9020965099334717:  37%|███▋      | 2410/6470 [29:52<58:50,  1.15it/s]
3|2411|Loss: 0.9020965099334717:  37%|███▋      | 2411/6470 [29:52<57:46,  1.17it/s]
3|2412|Loss: 1.0010969638824463:  37%|███▋      | 2411/6470 [29:52<57:46,  1.17it/s]
3|2412|Loss: 1.0010969638824463:  37%|███▋      | 2412/6470 [29:53<56:29,  1.20it/s]
3|2413|Loss: 0.8057630658149719:  37%|███▋      | 2412/6470 [29:53<56:29,  1.20it/s]
3|2413|Loss: 0.8057630658149719:  37%|███▋      | 2413/6470 [29:53<55:42,  1.21it/s]
3|2414|Loss: 0.8990150690078735:  37%|███▋      | 2413/6470 [29:54<55:42,  1.21it/s]
3|2414|Loss: 0.8990150690078735:  37%|███▋      | 2414/6470 [29:54<53:14,  1.27it/s]
3|2415|Loss: 0.8384726643562317:  37%|███▋      | 2414/6470 [29:55<53:14,  1.27it/s]
3|2415|Loss: 0.8384726643562317:  37%|███▋      | 2415/6470 [29:55<55:19,  1.22it/s]
3|2416|Loss: 0.7290749549865723:  37%|███▋      | 2415/6470 [29:56<55:19,  1.22it/s]
3|2416|Loss: 0.7290749549865723:  37%|███▋      | 2416/6470 [29:56<57:20,  1.18it/s]
3|2417|Loss: 0.8885933756828308:  37%|███▋      | 2416/6470 [29:56<57:20,  1.18it/s]
3|2417|Loss: 0.8885933756828308:  37%|███▋      | 2417/6470 [29:57<57:36,  1.17it/s]
3|2418|Loss: 0.8965322375297546:  37%|███▋      | 2417/6470 [29:57<57:36,  1.17it/s]
3|2418|Loss: 0.8965322375297546:  37%|███▋      | 2418/6470 [29:58<56:34,  1.19it/s]
3|2419|Loss: 0.8809742331504822:  37%|███▋      | 2418/6470 [29:58<56:34,  1.19it/s]
3|2419|Loss: 0.8809742331504822:  37%|███▋      | 2419/6470 [29:58<52:49,  1.28it/s]
3|2420|Loss: 0.8427199721336365:  37%|███▋      | 2419/6470 [29:59<52:49,  1.28it/s]
3|2420|Loss: 0.8427199721336365:  37%|███▋      | 2420/6470 [29:59<52:13,  1.29it/s]
3|2421|Loss: 0.9376198053359985:  37%|███▋      | 2420/6470 [29:59<52:13,  1.29it/s]
3|2421|Loss: 0.9376198053359985:  37%|███▋      | 2421/6470 [30:00<52:34,  1.28it/s]
3|2422|Loss: 0.8603322505950928:  37%|███▋      | 2421/6470 [30:00<52:34,  1.28it/s]
3|2422|Loss: 0.8603322505950928:  37%|███▋      | 2422/6470 [30:01<55:05,  1.22it/s]
3|2423|Loss: 0.7622102499008179:  37%|███▋      | 2422/6470 [30:01<55:05,  1.22it/s]
3|2423|Loss: 0.7622102499008179:  37%|███▋      | 2423/6470 [30:01<53:19,  1.27it/s]
3|2424|Loss: 0.8381993770599365:  37%|███▋      | 2423/6470 [30:02<53:19,  1.27it/s]
3|2424|Loss: 0.8381993770599365:  37%|███▋      | 2424/6470 [30:02<52:38,  1.28it/s]
3|2425|Loss: 0.9200531840324402:  37%|███▋      | 2424/6470 [30:03<52:38,  1.28it/s]
3|2425|Loss: 0.9200531840324402:  37%|███▋      | 2425/6470 [30:03<52:04,  1.29it/s]
3|2426|Loss: 0.9122135043144226:  37%|███▋      | 2425/6470 [30:03<52:04,  1.29it/s]
3|2426|Loss: 0.9122135043144226:  37%|███▋      | 2426/6470 [30:04<50:52,  1.32it/s]
3|2427|Loss: 0.9342432022094727:  37%|███▋      | 2426/6470 [30:04<50:52,  1.32it/s]
3|2427|Loss: 0.9342432022094727:  38%|███▊      | 2427/6470 [30:04<48:38,  1.39it/s]
3|2428|Loss: 0.8555288910865784:  38%|███▊      | 2427/6470 [30:05<48:38,  1.39it/s]
3|2428|Loss: 0.8555288910865784:  38%|███▊      | 2428/6470 [30:05<53:29,  1.26it/s]
3|2429|Loss: 0.8325128555297852:  38%|███▊      | 2428/6470 [30:06<53:29,  1.26it/s]
3|2429|Loss: 0.8325128555297852:  38%|███▊      | 2429/6470 [30:06<54:57,  1.23it/s]
3|2430|Loss: 0.753791093826294:  38%|███▊      | 2429/6470 [30:07<54:57,  1.23it/s] 
3|2430|Loss: 0.753791093826294:  38%|███▊      | 2430/6470 [30:07<52:39,  1.28it/s]
3|2431|Loss: 0.7911965847015381:  38%|███▊      | 2430/6470 [30:07<52:39,  1.28it/s]
3|2431|Loss: 0.7911965847015381:  38%|███▊      | 2431/6470 [30:08<56:27,  1.19it/s]
3|2432|Loss: 0.6882132291793823:  38%|███▊      | 2431/6470 [30:08<56:27,  1.19it/s]
3|2432|Loss: 0.6882132291793823:  38%|███▊      | 2432/6470 [30:09<58:22,  1.15it/s]
3|2433|Loss: 0.7009917497634888:  38%|███▊      | 2432/6470 [30:09<58:22,  1.15it/s]
3|2433|Loss: 0.7009917497634888:  38%|███▊      | 2433/6470 [30:10<56:54,  1.18it/s]
3|2434|Loss: 0.8312177658081055:  38%|███▊      | 2433/6470 [30:10<56:54,  1.18it/s]
3|2434|Loss: 0.8312177658081055:  38%|███▊      | 2434/6470 [30:10<55:37,  1.21it/s]
3|2435|Loss: 0.8461765050888062:  38%|███▊      | 2434/6470 [30:11<55:37,  1.21it/s]
3|2435|Loss: 0.8461765050888062:  38%|███▊      | 2435/6470 [30:11<55:09,  1.22it/s]
3|2436|Loss: 0.8246752619743347:  38%|███▊      | 2435/6470 [30:12<55:09,  1.22it/s]
3|2436|Loss: 0.8246752619743347:  38%|███▊      | 2436/6470 [30:12<54:52,  1.23it/s]
3|2437|Loss: 0.8680972456932068:  38%|███▊      | 2436/6470 [30:12<54:52,  1.23it/s]
3|2437|Loss: 0.8680972456932068:  38%|███▊      | 2437/6470 [30:13<54:02,  1.24it/s]
3|2438|Loss: 0.6644183397293091:  38%|███▊      | 2437/6470 [30:13<54:02,  1.24it/s]
3|2438|Loss: 0.6644183397293091:  38%|███▊      | 2438/6470 [30:13<50:17,  1.34it/s]
3|2439|Loss: 0.8199570775032043:  38%|███▊      | 2438/6470 [30:14<50:17,  1.34it/s]
3|2439|Loss: 0.8199570775032043:  38%|███▊      | 2439/6470 [30:14<51:16,  1.31it/s]
3|2440|Loss: 0.905972421169281:  38%|███▊      | 2439/6470 [30:15<51:16,  1.31it/s] 
3|2440|Loss: 0.905972421169281:  38%|███▊      | 2440/6470 [30:15<50:35,  1.33it/s]
3|2441|Loss: 0.8271886110305786:  38%|███▊      | 2440/6470 [30:15<50:35,  1.33it/s]
3|2441|Loss: 0.8271886110305786:  38%|███▊      | 2441/6470 [30:16<50:36,  1.33it/s]
3|2442|Loss: 0.8396708369255066:  38%|███▊      | 2441/6470 [30:16<50:36,  1.33it/s]
3|2442|Loss: 0.8396708369255066:  38%|███▊      | 2442/6470 [30:16<49:38,  1.35it/s]
3|2443|Loss: 0.743291974067688:  38%|███▊      | 2442/6470 [30:17<49:38,  1.35it/s] 
3|2443|Loss: 0.743291974067688:  38%|███▊      | 2443/6470 [30:17<48:53,  1.37it/s]
3|2444|Loss: 0.8232731819152832:  38%|███▊      | 2443/6470 [30:18<48:53,  1.37it/s]
3|2444|Loss: 0.8232731819152832:  38%|███▊      | 2444/6470 [30:18<51:34,  1.30it/s]
3|2445|Loss: 0.8542318940162659:  38%|███▊      | 2444/6470 [30:18<51:34,  1.30it/s]
3|2445|Loss: 0.8542318940162659:  38%|███▊      | 2445/6470 [30:19<53:26,  1.26it/s]
3|2446|Loss: 0.8613793849945068:  38%|███▊      | 2445/6470 [30:19<53:26,  1.26it/s]
3|2446|Loss: 0.8613793849945068:  38%|███▊      | 2446/6470 [30:20<53:57,  1.24it/s]
3|2447|Loss: 0.7934480309486389:  38%|███▊      | 2446/6470 [30:20<53:57,  1.24it/s]
3|2447|Loss: 0.7934480309486389:  38%|███▊      | 2447/6470 [30:20<49:49,  1.35it/s]
3|2448|Loss: 0.8149828910827637:  38%|███▊      | 2447/6470 [30:21<49:49,  1.35it/s]
3|2448|Loss: 0.8149828910827637:  38%|███▊      | 2448/6470 [30:21<49:08,  1.36it/s]
3|2449|Loss: 0.8135466575622559:  38%|███▊      | 2448/6470 [30:21<49:08,  1.36it/s]
3|2449|Loss: 0.8135466575622559:  38%|███▊      | 2449/6470 [30:22<49:33,  1.35it/s]
3|2450|Loss: 0.80264812707901:  38%|███▊      | 2449/6470 [30:22<49:33,  1.35it/s]  
3|2450|Loss: 0.80264812707901:  38%|███▊      | 2450/6470 [30:23<51:35,  1.30it/s]
3|2451|Loss: 0.7125853896141052:  38%|███▊      | 2450/6470 [30:23<51:35,  1.30it/s]
3|2451|Loss: 0.7125853896141052:  38%|███▊      | 2451/6470 [30:23<50:20,  1.33it/s]
3|2452|Loss: 0.8686822056770325:  38%|███▊      | 2451/6470 [30:24<50:20,  1.33it/s]
3|2452|Loss: 0.8686822056770325:  38%|███▊      | 2452/6470 [30:24<48:57,  1.37it/s]
3|2453|Loss: 0.7138301730155945:  38%|███▊      | 2452/6470 [30:24<48:57,  1.37it/s]
3|2453|Loss: 0.7138301730155945:  38%|███▊      | 2453/6470 [30:25<49:20,  1.36it/s]
3|2454|Loss: 0.8390145301818848:  38%|███▊      | 2453/6470 [30:25<49:20,  1.36it/s]
3|2454|Loss: 0.8390145301818848:  38%|███▊      | 2454/6470 [30:25<47:51,  1.40it/s]
3|2455|Loss: 0.7152771949768066:  38%|███▊      | 2454/6470 [30:26<47:51,  1.40it/s]
3|2455|Loss: 0.7152771949768066:  38%|███▊      | 2455/6470 [30:26<47:14,  1.42it/s]
3|2456|Loss: 0.9224456548690796:  38%|███▊      | 2455/6470 [30:26<47:14,  1.42it/s]
3|2456|Loss: 0.9224456548690796:  38%|███▊      | 2456/6470 [30:27<46:37,  1.43it/s]
3|2457|Loss: 0.9003881812095642:  38%|███▊      | 2456/6470 [30:27<46:37,  1.43it/s]
3|2457|Loss: 0.9003881812095642:  38%|███▊      | 2457/6470 [30:27<44:17,  1.51it/s]
3|2458|Loss: 0.8204003572463989:  38%|███▊      | 2457/6470 [30:28<44:17,  1.51it/s]
3|2458|Loss: 0.8204003572463989:  38%|███▊      | 2458/6470 [30:28<41:51,  1.60it/s]
3|2459|Loss: 0.6641295552253723:  38%|███▊      | 2458/6470 [30:28<41:51,  1.60it/s]
3|2459|Loss: 0.6641295552253723:  38%|███▊      | 2459/6470 [30:29<43:21,  1.54it/s]
3|2460|Loss: 0.7623811364173889:  38%|███▊      | 2459/6470 [30:29<43:21,  1.54it/s]
3|2460|Loss: 0.7623811364173889:  38%|███▊      | 2460/6470 [30:29<47:46,  1.40it/s]
3|2461|Loss: 0.8781923055648804:  38%|███▊      | 2460/6470 [30:30<47:46,  1.40it/s]
3|2461|Loss: 0.8781923055648804:  38%|███▊      | 2461/6470 [30:30<50:05,  1.33it/s]
3|2462|Loss: 0.9803867936134338:  38%|███▊      | 2461/6470 [30:31<50:05,  1.33it/s]
3|2462|Loss: 0.9803867936134338:  38%|███▊      | 2462/6470 [30:31<49:10,  1.36it/s]
3|2463|Loss: 0.8757328987121582:  38%|███▊      | 2462/6470 [30:31<49:10,  1.36it/s]
3|2463|Loss: 0.8757328987121582:  38%|███▊      | 2463/6470 [30:32<49:11,  1.36it/s]
3|2464|Loss: 0.854897141456604:  38%|███▊      | 2463/6470 [30:32<49:11,  1.36it/s] 
3|2464|Loss: 0.854897141456604:  38%|███▊      | 2464/6470 [30:32<49:34,  1.35it/s]
3|2465|Loss: 0.8349509239196777:  38%|███▊      | 2464/6470 [30:33<49:34,  1.35it/s]
3|2465|Loss: 0.8349509239196777:  38%|███▊      | 2465/6470 [30:33<50:33,  1.32it/s]
3|2466|Loss: 0.7390967607498169:  38%|███▊      | 2465/6470 [30:34<50:33,  1.32it/s]
3|2466|Loss: 0.7390967607498169:  38%|███▊      | 2466/6470 [30:34<51:21,  1.30it/s]
3|2467|Loss: 0.9695086479187012:  38%|███▊      | 2466/6470 [30:34<51:21,  1.30it/s]
3|2467|Loss: 0.9695086479187012:  38%|███▊      | 2467/6470 [30:35<52:54,  1.26it/s]
3|2468|Loss: 0.8294258117675781:  38%|███▊      | 2467/6470 [30:35<52:54,  1.26it/s]
3|2468|Loss: 0.8294258117675781:  38%|███▊      | 2468/6470 [30:36<50:04,  1.33it/s]
3|2469|Loss: 0.8871712684631348:  38%|███▊      | 2468/6470 [30:36<50:04,  1.33it/s]
3|2469|Loss: 0.8871712684631348:  38%|███▊      | 2469/6470 [30:36<47:31,  1.40it/s]
3|2470|Loss: 0.5940636992454529:  38%|███▊      | 2469/6470 [30:37<47:31,  1.40it/s]
3|2470|Loss: 0.5940636992454529:  38%|███▊      | 2470/6470 [30:37<52:44,  1.26it/s]
3|2471|Loss: 0.8368556499481201:  38%|███▊      | 2470/6470 [30:38<52:44,  1.26it/s]
3|2471|Loss: 0.8368556499481201:  38%|███▊      | 2471/6470 [30:38<53:56,  1.24it/s]
3|2472|Loss: 0.73643958568573:  38%|███▊      | 2471/6470 [30:38<53:56,  1.24it/s]  
3|2472|Loss: 0.73643958568573:  38%|███▊      | 2472/6470 [30:39<52:16,  1.27it/s]
3|2473|Loss: 0.885111391544342:  38%|███▊      | 2472/6470 [30:39<52:16,  1.27it/s]
3|2473|Loss: 0.885111391544342:  38%|███▊      | 2473/6470 [30:39<49:51,  1.34it/s]
3|2474|Loss: 0.8154335021972656:  38%|███▊      | 2473/6470 [30:40<49:51,  1.34it/s]
3|2474|Loss: 0.8154335021972656:  38%|███▊      | 2474/6470 [30:40<48:16,  1.38it/s]
3|2475|Loss: 0.8121500611305237:  38%|███▊      | 2474/6470 [30:40<48:16,  1.38it/s]
3|2475|Loss: 0.8121500611305237:  38%|███▊      | 2475/6470 [30:41<48:30,  1.37it/s]
3|2476|Loss: 0.8197193145751953:  38%|███▊      | 2475/6470 [30:41<48:30,  1.37it/s]
3|2476|Loss: 0.8197193145751953:  38%|███▊      | 2476/6470 [30:42<53:09,  1.25it/s]
3|2477|Loss: 0.9264952540397644:  38%|███▊      | 2476/6470 [30:42<53:09,  1.25it/s]
3|2477|Loss: 0.9264952540397644:  38%|███▊      | 2477/6470 [30:43<53:29,  1.24it/s]
3|2478|Loss: 0.8618131875991821:  38%|███▊      | 2477/6470 [30:43<53:29,  1.24it/s]
3|2478|Loss: 0.8618131875991821:  38%|███▊      | 2478/6470 [30:43<52:37,  1.26it/s]
3|2479|Loss: 0.9771705865859985:  38%|███▊      | 2478/6470 [30:44<52:37,  1.26it/s]
3|2479|Loss: 0.9771705865859985:  38%|███▊      | 2479/6470 [30:44<50:00,  1.33it/s]
3|2480|Loss: 0.755485475063324:  38%|███▊      | 2479/6470 [30:44<50:00,  1.33it/s] 
3|2480|Loss: 0.755485475063324:  38%|███▊      | 2480/6470 [30:45<48:39,  1.37it/s]
3|2481|Loss: 0.7787696123123169:  38%|███▊      | 2480/6470 [30:45<48:39,  1.37it/s]
3|2481|Loss: 0.7787696123123169:  38%|███▊      | 2481/6470 [30:45<49:47,  1.34it/s]
3|2482|Loss: 0.8545792698860168:  38%|███▊      | 2481/6470 [30:46<49:47,  1.34it/s]
3|2482|Loss: 0.8545792698860168:  38%|███▊      | 2482/6470 [30:46<50:40,  1.31it/s]
3|2483|Loss: 0.6978638172149658:  38%|███▊      | 2482/6470 [30:47<50:40,  1.31it/s]
3|2483|Loss: 0.6978638172149658:  38%|███▊      | 2483/6470 [30:47<55:25,  1.20it/s]
3|2484|Loss: 0.947924017906189:  38%|███▊      | 2483/6470 [30:48<55:25,  1.20it/s] 
3|2484|Loss: 0.947924017906189:  38%|███▊      | 2484/6470 [30:48<55:36,  1.19it/s]
3|2485|Loss: 0.8975827693939209:  38%|███▊      | 2484/6470 [30:48<55:36,  1.19it/s]
3|2485|Loss: 0.8975827693939209:  38%|███▊      | 2485/6470 [30:49<54:44,  1.21it/s]
3|2486|Loss: 0.8262965083122253:  38%|███▊      | 2485/6470 [30:49<54:44,  1.21it/s]
3|2486|Loss: 0.8262965083122253:  38%|███▊      | 2486/6470 [30:50<53:23,  1.24it/s]
3|2487|Loss: 0.8063246011734009:  38%|███▊      | 2486/6470 [30:50<53:23,  1.24it/s]
3|2487|Loss: 0.8063246011734009:  38%|███▊      | 2487/6470 [30:50<50:14,  1.32it/s]
3|2488|Loss: 0.7427148818969727:  38%|███▊      | 2487/6470 [30:51<50:14,  1.32it/s]
3|2488|Loss: 0.7427148818969727:  38%|███▊      | 2488/6470 [30:51<48:09,  1.38it/s]
3|2489|Loss: 0.991836667060852:  38%|███▊      | 2488/6470 [30:51<48:09,  1.38it/s] 
3|2489|Loss: 0.991836667060852:  38%|███▊      | 2489/6470 [30:52<47:25,  1.40it/s]
3|2490|Loss: 0.7380006909370422:  38%|███▊      | 2489/6470 [30:52<47:25,  1.40it/s]
3|2490|Loss: 0.7380006909370422:  38%|███▊      | 2490/6470 [30:53<51:26,  1.29it/s]
3|2491|Loss: 0.8025564551353455:  38%|███▊      | 2490/6470 [30:53<51:26,  1.29it/s]
3|2491|Loss: 0.8025564551353455:  39%|███▊      | 2491/6470 [30:53<50:38,  1.31it/s]
3|2492|Loss: 0.9482300877571106:  39%|███▊      | 2491/6470 [30:54<50:38,  1.31it/s]
3|2492|Loss: 0.9482300877571106:  39%|███▊      | 2492/6470 [30:54<48:41,  1.36it/s]
3|2493|Loss: 0.7416490316390991:  39%|███▊      | 2492/6470 [30:54<48:41,  1.36it/s]
3|2493|Loss: 0.7416490316390991:  39%|███▊      | 2493/6470 [30:55<50:45,  1.31it/s]
3|2494|Loss: 0.7725226879119873:  39%|███▊      | 2493/6470 [30:55<50:45,  1.31it/s]
3|2494|Loss: 0.7725226879119873:  39%|███▊      | 2494/6470 [30:56<51:01,  1.30it/s]
3|2495|Loss: 0.8510125875473022:  39%|███▊      | 2494/6470 [30:56<51:01,  1.30it/s]
3|2495|Loss: 0.8510125875473022:  39%|███▊      | 2495/6470 [30:56<50:54,  1.30it/s]
3|2496|Loss: 0.8543084859848022:  39%|███▊      | 2495/6470 [30:57<50:54,  1.30it/s]
3|2496|Loss: 0.8543084859848022:  39%|███▊      | 2496/6470 [30:57<51:27,  1.29it/s]
3|2497|Loss: 0.687091588973999:  39%|███▊      | 2496/6470 [30:57<51:27,  1.29it/s] 
3|2497|Loss: 0.687091588973999:  39%|███▊      | 2497/6470 [30:58<48:39,  1.36it/s]
3|2498|Loss: 1.0216460227966309:  39%|███▊      | 2497/6470 [30:58<48:39,  1.36it/s]
3|2498|Loss: 1.0216460227966309:  39%|███▊      | 2498/6470 [30:59<50:40,  1.31it/s]
3|2499|Loss: 0.655209481716156:  39%|███▊      | 2498/6470 [30:59<50:40,  1.31it/s] 
3|2499|Loss: 0.655209481716156:  39%|███▊      | 2499/6470 [31:00<54:53,  1.21it/s]
3|2500|Loss: 0.6956525444984436:  39%|███▊      | 2499/6470 [31:00<54:53,  1.21it/s]
3|2500|Loss: 0.6956525444984436:  39%|███▊      | 2500/6470 [31:00<51:10,  1.29it/s]
3|2501|Loss: 0.7035048604011536:  39%|███▊      | 2500/6470 [31:01<51:10,  1.29it/s]
3|2501|Loss: 0.7035048604011536:  39%|███▊      | 2501/6470 [31:01<48:22,  1.37it/s]
3|2502|Loss: 0.7894946336746216:  39%|███▊      | 2501/6470 [31:01<48:22,  1.37it/s]
3|2502|Loss: 0.7894946336746216:  39%|███▊      | 2502/6470 [31:02<48:55,  1.35it/s]
3|2503|Loss: 0.8758710622787476:  39%|███▊      | 2502/6470 [31:02<48:55,  1.35it/s]
3|2503|Loss: 0.8758710622787476:  39%|███▊      | 2503/6470 [31:02<48:34,  1.36it/s]
3|2504|Loss: 0.9472401142120361:  39%|███▊      | 2503/6470 [31:03<48:34,  1.36it/s]
3|2504|Loss: 0.9472401142120361:  39%|███▊      | 2504/6470 [31:03<44:54,  1.47it/s]
3|2505|Loss: 0.7769268751144409:  39%|███▊      | 2504/6470 [31:03<44:54,  1.47it/s]
3|2505|Loss: 0.7769268751144409:  39%|███▊      | 2505/6470 [31:04<45:20,  1.46it/s]
3|2506|Loss: 0.8913172483444214:  39%|███▊      | 2505/6470 [31:04<45:20,  1.46it/s]
3|2506|Loss: 0.8913172483444214:  39%|███▊      | 2506/6470 [31:04<45:50,  1.44it/s]
3|2507|Loss: 0.7168439030647278:  39%|███▊      | 2506/6470 [31:05<45:50,  1.44it/s]
3|2507|Loss: 0.7168439030647278:  39%|███▊      | 2507/6470 [31:05<48:48,  1.35it/s]
3|2508|Loss: 0.7354010343551636:  39%|███▊      | 2507/6470 [31:06<48:48,  1.35it/s]
3|2508|Loss: 0.7354010343551636:  39%|███▉      | 2508/6470 [31:06<51:52,  1.27it/s]
3|2509|Loss: 0.7872351408004761:  39%|███▉      | 2508/6470 [31:06<51:52,  1.27it/s]
3|2509|Loss: 0.7872351408004761:  39%|███▉      | 2509/6470 [31:07<47:21,  1.39it/s]
3|2510|Loss: 0.8897982835769653:  39%|███▉      | 2509/6470 [31:07<47:21,  1.39it/s]
3|2510|Loss: 0.8897982835769653:  39%|███▉      | 2510/6470 [31:07<44:36,  1.48it/s]
3|2511|Loss: 0.932923436164856:  39%|███▉      | 2510/6470 [31:08<44:36,  1.48it/s] 
3|2511|Loss: 0.932923436164856:  39%|███▉      | 2511/6470 [31:08<43:24,  1.52it/s]
3|2512|Loss: 0.9973691701889038:  39%|███▉      | 2511/6470 [31:08<43:24,  1.52it/s]
3|2512|Loss: 0.9973691701889038:  39%|███▉      | 2512/6470 [31:09<46:38,  1.41it/s]
3|2513|Loss: 0.8538785576820374:  39%|███▉      | 2512/6470 [31:09<46:38,  1.41it/s]
3|2513|Loss: 0.8538785576820374:  39%|███▉      | 2513/6470 [31:09<47:55,  1.38it/s]
3|2514|Loss: 0.8470877408981323:  39%|███▉      | 2513/6470 [31:10<47:55,  1.38it/s]
3|2514|Loss: 0.8470877408981323:  39%|███▉      | 2514/6470 [31:10<48:41,  1.35it/s]
3|2515|Loss: 0.8923549056053162:  39%|███▉      | 2514/6470 [31:11<48:41,  1.35it/s]
3|2515|Loss: 0.8923549056053162:  39%|███▉      | 2515/6470 [31:11<49:43,  1.33it/s]
3|2516|Loss: 0.8901378512382507:  39%|███▉      | 2515/6470 [31:11<49:43,  1.33it/s]
3|2516|Loss: 0.8901378512382507:  39%|███▉      | 2516/6470 [31:12<53:49,  1.22it/s]
3|2517|Loss: 1.0194804668426514:  39%|███▉      | 2516/6470 [31:12<53:49,  1.22it/s]
3|2517|Loss: 1.0194804668426514:  39%|███▉      | 2517/6470 [31:13<56:33,  1.16it/s]
3|2518|Loss: 0.874037504196167:  39%|███▉      | 2517/6470 [31:13<56:33,  1.16it/s] 
3|2518|Loss: 0.874037504196167:  39%|███▉      | 2518/6470 [31:14<58:12,  1.13it/s]
3|2519|Loss: 0.9716500639915466:  39%|███▉      | 2518/6470 [31:14<58:12,  1.13it/s]
3|2519|Loss: 0.9716500639915466:  39%|███▉      | 2519/6470 [31:15<55:33,  1.19it/s]
3|2520|Loss: 0.808427095413208:  39%|███▉      | 2519/6470 [31:15<55:33,  1.19it/s] 
3|2520|Loss: 0.808427095413208:  39%|███▉      | 2520/6470 [31:15<50:29,  1.30it/s]
3|2521|Loss: 0.9247750043869019:  39%|███▉      | 2520/6470 [31:16<50:29,  1.30it/s]
3|2521|Loss: 0.9247750043869019:  39%|███▉      | 2521/6470 [31:16<51:12,  1.29it/s]
3|2522|Loss: 0.8184196352958679:  39%|███▉      | 2521/6470 [31:16<51:12,  1.29it/s]
3|2522|Loss: 0.8184196352958679:  39%|███▉      | 2522/6470 [31:17<52:47,  1.25it/s]
3|2523|Loss: 0.7921072840690613:  39%|███▉      | 2522/6470 [31:17<52:47,  1.25it/s]
3|2523|Loss: 0.7921072840690613:  39%|███▉      | 2523/6470 [31:18<54:37,  1.20it/s]
3|2524|Loss: 0.9543989896774292:  39%|███▉      | 2523/6470 [31:18<54:37,  1.20it/s]
3|2524|Loss: 0.9543989896774292:  39%|███▉      | 2524/6470 [31:18<52:44,  1.25it/s]
3|2525|Loss: 0.8042364120483398:  39%|███▉      | 2524/6470 [31:19<52:44,  1.25it/s]
3|2525|Loss: 0.8042364120483398:  39%|███▉      | 2525/6470 [31:19<50:10,  1.31it/s]
3|2526|Loss: 0.8625860810279846:  39%|███▉      | 2525/6470 [31:20<50:10,  1.31it/s]
3|2526|Loss: 0.8625860810279846:  39%|███▉      | 2526/6470 [31:20<51:12,  1.28it/s]
3|2527|Loss: 1.0256530046463013:  39%|███▉      | 2526/6470 [31:20<51:12,  1.28it/s]
3|2527|Loss: 1.0256530046463013:  39%|███▉      | 2527/6470 [31:21<50:11,  1.31it/s]
3|2528|Loss: 0.7946916818618774:  39%|███▉      | 2527/6470 [31:21<50:11,  1.31it/s]
3|2528|Loss: 0.7946916818618774:  39%|███▉      | 2528/6470 [31:21<47:04,  1.40it/s]
3|2529|Loss: 0.8134947419166565:  39%|███▉      | 2528/6470 [31:22<47:04,  1.40it/s]
3|2529|Loss: 0.8134947419166565:  39%|███▉      | 2529/6470 [31:22<50:17,  1.31it/s]
3|2530|Loss: 0.8098515868186951:  39%|███▉      | 2529/6470 [31:23<50:17,  1.31it/s]
3|2530|Loss: 0.8098515868186951:  39%|███▉      | 2530/6470 [31:23<52:01,  1.26it/s]
3|2531|Loss: 0.727494478225708:  39%|███▉      | 2530/6470 [31:23<52:01,  1.26it/s] 
3|2531|Loss: 0.727494478225708:  39%|███▉      | 2531/6470 [31:24<51:55,  1.26it/s]
3|2532|Loss: 0.8148058652877808:  39%|███▉      | 2531/6470 [31:24<51:55,  1.26it/s]
3|2532|Loss: 0.8148058652877808:  39%|███▉      | 2532/6470 [31:24<47:46,  1.37it/s]
3|2533|Loss: 0.8548603653907776:  39%|███▉      | 2532/6470 [31:25<47:46,  1.37it/s]
3|2533|Loss: 0.8548603653907776:  39%|███▉      | 2533/6470 [31:25<50:00,  1.31it/s]
3|2534|Loss: 0.8394062519073486:  39%|███▉      | 2533/6470 [31:26<50:00,  1.31it/s]
3|2534|Loss: 0.8394062519073486:  39%|███▉      | 2534/6470 [31:26<47:07,  1.39it/s]
3|2535|Loss: 0.8046984672546387:  39%|███▉      | 2534/6470 [31:26<47:07,  1.39it/s]
3|2535|Loss: 0.8046984672546387:  39%|███▉      | 2535/6470 [31:26<42:58,  1.53it/s]
3|2536|Loss: 0.8364996910095215:  39%|███▉      | 2535/6470 [31:27<42:58,  1.53it/s]
3|2536|Loss: 0.8364996910095215:  39%|███▉      | 2536/6470 [31:27<42:27,  1.54it/s]
3|2537|Loss: 0.7999541163444519:  39%|███▉      | 2536/6470 [31:27<42:27,  1.54it/s]
3|2537|Loss: 0.7999541163444519:  39%|███▉      | 2537/6470 [31:28<45:07,  1.45it/s]
3|2538|Loss: 1.0188510417938232:  39%|███▉      | 2537/6470 [31:28<45:07,  1.45it/s]
3|2538|Loss: 1.0188510417938232:  39%|███▉      | 2538/6470 [31:28<46:51,  1.40it/s]
3|2539|Loss: 0.7171992063522339:  39%|███▉      | 2538/6470 [31:29<46:51,  1.40it/s]
3|2539|Loss: 0.7171992063522339:  39%|███▉      | 2539/6470 [31:29<49:19,  1.33it/s]
3|2540|Loss: 0.8203064203262329:  39%|███▉      | 2539/6470 [31:30<49:19,  1.33it/s]
3|2540|Loss: 0.8203064203262329:  39%|███▉      | 2540/6470 [31:30<50:18,  1.30it/s]
3|2541|Loss: 0.8018730282783508:  39%|███▉      | 2540/6470 [31:31<50:18,  1.30it/s]
3|2541|Loss: 0.8018730282783508:  39%|███▉      | 2541/6470 [31:31<49:26,  1.32it/s]
3|2542|Loss: 0.8087003231048584:  39%|███▉      | 2541/6470 [31:31<49:26,  1.32it/s]
3|2542|Loss: 0.8087003231048584:  39%|███▉      | 2542/6470 [31:32<49:54,  1.31it/s]
3|2543|Loss: 0.7978575229644775:  39%|███▉      | 2542/6470 [31:32<49:54,  1.31it/s]
3|2543|Loss: 0.7978575229644775:  39%|███▉      | 2543/6470 [31:32<51:30,  1.27it/s]
3|2544|Loss: 1.0019484758377075:  39%|███▉      | 2543/6470 [31:33<51:30,  1.27it/s]
3|2544|Loss: 1.0019484758377075:  39%|███▉      | 2544/6470 [31:33<49:58,  1.31it/s]
3|2545|Loss: 0.8158782124519348:  39%|███▉      | 2544/6470 [31:34<49:58,  1.31it/s]
3|2545|Loss: 0.8158782124519348:  39%|███▉      | 2545/6470 [31:34<51:50,  1.26it/s]
3|2546|Loss: 0.8197526335716248:  39%|███▉      | 2545/6470 [31:34<51:50,  1.26it/s]
3|2546|Loss: 0.8197526335716248:  39%|███▉      | 2546/6470 [31:35<51:31,  1.27it/s]
3|2547|Loss: 0.6557902693748474:  39%|███▉      | 2546/6470 [31:35<51:31,  1.27it/s]
3|2547|Loss: 0.6557902693748474:  39%|███▉      | 2547/6470 [31:35<47:45,  1.37it/s]
3|2548|Loss: 0.9503841400146484:  39%|███▉      | 2547/6470 [31:36<47:45,  1.37it/s]
3|2548|Loss: 0.9503841400146484:  39%|███▉      | 2548/6470 [31:36<47:15,  1.38it/s]
3|2549|Loss: 0.6981104016304016:  39%|███▉      | 2548/6470 [31:36<47:15,  1.38it/s]
3|2549|Loss: 0.6981104016304016:  39%|███▉      | 2549/6470 [31:37<44:47,  1.46it/s]
3|2550|Loss: 0.8763235807418823:  39%|███▉      | 2549/6470 [31:37<44:47,  1.46it/s]
3|2550|Loss: 0.8763235807418823:  39%|███▉      | 2550/6470 [31:37<44:10,  1.48it/s]
3|2551|Loss: 0.8757652640342712:  39%|███▉      | 2550/6470 [31:38<44:10,  1.48it/s]
3|2551|Loss: 0.8757652640342712:  39%|███▉      | 2551/6470 [31:38<43:23,  1.51it/s]
3|2552|Loss: 0.6831368803977966:  39%|███▉      | 2551/6470 [31:38<43:23,  1.51it/s]
3|2552|Loss: 0.6831368803977966:  39%|███▉      | 2552/6470 [31:39<46:33,  1.40it/s]
3|2553|Loss: 0.9440231919288635:  39%|███▉      | 2552/6470 [31:39<46:33,  1.40it/s]
3|2553|Loss: 0.9440231919288635:  39%|███▉      | 2553/6470 [31:39<45:03,  1.45it/s]
3|2554|Loss: 0.8264425992965698:  39%|███▉      | 2553/6470 [31:40<45:03,  1.45it/s]
3|2554|Loss: 0.8264425992965698:  39%|███▉      | 2554/6470 [31:40<45:58,  1.42it/s]
3|2555|Loss: 0.9077993035316467:  39%|███▉      | 2554/6470 [31:41<45:58,  1.42it/s]
3|2555|Loss: 0.9077993035316467:  39%|███▉      | 2555/6470 [31:41<43:54,  1.49it/s]
3|2556|Loss: 0.9595478773117065:  39%|███▉      | 2555/6470 [31:41<43:54,  1.49it/s]
3|2556|Loss: 0.9595478773117065:  40%|███▉      | 2556/6470 [31:41<41:11,  1.58it/s]
3|2557|Loss: 0.9097460508346558:  40%|███▉      | 2556/6470 [31:42<41:11,  1.58it/s]
3|2557|Loss: 0.9097460508346558:  40%|███▉      | 2557/6470 [31:42<44:56,  1.45it/s]
3|2558|Loss: 0.8853573203086853:  40%|███▉      | 2557/6470 [31:43<44:56,  1.45it/s]
3|2558|Loss: 0.8853573203086853:  40%|███▉      | 2558/6470 [31:43<46:26,  1.40it/s]
3|2559|Loss: 0.8348842859268188:  40%|███▉      | 2558/6470 [31:43<46:26,  1.40it/s]
3|2559|Loss: 0.8348842859268188:  40%|███▉      | 2559/6470 [31:44<43:21,  1.50it/s]
3|2560|Loss: 0.817021369934082:  40%|███▉      | 2559/6470 [31:44<43:21,  1.50it/s] 
3|2560|Loss: 0.817021369934082:  40%|███▉      | 2560/6470 [31:44<44:21,  1.47it/s]
3|2561|Loss: 0.744155764579773:  40%|███▉      | 2560/6470 [31:45<44:21,  1.47it/s]
3|2561|Loss: 0.744155764579773:  40%|███▉      | 2561/6470 [31:45<42:56,  1.52it/s]
3|2562|Loss: 0.7290343642234802:  40%|███▉      | 2561/6470 [31:45<42:56,  1.52it/s]
3|2562|Loss: 0.7290343642234802:  40%|███▉      | 2562/6470 [31:46<43:09,  1.51it/s]
3|2563|Loss: 0.7330642342567444:  40%|███▉      | 2562/6470 [31:46<43:09,  1.51it/s]
3|2563|Loss: 0.7330642342567444:  40%|███▉      | 2563/6470 [31:46<44:50,  1.45it/s]
3|2564|Loss: 0.9626263976097107:  40%|███▉      | 2563/6470 [31:47<44:50,  1.45it/s]
3|2564|Loss: 0.9626263976097107:  40%|███▉      | 2564/6470 [31:47<43:35,  1.49it/s]
3|2565|Loss: 0.9134253263473511:  40%|███▉      | 2564/6470 [31:47<43:35,  1.49it/s]
3|2565|Loss: 0.9134253263473511:  40%|███▉      | 2565/6470 [31:47<42:24,  1.53it/s]
3|2566|Loss: 0.913365364074707:  40%|███▉      | 2565/6470 [31:48<42:24,  1.53it/s] 
3|2566|Loss: 0.913365364074707:  40%|███▉      | 2566/6470 [31:48<45:51,  1.42it/s]
3|2567|Loss: 0.8471647500991821:  40%|███▉      | 2566/6470 [31:49<45:51,  1.42it/s]
3|2567|Loss: 0.8471647500991821:  40%|███▉      | 2567/6470 [31:49<45:48,  1.42it/s]
3|2568|Loss: 0.7167336344718933:  40%|███▉      | 2567/6470 [31:49<45:48,  1.42it/s]
3|2568|Loss: 0.7167336344718933:  40%|███▉      | 2568/6470 [31:50<45:46,  1.42it/s]
3|2569|Loss: 0.7474924325942993:  40%|███▉      | 2568/6470 [31:50<45:46,  1.42it/s]
3|2569|Loss: 0.7474924325942993:  40%|███▉      | 2569/6470 [31:51<49:39,  1.31it/s]
3|2570|Loss: 0.8564480543136597:  40%|███▉      | 2569/6470 [31:51<49:39,  1.31it/s]
3|2570|Loss: 0.8564480543136597:  40%|███▉      | 2570/6470 [31:51<50:53,  1.28it/s]
3|2571|Loss: 0.9495518207550049:  40%|███▉      | 2570/6470 [31:52<50:53,  1.28it/s]
3|2571|Loss: 0.9495518207550049:  40%|███▉      | 2571/6470 [31:52<51:03,  1.27it/s]
3|2572|Loss: 0.8813609480857849:  40%|███▉      | 2571/6470 [31:53<51:03,  1.27it/s]
3|2572|Loss: 0.8813609480857849:  40%|███▉      | 2572/6470 [31:53<50:22,  1.29it/s]
3|2573|Loss: 0.6767014265060425:  40%|███▉      | 2572/6470 [31:53<50:22,  1.29it/s]
3|2573|Loss: 0.6767014265060425:  40%|███▉      | 2573/6470 [31:54<49:05,  1.32it/s]
3|2574|Loss: 1.05562424659729:  40%|███▉      | 2573/6470 [31:54<49:05,  1.32it/s]  
3|2574|Loss: 1.05562424659729:  40%|███▉      | 2574/6470 [31:54<49:30,  1.31it/s]
3|2575|Loss: 0.7788315415382385:  40%|███▉      | 2574/6470 [31:55<49:30,  1.31it/s]
3|2575|Loss: 0.7788315415382385:  40%|███▉      | 2575/6470 [31:55<49:44,  1.30it/s]
3|2576|Loss: 0.74791020154953:  40%|███▉      | 2575/6470 [31:56<49:44,  1.30it/s]  
3|2576|Loss: 0.74791020154953:  40%|███▉      | 2576/6470 [31:56<48:00,  1.35it/s]
3|2577|Loss: 0.741144597530365:  40%|███▉      | 2576/6470 [31:56<48:00,  1.35it/s]
3|2577|Loss: 0.741144597530365:  40%|███▉      | 2577/6470 [31:57<46:29,  1.40it/s]
3|2578|Loss: 0.7991889715194702:  40%|███▉      | 2577/6470 [31:57<46:29,  1.40it/s]
3|2578|Loss: 0.7991889715194702:  40%|███▉      | 2578/6470 [31:57<45:56,  1.41it/s]
3|2579|Loss: 0.7820688486099243:  40%|███▉      | 2578/6470 [31:58<45:56,  1.41it/s]
3|2579|Loss: 0.7820688486099243:  40%|███▉      | 2579/6470 [31:58<48:50,  1.33it/s]
3|2580|Loss: 0.907699704170227:  40%|███▉      | 2579/6470 [31:59<48:50,  1.33it/s] 
3|2580|Loss: 0.907699704170227:  40%|███▉      | 2580/6470 [31:59<48:02,  1.35it/s]
3|2581|Loss: 0.8938971161842346:  40%|███▉      | 2580/6470 [31:59<48:02,  1.35it/s]
3|2581|Loss: 0.8938971161842346:  40%|███▉      | 2581/6470 [32:00<50:48,  1.28it/s]
3|2582|Loss: 0.9043059945106506:  40%|███▉      | 2581/6470 [32:00<50:48,  1.28it/s]
3|2582|Loss: 0.9043059945106506:  40%|███▉      | 2582/6470 [32:00<49:44,  1.30it/s]
3|2583|Loss: 0.8502635359764099:  40%|███▉      | 2582/6470 [32:01<49:44,  1.30it/s]
3|2583|Loss: 0.8502635359764099:  40%|███▉      | 2583/6470 [32:01<51:26,  1.26it/s]
3|2584|Loss: 0.8172169923782349:  40%|███▉      | 2583/6470 [32:02<51:26,  1.26it/s]
3|2584|Loss: 0.8172169923782349:  40%|███▉      | 2584/6470 [32:02<50:27,  1.28it/s]
3|2585|Loss: 0.8508566617965698:  40%|███▉      | 2584/6470 [32:02<50:27,  1.28it/s]
3|2585|Loss: 0.8508566617965698:  40%|███▉      | 2585/6470 [32:03<49:04,  1.32it/s]
3|2586|Loss: 0.9118728041648865:  40%|███▉      | 2585/6470 [32:03<49:04,  1.32it/s]
3|2586|Loss: 0.9118728041648865:  40%|███▉      | 2586/6470 [32:03<45:36,  1.42it/s]
3|2587|Loss: 0.9079848527908325:  40%|███▉      | 2586/6470 [32:04<45:36,  1.42it/s]
3|2587|Loss: 0.9079848527908325:  40%|███▉      | 2587/6470 [32:04<45:02,  1.44it/s]
3|2588|Loss: 0.83622145652771:  40%|███▉      | 2587/6470 [32:04<45:02,  1.44it/s]  
3|2588|Loss: 0.83622145652771:  40%|████      | 2588/6470 [32:05<45:11,  1.43it/s]
3|2589|Loss: 0.8274253606796265:  40%|████      | 2588/6470 [32:05<45:11,  1.43it/s]
3|2589|Loss: 0.8274253606796265:  40%|████      | 2589/6470 [32:05<42:16,  1.53it/s]
3|2590|Loss: 0.9774089455604553:  40%|████      | 2589/6470 [32:06<42:16,  1.53it/s]
3|2590|Loss: 0.9774089455604553:  40%|████      | 2590/6470 [32:06<40:18,  1.60it/s]
3|2591|Loss: 0.8570700883865356:  40%|████      | 2590/6470 [32:06<40:18,  1.60it/s]
3|2591|Loss: 0.8570700883865356:  40%|████      | 2591/6470 [32:06<39:06,  1.65it/s]
3|2592|Loss: 0.8031263947486877:  40%|████      | 2591/6470 [32:07<39:06,  1.65it/s]
3|2592|Loss: 0.8031263947486877:  40%|████      | 2592/6470 [32:07<41:34,  1.55it/s]
3|2593|Loss: 0.7282386422157288:  40%|████      | 2592/6470 [32:08<41:34,  1.55it/s]
3|2593|Loss: 0.7282386422157288:  40%|████      | 2593/6470 [32:08<41:50,  1.54it/s]
3|2594|Loss: 0.7873720526695251:  40%|████      | 2593/6470 [32:08<41:50,  1.54it/s]
3|2594|Loss: 0.7873720526695251:  40%|████      | 2594/6470 [32:08<41:45,  1.55it/s]
3|2595|Loss: 0.9584894776344299:  40%|████      | 2594/6470 [32:09<41:45,  1.55it/s]
3|2595|Loss: 0.9584894776344299:  40%|████      | 2595/6470 [32:09<43:32,  1.48it/s]
3|2596|Loss: 0.6923742294311523:  40%|████      | 2595/6470 [32:10<43:32,  1.48it/s]
3|2596|Loss: 0.6923742294311523:  40%|████      | 2596/6470 [32:10<45:30,  1.42it/s]
3|2597|Loss: 0.8391535878181458:  40%|████      | 2596/6470 [32:10<45:30,  1.42it/s]
3|2597|Loss: 0.8391535878181458:  40%|████      | 2597/6470 [32:11<44:28,  1.45it/s]
3|2598|Loss: 0.7642971277236938:  40%|████      | 2597/6470 [32:11<44:28,  1.45it/s]
3|2598|Loss: 0.7642971277236938:  40%|████      | 2598/6470 [32:11<43:43,  1.48it/s]
3|2599|Loss: 0.7345870137214661:  40%|████      | 2598/6470 [32:12<43:43,  1.48it/s]
3|2599|Loss: 0.7345870137214661:  40%|████      | 2599/6470 [32:12<43:51,  1.47it/s]
3|2600|Loss: 1.0718213319778442:  40%|████      | 2599/6470 [32:12<43:51,  1.47it/s]
3|2600|Loss: 1.0718213319778442:  40%|████      | 2600/6470 [32:13<44:16,  1.46it/s]
3|2601|Loss: 0.7962726354598999:  40%|████      | 2600/6470 [32:13<44:16,  1.46it/s]
3|2601|Loss: 0.7962726354598999:  40%|████      | 2601/6470 [32:13<45:26,  1.42it/s]
3|2602|Loss: 0.7716044187545776:  40%|████      | 2601/6470 [32:14<45:26,  1.42it/s]
3|2602|Loss: 0.7716044187545776:  40%|████      | 2602/6470 [32:14<41:02,  1.57it/s]
3|2603|Loss: 0.8390623927116394:  40%|████      | 2602/6470 [32:14<41:02,  1.57it/s]
3|2603|Loss: 0.8390623927116394:  40%|████      | 2603/6470 [32:15<42:47,  1.51it/s]
3|2604|Loss: 0.9511270523071289:  40%|████      | 2603/6470 [32:15<42:47,  1.51it/s]
3|2604|Loss: 0.9511270523071289:  40%|████      | 2604/6470 [32:15<46:26,  1.39it/s]
3|2605|Loss: 0.85392826795578:  40%|████      | 2604/6470 [32:16<46:26,  1.39it/s]  
3|2605|Loss: 0.85392826795578:  40%|████      | 2605/6470 [32:16<44:22,  1.45it/s]
3|2606|Loss: 0.8940452337265015:  40%|████      | 2605/6470 [32:16<44:22,  1.45it/s]
3|2606|Loss: 0.8940452337265015:  40%|████      | 2606/6470 [32:17<47:07,  1.37it/s]
3|2607|Loss: 0.7428209185600281:  40%|████      | 2606/6470 [32:17<47:07,  1.37it/s]
3|2607|Loss: 0.7428209185600281:  40%|████      | 2607/6470 [32:18<46:42,  1.38it/s]
3|2608|Loss: 0.7352858781814575:  40%|████      | 2607/6470 [32:18<46:42,  1.38it/s]
3|2608|Loss: 0.7352858781814575:  40%|████      | 2608/6470 [32:18<45:52,  1.40it/s]
3|2609|Loss: 0.9546863436698914:  40%|████      | 2608/6470 [32:19<45:52,  1.40it/s]
3|2609|Loss: 0.9546863436698914:  40%|████      | 2609/6470 [32:19<45:29,  1.41it/s]
3|2610|Loss: 0.6760405898094177:  40%|████      | 2609/6470 [32:19<45:29,  1.41it/s]
3|2610|Loss: 0.6760405898094177:  40%|████      | 2610/6470 [32:20<50:58,  1.26it/s]
3|2611|Loss: 0.8315577507019043:  40%|████      | 2610/6470 [32:20<50:58,  1.26it/s]
3|2611|Loss: 0.8315577507019043:  40%|████      | 2611/6470 [32:21<51:52,  1.24it/s]
3|2612|Loss: 0.7907818555831909:  40%|████      | 2611/6470 [32:21<51:52,  1.24it/s]
3|2612|Loss: 0.7907818555831909:  40%|████      | 2612/6470 [32:21<49:01,  1.31it/s]
3|2613|Loss: 0.8301961421966553:  40%|████      | 2612/6470 [32:22<49:01,  1.31it/s]
3|2613|Loss: 0.8301961421966553:  40%|████      | 2613/6470 [32:22<51:09,  1.26it/s]
3|2614|Loss: 0.7041911482810974:  40%|████      | 2613/6470 [32:23<51:09,  1.26it/s]
3|2614|Loss: 0.7041911482810974:  40%|████      | 2614/6470 [32:23<52:00,  1.24it/s]
3|2615|Loss: 0.863679051399231:  40%|████      | 2614/6470 [32:24<52:00,  1.24it/s] 
3|2615|Loss: 0.863679051399231:  40%|████      | 2615/6470 [32:24<51:09,  1.26it/s]
3|2616|Loss: 0.9275847673416138:  40%|████      | 2615/6470 [32:24<51:09,  1.26it/s]
3|2616|Loss: 0.9275847673416138:  40%|████      | 2616/6470 [32:25<51:10,  1.26it/s]
3|2617|Loss: 0.7879769206047058:  40%|████      | 2616/6470 [32:25<51:10,  1.26it/s]
3|2617|Loss: 0.7879769206047058:  40%|████      | 2617/6470 [32:26<51:09,  1.26it/s]
3|2618|Loss: 0.6905475854873657:  40%|████      | 2617/6470 [32:26<51:09,  1.26it/s]
3|2618|Loss: 0.6905475854873657:  40%|████      | 2618/6470 [32:26<47:24,  1.35it/s]
3|2619|Loss: 0.8182521462440491:  40%|████      | 2618/6470 [32:27<47:24,  1.35it/s]
3|2619|Loss: 0.8182521462440491:  40%|████      | 2619/6470 [32:27<45:29,  1.41it/s]
3|2620|Loss: 0.9709832668304443:  40%|████      | 2619/6470 [32:27<45:29,  1.41it/s]
3|2620|Loss: 0.9709832668304443:  40%|████      | 2620/6470 [32:28<46:46,  1.37it/s]
3|2621|Loss: 0.7351024746894836:  40%|████      | 2620/6470 [32:28<46:46,  1.37it/s]
3|2621|Loss: 0.7351024746894836:  41%|████      | 2621/6470 [32:28<46:19,  1.38it/s]
3|2622|Loss: 0.8886010050773621:  41%|████      | 2621/6470 [32:29<46:19,  1.38it/s]
3|2622|Loss: 0.8886010050773621:  41%|████      | 2622/6470 [32:29<46:57,  1.37it/s]
3|2623|Loss: 1.054855227470398:  41%|████      | 2622/6470 [32:29<46:57,  1.37it/s] 
3|2623|Loss: 1.054855227470398:  41%|████      | 2623/6470 [32:30<47:34,  1.35it/s]
3|2624|Loss: 0.75364750623703:  41%|████      | 2623/6470 [32:30<47:34,  1.35it/s] 
3|2624|Loss: 0.75364750623703:  41%|████      | 2624/6470 [32:31<52:24,  1.22it/s]
3|2625|Loss: 1.0188491344451904:  41%|████      | 2624/6470 [32:31<52:24,  1.22it/s]
3|2625|Loss: 1.0188491344451904:  41%|████      | 2625/6470 [32:32<51:09,  1.25it/s]
3|2626|Loss: 0.8445852994918823:  41%|████      | 2625/6470 [32:32<51:09,  1.25it/s]
3|2626|Loss: 0.8445852994918823:  41%|████      | 2626/6470 [32:32<49:21,  1.30it/s]
3|2627|Loss: 0.9552807807922363:  41%|████      | 2626/6470 [32:33<49:21,  1.30it/s]
3|2627|Loss: 0.9552807807922363:  41%|████      | 2627/6470 [32:33<50:20,  1.27it/s]
3|2628|Loss: 0.8358510136604309:  41%|████      | 2627/6470 [32:34<50:20,  1.27it/s]
3|2628|Loss: 0.8358510136604309:  41%|████      | 2628/6470 [32:34<52:07,  1.23it/s]
3|2629|Loss: 0.9440365433692932:  41%|████      | 2628/6470 [32:34<52:07,  1.23it/s]
3|2629|Loss: 0.9440365433692932:  41%|████      | 2629/6470 [32:35<50:02,  1.28it/s]
3|2630|Loss: 0.7770407199859619:  41%|████      | 2629/6470 [32:35<50:02,  1.28it/s]
3|2630|Loss: 0.7770407199859619:  41%|████      | 2630/6470 [32:35<50:00,  1.28it/s]
3|2631|Loss: 0.8191373348236084:  41%|████      | 2630/6470 [32:36<50:00,  1.28it/s]
3|2631|Loss: 0.8191373348236084:  41%|████      | 2631/6470 [32:36<48:04,  1.33it/s]
3|2632|Loss: 0.7595563530921936:  41%|████      | 2631/6470 [32:37<48:04,  1.33it/s]
3|2632|Loss: 0.7595563530921936:  41%|████      | 2632/6470 [32:37<51:02,  1.25it/s]
3|2633|Loss: 0.8106116056442261:  41%|████      | 2632/6470 [32:38<51:02,  1.25it/s]
3|2633|Loss: 0.8106116056442261:  41%|████      | 2633/6470 [32:38<53:23,  1.20it/s]
3|2634|Loss: 0.7288914918899536:  41%|████      | 2633/6470 [32:38<53:23,  1.20it/s]
3|2634|Loss: 0.7288914918899536:  41%|████      | 2634/6470 [32:39<55:17,  1.16it/s]
3|2635|Loss: 0.8119396567344666:  41%|████      | 2634/6470 [32:39<55:17,  1.16it/s]
3|2635|Loss: 0.8119396567344666:  41%|████      | 2635/6470 [32:40<54:00,  1.18it/s]
3|2636|Loss: 0.8296877145767212:  41%|████      | 2635/6470 [32:40<54:00,  1.18it/s]
3|2636|Loss: 0.8296877145767212:  41%|████      | 2636/6470 [32:41<55:08,  1.16it/s]
3|2637|Loss: 0.8105770945549011:  41%|████      | 2636/6470 [32:41<55:08,  1.16it/s]
3|2637|Loss: 0.8105770945549011:  41%|████      | 2637/6470 [32:41<51:50,  1.23it/s]
3|2638|Loss: 0.8091714382171631:  41%|████      | 2637/6470 [32:42<51:50,  1.23it/s]
3|2638|Loss: 0.8091714382171631:  41%|████      | 2638/6470 [32:42<48:54,  1.31it/s]
3|2639|Loss: 0.8982489109039307:  41%|████      | 2638/6470 [32:42<48:54,  1.31it/s]
3|2639|Loss: 0.8982489109039307:  41%|████      | 2639/6470 [32:43<48:59,  1.30it/s]
3|2640|Loss: 0.8958442807197571:  41%|████      | 2639/6470 [32:43<48:59,  1.30it/s]
3|2640|Loss: 0.8958442807197571:  41%|████      | 2640/6470 [32:44<51:01,  1.25it/s]
3|2641|Loss: 0.8274555802345276:  41%|████      | 2640/6470 [32:44<51:01,  1.25it/s]
3|2641|Loss: 0.8274555802345276:  41%|████      | 2641/6470 [32:44<49:38,  1.29it/s]
3|2642|Loss: 0.7310057878494263:  41%|████      | 2641/6470 [32:45<49:38,  1.29it/s]
3|2642|Loss: 0.7310057878494263:  41%|████      | 2642/6470 [32:45<44:46,  1.43it/s]
3|2643|Loss: 0.8386277556419373:  41%|████      | 2642/6470 [32:45<44:46,  1.43it/s]
3|2643|Loss: 0.8386277556419373:  41%|████      | 2643/6470 [32:46<45:20,  1.41it/s]
3|2644|Loss: 0.8347819447517395:  41%|████      | 2643/6470 [32:46<45:20,  1.41it/s]
3|2644|Loss: 0.8347819447517395:  41%|████      | 2644/6470 [32:46<49:13,  1.30it/s]
3|2645|Loss: 0.7970365285873413:  41%|████      | 2644/6470 [32:47<49:13,  1.30it/s]
3|2645|Loss: 0.7970365285873413:  41%|████      | 2645/6470 [32:47<53:36,  1.19it/s]
3|2646|Loss: 0.7644158005714417:  41%|████      | 2645/6470 [32:48<53:36,  1.19it/s]
3|2646|Loss: 0.7644158005714417:  41%|████      | 2646/6470 [32:48<56:08,  1.14it/s]
3|2647|Loss: 0.8599883317947388:  41%|████      | 2646/6470 [32:49<56:08,  1.14it/s]
3|2647|Loss: 0.8599883317947388:  41%|████      | 2647/6470 [32:49<53:06,  1.20it/s]
3|2648|Loss: 0.6886785626411438:  41%|████      | 2647/6470 [32:50<53:06,  1.20it/s]
3|2648|Loss: 0.6886785626411438:  41%|████      | 2648/6470 [32:50<52:25,  1.22it/s]
3|2649|Loss: 0.7878925800323486:  41%|████      | 2648/6470 [32:50<52:25,  1.22it/s]
3|2649|Loss: 0.7878925800323486:  41%|████      | 2649/6470 [32:51<49:01,  1.30it/s]
3|2650|Loss: 0.8582260012626648:  41%|████      | 2649/6470 [32:51<49:01,  1.30it/s]
3|2650|Loss: 0.8582260012626648:  41%|████      | 2650/6470 [32:52<51:29,  1.24it/s]
3|2651|Loss: 0.7945259809494019:  41%|████      | 2650/6470 [32:52<51:29,  1.24it/s]
3|2651|Loss: 0.7945259809494019:  41%|████      | 2651/6470 [32:52<52:13,  1.22it/s]
3|2652|Loss: 0.8248352408409119:  41%|████      | 2651/6470 [32:53<52:13,  1.22it/s]
3|2652|Loss: 0.8248352408409119:  41%|████      | 2652/6470 [32:53<51:12,  1.24it/s]
3|2653|Loss: 0.819708526134491:  41%|████      | 2652/6470 [32:54<51:12,  1.24it/s] 
3|2653|Loss: 0.819708526134491:  41%|████      | 2653/6470 [32:54<47:42,  1.33it/s]
3|2654|Loss: 0.7709704637527466:  41%|████      | 2653/6470 [32:54<47:42,  1.33it/s]
3|2654|Loss: 0.7709704637527466:  41%|████      | 2654/6470 [32:54<45:46,  1.39it/s]
3|2655|Loss: 0.7106215357780457:  41%|████      | 2654/6470 [32:55<45:46,  1.39it/s]
3|2655|Loss: 0.7106215357780457:  41%|████      | 2655/6470 [32:55<44:18,  1.43it/s]
3|2656|Loss: 0.7756459712982178:  41%|████      | 2655/6470 [32:55<44:18,  1.43it/s]
3|2656|Loss: 0.7756459712982178:  41%|████      | 2656/6470 [32:56<45:30,  1.40it/s]
3|2657|Loss: 0.8012270331382751:  41%|████      | 2656/6470 [32:56<45:30,  1.40it/s]
3|2657|Loss: 0.8012270331382751:  41%|████      | 2657/6470 [32:57<49:29,  1.28it/s]
3|2658|Loss: 0.8668049573898315:  41%|████      | 2657/6470 [32:57<49:29,  1.28it/s]
3|2658|Loss: 0.8668049573898315:  41%|████      | 2658/6470 [32:58<52:59,  1.20it/s]
3|2659|Loss: 0.9408842325210571:  41%|████      | 2658/6470 [32:58<52:59,  1.20it/s]
3|2659|Loss: 0.9408842325210571:  41%|████      | 2659/6470 [32:58<49:10,  1.29it/s]
3|2660|Loss: 0.9614625573158264:  41%|████      | 2659/6470 [32:59<49:10,  1.29it/s]
3|2660|Loss: 0.9614625573158264:  41%|████      | 2660/6470 [32:59<46:26,  1.37it/s]
3|2661|Loss: 0.7166950106620789:  41%|████      | 2660/6470 [32:59<46:26,  1.37it/s]
3|2661|Loss: 0.7166950106620789:  41%|████      | 2661/6470 [33:00<45:14,  1.40it/s]
3|2662|Loss: 0.7651939988136292:  41%|████      | 2661/6470 [33:00<45:14,  1.40it/s]
3|2662|Loss: 0.7651939988136292:  41%|████      | 2662/6470 [33:00<47:59,  1.32it/s]
3|2663|Loss: 0.8746901750564575:  41%|████      | 2662/6470 [33:01<47:59,  1.32it/s]
3|2663|Loss: 0.8746901750564575:  41%|████      | 2663/6470 [33:01<44:18,  1.43it/s]
3|2664|Loss: 0.8173754811286926:  41%|████      | 2663/6470 [33:01<44:18,  1.43it/s]
3|2664|Loss: 0.8173754811286926:  41%|████      | 2664/6470 [33:02<44:00,  1.44it/s]
3|2665|Loss: 0.7528907060623169:  41%|████      | 2664/6470 [33:02<44:00,  1.44it/s]
3|2665|Loss: 0.7528907060623169:  41%|████      | 2665/6470 [33:03<48:32,  1.31it/s]
3|2666|Loss: 0.8416699767112732:  41%|████      | 2665/6470 [33:03<48:32,  1.31it/s]
3|2666|Loss: 0.8416699767112732:  41%|████      | 2666/6470 [33:04<52:56,  1.20it/s]
3|2667|Loss: 0.9834244847297668:  41%|████      | 2666/6470 [33:04<52:56,  1.20it/s]
3|2667|Loss: 0.9834244847297668:  41%|████      | 2667/6470 [33:05<55:10,  1.15it/s]
3|2668|Loss: 0.83620285987854:  41%|████      | 2667/6470 [33:05<55:10,  1.15it/s]  
3|2668|Loss: 0.83620285987854:  41%|████      | 2668/6470 [33:06<55:30,  1.14it/s]
3|2669|Loss: 0.7672273516654968:  41%|████      | 2668/6470 [33:06<55:30,  1.14it/s]
3|2669|Loss: 0.7672273516654968:  41%|████▏     | 2669/6470 [33:06<53:39,  1.18it/s]
3|2670|Loss: 0.8224798440933228:  41%|████▏     | 2669/6470 [33:07<53:39,  1.18it/s]
3|2670|Loss: 0.8224798440933228:  41%|████▏     | 2670/6470 [33:07<50:57,  1.24it/s]
3|2671|Loss: 0.9795222878456116:  41%|████▏     | 2670/6470 [33:07<50:57,  1.24it/s]
3|2671|Loss: 0.9795222878456116:  41%|████▏     | 2671/6470 [33:08<49:53,  1.27it/s]
3|2672|Loss: 0.9125314354896545:  41%|████▏     | 2671/6470 [33:08<49:53,  1.27it/s]
3|2672|Loss: 0.9125314354896545:  41%|████▏     | 2672/6470 [33:08<48:12,  1.31it/s]
3|2673|Loss: 0.7862966656684875:  41%|████▏     | 2672/6470 [33:09<48:12,  1.31it/s]
3|2673|Loss: 0.7862966656684875:  41%|████▏     | 2673/6470 [33:09<46:28,  1.36it/s]
3|2674|Loss: 0.8870456218719482:  41%|████▏     | 2673/6470 [33:09<46:28,  1.36it/s]
3|2674|Loss: 0.8870456218719482:  41%|████▏     | 2674/6470 [33:10<45:11,  1.40it/s]
3|2675|Loss: 0.7916293740272522:  41%|████▏     | 2674/6470 [33:10<45:11,  1.40it/s]
3|2675|Loss: 0.7916293740272522:  41%|████▏     | 2675/6470 [33:11<46:41,  1.35it/s]
3|2676|Loss: 0.79886794090271:  41%|████▏     | 2675/6470 [33:11<46:41,  1.35it/s]  
3|2676|Loss: 0.79886794090271:  41%|████▏     | 2676/6470 [33:11<46:32,  1.36it/s]
3|2677|Loss: 0.9683793783187866:  41%|████▏     | 2676/6470 [33:12<46:32,  1.36it/s]
3|2677|Loss: 0.9683793783187866:  41%|████▏     | 2677/6470 [33:12<44:41,  1.41it/s]
3|2678|Loss: 0.8647273778915405:  41%|████▏     | 2677/6470 [33:12<44:41,  1.41it/s]
3|2678|Loss: 0.8647273778915405:  41%|████▏     | 2678/6470 [33:13<43:34,  1.45it/s]
3|2679|Loss: 0.8238003253936768:  41%|████▏     | 2678/6470 [33:13<43:34,  1.45it/s]
3|2679|Loss: 0.8238003253936768:  41%|████▏     | 2679/6470 [33:13<43:39,  1.45it/s]
3|2680|Loss: 0.7368367314338684:  41%|████▏     | 2679/6470 [33:14<43:39,  1.45it/s]
3|2680|Loss: 0.7368367314338684:  41%|████▏     | 2680/6470 [33:14<47:09,  1.34it/s]
3|2681|Loss: 0.846251368522644:  41%|████▏     | 2680/6470 [33:15<47:09,  1.34it/s] 
3|2681|Loss: 0.846251368522644:  41%|████▏     | 2681/6470 [33:15<45:50,  1.38it/s]
3|2682|Loss: 0.8074755668640137:  41%|████▏     | 2681/6470 [33:15<45:50,  1.38it/s]
3|2682|Loss: 0.8074755668640137:  41%|████▏     | 2682/6470 [33:15<43:39,  1.45it/s]
3|2683|Loss: 0.8978252410888672:  41%|████▏     | 2682/6470 [33:16<43:39,  1.45it/s]
3|2683|Loss: 0.8978252410888672:  41%|████▏     | 2683/6470 [33:16<43:27,  1.45it/s]
3|2684|Loss: 0.9925783276557922:  41%|████▏     | 2683/6470 [33:17<43:27,  1.45it/s]
3|2684|Loss: 0.9925783276557922:  41%|████▏     | 2684/6470 [33:17<43:37,  1.45it/s]
3|2685|Loss: 0.7939527630805969:  41%|████▏     | 2684/6470 [33:17<43:37,  1.45it/s]
3|2685|Loss: 0.7939527630805969:  41%|████▏     | 2685/6470 [33:18<47:15,  1.33it/s]
3|2686|Loss: 0.8420150876045227:  41%|████▏     | 2685/6470 [33:18<47:15,  1.33it/s]
3|2686|Loss: 0.8420150876045227:  42%|████▏     | 2686/6470 [33:18<46:00,  1.37it/s]
3|2687|Loss: 0.6895014643669128:  42%|████▏     | 2686/6470 [33:19<46:00,  1.37it/s]
3|2687|Loss: 0.6895014643669128:  42%|████▏     | 2687/6470 [33:19<44:01,  1.43it/s]
3|2688|Loss: 0.671762228012085:  42%|████▏     | 2687/6470 [33:19<44:01,  1.43it/s] 
3|2688|Loss: 0.671762228012085:  42%|████▏     | 2688/6470 [33:20<42:08,  1.50it/s]
3|2689|Loss: 0.9423392415046692:  42%|████▏     | 2688/6470 [33:20<42:08,  1.50it/s]
3|2689|Loss: 0.9423392415046692:  42%|████▏     | 2689/6470 [33:20<41:16,  1.53it/s]
3|2690|Loss: 0.761271595954895:  42%|████▏     | 2689/6470 [33:21<41:16,  1.53it/s] 
3|2690|Loss: 0.761271595954895:  42%|████▏     | 2690/6470 [33:21<45:02,  1.40it/s]
3|2691|Loss: 0.7953281998634338:  42%|████▏     | 2690/6470 [33:22<45:02,  1.40it/s]
3|2691|Loss: 0.7953281998634338:  42%|████▏     | 2691/6470 [33:22<45:08,  1.40it/s]
3|2692|Loss: 0.7420814633369446:  42%|████▏     | 2691/6470 [33:22<45:08,  1.40it/s]
3|2692|Loss: 0.7420814633369446:  42%|████▏     | 2692/6470 [33:22<42:34,  1.48it/s]
3|2693|Loss: 0.6670670509338379:  42%|████▏     | 2692/6470 [33:23<42:34,  1.48it/s]
3|2693|Loss: 0.6670670509338379:  42%|████▏     | 2693/6470 [33:23<42:00,  1.50it/s]
3|2694|Loss: 0.852400004863739:  42%|████▏     | 2693/6470 [33:23<42:00,  1.50it/s] 
3|2694|Loss: 0.852400004863739:  42%|████▏     | 2694/6470 [33:24<44:16,  1.42it/s]
3|2695|Loss: 0.7453492879867554:  42%|████▏     | 2694/6470 [33:24<44:16,  1.42it/s]
3|2695|Loss: 0.7453492879867554:  42%|████▏     | 2695/6470 [33:25<44:21,  1.42it/s]
3|2696|Loss: 0.9348984360694885:  42%|████▏     | 2695/6470 [33:25<44:21,  1.42it/s]
3|2696|Loss: 0.9348984360694885:  42%|████▏     | 2696/6470 [33:25<45:44,  1.38it/s]
3|2697|Loss: 0.6937631964683533:  42%|████▏     | 2696/6470 [33:26<45:44,  1.38it/s]
3|2697|Loss: 0.6937631964683533:  42%|████▏     | 2697/6470 [33:26<46:59,  1.34it/s]
3|2698|Loss: 0.833977222442627:  42%|████▏     | 2697/6470 [33:27<46:59,  1.34it/s] 
3|2698|Loss: 0.833977222442627:  42%|████▏     | 2698/6470 [33:27<48:58,  1.28it/s]
3|2699|Loss: 0.7184594869613647:  42%|████▏     | 2698/6470 [33:27<48:58,  1.28it/s]
3|2699|Loss: 0.7184594869613647:  42%|████▏     | 2699/6470 [33:28<45:45,  1.37it/s]
3|2700|Loss: 0.7234168648719788:  42%|████▏     | 2699/6470 [33:28<45:45,  1.37it/s]
3|2700|Loss: 0.7234168648719788:  42%|████▏     | 2700/6470 [33:28<45:48,  1.37it/s]
3|2701|Loss: 0.8850260376930237:  42%|████▏     | 2700/6470 [33:29<45:48,  1.37it/s]
3|2701|Loss: 0.8850260376930237:  42%|████▏     | 2701/6470 [33:29<43:10,  1.46it/s]
3|2702|Loss: 0.8026522397994995:  42%|████▏     | 2701/6470 [33:29<43:10,  1.46it/s]
3|2702|Loss: 0.8026522397994995:  42%|████▏     | 2702/6470 [33:30<42:05,  1.49it/s]
3|2703|Loss: 0.8665941953659058:  42%|████▏     | 2702/6470 [33:30<42:05,  1.49it/s]
3|2703|Loss: 0.8665941953659058:  42%|████▏     | 2703/6470 [33:30<43:35,  1.44it/s]
3|2704|Loss: 0.8397479057312012:  42%|████▏     | 2703/6470 [33:31<43:35,  1.44it/s]
3|2704|Loss: 0.8397479057312012:  42%|████▏     | 2704/6470 [33:31<44:09,  1.42it/s]
3|2705|Loss: 0.8892737627029419:  42%|████▏     | 2704/6470 [33:31<44:09,  1.42it/s]
3|2705|Loss: 0.8892737627029419:  42%|████▏     | 2705/6470 [33:32<45:57,  1.37it/s]
3|2706|Loss: 0.8965812921524048:  42%|████▏     | 2705/6470 [33:32<45:57,  1.37it/s]
3|2706|Loss: 0.8965812921524048:  42%|████▏     | 2706/6470 [33:33<48:59,  1.28it/s]
3|2707|Loss: 0.8667559623718262:  42%|████▏     | 2706/6470 [33:33<48:59,  1.28it/s]
3|2707|Loss: 0.8667559623718262:  42%|████▏     | 2707/6470 [33:33<47:47,  1.31it/s]
3|2708|Loss: 0.7164433598518372:  42%|████▏     | 2707/6470 [33:34<47:47,  1.31it/s]
3|2708|Loss: 0.7164433598518372:  42%|████▏     | 2708/6470 [33:34<49:24,  1.27it/s]
3|2709|Loss: 0.9245193600654602:  42%|████▏     | 2708/6470 [33:35<49:24,  1.27it/s]
3|2709|Loss: 0.9245193600654602:  42%|████▏     | 2709/6470 [33:35<51:43,  1.21it/s]
3|2710|Loss: 0.7970832586288452:  42%|████▏     | 2709/6470 [33:36<51:43,  1.21it/s]
3|2710|Loss: 0.7970832586288452:  42%|████▏     | 2710/6470 [33:36<48:13,  1.30it/s]
3|2711|Loss: 0.8037241697311401:  42%|████▏     | 2710/6470 [33:36<48:13,  1.30it/s]
3|2711|Loss: 0.8037241697311401:  42%|████▏     | 2711/6470 [33:37<50:09,  1.25it/s]
3|2712|Loss: 0.9103280901908875:  42%|████▏     | 2711/6470 [33:37<50:09,  1.25it/s]
3|2712|Loss: 0.9103280901908875:  42%|████▏     | 2712/6470 [33:38<50:29,  1.24it/s]
3|2713|Loss: 0.8407009243965149:  42%|████▏     | 2712/6470 [33:38<50:29,  1.24it/s]
3|2713|Loss: 0.8407009243965149:  42%|████▏     | 2713/6470 [33:38<52:59,  1.18it/s]
3|2714|Loss: 0.9277852773666382:  42%|████▏     | 2713/6470 [33:39<52:59,  1.18it/s]
3|2714|Loss: 0.9277852773666382:  42%|████▏     | 2714/6470 [33:39<52:14,  1.20it/s]
3|2715|Loss: 0.7990850210189819:  42%|████▏     | 2714/6470 [33:40<52:14,  1.20it/s]
3|2715|Loss: 0.7990850210189819:  42%|████▏     | 2715/6470 [33:40<51:32,  1.21it/s]
3|2716|Loss: 0.6911548972129822:  42%|████▏     | 2715/6470 [33:40<51:32,  1.21it/s]
3|2716|Loss: 0.6911548972129822:  42%|████▏     | 2716/6470 [33:41<50:23,  1.24it/s]
3|2717|Loss: 0.9738386869430542:  42%|████▏     | 2716/6470 [33:41<50:23,  1.24it/s]
3|2717|Loss: 0.9738386869430542:  42%|████▏     | 2717/6470 [33:42<49:46,  1.26it/s]
3|2718|Loss: 0.9472230672836304:  42%|████▏     | 2717/6470 [33:42<49:46,  1.26it/s]
3|2718|Loss: 0.9472230672836304:  42%|████▏     | 2718/6470 [33:42<49:45,  1.26it/s]
3|2719|Loss: 0.8964306116104126:  42%|████▏     | 2718/6470 [33:43<49:45,  1.26it/s]
3|2719|Loss: 0.8964306116104126:  42%|████▏     | 2719/6470 [33:43<50:18,  1.24it/s]
3|2720|Loss: 0.7247060537338257:  42%|████▏     | 2719/6470 [33:44<50:18,  1.24it/s]
3|2720|Loss: 0.7247060537338257:  42%|████▏     | 2720/6470 [33:44<51:24,  1.22it/s]
3|2721|Loss: 0.8874840140342712:  42%|████▏     | 2720/6470 [33:45<51:24,  1.22it/s]
3|2721|Loss: 0.8874840140342712:  42%|████▏     | 2721/6470 [33:45<51:58,  1.20it/s]
3|2722|Loss: 0.8687765002250671:  42%|████▏     | 2721/6470 [33:45<51:58,  1.20it/s]
3|2722|Loss: 0.8687765002250671:  42%|████▏     | 2722/6470 [33:46<50:00,  1.25it/s]
3|2723|Loss: 0.804894208908081:  42%|████▏     | 2722/6470 [33:46<50:00,  1.25it/s] 
3|2723|Loss: 0.804894208908081:  42%|████▏     | 2723/6470 [33:46<47:06,  1.33it/s]
3|2724|Loss: 0.8926149606704712:  42%|████▏     | 2723/6470 [33:47<47:06,  1.33it/s]
3|2724|Loss: 0.8926149606704712:  42%|████▏     | 2724/6470 [33:47<44:42,  1.40it/s]
3|2725|Loss: 0.8186166286468506:  42%|████▏     | 2724/6470 [33:47<44:42,  1.40it/s]
3|2725|Loss: 0.8186166286468506:  42%|████▏     | 2725/6470 [33:48<44:55,  1.39it/s]
3|2726|Loss: 0.8265095353126526:  42%|████▏     | 2725/6470 [33:48<44:55,  1.39it/s]
3|2726|Loss: 0.8265095353126526:  42%|████▏     | 2726/6470 [33:48<46:04,  1.35it/s]
3|2727|Loss: 0.7690609693527222:  42%|████▏     | 2726/6470 [33:49<46:04,  1.35it/s]
3|2727|Loss: 0.7690609693527222:  42%|████▏     | 2727/6470 [33:49<48:14,  1.29it/s]
3|2728|Loss: 0.7862115502357483:  42%|████▏     | 2727/6470 [33:50<48:14,  1.29it/s]
3|2728|Loss: 0.7862115502357483:  42%|████▏     | 2728/6470 [33:50<45:04,  1.38it/s]
3|2729|Loss: 0.9385223388671875:  42%|████▏     | 2728/6470 [33:50<45:04,  1.38it/s]
3|2729|Loss: 0.9385223388671875:  42%|████▏     | 2729/6470 [33:51<47:46,  1.31it/s]
3|2730|Loss: 0.7847077250480652:  42%|████▏     | 2729/6470 [33:51<47:46,  1.31it/s]
3|2730|Loss: 0.7847077250480652:  42%|████▏     | 2730/6470 [33:52<48:23,  1.29it/s]
3|2731|Loss: 0.890764594078064:  42%|████▏     | 2730/6470 [33:52<48:23,  1.29it/s] 
3|2731|Loss: 0.890764594078064:  42%|████▏     | 2731/6470 [33:52<47:25,  1.31it/s]
3|2732|Loss: 0.880207359790802:  42%|████▏     | 2731/6470 [33:53<47:25,  1.31it/s]
3|2732|Loss: 0.880207359790802:  42%|████▏     | 2732/6470 [33:53<47:15,  1.32it/s]
3|2733|Loss: 0.8549099564552307:  42%|████▏     | 2732/6470 [33:53<47:15,  1.32it/s]
3|2733|Loss: 0.8549099564552307:  42%|████▏     | 2733/6470 [33:54<47:45,  1.30it/s]
3|2734|Loss: 0.905381441116333:  42%|████▏     | 2733/6470 [33:54<47:45,  1.30it/s] 
3|2734|Loss: 0.905381441116333:  42%|████▏     | 2734/6470 [33:55<45:51,  1.36it/s]
3|2735|Loss: 0.8876757025718689:  42%|████▏     | 2734/6470 [33:55<45:51,  1.36it/s]
3|2735|Loss: 0.8876757025718689:  42%|████▏     | 2735/6470 [33:55<42:24,  1.47it/s]
3|2736|Loss: 0.8661915063858032:  42%|████▏     | 2735/6470 [33:55<42:24,  1.47it/s]
3|2736|Loss: 0.8661915063858032:  42%|████▏     | 2736/6470 [33:56<42:11,  1.47it/s]
3|2737|Loss: 0.8760588765144348:  42%|████▏     | 2736/6470 [33:56<42:11,  1.47it/s]
3|2737|Loss: 0.8760588765144348:  42%|████▏     | 2737/6470 [33:56<43:35,  1.43it/s]
3|2738|Loss: 0.686970591545105:  42%|████▏     | 2737/6470 [33:57<43:35,  1.43it/s] 
3|2738|Loss: 0.686970591545105:  42%|████▏     | 2738/6470 [33:57<47:24,  1.31it/s]
3|2739|Loss: 0.821950376033783:  42%|████▏     | 2738/6470 [33:58<47:24,  1.31it/s]
3|2739|Loss: 0.821950376033783:  42%|████▏     | 2739/6470 [33:58<47:01,  1.32it/s]
3|2740|Loss: 0.8906054496765137:  42%|████▏     | 2739/6470 [33:59<47:01,  1.32it/s]
3|2740|Loss: 0.8906054496765137:  42%|████▏     | 2740/6470 [33:59<48:13,  1.29it/s]
3|2741|Loss: 0.7354965806007385:  42%|████▏     | 2740/6470 [33:59<48:13,  1.29it/s]
3|2741|Loss: 0.7354965806007385:  42%|████▏     | 2741/6470 [34:00<50:22,  1.23it/s]
3|2742|Loss: 0.7239678502082825:  42%|████▏     | 2741/6470 [34:00<50:22,  1.23it/s]
3|2742|Loss: 0.7239678502082825:  42%|████▏     | 2742/6470 [34:01<49:01,  1.27it/s]
3|2743|Loss: 0.9066219925880432:  42%|████▏     | 2742/6470 [34:01<49:01,  1.27it/s]
3|2743|Loss: 0.9066219925880432:  42%|████▏     | 2743/6470 [34:01<48:14,  1.29it/s]
3|2744|Loss: 0.9221833348274231:  42%|████▏     | 2743/6470 [34:02<48:14,  1.29it/s]
3|2744|Loss: 0.9221833348274231:  42%|████▏     | 2744/6470 [34:02<47:57,  1.29it/s]
3|2745|Loss: 0.6979793310165405:  42%|████▏     | 2744/6470 [34:03<47:57,  1.29it/s]
3|2745|Loss: 0.6979793310165405:  42%|████▏     | 2745/6470 [34:03<48:26,  1.28it/s]
3|2746|Loss: 0.8896376490592957:  42%|████▏     | 2745/6470 [34:03<48:26,  1.28it/s]
3|2746|Loss: 0.8896376490592957:  42%|████▏     | 2746/6470 [34:04<50:42,  1.22it/s]
3|2747|Loss: 1.0121097564697266:  42%|████▏     | 2746/6470 [34:04<50:42,  1.22it/s]
3|2747|Loss: 1.0121097564697266:  42%|████▏     | 2747/6470 [34:04<48:22,  1.28it/s]
3|2748|Loss: 0.9639642834663391:  42%|████▏     | 2747/6470 [34:05<48:22,  1.28it/s]
3|2748|Loss: 0.9639642834663391:  42%|████▏     | 2748/6470 [34:05<45:21,  1.37it/s]
3|2749|Loss: 0.9635409712791443:  42%|████▏     | 2748/6470 [34:06<45:21,  1.37it/s]
3|2749|Loss: 0.9635409712791443:  42%|████▏     | 2749/6470 [34:06<47:50,  1.30it/s]
3|2750|Loss: 0.7335511445999146:  42%|████▏     | 2749/6470 [34:06<47:50,  1.30it/s]
3|2750|Loss: 0.7335511445999146:  43%|████▎     | 2750/6470 [34:07<48:03,  1.29it/s]
3|2751|Loss: 0.8390657305717468:  43%|████▎     | 2750/6470 [34:07<48:03,  1.29it/s]
3|2751|Loss: 0.8390657305717468:  43%|████▎     | 2751/6470 [34:08<49:41,  1.25it/s]
3|2752|Loss: 0.9533364176750183:  43%|████▎     | 2751/6470 [34:08<49:41,  1.25it/s]
3|2752|Loss: 0.9533364176750183:  43%|████▎     | 2752/6470 [34:08<48:05,  1.29it/s]
3|2753|Loss: 0.7228701710700989:  43%|████▎     | 2752/6470 [34:09<48:05,  1.29it/s]
3|2753|Loss: 0.7228701710700989:  43%|████▎     | 2753/6470 [34:09<51:51,  1.19it/s]
3|2754|Loss: 0.7077311873435974:  43%|████▎     | 2753/6470 [34:10<51:51,  1.19it/s]
3|2754|Loss: 0.7077311873435974:  43%|████▎     | 2754/6470 [34:10<51:56,  1.19it/s]
3|2755|Loss: 0.7470870614051819:  43%|████▎     | 2754/6470 [34:11<51:56,  1.19it/s]
3|2755|Loss: 0.7470870614051819:  43%|████▎     | 2755/6470 [34:11<51:20,  1.21it/s]
3|2756|Loss: 0.8473401069641113:  43%|████▎     | 2755/6470 [34:11<51:20,  1.21it/s]
3|2756|Loss: 0.8473401069641113:  43%|████▎     | 2756/6470 [34:12<50:23,  1.23it/s]
3|2757|Loss: 0.8724923133850098:  43%|████▎     | 2756/6470 [34:12<50:23,  1.23it/s]
3|2757|Loss: 0.8724923133850098:  43%|████▎     | 2757/6470 [34:13<49:34,  1.25it/s]
3|2758|Loss: 0.6820343732833862:  43%|████▎     | 2757/6470 [34:13<49:34,  1.25it/s]
3|2758|Loss: 0.6820343732833862:  43%|████▎     | 2758/6470 [34:13<46:26,  1.33it/s]
3|2759|Loss: 0.9675487875938416:  43%|████▎     | 2758/6470 [34:14<46:26,  1.33it/s]
3|2759|Loss: 0.9675487875938416:  43%|████▎     | 2759/6470 [34:14<46:22,  1.33it/s]
3|2760|Loss: 0.8484863638877869:  43%|████▎     | 2759/6470 [34:14<46:22,  1.33it/s]
3|2760|Loss: 0.8484863638877869:  43%|████▎     | 2760/6470 [34:15<47:34,  1.30it/s]
3|2761|Loss: 0.8636144399642944:  43%|████▎     | 2760/6470 [34:15<47:34,  1.30it/s]
3|2761|Loss: 0.8636144399642944:  43%|████▎     | 2761/6470 [34:15<47:37,  1.30it/s]
3|2762|Loss: 0.8853414058685303:  43%|████▎     | 2761/6470 [34:16<47:37,  1.30it/s]
3|2762|Loss: 0.8853414058685303:  43%|████▎     | 2762/6470 [34:16<47:11,  1.31it/s]
3|2763|Loss: 0.7786544561386108:  43%|████▎     | 2762/6470 [34:17<47:11,  1.31it/s]
3|2763|Loss: 0.7786544561386108:  43%|████▎     | 2763/6470 [34:17<48:37,  1.27it/s]
3|2764|Loss: 0.7639963626861572:  43%|████▎     | 2763/6470 [34:17<48:37,  1.27it/s]
3|2764|Loss: 0.7639963626861572:  43%|████▎     | 2764/6470 [34:18<46:24,  1.33it/s]
3|2765|Loss: 0.8713853359222412:  43%|████▎     | 2764/6470 [34:18<46:24,  1.33it/s]
3|2765|Loss: 0.8713853359222412:  43%|████▎     | 2765/6470 [34:18<42:37,  1.45it/s]
3|2766|Loss: 0.7893946766853333:  43%|████▎     | 2765/6470 [34:19<42:37,  1.45it/s]
3|2766|Loss: 0.7893946766853333:  43%|████▎     | 2766/6470 [34:19<41:30,  1.49it/s]
3|2767|Loss: 0.9610790014266968:  43%|████▎     | 2766/6470 [34:19<41:30,  1.49it/s]
3|2767|Loss: 0.9610790014266968:  43%|████▎     | 2767/6470 [34:20<42:47,  1.44it/s]
3|2768|Loss: 0.8751136660575867:  43%|████▎     | 2767/6470 [34:20<42:47,  1.44it/s]
3|2768|Loss: 0.8751136660575867:  43%|████▎     | 2768/6470 [34:21<46:10,  1.34it/s]
3|2769|Loss: 0.9619370102882385:  43%|████▎     | 2768/6470 [34:21<46:10,  1.34it/s]
3|2769|Loss: 0.9619370102882385:  43%|████▎     | 2769/6470 [34:21<48:42,  1.27it/s]
3|2770|Loss: 0.9449043869972229:  43%|████▎     | 2769/6470 [34:22<48:42,  1.27it/s]
3|2770|Loss: 0.9449043869972229:  43%|████▎     | 2770/6470 [34:22<45:18,  1.36it/s]
3|2771|Loss: 0.8352271318435669:  43%|████▎     | 2770/6470 [34:22<45:18,  1.36it/s]
3|2771|Loss: 0.8352271318435669:  43%|████▎     | 2771/6470 [34:23<45:30,  1.35it/s]
3|2772|Loss: 0.5911459922790527:  43%|████▎     | 2771/6470 [34:23<45:30,  1.35it/s]
3|2772|Loss: 0.5911459922790527:  43%|████▎     | 2772/6470 [34:23<44:19,  1.39it/s]
3|2773|Loss: 0.794560432434082:  43%|████▎     | 2772/6470 [34:24<44:19,  1.39it/s] 
3|2773|Loss: 0.794560432434082:  43%|████▎     | 2773/6470 [34:24<43:42,  1.41it/s]
3|2774|Loss: 0.895821213722229:  43%|████▎     | 2773/6470 [34:25<43:42,  1.41it/s]
3|2774|Loss: 0.895821213722229:  43%|████▎     | 2774/6470 [34:25<44:51,  1.37it/s]
3|2775|Loss: 0.8781428337097168:  43%|████▎     | 2774/6470 [34:25<44:51,  1.37it/s]
3|2775|Loss: 0.8781428337097168:  43%|████▎     | 2775/6470 [34:26<46:10,  1.33it/s]
3|2776|Loss: 0.7991529703140259:  43%|████▎     | 2775/6470 [34:26<46:10,  1.33it/s]
3|2776|Loss: 0.7991529703140259:  43%|████▎     | 2776/6470 [34:26<45:08,  1.36it/s]
3|2777|Loss: 0.7786791920661926:  43%|████▎     | 2776/6470 [34:27<45:08,  1.36it/s]
3|2777|Loss: 0.7786791920661926:  43%|████▎     | 2777/6470 [34:27<43:52,  1.40it/s]
3|2778|Loss: 0.7990098595619202:  43%|████▎     | 2777/6470 [34:28<43:52,  1.40it/s]
3|2778|Loss: 0.7990098595619202:  43%|████▎     | 2778/6470 [34:28<46:46,  1.32it/s]
3|2779|Loss: 0.8367120623588562:  43%|████▎     | 2778/6470 [34:28<46:46,  1.32it/s]
3|2779|Loss: 0.8367120623588562:  43%|████▎     | 2779/6470 [34:29<48:01,  1.28it/s]
3|2780|Loss: 0.8779386878013611:  43%|████▎     | 2779/6470 [34:29<48:01,  1.28it/s]
3|2780|Loss: 0.8779386878013611:  43%|████▎     | 2780/6470 [34:30<47:58,  1.28it/s]
3|2781|Loss: 0.9186875820159912:  43%|████▎     | 2780/6470 [34:30<47:58,  1.28it/s]
3|2781|Loss: 0.9186875820159912:  43%|████▎     | 2781/6470 [34:30<45:30,  1.35it/s]
3|2782|Loss: 0.7916826605796814:  43%|████▎     | 2781/6470 [34:31<45:30,  1.35it/s]
3|2782|Loss: 0.7916826605796814:  43%|████▎     | 2782/6470 [34:31<41:09,  1.49it/s]
3|2783|Loss: 0.7391040921211243:  43%|████▎     | 2782/6470 [34:31<41:09,  1.49it/s]
3|2783|Loss: 0.7391040921211243:  43%|████▎     | 2783/6470 [34:31<40:28,  1.52it/s]
3|2784|Loss: 0.8118893504142761:  43%|████▎     | 2783/6470 [34:32<40:28,  1.52it/s]
3|2784|Loss: 0.8118893504142761:  43%|████▎     | 2784/6470 [34:32<44:55,  1.37it/s]
3|2785|Loss: 0.8716716766357422:  43%|████▎     | 2784/6470 [34:33<44:55,  1.37it/s]
3|2785|Loss: 0.8716716766357422:  43%|████▎     | 2785/6470 [34:33<40:58,  1.50it/s]
3|2786|Loss: 0.832035481929779:  43%|████▎     | 2785/6470 [34:33<40:58,  1.50it/s] 
3|2786|Loss: 0.832035481929779:  43%|████▎     | 2786/6470 [34:34<42:40,  1.44it/s]
3|2787|Loss: 0.7966940999031067:  43%|████▎     | 2786/6470 [34:34<42:40,  1.44it/s]
3|2787|Loss: 0.7966940999031067:  43%|████▎     | 2787/6470 [34:34<44:23,  1.38it/s]
3|2788|Loss: 0.7174893021583557:  43%|████▎     | 2787/6470 [34:35<44:23,  1.38it/s]
3|2788|Loss: 0.7174893021583557:  43%|████▎     | 2788/6470 [34:35<45:39,  1.34it/s]
3|2789|Loss: 0.8627708554267883:  43%|████▎     | 2788/6470 [34:35<45:39,  1.34it/s]
3|2789|Loss: 0.8627708554267883:  43%|████▎     | 2789/6470 [34:36<42:39,  1.44it/s]
3|2790|Loss: 0.7687878608703613:  43%|████▎     | 2789/6470 [34:36<42:39,  1.44it/s]
3|2790|Loss: 0.7687878608703613:  43%|████▎     | 2790/6470 [34:37<45:35,  1.35it/s]
3|2791|Loss: 0.839678943157196:  43%|████▎     | 2790/6470 [34:37<45:35,  1.35it/s] 
3|2791|Loss: 0.839678943157196:  43%|████▎     | 2791/6470 [34:37<47:33,  1.29it/s]
3|2792|Loss: 0.8706765174865723:  43%|████▎     | 2791/6470 [34:38<47:33,  1.29it/s]
3|2792|Loss: 0.8706765174865723:  43%|████▎     | 2792/6470 [34:38<46:38,  1.31it/s]
3|2793|Loss: 0.8115365505218506:  43%|████▎     | 2792/6470 [34:38<46:38,  1.31it/s]
3|2793|Loss: 0.8115365505218506:  43%|████▎     | 2793/6470 [34:39<43:24,  1.41it/s]
3|2794|Loss: 0.8337494730949402:  43%|████▎     | 2793/6470 [34:39<43:24,  1.41it/s]
3|2794|Loss: 0.8337494730949402:  43%|████▎     | 2794/6470 [34:39<44:15,  1.38it/s]
3|2795|Loss: 0.9487723708152771:  43%|████▎     | 2794/6470 [34:40<44:15,  1.38it/s]
3|2795|Loss: 0.9487723708152771:  43%|████▎     | 2795/6470 [34:40<47:45,  1.28it/s]
3|2796|Loss: 0.8499606251716614:  43%|████▎     | 2795/6470 [34:41<47:45,  1.28it/s]
3|2796|Loss: 0.8499606251716614:  43%|████▎     | 2796/6470 [34:41<50:34,  1.21it/s]
3|2797|Loss: 0.9027471542358398:  43%|████▎     | 2796/6470 [34:42<50:34,  1.21it/s]
3|2797|Loss: 0.9027471542358398:  43%|████▎     | 2797/6470 [34:42<44:23,  1.38it/s]
3|2798|Loss: 0.770822286605835:  43%|████▎     | 2797/6470 [34:42<44:23,  1.38it/s] 
3|2798|Loss: 0.770822286605835:  43%|████▎     | 2798/6470 [34:43<45:05,  1.36it/s]
3|2799|Loss: 0.8347758650779724:  43%|████▎     | 2798/6470 [34:43<45:05,  1.36it/s]
3|2799|Loss: 0.8347758650779724:  43%|████▎     | 2799/6470 [34:43<45:15,  1.35it/s]
3|2800|Loss: 0.7367780208587646:  43%|████▎     | 2799/6470 [34:44<45:15,  1.35it/s]
3|2800|Loss: 0.7367780208587646:  43%|████▎     | 2800/6470 [34:44<44:44,  1.37it/s]
3|2801|Loss: 0.8814211487770081:  43%|████▎     | 2800/6470 [34:44<44:44,  1.37it/s]
3|2801|Loss: 0.8814211487770081:  43%|████▎     | 2801/6470 [34:45<45:52,  1.33it/s]
3|2802|Loss: 0.772806704044342:  43%|████▎     | 2801/6470 [34:45<45:52,  1.33it/s] 
3|2802|Loss: 0.772806704044342:  43%|████▎     | 2802/6470 [34:46<47:02,  1.30it/s]
3|2803|Loss: 0.9601691961288452:  43%|████▎     | 2802/6470 [34:46<47:02,  1.30it/s]
3|2803|Loss: 0.9601691961288452:  43%|████▎     | 2803/6470 [34:46<44:35,  1.37it/s]
3|2804|Loss: 0.7565191984176636:  43%|████▎     | 2803/6470 [34:47<44:35,  1.37it/s]
3|2804|Loss: 0.7565191984176636:  43%|████▎     | 2804/6470 [34:47<48:48,  1.25it/s]
3|2805|Loss: 0.8710421323776245:  43%|████▎     | 2804/6470 [34:48<48:48,  1.25it/s]
3|2805|Loss: 0.8710421323776245:  43%|████▎     | 2805/6470 [34:48<52:18,  1.17it/s]
3|2806|Loss: 0.8997641801834106:  43%|████▎     | 2805/6470 [34:49<52:18,  1.17it/s]
3|2806|Loss: 0.8997641801834106:  43%|████▎     | 2806/6470 [34:49<48:49,  1.25it/s]
3|2807|Loss: 0.8210837841033936:  43%|████▎     | 2806/6470 [34:49<48:49,  1.25it/s]
3|2807|Loss: 0.8210837841033936:  43%|████▎     | 2807/6470 [34:50<47:06,  1.30it/s]
3|2808|Loss: 0.7883948683738708:  43%|████▎     | 2807/6470 [34:50<47:06,  1.30it/s]
3|2808|Loss: 0.7883948683738708:  43%|████▎     | 2808/6470 [34:50<48:19,  1.26it/s]
3|2809|Loss: 1.074650764465332:  43%|████▎     | 2808/6470 [34:51<48:19,  1.26it/s] 
3|2809|Loss: 1.074650764465332:  43%|████▎     | 2809/6470 [34:51<50:35,  1.21it/s]
3|2810|Loss: 0.8457856178283691:  43%|████▎     | 2809/6470 [34:52<50:35,  1.21it/s]
3|2810|Loss: 0.8457856178283691:  43%|████▎     | 2810/6470 [34:52<45:56,  1.33it/s]
3|2811|Loss: 0.8070551156997681:  43%|████▎     | 2810/6470 [34:52<45:56,  1.33it/s]
3|2811|Loss: 0.8070551156997681:  43%|████▎     | 2811/6470 [34:52<41:16,  1.48it/s]
3|2812|Loss: 0.8633443117141724:  43%|████▎     | 2811/6470 [34:53<41:16,  1.48it/s]
3|2812|Loss: 0.8633443117141724:  43%|████▎     | 2812/6470 [34:53<42:27,  1.44it/s]
3|2813|Loss: 0.8992654085159302:  43%|████▎     | 2812/6470 [34:54<42:27,  1.44it/s]
3|2813|Loss: 0.8992654085159302:  43%|████▎     | 2813/6470 [34:54<42:31,  1.43it/s]
3|2814|Loss: 0.812857985496521:  43%|████▎     | 2813/6470 [34:54<42:31,  1.43it/s] 
3|2814|Loss: 0.812857985496521:  43%|████▎     | 2814/6470 [34:55<45:47,  1.33it/s]
3|2815|Loss: 0.7007113695144653:  43%|████▎     | 2814/6470 [34:55<45:47,  1.33it/s]
3|2815|Loss: 0.7007113695144653:  44%|████▎     | 2815/6470 [34:55<42:29,  1.43it/s]
3|2816|Loss: 0.8557738661766052:  44%|████▎     | 2815/6470 [34:56<42:29,  1.43it/s]
3|2816|Loss: 0.8557738661766052:  44%|████▎     | 2816/6470 [34:56<42:17,  1.44it/s]
3|2817|Loss: 0.6703155040740967:  44%|████▎     | 2816/6470 [34:56<42:17,  1.44it/s]
3|2817|Loss: 0.6703155040740967:  44%|████▎     | 2817/6470 [34:57<41:14,  1.48it/s]
3|2818|Loss: 0.9814633727073669:  44%|████▎     | 2817/6470 [34:57<41:14,  1.48it/s]
3|2818|Loss: 0.9814633727073669:  44%|████▎     | 2818/6470 [34:57<38:21,  1.59it/s]
3|2819|Loss: 0.8960268497467041:  44%|████▎     | 2818/6470 [34:58<38:21,  1.59it/s]
3|2819|Loss: 0.8960268497467041:  44%|████▎     | 2819/6470 [34:58<40:09,  1.52it/s]
3|2820|Loss: 0.8104137182235718:  44%|████▎     | 2819/6470 [34:58<40:09,  1.52it/s]
3|2820|Loss: 0.8104137182235718:  44%|████▎     | 2820/6470 [34:59<41:14,  1.48it/s]
3|2821|Loss: 0.8230841755867004:  44%|████▎     | 2820/6470 [34:59<41:14,  1.48it/s]
3|2821|Loss: 0.8230841755867004:  44%|████▎     | 2821/6470 [34:59<36:54,  1.65it/s]
3|2822|Loss: 0.7604116201400757:  44%|████▎     | 2821/6470 [34:59<36:54,  1.65it/s]
3|2822|Loss: 0.7604116201400757:  44%|████▎     | 2822/6470 [35:00<36:52,  1.65it/s]
3|2823|Loss: 0.7996799945831299:  44%|████▎     | 2822/6470 [35:00<36:52,  1.65it/s]
3|2823|Loss: 0.7996799945831299:  44%|████▎     | 2823/6470 [35:00<38:09,  1.59it/s]
3|2824|Loss: 0.9582768678665161:  44%|████▎     | 2823/6470 [35:01<38:09,  1.59it/s]
3|2824|Loss: 0.9582768678665161:  44%|████▎     | 2824/6470 [35:01<41:49,  1.45it/s]
3|2825|Loss: 0.9340371489524841:  44%|████▎     | 2824/6470 [35:02<41:49,  1.45it/s]
3|2825|Loss: 0.9340371489524841:  44%|████▎     | 2825/6470 [35:02<42:27,  1.43it/s]
3|2826|Loss: 0.7760599851608276:  44%|████▎     | 2825/6470 [35:02<42:27,  1.43it/s]
3|2826|Loss: 0.7760599851608276:  44%|████▎     | 2826/6470 [35:03<43:05,  1.41it/s]
3|2827|Loss: 0.8808180093765259:  44%|████▎     | 2826/6470 [35:03<43:05,  1.41it/s]
3|2827|Loss: 0.8808180093765259:  44%|████▎     | 2827/6470 [35:03<46:24,  1.31it/s]
3|2828|Loss: 0.9369233250617981:  44%|████▎     | 2827/6470 [35:04<46:24,  1.31it/s]
3|2828|Loss: 0.9369233250617981:  44%|████▎     | 2828/6470 [35:04<48:42,  1.25it/s]
3|2829|Loss: 0.6727679967880249:  44%|████▎     | 2828/6470 [35:05<48:42,  1.25it/s]
3|2829|Loss: 0.6727679967880249:  44%|████▎     | 2829/6470 [35:05<49:41,  1.22it/s]
3|2830|Loss: 0.8253543376922607:  44%|████▎     | 2829/6470 [35:06<49:41,  1.22it/s]
3|2830|Loss: 0.8253543376922607:  44%|████▎     | 2830/6470 [35:06<47:36,  1.27it/s]
3|2831|Loss: 0.7679910659790039:  44%|████▎     | 2830/6470 [35:06<47:36,  1.27it/s]
3|2831|Loss: 0.7679910659790039:  44%|████▍     | 2831/6470 [35:07<47:20,  1.28it/s]
3|2832|Loss: 0.8338185548782349:  44%|████▍     | 2831/6470 [35:07<47:20,  1.28it/s]
3|2832|Loss: 0.8338185548782349:  44%|████▍     | 2832/6470 [35:07<46:57,  1.29it/s]
3|2833|Loss: 0.7189350724220276:  44%|████▍     | 2832/6470 [35:08<46:57,  1.29it/s]
3|2833|Loss: 0.7189350724220276:  44%|████▍     | 2833/6470 [35:08<44:38,  1.36it/s]
3|2834|Loss: 0.7678582668304443:  44%|████▍     | 2833/6470 [35:08<44:38,  1.36it/s]
3|2834|Loss: 0.7678582668304443:  44%|████▍     | 2834/6470 [35:09<43:34,  1.39it/s]
3|2835|Loss: 0.7502440810203552:  44%|████▍     | 2834/6470 [35:09<43:34,  1.39it/s]
3|2835|Loss: 0.7502440810203552:  44%|████▍     | 2835/6470 [35:10<44:50,  1.35it/s]
3|2836|Loss: 0.8883707523345947:  44%|████▍     | 2835/6470 [35:10<44:50,  1.35it/s]
3|2836|Loss: 0.8883707523345947:  44%|████▍     | 2836/6470 [35:10<43:03,  1.41it/s]
3|2837|Loss: 0.7655768990516663:  44%|████▍     | 2836/6470 [35:11<43:03,  1.41it/s]
3|2837|Loss: 0.7655768990516663:  44%|████▍     | 2837/6470 [35:11<41:47,  1.45it/s]
3|2838|Loss: 0.7659045457839966:  44%|████▍     | 2837/6470 [35:11<41:47,  1.45it/s]
3|2838|Loss: 0.7659045457839966:  44%|████▍     | 2838/6470 [35:12<42:49,  1.41it/s]
3|2839|Loss: 0.7480113506317139:  44%|████▍     | 2838/6470 [35:12<42:49,  1.41it/s]
3|2839|Loss: 0.7480113506317139:  44%|████▍     | 2839/6470 [35:13<46:19,  1.31it/s]
3|2840|Loss: 0.8301251530647278:  44%|████▍     | 2839/6470 [35:13<46:19,  1.31it/s]
3|2840|Loss: 0.8301251530647278:  44%|████▍     | 2840/6470 [35:13<46:01,  1.31it/s]
3|2841|Loss: 0.8468615412712097:  44%|████▍     | 2840/6470 [35:14<46:01,  1.31it/s]
3|2841|Loss: 0.8468615412712097:  44%|████▍     | 2841/6470 [35:14<47:34,  1.27it/s]
3|2842|Loss: 0.7701379656791687:  44%|████▍     | 2841/6470 [35:14<47:34,  1.27it/s]
3|2842|Loss: 0.7701379656791687:  44%|████▍     | 2842/6470 [35:15<43:30,  1.39it/s]
3|2843|Loss: 0.8316730260848999:  44%|████▍     | 2842/6470 [35:15<43:30,  1.39it/s]
3|2843|Loss: 0.8316730260848999:  44%|████▍     | 2843/6470 [35:15<40:13,  1.50it/s]
3|2844|Loss: 0.8230367302894592:  44%|████▍     | 2843/6470 [35:16<40:13,  1.50it/s]
3|2844|Loss: 0.8230367302894592:  44%|████▍     | 2844/6470 [35:16<39:46,  1.52it/s]
3|2845|Loss: 0.8442429304122925:  44%|████▍     | 2844/6470 [35:16<39:46,  1.52it/s]
3|2845|Loss: 0.8442429304122925:  44%|████▍     | 2845/6470 [35:17<39:18,  1.54it/s]
3|2846|Loss: 0.8453395366668701:  44%|████▍     | 2845/6470 [35:17<39:18,  1.54it/s]
3|2846|Loss: 0.8453395366668701:  44%|████▍     | 2846/6470 [35:17<39:12,  1.54it/s]
3|2847|Loss: 0.6859755516052246:  44%|████▍     | 2846/6470 [35:17<39:12,  1.54it/s]
3|2847|Loss: 0.6859755516052246:  44%|████▍     | 2847/6470 [35:18<37:33,  1.61it/s]
3|2848|Loss: 0.9296735525131226:  44%|████▍     | 2847/6470 [35:18<37:33,  1.61it/s]
3|2848|Loss: 0.9296735525131226:  44%|████▍     | 2848/6470 [35:18<38:50,  1.55it/s]
3|2849|Loss: 0.8256548643112183:  44%|████▍     | 2848/6470 [35:19<38:50,  1.55it/s]
3|2849|Loss: 0.8256548643112183:  44%|████▍     | 2849/6470 [35:19<43:34,  1.39it/s]
3|2850|Loss: 0.9505950212478638:  44%|████▍     | 2849/6470 [35:20<43:34,  1.39it/s]
3|2850|Loss: 0.9505950212478638:  44%|████▍     | 2850/6470 [35:20<43:45,  1.38it/s]
3|2851|Loss: 0.7495800256729126:  44%|████▍     | 2850/6470 [35:20<43:45,  1.38it/s]
3|2851|Loss: 0.7495800256729126:  44%|████▍     | 2851/6470 [35:21<42:29,  1.42it/s]
3|2852|Loss: 0.8008327484130859:  44%|████▍     | 2851/6470 [35:21<42:29,  1.42it/s]
3|2852|Loss: 0.8008327484130859:  44%|████▍     | 2852/6470 [35:21<41:10,  1.46it/s]
3|2853|Loss: 0.9457393884658813:  44%|████▍     | 2852/6470 [35:22<41:10,  1.46it/s]
3|2853|Loss: 0.9457393884658813:  44%|████▍     | 2853/6470 [35:22<40:11,  1.50it/s]
3|2854|Loss: 0.906668484210968:  44%|████▍     | 2853/6470 [35:22<40:11,  1.50it/s] 
3|2854|Loss: 0.906668484210968:  44%|████▍     | 2854/6470 [35:23<43:21,  1.39it/s]
3|2855|Loss: 0.7661588191986084:  44%|████▍     | 2854/6470 [35:23<43:21,  1.39it/s]
3|2855|Loss: 0.7661588191986084:  44%|████▍     | 2855/6470 [35:24<45:42,  1.32it/s]
3|2856|Loss: 0.9261508584022522:  44%|████▍     | 2855/6470 [35:24<45:42,  1.32it/s]
3|2856|Loss: 0.9261508584022522:  44%|████▍     | 2856/6470 [35:24<46:43,  1.29it/s]
3|2857|Loss: 0.7794727087020874:  44%|████▍     | 2856/6470 [35:25<46:43,  1.29it/s]
3|2857|Loss: 0.7794727087020874:  44%|████▍     | 2857/6470 [35:25<49:37,  1.21it/s]
3|2858|Loss: 0.7247962355613708:  44%|████▍     | 2857/6470 [35:26<49:37,  1.21it/s]
3|2858|Loss: 0.7247962355613708:  44%|████▍     | 2858/6470 [35:26<48:03,  1.25it/s]
3|2859|Loss: 0.82138592004776:  44%|████▍     | 2858/6470 [35:27<48:03,  1.25it/s]  
3|2859|Loss: 0.82138592004776:  44%|████▍     | 2859/6470 [35:27<49:06,  1.23it/s]
3|2860|Loss: 0.8539721369743347:  44%|████▍     | 2859/6470 [35:28<49:06,  1.23it/s]
3|2860|Loss: 0.8539721369743347:  44%|████▍     | 2860/6470 [35:28<51:49,  1.16it/s]
3|2861|Loss: 0.9038538336753845:  44%|████▍     | 2860/6470 [35:28<51:49,  1.16it/s]
3|2861|Loss: 0.9038538336753845:  44%|████▍     | 2861/6470 [35:29<50:44,  1.19it/s]
3|2862|Loss: 0.8427415490150452:  44%|████▍     | 2861/6470 [35:29<50:44,  1.19it/s]
3|2862|Loss: 0.8427415490150452:  44%|████▍     | 2862/6470 [35:29<45:04,  1.33it/s]
3|2863|Loss: 0.76361483335495:  44%|████▍     | 2862/6470 [35:30<45:04,  1.33it/s]  
3|2863|Loss: 0.76361483335495:  44%|████▍     | 2863/6470 [35:30<43:54,  1.37it/s]
3|2864|Loss: 0.8673641681671143:  44%|████▍     | 2863/6470 [35:30<43:54,  1.37it/s]
3|2864|Loss: 0.8673641681671143:  44%|████▍     | 2864/6470 [35:31<45:04,  1.33it/s]
3|2865|Loss: 0.794306218624115:  44%|████▍     | 2864/6470 [35:31<45:04,  1.33it/s] 
3|2865|Loss: 0.794306218624115:  44%|████▍     | 2865/6470 [35:32<45:38,  1.32it/s]
3|2866|Loss: 1.026137351989746:  44%|████▍     | 2865/6470 [35:32<45:38,  1.32it/s]
3|2866|Loss: 1.026137351989746:  44%|████▍     | 2866/6470 [35:32<44:08,  1.36it/s]
3|2867|Loss: 0.9001703858375549:  44%|████▍     | 2866/6470 [35:33<44:08,  1.36it/s]
3|2867|Loss: 0.9001703858375549:  44%|████▍     | 2867/6470 [35:33<43:13,  1.39it/s]
3|2868|Loss: 0.9281595945358276:  44%|████▍     | 2867/6470 [35:33<43:13,  1.39it/s]
3|2868|Loss: 0.9281595945358276:  44%|████▍     | 2868/6470 [35:34<46:26,  1.29it/s]
3|2869|Loss: 0.7980232834815979:  44%|████▍     | 2868/6470 [35:34<46:26,  1.29it/s]
3|2869|Loss: 0.7980232834815979:  44%|████▍     | 2869/6470 [35:35<49:34,  1.21it/s]
3|2870|Loss: 0.8099433779716492:  44%|████▍     | 2869/6470 [35:35<49:34,  1.21it/s]
3|2870|Loss: 0.8099433779716492:  44%|████▍     | 2870/6470 [35:36<49:46,  1.21it/s]
3|2871|Loss: 0.8403620719909668:  44%|████▍     | 2870/6470 [35:36<49:46,  1.21it/s]
3|2871|Loss: 0.8403620719909668:  44%|████▍     | 2871/6470 [35:37<51:32,  1.16it/s]
3|2872|Loss: 0.9134671688079834:  44%|████▍     | 2871/6470 [35:37<51:32,  1.16it/s]
3|2872|Loss: 0.9134671688079834:  44%|████▍     | 2872/6470 [35:37<48:36,  1.23it/s]
3|2873|Loss: 1.095570683479309:  44%|████▍     | 2872/6470 [35:38<48:36,  1.23it/s] 
3|2873|Loss: 1.095570683479309:  44%|████▍     | 2873/6470 [35:38<46:50,  1.28it/s]
3|2874|Loss: 0.8424013257026672:  44%|████▍     | 2873/6470 [35:38<46:50,  1.28it/s]
3|2874|Loss: 0.8424013257026672:  44%|████▍     | 2874/6470 [35:39<47:36,  1.26it/s]
3|2875|Loss: 0.7646074891090393:  44%|████▍     | 2874/6470 [35:39<47:36,  1.26it/s]
3|2875|Loss: 0.7646074891090393:  44%|████▍     | 2875/6470 [35:40<47:18,  1.27it/s]
3|2876|Loss: 0.7983814477920532:  44%|████▍     | 2875/6470 [35:40<47:18,  1.27it/s]
3|2876|Loss: 0.7983814477920532:  44%|████▍     | 2876/6470 [35:40<47:12,  1.27it/s]
3|2877|Loss: 0.9312368035316467:  44%|████▍     | 2876/6470 [35:41<47:12,  1.27it/s]
3|2877|Loss: 0.9312368035316467:  44%|████▍     | 2877/6470 [35:41<45:50,  1.31it/s]
3|2878|Loss: 0.80329430103302:  44%|████▍     | 2877/6470 [35:41<45:50,  1.31it/s]  
3|2878|Loss: 0.80329430103302:  44%|████▍     | 2878/6470 [35:42<46:00,  1.30it/s]
3|2879|Loss: 0.7901598811149597:  44%|████▍     | 2878/6470 [35:42<46:00,  1.30it/s]
3|2879|Loss: 0.7901598811149597:  44%|████▍     | 2879/6470 [35:42<43:34,  1.37it/s]
3|2880|Loss: 1.0352978706359863:  44%|████▍     | 2879/6470 [35:43<43:34,  1.37it/s]
3|2880|Loss: 1.0352978706359863:  45%|████▍     | 2880/6470 [35:43<46:19,  1.29it/s]
3|2881|Loss: 0.9207231998443604:  45%|████▍     | 2880/6470 [35:44<46:19,  1.29it/s]
3|2881|Loss: 0.9207231998443604:  45%|████▍     | 2881/6470 [35:44<46:10,  1.30it/s]
3|2882|Loss: 0.7929828763008118:  45%|████▍     | 2881/6470 [35:45<46:10,  1.30it/s]
3|2882|Loss: 0.7929828763008118:  45%|████▍     | 2882/6470 [35:45<45:45,  1.31it/s]
3|2883|Loss: 0.8949254155158997:  45%|████▍     | 2882/6470 [35:45<45:45,  1.31it/s]
3|2883|Loss: 0.8949254155158997:  45%|████▍     | 2883/6470 [35:46<44:34,  1.34it/s]
3|2884|Loss: 0.7962183952331543:  45%|████▍     | 2883/6470 [35:46<44:34,  1.34it/s]
3|2884|Loss: 0.7962183952331543:  45%|████▍     | 2884/6470 [35:46<44:27,  1.34it/s]
3|2885|Loss: 0.9695102572441101:  45%|████▍     | 2884/6470 [35:47<44:27,  1.34it/s]
3|2885|Loss: 0.9695102572441101:  45%|████▍     | 2885/6470 [35:47<44:37,  1.34it/s]
3|2886|Loss: 0.9497848153114319:  45%|████▍     | 2885/6470 [35:47<44:37,  1.34it/s]
3|2886|Loss: 0.9497848153114319:  45%|████▍     | 2886/6470 [35:48<42:20,  1.41it/s]
3|2887|Loss: 0.8362769484519958:  45%|████▍     | 2886/6470 [35:48<42:20,  1.41it/s]
3|2887|Loss: 0.8362769484519958:  45%|████▍     | 2887/6470 [35:48<43:13,  1.38it/s]
3|2888|Loss: 0.7753538489341736:  45%|████▍     | 2887/6470 [35:49<43:13,  1.38it/s]
3|2888|Loss: 0.7753538489341736:  45%|████▍     | 2888/6470 [35:49<43:10,  1.38it/s]
3|2889|Loss: 0.8477922677993774:  45%|████▍     | 2888/6470 [35:50<43:10,  1.38it/s]
3|2889|Loss: 0.8477922677993774:  45%|████▍     | 2889/6470 [35:50<47:45,  1.25it/s]
3|2890|Loss: 0.9466699957847595:  45%|████▍     | 2889/6470 [35:51<47:45,  1.25it/s]
3|2890|Loss: 0.9466699957847595:  45%|████▍     | 2890/6470 [35:51<46:58,  1.27it/s]
3|2891|Loss: 0.8860172033309937:  45%|████▍     | 2890/6470 [35:51<46:58,  1.27it/s]
3|2891|Loss: 0.8860172033309937:  45%|████▍     | 2891/6470 [35:52<48:13,  1.24it/s]
3|2892|Loss: 0.9125253558158875:  45%|████▍     | 2891/6470 [35:52<48:13,  1.24it/s]
3|2892|Loss: 0.9125253558158875:  45%|████▍     | 2892/6470 [35:53<49:50,  1.20it/s]
3|2893|Loss: 0.8041656613349915:  45%|████▍     | 2892/6470 [35:53<49:50,  1.20it/s]
3|2893|Loss: 0.8041656613349915:  45%|████▍     | 2893/6470 [35:53<48:48,  1.22it/s]
3|2894|Loss: 0.7674044370651245:  45%|████▍     | 2893/6470 [35:54<48:48,  1.22it/s]
3|2894|Loss: 0.7674044370651245:  45%|████▍     | 2894/6470 [35:54<45:08,  1.32it/s]
3|2895|Loss: 0.8330675363540649:  45%|████▍     | 2894/6470 [35:54<45:08,  1.32it/s]
3|2895|Loss: 0.8330675363540649:  45%|████▍     | 2895/6470 [35:55<42:29,  1.40it/s]
3|2896|Loss: 0.8907788395881653:  45%|████▍     | 2895/6470 [35:55<42:29,  1.40it/s]
3|2896|Loss: 0.8907788395881653:  45%|████▍     | 2896/6470 [35:55<43:44,  1.36it/s]
3|2897|Loss: 0.8969045281410217:  45%|████▍     | 2896/6470 [35:56<43:44,  1.36it/s]
3|2897|Loss: 0.8969045281410217:  45%|████▍     | 2897/6470 [35:56<41:28,  1.44it/s]
3|2898|Loss: 0.8598849773406982:  45%|████▍     | 2897/6470 [35:56<41:28,  1.44it/s]
3|2898|Loss: 0.8598849773406982:  45%|████▍     | 2898/6470 [35:57<40:21,  1.48it/s]
3|2899|Loss: 0.9432404041290283:  45%|████▍     | 2898/6470 [35:57<40:21,  1.48it/s]
3|2899|Loss: 0.9432404041290283:  45%|████▍     | 2899/6470 [35:57<40:21,  1.47it/s]
3|2900|Loss: 0.8694911003112793:  45%|████▍     | 2899/6470 [35:58<40:21,  1.47it/s]
3|2900|Loss: 0.8694911003112793:  45%|████▍     | 2900/6470 [35:58<40:14,  1.48it/s]
3|2901|Loss: 0.7616660594940186:  45%|████▍     | 2900/6470 [35:58<40:14,  1.48it/s]
3|2901|Loss: 0.7616660594940186:  45%|████▍     | 2901/6470 [35:58<35:52,  1.66it/s]
3|2902|Loss: 0.7815805673599243:  45%|████▍     | 2901/6470 [35:59<35:52,  1.66it/s]
3|2902|Loss: 0.7815805673599243:  45%|████▍     | 2902/6470 [35:59<37:39,  1.58it/s]
3|2903|Loss: 0.8557597994804382:  45%|████▍     | 2902/6470 [36:00<37:39,  1.58it/s]
3|2903|Loss: 0.8557597994804382:  45%|████▍     | 2903/6470 [36:00<40:51,  1.46it/s]
3|2904|Loss: 0.8237491846084595:  45%|████▍     | 2903/6470 [36:00<40:51,  1.46it/s]
3|2904|Loss: 0.8237491846084595:  45%|████▍     | 2904/6470 [36:01<42:57,  1.38it/s]
3|2905|Loss: 0.8827371001243591:  45%|████▍     | 2904/6470 [36:01<42:57,  1.38it/s]
3|2905|Loss: 0.8827371001243591:  45%|████▍     | 2905/6470 [36:02<43:37,  1.36it/s]
3|2906|Loss: 0.7510086894035339:  45%|████▍     | 2905/6470 [36:02<43:37,  1.36it/s]
3|2906|Loss: 0.7510086894035339:  45%|████▍     | 2906/6470 [36:02<46:59,  1.26it/s]
3|2907|Loss: 0.9020344614982605:  45%|████▍     | 2906/6470 [36:03<46:59,  1.26it/s]
3|2907|Loss: 0.9020344614982605:  45%|████▍     | 2907/6470 [36:03<48:00,  1.24it/s]
3|2908|Loss: 0.8521819114685059:  45%|████▍     | 2907/6470 [36:04<48:00,  1.24it/s]
3|2908|Loss: 0.8521819114685059:  45%|████▍     | 2908/6470 [36:04<44:20,  1.34it/s]
3|2909|Loss: 0.8352338075637817:  45%|████▍     | 2908/6470 [36:04<44:20,  1.34it/s]
3|2909|Loss: 0.8352338075637817:  45%|████▍     | 2909/6470 [36:05<42:39,  1.39it/s]
3|2910|Loss: 0.7467687129974365:  45%|████▍     | 2909/6470 [36:05<42:39,  1.39it/s]
3|2910|Loss: 0.7467687129974365:  45%|████▍     | 2910/6470 [36:05<43:38,  1.36it/s]
3|2911|Loss: 0.8397078514099121:  45%|████▍     | 2910/6470 [36:06<43:38,  1.36it/s]
3|2911|Loss: 0.8397078514099121:  45%|████▍     | 2911/6470 [36:06<46:29,  1.28it/s]
3|2912|Loss: 0.7639625072479248:  45%|████▍     | 2911/6470 [36:07<46:29,  1.28it/s]
3|2912|Loss: 0.7639625072479248:  45%|████▌     | 2912/6470 [36:07<45:27,  1.30it/s]
3|2913|Loss: 0.887946367263794:  45%|████▌     | 2912/6470 [36:07<45:27,  1.30it/s] 
3|2913|Loss: 0.887946367263794:  45%|████▌     | 2913/6470 [36:08<42:27,  1.40it/s]
3|2914|Loss: 0.9459611177444458:  45%|████▌     | 2913/6470 [36:08<42:27,  1.40it/s]
3|2914|Loss: 0.9459611177444458:  45%|████▌     | 2914/6470 [36:08<42:17,  1.40it/s]
3|2915|Loss: 0.9972860217094421:  45%|████▌     | 2914/6470 [36:09<42:17,  1.40it/s]
3|2915|Loss: 0.9972860217094421:  45%|████▌     | 2915/6470 [36:09<47:21,  1.25it/s]
3|2916|Loss: 0.6707661747932434:  45%|████▌     | 2915/6470 [36:10<47:21,  1.25it/s]
3|2916|Loss: 0.6707661747932434:  45%|████▌     | 2916/6470 [36:10<51:52,  1.14it/s]
3|2917|Loss: 0.859761118888855:  45%|████▌     | 2916/6470 [36:11<51:52,  1.14it/s] 
3|2917|Loss: 0.859761118888855:  45%|████▌     | 2917/6470 [36:11<52:21,  1.13it/s]
3|2918|Loss: 0.7334900498390198:  45%|████▌     | 2917/6470 [36:12<52:21,  1.13it/s]
3|2918|Loss: 0.7334900498390198:  45%|████▌     | 2918/6470 [36:12<52:21,  1.13it/s]
3|2919|Loss: 0.9077399969100952:  45%|████▌     | 2918/6470 [36:13<52:21,  1.13it/s]
3|2919|Loss: 0.9077399969100952:  45%|████▌     | 2919/6470 [36:13<49:30,  1.20it/s]
3|2920|Loss: 0.7509592175483704:  45%|████▌     | 2919/6470 [36:13<49:30,  1.20it/s]
3|2920|Loss: 0.7509592175483704:  45%|████▌     | 2920/6470 [36:14<48:20,  1.22it/s]
3|2921|Loss: 0.8558023571968079:  45%|████▌     | 2920/6470 [36:14<48:20,  1.22it/s]
3|2921|Loss: 0.8558023571968079:  45%|████▌     | 2921/6470 [36:14<47:26,  1.25it/s]
3|2922|Loss: 0.9422246813774109:  45%|████▌     | 2921/6470 [36:15<47:26,  1.25it/s]
3|2922|Loss: 0.9422246813774109:  45%|████▌     | 2922/6470 [36:15<45:47,  1.29it/s]
3|2923|Loss: 0.8221701383590698:  45%|████▌     | 2922/6470 [36:15<45:47,  1.29it/s]
3|2923|Loss: 0.8221701383590698:  45%|████▌     | 2923/6470 [36:16<45:05,  1.31it/s]
3|2924|Loss: 0.7451006174087524:  45%|████▌     | 2923/6470 [36:16<45:05,  1.31it/s]
3|2924|Loss: 0.7451006174087524:  45%|████▌     | 2924/6470 [36:16<39:34,  1.49it/s]
3|2925|Loss: 0.7612639665603638:  45%|████▌     | 2924/6470 [36:17<39:34,  1.49it/s]
3|2925|Loss: 0.7612639665603638:  45%|████▌     | 2925/6470 [36:17<38:27,  1.54it/s]
3|2926|Loss: 0.7222296595573425:  45%|████▌     | 2925/6470 [36:17<38:27,  1.54it/s]
3|2926|Loss: 0.7222296595573425:  45%|████▌     | 2926/6470 [36:18<41:42,  1.42it/s]
3|2927|Loss: 0.9055196642875671:  45%|████▌     | 2926/6470 [36:18<41:42,  1.42it/s]
3|2927|Loss: 0.9055196642875671:  45%|████▌     | 2927/6470 [36:18<42:49,  1.38it/s]
3|2928|Loss: 0.7348849177360535:  45%|████▌     | 2927/6470 [36:19<42:49,  1.38it/s]
3|2928|Loss: 0.7348849177360535:  45%|████▌     | 2928/6470 [36:19<41:16,  1.43it/s]
3|2929|Loss: 0.7891315221786499:  45%|████▌     | 2928/6470 [36:20<41:16,  1.43it/s]
3|2929|Loss: 0.7891315221786499:  45%|████▌     | 2929/6470 [36:20<42:38,  1.38it/s]
3|2930|Loss: 0.9109386801719666:  45%|████▌     | 2929/6470 [36:20<42:38,  1.38it/s]
3|2930|Loss: 0.9109386801719666:  45%|████▌     | 2930/6470 [36:21<45:29,  1.30it/s]
3|2931|Loss: 0.958457350730896:  45%|████▌     | 2930/6470 [36:21<45:29,  1.30it/s] 
3|2931|Loss: 0.958457350730896:  45%|████▌     | 2931/6470 [36:22<45:26,  1.30it/s]
3|2932|Loss: 0.8426477909088135:  45%|████▌     | 2931/6470 [36:22<45:26,  1.30it/s]
3|2932|Loss: 0.8426477909088135:  45%|████▌     | 2932/6470 [36:22<42:48,  1.38it/s]
3|2933|Loss: 0.9534540772438049:  45%|████▌     | 2932/6470 [36:23<42:48,  1.38it/s]
3|2933|Loss: 0.9534540772438049:  45%|████▌     | 2933/6470 [36:23<42:42,  1.38it/s]
3|2934|Loss: 0.8856987953186035:  45%|████▌     | 2933/6470 [36:23<42:42,  1.38it/s]
3|2934|Loss: 0.8856987953186035:  45%|████▌     | 2934/6470 [36:24<43:36,  1.35it/s]
3|2935|Loss: 0.8785040378570557:  45%|████▌     | 2934/6470 [36:24<43:36,  1.35it/s]
3|2935|Loss: 0.8785040378570557:  45%|████▌     | 2935/6470 [36:24<44:29,  1.32it/s]
3|2936|Loss: 0.8125587105751038:  45%|████▌     | 2935/6470 [36:25<44:29,  1.32it/s]
3|2936|Loss: 0.8125587105751038:  45%|████▌     | 2936/6470 [36:25<41:26,  1.42it/s]
3|2937|Loss: 0.8894647359848022:  45%|████▌     | 2936/6470 [36:25<41:26,  1.42it/s]
3|2937|Loss: 0.8894647359848022:  45%|████▌     | 2937/6470 [36:26<43:39,  1.35it/s]
3|2938|Loss: 0.9361006021499634:  45%|████▌     | 2937/6470 [36:26<43:39,  1.35it/s]
3|2938|Loss: 0.9361006021499634:  45%|████▌     | 2938/6470 [36:27<44:14,  1.33it/s]
3|2939|Loss: 0.8927612900733948:  45%|████▌     | 2938/6470 [36:27<44:14,  1.33it/s]
3|2939|Loss: 0.8927612900733948:  45%|████▌     | 2939/6470 [36:28<46:57,  1.25it/s]
3|2940|Loss: 0.9174433350563049:  45%|████▌     | 2939/6470 [36:28<46:57,  1.25it/s]
3|2940|Loss: 0.9174433350563049:  45%|████▌     | 2940/6470 [36:28<48:58,  1.20it/s]
3|2941|Loss: 0.825194776058197:  45%|████▌     | 2940/6470 [36:29<48:58,  1.20it/s] 
3|2941|Loss: 0.825194776058197:  45%|████▌     | 2941/6470 [36:29<49:01,  1.20it/s]
3|2942|Loss: 0.8271802663803101:  45%|████▌     | 2941/6470 [36:30<49:01,  1.20it/s]
3|2942|Loss: 0.8271802663803101:  45%|████▌     | 2942/6470 [36:30<47:49,  1.23it/s]
3|2943|Loss: 0.8535739183425903:  45%|████▌     | 2942/6470 [36:31<47:49,  1.23it/s]
3|2943|Loss: 0.8535739183425903:  45%|████▌     | 2943/6470 [36:31<48:03,  1.22it/s]
3|2944|Loss: 0.8222665786743164:  45%|████▌     | 2943/6470 [36:31<48:03,  1.22it/s]
3|2944|Loss: 0.8222665786743164:  46%|████▌     | 2944/6470 [36:32<46:09,  1.27it/s]
3|2945|Loss: 0.7665446996688843:  46%|████▌     | 2944/6470 [36:32<46:09,  1.27it/s]
3|2945|Loss: 0.7665446996688843:  46%|████▌     | 2945/6470 [36:32<44:02,  1.33it/s]
3|2946|Loss: 0.8161932229995728:  46%|████▌     | 2945/6470 [36:33<44:02,  1.33it/s]
3|2946|Loss: 0.8161932229995728:  46%|████▌     | 2946/6470 [36:33<43:45,  1.34it/s]
3|2947|Loss: 0.8334353566169739:  46%|████▌     | 2946/6470 [36:33<43:45,  1.34it/s]
3|2947|Loss: 0.8334353566169739:  46%|████▌     | 2947/6470 [36:34<40:48,  1.44it/s]
3|2948|Loss: 0.9342648386955261:  46%|████▌     | 2947/6470 [36:34<40:48,  1.44it/s]
3|2948|Loss: 0.9342648386955261:  46%|████▌     | 2948/6470 [36:34<42:43,  1.37it/s]
3|2949|Loss: 0.7589988708496094:  46%|████▌     | 2948/6470 [36:35<42:43,  1.37it/s]
3|2949|Loss: 0.7589988708496094:  46%|████▌     | 2949/6470 [36:35<44:11,  1.33it/s]
3|2950|Loss: 0.8684564828872681:  46%|████▌     | 2949/6470 [36:36<44:11,  1.33it/s]
3|2950|Loss: 0.8684564828872681:  46%|████▌     | 2950/6470 [36:36<43:52,  1.34it/s]
3|2951|Loss: 0.8296235203742981:  46%|████▌     | 2950/6470 [36:36<43:52,  1.34it/s]
3|2951|Loss: 0.8296235203742981:  46%|████▌     | 2951/6470 [36:37<43:50,  1.34it/s]
3|2952|Loss: 0.7405139207839966:  46%|████▌     | 2951/6470 [36:37<43:50,  1.34it/s]
3|2952|Loss: 0.7405139207839966:  46%|████▌     | 2952/6470 [36:37<43:43,  1.34it/s]
3|2953|Loss: 0.7510038018226624:  46%|████▌     | 2952/6470 [36:38<43:43,  1.34it/s]
3|2953|Loss: 0.7510038018226624:  46%|████▌     | 2953/6470 [36:38<44:25,  1.32it/s]
3|2954|Loss: 0.7240492105484009:  46%|████▌     | 2953/6470 [36:39<44:25,  1.32it/s]
3|2954|Loss: 0.7240492105484009:  46%|████▌     | 2954/6470 [36:39<47:52,  1.22it/s]
3|2955|Loss: 1.030846118927002:  46%|████▌     | 2954/6470 [36:40<47:52,  1.22it/s] 
3|2955|Loss: 1.030846118927002:  46%|████▌     | 2955/6470 [36:40<49:22,  1.19it/s]
3|2956|Loss: 0.8774080276489258:  46%|████▌     | 2955/6470 [36:41<49:22,  1.19it/s]
3|2956|Loss: 0.8774080276489258:  46%|████▌     | 2956/6470 [36:41<48:18,  1.21it/s]
3|2957|Loss: 0.8655839562416077:  46%|████▌     | 2956/6470 [36:41<48:18,  1.21it/s]
3|2957|Loss: 0.8655839562416077:  46%|████▌     | 2957/6470 [36:42<49:20,  1.19it/s]
3|2958|Loss: 0.9190083146095276:  46%|████▌     | 2957/6470 [36:42<49:20,  1.19it/s]
3|2958|Loss: 0.9190083146095276:  46%|████▌     | 2958/6470 [36:43<48:22,  1.21it/s]
3|2959|Loss: 0.6910470128059387:  46%|████▌     | 2958/6470 [36:43<48:22,  1.21it/s]
3|2959|Loss: 0.6910470128059387:  46%|████▌     | 2959/6470 [36:43<47:51,  1.22it/s]
3|2960|Loss: 0.8207312226295471:  46%|████▌     | 2959/6470 [36:44<47:51,  1.22it/s]
3|2960|Loss: 0.8207312226295471:  46%|████▌     | 2960/6470 [36:44<47:36,  1.23it/s]
3|2961|Loss: 0.739240825176239:  46%|████▌     | 2960/6470 [36:44<47:36,  1.23it/s] 
3|2961|Loss: 0.739240825176239:  46%|████▌     | 2961/6470 [36:45<43:07,  1.36it/s]
3|2962|Loss: 0.7959371209144592:  46%|████▌     | 2961/6470 [36:45<43:07,  1.36it/s]
3|2962|Loss: 0.7959371209144592:  46%|████▌     | 2962/6470 [36:45<43:59,  1.33it/s]
3|2963|Loss: 0.8171789050102234:  46%|████▌     | 2962/6470 [36:46<43:59,  1.33it/s]
3|2963|Loss: 0.8171789050102234:  46%|████▌     | 2963/6470 [36:46<43:42,  1.34it/s]
3|2964|Loss: 0.8437435626983643:  46%|████▌     | 2963/6470 [36:47<43:42,  1.34it/s]
3|2964|Loss: 0.8437435626983643:  46%|████▌     | 2964/6470 [36:47<43:01,  1.36it/s]
3|2965|Loss: 0.8681631088256836:  46%|████▌     | 2964/6470 [36:47<43:01,  1.36it/s]
3|2965|Loss: 0.8681631088256836:  46%|████▌     | 2965/6470 [36:48<42:13,  1.38it/s]
3|2966|Loss: 0.8432935476303101:  46%|████▌     | 2965/6470 [36:48<42:13,  1.38it/s]
3|2966|Loss: 0.8432935476303101:  46%|████▌     | 2966/6470 [36:48<42:52,  1.36it/s]
3|2967|Loss: 0.7444190382957458:  46%|████▌     | 2966/6470 [36:49<42:52,  1.36it/s]
3|2967|Loss: 0.7444190382957458:  46%|████▌     | 2967/6470 [36:49<42:36,  1.37it/s]
3|2968|Loss: 0.7211667895317078:  46%|████▌     | 2967/6470 [36:49<42:36,  1.37it/s]
3|2968|Loss: 0.7211667895317078:  46%|████▌     | 2968/6470 [36:50<39:50,  1.46it/s]
3|2969|Loss: 0.9908260703086853:  46%|████▌     | 2968/6470 [36:50<39:50,  1.46it/s]
3|2969|Loss: 0.9908260703086853:  46%|████▌     | 2969/6470 [36:50<42:06,  1.39it/s]
3|2970|Loss: 0.7442429065704346:  46%|████▌     | 2969/6470 [36:51<42:06,  1.39it/s]
3|2970|Loss: 0.7442429065704346:  46%|████▌     | 2970/6470 [36:51<43:33,  1.34it/s]
3|2971|Loss: 0.7451624274253845:  46%|████▌     | 2970/6470 [36:52<43:33,  1.34it/s]
3|2971|Loss: 0.7451624274253845:  46%|████▌     | 2971/6470 [36:52<41:11,  1.42it/s]
3|2972|Loss: 0.7508833408355713:  46%|████▌     | 2971/6470 [36:52<41:11,  1.42it/s]
3|2972|Loss: 0.7508833408355713:  46%|████▌     | 2972/6470 [36:52<38:13,  1.53it/s]
3|2973|Loss: 0.9085635542869568:  46%|████▌     | 2972/6470 [36:53<38:13,  1.53it/s]
3|2973|Loss: 0.9085635542869568:  46%|████▌     | 2973/6470 [36:53<39:12,  1.49it/s]
3|2974|Loss: 0.8342857956886292:  46%|████▌     | 2973/6470 [36:54<39:12,  1.49it/s]
3|2974|Loss: 0.8342857956886292:  46%|████▌     | 2974/6470 [36:54<37:36,  1.55it/s]
3|2975|Loss: 0.9216712117195129:  46%|████▌     | 2974/6470 [36:54<37:36,  1.55it/s]
3|2975|Loss: 0.9216712117195129:  46%|████▌     | 2975/6470 [36:54<38:38,  1.51it/s]
3|2976|Loss: 0.7531319260597229:  46%|████▌     | 2975/6470 [36:55<38:38,  1.51it/s]
3|2976|Loss: 0.7531319260597229:  46%|████▌     | 2976/6470 [36:55<37:23,  1.56it/s]
3|2977|Loss: 0.6984487175941467:  46%|████▌     | 2976/6470 [36:55<37:23,  1.56it/s]
3|2977|Loss: 0.6984487175941467:  46%|████▌     | 2977/6470 [36:56<36:44,  1.58it/s]
3|2978|Loss: 0.8873963356018066:  46%|████▌     | 2977/6470 [36:56<36:44,  1.58it/s]
3|2978|Loss: 0.8873963356018066:  46%|████▌     | 2978/6470 [36:56<38:31,  1.51it/s]
3|2979|Loss: 0.9661461114883423:  46%|████▌     | 2978/6470 [36:57<38:31,  1.51it/s]
3|2979|Loss: 0.9661461114883423:  46%|████▌     | 2979/6470 [36:57<38:23,  1.52it/s]
3|2980|Loss: 0.9262651801109314:  46%|████▌     | 2979/6470 [36:57<38:23,  1.52it/s]
3|2980|Loss: 0.9262651801109314:  46%|████▌     | 2980/6470 [36:58<35:34,  1.63it/s]
3|2981|Loss: 0.7398815155029297:  46%|████▌     | 2980/6470 [36:58<35:34,  1.63it/s]
3|2981|Loss: 0.7398815155029297:  46%|████▌     | 2981/6470 [36:58<36:41,  1.59it/s]
3|2982|Loss: 0.8131163716316223:  46%|████▌     | 2981/6470 [36:59<36:41,  1.59it/s]
3|2982|Loss: 0.8131163716316223:  46%|████▌     | 2982/6470 [36:59<39:23,  1.48it/s]
3|2983|Loss: 0.6784815192222595:  46%|████▌     | 2982/6470 [36:59<39:23,  1.48it/s]
3|2983|Loss: 0.6784815192222595:  46%|████▌     | 2983/6470 [37:00<41:33,  1.40it/s]
3|2984|Loss: 0.8535086512565613:  46%|████▌     | 2983/6470 [37:00<41:33,  1.40it/s]
3|2984|Loss: 0.8535086512565613:  46%|████▌     | 2984/6470 [37:01<42:28,  1.37it/s]
3|2985|Loss: 0.8230619430541992:  46%|████▌     | 2984/6470 [37:01<42:28,  1.37it/s]
3|2985|Loss: 0.8230619430541992:  46%|████▌     | 2985/6470 [37:01<45:12,  1.28it/s]
3|2986|Loss: 0.9018967747688293:  46%|████▌     | 2985/6470 [37:02<45:12,  1.28it/s]
3|2986|Loss: 0.9018967747688293:  46%|████▌     | 2986/6470 [37:02<47:54,  1.21it/s]
3|2987|Loss: 0.9313380122184753:  46%|████▌     | 2986/6470 [37:03<47:54,  1.21it/s]
3|2987|Loss: 0.9313380122184753:  46%|████▌     | 2987/6470 [37:03<43:47,  1.33it/s]
3|2988|Loss: 0.8817863464355469:  46%|████▌     | 2987/6470 [37:03<43:47,  1.33it/s]
3|2988|Loss: 0.8817863464355469:  46%|████▌     | 2988/6470 [37:04<42:48,  1.36it/s]
3|2989|Loss: 0.7262906432151794:  46%|████▌     | 2988/6470 [37:04<42:48,  1.36it/s]
3|2989|Loss: 0.7262906432151794:  46%|████▌     | 2989/6470 [37:04<42:26,  1.37it/s]
3|2990|Loss: 0.6889990568161011:  46%|████▌     | 2989/6470 [37:05<42:26,  1.37it/s]
3|2990|Loss: 0.6889990568161011:  46%|████▌     | 2990/6470 [37:05<40:44,  1.42it/s]
3|2991|Loss: 0.8959184288978577:  46%|████▌     | 2990/6470 [37:05<40:44,  1.42it/s]
3|2991|Loss: 0.8959184288978577:  46%|████▌     | 2991/6470 [37:06<41:31,  1.40it/s]
3|2992|Loss: 0.6792503595352173:  46%|████▌     | 2991/6470 [37:06<41:31,  1.40it/s]
3|2992|Loss: 0.6792503595352173:  46%|████▌     | 2992/6470 [37:06<39:22,  1.47it/s]
3|2993|Loss: 1.0120916366577148:  46%|████▌     | 2992/6470 [37:07<39:22,  1.47it/s]
3|2993|Loss: 1.0120916366577148:  46%|████▋     | 2993/6470 [37:07<38:07,  1.52it/s]
3|2994|Loss: 0.6676246523857117:  46%|████▋     | 2993/6470 [37:07<38:07,  1.52it/s]
3|2994|Loss: 0.6676246523857117:  46%|████▋     | 2994/6470 [37:08<38:38,  1.50it/s]
3|2995|Loss: 0.7656436562538147:  46%|████▋     | 2994/6470 [37:08<38:38,  1.50it/s]
3|2995|Loss: 0.7656436562538147:  46%|████▋     | 2995/6470 [37:08<40:00,  1.45it/s]
3|2996|Loss: 0.987521767616272:  46%|████▋     | 2995/6470 [37:09<40:00,  1.45it/s] 
3|2996|Loss: 0.987521767616272:  46%|████▋     | 2996/6470 [37:09<44:09,  1.31it/s]
3|2997|Loss: 0.7759026885032654:  46%|████▋     | 2996/6470 [37:10<44:09,  1.31it/s]
3|2997|Loss: 0.7759026885032654:  46%|████▋     | 2997/6470 [37:10<45:26,  1.27it/s]
3|2998|Loss: 0.8250856995582581:  46%|████▋     | 2997/6470 [37:11<45:26,  1.27it/s]
3|2998|Loss: 0.8250856995582581:  46%|████▋     | 2998/6470 [37:11<48:26,  1.19it/s]
3|2999|Loss: 0.7550384998321533:  46%|████▋     | 2998/6470 [37:12<48:26,  1.19it/s]
3|2999|Loss: 0.7550384998321533:  46%|████▋     | 2999/6470 [37:12<50:24,  1.15it/s]
3|3000|Loss: 0.544703483581543:  46%|████▋     | 2999/6470 [37:13<50:24,  1.15it/s] 
3|3000|Loss: 0.544703483581543:  46%|████▋     | 3000/6470 [37:13<49:30,  1.17it/s]
3|3001|Loss: 0.8834742903709412:  46%|████▋     | 3000/6470 [37:13<49:30,  1.17it/s]
3|3001|Loss: 0.8834742903709412:  46%|████▋     | 3001/6470 [37:14<48:42,  1.19it/s]
3|3002|Loss: 0.8401409387588501:  46%|████▋     | 3001/6470 [37:14<48:42,  1.19it/s]
3|3002|Loss: 0.8401409387588501:  46%|████▋     | 3002/6470 [37:14<46:09,  1.25it/s]
3|3003|Loss: 0.8894010186195374:  46%|████▋     | 3002/6470 [37:15<46:09,  1.25it/s]
3|3003|Loss: 0.8894010186195374:  46%|████▋     | 3003/6470 [37:15<46:51,  1.23it/s]
3|3004|Loss: 1.039910078048706:  46%|████▋     | 3003/6470 [37:16<46:51,  1.23it/s] 
3|3004|Loss: 1.039910078048706:  46%|████▋     | 3004/6470 [37:16<48:40,  1.19it/s]
3|3005|Loss: 0.8006994724273682:  46%|████▋     | 3004/6470 [37:17<48:40,  1.19it/s]
3|3005|Loss: 0.8006994724273682:  46%|████▋     | 3005/6470 [37:17<45:33,  1.27it/s]
3|3006|Loss: 0.8114397525787354:  46%|████▋     | 3005/6470 [37:17<45:33,  1.27it/s]
3|3006|Loss: 0.8114397525787354:  46%|████▋     | 3006/6470 [37:18<45:59,  1.26it/s]
3|3007|Loss: 0.7188636064529419:  46%|████▋     | 3006/6470 [37:18<45:59,  1.26it/s]
3|3007|Loss: 0.7188636064529419:  46%|████▋     | 3007/6470 [37:18<46:55,  1.23it/s]
3|3008|Loss: 0.8876221179962158:  46%|████▋     | 3007/6470 [37:19<46:55,  1.23it/s]
3|3008|Loss: 0.8876221179962158:  46%|████▋     | 3008/6470 [37:19<47:07,  1.22it/s]
3|3009|Loss: 0.8986207842826843:  46%|████▋     | 3008/6470 [37:20<47:07,  1.22it/s]
3|3009|Loss: 0.8986207842826843:  47%|████▋     | 3009/6470 [37:20<45:18,  1.27it/s]
3|3010|Loss: 0.6470947861671448:  47%|████▋     | 3009/6470 [37:20<45:18,  1.27it/s]
3|3010|Loss: 0.6470947861671448:  47%|████▋     | 3010/6470 [37:21<45:08,  1.28it/s]
3|3011|Loss: 0.9088947176933289:  47%|████▋     | 3010/6470 [37:21<45:08,  1.28it/s]
3|3011|Loss: 0.9088947176933289:  47%|████▋     | 3011/6470 [37:22<45:29,  1.27it/s]
3|3012|Loss: 0.8065075278282166:  47%|████▋     | 3011/6470 [37:22<45:29,  1.27it/s]
3|3012|Loss: 0.8065075278282166:  47%|████▋     | 3012/6470 [37:22<43:18,  1.33it/s]
3|3013|Loss: 0.8781653046607971:  47%|████▋     | 3012/6470 [37:23<43:18,  1.33it/s]
3|3013|Loss: 0.8781653046607971:  47%|████▋     | 3013/6470 [37:23<44:46,  1.29it/s]
3|3014|Loss: 0.7209963798522949:  47%|████▋     | 3013/6470 [37:24<44:46,  1.29it/s]
3|3014|Loss: 0.7209963798522949:  47%|████▋     | 3014/6470 [37:24<43:48,  1.32it/s]
3|3015|Loss: 0.7551246285438538:  47%|████▋     | 3014/6470 [37:24<43:48,  1.32it/s]
3|3015|Loss: 0.7551246285438538:  47%|████▋     | 3015/6470 [37:25<42:20,  1.36it/s]
3|3016|Loss: 0.9891729354858398:  47%|████▋     | 3015/6470 [37:25<42:20,  1.36it/s]
3|3016|Loss: 0.9891729354858398:  47%|████▋     | 3016/6470 [37:25<39:19,  1.46it/s]
3|3017|Loss: 0.8001558780670166:  47%|████▋     | 3016/6470 [37:25<39:19,  1.46it/s]
3|3017|Loss: 0.8001558780670166:  47%|████▋     | 3017/6470 [37:26<40:31,  1.42it/s]
3|3018|Loss: 0.9435691237449646:  47%|████▋     | 3017/6470 [37:26<40:31,  1.42it/s]
3|3018|Loss: 0.9435691237449646:  47%|████▋     | 3018/6470 [37:27<40:28,  1.42it/s]
3|3019|Loss: 0.8028327226638794:  47%|████▋     | 3018/6470 [37:27<40:28,  1.42it/s]
3|3019|Loss: 0.8028327226638794:  47%|████▋     | 3019/6470 [37:27<39:45,  1.45it/s]
3|3020|Loss: 0.7465499043464661:  47%|████▋     | 3019/6470 [37:28<39:45,  1.45it/s]
3|3020|Loss: 0.7465499043464661:  47%|████▋     | 3020/6470 [37:28<36:41,  1.57it/s]
3|3021|Loss: 0.9150758385658264:  47%|████▋     | 3020/6470 [37:28<36:41,  1.57it/s]
3|3021|Loss: 0.9150758385658264:  47%|████▋     | 3021/6470 [37:28<35:14,  1.63it/s]
3|3022|Loss: 0.6771822571754456:  47%|████▋     | 3021/6470 [37:29<35:14,  1.63it/s]
3|3022|Loss: 0.6771822571754456:  47%|████▋     | 3022/6470 [37:29<36:17,  1.58it/s]
3|3023|Loss: 0.8254520893096924:  47%|████▋     | 3022/6470 [37:29<36:17,  1.58it/s]
3|3023|Loss: 0.8254520893096924:  47%|████▋     | 3023/6470 [37:30<36:39,  1.57it/s]
3|3024|Loss: 1.213974118232727:  47%|████▋     | 3023/6470 [37:30<36:39,  1.57it/s] 
3|3024|Loss: 1.213974118232727:  47%|████▋     | 3024/6470 [37:30<40:26,  1.42it/s]
3|3025|Loss: 0.8337662816047668:  47%|████▋     | 3024/6470 [37:31<40:26,  1.42it/s]
3|3025|Loss: 0.8337662816047668:  47%|████▋     | 3025/6470 [37:31<40:06,  1.43it/s]
3|3026|Loss: 0.7882165312767029:  47%|████▋     | 3025/6470 [37:32<40:06,  1.43it/s]
3|3026|Loss: 0.7882165312767029:  47%|████▋     | 3026/6470 [37:32<42:18,  1.36it/s]
3|3027|Loss: 0.7840662598609924:  47%|████▋     | 3026/6470 [37:32<42:18,  1.36it/s]
3|3027|Loss: 0.7840662598609924:  47%|████▋     | 3027/6470 [37:33<41:06,  1.40it/s]
3|3028|Loss: 0.7294933199882507:  47%|████▋     | 3027/6470 [37:33<41:06,  1.40it/s]
3|3028|Loss: 0.7294933199882507:  47%|████▋     | 3028/6470 [37:33<41:54,  1.37it/s]
3|3029|Loss: 0.9338698983192444:  47%|████▋     | 3028/6470 [37:34<41:54,  1.37it/s]
3|3029|Loss: 0.9338698983192444:  47%|████▋     | 3029/6470 [37:34<42:38,  1.35it/s]
3|3030|Loss: 0.881052553653717:  47%|████▋     | 3029/6470 [37:35<42:38,  1.35it/s] 
3|3030|Loss: 0.881052553653717:  47%|████▋     | 3030/6470 [37:35<44:21,  1.29it/s]
3|3031|Loss: 0.8717167973518372:  47%|████▋     | 3030/6470 [37:35<44:21,  1.29it/s]
3|3031|Loss: 0.8717167973518372:  47%|████▋     | 3031/6470 [37:36<46:25,  1.23it/s]
3|3032|Loss: 0.7694560289382935:  47%|████▋     | 3031/6470 [37:36<46:25,  1.23it/s]
3|3032|Loss: 0.7694560289382935:  47%|████▋     | 3032/6470 [37:37<46:32,  1.23it/s]
3|3033|Loss: 1.0192346572875977:  47%|████▋     | 3032/6470 [37:37<46:32,  1.23it/s]
3|3033|Loss: 1.0192346572875977:  47%|████▋     | 3033/6470 [37:38<47:32,  1.20it/s]
3|3034|Loss: 0.9003280997276306:  47%|████▋     | 3033/6470 [37:38<47:32,  1.20it/s]
3|3034|Loss: 0.9003280997276306:  47%|████▋     | 3034/6470 [37:38<46:38,  1.23it/s]
3|3035|Loss: 0.7583330273628235:  47%|████▋     | 3034/6470 [37:39<46:38,  1.23it/s]
3|3035|Loss: 0.7583330273628235:  47%|████▋     | 3035/6470 [37:39<46:04,  1.24it/s]
3|3036|Loss: 0.9728984236717224:  47%|████▋     | 3035/6470 [37:40<46:04,  1.24it/s]
3|3036|Loss: 0.9728984236717224:  47%|████▋     | 3036/6470 [37:40<44:32,  1.28it/s]
3|3037|Loss: 0.9490723609924316:  47%|████▋     | 3036/6470 [37:40<44:32,  1.28it/s]
3|3037|Loss: 0.9490723609924316:  47%|████▋     | 3037/6470 [37:41<43:58,  1.30it/s]
3|3038|Loss: 0.9610190391540527:  47%|████▋     | 3037/6470 [37:41<43:58,  1.30it/s]
3|3038|Loss: 0.9610190391540527:  47%|████▋     | 3038/6470 [37:41<45:13,  1.26it/s]
3|3039|Loss: 0.7763075232505798:  47%|████▋     | 3038/6470 [37:42<45:13,  1.26it/s]
3|3039|Loss: 0.7763075232505798:  47%|████▋     | 3039/6470 [37:42<45:06,  1.27it/s]
3|3040|Loss: 0.7745810747146606:  47%|████▋     | 3039/6470 [37:43<45:06,  1.27it/s]
3|3040|Loss: 0.7745810747146606:  47%|████▋     | 3040/6470 [37:43<44:27,  1.29it/s]
3|3041|Loss: 0.8434770703315735:  47%|████▋     | 3040/6470 [37:43<44:27,  1.29it/s]
3|3041|Loss: 0.8434770703315735:  47%|████▋     | 3041/6470 [37:44<42:51,  1.33it/s]
3|3042|Loss: 0.7592176198959351:  47%|████▋     | 3041/6470 [37:44<42:51,  1.33it/s]
3|3042|Loss: 0.7592176198959351:  47%|████▋     | 3042/6470 [37:44<41:20,  1.38it/s]
3|3043|Loss: 0.7606744170188904:  47%|████▋     | 3042/6470 [37:45<41:20,  1.38it/s]
3|3043|Loss: 0.7606744170188904:  47%|████▋     | 3043/6470 [37:45<38:33,  1.48it/s]
3|3044|Loss: 0.9012357592582703:  47%|████▋     | 3043/6470 [37:45<38:33,  1.48it/s]
3|3044|Loss: 0.9012357592582703:  47%|████▋     | 3044/6470 [37:46<39:06,  1.46it/s]
3|3045|Loss: 0.7593538165092468:  47%|████▋     | 3044/6470 [37:46<39:06,  1.46it/s]
3|3045|Loss: 0.7593538165092468:  47%|████▋     | 3045/6470 [37:46<38:42,  1.47it/s]
3|3046|Loss: 1.2095712423324585:  47%|████▋     | 3045/6470 [37:47<38:42,  1.47it/s]
3|3046|Loss: 1.2095712423324585:  47%|████▋     | 3046/6470 [37:47<36:28,  1.56it/s]
3|3047|Loss: 0.9163119196891785:  47%|████▋     | 3046/6470 [37:47<36:28,  1.56it/s]
3|3047|Loss: 0.9163119196891785:  47%|████▋     | 3047/6470 [37:48<37:33,  1.52it/s]
3|3048|Loss: 0.9122908711433411:  47%|████▋     | 3047/6470 [37:48<37:33,  1.52it/s]
3|3048|Loss: 0.9122908711433411:  47%|████▋     | 3048/6470 [37:48<42:03,  1.36it/s]
3|3049|Loss: 0.8060823678970337:  47%|████▋     | 3048/6470 [37:49<42:03,  1.36it/s]
3|3049|Loss: 0.8060823678970337:  47%|████▋     | 3049/6470 [37:49<42:15,  1.35it/s]
3|3050|Loss: 0.6011191606521606:  47%|████▋     | 3049/6470 [37:50<42:15,  1.35it/s]
3|3050|Loss: 0.6011191606521606:  47%|████▋     | 3050/6470 [37:50<44:18,  1.29it/s]
3|3051|Loss: 0.982734203338623:  47%|████▋     | 3050/6470 [37:50<44:18,  1.29it/s] 
3|3051|Loss: 0.982734203338623:  47%|████▋     | 3051/6470 [37:51<41:25,  1.38it/s]
3|3052|Loss: 0.7218703031539917:  47%|████▋     | 3051/6470 [37:51<41:25,  1.38it/s]
3|3052|Loss: 0.7218703031539917:  47%|████▋     | 3052/6470 [37:52<43:27,  1.31it/s]
3|3053|Loss: 0.7948745489120483:  47%|████▋     | 3052/6470 [37:52<43:27,  1.31it/s]
3|3053|Loss: 0.7948745489120483:  47%|████▋     | 3053/6470 [37:52<46:37,  1.22it/s]
3|3054|Loss: 0.7210458517074585:  47%|████▋     | 3053/6470 [37:53<46:37,  1.22it/s]
3|3054|Loss: 0.7210458517074585:  47%|████▋     | 3054/6470 [37:53<44:27,  1.28it/s]
3|3055|Loss: 0.8688628077507019:  47%|████▋     | 3054/6470 [37:54<44:27,  1.28it/s]
3|3055|Loss: 0.8688628077507019:  47%|████▋     | 3055/6470 [37:54<44:13,  1.29it/s]
3|3056|Loss: 0.8938575387001038:  47%|████▋     | 3055/6470 [37:54<44:13,  1.29it/s]
3|3056|Loss: 0.8938575387001038:  47%|████▋     | 3056/6470 [37:55<42:09,  1.35it/s]
3|3057|Loss: 0.8887287974357605:  47%|████▋     | 3056/6470 [37:55<42:09,  1.35it/s]
3|3057|Loss: 0.8887287974357605:  47%|████▋     | 3057/6470 [37:55<41:49,  1.36it/s]
3|3058|Loss: 1.0783363580703735:  47%|████▋     | 3057/6470 [37:56<41:49,  1.36it/s]
3|3058|Loss: 1.0783363580703735:  47%|████▋     | 3058/6470 [37:56<40:15,  1.41it/s]
3|3059|Loss: 0.8505239486694336:  47%|████▋     | 3058/6470 [37:56<40:15,  1.41it/s]
3|3059|Loss: 0.8505239486694336:  47%|████▋     | 3059/6470 [37:57<43:26,  1.31it/s]
3|3060|Loss: 0.7997310757637024:  47%|████▋     | 3059/6470 [37:57<43:26,  1.31it/s]
3|3060|Loss: 0.7997310757637024:  47%|████▋     | 3060/6470 [37:58<44:56,  1.26it/s]
3|3061|Loss: 0.9351407885551453:  47%|████▋     | 3060/6470 [37:58<44:56,  1.26it/s]
3|3061|Loss: 0.9351407885551453:  47%|████▋     | 3061/6470 [37:58<45:00,  1.26it/s]
3|3062|Loss: 0.8824383616447449:  47%|████▋     | 3061/6470 [37:59<45:00,  1.26it/s]
3|3062|Loss: 0.8824383616447449:  47%|████▋     | 3062/6470 [37:59<46:50,  1.21it/s]
3|3063|Loss: 0.8299787640571594:  47%|████▋     | 3062/6470 [38:00<46:50,  1.21it/s]
3|3063|Loss: 0.8299787640571594:  47%|████▋     | 3063/6470 [38:00<48:05,  1.18it/s]
3|3064|Loss: 0.809384822845459:  47%|████▋     | 3063/6470 [38:01<48:05,  1.18it/s] 
3|3064|Loss: 0.809384822845459:  47%|████▋     | 3064/6470 [38:01<48:55,  1.16it/s]
3|3065|Loss: 0.8345951437950134:  47%|████▋     | 3064/6470 [38:02<48:55,  1.16it/s]
3|3065|Loss: 0.8345951437950134:  47%|████▋     | 3065/6470 [38:02<46:45,  1.21it/s]
3|3066|Loss: 0.7927002310752869:  47%|████▋     | 3065/6470 [38:02<46:45,  1.21it/s]
3|3066|Loss: 0.7927002310752869:  47%|████▋     | 3066/6470 [38:03<47:41,  1.19it/s]
3|3067|Loss: 0.801459789276123:  47%|████▋     | 3066/6470 [38:03<47:41,  1.19it/s] 
3|3067|Loss: 0.801459789276123:  47%|████▋     | 3067/6470 [38:04<47:18,  1.20it/s]
3|3068|Loss: 0.797924280166626:  47%|████▋     | 3067/6470 [38:04<47:18,  1.20it/s]
3|3068|Loss: 0.797924280166626:  47%|████▋     | 3068/6470 [38:05<48:11,  1.18it/s]
3|3069|Loss: 1.01667058467865:  47%|████▋     | 3068/6470 [38:05<48:11,  1.18it/s] 
3|3069|Loss: 1.01667058467865:  47%|████▋     | 3069/6470 [38:05<46:08,  1.23it/s]
3|3070|Loss: 0.7950022220611572:  47%|████▋     | 3069/6470 [38:06<46:08,  1.23it/s]
3|3070|Loss: 0.7950022220611572:  47%|████▋     | 3070/6470 [38:06<44:47,  1.26it/s]
3|3071|Loss: 0.8215636610984802:  47%|████▋     | 3070/6470 [38:06<44:47,  1.26it/s]
3|3071|Loss: 0.8215636610984802:  47%|████▋     | 3071/6470 [38:07<46:27,  1.22it/s]
3|3072|Loss: 0.6522713899612427:  47%|████▋     | 3071/6470 [38:07<46:27,  1.22it/s]
3|3072|Loss: 0.6522713899612427:  47%|████▋     | 3072/6470 [38:08<46:40,  1.21it/s]
3|3073|Loss: 0.8798226118087769:  47%|████▋     | 3072/6470 [38:08<46:40,  1.21it/s]
3|3073|Loss: 0.8798226118087769:  47%|████▋     | 3073/6470 [38:09<46:28,  1.22it/s]
3|3074|Loss: 0.8895480632781982:  47%|████▋     | 3073/6470 [38:09<46:28,  1.22it/s]
3|3074|Loss: 0.8895480632781982:  48%|████▊     | 3074/6470 [38:09<45:28,  1.24it/s]
3|3075|Loss: 0.9680289626121521:  48%|████▊     | 3074/6470 [38:10<45:28,  1.24it/s]
3|3075|Loss: 0.9680289626121521:  48%|████▊     | 3075/6470 [38:10<42:02,  1.35it/s]
3|3076|Loss: 0.8328743577003479:  48%|████▊     | 3075/6470 [38:10<42:02,  1.35it/s]
3|3076|Loss: 0.8328743577003479:  48%|████▊     | 3076/6470 [38:11<44:23,  1.27it/s]
3|3077|Loss: 0.9577494263648987:  48%|████▊     | 3076/6470 [38:11<44:23,  1.27it/s]
3|3077|Loss: 0.9577494263648987:  48%|████▊     | 3077/6470 [38:11<42:11,  1.34it/s]
3|3078|Loss: 0.5712230205535889:  48%|████▊     | 3077/6470 [38:12<42:11,  1.34it/s]
3|3078|Loss: 0.5712230205535889:  48%|████▊     | 3078/6470 [38:12<44:07,  1.28it/s]
3|3079|Loss: 0.9511532783508301:  48%|████▊     | 3078/6470 [38:13<44:07,  1.28it/s]
3|3079|Loss: 0.9511532783508301:  48%|████▊     | 3079/6470 [38:13<44:54,  1.26it/s]
3|3080|Loss: 0.7948296070098877:  48%|████▊     | 3079/6470 [38:14<44:54,  1.26it/s]
3|3080|Loss: 0.7948296070098877:  48%|████▊     | 3080/6470 [38:14<43:51,  1.29it/s]
3|3081|Loss: 0.8780218958854675:  48%|████▊     | 3080/6470 [38:14<43:51,  1.29it/s]
3|3081|Loss: 0.8780218958854675:  48%|████▊     | 3081/6470 [38:14<41:04,  1.38it/s]
3|3082|Loss: 0.8487244248390198:  48%|████▊     | 3081/6470 [38:15<41:04,  1.38it/s]
3|3082|Loss: 0.8487244248390198:  48%|████▊     | 3082/6470 [38:15<43:03,  1.31it/s]
3|3083|Loss: 0.6354369521141052:  48%|████▊     | 3082/6470 [38:16<43:03,  1.31it/s]
3|3083|Loss: 0.6354369521141052:  48%|████▊     | 3083/6470 [38:16<42:40,  1.32it/s]
3|3084|Loss: 0.8804249167442322:  48%|████▊     | 3083/6470 [38:16<42:40,  1.32it/s]
3|3084|Loss: 0.8804249167442322:  48%|████▊     | 3084/6470 [38:17<40:10,  1.40it/s]
3|3085|Loss: 0.896686315536499:  48%|████▊     | 3084/6470 [38:17<40:10,  1.40it/s] 
3|3085|Loss: 0.896686315536499:  48%|████▊     | 3085/6470 [38:17<40:21,  1.40it/s]
3|3086|Loss: 0.6687198877334595:  48%|████▊     | 3085/6470 [38:18<40:21,  1.40it/s]
3|3086|Loss: 0.6687198877334595:  48%|████▊     | 3086/6470 [38:18<40:08,  1.41it/s]
3|3087|Loss: 0.8225776553153992:  48%|████▊     | 3086/6470 [38:18<40:08,  1.41it/s]
3|3087|Loss: 0.8225776553153992:  48%|████▊     | 3087/6470 [38:19<41:11,  1.37it/s]
3|3088|Loss: 0.8821329474449158:  48%|████▊     | 3087/6470 [38:19<41:11,  1.37it/s]
3|3088|Loss: 0.8821329474449158:  48%|████▊     | 3088/6470 [38:20<43:31,  1.29it/s]
3|3089|Loss: 0.8070352077484131:  48%|████▊     | 3088/6470 [38:20<43:31,  1.29it/s]
3|3089|Loss: 0.8070352077484131:  48%|████▊     | 3089/6470 [38:21<44:21,  1.27it/s]
3|3090|Loss: 0.8125150203704834:  48%|████▊     | 3089/6470 [38:21<44:21,  1.27it/s]
3|3090|Loss: 0.8125150203704834:  48%|████▊     | 3090/6470 [38:21<42:50,  1.31it/s]
3|3091|Loss: 0.9316562414169312:  48%|████▊     | 3090/6470 [38:22<42:50,  1.31it/s]
3|3091|Loss: 0.9316562414169312:  48%|████▊     | 3091/6470 [38:22<43:48,  1.29it/s]
3|3092|Loss: 0.8709129691123962:  48%|████▊     | 3091/6470 [38:22<43:48,  1.29it/s]
3|3092|Loss: 0.8709129691123962:  48%|████▊     | 3092/6470 [38:23<44:03,  1.28it/s]
3|3093|Loss: 0.6759792566299438:  48%|████▊     | 3092/6470 [38:23<44:03,  1.28it/s]
3|3093|Loss: 0.6759792566299438:  48%|████▊     | 3093/6470 [38:24<45:51,  1.23it/s]
3|3094|Loss: 0.7767367959022522:  48%|████▊     | 3093/6470 [38:24<45:51,  1.23it/s]
3|3094|Loss: 0.7767367959022522:  48%|████▊     | 3094/6470 [38:25<47:27,  1.19it/s]
3|3095|Loss: 0.9073733687400818:  48%|████▊     | 3094/6470 [38:25<47:27,  1.19it/s]
3|3095|Loss: 0.9073733687400818:  48%|████▊     | 3095/6470 [38:25<46:54,  1.20it/s]
3|3096|Loss: 0.6804282069206238:  48%|████▊     | 3095/6470 [38:26<46:54,  1.20it/s]
3|3096|Loss: 0.6804282069206238:  48%|████▊     | 3096/6470 [38:26<43:30,  1.29it/s]
3|3097|Loss: 0.7254090905189514:  48%|████▊     | 3096/6470 [38:26<43:30,  1.29it/s]
3|3097|Loss: 0.7254090905189514:  48%|████▊     | 3097/6470 [38:26<37:14,  1.51it/s]
3|3098|Loss: 0.7108542323112488:  48%|████▊     | 3097/6470 [38:27<37:14,  1.51it/s]
3|3098|Loss: 0.7108542323112488:  48%|████▊     | 3098/6470 [38:27<35:04,  1.60it/s]
3|3099|Loss: 0.8659718036651611:  48%|████▊     | 3098/6470 [38:27<35:04,  1.60it/s]
3|3099|Loss: 0.8659718036651611:  48%|████▊     | 3099/6470 [38:28<38:39,  1.45it/s]
3|3100|Loss: 0.7520672678947449:  48%|████▊     | 3099/6470 [38:28<38:39,  1.45it/s]
3|3100|Loss: 0.7520672678947449:  48%|████▊     | 3100/6470 [38:28<37:26,  1.50it/s]
3|3101|Loss: 0.8813540935516357:  48%|████▊     | 3100/6470 [38:29<37:26,  1.50it/s]
3|3101|Loss: 0.8813540935516357:  48%|████▊     | 3101/6470 [38:29<35:30,  1.58it/s]
3|3102|Loss: 0.8562668561935425:  48%|████▊     | 3101/6470 [38:29<35:30,  1.58it/s]
3|3102|Loss: 0.8562668561935425:  48%|████▊     | 3102/6470 [38:30<37:47,  1.49it/s]
3|3103|Loss: 0.7999816536903381:  48%|████▊     | 3102/6470 [38:30<37:47,  1.49it/s]
3|3103|Loss: 0.7999816536903381:  48%|████▊     | 3103/6470 [38:31<39:54,  1.41it/s]
3|3104|Loss: 0.7973347306251526:  48%|████▊     | 3103/6470 [38:31<39:54,  1.41it/s]
3|3104|Loss: 0.7973347306251526:  48%|████▊     | 3104/6470 [38:31<37:05,  1.51it/s]
3|3105|Loss: 0.7393388152122498:  48%|████▊     | 3104/6470 [38:32<37:05,  1.51it/s]
3|3105|Loss: 0.7393388152122498:  48%|████▊     | 3105/6470 [38:32<38:56,  1.44it/s]
3|3106|Loss: 0.7752024531364441:  48%|████▊     | 3105/6470 [38:32<38:56,  1.44it/s]
3|3106|Loss: 0.7752024531364441:  48%|████▊     | 3106/6470 [38:33<42:45,  1.31it/s]
3|3107|Loss: 0.869124174118042:  48%|████▊     | 3106/6470 [38:33<42:45,  1.31it/s] 
3|3107|Loss: 0.869124174118042:  48%|████▊     | 3107/6470 [38:34<43:47,  1.28it/s]
3|3108|Loss: 0.7347404956817627:  48%|████▊     | 3107/6470 [38:34<43:47,  1.28it/s]
3|3108|Loss: 0.7347404956817627:  48%|████▊     | 3108/6470 [38:34<42:17,  1.32it/s]
3|3109|Loss: 0.9263898730278015:  48%|████▊     | 3108/6470 [38:35<42:17,  1.32it/s]
3|3109|Loss: 0.9263898730278015:  48%|████▊     | 3109/6470 [38:35<41:58,  1.33it/s]
3|3110|Loss: 0.8766263127326965:  48%|████▊     | 3109/6470 [38:36<41:58,  1.33it/s]
3|3110|Loss: 0.8766263127326965:  48%|████▊     | 3110/6470 [38:36<42:25,  1.32it/s]
3|3111|Loss: 0.9539373517036438:  48%|████▊     | 3110/6470 [38:36<42:25,  1.32it/s]
3|3111|Loss: 0.9539373517036438:  48%|████▊     | 3111/6470 [38:37<42:36,  1.31it/s]
3|3112|Loss: 0.7484216690063477:  48%|████▊     | 3111/6470 [38:37<42:36,  1.31it/s]
3|3112|Loss: 0.7484216690063477:  48%|████▊     | 3112/6470 [38:37<40:35,  1.38it/s]
3|3113|Loss: 0.8215298652648926:  48%|████▊     | 3112/6470 [38:38<40:35,  1.38it/s]
3|3113|Loss: 0.8215298652648926:  48%|████▊     | 3113/6470 [38:38<40:27,  1.38it/s]
3|3114|Loss: 0.7301801443099976:  48%|████▊     | 3113/6470 [38:38<40:27,  1.38it/s]
3|3114|Loss: 0.7301801443099976:  48%|████▊     | 3114/6470 [38:39<41:15,  1.36it/s]
3|3115|Loss: 0.8013787865638733:  48%|████▊     | 3114/6470 [38:39<41:15,  1.36it/s]
3|3115|Loss: 0.8013787865638733:  48%|████▊     | 3115/6470 [38:39<38:25,  1.46it/s]
3|3116|Loss: 0.7263233661651611:  48%|████▊     | 3115/6470 [38:40<38:25,  1.46it/s]
3|3116|Loss: 0.7263233661651611:  48%|████▊     | 3116/6470 [38:40<37:27,  1.49it/s]
3|3117|Loss: 0.8268646001815796:  48%|████▊     | 3116/6470 [38:40<37:27,  1.49it/s]
3|3117|Loss: 0.8268646001815796:  48%|████▊     | 3117/6470 [38:41<36:31,  1.53it/s]
3|3118|Loss: 0.9693776965141296:  48%|████▊     | 3117/6470 [38:41<36:31,  1.53it/s]
3|3118|Loss: 0.9693776965141296:  48%|████▊     | 3118/6470 [38:41<38:25,  1.45it/s]
3|3119|Loss: 0.8691603541374207:  48%|████▊     | 3118/6470 [38:42<38:25,  1.45it/s]
3|3119|Loss: 0.8691603541374207:  48%|████▊     | 3119/6470 [38:42<41:44,  1.34it/s]
3|3120|Loss: 0.7326651215553284:  48%|████▊     | 3119/6470 [38:43<41:44,  1.34it/s]
3|3120|Loss: 0.7326651215553284:  48%|████▊     | 3120/6470 [38:43<42:59,  1.30it/s]
3|3121|Loss: 0.7051562666893005:  48%|████▊     | 3120/6470 [38:43<42:59,  1.30it/s]
3|3121|Loss: 0.7051562666893005:  48%|████▊     | 3121/6470 [38:44<42:18,  1.32it/s]
3|3122|Loss: 0.9261391162872314:  48%|████▊     | 3121/6470 [38:44<42:18,  1.32it/s]
3|3122|Loss: 0.9261391162872314:  48%|████▊     | 3122/6470 [38:45<44:03,  1.27it/s]
3|3123|Loss: 0.9538434147834778:  48%|████▊     | 3122/6470 [38:45<44:03,  1.27it/s]
3|3123|Loss: 0.9538434147834778:  48%|████▊     | 3123/6470 [38:45<41:41,  1.34it/s]
3|3124|Loss: 0.9572749733924866:  48%|████▊     | 3123/6470 [38:46<41:41,  1.34it/s]
3|3124|Loss: 0.9572749733924866:  48%|████▊     | 3124/6470 [38:46<39:30,  1.41it/s]
3|3125|Loss: 0.9887599945068359:  48%|████▊     | 3124/6470 [38:46<39:30,  1.41it/s]
3|3125|Loss: 0.9887599945068359:  48%|████▊     | 3125/6470 [38:47<38:13,  1.46it/s]
3|3126|Loss: 0.9437288045883179:  48%|████▊     | 3125/6470 [38:47<38:13,  1.46it/s]
3|3126|Loss: 0.9437288045883179:  48%|████▊     | 3126/6470 [38:47<35:17,  1.58it/s]
3|3127|Loss: 0.7278582453727722:  48%|████▊     | 3126/6470 [38:47<35:17,  1.58it/s]
3|3127|Loss: 0.7278582453727722:  48%|████▊     | 3127/6470 [38:48<36:50,  1.51it/s]
3|3128|Loss: 0.7392895221710205:  48%|████▊     | 3127/6470 [38:48<36:50,  1.51it/s]
3|3128|Loss: 0.7392895221710205:  48%|████▊     | 3128/6470 [38:48<36:32,  1.52it/s]
3|3129|Loss: 0.7780855298042297:  48%|████▊     | 3128/6470 [38:49<36:32,  1.52it/s]
3|3129|Loss: 0.7780855298042297:  48%|████▊     | 3129/6470 [38:49<37:36,  1.48it/s]
3|3130|Loss: 0.7973144054412842:  48%|████▊     | 3129/6470 [38:50<37:36,  1.48it/s]
3|3130|Loss: 0.7973144054412842:  48%|████▊     | 3130/6470 [38:50<38:11,  1.46it/s]
3|3131|Loss: 0.9166058301925659:  48%|████▊     | 3130/6470 [38:50<38:11,  1.46it/s]
3|3131|Loss: 0.9166058301925659:  48%|████▊     | 3131/6470 [38:51<40:41,  1.37it/s]
3|3132|Loss: 0.8927844166755676:  48%|████▊     | 3131/6470 [38:51<40:41,  1.37it/s]
3|3132|Loss: 0.8927844166755676:  48%|████▊     | 3132/6470 [38:52<42:21,  1.31it/s]
3|3133|Loss: 0.6720585823059082:  48%|████▊     | 3132/6470 [38:52<42:21,  1.31it/s]
3|3133|Loss: 0.6720585823059082:  48%|████▊     | 3133/6470 [38:52<39:53,  1.39it/s]
3|3134|Loss: 0.943173885345459:  48%|████▊     | 3133/6470 [38:53<39:53,  1.39it/s] 
3|3134|Loss: 0.943173885345459:  48%|████▊     | 3134/6470 [38:53<41:39,  1.33it/s]
3|3135|Loss: 0.7572436928749084:  48%|████▊     | 3134/6470 [38:53<41:39,  1.33it/s]
3|3135|Loss: 0.7572436928749084:  48%|████▊     | 3135/6470 [38:53<37:08,  1.50it/s]
3|3136|Loss: 0.9382984042167664:  48%|████▊     | 3135/6470 [38:54<37:08,  1.50it/s]
3|3136|Loss: 0.9382984042167664:  48%|████▊     | 3136/6470 [38:54<38:01,  1.46it/s]
3|3137|Loss: 0.9349256157875061:  48%|████▊     | 3136/6470 [38:55<38:01,  1.46it/s]
3|3137|Loss: 0.9349256157875061:  48%|████▊     | 3137/6470 [38:55<41:38,  1.33it/s]
3|3138|Loss: 0.8293073773384094:  48%|████▊     | 3137/6470 [38:55<41:38,  1.33it/s]
3|3138|Loss: 0.8293073773384094:  49%|████▊     | 3138/6470 [38:56<40:31,  1.37it/s]
3|3139|Loss: 0.8476981520652771:  49%|████▊     | 3138/6470 [38:56<40:31,  1.37it/s]
3|3139|Loss: 0.8476981520652771:  49%|████▊     | 3139/6470 [38:57<42:43,  1.30it/s]
3|3140|Loss: 0.8525344729423523:  49%|████▊     | 3139/6470 [38:57<42:43,  1.30it/s]
3|3140|Loss: 0.8525344729423523:  49%|████▊     | 3140/6470 [38:57<38:06,  1.46it/s]
3|3141|Loss: 0.6201894283294678:  49%|████▊     | 3140/6470 [38:57<38:06,  1.46it/s]
3|3141|Loss: 0.6201894283294678:  49%|████▊     | 3141/6470 [38:58<37:47,  1.47it/s]
3|3142|Loss: 0.8881779909133911:  49%|████▊     | 3141/6470 [38:58<37:47,  1.47it/s]
3|3142|Loss: 0.8881779909133911:  49%|████▊     | 3142/6470 [38:59<39:28,  1.41it/s]
3|3143|Loss: 0.9323024153709412:  49%|████▊     | 3142/6470 [38:59<39:28,  1.41it/s]
3|3143|Loss: 0.9323024153709412:  49%|████▊     | 3143/6470 [38:59<39:06,  1.42it/s]
3|3144|Loss: 0.8981106281280518:  49%|████▊     | 3143/6470 [39:00<39:06,  1.42it/s]
3|3144|Loss: 0.8981106281280518:  49%|████▊     | 3144/6470 [39:00<40:05,  1.38it/s]
3|3145|Loss: 1.0101385116577148:  49%|████▊     | 3144/6470 [39:00<40:05,  1.38it/s]
3|3145|Loss: 1.0101385116577148:  49%|████▊     | 3145/6470 [39:01<41:21,  1.34it/s]
3|3146|Loss: 0.9395642280578613:  49%|████▊     | 3145/6470 [39:01<41:21,  1.34it/s]
3|3146|Loss: 0.9395642280578613:  49%|████▊     | 3146/6470 [39:02<43:34,  1.27it/s]
3|3147|Loss: 0.8637422323226929:  49%|████▊     | 3146/6470 [39:02<43:34,  1.27it/s]
3|3147|Loss: 0.8637422323226929:  49%|████▊     | 3147/6470 [39:03<46:00,  1.20it/s]
3|3148|Loss: 0.9523018598556519:  49%|████▊     | 3147/6470 [39:03<46:00,  1.20it/s]
3|3148|Loss: 0.9523018598556519:  49%|████▊     | 3148/6470 [39:03<44:00,  1.26it/s]
3|3149|Loss: 0.8670589923858643:  49%|████▊     | 3148/6470 [39:04<44:00,  1.26it/s]
3|3149|Loss: 0.8670589923858643:  49%|████▊     | 3149/6470 [39:04<40:59,  1.35it/s]
3|3150|Loss: 0.681978702545166:  49%|████▊     | 3149/6470 [39:04<40:59,  1.35it/s] 
3|3150|Loss: 0.681978702545166:  49%|████▊     | 3150/6470 [39:05<43:13,  1.28it/s]
3|3151|Loss: 0.8927017450332642:  49%|████▊     | 3150/6470 [39:05<43:13,  1.28it/s]
3|3151|Loss: 0.8927017450332642:  49%|████▊     | 3151/6470 [39:06<46:16,  1.20it/s]
3|3152|Loss: 0.9602757096290588:  49%|████▊     | 3151/6470 [39:06<46:16,  1.20it/s]
3|3152|Loss: 0.9602757096290588:  49%|████▊     | 3152/6470 [39:06<43:42,  1.27it/s]
3|3153|Loss: 0.7076303958892822:  49%|████▊     | 3152/6470 [39:07<43:42,  1.27it/s]
3|3153|Loss: 0.7076303958892822:  49%|████▊     | 3153/6470 [39:07<38:47,  1.43it/s]
3|3154|Loss: 0.7250227928161621:  49%|████▊     | 3153/6470 [39:07<38:47,  1.43it/s]
3|3154|Loss: 0.7250227928161621:  49%|████▊     | 3154/6470 [39:08<37:44,  1.46it/s]
3|3155|Loss: 0.7636613845825195:  49%|████▊     | 3154/6470 [39:08<37:44,  1.46it/s]
3|3155|Loss: 0.7636613845825195:  49%|████▉     | 3155/6470 [39:08<35:06,  1.57it/s]
3|3156|Loss: 0.8270198106765747:  49%|████▉     | 3155/6470 [39:09<35:06,  1.57it/s]
3|3156|Loss: 0.8270198106765747:  49%|████▉     | 3156/6470 [39:09<35:58,  1.54it/s]
3|3157|Loss: 0.8392159938812256:  49%|████▉     | 3156/6470 [39:09<35:58,  1.54it/s]
3|3157|Loss: 0.8392159938812256:  49%|████▉     | 3157/6470 [39:10<40:08,  1.38it/s]
3|3158|Loss: 0.9841899275779724:  49%|████▉     | 3157/6470 [39:10<40:08,  1.38it/s]
3|3158|Loss: 0.9841899275779724:  49%|████▉     | 3158/6470 [39:10<40:13,  1.37it/s]
3|3159|Loss: 0.9399970173835754:  49%|████▉     | 3158/6470 [39:11<40:13,  1.37it/s]
3|3159|Loss: 0.9399970173835754:  49%|████▉     | 3159/6470 [39:11<40:51,  1.35it/s]
3|3160|Loss: 0.8122387528419495:  49%|████▉     | 3159/6470 [39:12<40:51,  1.35it/s]
3|3160|Loss: 0.8122387528419495:  49%|████▉     | 3160/6470 [39:12<39:54,  1.38it/s]
3|3161|Loss: 0.8388932943344116:  49%|████▉     | 3160/6470 [39:12<39:54,  1.38it/s]
3|3161|Loss: 0.8388932943344116:  49%|████▉     | 3161/6470 [39:13<39:15,  1.40it/s]
3|3162|Loss: 0.9971065521240234:  49%|████▉     | 3161/6470 [39:13<39:15,  1.40it/s]
3|3162|Loss: 0.9971065521240234:  49%|████▉     | 3162/6470 [39:13<40:29,  1.36it/s]
3|3163|Loss: 0.8088505268096924:  49%|████▉     | 3162/6470 [39:14<40:29,  1.36it/s]
3|3163|Loss: 0.8088505268096924:  49%|████▉     | 3163/6470 [39:14<42:27,  1.30it/s]
3|3164|Loss: 0.9289044737815857:  49%|████▉     | 3163/6470 [39:15<42:27,  1.30it/s]
3|3164|Loss: 0.9289044737815857:  49%|████▉     | 3164/6470 [39:15<44:29,  1.24it/s]
3|3165|Loss: 0.88560551404953:  49%|████▉     | 3164/6470 [39:16<44:29,  1.24it/s]  
3|3165|Loss: 0.88560551404953:  49%|████▉     | 3165/6470 [39:16<46:35,  1.18it/s]
3|3166|Loss: 0.7368850708007812:  49%|████▉     | 3165/6470 [39:17<46:35,  1.18it/s]
3|3166|Loss: 0.7368850708007812:  49%|████▉     | 3166/6470 [39:17<46:19,  1.19it/s]
3|3167|Loss: 0.8194023966789246:  49%|████▉     | 3166/6470 [39:17<46:19,  1.19it/s]
3|3167|Loss: 0.8194023966789246:  49%|████▉     | 3167/6470 [39:18<45:55,  1.20it/s]
3|3168|Loss: 0.9549254775047302:  49%|████▉     | 3167/6470 [39:18<45:55,  1.20it/s]
3|3168|Loss: 0.9549254775047302:  49%|████▉     | 3168/6470 [39:19<45:31,  1.21it/s]
3|3169|Loss: 0.9060593843460083:  49%|████▉     | 3168/6470 [39:19<45:31,  1.21it/s]
3|3169|Loss: 0.9060593843460083:  49%|████▉     | 3169/6470 [39:19<46:41,  1.18it/s]
3|3170|Loss: 0.7632287740707397:  49%|████▉     | 3169/6470 [39:20<46:41,  1.18it/s]
3|3170|Loss: 0.7632287740707397:  49%|████▉     | 3170/6470 [39:20<43:56,  1.25it/s]
3|3171|Loss: 0.8785242438316345:  49%|████▉     | 3170/6470 [39:20<43:56,  1.25it/s]
3|3171|Loss: 0.8785242438316345:  49%|████▉     | 3171/6470 [39:21<41:31,  1.32it/s]
3|3172|Loss: 0.629895031452179:  49%|████▉     | 3171/6470 [39:21<41:31,  1.32it/s] 
3|3172|Loss: 0.629895031452179:  49%|████▉     | 3172/6470 [39:21<35:37,  1.54it/s]
3|3173|Loss: 0.8459107279777527:  49%|████▉     | 3172/6470 [39:21<35:37,  1.54it/s]
3|3173|Loss: 0.8459107279777527:  49%|████▉     | 3173/6470 [39:22<32:13,  1.70it/s]
3|3174|Loss: 0.7190628051757812:  49%|████▉     | 3173/6470 [39:22<32:13,  1.70it/s]
3|3174|Loss: 0.7190628051757812:  49%|████▉     | 3174/6470 [39:22<33:08,  1.66it/s]
3|3175|Loss: 0.9150280952453613:  49%|████▉     | 3174/6470 [39:23<33:08,  1.66it/s]
3|3175|Loss: 0.9150280952453613:  49%|████▉     | 3175/6470 [39:23<36:15,  1.51it/s]
3|3176|Loss: 0.8084403276443481:  49%|████▉     | 3175/6470 [39:23<36:15,  1.51it/s]
3|3176|Loss: 0.8084403276443481:  49%|████▉     | 3176/6470 [39:24<37:02,  1.48it/s]
3|3177|Loss: 0.9043676853179932:  49%|████▉     | 3176/6470 [39:24<37:02,  1.48it/s]
3|3177|Loss: 0.9043676853179932:  49%|████▉     | 3177/6470 [39:25<39:20,  1.40it/s]
3|3178|Loss: 0.8120365142822266:  49%|████▉     | 3177/6470 [39:25<39:20,  1.40it/s]
3|3178|Loss: 0.8120365142822266:  49%|████▉     | 3178/6470 [39:25<41:40,  1.32it/s]
3|3179|Loss: 0.6244868636131287:  49%|████▉     | 3178/6470 [39:26<41:40,  1.32it/s]
3|3179|Loss: 0.6244868636131287:  49%|████▉     | 3179/6470 [39:26<40:45,  1.35it/s]
3|3180|Loss: 0.8537463545799255:  49%|████▉     | 3179/6470 [39:27<40:45,  1.35it/s]
3|3180|Loss: 0.8537463545799255:  49%|████▉     | 3180/6470 [39:27<41:28,  1.32it/s]
3|3181|Loss: 0.827794075012207:  49%|████▉     | 3180/6470 [39:27<41:28,  1.32it/s] 
3|3181|Loss: 0.827794075012207:  49%|████▉     | 3181/6470 [39:28<41:39,  1.32it/s]
3|3182|Loss: 0.9163190126419067:  49%|████▉     | 3181/6470 [39:28<41:39,  1.32it/s]
3|3182|Loss: 0.9163190126419067:  49%|████▉     | 3182/6470 [39:28<40:56,  1.34it/s]
3|3183|Loss: 0.799317479133606:  49%|████▉     | 3182/6470 [39:29<40:56,  1.34it/s] 
3|3183|Loss: 0.799317479133606:  49%|████▉     | 3183/6470 [39:29<39:22,  1.39it/s]
3|3184|Loss: 0.9866538047790527:  49%|████▉     | 3183/6470 [39:29<39:22,  1.39it/s]
3|3184|Loss: 0.9866538047790527:  49%|████▉     | 3184/6470 [39:30<38:31,  1.42it/s]
3|3185|Loss: 0.9245837330818176:  49%|████▉     | 3184/6470 [39:30<38:31,  1.42it/s]
3|3185|Loss: 0.9245837330818176:  49%|████▉     | 3185/6470 [39:31<39:53,  1.37it/s]
3|3186|Loss: 0.823708713054657:  49%|████▉     | 3185/6470 [39:31<39:53,  1.37it/s] 
3|3186|Loss: 0.823708713054657:  49%|████▉     | 3186/6470 [39:31<41:03,  1.33it/s]
3|3187|Loss: 0.7619884014129639:  49%|████▉     | 3186/6470 [39:32<41:03,  1.33it/s]
3|3187|Loss: 0.7619884014129639:  49%|████▉     | 3187/6470 [39:32<43:37,  1.25it/s]
3|3188|Loss: 0.738128662109375:  49%|████▉     | 3187/6470 [39:33<43:37,  1.25it/s] 
3|3188|Loss: 0.738128662109375:  49%|████▉     | 3188/6470 [39:33<44:02,  1.24it/s]
3|3189|Loss: 0.7991511821746826:  49%|████▉     | 3188/6470 [39:33<44:02,  1.24it/s]
3|3189|Loss: 0.7991511821746826:  49%|████▉     | 3189/6470 [39:34<43:23,  1.26it/s]
3|3190|Loss: 0.8257977962493896:  49%|████▉     | 3189/6470 [39:34<43:23,  1.26it/s]
3|3190|Loss: 0.8257977962493896:  49%|████▉     | 3190/6470 [39:35<43:16,  1.26it/s]
3|3191|Loss: 0.9684118628501892:  49%|████▉     | 3190/6470 [39:35<43:16,  1.26it/s]
3|3191|Loss: 0.9684118628501892:  49%|████▉     | 3191/6470 [39:35<40:56,  1.33it/s]
3|3192|Loss: 0.797405481338501:  49%|████▉     | 3191/6470 [39:36<40:56,  1.33it/s] 
3|3192|Loss: 0.797405481338501:  49%|████▉     | 3192/6470 [39:36<39:19,  1.39it/s]
3|3193|Loss: 0.8215846419334412:  49%|████▉     | 3192/6470 [39:36<39:19,  1.39it/s]
3|3193|Loss: 0.8215846419334412:  49%|████▉     | 3193/6470 [39:37<41:31,  1.32it/s]
3|3194|Loss: 0.7675508260726929:  49%|████▉     | 3193/6470 [39:37<41:31,  1.32it/s]
3|3194|Loss: 0.7675508260726929:  49%|████▉     | 3194/6470 [39:37<40:21,  1.35it/s]
3|3195|Loss: 0.8034641742706299:  49%|████▉     | 3194/6470 [39:38<40:21,  1.35it/s]
3|3195|Loss: 0.8034641742706299:  49%|████▉     | 3195/6470 [39:38<38:27,  1.42it/s]
3|3196|Loss: 0.7659682631492615:  49%|████▉     | 3195/6470 [39:38<38:27,  1.42it/s]
3|3196|Loss: 0.7659682631492615:  49%|████▉     | 3196/6470 [39:39<38:02,  1.43it/s]
3|3197|Loss: 0.9445944428443909:  49%|████▉     | 3196/6470 [39:39<38:02,  1.43it/s]
3|3197|Loss: 0.9445944428443909:  49%|████▉     | 3197/6470 [39:39<38:04,  1.43it/s]
3|3198|Loss: 0.9145849943161011:  49%|████▉     | 3197/6470 [39:40<38:04,  1.43it/s]
3|3198|Loss: 0.9145849943161011:  49%|████▉     | 3198/6470 [39:40<37:03,  1.47it/s]
3|3199|Loss: 0.8424063324928284:  49%|████▉     | 3198/6470 [39:40<37:03,  1.47it/s]
3|3199|Loss: 0.8424063324928284:  49%|████▉     | 3199/6470 [39:41<36:02,  1.51it/s]
3|3200|Loss: 0.8428238034248352:  49%|████▉     | 3199/6470 [39:41<36:02,  1.51it/s]
3|3200|Loss: 0.8428238034248352:  49%|████▉     | 3200/6470 [39:41<36:26,  1.50it/s]
3|3201|Loss: 0.971460223197937:  49%|████▉     | 3200/6470 [39:42<36:26,  1.50it/s] 
3|3201|Loss: 0.971460223197937:  49%|████▉     | 3201/6470 [39:42<37:56,  1.44it/s]
3|3202|Loss: 0.7234863042831421:  49%|████▉     | 3201/6470 [39:43<37:56,  1.44it/s]
3|3202|Loss: 0.7234863042831421:  49%|████▉     | 3202/6470 [39:43<37:47,  1.44it/s]
3|3203|Loss: 0.8119733333587646:  49%|████▉     | 3202/6470 [39:43<37:47,  1.44it/s]
3|3203|Loss: 0.8119733333587646:  50%|████▉     | 3203/6470 [39:44<38:53,  1.40it/s]
3|3204|Loss: 0.9309059381484985:  50%|████▉     | 3203/6470 [39:44<38:53,  1.40it/s]
3|3204|Loss: 0.9309059381484985:  50%|████▉     | 3204/6470 [39:44<38:49,  1.40it/s]
3|3205|Loss: 0.8368903398513794:  50%|████▉     | 3204/6470 [39:45<38:49,  1.40it/s]
3|3205|Loss: 0.8368903398513794:  50%|████▉     | 3205/6470 [39:45<39:20,  1.38it/s]
3|3206|Loss: 0.8598042130470276:  50%|████▉     | 3205/6470 [39:45<39:20,  1.38it/s]
3|3206|Loss: 0.8598042130470276:  50%|████▉     | 3206/6470 [39:46<40:07,  1.36it/s]
3|3207|Loss: 0.7901463508605957:  50%|████▉     | 3206/6470 [39:46<40:07,  1.36it/s]
3|3207|Loss: 0.7901463508605957:  50%|████▉     | 3207/6470 [39:47<40:54,  1.33it/s]
3|3208|Loss: 0.8355898261070251:  50%|████▉     | 3207/6470 [39:47<40:54,  1.33it/s]
3|3208|Loss: 0.8355898261070251:  50%|████▉     | 3208/6470 [39:47<40:09,  1.35it/s]
3|3209|Loss: 0.8470624685287476:  50%|████▉     | 3208/6470 [39:48<40:09,  1.35it/s]
3|3209|Loss: 0.8470624685287476:  50%|████▉     | 3209/6470 [39:48<41:32,  1.31it/s]
3|3210|Loss: 0.6367263197898865:  50%|████▉     | 3209/6470 [39:49<41:32,  1.31it/s]
3|3210|Loss: 0.6367263197898865:  50%|████▉     | 3210/6470 [39:49<43:30,  1.25it/s]
3|3211|Loss: 0.9671388268470764:  50%|████▉     | 3210/6470 [39:49<43:30,  1.25it/s]
3|3211|Loss: 0.9671388268470764:  50%|████▉     | 3211/6470 [39:50<41:52,  1.30it/s]
3|3212|Loss: 0.9410297274589539:  50%|████▉     | 3211/6470 [39:50<41:52,  1.30it/s]
3|3212|Loss: 0.9410297274589539:  50%|████▉     | 3212/6470 [39:51<41:49,  1.30it/s]
3|3213|Loss: 0.8757264614105225:  50%|████▉     | 3212/6470 [39:51<41:49,  1.30it/s]
3|3213|Loss: 0.8757264614105225:  50%|████▉     | 3213/6470 [39:51<41:59,  1.29it/s]
3|3214|Loss: 0.9374498128890991:  50%|████▉     | 3213/6470 [39:52<41:59,  1.29it/s]
3|3214|Loss: 0.9374498128890991:  50%|████▉     | 3214/6470 [39:52<41:06,  1.32it/s]
3|3215|Loss: 0.7455909252166748:  50%|████▉     | 3214/6470 [39:52<41:06,  1.32it/s]
3|3215|Loss: 0.7455909252166748:  50%|████▉     | 3215/6470 [39:53<39:22,  1.38it/s]
3|3216|Loss: 0.8158692121505737:  50%|████▉     | 3215/6470 [39:53<39:22,  1.38it/s]
3|3216|Loss: 0.8158692121505737:  50%|████▉     | 3216/6470 [39:53<38:08,  1.42it/s]
3|3217|Loss: 0.7871413230895996:  50%|████▉     | 3216/6470 [39:54<38:08,  1.42it/s]
3|3217|Loss: 0.7871413230895996:  50%|████▉     | 3217/6470 [39:54<39:16,  1.38it/s]
3|3218|Loss: 0.8576236367225647:  50%|████▉     | 3217/6470 [39:54<39:16,  1.38it/s]
3|3218|Loss: 0.8576236367225647:  50%|████▉     | 3218/6470 [39:55<37:31,  1.44it/s]
3|3219|Loss: 0.8060213327407837:  50%|████▉     | 3218/6470 [39:55<37:31,  1.44it/s]
3|3219|Loss: 0.8060213327407837:  50%|████▉     | 3219/6470 [39:55<35:18,  1.53it/s]
3|3220|Loss: 1.0371315479278564:  50%|████▉     | 3219/6470 [39:56<35:18,  1.53it/s]
3|3220|Loss: 1.0371315479278564:  50%|████▉     | 3220/6470 [39:56<34:24,  1.57it/s]
3|3221|Loss: 1.2768449783325195:  50%|████▉     | 3220/6470 [39:56<34:24,  1.57it/s]
3|3221|Loss: 1.2768449783325195:  50%|████▉     | 3221/6470 [39:57<44:27,  1.22it/s]
3|3222|Loss: 0.7340055108070374:  50%|████▉     | 3221/6470 [39:58<44:27,  1.22it/s]
3|3222|Loss: 0.7340055108070374:  50%|████▉     | 3222/6470 [39:58<44:25,  1.22it/s]
3|3223|Loss: 0.8902772068977356:  50%|████▉     | 3222/6470 [39:58<44:25,  1.22it/s]
3|3223|Loss: 0.8902772068977356:  50%|████▉     | 3223/6470 [39:59<45:37,  1.19it/s]
3|3224|Loss: 0.8024658560752869:  50%|████▉     | 3223/6470 [39:59<45:37,  1.19it/s]
3|3224|Loss: 0.8024658560752869:  50%|████▉     | 3224/6470 [40:00<43:36,  1.24it/s]
3|3225|Loss: 0.895614743232727:  50%|████▉     | 3224/6470 [40:00<43:36,  1.24it/s] 
3|3225|Loss: 0.895614743232727:  50%|████▉     | 3225/6470 [40:00<43:03,  1.26it/s]
3|3226|Loss: 0.8279063701629639:  50%|████▉     | 3225/6470 [40:01<43:03,  1.26it/s]
3|3226|Loss: 0.8279063701629639:  50%|████▉     | 3226/6470 [40:01<43:50,  1.23it/s]
3|3227|Loss: 0.7999746799468994:  50%|████▉     | 3226/6470 [40:02<43:50,  1.23it/s]
3|3227|Loss: 0.7999746799468994:  50%|████▉     | 3227/6470 [40:02<44:58,  1.20it/s]
3|3228|Loss: 0.6781503558158875:  50%|████▉     | 3227/6470 [40:03<44:58,  1.20it/s]
3|3228|Loss: 0.6781503558158875:  50%|████▉     | 3228/6470 [40:03<48:13,  1.12it/s]
3|3229|Loss: 0.7901473641395569:  50%|████▉     | 3228/6470 [40:04<48:13,  1.12it/s]
3|3229|Loss: 0.7901473641395569:  50%|████▉     | 3229/6470 [40:04<49:02,  1.10it/s]
3|3230|Loss: 0.8447547554969788:  50%|████▉     | 3229/6470 [40:05<49:02,  1.10it/s]
3|3230|Loss: 0.8447547554969788:  50%|████▉     | 3230/6470 [40:05<49:26,  1.09it/s]
3|3231|Loss: 0.8401411771774292:  50%|████▉     | 3230/6470 [40:06<49:26,  1.09it/s]
3|3231|Loss: 0.8401411771774292:  50%|████▉     | 3231/6470 [40:06<50:42,  1.06it/s]
3|3232|Loss: 0.8726697564125061:  50%|████▉     | 3231/6470 [40:06<50:42,  1.06it/s]
3|3232|Loss: 0.8726697564125061:  50%|████▉     | 3232/6470 [40:07<49:17,  1.09it/s]
3|3233|Loss: 0.7794221043586731:  50%|████▉     | 3232/6470 [40:07<49:17,  1.09it/s]
3|3233|Loss: 0.7794221043586731:  50%|████▉     | 3233/6470 [40:08<47:49,  1.13it/s]
3|3234|Loss: 0.8277788758277893:  50%|████▉     | 3233/6470 [40:08<47:49,  1.13it/s]
3|3234|Loss: 0.8277788758277893:  50%|████▉     | 3234/6470 [40:09<48:05,  1.12it/s]
3|3235|Loss: 0.7752741575241089:  50%|████▉     | 3234/6470 [40:09<48:05,  1.12it/s]
3|3235|Loss: 0.7752741575241089:  50%|█████     | 3235/6470 [40:10<51:31,  1.05it/s]
3|3236|Loss: 0.6936299204826355:  50%|█████     | 3235/6470 [40:10<51:31,  1.05it/s]
3|3236|Loss: 0.6936299204826355:  50%|█████     | 3236/6470 [40:11<52:52,  1.02it/s]
3|3237|Loss: 0.8751323223114014:  50%|█████     | 3236/6470 [40:11<52:52,  1.02it/s]
3|3237|Loss: 0.8751323223114014:  50%|█████     | 3237/6470 [40:12<50:36,  1.06it/s]
3|3238|Loss: 0.7765755653381348:  50%|█████     | 3237/6470 [40:12<50:36,  1.06it/s]
3|3238|Loss: 0.7765755653381348:  50%|█████     | 3238/6470 [40:12<45:10,  1.19it/s]
3|3239|Loss: 0.9019271731376648:  50%|█████     | 3238/6470 [40:12<45:10,  1.19it/s]
3|3239|Loss: 0.9019271731376648:  50%|█████     | 3239/6470 [40:13<43:27,  1.24it/s]
3|3240|Loss: 0.8578989505767822:  50%|█████     | 3239/6470 [40:13<43:27,  1.24it/s]
3|3240|Loss: 0.8578989505767822:  50%|█████     | 3240/6470 [40:14<41:56,  1.28it/s]
3|3241|Loss: 0.7740685343742371:  50%|█████     | 3240/6470 [40:14<41:56,  1.28it/s]
3|3241|Loss: 0.7740685343742371:  50%|█████     | 3241/6470 [40:14<40:34,  1.33it/s]
3|3242|Loss: 0.8156537413597107:  50%|█████     | 3241/6470 [40:15<40:34,  1.33it/s]
3|3242|Loss: 0.8156537413597107:  50%|█████     | 3242/6470 [40:15<40:50,  1.32it/s]
3|3243|Loss: 0.8503381609916687:  50%|█████     | 3242/6470 [40:15<40:50,  1.32it/s]
3|3243|Loss: 0.8503381609916687:  50%|█████     | 3243/6470 [40:16<40:14,  1.34it/s]
3|3244|Loss: 0.8110959529876709:  50%|█████     | 3243/6470 [40:16<40:14,  1.34it/s]
3|3244|Loss: 0.8110959529876709:  50%|█████     | 3244/6470 [40:16<38:39,  1.39it/s]
3|3245|Loss: 0.8494742512702942:  50%|█████     | 3244/6470 [40:17<38:39,  1.39it/s]
3|3245|Loss: 0.8494742512702942:  50%|█████     | 3245/6470 [40:17<38:37,  1.39it/s]
3|3246|Loss: 0.8516478538513184:  50%|█████     | 3245/6470 [40:18<38:37,  1.39it/s]
3|3246|Loss: 0.8516478538513184:  50%|█████     | 3246/6470 [40:18<39:05,  1.37it/s]
3|3247|Loss: 0.7762489318847656:  50%|█████     | 3246/6470 [40:18<39:05,  1.37it/s]
3|3247|Loss: 0.7762489318847656:  50%|█████     | 3247/6470 [40:19<41:24,  1.30it/s]
3|3248|Loss: 0.8450687527656555:  50%|█████     | 3247/6470 [40:19<41:24,  1.30it/s]
3|3248|Loss: 0.8450687527656555:  50%|█████     | 3248/6470 [40:19<40:16,  1.33it/s]
3|3249|Loss: 0.7933967113494873:  50%|█████     | 3248/6470 [40:20<40:16,  1.33it/s]
3|3249|Loss: 0.7933967113494873:  50%|█████     | 3249/6470 [40:20<40:27,  1.33it/s]
3|3250|Loss: 0.8336166739463806:  50%|█████     | 3249/6470 [40:21<40:27,  1.33it/s]
3|3250|Loss: 0.8336166739463806:  50%|█████     | 3250/6470 [40:21<39:50,  1.35it/s]
3|3251|Loss: 0.7846452593803406:  50%|█████     | 3250/6470 [40:21<39:50,  1.35it/s]
3|3251|Loss: 0.7846452593803406:  50%|█████     | 3251/6470 [40:22<41:57,  1.28it/s]
3|3252|Loss: 0.7960168123245239:  50%|█████     | 3251/6470 [40:22<41:57,  1.28it/s]
3|3252|Loss: 0.7960168123245239:  50%|█████     | 3252/6470 [40:23<42:47,  1.25it/s]
3|3253|Loss: 0.8858206272125244:  50%|█████     | 3252/6470 [40:23<42:47,  1.25it/s]
3|3253|Loss: 0.8858206272125244:  50%|█████     | 3253/6470 [40:23<42:32,  1.26it/s]
3|3254|Loss: 0.7591387629508972:  50%|█████     | 3253/6470 [40:24<42:32,  1.26it/s]
3|3254|Loss: 0.7591387629508972:  50%|█████     | 3254/6470 [40:24<41:48,  1.28it/s]
3|3255|Loss: 0.7857953310012817:  50%|█████     | 3254/6470 [40:25<41:48,  1.28it/s]
3|3255|Loss: 0.7857953310012817:  50%|█████     | 3255/6470 [40:25<38:47,  1.38it/s]
3|3256|Loss: 0.7744593024253845:  50%|█████     | 3255/6470 [40:25<38:47,  1.38it/s]
3|3256|Loss: 0.7744593024253845:  50%|█████     | 3256/6470 [40:25<35:50,  1.49it/s]
3|3257|Loss: 0.8006837964057922:  50%|█████     | 3256/6470 [40:26<35:50,  1.49it/s]
3|3257|Loss: 0.8006837964057922:  50%|█████     | 3257/6470 [40:26<35:41,  1.50it/s]
3|3258|Loss: 0.9114835262298584:  50%|█████     | 3257/6470 [40:26<35:41,  1.50it/s]
3|3258|Loss: 0.9114835262298584:  50%|█████     | 3258/6470 [40:27<35:18,  1.52it/s]
3|3259|Loss: 0.7510035634040833:  50%|█████     | 3258/6470 [40:27<35:18,  1.52it/s]
3|3259|Loss: 0.7510035634040833:  50%|█████     | 3259/6470 [40:27<37:07,  1.44it/s]
3|3260|Loss: 0.8511345386505127:  50%|█████     | 3259/6470 [40:28<37:07,  1.44it/s]
3|3260|Loss: 0.8511345386505127:  50%|█████     | 3260/6470 [40:28<38:32,  1.39it/s]
3|3261|Loss: 0.7859987020492554:  50%|█████     | 3260/6470 [40:29<38:32,  1.39it/s]
3|3261|Loss: 0.7859987020492554:  50%|█████     | 3261/6470 [40:29<41:10,  1.30it/s]
3|3262|Loss: 0.7572822570800781:  50%|█████     | 3261/6470 [40:29<41:10,  1.30it/s]
3|3262|Loss: 0.7572822570800781:  50%|█████     | 3262/6470 [40:30<38:33,  1.39it/s]
3|3263|Loss: 0.8751943707466125:  50%|█████     | 3262/6470 [40:30<38:33,  1.39it/s]
3|3263|Loss: 0.8751943707466125:  50%|█████     | 3263/6470 [40:31<40:41,  1.31it/s]
3|3264|Loss: 0.865311861038208:  50%|█████     | 3263/6470 [40:31<40:41,  1.31it/s] 
3|3264|Loss: 0.865311861038208:  50%|█████     | 3264/6470 [40:31<38:46,  1.38it/s]
3|3265|Loss: 0.7732623815536499:  50%|█████     | 3264/6470 [40:32<38:46,  1.38it/s]
3|3265|Loss: 0.7732623815536499:  50%|█████     | 3265/6470 [40:32<40:04,  1.33it/s]
3|3266|Loss: 0.7707446813583374:  50%|█████     | 3265/6470 [40:32<40:04,  1.33it/s]
3|3266|Loss: 0.7707446813583374:  50%|█████     | 3266/6470 [40:33<40:07,  1.33it/s]
3|3267|Loss: 0.8200161457061768:  50%|█████     | 3266/6470 [40:34<40:07,  1.33it/s]
3|3267|Loss: 0.8200161457061768:  50%|█████     | 3267/6470 [40:34<50:41,  1.05it/s]
3|3268|Loss: 0.8064661622047424:  50%|█████     | 3267/6470 [40:35<50:41,  1.05it/s]
3|3268|Loss: 0.8064661622047424:  51%|█████     | 3268/6470 [40:35<49:34,  1.08it/s]
3|3269|Loss: 0.9733167290687561:  51%|█████     | 3268/6470 [40:35<49:34,  1.08it/s]
3|3269|Loss: 0.9733167290687561:  51%|█████     | 3269/6470 [40:36<47:31,  1.12it/s]
3|3270|Loss: 0.9606366753578186:  51%|█████     | 3269/6470 [40:36<47:31,  1.12it/s]
3|3270|Loss: 0.9606366753578186:  51%|█████     | 3270/6470 [40:37<47:36,  1.12it/s]
3|3271|Loss: 0.8867939710617065:  51%|█████     | 3270/6470 [40:37<47:36,  1.12it/s]
3|3271|Loss: 0.8867939710617065:  51%|█████     | 3271/6470 [40:38<46:34,  1.14it/s]
3|3272|Loss: 0.9804896116256714:  51%|█████     | 3271/6470 [40:38<46:34,  1.14it/s]
3|3272|Loss: 0.9804896116256714:  51%|█████     | 3272/6470 [40:38<45:23,  1.17it/s]
3|3273|Loss: 0.9800723791122437:  51%|█████     | 3272/6470 [40:39<45:23,  1.17it/s]
3|3273|Loss: 0.9800723791122437:  51%|█████     | 3273/6470 [40:39<44:37,  1.19it/s]
3|3274|Loss: 0.8393346667289734:  51%|█████     | 3273/6470 [40:40<44:37,  1.19it/s]
3|3274|Loss: 0.8393346667289734:  51%|█████     | 3274/6470 [40:40<42:48,  1.24it/s]
3|3275|Loss: 0.8140292167663574:  51%|█████     | 3274/6470 [40:40<42:48,  1.24it/s]
3|3275|Loss: 0.8140292167663574:  51%|█████     | 3275/6470 [40:41<42:30,  1.25it/s]
3|3276|Loss: 0.736475944519043:  51%|█████     | 3275/6470 [40:41<42:30,  1.25it/s] 
3|3276|Loss: 0.736475944519043:  51%|█████     | 3276/6470 [40:41<40:16,  1.32it/s]
3|3277|Loss: 0.7396672368049622:  51%|█████     | 3276/6470 [40:42<40:16,  1.32it/s]
3|3277|Loss: 0.7396672368049622:  51%|█████     | 3277/6470 [40:42<39:37,  1.34it/s]
3|3278|Loss: 0.8876594305038452:  51%|█████     | 3277/6470 [40:42<39:37,  1.34it/s]
3|3278|Loss: 0.8876594305038452:  51%|█████     | 3278/6470 [40:43<41:09,  1.29it/s]
3|3279|Loss: 0.7772751450538635:  51%|█████     | 3278/6470 [40:43<41:09,  1.29it/s]
3|3279|Loss: 0.7772751450538635:  51%|█████     | 3279/6470 [40:44<41:45,  1.27it/s]
3|3280|Loss: 0.8703471422195435:  51%|█████     | 3279/6470 [40:44<41:45,  1.27it/s]
3|3280|Loss: 0.8703471422195435:  51%|█████     | 3280/6470 [40:45<42:48,  1.24it/s]
3|3281|Loss: 0.9411036372184753:  51%|█████     | 3280/6470 [40:45<42:48,  1.24it/s]
3|3281|Loss: 0.9411036372184753:  51%|█████     | 3281/6470 [40:45<41:32,  1.28it/s]
3|3282|Loss: 1.080946922302246:  51%|█████     | 3281/6470 [40:46<41:32,  1.28it/s] 
3|3282|Loss: 1.080946922302246:  51%|█████     | 3282/6470 [40:46<40:05,  1.33it/s]
3|3283|Loss: 0.8701480627059937:  51%|█████     | 3282/6470 [40:46<40:05,  1.33it/s]
3|3283|Loss: 0.8701480627059937:  51%|█████     | 3283/6470 [40:47<37:11,  1.43it/s]
3|3284|Loss: 1.0376193523406982:  51%|█████     | 3283/6470 [40:47<37:11,  1.43it/s]
3|3284|Loss: 1.0376193523406982:  51%|█████     | 3284/6470 [40:47<39:11,  1.35it/s]
3|3285|Loss: 0.8144093751907349:  51%|█████     | 3284/6470 [40:48<39:11,  1.35it/s]
3|3285|Loss: 0.8144093751907349:  51%|█████     | 3285/6470 [40:48<40:05,  1.32it/s]
3|3286|Loss: 0.8688620328903198:  51%|█████     | 3285/6470 [40:49<40:05,  1.32it/s]
3|3286|Loss: 0.8688620328903198:  51%|█████     | 3286/6470 [40:49<40:48,  1.30it/s]
3|3287|Loss: 0.9247891902923584:  51%|█████     | 3286/6470 [40:49<40:48,  1.30it/s]
3|3287|Loss: 0.9247891902923584:  51%|█████     | 3287/6470 [40:50<42:43,  1.24it/s]
3|3288|Loss: 0.7811253666877747:  51%|█████     | 3287/6470 [40:50<42:43,  1.24it/s]
3|3288|Loss: 0.7811253666877747:  51%|█████     | 3288/6470 [40:51<44:39,  1.19it/s]
3|3289|Loss: 0.8994659185409546:  51%|█████     | 3288/6470 [40:51<44:39,  1.19it/s]
3|3289|Loss: 0.8994659185409546:  51%|█████     | 3289/6470 [40:51<42:43,  1.24it/s]
3|3290|Loss: 0.7325708270072937:  51%|█████     | 3289/6470 [40:52<42:43,  1.24it/s]
3|3290|Loss: 0.7325708270072937:  51%|█████     | 3290/6470 [40:52<41:24,  1.28it/s]
3|3291|Loss: 0.8062401413917542:  51%|█████     | 3290/6470 [40:53<41:24,  1.28it/s]
3|3291|Loss: 0.8062401413917542:  51%|█████     | 3291/6470 [40:53<38:52,  1.36it/s]
3|3292|Loss: 1.0129314661026:  51%|█████     | 3291/6470 [40:53<38:52,  1.36it/s]   
3|3292|Loss: 1.0129314661026:  51%|█████     | 3292/6470 [40:53<37:25,  1.42it/s]
3|3293|Loss: 0.6904140710830688:  51%|█████     | 3292/6470 [40:54<37:25,  1.42it/s]
3|3293|Loss: 0.6904140710830688:  51%|█████     | 3293/6470 [40:54<37:52,  1.40it/s]
3|3294|Loss: 0.8378546237945557:  51%|█████     | 3293/6470 [40:55<37:52,  1.40it/s]
3|3294|Loss: 0.8378546237945557:  51%|█████     | 3294/6470 [40:55<38:28,  1.38it/s]
3|3295|Loss: 0.8058103919029236:  51%|█████     | 3294/6470 [40:55<38:28,  1.38it/s]
3|3295|Loss: 0.8058103919029236:  51%|█████     | 3295/6470 [40:56<37:53,  1.40it/s]
3|3296|Loss: 0.9248915314674377:  51%|█████     | 3295/6470 [40:56<37:53,  1.40it/s]
3|3296|Loss: 0.9248915314674377:  51%|█████     | 3296/6470 [40:56<36:30,  1.45it/s]
3|3297|Loss: 0.7512838840484619:  51%|█████     | 3296/6470 [40:57<36:30,  1.45it/s]
3|3297|Loss: 0.7512838840484619:  51%|█████     | 3297/6470 [40:57<36:24,  1.45it/s]
3|3298|Loss: 0.8973676562309265:  51%|█████     | 3297/6470 [40:57<36:24,  1.45it/s]
3|3298|Loss: 0.8973676562309265:  51%|█████     | 3298/6470 [40:58<36:40,  1.44it/s]
3|3299|Loss: 0.8516696691513062:  51%|█████     | 3298/6470 [40:58<36:40,  1.44it/s]
3|3299|Loss: 0.8516696691513062:  51%|█████     | 3299/6470 [40:58<35:02,  1.51it/s]
3|3300|Loss: 0.8435108661651611:  51%|█████     | 3299/6470 [40:59<35:02,  1.51it/s]
3|3300|Loss: 0.8435108661651611:  51%|█████     | 3300/6470 [40:59<31:29,  1.68it/s]
3|3301|Loss: 0.6813040971755981:  51%|█████     | 3300/6470 [40:59<31:29,  1.68it/s]
3|3301|Loss: 0.6813040971755981:  51%|█████     | 3301/6470 [40:59<30:57,  1.71it/s]
3|3302|Loss: 0.8674026131629944:  51%|█████     | 3301/6470 [41:00<30:57,  1.71it/s]
3|3302|Loss: 0.8674026131629944:  51%|█████     | 3302/6470 [41:00<32:10,  1.64it/s]
3|3303|Loss: 0.947528600692749:  51%|█████     | 3302/6470 [41:00<32:10,  1.64it/s] 
3|3303|Loss: 0.947528600692749:  51%|█████     | 3303/6470 [41:01<34:50,  1.51it/s]
3|3304|Loss: 0.9081202745437622:  51%|█████     | 3303/6470 [41:01<34:50,  1.51it/s]
3|3304|Loss: 0.9081202745437622:  51%|█████     | 3304/6470 [41:01<32:53,  1.60it/s]
3|3305|Loss: 0.7944741249084473:  51%|█████     | 3304/6470 [41:02<32:53,  1.60it/s]
3|3305|Loss: 0.7944741249084473:  51%|█████     | 3305/6470 [41:02<34:13,  1.54it/s]
3|3306|Loss: 0.8313709497451782:  51%|█████     | 3305/6470 [41:02<34:13,  1.54it/s]
3|3306|Loss: 0.8313709497451782:  51%|█████     | 3306/6470 [41:03<34:18,  1.54it/s]
3|3307|Loss: 0.8886902332305908:  51%|█████     | 3306/6470 [41:03<34:18,  1.54it/s]
3|3307|Loss: 0.8886902332305908:  51%|█████     | 3307/6470 [41:03<34:05,  1.55it/s]
3|3308|Loss: 0.9179954528808594:  51%|█████     | 3307/6470 [41:04<34:05,  1.55it/s]
3|3308|Loss: 0.9179954528808594:  51%|█████     | 3308/6470 [41:04<35:54,  1.47it/s]
3|3309|Loss: 0.8744555711746216:  51%|█████     | 3308/6470 [41:04<35:54,  1.47it/s]
3|3309|Loss: 0.8744555711746216:  51%|█████     | 3309/6470 [41:05<37:44,  1.40it/s]
3|3310|Loss: 1.0728200674057007:  51%|█████     | 3309/6470 [41:05<37:44,  1.40it/s]
3|3310|Loss: 1.0728200674057007:  51%|█████     | 3310/6470 [41:06<38:55,  1.35it/s]
3|3311|Loss: 0.9008178114891052:  51%|█████     | 3310/6470 [41:06<38:55,  1.35it/s]
3|3311|Loss: 0.9008178114891052:  51%|█████     | 3311/6470 [41:06<39:59,  1.32it/s]
3|3312|Loss: 0.639842689037323:  51%|█████     | 3311/6470 [41:07<39:59,  1.32it/s] 
3|3312|Loss: 0.639842689037323:  51%|█████     | 3312/6470 [41:07<43:17,  1.22it/s]
3|3313|Loss: 0.912516713142395:  51%|█████     | 3312/6470 [41:08<43:17,  1.22it/s]
3|3313|Loss: 0.912516713142395:  51%|█████     | 3313/6470 [41:08<43:42,  1.20it/s]
3|3314|Loss: 0.8038508296012878:  51%|█████     | 3313/6470 [41:09<43:42,  1.20it/s]
3|3314|Loss: 0.8038508296012878:  51%|█████     | 3314/6470 [41:09<42:55,  1.23it/s]
3|3315|Loss: 0.8913159966468811:  51%|█████     | 3314/6470 [41:09<42:55,  1.23it/s]
3|3315|Loss: 0.8913159966468811:  51%|█████     | 3315/6470 [41:10<40:06,  1.31it/s]
3|3316|Loss: 0.5459425449371338:  51%|█████     | 3315/6470 [41:10<40:06,  1.31it/s]
3|3316|Loss: 0.5459425449371338:  51%|█████▏    | 3316/6470 [41:10<41:20,  1.27it/s]
3|3317|Loss: 0.8098671436309814:  51%|█████▏    | 3316/6470 [41:11<41:20,  1.27it/s]
3|3317|Loss: 0.8098671436309814:  51%|█████▏    | 3317/6470 [41:11<40:25,  1.30it/s]
3|3318|Loss: 0.8193351030349731:  51%|█████▏    | 3317/6470 [41:12<40:25,  1.30it/s]
3|3318|Loss: 0.8193351030349731:  51%|█████▏    | 3318/6470 [41:12<40:16,  1.30it/s]
3|3319|Loss: 0.8840354084968567:  51%|█████▏    | 3318/6470 [41:12<40:16,  1.30it/s]
3|3319|Loss: 0.8840354084968567:  51%|█████▏    | 3319/6470 [41:13<41:55,  1.25it/s]
3|3320|Loss: 0.8638139963150024:  51%|█████▏    | 3319/6470 [41:13<41:55,  1.25it/s]
3|3320|Loss: 0.8638139963150024:  51%|█████▏    | 3320/6470 [41:14<42:01,  1.25it/s]
3|3321|Loss: 0.6198996901512146:  51%|█████▏    | 3320/6470 [41:14<42:01,  1.25it/s]
3|3321|Loss: 0.6198996901512146:  51%|█████▏    | 3321/6470 [41:14<36:43,  1.43it/s]
3|3322|Loss: 0.8369365930557251:  51%|█████▏    | 3321/6470 [41:14<36:43,  1.43it/s]
3|3322|Loss: 0.8369365930557251:  51%|█████▏    | 3322/6470 [41:15<37:52,  1.39it/s]
3|3323|Loss: 0.5073287487030029:  51%|█████▏    | 3322/6470 [41:15<37:52,  1.39it/s]
3|3323|Loss: 0.5073287487030029:  51%|█████▏    | 3323/6470 [41:16<39:19,  1.33it/s]
3|3324|Loss: 0.7814528346061707:  51%|█████▏    | 3323/6470 [41:16<39:19,  1.33it/s]
3|3324|Loss: 0.7814528346061707:  51%|█████▏    | 3324/6470 [41:17<41:21,  1.27it/s]
3|3325|Loss: 0.8150126934051514:  51%|█████▏    | 3324/6470 [41:17<41:21,  1.27it/s]
3|3325|Loss: 0.8150126934051514:  51%|█████▏    | 3325/6470 [41:17<40:31,  1.29it/s]
3|3326|Loss: 0.8633922338485718:  51%|█████▏    | 3325/6470 [41:18<40:31,  1.29it/s]
3|3326|Loss: 0.8633922338485718:  51%|█████▏    | 3326/6470 [41:18<39:23,  1.33it/s]
3|3327|Loss: 0.8127476572990417:  51%|█████▏    | 3326/6470 [41:18<39:23,  1.33it/s]
3|3327|Loss: 0.8127476572990417:  51%|█████▏    | 3327/6470 [41:19<38:07,  1.37it/s]
3|3328|Loss: 0.8641898036003113:  51%|█████▏    | 3327/6470 [41:19<38:07,  1.37it/s]
3|3328|Loss: 0.8641898036003113:  51%|█████▏    | 3328/6470 [41:19<38:16,  1.37it/s]
3|3329|Loss: 0.8787944316864014:  51%|█████▏    | 3328/6470 [41:20<38:16,  1.37it/s]
3|3329|Loss: 0.8787944316864014:  51%|█████▏    | 3329/6470 [41:20<39:06,  1.34it/s]
3|3330|Loss: 0.7833892703056335:  51%|█████▏    | 3329/6470 [41:21<39:06,  1.34it/s]
3|3330|Loss: 0.7833892703056335:  51%|█████▏    | 3330/6470 [41:21<39:27,  1.33it/s]
3|3331|Loss: 0.8129493594169617:  51%|█████▏    | 3330/6470 [41:21<39:27,  1.33it/s]
3|3331|Loss: 0.8129493594169617:  51%|█████▏    | 3331/6470 [41:22<38:39,  1.35it/s]
3|3332|Loss: 0.8330102562904358:  51%|█████▏    | 3331/6470 [41:22<38:39,  1.35it/s]
3|3332|Loss: 0.8330102562904358:  51%|█████▏    | 3332/6470 [41:23<40:01,  1.31it/s]
3|3333|Loss: 0.8919046521186829:  51%|█████▏    | 3332/6470 [41:23<40:01,  1.31it/s]
3|3333|Loss: 0.8919046521186829:  52%|█████▏    | 3333/6470 [41:23<39:36,  1.32it/s]
3|3334|Loss: 0.6964303851127625:  52%|█████▏    | 3333/6470 [41:24<39:36,  1.32it/s]
3|3334|Loss: 0.6964303851127625:  52%|█████▏    | 3334/6470 [41:24<39:51,  1.31it/s]
3|3335|Loss: 0.8398790955543518:  52%|█████▏    | 3334/6470 [41:24<39:51,  1.31it/s]
3|3335|Loss: 0.8398790955543518:  52%|█████▏    | 3335/6470 [41:25<40:23,  1.29it/s]
3|3336|Loss: 0.7710009813308716:  52%|█████▏    | 3335/6470 [41:25<40:23,  1.29it/s]
3|3336|Loss: 0.7710009813308716:  52%|█████▏    | 3336/6470 [41:26<40:23,  1.29it/s]
3|3337|Loss: 0.7457665801048279:  52%|█████▏    | 3336/6470 [41:26<40:23,  1.29it/s]
3|3337|Loss: 0.7457665801048279:  52%|█████▏    | 3337/6470 [41:26<40:35,  1.29it/s]
3|3338|Loss: 0.9386916160583496:  52%|█████▏    | 3337/6470 [41:27<40:35,  1.29it/s]
3|3338|Loss: 0.9386916160583496:  52%|█████▏    | 3338/6470 [41:27<40:19,  1.29it/s]
3|3339|Loss: 0.8867419958114624:  52%|█████▏    | 3338/6470 [41:28<40:19,  1.29it/s]
3|3339|Loss: 0.8867419958114624:  52%|█████▏    | 3339/6470 [41:28<40:22,  1.29it/s]
3|3340|Loss: 0.9672624468803406:  52%|█████▏    | 3339/6470 [41:28<40:22,  1.29it/s]
3|3340|Loss: 0.9672624468803406:  52%|█████▏    | 3340/6470 [41:29<39:12,  1.33it/s]
3|3341|Loss: 0.802776038646698:  52%|█████▏    | 3340/6470 [41:29<39:12,  1.33it/s] 
3|3341|Loss: 0.802776038646698:  52%|█████▏    | 3341/6470 [41:29<39:26,  1.32it/s]
3|3342|Loss: 0.7963617444038391:  52%|█████▏    | 3341/6470 [41:30<39:26,  1.32it/s]
3|3342|Loss: 0.7963617444038391:  52%|█████▏    | 3342/6470 [41:30<39:53,  1.31it/s]
3|3343|Loss: 0.996678352355957:  52%|█████▏    | 3342/6470 [41:31<39:53,  1.31it/s] 
3|3343|Loss: 0.996678352355957:  52%|█████▏    | 3343/6470 [41:31<40:23,  1.29it/s]
3|3344|Loss: 0.9642853140830994:  52%|█████▏    | 3343/6470 [41:31<40:23,  1.29it/s]
3|3344|Loss: 0.9642853140830994:  52%|█████▏    | 3344/6470 [41:32<38:18,  1.36it/s]
3|3345|Loss: 0.7740808725357056:  52%|█████▏    | 3344/6470 [41:32<38:18,  1.36it/s]
3|3345|Loss: 0.7740808725357056:  52%|█████▏    | 3345/6470 [41:32<38:40,  1.35it/s]
3|3346|Loss: 0.7853333950042725:  52%|█████▏    | 3345/6470 [41:33<38:40,  1.35it/s]
3|3346|Loss: 0.7853333950042725:  52%|█████▏    | 3346/6470 [41:33<40:21,  1.29it/s]
3|3347|Loss: 0.8143725991249084:  52%|█████▏    | 3346/6470 [41:34<40:21,  1.29it/s]
3|3347|Loss: 0.8143725991249084:  52%|█████▏    | 3347/6470 [41:34<37:44,  1.38it/s]
3|3348|Loss: 0.642536461353302:  52%|█████▏    | 3347/6470 [41:34<37:44,  1.38it/s] 
3|3348|Loss: 0.642536461353302:  52%|█████▏    | 3348/6470 [41:34<36:10,  1.44it/s]
3|3349|Loss: 0.72733074426651:  52%|█████▏    | 3348/6470 [41:35<36:10,  1.44it/s] 
3|3349|Loss: 0.72733074426651:  52%|█████▏    | 3349/6470 [41:35<37:07,  1.40it/s]
3|3350|Loss: 0.7617929577827454:  52%|█████▏    | 3349/6470 [41:36<37:07,  1.40it/s]
3|3350|Loss: 0.7617929577827454:  52%|█████▏    | 3350/6470 [41:36<35:41,  1.46it/s]
3|3351|Loss: 0.7825328707695007:  52%|█████▏    | 3350/6470 [41:36<35:41,  1.46it/s]
3|3351|Loss: 0.7825328707695007:  52%|█████▏    | 3351/6470 [41:37<35:33,  1.46it/s]
3|3352|Loss: 0.7077780961990356:  52%|█████▏    | 3351/6470 [41:37<35:33,  1.46it/s]
3|3352|Loss: 0.7077780961990356:  52%|█████▏    | 3352/6470 [41:38<40:18,  1.29it/s]
3|3353|Loss: 0.9557135701179504:  52%|█████▏    | 3352/6470 [41:38<40:18,  1.29it/s]
3|3353|Loss: 0.9557135701179504:  52%|█████▏    | 3353/6470 [41:38<42:15,  1.23it/s]
3|3354|Loss: 0.5769628882408142:  52%|█████▏    | 3353/6470 [41:39<42:15,  1.23it/s]
3|3354|Loss: 0.5769628882408142:  52%|█████▏    | 3354/6470 [41:39<38:47,  1.34it/s]
3|3355|Loss: 0.8785902261734009:  52%|█████▏    | 3354/6470 [41:39<38:47,  1.34it/s]
3|3355|Loss: 0.8785902261734009:  52%|█████▏    | 3355/6470 [41:40<38:08,  1.36it/s]
3|3356|Loss: 0.8102145195007324:  52%|█████▏    | 3355/6470 [41:40<38:08,  1.36it/s]
3|3356|Loss: 0.8102145195007324:  52%|█████▏    | 3356/6470 [41:40<38:24,  1.35it/s]
3|3357|Loss: 0.847307562828064:  52%|█████▏    | 3356/6470 [41:41<38:24,  1.35it/s] 
3|3357|Loss: 0.847307562828064:  52%|█████▏    | 3357/6470 [41:41<37:48,  1.37it/s]
3|3358|Loss: 0.8482237458229065:  52%|█████▏    | 3357/6470 [41:42<37:48,  1.37it/s]
3|3358|Loss: 0.8482237458229065:  52%|█████▏    | 3358/6470 [41:42<36:10,  1.43it/s]
3|3359|Loss: 0.6133105754852295:  52%|█████▏    | 3358/6470 [41:42<36:10,  1.43it/s]
3|3359|Loss: 0.6133105754852295:  52%|█████▏    | 3359/6470 [41:42<33:56,  1.53it/s]
3|3360|Loss: 0.8249040246009827:  52%|█████▏    | 3359/6470 [41:43<33:56,  1.53it/s]
3|3360|Loss: 0.8249040246009827:  52%|█████▏    | 3360/6470 [41:43<32:45,  1.58it/s]
3|3361|Loss: 0.8646101355552673:  52%|█████▏    | 3360/6470 [41:43<32:45,  1.58it/s]
3|3361|Loss: 0.8646101355552673:  52%|█████▏    | 3361/6470 [41:44<32:58,  1.57it/s]
3|3362|Loss: 0.7219039797782898:  52%|█████▏    | 3361/6470 [41:44<32:58,  1.57it/s]
3|3362|Loss: 0.7219039797782898:  52%|█████▏    | 3362/6470 [41:44<31:34,  1.64it/s]
3|3363|Loss: 0.8977358341217041:  52%|█████▏    | 3362/6470 [41:45<31:34,  1.64it/s]
3|3363|Loss: 0.8977358341217041:  52%|█████▏    | 3363/6470 [41:45<35:26,  1.46it/s]
3|3364|Loss: 0.8494697213172913:  52%|█████▏    | 3363/6470 [41:45<35:26,  1.46it/s]
3|3364|Loss: 0.8494697213172913:  52%|█████▏    | 3364/6470 [41:46<39:03,  1.33it/s]
3|3365|Loss: 0.6626622080802917:  52%|█████▏    | 3364/6470 [41:46<39:03,  1.33it/s]
3|3365|Loss: 0.6626622080802917:  52%|█████▏    | 3365/6470 [41:47<36:55,  1.40it/s]
3|3366|Loss: 0.9144287109375:  52%|█████▏    | 3365/6470 [41:47<36:55,  1.40it/s]   
3|3366|Loss: 0.9144287109375:  52%|█████▏    | 3366/6470 [41:47<37:25,  1.38it/s]
3|3367|Loss: 0.6495895981788635:  52%|█████▏    | 3366/6470 [41:48<37:25,  1.38it/s]
3|3367|Loss: 0.6495895981788635:  52%|█████▏    | 3367/6470 [41:48<37:56,  1.36it/s]
3|3368|Loss: 0.8089200854301453:  52%|█████▏    | 3367/6470 [41:48<37:56,  1.36it/s]
3|3368|Loss: 0.8089200854301453:  52%|█████▏    | 3368/6470 [41:49<40:44,  1.27it/s]
3|3369|Loss: 0.8865286707878113:  52%|█████▏    | 3368/6470 [41:49<40:44,  1.27it/s]
3|3369|Loss: 0.8865286707878113:  52%|█████▏    | 3369/6470 [41:50<41:32,  1.24it/s]
3|3370|Loss: 0.8566508293151855:  52%|█████▏    | 3369/6470 [41:50<41:32,  1.24it/s]
3|3370|Loss: 0.8566508293151855:  52%|█████▏    | 3370/6470 [41:51<42:04,  1.23it/s]
3|3371|Loss: 0.9331183433532715:  52%|█████▏    | 3370/6470 [41:51<42:04,  1.23it/s]
3|3371|Loss: 0.9331183433532715:  52%|█████▏    | 3371/6470 [41:51<42:13,  1.22it/s]
3|3372|Loss: 0.8927687406539917:  52%|█████▏    | 3371/6470 [41:52<42:13,  1.22it/s]
3|3372|Loss: 0.8927687406539917:  52%|█████▏    | 3372/6470 [41:52<40:43,  1.27it/s]
3|3373|Loss: 0.9825384020805359:  52%|█████▏    | 3372/6470 [41:53<40:43,  1.27it/s]
3|3373|Loss: 0.9825384020805359:  52%|█████▏    | 3373/6470 [41:53<40:27,  1.28it/s]
3|3374|Loss: 0.7924033403396606:  52%|█████▏    | 3373/6470 [41:53<40:27,  1.28it/s]
3|3374|Loss: 0.7924033403396606:  52%|█████▏    | 3374/6470 [41:54<38:22,  1.34it/s]
3|3375|Loss: 0.6996175050735474:  52%|█████▏    | 3374/6470 [41:54<38:22,  1.34it/s]
3|3375|Loss: 0.6996175050735474:  52%|█████▏    | 3375/6470 [41:54<36:30,  1.41it/s]
3|3376|Loss: 0.8293261528015137:  52%|█████▏    | 3375/6470 [41:55<36:30,  1.41it/s]
3|3376|Loss: 0.8293261528015137:  52%|█████▏    | 3376/6470 [41:55<36:48,  1.40it/s]
3|3377|Loss: 0.949552595615387:  52%|█████▏    | 3376/6470 [41:55<36:48,  1.40it/s] 
3|3377|Loss: 0.949552595615387:  52%|█████▏    | 3377/6470 [41:56<37:28,  1.38it/s]
3|3378|Loss: 0.7345484495162964:  52%|█████▏    | 3377/6470 [41:56<37:28,  1.38it/s]
3|3378|Loss: 0.7345484495162964:  52%|█████▏    | 3378/6470 [41:56<38:26,  1.34it/s]
3|3379|Loss: 0.9421123266220093:  52%|█████▏    | 3378/6470 [41:57<38:26,  1.34it/s]
3|3379|Loss: 0.9421123266220093:  52%|█████▏    | 3379/6470 [41:57<37:18,  1.38it/s]
3|3380|Loss: 0.8942493796348572:  52%|█████▏    | 3379/6470 [41:58<37:18,  1.38it/s]
3|3380|Loss: 0.8942493796348572:  52%|█████▏    | 3380/6470 [41:58<36:10,  1.42it/s]
3|3381|Loss: 0.9310507774353027:  52%|█████▏    | 3380/6470 [41:58<36:10,  1.42it/s]
3|3381|Loss: 0.9310507774353027:  52%|█████▏    | 3381/6470 [41:58<35:52,  1.44it/s]
3|3382|Loss: 0.747715175151825:  52%|█████▏    | 3381/6470 [41:59<35:52,  1.44it/s] 
3|3382|Loss: 0.747715175151825:  52%|█████▏    | 3382/6470 [41:59<36:55,  1.39it/s]
3|3383|Loss: 0.6445993781089783:  52%|█████▏    | 3382/6470 [42:00<36:55,  1.39it/s]
3|3383|Loss: 0.6445993781089783:  52%|█████▏    | 3383/6470 [42:00<37:23,  1.38it/s]
3|3384|Loss: 0.9830800890922546:  52%|█████▏    | 3383/6470 [42:00<37:23,  1.38it/s]
3|3384|Loss: 0.9830800890922546:  52%|█████▏    | 3384/6470 [42:01<39:43,  1.29it/s]
3|3385|Loss: 0.9583452343940735:  52%|█████▏    | 3384/6470 [42:01<39:43,  1.29it/s]
3|3385|Loss: 0.9583452343940735:  52%|█████▏    | 3385/6470 [42:02<38:46,  1.33it/s]
3|3386|Loss: 0.806569516658783:  52%|█████▏    | 3385/6470 [42:02<38:46,  1.33it/s] 
3|3386|Loss: 0.806569516658783:  52%|█████▏    | 3386/6470 [42:02<35:22,  1.45it/s]
3|3387|Loss: 0.8011572957038879:  52%|█████▏    | 3386/6470 [42:02<35:22,  1.45it/s]
3|3387|Loss: 0.8011572957038879:  52%|█████▏    | 3387/6470 [42:03<34:02,  1.51it/s]
3|3388|Loss: 0.8708035945892334:  52%|█████▏    | 3387/6470 [42:03<34:02,  1.51it/s]
3|3388|Loss: 0.8708035945892334:  52%|█████▏    | 3388/6470 [42:04<37:08,  1.38it/s]
3|3389|Loss: 0.883340060710907:  52%|█████▏    | 3388/6470 [42:04<37:08,  1.38it/s] 
3|3389|Loss: 0.883340060710907:  52%|█████▏    | 3389/6470 [42:04<38:59,  1.32it/s]
3|3390|Loss: 0.9188092350959778:  52%|█████▏    | 3389/6470 [42:05<38:59,  1.32it/s]
3|3390|Loss: 0.9188092350959778:  52%|█████▏    | 3390/6470 [42:05<38:02,  1.35it/s]
3|3391|Loss: 0.9223566055297852:  52%|█████▏    | 3390/6470 [42:05<38:02,  1.35it/s]
3|3391|Loss: 0.9223566055297852:  52%|█████▏    | 3391/6470 [42:06<35:13,  1.46it/s]
3|3392|Loss: 0.8620211482048035:  52%|█████▏    | 3391/6470 [42:06<35:13,  1.46it/s]
3|3392|Loss: 0.8620211482048035:  52%|█████▏    | 3392/6470 [42:06<34:27,  1.49it/s]
3|3393|Loss: 0.9059745669364929:  52%|█████▏    | 3392/6470 [42:07<34:27,  1.49it/s]
3|3393|Loss: 0.9059745669364929:  52%|█████▏    | 3393/6470 [42:07<35:47,  1.43it/s]
3|3394|Loss: 0.9350184798240662:  52%|█████▏    | 3393/6470 [42:08<35:47,  1.43it/s]
3|3394|Loss: 0.9350184798240662:  52%|█████▏    | 3394/6470 [42:08<37:23,  1.37it/s]
3|3395|Loss: 0.8280157446861267:  52%|█████▏    | 3394/6470 [42:08<37:23,  1.37it/s]
3|3395|Loss: 0.8280157446861267:  52%|█████▏    | 3395/6470 [42:09<36:32,  1.40it/s]
3|3396|Loss: 0.8747406005859375:  52%|█████▏    | 3395/6470 [42:09<36:32,  1.40it/s]
3|3396|Loss: 0.8747406005859375:  52%|█████▏    | 3396/6470 [42:09<37:01,  1.38it/s]
3|3397|Loss: 0.9289891719818115:  52%|█████▏    | 3396/6470 [42:10<37:01,  1.38it/s]
3|3397|Loss: 0.9289891719818115:  53%|█████▎    | 3397/6470 [42:10<38:45,  1.32it/s]
3|3398|Loss: 1.0200749635696411:  53%|█████▎    | 3397/6470 [42:11<38:45,  1.32it/s]
3|3398|Loss: 1.0200749635696411:  53%|█████▎    | 3398/6470 [42:11<39:39,  1.29it/s]
3|3399|Loss: 0.9039948582649231:  53%|█████▎    | 3398/6470 [42:11<39:39,  1.29it/s]
3|3399|Loss: 0.9039948582649231:  53%|█████▎    | 3399/6470 [42:12<36:06,  1.42it/s]
3|3400|Loss: 0.7467068433761597:  53%|█████▎    | 3399/6470 [42:12<36:06,  1.42it/s]
3|3400|Loss: 0.7467068433761597:  53%|█████▎    | 3400/6470 [42:12<36:11,  1.41it/s]
3|3401|Loss: 0.7955203652381897:  53%|█████▎    | 3400/6470 [42:13<36:11,  1.41it/s]
3|3401|Loss: 0.7955203652381897:  53%|█████▎    | 3401/6470 [42:13<36:17,  1.41it/s]
3|3402|Loss: 0.7920926213264465:  53%|█████▎    | 3401/6470 [42:13<36:17,  1.41it/s]
3|3402|Loss: 0.7920926213264465:  53%|█████▎    | 3402/6470 [42:14<38:51,  1.32it/s]
3|3403|Loss: 0.7416276335716248:  53%|█████▎    | 3402/6470 [42:14<38:51,  1.32it/s]
3|3403|Loss: 0.7416276335716248:  53%|█████▎    | 3403/6470 [42:15<40:29,  1.26it/s]
3|3404|Loss: 0.7623797655105591:  53%|█████▎    | 3403/6470 [42:15<40:29,  1.26it/s]
3|3404|Loss: 0.7623797655105591:  53%|█████▎    | 3404/6470 [42:15<40:10,  1.27it/s]
3|3405|Loss: 0.8654879331588745:  53%|█████▎    | 3404/6470 [42:16<40:10,  1.27it/s]
3|3405|Loss: 0.8654879331588745:  53%|█████▎    | 3405/6470 [42:16<40:51,  1.25it/s]
3|3406|Loss: 0.7671613693237305:  53%|█████▎    | 3405/6470 [42:17<40:51,  1.25it/s]
3|3406|Loss: 0.7671613693237305:  53%|█████▎    | 3406/6470 [42:17<40:27,  1.26it/s]
3|3407|Loss: 0.9419885873794556:  53%|█████▎    | 3406/6470 [42:17<40:27,  1.26it/s]
3|3407|Loss: 0.9419885873794556:  53%|█████▎    | 3407/6470 [42:18<40:07,  1.27it/s]
3|3408|Loss: 0.8349686861038208:  53%|█████▎    | 3407/6470 [42:18<40:07,  1.27it/s]
3|3408|Loss: 0.8349686861038208:  53%|█████▎    | 3408/6470 [42:19<38:46,  1.32it/s]
3|3409|Loss: 0.7253576517105103:  53%|█████▎    | 3408/6470 [42:19<38:46,  1.32it/s]
3|3409|Loss: 0.7253576517105103:  53%|█████▎    | 3409/6470 [42:19<37:33,  1.36it/s]
3|3410|Loss: 0.9443614482879639:  53%|█████▎    | 3409/6470 [42:20<37:33,  1.36it/s]
3|3410|Loss: 0.9443614482879639:  53%|█████▎    | 3410/6470 [42:20<35:34,  1.43it/s]
3|3411|Loss: 0.8525425791740417:  53%|█████▎    | 3410/6470 [42:20<35:34,  1.43it/s]
3|3411|Loss: 0.8525425791740417:  53%|█████▎    | 3411/6470 [42:21<36:18,  1.40it/s]
3|3412|Loss: 0.6740748286247253:  53%|█████▎    | 3411/6470 [42:21<36:18,  1.40it/s]
3|3412|Loss: 0.6740748286247253:  53%|█████▎    | 3412/6470 [42:21<36:05,  1.41it/s]
3|3413|Loss: 0.8762890696525574:  53%|█████▎    | 3412/6470 [42:22<36:05,  1.41it/s]
3|3413|Loss: 0.8762890696525574:  53%|█████▎    | 3413/6470 [42:22<35:59,  1.42it/s]
3|3414|Loss: 1.0226771831512451:  53%|█████▎    | 3413/6470 [42:22<35:59,  1.42it/s]
3|3414|Loss: 1.0226771831512451:  53%|█████▎    | 3414/6470 [42:23<38:49,  1.31it/s]
3|3415|Loss: 1.0065475702285767:  53%|█████▎    | 3414/6470 [42:23<38:49,  1.31it/s]
3|3415|Loss: 1.0065475702285767:  53%|█████▎    | 3415/6470 [42:24<39:47,  1.28it/s]
3|3416|Loss: 0.8210116624832153:  53%|█████▎    | 3415/6470 [42:24<39:47,  1.28it/s]
3|3416|Loss: 0.8210116624832153:  53%|█████▎    | 3416/6470 [42:24<38:47,  1.31it/s]
3|3417|Loss: 0.7889100313186646:  53%|█████▎    | 3416/6470 [42:25<38:47,  1.31it/s]
3|3417|Loss: 0.7889100313186646:  53%|█████▎    | 3417/6470 [42:25<37:18,  1.36it/s]
3|3418|Loss: 0.8562559485435486:  53%|█████▎    | 3417/6470 [42:26<37:18,  1.36it/s]
3|3418|Loss: 0.8562559485435486:  53%|█████▎    | 3418/6470 [42:26<39:28,  1.29it/s]
3|3419|Loss: 0.8847426176071167:  53%|█████▎    | 3418/6470 [42:26<39:28,  1.29it/s]
3|3419|Loss: 0.8847426176071167:  53%|█████▎    | 3419/6470 [42:27<39:54,  1.27it/s]
3|3420|Loss: 0.8699355125427246:  53%|█████▎    | 3419/6470 [42:27<39:54,  1.27it/s]
3|3420|Loss: 0.8699355125427246:  53%|█████▎    | 3420/6470 [42:28<41:30,  1.22it/s]
3|3421|Loss: 0.8108645081520081:  53%|█████▎    | 3420/6470 [42:28<41:30,  1.22it/s]
3|3421|Loss: 0.8108645081520081:  53%|█████▎    | 3421/6470 [42:29<42:50,  1.19it/s]
3|3422|Loss: 1.025541067123413:  53%|█████▎    | 3421/6470 [42:29<42:50,  1.19it/s] 
3|3422|Loss: 1.025541067123413:  53%|█████▎    | 3422/6470 [42:29<41:43,  1.22it/s]
3|3423|Loss: 0.8188019394874573:  53%|█████▎    | 3422/6470 [42:30<41:43,  1.22it/s]
3|3423|Loss: 0.8188019394874573:  53%|█████▎    | 3423/6470 [42:30<38:34,  1.32it/s]
3|3424|Loss: 0.8697566986083984:  53%|█████▎    | 3423/6470 [42:30<38:34,  1.32it/s]
3|3424|Loss: 0.8697566986083984:  53%|█████▎    | 3424/6470 [42:31<39:35,  1.28it/s]
3|3425|Loss: 0.9582420587539673:  53%|█████▎    | 3424/6470 [42:31<39:35,  1.28it/s]
3|3425|Loss: 0.9582420587539673:  53%|█████▎    | 3425/6470 [42:32<39:27,  1.29it/s]
3|3426|Loss: 0.8536087274551392:  53%|█████▎    | 3425/6470 [42:32<39:27,  1.29it/s]
3|3426|Loss: 0.8536087274551392:  53%|█████▎    | 3426/6470 [42:32<41:10,  1.23it/s]
3|3427|Loss: 0.8173839449882507:  53%|█████▎    | 3426/6470 [42:33<41:10,  1.23it/s]
3|3427|Loss: 0.8173839449882507:  53%|█████▎    | 3427/6470 [42:33<39:11,  1.29it/s]
3|3428|Loss: 0.7446243166923523:  53%|█████▎    | 3427/6470 [42:34<39:11,  1.29it/s]
3|3428|Loss: 0.7446243166923523:  53%|█████▎    | 3428/6470 [42:34<40:55,  1.24it/s]
3|3429|Loss: 0.7730677723884583:  53%|█████▎    | 3428/6470 [42:35<40:55,  1.24it/s]
3|3429|Loss: 0.7730677723884583:  53%|█████▎    | 3429/6470 [42:35<43:14,  1.17it/s]
3|3430|Loss: 0.723698616027832:  53%|█████▎    | 3429/6470 [42:35<43:14,  1.17it/s] 
3|3430|Loss: 0.723698616027832:  53%|█████▎    | 3430/6470 [42:36<42:56,  1.18it/s]
3|3431|Loss: 0.8674253225326538:  53%|█████▎    | 3430/6470 [42:36<42:56,  1.18it/s]
3|3431|Loss: 0.8674253225326538:  53%|█████▎    | 3431/6470 [42:37<43:25,  1.17it/s]
3|3432|Loss: 0.9553178548812866:  53%|█████▎    | 3431/6470 [42:37<43:25,  1.17it/s]
3|3432|Loss: 0.9553178548812866:  53%|█████▎    | 3432/6470 [42:38<43:35,  1.16it/s]
3|3433|Loss: 0.841753363609314:  53%|█████▎    | 3432/6470 [42:38<43:35,  1.16it/s] 
3|3433|Loss: 0.841753363609314:  53%|█████▎    | 3433/6470 [42:38<44:04,  1.15it/s]
3|3434|Loss: 0.9004576206207275:  53%|█████▎    | 3433/6470 [42:39<44:04,  1.15it/s]
3|3434|Loss: 0.9004576206207275:  53%|█████▎    | 3434/6470 [42:39<42:41,  1.19it/s]
3|3435|Loss: 0.8709654808044434:  53%|█████▎    | 3434/6470 [42:40<42:41,  1.19it/s]
3|3435|Loss: 0.8709654808044434:  53%|█████▎    | 3435/6470 [42:40<36:27,  1.39it/s]
3|3436|Loss: 0.8521739840507507:  53%|█████▎    | 3435/6470 [42:40<36:27,  1.39it/s]
3|3436|Loss: 0.8521739840507507:  53%|█████▎    | 3436/6470 [42:40<35:50,  1.41it/s]
3|3437|Loss: 0.9615780115127563:  53%|█████▎    | 3436/6470 [42:41<35:50,  1.41it/s]
3|3437|Loss: 0.9615780115127563:  53%|█████▎    | 3437/6470 [42:41<37:50,  1.34it/s]
3|3438|Loss: 0.8896753787994385:  53%|█████▎    | 3437/6470 [42:42<37:50,  1.34it/s]
3|3438|Loss: 0.8896753787994385:  53%|█████▎    | 3438/6470 [42:42<38:35,  1.31it/s]
3|3439|Loss: 1.1403372287750244:  53%|█████▎    | 3438/6470 [42:42<38:35,  1.31it/s]
3|3439|Loss: 1.1403372287750244:  53%|█████▎    | 3439/6470 [42:43<35:30,  1.42it/s]
3|3440|Loss: 0.8557145595550537:  53%|█████▎    | 3439/6470 [42:43<35:30,  1.42it/s]
3|3440|Loss: 0.8557145595550537:  53%|█████▎    | 3440/6470 [42:43<36:44,  1.37it/s]
3|3441|Loss: 0.7724037766456604:  53%|█████▎    | 3440/6470 [42:44<36:44,  1.37it/s]
3|3441|Loss: 0.7724037766456604:  53%|█████▎    | 3441/6470 [42:44<38:44,  1.30it/s]
3|3442|Loss: 0.6787176132202148:  53%|█████▎    | 3441/6470 [42:45<38:44,  1.30it/s]
3|3442|Loss: 0.6787176132202148:  53%|█████▎    | 3442/6470 [42:45<41:46,  1.21it/s]
3|3443|Loss: 0.9542360901832581:  53%|█████▎    | 3442/6470 [42:46<41:46,  1.21it/s]
3|3443|Loss: 0.9542360901832581:  53%|█████▎    | 3443/6470 [42:46<43:42,  1.15it/s]
3|3444|Loss: 0.8394871354103088:  53%|█████▎    | 3443/6470 [42:47<43:42,  1.15it/s]
3|3444|Loss: 0.8394871354103088:  53%|█████▎    | 3444/6470 [42:47<40:33,  1.24it/s]
3|3445|Loss: 0.8795013427734375:  53%|█████▎    | 3444/6470 [42:47<40:33,  1.24it/s]
3|3445|Loss: 0.8795013427734375:  53%|█████▎    | 3445/6470 [42:47<38:25,  1.31it/s]
3|3446|Loss: 0.8163882493972778:  53%|█████▎    | 3445/6470 [42:48<38:25,  1.31it/s]
3|3446|Loss: 0.8163882493972778:  53%|█████▎    | 3446/6470 [42:48<38:38,  1.30it/s]
3|3447|Loss: 0.8539180159568787:  53%|█████▎    | 3446/6470 [42:49<38:38,  1.30it/s]
3|3447|Loss: 0.8539180159568787:  53%|█████▎    | 3447/6470 [42:49<38:57,  1.29it/s]
3|3448|Loss: 0.784935712814331:  53%|█████▎    | 3447/6470 [42:49<38:57,  1.29it/s] 
3|3448|Loss: 0.784935712814331:  53%|█████▎    | 3448/6470 [42:49<34:26,  1.46it/s]
3|3449|Loss: 0.8025233149528503:  53%|█████▎    | 3448/6470 [42:50<34:26,  1.46it/s]
3|3449|Loss: 0.8025233149528503:  53%|█████▎    | 3449/6470 [42:50<32:31,  1.55it/s]
3|3450|Loss: 0.8085280656814575:  53%|█████▎    | 3449/6470 [42:50<32:31,  1.55it/s]
3|3450|Loss: 0.8085280656814575:  53%|█████▎    | 3450/6470 [42:51<31:01,  1.62it/s]
3|3451|Loss: 0.9604471325874329:  53%|█████▎    | 3450/6470 [42:51<31:01,  1.62it/s]
3|3451|Loss: 0.9604471325874329:  53%|█████▎    | 3451/6470 [42:51<30:48,  1.63it/s]
3|3452|Loss: 0.7152238488197327:  53%|█████▎    | 3451/6470 [42:51<30:48,  1.63it/s]
3|3452|Loss: 0.7152238488197327:  53%|█████▎    | 3452/6470 [42:52<27:46,  1.81it/s]
3|3453|Loss: 0.9035865664482117:  53%|█████▎    | 3452/6470 [42:52<27:46,  1.81it/s]
3|3453|Loss: 0.9035865664482117:  53%|█████▎    | 3453/6470 [42:52<29:49,  1.69it/s]
3|3454|Loss: 0.8581591248512268:  53%|█████▎    | 3453/6470 [42:53<29:49,  1.69it/s]
3|3454|Loss: 0.8581591248512268:  53%|█████▎    | 3454/6470 [42:53<32:29,  1.55it/s]
3|3455|Loss: 0.8112757802009583:  53%|█████▎    | 3454/6470 [42:53<32:29,  1.55it/s]
3|3455|Loss: 0.8112757802009583:  53%|█████▎    | 3455/6470 [42:54<32:58,  1.52it/s]
3|3456|Loss: 1.1710290908813477:  53%|█████▎    | 3455/6470 [42:54<32:58,  1.52it/s]
3|3456|Loss: 1.1710290908813477:  53%|█████▎    | 3456/6470 [42:54<34:38,  1.45it/s]
3|3457|Loss: 0.8803914785385132:  53%|█████▎    | 3456/6470 [42:55<34:38,  1.45it/s]
3|3457|Loss: 0.8803914785385132:  53%|█████▎    | 3457/6470 [42:55<36:12,  1.39it/s]
3|3458|Loss: 0.8113430142402649:  53%|█████▎    | 3457/6470 [42:56<36:12,  1.39it/s]
3|3458|Loss: 0.8113430142402649:  53%|█████▎    | 3458/6470 [42:56<37:29,  1.34it/s]
3|3459|Loss: 0.8762797117233276:  53%|█████▎    | 3458/6470 [42:57<37:29,  1.34it/s]
3|3459|Loss: 0.8762797117233276:  53%|█████▎    | 3459/6470 [42:57<40:42,  1.23it/s]
3|3460|Loss: 0.8632776737213135:  53%|█████▎    | 3459/6470 [42:58<40:42,  1.23it/s]
3|3460|Loss: 0.8632776737213135:  53%|█████▎    | 3460/6470 [42:58<41:03,  1.22it/s]
3|3461|Loss: 0.7181911468505859:  53%|█████▎    | 3460/6470 [42:58<41:03,  1.22it/s]
3|3461|Loss: 0.7181911468505859:  53%|█████▎    | 3461/6470 [42:59<39:26,  1.27it/s]
3|3462|Loss: 0.7333829998970032:  53%|█████▎    | 3461/6470 [42:59<39:26,  1.27it/s]
3|3462|Loss: 0.7333829998970032:  54%|█████▎    | 3462/6470 [42:59<33:43,  1.49it/s]
3|3463|Loss: 0.8835952281951904:  54%|█████▎    | 3462/6470 [42:59<33:43,  1.49it/s]
3|3463|Loss: 0.8835952281951904:  54%|█████▎    | 3463/6470 [43:00<32:02,  1.56it/s]
3|3464|Loss: 0.8164910674095154:  54%|█████▎    | 3463/6470 [43:00<32:02,  1.56it/s]
3|3464|Loss: 0.8164910674095154:  54%|█████▎    | 3464/6470 [43:00<35:35,  1.41it/s]
3|3465|Loss: 0.7930179238319397:  54%|█████▎    | 3464/6470 [43:01<35:35,  1.41it/s]
3|3465|Loss: 0.7930179238319397:  54%|█████▎    | 3465/6470 [43:01<37:26,  1.34it/s]
3|3466|Loss: 0.6897895336151123:  54%|█████▎    | 3465/6470 [43:02<37:26,  1.34it/s]
3|3466|Loss: 0.6897895336151123:  54%|█████▎    | 3466/6470 [43:02<35:09,  1.42it/s]
3|3467|Loss: 0.7588107585906982:  54%|█████▎    | 3466/6470 [43:02<35:09,  1.42it/s]
3|3467|Loss: 0.7588107585906982:  54%|█████▎    | 3467/6470 [43:02<33:35,  1.49it/s]
3|3468|Loss: 0.7981545329093933:  54%|█████▎    | 3467/6470 [43:03<33:35,  1.49it/s]
3|3468|Loss: 0.7981545329093933:  54%|█████▎    | 3468/6470 [43:03<31:36,  1.58it/s]
3|3469|Loss: 0.8959844708442688:  54%|█████▎    | 3468/6470 [43:03<31:36,  1.58it/s]
3|3469|Loss: 0.8959844708442688:  54%|█████▎    | 3469/6470 [43:04<31:30,  1.59it/s]
3|3470|Loss: 0.8020415902137756:  54%|█████▎    | 3469/6470 [43:04<31:30,  1.59it/s]
3|3470|Loss: 0.8020415902137756:  54%|█████▎    | 3470/6470 [43:04<33:05,  1.51it/s]
3|3471|Loss: 0.9032223224639893:  54%|█████▎    | 3470/6470 [43:05<33:05,  1.51it/s]
3|3471|Loss: 0.9032223224639893:  54%|█████▎    | 3471/6470 [43:05<33:42,  1.48it/s]
3|3472|Loss: 0.819365918636322:  54%|█████▎    | 3471/6470 [43:05<33:42,  1.48it/s] 
3|3472|Loss: 0.819365918636322:  54%|█████▎    | 3472/6470 [43:06<35:07,  1.42it/s]
3|3473|Loss: 0.8865172266960144:  54%|█████▎    | 3472/6470 [43:06<35:07,  1.42it/s]
3|3473|Loss: 0.8865172266960144:  54%|█████▎    | 3473/6470 [43:07<34:36,  1.44it/s]
3|3474|Loss: 0.7139946818351746:  54%|█████▎    | 3473/6470 [43:07<34:36,  1.44it/s]
3|3474|Loss: 0.7139946818351746:  54%|█████▎    | 3474/6470 [43:07<34:02,  1.47it/s]
3|3475|Loss: 0.8165773153305054:  54%|█████▎    | 3474/6470 [43:08<34:02,  1.47it/s]
3|3475|Loss: 0.8165773153305054:  54%|█████▎    | 3475/6470 [43:08<33:52,  1.47it/s]
3|3476|Loss: 0.8321435451507568:  54%|█████▎    | 3475/6470 [43:08<33:52,  1.47it/s]
3|3476|Loss: 0.8321435451507568:  54%|█████▎    | 3476/6470 [43:09<34:07,  1.46it/s]
3|3477|Loss: 0.7491540312767029:  54%|█████▎    | 3476/6470 [43:09<34:07,  1.46it/s]
3|3477|Loss: 0.7491540312767029:  54%|█████▎    | 3477/6470 [43:09<36:49,  1.35it/s]
3|3478|Loss: 0.9331356287002563:  54%|█████▎    | 3477/6470 [43:10<36:49,  1.35it/s]
3|3478|Loss: 0.9331356287002563:  54%|█████▍    | 3478/6470 [43:10<39:26,  1.26it/s]
3|3479|Loss: 0.9028527736663818:  54%|█████▍    | 3478/6470 [43:11<39:26,  1.26it/s]
3|3479|Loss: 0.9028527736663818:  54%|█████▍    | 3479/6470 [43:11<39:31,  1.26it/s]
3|3480|Loss: 0.9239726066589355:  54%|█████▍    | 3479/6470 [43:12<39:31,  1.26it/s]
3|3480|Loss: 0.9239726066589355:  54%|█████▍    | 3480/6470 [43:12<37:54,  1.31it/s]
3|3481|Loss: 0.6666027307510376:  54%|█████▍    | 3480/6470 [43:12<37:54,  1.31it/s]
3|3481|Loss: 0.6666027307510376:  54%|█████▍    | 3481/6470 [43:12<35:30,  1.40it/s]
3|3482|Loss: 0.7464478015899658:  54%|█████▍    | 3481/6470 [43:13<35:30,  1.40it/s]
3|3482|Loss: 0.7464478015899658:  54%|█████▍    | 3482/6470 [43:13<37:57,  1.31it/s]
3|3483|Loss: 0.7662518620491028:  54%|█████▍    | 3482/6470 [43:14<37:57,  1.31it/s]
3|3483|Loss: 0.7662518620491028:  54%|█████▍    | 3483/6470 [43:14<36:34,  1.36it/s]
3|3484|Loss: 0.701984703540802:  54%|█████▍    | 3483/6470 [43:14<36:34,  1.36it/s] 
3|3484|Loss: 0.701984703540802:  54%|█████▍    | 3484/6470 [43:15<35:30,  1.40it/s]
3|3485|Loss: 0.8630058169364929:  54%|█████▍    | 3484/6470 [43:15<35:30,  1.40it/s]
3|3485|Loss: 0.8630058169364929:  54%|█████▍    | 3485/6470 [43:15<36:02,  1.38it/s]
3|3486|Loss: 0.7031965851783752:  54%|█████▍    | 3485/6470 [43:16<36:02,  1.38it/s]
3|3486|Loss: 0.7031965851783752:  54%|█████▍    | 3486/6470 [43:16<33:36,  1.48it/s]
3|3487|Loss: 0.8452491164207458:  54%|█████▍    | 3486/6470 [43:16<33:36,  1.48it/s]
3|3487|Loss: 0.8452491164207458:  54%|█████▍    | 3487/6470 [43:17<34:04,  1.46it/s]
3|3488|Loss: 0.8795310854911804:  54%|█████▍    | 3487/6470 [43:17<34:04,  1.46it/s]
3|3488|Loss: 0.8795310854911804:  54%|█████▍    | 3488/6470 [43:17<34:20,  1.45it/s]
3|3489|Loss: 0.7529041767120361:  54%|█████▍    | 3488/6470 [43:18<34:20,  1.45it/s]
3|3489|Loss: 0.7529041767120361:  54%|█████▍    | 3489/6470 [43:18<35:08,  1.41it/s]
3|3490|Loss: 0.7506775259971619:  54%|█████▍    | 3489/6470 [43:19<35:08,  1.41it/s]
3|3490|Loss: 0.7506775259971619:  54%|█████▍    | 3490/6470 [43:19<35:54,  1.38it/s]
3|3491|Loss: 0.8062167167663574:  54%|█████▍    | 3490/6470 [43:19<35:54,  1.38it/s]
3|3491|Loss: 0.8062167167663574:  54%|█████▍    | 3491/6470 [43:20<34:56,  1.42it/s]
3|3492|Loss: 0.8180972933769226:  54%|█████▍    | 3491/6470 [43:20<34:56,  1.42it/s]
3|3492|Loss: 0.8180972933769226:  54%|█████▍    | 3492/6470 [43:20<35:10,  1.41it/s]
3|3493|Loss: 0.9339379668235779:  54%|█████▍    | 3492/6470 [43:21<35:10,  1.41it/s]
3|3493|Loss: 0.9339379668235779:  54%|█████▍    | 3493/6470 [43:21<34:40,  1.43it/s]
3|3494|Loss: 0.8156408667564392:  54%|█████▍    | 3493/6470 [43:21<34:40,  1.43it/s]
3|3494|Loss: 0.8156408667564392:  54%|█████▍    | 3494/6470 [43:22<35:16,  1.41it/s]
3|3495|Loss: 0.6895594000816345:  54%|█████▍    | 3494/6470 [43:22<35:16,  1.41it/s]
3|3495|Loss: 0.6895594000816345:  54%|█████▍    | 3495/6470 [43:23<38:20,  1.29it/s]
3|3496|Loss: 0.8867326974868774:  54%|█████▍    | 3495/6470 [43:23<38:20,  1.29it/s]
3|3496|Loss: 0.8867326974868774:  54%|█████▍    | 3496/6470 [43:24<41:24,  1.20it/s]
3|3497|Loss: 0.7921142578125:  54%|█████▍    | 3496/6470 [43:24<41:24,  1.20it/s]   
3|3497|Loss: 0.7921142578125:  54%|█████▍    | 3497/6470 [43:24<39:55,  1.24it/s]
3|3498|Loss: 0.847432017326355:  54%|█████▍    | 3497/6470 [43:25<39:55,  1.24it/s]
3|3498|Loss: 0.847432017326355:  54%|█████▍    | 3498/6470 [43:25<39:34,  1.25it/s]
3|3499|Loss: 0.7385391592979431:  54%|█████▍    | 3498/6470 [43:25<39:34,  1.25it/s]
3|3499|Loss: 0.7385391592979431:  54%|█████▍    | 3499/6470 [43:26<38:05,  1.30it/s]
3|3500|Loss: 0.9015766382217407:  54%|█████▍    | 3499/6470 [43:26<38:05,  1.30it/s]
3|3500|Loss: 0.9015766382217407:  54%|█████▍    | 3500/6470 [43:27<37:42,  1.31it/s]
3|3501|Loss: 0.8251342177391052:  54%|█████▍    | 3500/6470 [43:27<37:42,  1.31it/s]
3|3501|Loss: 0.8251342177391052:  54%|█████▍    | 3501/6470 [43:27<35:37,  1.39it/s]
3|3502|Loss: 0.8056086301803589:  54%|█████▍    | 3501/6470 [43:28<35:37,  1.39it/s]
3|3502|Loss: 0.8056086301803589:  54%|█████▍    | 3502/6470 [43:28<36:05,  1.37it/s]
3|3503|Loss: 0.8747500777244568:  54%|█████▍    | 3502/6470 [43:28<36:05,  1.37it/s]
3|3503|Loss: 0.8747500777244568:  54%|█████▍    | 3503/6470 [43:29<38:25,  1.29it/s]
3|3504|Loss: 0.8301953077316284:  54%|█████▍    | 3503/6470 [43:29<38:25,  1.29it/s]
3|3504|Loss: 0.8301953077316284:  54%|█████▍    | 3504/6470 [43:29<35:47,  1.38it/s]
3|3505|Loss: 0.949837863445282:  54%|█████▍    | 3504/6470 [43:30<35:47,  1.38it/s] 
3|3505|Loss: 0.949837863445282:  54%|█████▍    | 3505/6470 [43:30<34:34,  1.43it/s]
3|3506|Loss: 0.7839206457138062:  54%|█████▍    | 3505/6470 [43:30<34:34,  1.43it/s]
3|3506|Loss: 0.7839206457138062:  54%|█████▍    | 3506/6470 [43:31<35:52,  1.38it/s]
3|3507|Loss: 0.7187125086784363:  54%|█████▍    | 3506/6470 [43:31<35:52,  1.38it/s]
3|3507|Loss: 0.7187125086784363:  54%|█████▍    | 3507/6470 [43:31<33:37,  1.47it/s]
3|3508|Loss: 0.8446609973907471:  54%|█████▍    | 3507/6470 [43:32<33:37,  1.47it/s]
3|3508|Loss: 0.8446609973907471:  54%|█████▍    | 3508/6470 [43:32<31:07,  1.59it/s]
3|3509|Loss: 0.8667104244232178:  54%|█████▍    | 3508/6470 [43:32<31:07,  1.59it/s]
3|3509|Loss: 0.8667104244232178:  54%|█████▍    | 3509/6470 [43:32<30:26,  1.62it/s]
3|3510|Loss: 0.7390488982200623:  54%|█████▍    | 3509/6470 [43:33<30:26,  1.62it/s]
3|3510|Loss: 0.7390488982200623:  54%|█████▍    | 3510/6470 [43:33<31:18,  1.58it/s]
3|3511|Loss: 0.7633436918258667:  54%|█████▍    | 3510/6470 [43:34<31:18,  1.58it/s]
3|3511|Loss: 0.7633436918258667:  54%|█████▍    | 3511/6470 [43:34<32:56,  1.50it/s]
3|3512|Loss: 0.8614464998245239:  54%|█████▍    | 3511/6470 [43:34<32:56,  1.50it/s]
3|3512|Loss: 0.8614464998245239:  54%|█████▍    | 3512/6470 [43:35<35:52,  1.37it/s]
3|3513|Loss: 0.6853019595146179:  54%|█████▍    | 3512/6470 [43:35<35:52,  1.37it/s]
3|3513|Loss: 0.6853019595146179:  54%|█████▍    | 3513/6470 [43:36<39:27,  1.25it/s]
3|3514|Loss: 0.909803032875061:  54%|█████▍    | 3513/6470 [43:36<39:27,  1.25it/s] 
3|3514|Loss: 0.909803032875061:  54%|█████▍    | 3514/6470 [43:36<38:33,  1.28it/s]
3|3515|Loss: 0.8708788752555847:  54%|█████▍    | 3514/6470 [43:37<38:33,  1.28it/s]
3|3515|Loss: 0.8708788752555847:  54%|█████▍    | 3515/6470 [43:37<38:15,  1.29it/s]
3|3516|Loss: 0.8107578158378601:  54%|█████▍    | 3515/6470 [43:38<38:15,  1.29it/s]
3|3516|Loss: 0.8107578158378601:  54%|█████▍    | 3516/6470 [43:38<36:26,  1.35it/s]
3|3517|Loss: 0.8353421092033386:  54%|█████▍    | 3516/6470 [43:38<36:26,  1.35it/s]
3|3517|Loss: 0.8353421092033386:  54%|█████▍    | 3517/6470 [43:39<36:50,  1.34it/s]
3|3518|Loss: 0.799384355545044:  54%|█████▍    | 3517/6470 [43:39<36:50,  1.34it/s] 
3|3518|Loss: 0.799384355545044:  54%|█████▍    | 3518/6470 [43:39<36:01,  1.37it/s]
3|3519|Loss: 0.6828311681747437:  54%|█████▍    | 3518/6470 [43:40<36:01,  1.37it/s]
3|3519|Loss: 0.6828311681747437:  54%|█████▍    | 3519/6470 [43:40<35:05,  1.40it/s]
3|3520|Loss: 0.7978036999702454:  54%|█████▍    | 3519/6470 [43:40<35:05,  1.40it/s]
3|3520|Loss: 0.7978036999702454:  54%|█████▍    | 3520/6470 [43:41<34:54,  1.41it/s]
3|3521|Loss: 0.9539668560028076:  54%|█████▍    | 3520/6470 [43:41<34:54,  1.41it/s]
3|3521|Loss: 0.9539668560028076:  54%|█████▍    | 3521/6470 [43:42<36:00,  1.37it/s]
3|3522|Loss: 0.9055612683296204:  54%|█████▍    | 3521/6470 [43:42<36:00,  1.37it/s]
3|3522|Loss: 0.9055612683296204:  54%|█████▍    | 3522/6470 [43:42<37:39,  1.30it/s]
3|3523|Loss: 0.9795688390731812:  54%|█████▍    | 3522/6470 [43:43<37:39,  1.30it/s]
3|3523|Loss: 0.9795688390731812:  54%|█████▍    | 3523/6470 [43:43<38:23,  1.28it/s]
3|3524|Loss: 0.7547960877418518:  54%|█████▍    | 3523/6470 [43:44<38:23,  1.28it/s]
3|3524|Loss: 0.7547960877418518:  54%|█████▍    | 3524/6470 [43:44<38:32,  1.27it/s]
3|3525|Loss: 0.8077951073646545:  54%|█████▍    | 3524/6470 [43:44<38:32,  1.27it/s]
3|3525|Loss: 0.8077951073646545:  54%|█████▍    | 3525/6470 [43:45<39:17,  1.25it/s]
3|3526|Loss: 0.8823270201683044:  54%|█████▍    | 3525/6470 [43:45<39:17,  1.25it/s]
3|3526|Loss: 0.8823270201683044:  54%|█████▍    | 3526/6470 [43:46<40:29,  1.21it/s]
3|3527|Loss: 0.822277843952179:  54%|█████▍    | 3526/6470 [43:46<40:29,  1.21it/s] 
3|3527|Loss: 0.822277843952179:  55%|█████▍    | 3527/6470 [43:47<41:35,  1.18it/s]
3|3528|Loss: 0.8972969055175781:  55%|█████▍    | 3527/6470 [43:47<41:35,  1.18it/s]
3|3528|Loss: 0.8972969055175781:  55%|█████▍    | 3528/6470 [43:47<40:01,  1.22it/s]
3|3529|Loss: 0.8884422779083252:  55%|█████▍    | 3528/6470 [43:48<40:01,  1.22it/s]
3|3529|Loss: 0.8884422779083252:  55%|█████▍    | 3529/6470 [43:48<38:02,  1.29it/s]
3|3530|Loss: 0.8355334401130676:  55%|█████▍    | 3529/6470 [43:48<38:02,  1.29it/s]
3|3530|Loss: 0.8355334401130676:  55%|█████▍    | 3530/6470 [43:49<36:30,  1.34it/s]
3|3531|Loss: 0.9068151116371155:  55%|█████▍    | 3530/6470 [43:49<36:30,  1.34it/s]
3|3531|Loss: 0.9068151116371155:  55%|█████▍    | 3531/6470 [43:50<38:37,  1.27it/s]
3|3532|Loss: 0.9051482677459717:  55%|█████▍    | 3531/6470 [43:50<38:37,  1.27it/s]
3|3532|Loss: 0.9051482677459717:  55%|█████▍    | 3532/6470 [43:50<36:21,  1.35it/s]
3|3533|Loss: 0.8746557235717773:  55%|█████▍    | 3532/6470 [43:51<36:21,  1.35it/s]
3|3533|Loss: 0.8746557235717773:  55%|█████▍    | 3533/6470 [43:51<34:10,  1.43it/s]
3|3534|Loss: 0.9180243015289307:  55%|█████▍    | 3533/6470 [43:51<34:10,  1.43it/s]
3|3534|Loss: 0.9180243015289307:  55%|█████▍    | 3534/6470 [43:51<33:34,  1.46it/s]
3|3535|Loss: 0.8147620558738708:  55%|█████▍    | 3534/6470 [43:52<33:34,  1.46it/s]
3|3535|Loss: 0.8147620558738708:  55%|█████▍    | 3535/6470 [43:52<33:35,  1.46it/s]
3|3536|Loss: 0.8203855156898499:  55%|█████▍    | 3535/6470 [43:53<33:35,  1.46it/s]
3|3536|Loss: 0.8203855156898499:  55%|█████▍    | 3536/6470 [43:53<33:53,  1.44it/s]
3|3537|Loss: 0.8000421524047852:  55%|█████▍    | 3536/6470 [43:53<33:53,  1.44it/s]
3|3537|Loss: 0.8000421524047852:  55%|█████▍    | 3537/6470 [43:54<36:30,  1.34it/s]
3|3538|Loss: 0.7112477421760559:  55%|█████▍    | 3537/6470 [43:54<36:30,  1.34it/s]
3|3538|Loss: 0.7112477421760559:  55%|█████▍    | 3538/6470 [43:54<36:00,  1.36it/s]
3|3539|Loss: 1.014965534210205:  55%|█████▍    | 3538/6470 [43:55<36:00,  1.36it/s] 
3|3539|Loss: 1.014965534210205:  55%|█████▍    | 3539/6470 [43:55<37:34,  1.30it/s]
3|3540|Loss: 0.7613975405693054:  55%|█████▍    | 3539/6470 [43:56<37:34,  1.30it/s]
3|3540|Loss: 0.7613975405693054:  55%|█████▍    | 3540/6470 [43:56<37:05,  1.32it/s]
3|3541|Loss: 0.7456015348434448:  55%|█████▍    | 3540/6470 [43:56<37:05,  1.32it/s]
3|3541|Loss: 0.7456015348434448:  55%|█████▍    | 3541/6470 [43:57<39:07,  1.25it/s]
3|3542|Loss: 0.6726443767547607:  55%|█████▍    | 3541/6470 [43:57<39:07,  1.25it/s]
3|3542|Loss: 0.6726443767547607:  55%|█████▍    | 3542/6470 [43:58<36:56,  1.32it/s]
3|3543|Loss: 0.6770971417427063:  55%|█████▍    | 3542/6470 [43:58<36:56,  1.32it/s]
3|3543|Loss: 0.6770971417427063:  55%|█████▍    | 3543/6470 [43:58<36:23,  1.34it/s]
3|3544|Loss: 0.9082119464874268:  55%|█████▍    | 3543/6470 [43:59<36:23,  1.34it/s]
3|3544|Loss: 0.9082119464874268:  55%|█████▍    | 3544/6470 [43:59<39:50,  1.22it/s]
3|3545|Loss: 0.7709851861000061:  55%|█████▍    | 3544/6470 [44:00<39:50,  1.22it/s]
3|3545|Loss: 0.7709851861000061:  55%|█████▍    | 3545/6470 [44:00<39:55,  1.22it/s]
3|3546|Loss: 0.857742965221405:  55%|█████▍    | 3545/6470 [44:01<39:55,  1.22it/s] 
3|3546|Loss: 0.857742965221405:  55%|█████▍    | 3546/6470 [44:01<40:20,  1.21it/s]
3|3547|Loss: 0.7572063207626343:  55%|█████▍    | 3546/6470 [44:01<40:20,  1.21it/s]
3|3547|Loss: 0.7572063207626343:  55%|█████▍    | 3547/6470 [44:02<39:29,  1.23it/s]
3|3548|Loss: 0.7544402480125427:  55%|█████▍    | 3547/6470 [44:02<39:29,  1.23it/s]
3|3548|Loss: 0.7544402480125427:  55%|█████▍    | 3548/6470 [44:02<37:09,  1.31it/s]
3|3549|Loss: 0.8203718662261963:  55%|█████▍    | 3548/6470 [44:03<37:09,  1.31it/s]
3|3549|Loss: 0.8203718662261963:  55%|█████▍    | 3549/6470 [44:03<34:00,  1.43it/s]
3|3550|Loss: 0.7937392592430115:  55%|█████▍    | 3549/6470 [44:03<34:00,  1.43it/s]
3|3550|Loss: 0.7937392592430115:  55%|█████▍    | 3550/6470 [44:04<36:14,  1.34it/s]
3|3551|Loss: 0.7848216891288757:  55%|█████▍    | 3550/6470 [44:04<36:14,  1.34it/s]
3|3551|Loss: 0.7848216891288757:  55%|█████▍    | 3551/6470 [44:05<37:19,  1.30it/s]
3|3552|Loss: 0.9472715854644775:  55%|█████▍    | 3551/6470 [44:05<37:19,  1.30it/s]
3|3552|Loss: 0.9472715854644775:  55%|█████▍    | 3552/6470 [44:06<39:30,  1.23it/s]
3|3553|Loss: 0.8493568301200867:  55%|█████▍    | 3552/6470 [44:06<39:30,  1.23it/s]
3|3553|Loss: 0.8493568301200867:  55%|█████▍    | 3553/6470 [44:06<41:00,  1.19it/s]
3|3554|Loss: 1.0654511451721191:  55%|█████▍    | 3553/6470 [44:07<41:00,  1.19it/s]
3|3554|Loss: 1.0654511451721191:  55%|█████▍    | 3554/6470 [44:07<39:14,  1.24it/s]
3|3555|Loss: 0.8626177310943604:  55%|█████▍    | 3554/6470 [44:07<39:14,  1.24it/s]
3|3555|Loss: 0.8626177310943604:  55%|█████▍    | 3555/6470 [44:08<34:03,  1.43it/s]
3|3556|Loss: 0.8821500539779663:  55%|█████▍    | 3555/6470 [44:08<34:03,  1.43it/s]
3|3556|Loss: 0.8821500539779663:  55%|█████▍    | 3556/6470 [44:08<33:17,  1.46it/s]
3|3557|Loss: 0.715761661529541:  55%|█████▍    | 3556/6470 [44:09<33:17,  1.46it/s] 
3|3557|Loss: 0.715761661529541:  55%|█████▍    | 3557/6470 [44:09<31:00,  1.57it/s]
3|3558|Loss: 0.9111864566802979:  55%|█████▍    | 3557/6470 [44:09<31:00,  1.57it/s]
3|3558|Loss: 0.9111864566802979:  55%|█████▍    | 3558/6470 [44:09<31:39,  1.53it/s]
3|3559|Loss: 0.8871021270751953:  55%|█████▍    | 3558/6470 [44:10<31:39,  1.53it/s]
3|3559|Loss: 0.8871021270751953:  55%|█████▌    | 3559/6470 [44:10<31:40,  1.53it/s]
3|3560|Loss: 0.7731390595436096:  55%|█████▌    | 3559/6470 [44:11<31:40,  1.53it/s]
3|3560|Loss: 0.7731390595436096:  55%|█████▌    | 3560/6470 [44:11<34:32,  1.40it/s]
3|3561|Loss: 0.8234773278236389:  55%|█████▌    | 3560/6470 [44:11<34:32,  1.40it/s]
3|3561|Loss: 0.8234773278236389:  55%|█████▌    | 3561/6470 [44:12<34:25,  1.41it/s]
3|3562|Loss: 0.756748616695404:  55%|█████▌    | 3561/6470 [44:12<34:25,  1.41it/s] 
3|3562|Loss: 0.756748616695404:  55%|█████▌    | 3562/6470 [44:12<34:07,  1.42it/s]
3|3563|Loss: 0.9585155844688416:  55%|█████▌    | 3562/6470 [44:13<34:07,  1.42it/s]
3|3563|Loss: 0.9585155844688416:  55%|█████▌    | 3563/6470 [44:13<31:26,  1.54it/s]
3|3564|Loss: 0.8155325651168823:  55%|█████▌    | 3563/6470 [44:13<31:26,  1.54it/s]
3|3564|Loss: 0.8155325651168823:  55%|█████▌    | 3564/6470 [44:14<31:55,  1.52it/s]
3|3565|Loss: 0.910099983215332:  55%|█████▌    | 3564/6470 [44:14<31:55,  1.52it/s] 
3|3565|Loss: 0.910099983215332:  55%|█████▌    | 3565/6470 [44:14<35:46,  1.35it/s]
3|3566|Loss: 0.7848846912384033:  55%|█████▌    | 3565/6470 [44:15<35:46,  1.35it/s]
3|3566|Loss: 0.7848846912384033:  55%|█████▌    | 3566/6470 [44:15<37:34,  1.29it/s]
3|3567|Loss: 0.6780704259872437:  55%|█████▌    | 3566/6470 [44:16<37:34,  1.29it/s]
3|3567|Loss: 0.6780704259872437:  55%|█████▌    | 3567/6470 [44:16<34:16,  1.41it/s]
3|3568|Loss: 0.7667885422706604:  55%|█████▌    | 3567/6470 [44:16<34:16,  1.41it/s]
3|3568|Loss: 0.7667885422706604:  55%|█████▌    | 3568/6470 [44:17<33:55,  1.43it/s]
3|3569|Loss: 0.91926109790802:  55%|█████▌    | 3568/6470 [44:17<33:55,  1.43it/s]  
3|3569|Loss: 0.91926109790802:  55%|█████▌    | 3569/6470 [44:17<33:47,  1.43it/s]
3|3570|Loss: 0.8009250164031982:  55%|█████▌    | 3569/6470 [44:18<33:47,  1.43it/s]
3|3570|Loss: 0.8009250164031982:  55%|█████▌    | 3570/6470 [44:18<35:01,  1.38it/s]
3|3571|Loss: 0.7907059192657471:  55%|█████▌    | 3570/6470 [44:19<35:01,  1.38it/s]
3|3571|Loss: 0.7907059192657471:  55%|█████▌    | 3571/6470 [44:19<37:38,  1.28it/s]
3|3572|Loss: 0.8064144849777222:  55%|█████▌    | 3571/6470 [44:19<37:38,  1.28it/s]
3|3572|Loss: 0.8064144849777222:  55%|█████▌    | 3572/6470 [44:20<38:23,  1.26it/s]
3|3573|Loss: 0.8339856266975403:  55%|█████▌    | 3572/6470 [44:20<38:23,  1.26it/s]
3|3573|Loss: 0.8339856266975403:  55%|█████▌    | 3573/6470 [44:21<39:07,  1.23it/s]
3|3574|Loss: 0.7614346742630005:  55%|█████▌    | 3573/6470 [44:21<39:07,  1.23it/s]
3|3574|Loss: 0.7614346742630005:  55%|█████▌    | 3574/6470 [44:21<37:09,  1.30it/s]
3|3575|Loss: 0.9931601881980896:  55%|█████▌    | 3574/6470 [44:22<37:09,  1.30it/s]
3|3575|Loss: 0.9931601881980896:  55%|█████▌    | 3575/6470 [44:22<35:43,  1.35it/s]
3|3576|Loss: 0.8653659224510193:  55%|█████▌    | 3575/6470 [44:22<35:43,  1.35it/s]
3|3576|Loss: 0.8653659224510193:  55%|█████▌    | 3576/6470 [44:23<36:17,  1.33it/s]
3|3577|Loss: 0.7355306148529053:  55%|█████▌    | 3576/6470 [44:23<36:17,  1.33it/s]
3|3577|Loss: 0.7355306148529053:  55%|█████▌    | 3577/6470 [44:23<34:58,  1.38it/s]
3|3578|Loss: 0.9142863750457764:  55%|█████▌    | 3577/6470 [44:24<34:58,  1.38it/s]
3|3578|Loss: 0.9142863750457764:  55%|█████▌    | 3578/6470 [44:24<34:59,  1.38it/s]
3|3579|Loss: 0.9735270738601685:  55%|█████▌    | 3578/6470 [44:25<34:59,  1.38it/s]
3|3579|Loss: 0.9735270738601685:  55%|█████▌    | 3579/6470 [44:25<34:18,  1.40it/s]
3|3580|Loss: 0.6923009157180786:  55%|█████▌    | 3579/6470 [44:25<34:18,  1.40it/s]
3|3580|Loss: 0.6923009157180786:  55%|█████▌    | 3580/6470 [44:25<30:14,  1.59it/s]
3|3581|Loss: 0.8019054532051086:  55%|█████▌    | 3580/6470 [44:26<30:14,  1.59it/s]
3|3581|Loss: 0.8019054532051086:  55%|█████▌    | 3581/6470 [44:26<31:30,  1.53it/s]
3|3582|Loss: 0.7588062286376953:  55%|█████▌    | 3581/6470 [44:26<31:30,  1.53it/s]
3|3582|Loss: 0.7588062286376953:  55%|█████▌    | 3582/6470 [44:27<33:43,  1.43it/s]
3|3583|Loss: 0.8003257513046265:  55%|█████▌    | 3582/6470 [44:27<33:43,  1.43it/s]
3|3583|Loss: 0.8003257513046265:  55%|█████▌    | 3583/6470 [44:28<36:31,  1.32it/s]
3|3584|Loss: 0.790244460105896:  55%|█████▌    | 3583/6470 [44:28<36:31,  1.32it/s] 
3|3584|Loss: 0.790244460105896:  55%|█████▌    | 3584/6470 [44:29<38:10,  1.26it/s]
3|3585|Loss: 0.8335462808609009:  55%|█████▌    | 3584/6470 [44:29<38:10,  1.26it/s]
3|3585|Loss: 0.8335462808609009:  55%|█████▌    | 3585/6470 [44:29<36:56,  1.30it/s]
3|3586|Loss: 0.931329607963562:  55%|█████▌    | 3585/6470 [44:30<36:56,  1.30it/s] 
3|3586|Loss: 0.931329607963562:  55%|█████▌    | 3586/6470 [44:30<36:52,  1.30it/s]
3|3587|Loss: 0.8649821877479553:  55%|█████▌    | 3586/6470 [44:30<36:52,  1.30it/s]
3|3587|Loss: 0.8649821877479553:  55%|█████▌    | 3587/6470 [44:31<37:15,  1.29it/s]
3|3588|Loss: 0.7917855978012085:  55%|█████▌    | 3587/6470 [44:31<37:15,  1.29it/s]
3|3588|Loss: 0.7917855978012085:  55%|█████▌    | 3588/6470 [44:32<36:09,  1.33it/s]
3|3589|Loss: 0.9773000478744507:  55%|█████▌    | 3588/6470 [44:32<36:09,  1.33it/s]
3|3589|Loss: 0.9773000478744507:  55%|█████▌    | 3589/6470 [44:32<34:17,  1.40it/s]
3|3590|Loss: 0.8277669548988342:  55%|█████▌    | 3589/6470 [44:33<34:17,  1.40it/s]
3|3590|Loss: 0.8277669548988342:  55%|█████▌    | 3590/6470 [44:33<32:49,  1.46it/s]
3|3591|Loss: 0.8894643187522888:  55%|█████▌    | 3590/6470 [44:33<32:49,  1.46it/s]
3|3591|Loss: 0.8894643187522888:  56%|█████▌    | 3591/6470 [44:34<33:29,  1.43it/s]
3|3592|Loss: 0.7512930631637573:  56%|█████▌    | 3591/6470 [44:34<33:29,  1.43it/s]
3|3592|Loss: 0.7512930631637573:  56%|█████▌    | 3592/6470 [44:34<33:20,  1.44it/s]
3|3593|Loss: 0.7977474927902222:  56%|█████▌    | 3592/6470 [44:35<33:20,  1.44it/s]
3|3593|Loss: 0.7977474927902222:  56%|█████▌    | 3593/6470 [44:35<33:54,  1.41it/s]
3|3594|Loss: 1.1848807334899902:  56%|█████▌    | 3593/6470 [44:35<33:54,  1.41it/s]
3|3594|Loss: 1.1848807334899902:  56%|█████▌    | 3594/6470 [44:36<32:01,  1.50it/s]
3|3595|Loss: 0.7943210601806641:  56%|█████▌    | 3594/6470 [44:36<32:01,  1.50it/s]
3|3595|Loss: 0.7943210601806641:  56%|█████▌    | 3595/6470 [44:36<30:08,  1.59it/s]
3|3596|Loss: 0.933914840221405:  56%|█████▌    | 3595/6470 [44:36<30:08,  1.59it/s] 
3|3596|Loss: 0.933914840221405:  56%|█████▌    | 3596/6470 [44:37<31:47,  1.51it/s]
3|3597|Loss: 0.794567883014679:  56%|█████▌    | 3596/6470 [44:37<31:47,  1.51it/s]
3|3597|Loss: 0.794567883014679:  56%|█████▌    | 3597/6470 [44:38<33:38,  1.42it/s]
3|3598|Loss: 0.7016386985778809:  56%|█████▌    | 3597/6470 [44:38<33:38,  1.42it/s]
3|3598|Loss: 0.7016386985778809:  56%|█████▌    | 3598/6470 [44:38<36:10,  1.32it/s]
3|3599|Loss: 0.8351048827171326:  56%|█████▌    | 3598/6470 [44:39<36:10,  1.32it/s]
3|3599|Loss: 0.8351048827171326:  56%|█████▌    | 3599/6470 [44:39<35:55,  1.33it/s]
3|3600|Loss: 0.7162461280822754:  56%|█████▌    | 3599/6470 [44:40<35:55,  1.33it/s]
3|3600|Loss: 0.7162461280822754:  56%|█████▌    | 3600/6470 [44:40<34:56,  1.37it/s]
3|3601|Loss: 0.9050036668777466:  56%|█████▌    | 3600/6470 [44:40<34:56,  1.37it/s]
3|3601|Loss: 0.9050036668777466:  56%|█████▌    | 3601/6470 [44:41<35:04,  1.36it/s]
3|3602|Loss: 0.931056022644043:  56%|█████▌    | 3601/6470 [44:41<35:04,  1.36it/s] 
3|3602|Loss: 0.931056022644043:  56%|█████▌    | 3602/6470 [44:41<34:03,  1.40it/s]
3|3603|Loss: 0.7863267064094543:  56%|█████▌    | 3602/6470 [44:42<34:03,  1.40it/s]
3|3603|Loss: 0.7863267064094543:  56%|█████▌    | 3603/6470 [44:42<35:51,  1.33it/s]
3|3604|Loss: 0.6647211909294128:  56%|█████▌    | 3603/6470 [44:43<35:51,  1.33it/s]
3|3604|Loss: 0.6647211909294128:  56%|█████▌    | 3604/6470 [44:43<35:32,  1.34it/s]
3|3605|Loss: 0.9132649898529053:  56%|█████▌    | 3604/6470 [44:43<35:32,  1.34it/s]
3|3605|Loss: 0.9132649898529053:  56%|█████▌    | 3605/6470 [44:43<32:20,  1.48it/s]
3|3606|Loss: 0.8431971073150635:  56%|█████▌    | 3605/6470 [44:44<32:20,  1.48it/s]
3|3606|Loss: 0.8431971073150635:  56%|█████▌    | 3606/6470 [44:44<31:27,  1.52it/s]
3|3607|Loss: 0.7849743962287903:  56%|█████▌    | 3606/6470 [44:44<31:27,  1.52it/s]
3|3607|Loss: 0.7849743962287903:  56%|█████▌    | 3607/6470 [44:45<33:23,  1.43it/s]
3|3608|Loss: 0.8070855736732483:  56%|█████▌    | 3607/6470 [44:45<33:23,  1.43it/s]
3|3608|Loss: 0.8070855736732483:  56%|█████▌    | 3608/6470 [44:46<33:26,  1.43it/s]
3|3609|Loss: 0.9737737774848938:  56%|█████▌    | 3608/6470 [44:46<33:26,  1.43it/s]
3|3609|Loss: 0.9737737774848938:  56%|█████▌    | 3609/6470 [44:46<34:15,  1.39it/s]
3|3610|Loss: 0.8380356431007385:  56%|█████▌    | 3609/6470 [44:47<34:15,  1.39it/s]
3|3610|Loss: 0.8380356431007385:  56%|█████▌    | 3610/6470 [44:47<36:44,  1.30it/s]
3|3611|Loss: 0.7981355786323547:  56%|█████▌    | 3610/6470 [44:48<36:44,  1.30it/s]
3|3611|Loss: 0.7981355786323547:  56%|█████▌    | 3611/6470 [44:48<36:08,  1.32it/s]
3|3612|Loss: 0.9037814140319824:  56%|█████▌    | 3611/6470 [44:48<36:08,  1.32it/s]
3|3612|Loss: 0.9037814140319824:  56%|█████▌    | 3612/6470 [44:49<35:04,  1.36it/s]
3|3613|Loss: 0.817650556564331:  56%|█████▌    | 3612/6470 [44:49<35:04,  1.36it/s] 
3|3613|Loss: 0.817650556564331:  56%|█████▌    | 3613/6470 [44:49<36:19,  1.31it/s]
3|3614|Loss: 0.7967090606689453:  56%|█████▌    | 3613/6470 [44:50<36:19,  1.31it/s]
3|3614|Loss: 0.7967090606689453:  56%|█████▌    | 3614/6470 [44:50<36:32,  1.30it/s]
3|3615|Loss: 0.7342360019683838:  56%|█████▌    | 3614/6470 [44:51<36:32,  1.30it/s]
3|3615|Loss: 0.7342360019683838:  56%|█████▌    | 3615/6470 [44:51<33:03,  1.44it/s]
3|3616|Loss: 0.9367566704750061:  56%|█████▌    | 3615/6470 [44:51<33:03,  1.44it/s]
3|3616|Loss: 0.9367566704750061:  56%|█████▌    | 3616/6470 [44:51<32:58,  1.44it/s]
3|3617|Loss: 0.8382664322853088:  56%|█████▌    | 3616/6470 [44:52<32:58,  1.44it/s]
3|3617|Loss: 0.8382664322853088:  56%|█████▌    | 3617/6470 [44:52<34:01,  1.40it/s]
3|3618|Loss: 0.913124144077301:  56%|█████▌    | 3617/6470 [44:53<34:01,  1.40it/s] 
3|3618|Loss: 0.913124144077301:  56%|█████▌    | 3618/6470 [44:53<35:14,  1.35it/s]
3|3619|Loss: 0.7028234601020813:  56%|█████▌    | 3618/6470 [44:53<35:14,  1.35it/s]
3|3619|Loss: 0.7028234601020813:  56%|█████▌    | 3619/6470 [44:53<30:54,  1.54it/s]
3|3620|Loss: 0.851485013961792:  56%|█████▌    | 3619/6470 [44:54<30:54,  1.54it/s] 
3|3620|Loss: 0.851485013961792:  56%|█████▌    | 3620/6470 [44:54<28:35,  1.66it/s]
3|3621|Loss: 0.9830556511878967:  56%|█████▌    | 3620/6470 [44:54<28:35,  1.66it/s]
3|3621|Loss: 0.9830556511878967:  56%|█████▌    | 3621/6470 [44:54<27:42,  1.71it/s]
3|3622|Loss: 0.8669653534889221:  56%|█████▌    | 3621/6470 [44:55<27:42,  1.71it/s]
3|3622|Loss: 0.8669653534889221:  56%|█████▌    | 3622/6470 [44:55<29:32,  1.61it/s]
3|3623|Loss: 0.8353157639503479:  56%|█████▌    | 3622/6470 [44:56<29:32,  1.61it/s]
3|3623|Loss: 0.8353157639503479:  56%|█████▌    | 3623/6470 [44:56<31:43,  1.50it/s]
3|3624|Loss: 0.7592527866363525:  56%|█████▌    | 3623/6470 [44:56<31:43,  1.50it/s]
3|3624|Loss: 0.7592527866363525:  56%|█████▌    | 3624/6470 [44:57<30:35,  1.55it/s]
3|3625|Loss: 0.7848711013793945:  56%|█████▌    | 3624/6470 [44:57<30:35,  1.55it/s]
3|3625|Loss: 0.7848711013793945:  56%|█████▌    | 3625/6470 [44:57<33:14,  1.43it/s]
3|3626|Loss: 0.8247578144073486:  56%|█████▌    | 3625/6470 [44:58<33:14,  1.43it/s]
3|3626|Loss: 0.8247578144073486:  56%|█████▌    | 3626/6470 [44:58<34:54,  1.36it/s]
3|3627|Loss: 0.7664448618888855:  56%|█████▌    | 3626/6470 [44:59<34:54,  1.36it/s]
3|3627|Loss: 0.7664448618888855:  56%|█████▌    | 3627/6470 [44:59<34:24,  1.38it/s]
3|3628|Loss: 0.9382104873657227:  56%|█████▌    | 3627/6470 [44:59<34:24,  1.38it/s]
3|3628|Loss: 0.9382104873657227:  56%|█████▌    | 3628/6470 [44:59<33:04,  1.43it/s]
3|3629|Loss: 0.7335817217826843:  56%|█████▌    | 3628/6470 [45:00<33:04,  1.43it/s]
3|3629|Loss: 0.7335817217826843:  56%|█████▌    | 3629/6470 [45:00<33:39,  1.41it/s]
3|3630|Loss: 0.7138870358467102:  56%|█████▌    | 3629/6470 [45:01<33:39,  1.41it/s]
3|3630|Loss: 0.7138870358467102:  56%|█████▌    | 3630/6470 [45:01<34:45,  1.36it/s]
3|3631|Loss: 0.8938869833946228:  56%|█████▌    | 3630/6470 [45:01<34:45,  1.36it/s]
3|3631|Loss: 0.8938869833946228:  56%|█████▌    | 3631/6470 [45:02<33:57,  1.39it/s]
3|3632|Loss: 0.8758155703544617:  56%|█████▌    | 3631/6470 [45:02<33:57,  1.39it/s]
3|3632|Loss: 0.8758155703544617:  56%|█████▌    | 3632/6470 [45:02<33:25,  1.41it/s]
3|3633|Loss: 1.1125659942626953:  56%|█████▌    | 3632/6470 [45:03<33:25,  1.41it/s]
3|3633|Loss: 1.1125659942626953:  56%|█████▌    | 3633/6470 [45:03<34:24,  1.37it/s]
3|3634|Loss: 0.7733530402183533:  56%|█████▌    | 3633/6470 [45:04<34:24,  1.37it/s]
3|3634|Loss: 0.7733530402183533:  56%|█████▌    | 3634/6470 [45:04<38:12,  1.24it/s]
3|3635|Loss: 0.9117660522460938:  56%|█████▌    | 3634/6470 [45:05<38:12,  1.24it/s]
3|3635|Loss: 0.9117660522460938:  56%|█████▌    | 3635/6470 [45:05<38:28,  1.23it/s]
3|3636|Loss: 0.7782590389251709:  56%|█████▌    | 3635/6470 [45:05<38:28,  1.23it/s]
3|3636|Loss: 0.7782590389251709:  56%|█████▌    | 3636/6470 [45:06<38:13,  1.24it/s]
3|3637|Loss: 0.9569347500801086:  56%|█████▌    | 3636/6470 [45:06<38:13,  1.24it/s]
3|3637|Loss: 0.9569347500801086:  56%|█████▌    | 3637/6470 [45:06<35:31,  1.33it/s]
3|3638|Loss: 0.8174892663955688:  56%|█████▌    | 3637/6470 [45:07<35:31,  1.33it/s]
3|3638|Loss: 0.8174892663955688:  56%|█████▌    | 3638/6470 [45:07<32:36,  1.45it/s]
3|3639|Loss: 0.8420624136924744:  56%|█████▌    | 3638/6470 [45:07<32:36,  1.45it/s]
3|3639|Loss: 0.8420624136924744:  56%|█████▌    | 3639/6470 [45:08<33:06,  1.43it/s]
3|3640|Loss: 0.8187317252159119:  56%|█████▌    | 3639/6470 [45:08<33:06,  1.43it/s]
3|3640|Loss: 0.8187317252159119:  56%|█████▋    | 3640/6470 [45:08<33:55,  1.39it/s]
3|3641|Loss: 0.8298049569129944:  56%|█████▋    | 3640/6470 [45:09<33:55,  1.39it/s]
3|3641|Loss: 0.8298049569129944:  56%|█████▋    | 3641/6470 [45:09<32:04,  1.47it/s]
3|3642|Loss: 0.9816752076148987:  56%|█████▋    | 3641/6470 [45:09<32:04,  1.47it/s]
3|3642|Loss: 0.9816752076148987:  56%|█████▋    | 3642/6470 [45:10<32:31,  1.45it/s]
3|3643|Loss: 0.8265875577926636:  56%|█████▋    | 3642/6470 [45:10<32:31,  1.45it/s]
3|3643|Loss: 0.8265875577926636:  56%|█████▋    | 3643/6470 [45:11<34:03,  1.38it/s]
3|3644|Loss: 0.9677299857139587:  56%|█████▋    | 3643/6470 [45:11<34:03,  1.38it/s]
3|3644|Loss: 0.9677299857139587:  56%|█████▋    | 3644/6470 [45:11<32:25,  1.45it/s]
3|3645|Loss: 0.7876030802726746:  56%|█████▋    | 3644/6470 [45:12<32:25,  1.45it/s]
3|3645|Loss: 0.7876030802726746:  56%|█████▋    | 3645/6470 [45:12<32:32,  1.45it/s]
3|3646|Loss: 0.6247671842575073:  56%|█████▋    | 3645/6470 [45:12<32:32,  1.45it/s]
3|3646|Loss: 0.6247671842575073:  56%|█████▋    | 3646/6470 [45:13<33:50,  1.39it/s]
3|3647|Loss: 0.7964682579040527:  56%|█████▋    | 3646/6470 [45:13<33:50,  1.39it/s]
3|3647|Loss: 0.7964682579040527:  56%|█████▋    | 3647/6470 [45:13<32:08,  1.46it/s]
3|3648|Loss: 0.5958240628242493:  56%|█████▋    | 3647/6470 [45:14<32:08,  1.46it/s]
3|3648|Loss: 0.5958240628242493:  56%|█████▋    | 3648/6470 [45:14<30:36,  1.54it/s]
3|3649|Loss: 1.011839747428894:  56%|█████▋    | 3648/6470 [45:14<30:36,  1.54it/s] 
3|3649|Loss: 1.011839747428894:  56%|█████▋    | 3649/6470 [45:14<30:54,  1.52it/s]
3|3650|Loss: 0.8461456298828125:  56%|█████▋    | 3649/6470 [45:15<30:54,  1.52it/s]
3|3650|Loss: 0.8461456298828125:  56%|█████▋    | 3650/6470 [45:15<31:14,  1.50it/s]
3|3651|Loss: 0.7777846455574036:  56%|█████▋    | 3650/6470 [45:16<31:14,  1.50it/s]
3|3651|Loss: 0.7777846455574036:  56%|█████▋    | 3651/6470 [45:16<31:43,  1.48it/s]
3|3652|Loss: 0.8037792444229126:  56%|█████▋    | 3651/6470 [45:16<31:43,  1.48it/s]
3|3652|Loss: 0.8037792444229126:  56%|█████▋    | 3652/6470 [45:17<32:55,  1.43it/s]
3|3653|Loss: 0.9635614156723022:  56%|█████▋    | 3652/6470 [45:17<32:55,  1.43it/s]
3|3653|Loss: 0.9635614156723022:  56%|█████▋    | 3653/6470 [45:17<33:45,  1.39it/s]
3|3654|Loss: 0.8083817362785339:  56%|█████▋    | 3653/6470 [45:18<33:45,  1.39it/s]
3|3654|Loss: 0.8083817362785339:  56%|█████▋    | 3654/6470 [45:18<34:46,  1.35it/s]
3|3655|Loss: 0.8890535235404968:  56%|█████▋    | 3654/6470 [45:19<34:46,  1.35it/s]
3|3655|Loss: 0.8890535235404968:  56%|█████▋    | 3655/6470 [45:19<34:22,  1.37it/s]
3|3656|Loss: 0.8432063460350037:  56%|█████▋    | 3655/6470 [45:19<34:22,  1.37it/s]
3|3656|Loss: 0.8432063460350037:  57%|█████▋    | 3656/6470 [45:19<32:19,  1.45it/s]
3|3657|Loss: 0.8011223673820496:  57%|█████▋    | 3656/6470 [45:20<32:19,  1.45it/s]
3|3657|Loss: 0.8011223673820496:  57%|█████▋    | 3657/6470 [45:20<34:29,  1.36it/s]
3|3658|Loss: 0.8235852718353271:  57%|█████▋    | 3657/6470 [45:21<34:29,  1.36it/s]
3|3658|Loss: 0.8235852718353271:  57%|█████▋    | 3658/6470 [45:21<35:03,  1.34it/s]
3|3659|Loss: 0.7501606345176697:  57%|█████▋    | 3658/6470 [45:21<35:03,  1.34it/s]
3|3659|Loss: 0.7501606345176697:  57%|█████▋    | 3659/6470 [45:22<31:10,  1.50it/s]
3|3660|Loss: 0.8207374215126038:  57%|█████▋    | 3659/6470 [45:22<31:10,  1.50it/s]
3|3660|Loss: 0.8207374215126038:  57%|█████▋    | 3660/6470 [45:22<31:48,  1.47it/s]
3|3661|Loss: 0.7792897820472717:  57%|█████▋    | 3660/6470 [45:23<31:48,  1.47it/s]
3|3661|Loss: 0.7792897820472717:  57%|█████▋    | 3661/6470 [45:23<32:18,  1.45it/s]
3|3662|Loss: 0.8272358179092407:  57%|█████▋    | 3661/6470 [45:23<32:18,  1.45it/s]
3|3662|Loss: 0.8272358179092407:  57%|█████▋    | 3662/6470 [45:24<32:57,  1.42it/s]
3|3663|Loss: 0.8503667712211609:  57%|█████▋    | 3662/6470 [45:24<32:57,  1.42it/s]
3|3663|Loss: 0.8503667712211609:  57%|█████▋    | 3663/6470 [45:25<34:13,  1.37it/s]
3|3664|Loss: 0.8343716263771057:  57%|█████▋    | 3663/6470 [45:25<34:13,  1.37it/s]
3|3664|Loss: 0.8343716263771057:  57%|█████▋    | 3664/6470 [45:25<36:05,  1.30it/s]
3|3665|Loss: 0.924852192401886:  57%|█████▋    | 3664/6470 [45:26<36:05,  1.30it/s] 
3|3665|Loss: 0.924852192401886:  57%|█████▋    | 3665/6470 [45:26<36:08,  1.29it/s]
3|3666|Loss: 0.950053334236145:  57%|█████▋    | 3665/6470 [45:27<36:08,  1.29it/s]
3|3666|Loss: 0.950053334236145:  57%|█████▋    | 3666/6470 [45:27<35:15,  1.33it/s]
3|3667|Loss: 0.8238582015037537:  57%|█████▋    | 3666/6470 [45:27<35:15,  1.33it/s]
3|3667|Loss: 0.8238582015037537:  57%|█████▋    | 3667/6470 [45:28<35:32,  1.31it/s]
3|3668|Loss: 0.8850215673446655:  57%|█████▋    | 3667/6470 [45:28<35:32,  1.31it/s]
3|3668|Loss: 0.8850215673446655:  57%|█████▋    | 3668/6470 [45:28<34:51,  1.34it/s]
3|3669|Loss: 0.7223788499832153:  57%|█████▋    | 3668/6470 [45:29<34:51,  1.34it/s]
3|3669|Loss: 0.7223788499832153:  57%|█████▋    | 3669/6470 [45:29<36:38,  1.27it/s]
3|3670|Loss: 0.7029321193695068:  57%|█████▋    | 3669/6470 [45:30<36:38,  1.27it/s]
3|3670|Loss: 0.7029321193695068:  57%|█████▋    | 3670/6470 [45:30<35:35,  1.31it/s]
3|3671|Loss: 0.9039931297302246:  57%|█████▋    | 3670/6470 [45:30<35:35,  1.31it/s]
3|3671|Loss: 0.9039931297302246:  57%|█████▋    | 3671/6470 [45:31<35:42,  1.31it/s]
3|3672|Loss: 0.7795642614364624:  57%|█████▋    | 3671/6470 [45:31<35:42,  1.31it/s]
3|3672|Loss: 0.7795642614364624:  57%|█████▋    | 3672/6470 [45:32<36:46,  1.27it/s]
3|3673|Loss: 0.7506948113441467:  57%|█████▋    | 3672/6470 [45:32<36:46,  1.27it/s]
3|3673|Loss: 0.7506948113441467:  57%|█████▋    | 3673/6470 [45:32<35:17,  1.32it/s]
3|3674|Loss: 0.8922224044799805:  57%|█████▋    | 3673/6470 [45:33<35:17,  1.32it/s]
3|3674|Loss: 0.8922224044799805:  57%|█████▋    | 3674/6470 [45:33<35:15,  1.32it/s]
3|3675|Loss: 0.9164596199989319:  57%|█████▋    | 3674/6470 [45:33<35:15,  1.32it/s]
3|3675|Loss: 0.9164596199989319:  57%|█████▋    | 3675/6470 [45:34<34:44,  1.34it/s]
3|3676|Loss: 0.8542616367340088:  57%|█████▋    | 3675/6470 [45:34<34:44,  1.34it/s]
3|3676|Loss: 0.8542616367340088:  57%|█████▋    | 3676/6470 [45:34<33:54,  1.37it/s]
3|3677|Loss: 0.6881413459777832:  57%|█████▋    | 3676/6470 [45:35<33:54,  1.37it/s]
3|3677|Loss: 0.6881413459777832:  57%|█████▋    | 3677/6470 [45:35<37:15,  1.25it/s]
3|3678|Loss: 0.8950874209403992:  57%|█████▋    | 3677/6470 [45:36<37:15,  1.25it/s]
3|3678|Loss: 0.8950874209403992:  57%|█████▋    | 3678/6470 [45:36<35:52,  1.30it/s]
3|3679|Loss: 0.7817259430885315:  57%|█████▋    | 3678/6470 [45:36<35:52,  1.30it/s]
3|3679|Loss: 0.7817259430885315:  57%|█████▋    | 3679/6470 [45:37<35:40,  1.30it/s]
3|3680|Loss: 0.7887553572654724:  57%|█████▋    | 3679/6470 [45:37<35:40,  1.30it/s]
3|3680|Loss: 0.7887553572654724:  57%|█████▋    | 3680/6470 [45:38<36:59,  1.26it/s]
3|3681|Loss: 0.9218817353248596:  57%|█████▋    | 3680/6470 [45:38<36:59,  1.26it/s]
3|3681|Loss: 0.9218817353248596:  57%|█████▋    | 3681/6470 [45:39<37:08,  1.25it/s]
3|3682|Loss: 0.6997758150100708:  57%|█████▋    | 3681/6470 [45:39<37:08,  1.25it/s]
3|3682|Loss: 0.6997758150100708:  57%|█████▋    | 3682/6470 [45:40<47:51,  1.03s/it]
3|3683|Loss: 0.9027202129364014:  57%|█████▋    | 3682/6470 [45:41<47:51,  1.03s/it]
3|3683|Loss: 0.9027202129364014:  57%|█████▋    | 3683/6470 [45:41<48:44,  1.05s/it]
3|3684|Loss: 0.8162277340888977:  57%|█████▋    | 3683/6470 [45:42<48:44,  1.05s/it]
3|3684|Loss: 0.8162277340888977:  57%|█████▋    | 3684/6470 [45:42<43:27,  1.07it/s]
3|3685|Loss: 0.7617473602294922:  57%|█████▋    | 3684/6470 [45:42<43:27,  1.07it/s]
3|3685|Loss: 0.7617473602294922:  57%|█████▋    | 3685/6470 [45:43<40:56,  1.13it/s]
3|3686|Loss: 0.759519100189209:  57%|█████▋    | 3685/6470 [45:43<40:56,  1.13it/s] 
3|3686|Loss: 0.759519100189209:  57%|█████▋    | 3686/6470 [45:44<41:29,  1.12it/s]
3|3687|Loss: 0.7783255577087402:  57%|█████▋    | 3686/6470 [45:44<41:29,  1.12it/s]
3|3687|Loss: 0.7783255577087402:  57%|█████▋    | 3687/6470 [45:44<39:10,  1.18it/s]
3|3688|Loss: 0.7781298756599426:  57%|█████▋    | 3687/6470 [45:45<39:10,  1.18it/s]
3|3688|Loss: 0.7781298756599426:  57%|█████▋    | 3688/6470 [45:45<37:47,  1.23it/s]
3|3689|Loss: 0.9443102478981018:  57%|█████▋    | 3688/6470 [45:45<37:47,  1.23it/s]
3|3689|Loss: 0.9443102478981018:  57%|█████▋    | 3689/6470 [45:46<36:59,  1.25it/s]
3|3690|Loss: 0.911870002746582:  57%|█████▋    | 3689/6470 [45:46<36:59,  1.25it/s] 
3|3690|Loss: 0.911870002746582:  57%|█████▋    | 3690/6470 [45:47<38:00,  1.22it/s]
3|3691|Loss: 0.7968599200248718:  57%|█████▋    | 3690/6470 [45:47<38:00,  1.22it/s]
3|3691|Loss: 0.7968599200248718:  57%|█████▋    | 3691/6470 [45:47<37:49,  1.22it/s]
3|3692|Loss: 0.8464354276657104:  57%|█████▋    | 3691/6470 [45:48<37:49,  1.22it/s]
3|3692|Loss: 0.8464354276657104:  57%|█████▋    | 3692/6470 [45:48<37:31,  1.23it/s]
3|3693|Loss: 0.8654308319091797:  57%|█████▋    | 3692/6470 [45:49<37:31,  1.23it/s]
3|3693|Loss: 0.8654308319091797:  57%|█████▋    | 3693/6470 [45:49<36:47,  1.26it/s]
3|3694|Loss: 0.8352155089378357:  57%|█████▋    | 3693/6470 [45:49<36:47,  1.26it/s]
3|3694|Loss: 0.8352155089378357:  57%|█████▋    | 3694/6470 [45:50<36:43,  1.26it/s]
3|3695|Loss: 0.9026819467544556:  57%|█████▋    | 3694/6470 [45:50<36:43,  1.26it/s]
3|3695|Loss: 0.9026819467544556:  57%|█████▋    | 3695/6470 [45:51<36:37,  1.26it/s]
3|3696|Loss: 0.7931055426597595:  57%|█████▋    | 3695/6470 [45:51<36:37,  1.26it/s]
3|3696|Loss: 0.7931055426597595:  57%|█████▋    | 3696/6470 [45:51<36:00,  1.28it/s]
3|3697|Loss: 0.7780461311340332:  57%|█████▋    | 3696/6470 [45:52<36:00,  1.28it/s]
3|3697|Loss: 0.7780461311340332:  57%|█████▋    | 3697/6470 [45:52<37:12,  1.24it/s]
3|3698|Loss: 0.8264123201370239:  57%|█████▋    | 3697/6470 [45:53<37:12,  1.24it/s]
3|3698|Loss: 0.8264123201370239:  57%|█████▋    | 3698/6470 [45:53<35:13,  1.31it/s]
3|3699|Loss: 0.7212209105491638:  57%|█████▋    | 3698/6470 [45:53<35:13,  1.31it/s]
3|3699|Loss: 0.7212209105491638:  57%|█████▋    | 3699/6470 [45:54<33:58,  1.36it/s]
3|3700|Loss: 0.7499560117721558:  57%|█████▋    | 3699/6470 [45:54<33:58,  1.36it/s]
3|3700|Loss: 0.7499560117721558:  57%|█████▋    | 3700/6470 [45:54<35:44,  1.29it/s]
3|3701|Loss: 0.8435117602348328:  57%|█████▋    | 3700/6470 [45:55<35:44,  1.29it/s]
3|3701|Loss: 0.8435117602348328:  57%|█████▋    | 3701/6470 [45:55<36:33,  1.26it/s]
3|3702|Loss: 0.8926948308944702:  57%|█████▋    | 3701/6470 [45:56<36:33,  1.26it/s]
3|3702|Loss: 0.8926948308944702:  57%|█████▋    | 3702/6470 [45:56<36:48,  1.25it/s]
3|3703|Loss: 0.8480017781257629:  57%|█████▋    | 3702/6470 [45:57<36:48,  1.25it/s]
3|3703|Loss: 0.8480017781257629:  57%|█████▋    | 3703/6470 [45:57<37:55,  1.22it/s]
3|3704|Loss: 0.8242337703704834:  57%|█████▋    | 3703/6470 [45:57<37:55,  1.22it/s]
3|3704|Loss: 0.8242337703704834:  57%|█████▋    | 3704/6470 [45:58<40:07,  1.15it/s]
3|3705|Loss: 0.8829682469367981:  57%|█████▋    | 3704/6470 [45:58<40:07,  1.15it/s]
3|3705|Loss: 0.8829682469367981:  57%|█████▋    | 3705/6470 [45:59<38:14,  1.21it/s]
3|3706|Loss: 0.7949076294898987:  57%|█████▋    | 3705/6470 [45:59<38:14,  1.21it/s]
3|3706|Loss: 0.7949076294898987:  57%|█████▋    | 3706/6470 [45:59<35:06,  1.31it/s]
3|3707|Loss: 0.828521192073822:  57%|█████▋    | 3706/6470 [46:00<35:06,  1.31it/s] 
3|3707|Loss: 0.828521192073822:  57%|█████▋    | 3707/6470 [46:00<35:03,  1.31it/s]
3|3708|Loss: 0.8733633160591125:  57%|█████▋    | 3707/6470 [46:00<35:03,  1.31it/s]
3|3708|Loss: 0.8733633160591125:  57%|█████▋    | 3708/6470 [46:01<35:45,  1.29it/s]
3|3709|Loss: 0.8109598755836487:  57%|█████▋    | 3708/6470 [46:01<35:45,  1.29it/s]
3|3709|Loss: 0.8109598755836487:  57%|█████▋    | 3709/6470 [46:02<37:21,  1.23it/s]
3|3710|Loss: 0.8804414868354797:  57%|█████▋    | 3709/6470 [46:02<37:21,  1.23it/s]
3|3710|Loss: 0.8804414868354797:  57%|█████▋    | 3710/6470 [46:03<37:33,  1.22it/s]
3|3711|Loss: 0.8170387148857117:  57%|█████▋    | 3710/6470 [46:03<37:33,  1.22it/s]
3|3711|Loss: 0.8170387148857117:  57%|█████▋    | 3711/6470 [46:03<36:00,  1.28it/s]
3|3712|Loss: 0.7275166511535645:  57%|█████▋    | 3711/6470 [46:04<36:00,  1.28it/s]
3|3712|Loss: 0.7275166511535645:  57%|█████▋    | 3712/6470 [46:04<35:30,  1.29it/s]
3|3713|Loss: 0.7179513573646545:  57%|█████▋    | 3712/6470 [46:04<35:30,  1.29it/s]
3|3713|Loss: 0.7179513573646545:  57%|█████▋    | 3713/6470 [46:05<34:33,  1.33it/s]
3|3714|Loss: 0.8234398365020752:  57%|█████▋    | 3713/6470 [46:05<34:33,  1.33it/s]
3|3714|Loss: 0.8234398365020752:  57%|█████▋    | 3714/6470 [46:05<33:27,  1.37it/s]
3|3715|Loss: 0.7403844594955444:  57%|█████▋    | 3714/6470 [46:06<33:27,  1.37it/s]
3|3715|Loss: 0.7403844594955444:  57%|█████▋    | 3715/6470 [46:06<35:41,  1.29it/s]
3|3716|Loss: 0.9918866753578186:  57%|█████▋    | 3715/6470 [46:07<35:41,  1.29it/s]
3|3716|Loss: 0.9918866753578186:  57%|█████▋    | 3716/6470 [46:07<36:25,  1.26it/s]
3|3717|Loss: 0.9382634162902832:  57%|█████▋    | 3716/6470 [46:07<36:25,  1.26it/s]
3|3717|Loss: 0.9382634162902832:  57%|█████▋    | 3717/6470 [46:08<34:06,  1.34it/s]
3|3718|Loss: 0.7556847333908081:  57%|█████▋    | 3717/6470 [46:08<34:06,  1.34it/s]
3|3718|Loss: 0.7556847333908081:  57%|█████▋    | 3718/6470 [46:08<32:43,  1.40it/s]
3|3719|Loss: 0.9577307105064392:  57%|█████▋    | 3718/6470 [46:09<32:43,  1.40it/s]
3|3719|Loss: 0.9577307105064392:  57%|█████▋    | 3719/6470 [46:09<32:12,  1.42it/s]
3|3720|Loss: 0.7934529781341553:  57%|█████▋    | 3719/6470 [46:09<32:12,  1.42it/s]
3|3720|Loss: 0.7934529781341553:  57%|█████▋    | 3720/6470 [46:10<31:21,  1.46it/s]
3|3721|Loss: 0.8425586223602295:  57%|█████▋    | 3720/6470 [46:10<31:21,  1.46it/s]
3|3721|Loss: 0.8425586223602295:  58%|█████▊    | 3721/6470 [46:11<34:02,  1.35it/s]
3|3722|Loss: 0.8211661577224731:  58%|█████▊    | 3721/6470 [46:11<34:02,  1.35it/s]
3|3722|Loss: 0.8211661577224731:  58%|█████▊    | 3722/6470 [46:11<34:51,  1.31it/s]
3|3723|Loss: 0.7507539987564087:  58%|█████▊    | 3722/6470 [46:12<34:51,  1.31it/s]
3|3723|Loss: 0.7507539987564087:  58%|█████▊    | 3723/6470 [46:12<38:09,  1.20it/s]
3|3724|Loss: 0.8720771670341492:  58%|█████▊    | 3723/6470 [46:13<38:09,  1.20it/s]
3|3724|Loss: 0.8720771670341492:  58%|█████▊    | 3724/6470 [46:13<36:33,  1.25it/s]
3|3725|Loss: 0.8113996982574463:  58%|█████▊    | 3724/6470 [46:13<36:33,  1.25it/s]
3|3725|Loss: 0.8113996982574463:  58%|█████▊    | 3725/6470 [46:14<34:22,  1.33it/s]
3|3726|Loss: 0.8293150663375854:  58%|█████▊    | 3725/6470 [46:14<34:22,  1.33it/s]
3|3726|Loss: 0.8293150663375854:  58%|█████▊    | 3726/6470 [46:15<35:35,  1.28it/s]
3|3727|Loss: 0.9241234064102173:  58%|█████▊    | 3726/6470 [46:15<35:35,  1.28it/s]
3|3727|Loss: 0.9241234064102173:  58%|█████▊    | 3727/6470 [46:15<34:38,  1.32it/s]
3|3728|Loss: 0.8152523636817932:  58%|█████▊    | 3727/6470 [46:16<34:38,  1.32it/s]
3|3728|Loss: 0.8152523636817932:  58%|█████▊    | 3728/6470 [46:16<32:50,  1.39it/s]
3|3729|Loss: 0.9017210602760315:  58%|█████▊    | 3728/6470 [46:16<32:50,  1.39it/s]
3|3729|Loss: 0.9017210602760315:  58%|█████▊    | 3729/6470 [46:17<35:02,  1.30it/s]
3|3730|Loss: 0.856263279914856:  58%|█████▊    | 3729/6470 [46:17<35:02,  1.30it/s] 
3|3730|Loss: 0.856263279914856:  58%|█████▊    | 3730/6470 [46:18<35:33,  1.28it/s]
3|3731|Loss: 0.859401285648346:  58%|█████▊    | 3730/6470 [46:18<35:33,  1.28it/s]
3|3731|Loss: 0.859401285648346:  58%|█████▊    | 3731/6470 [46:18<33:57,  1.34it/s]
3|3732|Loss: 0.851960301399231:  58%|█████▊    | 3731/6470 [46:19<33:57,  1.34it/s]
3|3732|Loss: 0.851960301399231:  58%|█████▊    | 3732/6470 [46:19<33:41,  1.35it/s]
3|3733|Loss: 1.0430139303207397:  58%|█████▊    | 3732/6470 [46:19<33:41,  1.35it/s]
3|3733|Loss: 1.0430139303207397:  58%|█████▊    | 3733/6470 [46:20<33:48,  1.35it/s]
3|3734|Loss: 0.8517125248908997:  58%|█████▊    | 3733/6470 [46:20<33:48,  1.35it/s]
3|3734|Loss: 0.8517125248908997:  58%|█████▊    | 3734/6470 [46:20<32:58,  1.38it/s]
3|3735|Loss: 0.775063157081604:  58%|█████▊    | 3734/6470 [46:21<32:58,  1.38it/s] 
3|3735|Loss: 0.775063157081604:  58%|█████▊    | 3735/6470 [46:21<33:22,  1.37it/s]
3|3736|Loss: 0.7843850255012512:  58%|█████▊    | 3735/6470 [46:22<33:22,  1.37it/s]
3|3736|Loss: 0.7843850255012512:  58%|█████▊    | 3736/6470 [46:22<35:35,  1.28it/s]
3|3737|Loss: 0.7964601516723633:  58%|█████▊    | 3736/6470 [46:22<35:35,  1.28it/s]
3|3737|Loss: 0.7964601516723633:  58%|█████▊    | 3737/6470 [46:23<34:55,  1.30it/s]
3|3738|Loss: 0.8934497237205505:  58%|█████▊    | 3737/6470 [46:23<34:55,  1.30it/s]
3|3738|Loss: 0.8934497237205505:  58%|█████▊    | 3738/6470 [46:23<33:43,  1.35it/s]
3|3739|Loss: 0.7964368462562561:  58%|█████▊    | 3738/6470 [46:24<33:43,  1.35it/s]
3|3739|Loss: 0.7964368462562561:  58%|█████▊    | 3739/6470 [46:24<34:00,  1.34it/s]
3|3740|Loss: 0.7801957726478577:  58%|█████▊    | 3739/6470 [46:25<34:00,  1.34it/s]
3|3740|Loss: 0.7801957726478577:  58%|█████▊    | 3740/6470 [46:25<32:41,  1.39it/s]
3|3741|Loss: 0.8835772275924683:  58%|█████▊    | 3740/6470 [46:25<32:41,  1.39it/s]
3|3741|Loss: 0.8835772275924683:  58%|█████▊    | 3741/6470 [46:26<33:21,  1.36it/s]
3|3742|Loss: 0.9892714619636536:  58%|█████▊    | 3741/6470 [46:26<33:21,  1.36it/s]
3|3742|Loss: 0.9892714619636536:  58%|█████▊    | 3742/6470 [46:26<34:58,  1.30it/s]
3|3743|Loss: 0.7349323630332947:  58%|█████▊    | 3742/6470 [46:27<34:58,  1.30it/s]
3|3743|Loss: 0.7349323630332947:  58%|█████▊    | 3743/6470 [46:27<33:03,  1.37it/s]
3|3744|Loss: 0.9350779056549072:  58%|█████▊    | 3743/6470 [46:27<33:03,  1.37it/s]
3|3744|Loss: 0.9350779056549072:  58%|█████▊    | 3744/6470 [46:28<32:36,  1.39it/s]
3|3745|Loss: 0.8814034461975098:  58%|█████▊    | 3744/6470 [46:28<32:36,  1.39it/s]
3|3745|Loss: 0.8814034461975098:  58%|█████▊    | 3745/6470 [46:29<34:09,  1.33it/s]
3|3746|Loss: 0.7425235509872437:  58%|█████▊    | 3745/6470 [46:29<34:09,  1.33it/s]
3|3746|Loss: 0.7425235509872437:  58%|█████▊    | 3746/6470 [46:29<34:50,  1.30it/s]
3|3747|Loss: 0.9309532642364502:  58%|█████▊    | 3746/6470 [46:30<34:50,  1.30it/s]
3|3747|Loss: 0.9309532642364502:  58%|█████▊    | 3747/6470 [46:30<33:18,  1.36it/s]
3|3748|Loss: 0.8395581841468811:  58%|█████▊    | 3747/6470 [46:30<33:18,  1.36it/s]
3|3748|Loss: 0.8395581841468811:  58%|█████▊    | 3748/6470 [46:31<33:12,  1.37it/s]
3|3749|Loss: 0.840605616569519:  58%|█████▊    | 3748/6470 [46:31<33:12,  1.37it/s] 
3|3749|Loss: 0.840605616569519:  58%|█████▊    | 3749/6470 [46:31<32:17,  1.40it/s]
3|3750|Loss: 0.897092878818512:  58%|█████▊    | 3749/6470 [46:32<32:17,  1.40it/s]
3|3750|Loss: 0.897092878818512:  58%|█████▊    | 3750/6470 [46:32<30:10,  1.50it/s]
3|3751|Loss: 0.85948246717453:  58%|█████▊    | 3750/6470 [46:32<30:10,  1.50it/s] 
3|3751|Loss: 0.85948246717453:  58%|█████▊    | 3751/6470 [46:33<30:54,  1.47it/s]
3|3752|Loss: 0.8843861818313599:  58%|█████▊    | 3751/6470 [46:33<30:54,  1.47it/s]
3|3752|Loss: 0.8843861818313599:  58%|█████▊    | 3752/6470 [46:33<30:55,  1.46it/s]
3|3753|Loss: 0.9196822643280029:  58%|█████▊    | 3752/6470 [46:34<30:55,  1.46it/s]
3|3753|Loss: 0.9196822643280029:  58%|█████▊    | 3753/6470 [46:34<30:37,  1.48it/s]
3|3754|Loss: 0.8716092705726624:  58%|█████▊    | 3753/6470 [46:35<30:37,  1.48it/s]
3|3754|Loss: 0.8716092705726624:  58%|█████▊    | 3754/6470 [46:35<31:32,  1.43it/s]
3|3755|Loss: 0.7501933574676514:  58%|█████▊    | 3754/6470 [46:35<31:32,  1.43it/s]
3|3755|Loss: 0.7501933574676514:  58%|█████▊    | 3755/6470 [46:36<31:45,  1.43it/s]
3|3756|Loss: 0.9215916395187378:  58%|█████▊    | 3755/6470 [46:36<31:45,  1.43it/s]
3|3756|Loss: 0.9215916395187378:  58%|█████▊    | 3756/6470 [46:36<29:54,  1.51it/s]
3|3757|Loss: 0.8997083902359009:  58%|█████▊    | 3756/6470 [46:37<29:54,  1.51it/s]
3|3757|Loss: 0.8997083902359009:  58%|█████▊    | 3757/6470 [46:37<32:13,  1.40it/s]
3|3758|Loss: 0.9005056619644165:  58%|█████▊    | 3757/6470 [46:37<32:13,  1.40it/s]
3|3758|Loss: 0.9005056619644165:  58%|█████▊    | 3758/6470 [46:38<30:49,  1.47it/s]
3|3759|Loss: 0.7909107208251953:  58%|█████▊    | 3758/6470 [46:38<30:49,  1.47it/s]
3|3759|Loss: 0.7909107208251953:  58%|█████▊    | 3759/6470 [46:38<33:21,  1.35it/s]
3|3760|Loss: 0.7300027012825012:  58%|█████▊    | 3759/6470 [46:39<33:21,  1.35it/s]
3|3760|Loss: 0.7300027012825012:  58%|█████▊    | 3760/6470 [46:39<36:02,  1.25it/s]
3|3761|Loss: 0.855121910572052:  58%|█████▊    | 3760/6470 [46:40<36:02,  1.25it/s] 
3|3761|Loss: 0.855121910572052:  58%|█████▊    | 3761/6470 [46:40<33:56,  1.33it/s]
3|3762|Loss: 0.8509021997451782:  58%|█████▊    | 3761/6470 [46:40<33:56,  1.33it/s]
3|3762|Loss: 0.8509021997451782:  58%|█████▊    | 3762/6470 [46:41<33:49,  1.33it/s]
3|3763|Loss: 0.8035737872123718:  58%|█████▊    | 3762/6470 [46:41<33:49,  1.33it/s]
3|3763|Loss: 0.8035737872123718:  58%|█████▊    | 3763/6470 [46:41<32:38,  1.38it/s]
3|3764|Loss: 0.9396239519119263:  58%|█████▊    | 3763/6470 [46:42<32:38,  1.38it/s]
3|3764|Loss: 0.9396239519119263:  58%|█████▊    | 3764/6470 [46:42<33:14,  1.36it/s]
3|3765|Loss: 0.8960047960281372:  58%|█████▊    | 3764/6470 [46:43<33:14,  1.36it/s]
3|3765|Loss: 0.8960047960281372:  58%|█████▊    | 3765/6470 [46:43<32:31,  1.39it/s]
3|3766|Loss: 0.9224643707275391:  58%|█████▊    | 3765/6470 [46:43<32:31,  1.39it/s]
3|3766|Loss: 0.9224643707275391:  58%|█████▊    | 3766/6470 [46:44<34:12,  1.32it/s]
3|3767|Loss: 0.7227175831794739:  58%|█████▊    | 3766/6470 [46:44<34:12,  1.32it/s]
3|3767|Loss: 0.7227175831794739:  58%|█████▊    | 3767/6470 [46:44<30:08,  1.49it/s]
3|3768|Loss: 0.8466350436210632:  58%|█████▊    | 3767/6470 [46:45<30:08,  1.49it/s]
3|3768|Loss: 0.8466350436210632:  58%|█████▊    | 3768/6470 [46:45<29:11,  1.54it/s]
3|3769|Loss: 0.8740893006324768:  58%|█████▊    | 3768/6470 [46:45<29:11,  1.54it/s]
3|3769|Loss: 0.8740893006324768:  58%|█████▊    | 3769/6470 [46:45<28:15,  1.59it/s]
3|3770|Loss: 0.968851625919342:  58%|█████▊    | 3769/6470 [46:46<28:15,  1.59it/s] 
3|3770|Loss: 0.968851625919342:  58%|█████▊    | 3770/6470 [46:46<29:24,  1.53it/s]
3|3771|Loss: 0.9475283026695251:  58%|█████▊    | 3770/6470 [46:47<29:24,  1.53it/s]
3|3771|Loss: 0.9475283026695251:  58%|█████▊    | 3771/6470 [46:47<31:17,  1.44it/s]
3|3772|Loss: 0.7541326880455017:  58%|█████▊    | 3771/6470 [46:47<31:17,  1.44it/s]
3|3772|Loss: 0.7541326880455017:  58%|█████▊    | 3772/6470 [46:47<29:13,  1.54it/s]
3|3773|Loss: 0.765441358089447:  58%|█████▊    | 3772/6470 [46:48<29:13,  1.54it/s] 
3|3773|Loss: 0.765441358089447:  58%|█████▊    | 3773/6470 [46:48<28:03,  1.60it/s]
3|3774|Loss: 0.8576934337615967:  58%|█████▊    | 3773/6470 [46:48<28:03,  1.60it/s]
3|3774|Loss: 0.8576934337615967:  58%|█████▊    | 3774/6470 [46:49<28:20,  1.59it/s]
3|3775|Loss: 0.6141179800033569:  58%|█████▊    | 3774/6470 [46:49<28:20,  1.59it/s]
3|3775|Loss: 0.6141179800033569:  58%|█████▊    | 3775/6470 [46:49<30:09,  1.49it/s]
3|3776|Loss: 0.8890441060066223:  58%|█████▊    | 3775/6470 [46:50<30:09,  1.49it/s]
3|3776|Loss: 0.8890441060066223:  58%|█████▊    | 3776/6470 [46:50<33:17,  1.35it/s]
3|3777|Loss: 0.5862281322479248:  58%|█████▊    | 3776/6470 [46:51<33:17,  1.35it/s]
3|3777|Loss: 0.5862281322479248:  58%|█████▊    | 3777/6470 [46:51<37:30,  1.20it/s]
3|3778|Loss: 0.8537805080413818:  58%|█████▊    | 3777/6470 [46:52<37:30,  1.20it/s]
3|3778|Loss: 0.8537805080413818:  58%|█████▊    | 3778/6470 [46:52<36:25,  1.23it/s]
3|3779|Loss: 0.8636729717254639:  58%|█████▊    | 3778/6470 [46:53<36:25,  1.23it/s]
3|3779|Loss: 0.8636729717254639:  58%|█████▊    | 3779/6470 [46:53<35:53,  1.25it/s]
3|3780|Loss: 0.7419014573097229:  58%|█████▊    | 3779/6470 [46:53<35:53,  1.25it/s]
3|3780|Loss: 0.7419014573097229:  58%|█████▊    | 3780/6470 [46:54<35:49,  1.25it/s]
3|3781|Loss: 0.8614550828933716:  58%|█████▊    | 3780/6470 [46:54<35:49,  1.25it/s]
3|3781|Loss: 0.8614550828933716:  58%|█████▊    | 3781/6470 [46:54<35:32,  1.26it/s]
3|3782|Loss: 0.7321553826332092:  58%|█████▊    | 3781/6470 [46:55<35:32,  1.26it/s]
3|3782|Loss: 0.7321553826332092:  58%|█████▊    | 3782/6470 [46:55<36:32,  1.23it/s]
3|3783|Loss: 0.8553516864776611:  58%|█████▊    | 3782/6470 [46:56<36:32,  1.23it/s]
3|3783|Loss: 0.8553516864776611:  58%|█████▊    | 3783/6470 [46:56<37:44,  1.19it/s]
3|3784|Loss: 1.0122798681259155:  58%|█████▊    | 3783/6470 [46:57<37:44,  1.19it/s]
3|3784|Loss: 1.0122798681259155:  58%|█████▊    | 3784/6470 [46:57<33:01,  1.36it/s]
3|3785|Loss: 0.7483963966369629:  58%|█████▊    | 3784/6470 [46:57<33:01,  1.36it/s]
3|3785|Loss: 0.7483963966369629:  59%|█████▊    | 3785/6470 [46:57<31:15,  1.43it/s]
3|3786|Loss: 0.7780799269676208:  59%|█████▊    | 3785/6470 [46:58<31:15,  1.43it/s]
3|3786|Loss: 0.7780799269676208:  59%|█████▊    | 3786/6470 [46:58<30:56,  1.45it/s]
3|3787|Loss: 0.8060966730117798:  59%|█████▊    | 3786/6470 [46:58<30:56,  1.45it/s]
3|3787|Loss: 0.8060966730117798:  59%|█████▊    | 3787/6470 [46:59<32:13,  1.39it/s]
3|3788|Loss: 0.7247045636177063:  59%|█████▊    | 3787/6470 [46:59<32:13,  1.39it/s]
3|3788|Loss: 0.7247045636177063:  59%|█████▊    | 3788/6470 [46:59<28:55,  1.55it/s]
3|3789|Loss: 0.7233779430389404:  59%|█████▊    | 3788/6470 [47:00<28:55,  1.55it/s]
3|3789|Loss: 0.7233779430389404:  59%|█████▊    | 3789/6470 [47:00<29:43,  1.50it/s]
3|3790|Loss: 0.9141822457313538:  59%|█████▊    | 3789/6470 [47:00<29:43,  1.50it/s]
3|3790|Loss: 0.9141822457313538:  59%|█████▊    | 3790/6470 [47:01<30:16,  1.48it/s]
3|3791|Loss: 0.8420702219009399:  59%|█████▊    | 3790/6470 [47:01<30:16,  1.48it/s]
3|3791|Loss: 0.8420702219009399:  59%|█████▊    | 3791/6470 [47:01<30:20,  1.47it/s]
3|3792|Loss: 0.9147450923919678:  59%|█████▊    | 3791/6470 [47:02<30:20,  1.47it/s]
3|3792|Loss: 0.9147450923919678:  59%|█████▊    | 3792/6470 [47:02<31:40,  1.41it/s]
3|3793|Loss: 0.8288261294364929:  59%|█████▊    | 3792/6470 [47:03<31:40,  1.41it/s]
3|3793|Loss: 0.8288261294364929:  59%|█████▊    | 3793/6470 [47:03<30:22,  1.47it/s]
3|3794|Loss: 0.8652377724647522:  59%|█████▊    | 3793/6470 [47:03<30:22,  1.47it/s]
3|3794|Loss: 0.8652377724647522:  59%|█████▊    | 3794/6470 [47:04<34:17,  1.30it/s]
3|3795|Loss: 0.9486977458000183:  59%|█████▊    | 3794/6470 [47:04<34:17,  1.30it/s]
3|3795|Loss: 0.9486977458000183:  59%|█████▊    | 3795/6470 [47:05<36:22,  1.23it/s]
3|3796|Loss: 0.8591684103012085:  59%|█████▊    | 3795/6470 [47:05<36:22,  1.23it/s]
3|3796|Loss: 0.8591684103012085:  59%|█████▊    | 3796/6470 [47:05<33:22,  1.34it/s]
3|3797|Loss: 0.8497673273086548:  59%|█████▊    | 3796/6470 [47:06<33:22,  1.34it/s]
3|3797|Loss: 0.8497673273086548:  59%|█████▊    | 3797/6470 [47:06<32:10,  1.38it/s]
3|3798|Loss: 0.7926463484764099:  59%|█████▊    | 3797/6470 [47:06<32:10,  1.38it/s]
3|3798|Loss: 0.7926463484764099:  59%|█████▊    | 3798/6470 [47:07<32:32,  1.37it/s]
3|3799|Loss: 0.9344421029090881:  59%|█████▊    | 3798/6470 [47:07<32:32,  1.37it/s]
3|3799|Loss: 0.9344421029090881:  59%|█████▊    | 3799/6470 [47:08<34:29,  1.29it/s]
3|3800|Loss: 0.5969989895820618:  59%|█████▊    | 3799/6470 [47:08<34:29,  1.29it/s]
3|3800|Loss: 0.5969989895820618:  59%|█████▊    | 3800/6470 [47:08<34:00,  1.31it/s]
3|3801|Loss: 0.9259253144264221:  59%|█████▊    | 3800/6470 [47:09<34:00,  1.31it/s]
3|3801|Loss: 0.9259253144264221:  59%|█████▊    | 3801/6470 [47:09<31:25,  1.42it/s]
3|3802|Loss: 0.7302660942077637:  59%|█████▊    | 3801/6470 [47:09<31:25,  1.42it/s]
3|3802|Loss: 0.7302660942077637:  59%|█████▉    | 3802/6470 [47:09<29:57,  1.48it/s]
3|3803|Loss: 0.7541107535362244:  59%|█████▉    | 3802/6470 [47:10<29:57,  1.48it/s]
3|3803|Loss: 0.7541107535362244:  59%|█████▉    | 3803/6470 [47:10<32:12,  1.38it/s]
3|3804|Loss: 0.903399646282196:  59%|█████▉    | 3803/6470 [47:11<32:12,  1.38it/s] 
3|3804|Loss: 0.903399646282196:  59%|█████▉    | 3804/6470 [47:11<30:27,  1.46it/s]
3|3805|Loss: 0.7258040904998779:  59%|█████▉    | 3804/6470 [47:11<30:27,  1.46it/s]
3|3805|Loss: 0.7258040904998779:  59%|█████▉    | 3805/6470 [47:12<32:11,  1.38it/s]
3|3806|Loss: 0.6847532987594604:  59%|█████▉    | 3805/6470 [47:12<32:11,  1.38it/s]
3|3806|Loss: 0.6847532987594604:  59%|█████▉    | 3806/6470 [47:13<33:47,  1.31it/s]
3|3807|Loss: 0.944033682346344:  59%|█████▉    | 3806/6470 [47:13<33:47,  1.31it/s] 
3|3807|Loss: 0.944033682346344:  59%|█████▉    | 3807/6470 [47:13<33:21,  1.33it/s]
3|3808|Loss: 0.7946955561637878:  59%|█████▉    | 3807/6470 [47:14<33:21,  1.33it/s]
3|3808|Loss: 0.7946955561637878:  59%|█████▉    | 3808/6470 [47:14<31:54,  1.39it/s]
3|3809|Loss: 0.9214457273483276:  59%|█████▉    | 3808/6470 [47:14<31:54,  1.39it/s]
3|3809|Loss: 0.9214457273483276:  59%|█████▉    | 3809/6470 [47:15<31:23,  1.41it/s]
3|3810|Loss: 0.6973883509635925:  59%|█████▉    | 3809/6470 [47:15<31:23,  1.41it/s]
3|3810|Loss: 0.6973883509635925:  59%|█████▉    | 3810/6470 [47:15<31:49,  1.39it/s]
3|3811|Loss: 0.8483137488365173:  59%|█████▉    | 3810/6470 [47:16<31:49,  1.39it/s]
3|3811|Loss: 0.8483137488365173:  59%|█████▉    | 3811/6470 [47:16<32:40,  1.36it/s]
3|3812|Loss: 0.8435183167457581:  59%|█████▉    | 3811/6470 [47:17<32:40,  1.36it/s]
3|3812|Loss: 0.8435183167457581:  59%|█████▉    | 3812/6470 [47:17<32:31,  1.36it/s]
3|3813|Loss: 0.7578118443489075:  59%|█████▉    | 3812/6470 [47:17<32:31,  1.36it/s]
3|3813|Loss: 0.7578118443489075:  59%|█████▉    | 3813/6470 [47:18<32:56,  1.34it/s]
3|3814|Loss: 0.722985029220581:  59%|█████▉    | 3813/6470 [47:18<32:56,  1.34it/s] 
3|3814|Loss: 0.722985029220581:  59%|█████▉    | 3814/6470 [47:18<32:14,  1.37it/s]
3|3815|Loss: 0.8566102981567383:  59%|█████▉    | 3814/6470 [47:19<32:14,  1.37it/s]
3|3815|Loss: 0.8566102981567383:  59%|█████▉    | 3815/6470 [47:19<32:58,  1.34it/s]
3|3816|Loss: 0.6792697310447693:  59%|█████▉    | 3815/6470 [47:20<32:58,  1.34it/s]
3|3816|Loss: 0.6792697310447693:  59%|█████▉    | 3816/6470 [47:20<31:47,  1.39it/s]
3|3817|Loss: 0.7027841806411743:  59%|█████▉    | 3816/6470 [47:20<31:47,  1.39it/s]
3|3817|Loss: 0.7027841806411743:  59%|█████▉    | 3817/6470 [47:20<31:47,  1.39it/s]
3|3818|Loss: 0.7681466937065125:  59%|█████▉    | 3817/6470 [47:21<31:47,  1.39it/s]
3|3818|Loss: 0.7681466937065125:  59%|█████▉    | 3818/6470 [47:21<33:48,  1.31it/s]
3|3819|Loss: 0.8941894769668579:  59%|█████▉    | 3818/6470 [47:22<33:48,  1.31it/s]
3|3819|Loss: 0.8941894769668579:  59%|█████▉    | 3819/6470 [47:22<32:40,  1.35it/s]
3|3820|Loss: 0.8685156106948853:  59%|█████▉    | 3819/6470 [47:22<32:40,  1.35it/s]
3|3820|Loss: 0.8685156106948853:  59%|█████▉    | 3820/6470 [47:23<34:16,  1.29it/s]
3|3821|Loss: 0.8377053737640381:  59%|█████▉    | 3820/6470 [47:23<34:16,  1.29it/s]
3|3821|Loss: 0.8377053737640381:  59%|█████▉    | 3821/6470 [47:24<34:35,  1.28it/s]
3|3822|Loss: 0.7284112572669983:  59%|█████▉    | 3821/6470 [47:24<34:35,  1.28it/s]
3|3822|Loss: 0.7284112572669983:  59%|█████▉    | 3822/6470 [47:25<34:39,  1.27it/s]
3|3823|Loss: 0.7559869885444641:  59%|█████▉    | 3822/6470 [47:25<34:39,  1.27it/s]
3|3823|Loss: 0.7559869885444641:  59%|█████▉    | 3823/6470 [47:25<34:54,  1.26it/s]
3|3824|Loss: 0.7006288766860962:  59%|█████▉    | 3823/6470 [47:26<34:54,  1.26it/s]
3|3824|Loss: 0.7006288766860962:  59%|█████▉    | 3824/6470 [47:26<36:11,  1.22it/s]
3|3825|Loss: 0.883718729019165:  59%|█████▉    | 3824/6470 [47:27<36:11,  1.22it/s] 
3|3825|Loss: 0.883718729019165:  59%|█████▉    | 3825/6470 [47:27<35:59,  1.23it/s]
3|3826|Loss: 0.8693675398826599:  59%|█████▉    | 3825/6470 [47:27<35:59,  1.23it/s]
3|3826|Loss: 0.8693675398826599:  59%|█████▉    | 3826/6470 [47:28<34:40,  1.27it/s]
3|3827|Loss: 0.8993641138076782:  59%|█████▉    | 3826/6470 [47:28<34:40,  1.27it/s]
3|3827|Loss: 0.8993641138076782:  59%|█████▉    | 3827/6470 [47:29<34:39,  1.27it/s]
3|3828|Loss: 0.7503638863563538:  59%|█████▉    | 3827/6470 [47:29<34:39,  1.27it/s]
3|3828|Loss: 0.7503638863563538:  59%|█████▉    | 3828/6470 [47:29<33:47,  1.30it/s]
3|3829|Loss: 0.6504718661308289:  59%|█████▉    | 3828/6470 [47:30<33:47,  1.30it/s]
3|3829|Loss: 0.6504718661308289:  59%|█████▉    | 3829/6470 [47:30<34:02,  1.29it/s]
3|3830|Loss: 0.7963584065437317:  59%|█████▉    | 3829/6470 [47:30<34:02,  1.29it/s]
3|3830|Loss: 0.7963584065437317:  59%|█████▉    | 3830/6470 [47:31<33:20,  1.32it/s]
3|3831|Loss: 0.7564868927001953:  59%|█████▉    | 3830/6470 [47:31<33:20,  1.32it/s]
3|3831|Loss: 0.7564868927001953:  59%|█████▉    | 3831/6470 [47:31<30:08,  1.46it/s]
3|3832|Loss: 0.8349367380142212:  59%|█████▉    | 3831/6470 [47:32<30:08,  1.46it/s]
3|3832|Loss: 0.8349367380142212:  59%|█████▉    | 3832/6470 [47:32<30:41,  1.43it/s]
3|3833|Loss: 0.897130012512207:  59%|█████▉    | 3832/6470 [47:32<30:41,  1.43it/s] 
3|3833|Loss: 0.897130012512207:  59%|█████▉    | 3833/6470 [47:33<30:59,  1.42it/s]
3|3834|Loss: 0.8859659433364868:  59%|█████▉    | 3833/6470 [47:33<30:59,  1.42it/s]
3|3834|Loss: 0.8859659433364868:  59%|█████▉    | 3834/6470 [47:33<29:26,  1.49it/s]
3|3835|Loss: 0.837958037853241:  59%|█████▉    | 3834/6470 [47:34<29:26,  1.49it/s] 
3|3835|Loss: 0.837958037853241:  59%|█████▉    | 3835/6470 [47:34<28:57,  1.52it/s]
3|3836|Loss: 0.9194729328155518:  59%|█████▉    | 3835/6470 [47:34<28:57,  1.52it/s]
3|3836|Loss: 0.9194729328155518:  59%|█████▉    | 3836/6470 [47:35<29:01,  1.51it/s]
3|3837|Loss: 0.8477998971939087:  59%|█████▉    | 3836/6470 [47:35<29:01,  1.51it/s]
3|3837|Loss: 0.8477998971939087:  59%|█████▉    | 3837/6470 [47:35<30:28,  1.44it/s]
3|3838|Loss: 0.8087681531906128:  59%|█████▉    | 3837/6470 [47:36<30:28,  1.44it/s]
3|3838|Loss: 0.8087681531906128:  59%|█████▉    | 3838/6470 [47:36<30:28,  1.44it/s]
3|3839|Loss: 0.8101334571838379:  59%|█████▉    | 3838/6470 [47:37<30:28,  1.44it/s]
3|3839|Loss: 0.8101334571838379:  59%|█████▉    | 3839/6470 [47:37<33:03,  1.33it/s]
3|3840|Loss: 0.8566761016845703:  59%|█████▉    | 3839/6470 [47:37<33:03,  1.33it/s]
3|3840|Loss: 0.8566761016845703:  59%|█████▉    | 3840/6470 [47:38<34:01,  1.29it/s]
3|3841|Loss: 0.9914271235466003:  59%|█████▉    | 3840/6470 [47:38<34:01,  1.29it/s]
3|3841|Loss: 0.9914271235466003:  59%|█████▉    | 3841/6470 [47:39<34:17,  1.28it/s]
3|3842|Loss: 0.746593177318573:  59%|█████▉    | 3841/6470 [47:39<34:17,  1.28it/s] 
3|3842|Loss: 0.746593177318573:  59%|█████▉    | 3842/6470 [47:39<35:14,  1.24it/s]
3|3843|Loss: 0.760972797870636:  59%|█████▉    | 3842/6470 [47:40<35:14,  1.24it/s]
3|3843|Loss: 0.760972797870636:  59%|█████▉    | 3843/6470 [47:40<35:13,  1.24it/s]
3|3844|Loss: 0.7848175168037415:  59%|█████▉    | 3843/6470 [47:41<35:13,  1.24it/s]
3|3844|Loss: 0.7848175168037415:  59%|█████▉    | 3844/6470 [47:41<35:03,  1.25it/s]
3|3845|Loss: 0.8865846991539001:  59%|█████▉    | 3844/6470 [47:41<35:03,  1.25it/s]
3|3845|Loss: 0.8865846991539001:  59%|█████▉    | 3845/6470 [47:42<35:06,  1.25it/s]
3|3846|Loss: 0.8791983127593994:  59%|█████▉    | 3845/6470 [47:42<35:06,  1.25it/s]
3|3846|Loss: 0.8791983127593994:  59%|█████▉    | 3846/6470 [47:43<33:20,  1.31it/s]
3|3847|Loss: 0.7853008508682251:  59%|█████▉    | 3846/6470 [47:43<33:20,  1.31it/s]
3|3847|Loss: 0.7853008508682251:  59%|█████▉    | 3847/6470 [47:43<31:19,  1.40it/s]
3|3848|Loss: 0.7562761902809143:  59%|█████▉    | 3847/6470 [47:44<31:19,  1.40it/s]
3|3848|Loss: 0.7562761902809143:  59%|█████▉    | 3848/6470 [47:44<32:35,  1.34it/s]
3|3849|Loss: 0.7248899340629578:  59%|█████▉    | 3848/6470 [47:44<32:35,  1.34it/s]
3|3849|Loss: 0.7248899340629578:  59%|█████▉    | 3849/6470 [47:45<30:49,  1.42it/s]
3|3850|Loss: 0.8338387608528137:  59%|█████▉    | 3849/6470 [47:45<30:49,  1.42it/s]
3|3850|Loss: 0.8338387608528137:  60%|█████▉    | 3850/6470 [47:45<29:32,  1.48it/s]
3|3851|Loss: 0.8665722608566284:  60%|█████▉    | 3850/6470 [47:46<29:32,  1.48it/s]
3|3851|Loss: 0.8665722608566284:  60%|█████▉    | 3851/6470 [47:46<30:08,  1.45it/s]
3|3852|Loss: 0.6740196943283081:  60%|█████▉    | 3851/6470 [47:46<30:08,  1.45it/s]
3|3852|Loss: 0.6740196943283081:  60%|█████▉    | 3852/6470 [47:47<29:57,  1.46it/s]
3|3853|Loss: 0.8285766243934631:  60%|█████▉    | 3852/6470 [47:47<29:57,  1.46it/s]
3|3853|Loss: 0.8285766243934631:  60%|█████▉    | 3853/6470 [47:47<30:32,  1.43it/s]
3|3854|Loss: 0.7872530221939087:  60%|█████▉    | 3853/6470 [47:48<30:32,  1.43it/s]
3|3854|Loss: 0.7872530221939087:  60%|█████▉    | 3854/6470 [47:48<30:12,  1.44it/s]
3|3855|Loss: 0.8912331461906433:  60%|█████▉    | 3854/6470 [47:48<30:12,  1.44it/s]
3|3855|Loss: 0.8912331461906433:  60%|█████▉    | 3855/6470 [47:49<31:47,  1.37it/s]
3|3856|Loss: 0.8862666487693787:  60%|█████▉    | 3855/6470 [47:49<31:47,  1.37it/s]
3|3856|Loss: 0.8862666487693787:  60%|█████▉    | 3856/6470 [47:49<30:33,  1.43it/s]
3|3857|Loss: 0.628154456615448:  60%|█████▉    | 3856/6470 [47:50<30:33,  1.43it/s] 
3|3857|Loss: 0.628154456615448:  60%|█████▉    | 3857/6470 [47:50<31:43,  1.37it/s]
3|3858|Loss: 0.8398709893226624:  60%|█████▉    | 3857/6470 [47:51<31:43,  1.37it/s]
3|3858|Loss: 0.8398709893226624:  60%|█████▉    | 3858/6470 [47:51<32:36,  1.33it/s]
3|3859|Loss: 0.8367749452590942:  60%|█████▉    | 3858/6470 [47:51<32:36,  1.33it/s]
3|3859|Loss: 0.8367749452590942:  60%|█████▉    | 3859/6470 [47:52<34:21,  1.27it/s]
3|3860|Loss: 1.0248438119888306:  60%|█████▉    | 3859/6470 [47:52<34:21,  1.27it/s]
3|3860|Loss: 1.0248438119888306:  60%|█████▉    | 3860/6470 [47:53<32:20,  1.34it/s]
3|3861|Loss: 0.7760774493217468:  60%|█████▉    | 3860/6470 [47:53<32:20,  1.34it/s]
3|3861|Loss: 0.7760774493217468:  60%|█████▉    | 3861/6470 [47:53<32:25,  1.34it/s]
3|3862|Loss: 0.8908926844596863:  60%|█████▉    | 3861/6470 [47:54<32:25,  1.34it/s]
3|3862|Loss: 0.8908926844596863:  60%|█████▉    | 3862/6470 [47:54<34:40,  1.25it/s]
3|3863|Loss: 0.8415418267250061:  60%|█████▉    | 3862/6470 [47:55<34:40,  1.25it/s]
3|3863|Loss: 0.8415418267250061:  60%|█████▉    | 3863/6470 [47:55<36:43,  1.18it/s]
3|3864|Loss: 0.8744555711746216:  60%|█████▉    | 3863/6470 [47:56<36:43,  1.18it/s]
3|3864|Loss: 0.8744555711746216:  60%|█████▉    | 3864/6470 [47:56<35:37,  1.22it/s]
3|3865|Loss: 0.6160494089126587:  60%|█████▉    | 3864/6470 [47:56<35:37,  1.22it/s]
3|3865|Loss: 0.6160494089126587:  60%|█████▉    | 3865/6470 [47:57<37:37,  1.15it/s]
3|3866|Loss: 0.8795426487922668:  60%|█████▉    | 3865/6470 [47:57<37:37,  1.15it/s]
3|3866|Loss: 0.8795426487922668:  60%|█████▉    | 3866/6470 [47:58<35:56,  1.21it/s]
3|3867|Loss: 0.8304895758628845:  60%|█████▉    | 3866/6470 [47:58<35:56,  1.21it/s]
3|3867|Loss: 0.8304895758628845:  60%|█████▉    | 3867/6470 [47:58<34:08,  1.27it/s]
3|3868|Loss: 0.7589623332023621:  60%|█████▉    | 3867/6470 [47:59<34:08,  1.27it/s]
3|3868|Loss: 0.7589623332023621:  60%|█████▉    | 3868/6470 [47:59<33:51,  1.28it/s]
3|3869|Loss: 0.9377985596656799:  60%|█████▉    | 3868/6470 [48:00<33:51,  1.28it/s]
3|3869|Loss: 0.9377985596656799:  60%|█████▉    | 3869/6470 [48:00<34:52,  1.24it/s]
3|3870|Loss: 0.8047128319740295:  60%|█████▉    | 3869/6470 [48:00<34:52,  1.24it/s]
3|3870|Loss: 0.8047128319740295:  60%|█████▉    | 3870/6470 [48:01<33:34,  1.29it/s]
3|3871|Loss: 0.798746645450592:  60%|█████▉    | 3870/6470 [48:01<33:34,  1.29it/s] 
3|3871|Loss: 0.798746645450592:  60%|█████▉    | 3871/6470 [48:01<34:22,  1.26it/s]
3|3872|Loss: 0.7675108909606934:  60%|█████▉    | 3871/6470 [48:02<34:22,  1.26it/s]
3|3872|Loss: 0.7675108909606934:  60%|█████▉    | 3872/6470 [48:02<34:17,  1.26it/s]
3|3873|Loss: 0.793808102607727:  60%|█████▉    | 3872/6470 [48:03<34:17,  1.26it/s] 
3|3873|Loss: 0.793808102607727:  60%|█████▉    | 3873/6470 [48:03<32:26,  1.33it/s]
3|3874|Loss: 0.8179040551185608:  60%|█████▉    | 3873/6470 [48:03<32:26,  1.33it/s]
3|3874|Loss: 0.8179040551185608:  60%|█████▉    | 3874/6470 [48:04<31:34,  1.37it/s]
3|3875|Loss: 0.821625292301178:  60%|█████▉    | 3874/6470 [48:04<31:34,  1.37it/s] 
3|3875|Loss: 0.821625292301178:  60%|█████▉    | 3875/6470 [48:04<33:33,  1.29it/s]
3|3876|Loss: 0.8621218800544739:  60%|█████▉    | 3875/6470 [48:05<33:33,  1.29it/s]
3|3876|Loss: 0.8621218800544739:  60%|█████▉    | 3876/6470 [48:05<32:48,  1.32it/s]
3|3877|Loss: 0.7095458507537842:  60%|█████▉    | 3876/6470 [48:06<32:48,  1.32it/s]
3|3877|Loss: 0.7095458507537842:  60%|█████▉    | 3877/6470 [48:06<29:40,  1.46it/s]
3|3878|Loss: 0.769405722618103:  60%|█████▉    | 3877/6470 [48:06<29:40,  1.46it/s] 
3|3878|Loss: 0.769405722618103:  60%|█████▉    | 3878/6470 [48:06<28:32,  1.51it/s]
3|3879|Loss: 0.7036231160163879:  60%|█████▉    | 3878/6470 [48:07<28:32,  1.51it/s]
3|3879|Loss: 0.7036231160163879:  60%|█████▉    | 3879/6470 [48:07<29:18,  1.47it/s]
3|3880|Loss: 0.789533257484436:  60%|█████▉    | 3879/6470 [48:08<29:18,  1.47it/s] 
3|3880|Loss: 0.789533257484436:  60%|█████▉    | 3880/6470 [48:08<32:38,  1.32it/s]
3|3881|Loss: 0.7577461004257202:  60%|█████▉    | 3880/6470 [48:08<32:38,  1.32it/s]
3|3881|Loss: 0.7577461004257202:  60%|█████▉    | 3881/6470 [48:09<31:58,  1.35it/s]
3|3882|Loss: 0.8542919158935547:  60%|█████▉    | 3881/6470 [48:09<31:58,  1.35it/s]
3|3882|Loss: 0.8542919158935547:  60%|██████    | 3882/6470 [48:10<33:37,  1.28it/s]
3|3883|Loss: 0.8058599829673767:  60%|██████    | 3882/6470 [48:10<33:37,  1.28it/s]
3|3883|Loss: 0.8058599829673767:  60%|██████    | 3883/6470 [48:10<34:27,  1.25it/s]
3|3884|Loss: 0.773025631904602:  60%|██████    | 3883/6470 [48:11<34:27,  1.25it/s] 
3|3884|Loss: 0.773025631904602:  60%|██████    | 3884/6470 [48:11<32:40,  1.32it/s]
3|3885|Loss: 0.7519980669021606:  60%|██████    | 3884/6470 [48:12<32:40,  1.32it/s]
3|3885|Loss: 0.7519980669021606:  60%|██████    | 3885/6470 [48:12<34:45,  1.24it/s]
3|3886|Loss: 0.8648326396942139:  60%|██████    | 3885/6470 [48:12<34:45,  1.24it/s]
3|3886|Loss: 0.8648326396942139:  60%|██████    | 3886/6470 [48:13<35:55,  1.20it/s]
3|3887|Loss: 0.8090107440948486:  60%|██████    | 3886/6470 [48:13<35:55,  1.20it/s]
3|3887|Loss: 0.8090107440948486:  60%|██████    | 3887/6470 [48:14<34:35,  1.24it/s]
3|3888|Loss: 0.7975733280181885:  60%|██████    | 3887/6470 [48:14<34:35,  1.24it/s]
3|3888|Loss: 0.7975733280181885:  60%|██████    | 3888/6470 [48:14<29:55,  1.44it/s]
3|3889|Loss: 0.753559410572052:  60%|██████    | 3888/6470 [48:14<29:55,  1.44it/s] 
3|3889|Loss: 0.753559410572052:  60%|██████    | 3889/6470 [48:15<29:47,  1.44it/s]
3|3890|Loss: 0.7970477938652039:  60%|██████    | 3889/6470 [48:15<29:47,  1.44it/s]
3|3890|Loss: 0.7970477938652039:  60%|██████    | 3890/6470 [48:15<28:22,  1.52it/s]
3|3891|Loss: 0.8531694412231445:  60%|██████    | 3890/6470 [48:16<28:22,  1.52it/s]
3|3891|Loss: 0.8531694412231445:  60%|██████    | 3891/6470 [48:16<29:59,  1.43it/s]
3|3892|Loss: 0.9079390168190002:  60%|██████    | 3891/6470 [48:17<29:59,  1.43it/s]
3|3892|Loss: 0.9079390168190002:  60%|██████    | 3892/6470 [48:17<29:14,  1.47it/s]
3|3893|Loss: 0.8816850185394287:  60%|██████    | 3892/6470 [48:17<29:14,  1.47it/s]
3|3893|Loss: 0.8816850185394287:  60%|██████    | 3893/6470 [48:18<30:06,  1.43it/s]
3|3894|Loss: 0.9186326861381531:  60%|██████    | 3893/6470 [48:18<30:06,  1.43it/s]
3|3894|Loss: 0.9186326861381531:  60%|██████    | 3894/6470 [48:18<31:09,  1.38it/s]
3|3895|Loss: 0.762032687664032:  60%|██████    | 3894/6470 [48:19<31:09,  1.38it/s] 
3|3895|Loss: 0.762032687664032:  60%|██████    | 3895/6470 [48:19<33:10,  1.29it/s]
3|3896|Loss: 0.7046335339546204:  60%|██████    | 3895/6470 [48:20<33:10,  1.29it/s]
3|3896|Loss: 0.7046335339546204:  60%|██████    | 3896/6470 [48:20<33:51,  1.27it/s]
3|3897|Loss: 0.8297258019447327:  60%|██████    | 3896/6470 [48:20<33:51,  1.27it/s]
3|3897|Loss: 0.8297258019447327:  60%|██████    | 3897/6470 [48:21<33:35,  1.28it/s]
3|3898|Loss: 0.9397439956665039:  60%|██████    | 3897/6470 [48:21<33:35,  1.28it/s]
3|3898|Loss: 0.9397439956665039:  60%|██████    | 3898/6470 [48:21<31:28,  1.36it/s]
3|3899|Loss: 0.7909069657325745:  60%|██████    | 3898/6470 [48:22<31:28,  1.36it/s]
3|3899|Loss: 0.7909069657325745:  60%|██████    | 3899/6470 [48:22<29:49,  1.44it/s]
3|3900|Loss: 0.8735519051551819:  60%|██████    | 3899/6470 [48:22<29:49,  1.44it/s]
3|3900|Loss: 0.8735519051551819:  60%|██████    | 3900/6470 [48:23<32:22,  1.32it/s]
3|3901|Loss: 0.5585495233535767:  60%|██████    | 3900/6470 [48:23<32:22,  1.32it/s]
3|3901|Loss: 0.5585495233535767:  60%|██████    | 3901/6470 [48:24<33:00,  1.30it/s]
3|3902|Loss: 0.892552375793457:  60%|██████    | 3901/6470 [48:24<33:00,  1.30it/s] 
3|3902|Loss: 0.892552375793457:  60%|██████    | 3902/6470 [48:24<32:20,  1.32it/s]
3|3903|Loss: 0.795857846736908:  60%|██████    | 3902/6470 [48:25<32:20,  1.32it/s]
3|3903|Loss: 0.795857846736908:  60%|██████    | 3903/6470 [48:25<30:43,  1.39it/s]
3|3904|Loss: 0.9016386866569519:  60%|██████    | 3903/6470 [48:25<30:43,  1.39it/s]
3|3904|Loss: 0.9016386866569519:  60%|██████    | 3904/6470 [48:26<32:40,  1.31it/s]
3|3905|Loss: 0.8775935173034668:  60%|██████    | 3904/6470 [48:26<32:40,  1.31it/s]
3|3905|Loss: 0.8775935173034668:  60%|██████    | 3905/6470 [48:27<32:19,  1.32it/s]
3|3906|Loss: 0.8201325535774231:  60%|██████    | 3905/6470 [48:27<32:19,  1.32it/s]
3|3906|Loss: 0.8201325535774231:  60%|██████    | 3906/6470 [48:28<33:58,  1.26it/s]
3|3907|Loss: 0.8830118179321289:  60%|██████    | 3906/6470 [48:28<33:58,  1.26it/s]
3|3907|Loss: 0.8830118179321289:  60%|██████    | 3907/6470 [48:29<36:38,  1.17it/s]
3|3908|Loss: 0.8027862310409546:  60%|██████    | 3907/6470 [48:29<36:38,  1.17it/s]
3|3908|Loss: 0.8027862310409546:  60%|██████    | 3908/6470 [48:30<37:55,  1.13it/s]
3|3909|Loss: 0.640017569065094:  60%|██████    | 3908/6470 [48:30<37:55,  1.13it/s] 
3|3909|Loss: 0.640017569065094:  60%|██████    | 3909/6470 [48:30<37:10,  1.15it/s]
3|3910|Loss: 0.7874220013618469:  60%|██████    | 3909/6470 [48:31<37:10,  1.15it/s]
3|3910|Loss: 0.7874220013618469:  60%|██████    | 3910/6470 [48:31<34:27,  1.24it/s]
3|3911|Loss: 0.557670533657074:  60%|██████    | 3910/6470 [48:31<34:27,  1.24it/s] 
3|3911|Loss: 0.557670533657074:  60%|██████    | 3911/6470 [48:32<37:14,  1.15it/s]
3|3912|Loss: 0.8669754266738892:  60%|██████    | 3911/6470 [48:33<37:14,  1.15it/s]
3|3912|Loss: 0.8669754266738892:  60%|██████    | 3912/6470 [48:33<37:42,  1.13it/s]
3|3913|Loss: 0.8748647570610046:  60%|██████    | 3912/6470 [48:33<37:42,  1.13it/s]
3|3913|Loss: 0.8748647570610046:  60%|██████    | 3913/6470 [48:34<35:30,  1.20it/s]
3|3914|Loss: 0.8268378973007202:  60%|██████    | 3913/6470 [48:34<35:30,  1.20it/s]
3|3914|Loss: 0.8268378973007202:  60%|██████    | 3914/6470 [48:34<33:14,  1.28it/s]
3|3915|Loss: 0.8845812678337097:  60%|██████    | 3914/6470 [48:35<33:14,  1.28it/s]
3|3915|Loss: 0.8845812678337097:  61%|██████    | 3915/6470 [48:35<35:01,  1.22it/s]
3|3916|Loss: 0.7718960642814636:  61%|██████    | 3915/6470 [48:36<35:01,  1.22it/s]
3|3916|Loss: 0.7718960642814636:  61%|██████    | 3916/6470 [48:36<32:38,  1.30it/s]
3|3917|Loss: 1.0328420400619507:  61%|██████    | 3916/6470 [48:36<32:38,  1.30it/s]
3|3917|Loss: 1.0328420400619507:  61%|██████    | 3917/6470 [48:36<29:37,  1.44it/s]
3|3918|Loss: 0.9268726110458374:  61%|██████    | 3917/6470 [48:37<29:37,  1.44it/s]
3|3918|Loss: 0.9268726110458374:  61%|██████    | 3918/6470 [48:37<31:20,  1.36it/s]
3|3919|Loss: 0.8272682428359985:  61%|██████    | 3918/6470 [48:38<31:20,  1.36it/s]
3|3919|Loss: 0.8272682428359985:  61%|██████    | 3919/6470 [48:38<31:55,  1.33it/s]
3|3920|Loss: 0.8772773146629333:  61%|██████    | 3919/6470 [48:38<31:55,  1.33it/s]
3|3920|Loss: 0.8772773146629333:  61%|██████    | 3920/6470 [48:39<32:19,  1.32it/s]
3|3921|Loss: 0.8009092807769775:  61%|██████    | 3920/6470 [48:39<32:19,  1.32it/s]
3|3921|Loss: 0.8009092807769775:  61%|██████    | 3921/6470 [48:40<32:39,  1.30it/s]
3|3922|Loss: 0.7899764180183411:  61%|██████    | 3921/6470 [48:40<32:39,  1.30it/s]
3|3922|Loss: 0.7899764180183411:  61%|██████    | 3922/6470 [48:40<31:27,  1.35it/s]
3|3923|Loss: 0.7697727084159851:  61%|██████    | 3922/6470 [48:41<31:27,  1.35it/s]
3|3923|Loss: 0.7697727084159851:  61%|██████    | 3923/6470 [48:41<30:51,  1.38it/s]
3|3924|Loss: 0.8568525314331055:  61%|██████    | 3923/6470 [48:41<30:51,  1.38it/s]
3|3924|Loss: 0.8568525314331055:  61%|██████    | 3924/6470 [48:42<33:17,  1.27it/s]
3|3925|Loss: 0.8232858180999756:  61%|██████    | 3924/6470 [48:42<33:17,  1.27it/s]
3|3925|Loss: 0.8232858180999756:  61%|██████    | 3925/6470 [48:43<33:49,  1.25it/s]
3|3926|Loss: 0.8571774959564209:  61%|██████    | 3925/6470 [48:43<33:49,  1.25it/s]
3|3926|Loss: 0.8571774959564209:  61%|██████    | 3926/6470 [48:43<32:49,  1.29it/s]
3|3927|Loss: 0.8621559739112854:  61%|██████    | 3926/6470 [48:44<32:49,  1.29it/s]
3|3927|Loss: 0.8621559739112854:  61%|██████    | 3927/6470 [48:44<31:33,  1.34it/s]
3|3928|Loss: 0.9334418773651123:  61%|██████    | 3927/6470 [48:44<31:33,  1.34it/s]
3|3928|Loss: 0.9334418773651123:  61%|██████    | 3928/6470 [48:45<29:16,  1.45it/s]
3|3929|Loss: 0.8458313941955566:  61%|██████    | 3928/6470 [48:45<29:16,  1.45it/s]
3|3929|Loss: 0.8458313941955566:  61%|██████    | 3929/6470 [48:45<29:24,  1.44it/s]
3|3930|Loss: 0.6496381163597107:  61%|██████    | 3929/6470 [48:46<29:24,  1.44it/s]
3|3930|Loss: 0.6496381163597107:  61%|██████    | 3930/6470 [48:46<29:15,  1.45it/s]
3|3931|Loss: 0.9648153781890869:  61%|██████    | 3930/6470 [48:46<29:15,  1.45it/s]
3|3931|Loss: 0.9648153781890869:  61%|██████    | 3931/6470 [48:47<30:12,  1.40it/s]
3|3932|Loss: 0.849947988986969:  61%|██████    | 3931/6470 [48:47<30:12,  1.40it/s] 
3|3932|Loss: 0.849947988986969:  61%|██████    | 3932/6470 [48:48<29:58,  1.41it/s]
3|3933|Loss: 0.7911508083343506:  61%|██████    | 3932/6470 [48:48<29:58,  1.41it/s]
3|3933|Loss: 0.7911508083343506:  61%|██████    | 3933/6470 [48:48<29:04,  1.45it/s]
3|3934|Loss: 0.8529103398323059:  61%|██████    | 3933/6470 [48:49<29:04,  1.45it/s]
3|3934|Loss: 0.8529103398323059:  61%|██████    | 3934/6470 [48:49<29:57,  1.41it/s]
3|3935|Loss: 0.6573805212974548:  61%|██████    | 3934/6470 [48:49<29:57,  1.41it/s]
3|3935|Loss: 0.6573805212974548:  61%|██████    | 3935/6470 [48:50<29:14,  1.44it/s]
3|3936|Loss: 0.96406090259552:  61%|██████    | 3935/6470 [48:50<29:14,  1.44it/s]  
3|3936|Loss: 0.96406090259552:  61%|██████    | 3936/6470 [48:50<31:30,  1.34it/s]
3|3937|Loss: 0.8846791982650757:  61%|██████    | 3936/6470 [48:51<31:30,  1.34it/s]
3|3937|Loss: 0.8846791982650757:  61%|██████    | 3937/6470 [48:51<31:11,  1.35it/s]
3|3938|Loss: 0.8233502507209778:  61%|██████    | 3937/6470 [48:52<31:11,  1.35it/s]
3|3938|Loss: 0.8233502507209778:  61%|██████    | 3938/6470 [48:52<31:43,  1.33it/s]
3|3939|Loss: 0.8968936204910278:  61%|██████    | 3938/6470 [48:52<31:43,  1.33it/s]
3|3939|Loss: 0.8968936204910278:  61%|██████    | 3939/6470 [48:53<33:23,  1.26it/s]
3|3940|Loss: 0.9573191404342651:  61%|██████    | 3939/6470 [48:53<33:23,  1.26it/s]
3|3940|Loss: 0.9573191404342651:  61%|██████    | 3940/6470 [48:54<33:47,  1.25it/s]
3|3941|Loss: 0.938310444355011:  61%|██████    | 3940/6470 [48:54<33:47,  1.25it/s] 
3|3941|Loss: 0.938310444355011:  61%|██████    | 3941/6470 [48:54<30:38,  1.38it/s]
3|3942|Loss: 0.9399380683898926:  61%|██████    | 3941/6470 [48:55<30:38,  1.38it/s]
3|3942|Loss: 0.9399380683898926:  61%|██████    | 3942/6470 [48:55<30:28,  1.38it/s]
3|3943|Loss: 0.75607830286026:  61%|██████    | 3942/6470 [48:55<30:28,  1.38it/s]  
3|3943|Loss: 0.75607830286026:  61%|██████    | 3943/6470 [48:56<30:25,  1.38it/s]
3|3944|Loss: 0.9167977571487427:  61%|██████    | 3943/6470 [48:56<30:25,  1.38it/s]
3|3944|Loss: 0.9167977571487427:  61%|██████    | 3944/6470 [48:56<29:23,  1.43it/s]
3|3945|Loss: 1.0009597539901733:  61%|██████    | 3944/6470 [48:57<29:23,  1.43it/s]
3|3945|Loss: 1.0009597539901733:  61%|██████    | 3945/6470 [48:57<31:26,  1.34it/s]
3|3946|Loss: 0.8717138767242432:  61%|██████    | 3945/6470 [48:58<31:26,  1.34it/s]
3|3946|Loss: 0.8717138767242432:  61%|██████    | 3946/6470 [48:58<31:08,  1.35it/s]
3|3947|Loss: 0.8042388558387756:  61%|██████    | 3946/6470 [48:58<31:08,  1.35it/s]
3|3947|Loss: 0.8042388558387756:  61%|██████    | 3947/6470 [48:58<28:15,  1.49it/s]
3|3948|Loss: 0.7788431644439697:  61%|██████    | 3947/6470 [48:59<28:15,  1.49it/s]
3|3948|Loss: 0.7788431644439697:  61%|██████    | 3948/6470 [48:59<28:28,  1.48it/s]
3|3949|Loss: 0.8739861249923706:  61%|██████    | 3948/6470 [48:59<28:28,  1.48it/s]
3|3949|Loss: 0.8739861249923706:  61%|██████    | 3949/6470 [49:00<28:57,  1.45it/s]
3|3950|Loss: 0.7305039167404175:  61%|██████    | 3949/6470 [49:00<28:57,  1.45it/s]
3|3950|Loss: 0.7305039167404175:  61%|██████    | 3950/6470 [49:00<29:10,  1.44it/s]
3|3951|Loss: 0.6958256959915161:  61%|██████    | 3950/6470 [49:01<29:10,  1.44it/s]
3|3951|Loss: 0.6958256959915161:  61%|██████    | 3951/6470 [49:01<30:44,  1.37it/s]
3|3952|Loss: 0.8015735745429993:  61%|██████    | 3951/6470 [49:02<30:44,  1.37it/s]
3|3952|Loss: 0.8015735745429993:  61%|██████    | 3952/6470 [49:02<30:05,  1.39it/s]
3|3953|Loss: 0.8285316228866577:  61%|██████    | 3952/6470 [49:02<30:05,  1.39it/s]
3|3953|Loss: 0.8285316228866577:  61%|██████    | 3953/6470 [49:03<30:35,  1.37it/s]
3|3954|Loss: 0.8136717081069946:  61%|██████    | 3953/6470 [49:03<30:35,  1.37it/s]
3|3954|Loss: 0.8136717081069946:  61%|██████    | 3954/6470 [49:04<31:12,  1.34it/s]
3|3955|Loss: 0.9116926193237305:  61%|██████    | 3954/6470 [49:04<31:12,  1.34it/s]
3|3955|Loss: 0.9116926193237305:  61%|██████    | 3955/6470 [49:04<31:24,  1.33it/s]
3|3956|Loss: 0.7200101613998413:  61%|██████    | 3955/6470 [49:05<31:24,  1.33it/s]
3|3956|Loss: 0.7200101613998413:  61%|██████    | 3956/6470 [49:05<30:30,  1.37it/s]
3|3957|Loss: 0.9109373092651367:  61%|██████    | 3956/6470 [49:05<30:30,  1.37it/s]
3|3957|Loss: 0.9109373092651367:  61%|██████    | 3957/6470 [49:06<33:04,  1.27it/s]
3|3958|Loss: 1.03361976146698:  61%|██████    | 3957/6470 [49:06<33:04,  1.27it/s]  
3|3958|Loss: 1.03361976146698:  61%|██████    | 3958/6470 [49:06<30:26,  1.38it/s]
3|3959|Loss: 0.7468789219856262:  61%|██████    | 3958/6470 [49:07<30:26,  1.38it/s]
3|3959|Loss: 0.7468789219856262:  61%|██████    | 3959/6470 [49:07<30:30,  1.37it/s]
3|3960|Loss: 0.8710274696350098:  61%|██████    | 3959/6470 [49:08<30:30,  1.37it/s]
3|3960|Loss: 0.8710274696350098:  61%|██████    | 3960/6470 [49:08<32:42,  1.28it/s]
3|3961|Loss: 0.8274787664413452:  61%|██████    | 3960/6470 [49:09<32:42,  1.28it/s]
3|3961|Loss: 0.8274787664413452:  61%|██████    | 3961/6470 [49:09<31:16,  1.34it/s]
3|3962|Loss: 0.840152382850647:  61%|██████    | 3961/6470 [49:09<31:16,  1.34it/s] 
3|3962|Loss: 0.840152382850647:  61%|██████    | 3962/6470 [49:09<30:54,  1.35it/s]
3|3963|Loss: 0.8018450140953064:  61%|██████    | 3962/6470 [49:10<30:54,  1.35it/s]
3|3963|Loss: 0.8018450140953064:  61%|██████▏   | 3963/6470 [49:10<32:50,  1.27it/s]
3|3964|Loss: 0.7003148198127747:  61%|██████▏   | 3963/6470 [49:11<32:50,  1.27it/s]
3|3964|Loss: 0.7003148198127747:  61%|██████▏   | 3964/6470 [49:11<34:25,  1.21it/s]
3|3965|Loss: 0.7646172642707825:  61%|██████▏   | 3964/6470 [49:12<34:25,  1.21it/s]
3|3965|Loss: 0.7646172642707825:  61%|██████▏   | 3965/6470 [49:12<33:01,  1.26it/s]
3|3966|Loss: 0.7980865240097046:  61%|██████▏   | 3965/6470 [49:12<33:01,  1.26it/s]
3|3966|Loss: 0.7980865240097046:  61%|██████▏   | 3966/6470 [49:13<32:39,  1.28it/s]
3|3967|Loss: 0.8503312468528748:  61%|██████▏   | 3966/6470 [49:13<32:39,  1.28it/s]
3|3967|Loss: 0.8503312468528748:  61%|██████▏   | 3967/6470 [49:14<32:16,  1.29it/s]
3|3968|Loss: 0.8932132124900818:  61%|██████▏   | 3967/6470 [49:14<32:16,  1.29it/s]
3|3968|Loss: 0.8932132124900818:  61%|██████▏   | 3968/6470 [49:14<33:01,  1.26it/s]
3|3969|Loss: 0.904363751411438:  61%|██████▏   | 3968/6470 [49:15<33:01,  1.26it/s] 
3|3969|Loss: 0.904363751411438:  61%|██████▏   | 3969/6470 [49:15<34:12,  1.22it/s]
3|3970|Loss: 0.6061423420906067:  61%|██████▏   | 3969/6470 [49:16<34:12,  1.22it/s]
3|3970|Loss: 0.6061423420906067:  61%|██████▏   | 3970/6470 [49:16<31:13,  1.33it/s]
3|3971|Loss: 0.8543848991394043:  61%|██████▏   | 3970/6470 [49:16<31:13,  1.33it/s]
3|3971|Loss: 0.8543848991394043:  61%|██████▏   | 3971/6470 [49:16<29:07,  1.43it/s]
3|3972|Loss: 0.7615018486976624:  61%|██████▏   | 3971/6470 [49:17<29:07,  1.43it/s]
3|3972|Loss: 0.7615018486976624:  61%|██████▏   | 3972/6470 [49:17<31:57,  1.30it/s]
3|3973|Loss: 0.7148868441581726:  61%|██████▏   | 3972/6470 [49:18<31:57,  1.30it/s]
3|3973|Loss: 0.7148868441581726:  61%|██████▏   | 3973/6470 [49:18<33:31,  1.24it/s]
3|3974|Loss: 0.7093743681907654:  61%|██████▏   | 3973/6470 [49:19<33:31,  1.24it/s]
3|3974|Loss: 0.7093743681907654:  61%|██████▏   | 3974/6470 [49:19<32:36,  1.28it/s]
3|3975|Loss: 0.9212337732315063:  61%|██████▏   | 3974/6470 [49:19<32:36,  1.28it/s]
3|3975|Loss: 0.9212337732315063:  61%|██████▏   | 3975/6470 [49:20<32:38,  1.27it/s]
3|3976|Loss: 0.746570885181427:  61%|██████▏   | 3975/6470 [49:20<32:38,  1.27it/s] 
3|3976|Loss: 0.746570885181427:  61%|██████▏   | 3976/6470 [49:20<31:30,  1.32it/s]
3|3977|Loss: 0.7907371520996094:  61%|██████▏   | 3976/6470 [49:21<31:30,  1.32it/s]
3|3977|Loss: 0.7907371520996094:  61%|██████▏   | 3977/6470 [49:21<30:26,  1.36it/s]
3|3978|Loss: 0.8255854845046997:  61%|██████▏   | 3977/6470 [49:21<30:26,  1.36it/s]
3|3978|Loss: 0.8255854845046997:  61%|██████▏   | 3978/6470 [49:22<28:36,  1.45it/s]
3|3979|Loss: 0.7776114344596863:  61%|██████▏   | 3978/6470 [49:22<28:36,  1.45it/s]
3|3979|Loss: 0.7776114344596863:  61%|██████▏   | 3979/6470 [49:22<28:46,  1.44it/s]
3|3980|Loss: 0.7172356247901917:  61%|██████▏   | 3979/6470 [49:23<28:46,  1.44it/s]
3|3980|Loss: 0.7172356247901917:  62%|██████▏   | 3980/6470 [49:23<29:26,  1.41it/s]
3|3981|Loss: 0.7400813102722168:  62%|██████▏   | 3980/6470 [49:24<29:26,  1.41it/s]
3|3981|Loss: 0.7400813102722168:  62%|██████▏   | 3981/6470 [49:24<29:21,  1.41it/s]
3|3982|Loss: 1.0205198526382446:  62%|██████▏   | 3981/6470 [49:24<29:21,  1.41it/s]
3|3982|Loss: 1.0205198526382446:  62%|██████▏   | 3982/6470 [49:25<29:05,  1.43it/s]
3|3983|Loss: 0.7761489152908325:  62%|██████▏   | 3982/6470 [49:25<29:05,  1.43it/s]
3|3983|Loss: 0.7761489152908325:  62%|██████▏   | 3983/6470 [49:25<27:39,  1.50it/s]
3|3984|Loss: 0.8578004837036133:  62%|██████▏   | 3983/6470 [49:26<27:39,  1.50it/s]
3|3984|Loss: 0.8578004837036133:  62%|██████▏   | 3984/6470 [49:26<28:44,  1.44it/s]
3|3985|Loss: 0.8644116520881653:  62%|██████▏   | 3984/6470 [49:26<28:44,  1.44it/s]
3|3985|Loss: 0.8644116520881653:  62%|██████▏   | 3985/6470 [49:27<30:10,  1.37it/s]
3|3986|Loss: 0.8044044375419617:  62%|██████▏   | 3985/6470 [49:27<30:10,  1.37it/s]
3|3986|Loss: 0.8044044375419617:  62%|██████▏   | 3986/6470 [49:28<31:55,  1.30it/s]
3|3987|Loss: 0.8666965961456299:  62%|██████▏   | 3986/6470 [49:28<31:55,  1.30it/s]
3|3987|Loss: 0.8666965961456299:  62%|██████▏   | 3987/6470 [49:28<32:21,  1.28it/s]
3|3988|Loss: 0.6672577261924744:  62%|██████▏   | 3987/6470 [49:29<32:21,  1.28it/s]
3|3988|Loss: 0.6672577261924744:  62%|██████▏   | 3988/6470 [49:29<30:46,  1.34it/s]
3|3989|Loss: 0.8511231541633606:  62%|██████▏   | 3988/6470 [49:29<30:46,  1.34it/s]
3|3989|Loss: 0.8511231541633606:  62%|██████▏   | 3989/6470 [49:30<31:12,  1.32it/s]
3|3990|Loss: 0.9191673398017883:  62%|██████▏   | 3989/6470 [49:30<31:12,  1.32it/s]
3|3990|Loss: 0.9191673398017883:  62%|██████▏   | 3990/6470 [49:31<31:28,  1.31it/s]
3|3991|Loss: 0.8560869097709656:  62%|██████▏   | 3990/6470 [49:31<31:28,  1.31it/s]
3|3991|Loss: 0.8560869097709656:  62%|██████▏   | 3991/6470 [49:31<31:53,  1.30it/s]
3|3992|Loss: 0.7930980920791626:  62%|██████▏   | 3991/6470 [49:32<31:53,  1.30it/s]
3|3992|Loss: 0.7930980920791626:  62%|██████▏   | 3992/6470 [49:32<33:29,  1.23it/s]
3|3993|Loss: 0.9275955557823181:  62%|██████▏   | 3992/6470 [49:33<33:29,  1.23it/s]
3|3993|Loss: 0.9275955557823181:  62%|██████▏   | 3993/6470 [49:33<33:19,  1.24it/s]
3|3994|Loss: 0.9041405916213989:  62%|██████▏   | 3993/6470 [49:34<33:19,  1.24it/s]
3|3994|Loss: 0.9041405916213989:  62%|██████▏   | 3994/6470 [49:34<32:45,  1.26it/s]
3|3995|Loss: 0.8449780344963074:  62%|██████▏   | 3994/6470 [49:34<32:45,  1.26it/s]
3|3995|Loss: 0.8449780344963074:  62%|██████▏   | 3995/6470 [49:35<32:40,  1.26it/s]
3|3996|Loss: 0.8939230442047119:  62%|██████▏   | 3995/6470 [49:35<32:40,  1.26it/s]
3|3996|Loss: 0.8939230442047119:  62%|██████▏   | 3996/6470 [49:35<31:41,  1.30it/s]
3|3997|Loss: 0.7696921825408936:  62%|██████▏   | 3996/6470 [49:36<31:41,  1.30it/s]
3|3997|Loss: 0.7696921825408936:  62%|██████▏   | 3997/6470 [49:36<28:36,  1.44it/s]
3|3998|Loss: 0.9598174095153809:  62%|██████▏   | 3997/6470 [49:36<28:36,  1.44it/s]
3|3998|Loss: 0.9598174095153809:  62%|██████▏   | 3998/6470 [49:36<27:10,  1.52it/s]
3|3999|Loss: 0.6686490774154663:  62%|██████▏   | 3998/6470 [49:37<27:10,  1.52it/s]
3|3999|Loss: 0.6686490774154663:  62%|██████▏   | 3999/6470 [49:37<27:25,  1.50it/s]
3|4000|Loss: 0.9546137452125549:  62%|██████▏   | 3999/6470 [49:38<27:25,  1.50it/s]
3|4000|Loss: 0.9546137452125549:  62%|██████▏   | 4000/6470 [49:38<27:26,  1.50it/s]
3|4001|Loss: 0.8500145673751831:  62%|██████▏   | 4000/6470 [49:38<27:26,  1.50it/s]
3|4001|Loss: 0.8500145673751831:  62%|██████▏   | 4001/6470 [49:39<28:42,  1.43it/s]
3|4002|Loss: 0.9316229820251465:  62%|██████▏   | 4001/6470 [49:39<28:42,  1.43it/s]
3|4002|Loss: 0.9316229820251465:  62%|██████▏   | 4002/6470 [49:39<30:10,  1.36it/s]
3|4003|Loss: 0.8018506765365601:  62%|██████▏   | 4002/6470 [49:40<30:10,  1.36it/s]
3|4003|Loss: 0.8018506765365601:  62%|██████▏   | 4003/6470 [49:40<32:26,  1.27it/s]
3|4004|Loss: 1.0405420064926147:  62%|██████▏   | 4003/6470 [49:41<32:26,  1.27it/s]
3|4004|Loss: 1.0405420064926147:  62%|██████▏   | 4004/6470 [49:41<31:11,  1.32it/s]
3|4005|Loss: 0.8690120577812195:  62%|██████▏   | 4004/6470 [49:41<31:11,  1.32it/s]
3|4005|Loss: 0.8690120577812195:  62%|██████▏   | 4005/6470 [49:42<29:47,  1.38it/s]
3|4006|Loss: 0.8976777195930481:  62%|██████▏   | 4005/6470 [49:42<29:47,  1.38it/s]
3|4006|Loss: 0.8976777195930481:  62%|██████▏   | 4006/6470 [49:42<29:01,  1.41it/s]
3|4007|Loss: 0.8907822966575623:  62%|██████▏   | 4006/6470 [49:43<29:01,  1.41it/s]
3|4007|Loss: 0.8907822966575623:  62%|██████▏   | 4007/6470 [49:43<27:45,  1.48it/s]
3|4008|Loss: 0.7114751935005188:  62%|██████▏   | 4007/6470 [49:43<27:45,  1.48it/s]
3|4008|Loss: 0.7114751935005188:  62%|██████▏   | 4008/6470 [49:43<24:43,  1.66it/s]
3|4009|Loss: 0.8321756720542908:  62%|██████▏   | 4008/6470 [49:44<24:43,  1.66it/s]
3|4009|Loss: 0.8321756720542908:  62%|██████▏   | 4009/6470 [49:44<23:54,  1.72it/s]
3|4010|Loss: 0.8685637712478638:  62%|██████▏   | 4009/6470 [49:44<23:54,  1.72it/s]
3|4010|Loss: 0.8685637712478638:  62%|██████▏   | 4010/6470 [49:45<27:52,  1.47it/s]
3|4011|Loss: 0.8904771208763123:  62%|██████▏   | 4010/6470 [49:45<27:52,  1.47it/s]
3|4011|Loss: 0.8904771208763123:  62%|██████▏   | 4011/6470 [49:46<29:50,  1.37it/s]
3|4012|Loss: 0.9170436263084412:  62%|██████▏   | 4011/6470 [49:46<29:50,  1.37it/s]
3|4012|Loss: 0.9170436263084412:  62%|██████▏   | 4012/6470 [49:46<29:32,  1.39it/s]
3|4013|Loss: 0.8282635807991028:  62%|██████▏   | 4012/6470 [49:47<29:32,  1.39it/s]
3|4013|Loss: 0.8282635807991028:  62%|██████▏   | 4013/6470 [49:47<31:17,  1.31it/s]
3|4014|Loss: 0.8123852014541626:  62%|██████▏   | 4013/6470 [49:48<31:17,  1.31it/s]
3|4014|Loss: 0.8123852014541626:  62%|██████▏   | 4014/6470 [49:48<33:23,  1.23it/s]
3|4015|Loss: 0.7560943365097046:  62%|██████▏   | 4014/6470 [49:49<33:23,  1.23it/s]
3|4015|Loss: 0.7560943365097046:  62%|██████▏   | 4015/6470 [49:49<33:12,  1.23it/s]
3|4016|Loss: 1.0263384580612183:  62%|██████▏   | 4015/6470 [49:49<33:12,  1.23it/s]
3|4016|Loss: 1.0263384580612183:  62%|██████▏   | 4016/6470 [49:50<32:30,  1.26it/s]
3|4017|Loss: 0.8315280079841614:  62%|██████▏   | 4016/6470 [49:50<32:30,  1.26it/s]
3|4017|Loss: 0.8315280079841614:  62%|██████▏   | 4017/6470 [49:50<31:26,  1.30it/s]
3|4018|Loss: 0.5965360999107361:  62%|██████▏   | 4017/6470 [49:51<31:26,  1.30it/s]
3|4018|Loss: 0.5965360999107361:  62%|██████▏   | 4018/6470 [49:52<44:38,  1.09s/it]
3|4019|Loss: 0.6853249073028564:  62%|██████▏   | 4018/6470 [49:53<44:38,  1.09s/it]
3|4019|Loss: 0.6853249073028564:  62%|██████▏   | 4019/6470 [49:53<41:20,  1.01s/it]
3|4020|Loss: 0.8244806528091431:  62%|██████▏   | 4019/6470 [49:53<41:20,  1.01s/it]
3|4020|Loss: 0.8244806528091431:  62%|██████▏   | 4020/6470 [49:54<36:36,  1.12it/s]
3|4021|Loss: 0.7405088543891907:  62%|██████▏   | 4020/6470 [49:54<36:36,  1.12it/s]
3|4021|Loss: 0.7405088543891907:  62%|██████▏   | 4021/6470 [49:55<36:58,  1.10it/s]
3|4022|Loss: 0.8526485562324524:  62%|██████▏   | 4021/6470 [49:55<36:58,  1.10it/s]
3|4022|Loss: 0.8526485562324524:  62%|██████▏   | 4022/6470 [49:56<36:56,  1.10it/s]
3|4023|Loss: 0.9634835124015808:  62%|██████▏   | 4022/6470 [49:56<36:56,  1.10it/s]
3|4023|Loss: 0.9634835124015808:  62%|██████▏   | 4023/6470 [49:56<37:18,  1.09it/s]
3|4024|Loss: 0.774857759475708:  62%|██████▏   | 4023/6470 [49:57<37:18,  1.09it/s] 
3|4024|Loss: 0.774857759475708:  62%|██████▏   | 4024/6470 [49:57<35:25,  1.15it/s]
3|4025|Loss: 0.8659076690673828:  62%|██████▏   | 4024/6470 [49:58<35:25,  1.15it/s]
3|4025|Loss: 0.8659076690673828:  62%|██████▏   | 4025/6470 [49:58<35:19,  1.15it/s]
3|4026|Loss: 0.7849788069725037:  62%|██████▏   | 4025/6470 [49:59<35:19,  1.15it/s]
3|4026|Loss: 0.7849788069725037:  62%|██████▏   | 4026/6470 [49:59<33:14,  1.23it/s]
3|4027|Loss: 0.7643076777458191:  62%|██████▏   | 4026/6470 [49:59<33:14,  1.23it/s]
3|4027|Loss: 0.7643076777458191:  62%|██████▏   | 4027/6470 [49:59<29:30,  1.38it/s]
3|4028|Loss: 0.8887850642204285:  62%|██████▏   | 4027/6470 [50:00<29:30,  1.38it/s]
3|4028|Loss: 0.8887850642204285:  62%|██████▏   | 4028/6470 [50:00<31:00,  1.31it/s]
3|4029|Loss: 1.027787685394287:  62%|██████▏   | 4028/6470 [50:01<31:00,  1.31it/s] 
3|4029|Loss: 1.027787685394287:  62%|██████▏   | 4029/6470 [50:01<30:57,  1.31it/s]
3|4030|Loss: 0.7051400542259216:  62%|██████▏   | 4029/6470 [50:01<30:57,  1.31it/s]
3|4030|Loss: 0.7051400542259216:  62%|██████▏   | 4030/6470 [50:02<29:56,  1.36it/s]
3|4031|Loss: 0.7333860993385315:  62%|██████▏   | 4030/6470 [50:02<29:56,  1.36it/s]
3|4031|Loss: 0.7333860993385315:  62%|██████▏   | 4031/6470 [50:02<29:54,  1.36it/s]
3|4032|Loss: 0.8062109351158142:  62%|██████▏   | 4031/6470 [50:03<29:54,  1.36it/s]
3|4032|Loss: 0.8062109351158142:  62%|██████▏   | 4032/6470 [50:03<30:37,  1.33it/s]
3|4033|Loss: 0.813570499420166:  62%|██████▏   | 4032/6470 [50:04<30:37,  1.33it/s] 
3|4033|Loss: 0.813570499420166:  62%|██████▏   | 4033/6470 [50:04<30:22,  1.34it/s]
3|4034|Loss: 0.6530024409294128:  62%|██████▏   | 4033/6470 [50:04<30:22,  1.34it/s]
3|4034|Loss: 0.6530024409294128:  62%|██████▏   | 4034/6470 [50:04<27:32,  1.47it/s]
3|4035|Loss: 0.755603015422821:  62%|██████▏   | 4034/6470 [50:05<27:32,  1.47it/s] 
3|4035|Loss: 0.755603015422821:  62%|██████▏   | 4035/6470 [50:05<26:15,  1.55it/s]
3|4036|Loss: 0.88612961769104:  62%|██████▏   | 4035/6470 [50:05<26:15,  1.55it/s] 
3|4036|Loss: 0.88612961769104:  62%|██████▏   | 4036/6470 [50:06<27:09,  1.49it/s]
3|4037|Loss: 0.8330768942832947:  62%|██████▏   | 4036/6470 [50:06<27:09,  1.49it/s]
3|4037|Loss: 0.8330768942832947:  62%|██████▏   | 4037/6470 [50:06<26:31,  1.53it/s]
3|4038|Loss: 0.9551762938499451:  62%|██████▏   | 4037/6470 [50:07<26:31,  1.53it/s]
3|4038|Loss: 0.9551762938499451:  62%|██████▏   | 4038/6470 [50:07<26:22,  1.54it/s]
3|4039|Loss: 0.8628280162811279:  62%|██████▏   | 4038/6470 [50:07<26:22,  1.54it/s]
3|4039|Loss: 0.8628280162811279:  62%|██████▏   | 4039/6470 [50:08<25:41,  1.58it/s]
3|4040|Loss: 0.7013528943061829:  62%|██████▏   | 4039/6470 [50:08<25:41,  1.58it/s]
3|4040|Loss: 0.7013528943061829:  62%|██████▏   | 4040/6470 [50:08<25:45,  1.57it/s]
3|4041|Loss: 0.84150630235672:  62%|██████▏   | 4040/6470 [50:09<25:45,  1.57it/s]  
3|4041|Loss: 0.84150630235672:  62%|██████▏   | 4041/6470 [50:09<25:08,  1.61it/s]
3|4042|Loss: 0.86391681432724:  62%|██████▏   | 4041/6470 [50:09<25:08,  1.61it/s]
3|4042|Loss: 0.86391681432724:  62%|██████▏   | 4042/6470 [50:09<25:48,  1.57it/s]
3|4043|Loss: 0.5474553108215332:  62%|██████▏   | 4042/6470 [50:10<25:48,  1.57it/s]
3|4043|Loss: 0.5474553108215332:  62%|██████▏   | 4043/6470 [50:10<26:43,  1.51it/s]
3|4044|Loss: 0.639137327671051:  62%|██████▏   | 4043/6470 [50:11<26:43,  1.51it/s] 
3|4044|Loss: 0.639137327671051:  63%|██████▎   | 4044/6470 [50:11<27:46,  1.46it/s]
3|4045|Loss: 0.7639610171318054:  63%|██████▎   | 4044/6470 [50:11<27:46,  1.46it/s]
3|4045|Loss: 0.7639610171318054:  63%|██████▎   | 4045/6470 [50:11<26:28,  1.53it/s]
3|4046|Loss: 0.8570611476898193:  63%|██████▎   | 4045/6470 [50:12<26:28,  1.53it/s]
3|4046|Loss: 0.8570611476898193:  63%|██████▎   | 4046/6470 [50:12<27:07,  1.49it/s]
3|4047|Loss: 0.8769020438194275:  63%|██████▎   | 4046/6470 [50:13<27:07,  1.49it/s]
3|4047|Loss: 0.8769020438194275:  63%|██████▎   | 4047/6470 [50:13<27:23,  1.47it/s]
3|4048|Loss: 0.8897544145584106:  63%|██████▎   | 4047/6470 [50:13<27:23,  1.47it/s]
3|4048|Loss: 0.8897544145584106:  63%|██████▎   | 4048/6470 [50:14<26:46,  1.51it/s]
3|4049|Loss: 0.8596159815788269:  63%|██████▎   | 4048/6470 [50:14<26:46,  1.51it/s]
3|4049|Loss: 0.8596159815788269:  63%|██████▎   | 4049/6470 [50:14<28:29,  1.42it/s]
3|4050|Loss: 0.8294633030891418:  63%|██████▎   | 4049/6470 [50:15<28:29,  1.42it/s]
3|4050|Loss: 0.8294633030891418:  63%|██████▎   | 4050/6470 [50:15<29:38,  1.36it/s]
3|4051|Loss: 0.858901858329773:  63%|██████▎   | 4050/6470 [50:15<29:38,  1.36it/s] 
3|4051|Loss: 0.858901858329773:  63%|██████▎   | 4051/6470 [50:16<27:15,  1.48it/s]
3|4052|Loss: 0.8446352481842041:  63%|██████▎   | 4051/6470 [50:16<27:15,  1.48it/s]
3|4052|Loss: 0.8446352481842041:  63%|██████▎   | 4052/6470 [50:16<26:16,  1.53it/s]
3|4053|Loss: 0.8839809894561768:  63%|██████▎   | 4052/6470 [50:17<26:16,  1.53it/s]
3|4053|Loss: 0.8839809894561768:  63%|██████▎   | 4053/6470 [50:17<28:47,  1.40it/s]
3|4054|Loss: 0.821398138999939:  63%|██████▎   | 4053/6470 [50:18<28:47,  1.40it/s] 
3|4054|Loss: 0.821398138999939:  63%|██████▎   | 4054/6470 [50:18<27:41,  1.45it/s]
3|4055|Loss: 0.9101596474647522:  63%|██████▎   | 4054/6470 [50:18<27:41,  1.45it/s]
3|4055|Loss: 0.9101596474647522:  63%|██████▎   | 4055/6470 [50:18<28:16,  1.42it/s]
3|4056|Loss: 0.6943296194076538:  63%|██████▎   | 4055/6470 [50:19<28:16,  1.42it/s]
3|4056|Loss: 0.6943296194076538:  63%|██████▎   | 4056/6470 [50:19<28:08,  1.43it/s]
3|4057|Loss: 0.7691718339920044:  63%|██████▎   | 4056/6470 [50:20<28:08,  1.43it/s]
3|4057|Loss: 0.7691718339920044:  63%|██████▎   | 4057/6470 [50:20<27:14,  1.48it/s]
3|4058|Loss: 0.8827307224273682:  63%|██████▎   | 4057/6470 [50:20<27:14,  1.48it/s]
3|4058|Loss: 0.8827307224273682:  63%|██████▎   | 4058/6470 [50:21<29:10,  1.38it/s]
3|4059|Loss: 0.9059789180755615:  63%|██████▎   | 4058/6470 [50:21<29:10,  1.38it/s]
3|4059|Loss: 0.9059789180755615:  63%|██████▎   | 4059/6470 [50:21<30:15,  1.33it/s]
3|4060|Loss: 1.1803025007247925:  63%|██████▎   | 4059/6470 [50:22<30:15,  1.33it/s]
3|4060|Loss: 1.1803025007247925:  63%|██████▎   | 4060/6470 [50:22<30:58,  1.30it/s]
3|4061|Loss: 0.7843085527420044:  63%|██████▎   | 4060/6470 [50:23<30:58,  1.30it/s]
3|4061|Loss: 0.7843085527420044:  63%|██████▎   | 4061/6470 [50:23<30:03,  1.34it/s]
3|4062|Loss: 0.8134476542472839:  63%|██████▎   | 4061/6470 [50:23<30:03,  1.34it/s]
3|4062|Loss: 0.8134476542472839:  63%|██████▎   | 4062/6470 [50:24<28:34,  1.40it/s]
3|4063|Loss: 0.753705620765686:  63%|██████▎   | 4062/6470 [50:24<28:34,  1.40it/s] 
3|4063|Loss: 0.753705620765686:  63%|██████▎   | 4063/6470 [50:24<27:33,  1.46it/s]
3|4064|Loss: 0.8627512454986572:  63%|██████▎   | 4063/6470 [50:25<27:33,  1.46it/s]
3|4064|Loss: 0.8627512454986572:  63%|██████▎   | 4064/6470 [50:25<28:30,  1.41it/s]
3|4065|Loss: 0.8874445557594299:  63%|██████▎   | 4064/6470 [50:25<28:30,  1.41it/s]
3|4065|Loss: 0.8874445557594299:  63%|██████▎   | 4065/6470 [50:26<29:26,  1.36it/s]
3|4066|Loss: 0.7637852430343628:  63%|██████▎   | 4065/6470 [50:26<29:26,  1.36it/s]
3|4066|Loss: 0.7637852430343628:  63%|██████▎   | 4066/6470 [50:26<29:00,  1.38it/s]
3|4067|Loss: 0.7484727501869202:  63%|██████▎   | 4066/6470 [50:27<29:00,  1.38it/s]
3|4067|Loss: 0.7484727501869202:  63%|██████▎   | 4067/6470 [50:27<30:12,  1.33it/s]
3|4068|Loss: 0.8616235256195068:  63%|██████▎   | 4067/6470 [50:28<30:12,  1.33it/s]
3|4068|Loss: 0.8616235256195068:  63%|██████▎   | 4068/6470 [50:28<31:26,  1.27it/s]
3|4069|Loss: 0.7971948385238647:  63%|██████▎   | 4068/6470 [50:29<31:26,  1.27it/s]
3|4069|Loss: 0.7971948385238647:  63%|██████▎   | 4069/6470 [50:29<31:25,  1.27it/s]
3|4070|Loss: 0.7317919135093689:  63%|██████▎   | 4069/6470 [50:29<31:25,  1.27it/s]
3|4070|Loss: 0.7317919135093689:  63%|██████▎   | 4070/6470 [50:30<30:22,  1.32it/s]
3|4071|Loss: 0.8397451639175415:  63%|██████▎   | 4070/6470 [50:30<30:22,  1.32it/s]
3|4071|Loss: 0.8397451639175415:  63%|██████▎   | 4071/6470 [50:30<31:35,  1.27it/s]
3|4072|Loss: 0.7737354636192322:  63%|██████▎   | 4071/6470 [50:31<31:35,  1.27it/s]
3|4072|Loss: 0.7737354636192322:  63%|██████▎   | 4072/6470 [50:31<33:04,  1.21it/s]
3|4073|Loss: 0.8782978653907776:  63%|██████▎   | 4072/6470 [50:32<33:04,  1.21it/s]
3|4073|Loss: 0.8782978653907776:  63%|██████▎   | 4073/6470 [50:32<31:47,  1.26it/s]
3|4074|Loss: 0.9377481341362:  63%|██████▎   | 4073/6470 [50:32<31:47,  1.26it/s]   
3|4074|Loss: 0.9377481341362:  63%|██████▎   | 4074/6470 [50:33<29:52,  1.34it/s]
3|4075|Loss: 0.7107344269752502:  63%|██████▎   | 4074/6470 [50:33<29:52,  1.34it/s]
3|4075|Loss: 0.7107344269752502:  63%|██████▎   | 4075/6470 [50:33<27:32,  1.45it/s]
3|4076|Loss: 0.9008992314338684:  63%|██████▎   | 4075/6470 [50:34<27:32,  1.45it/s]
3|4076|Loss: 0.9008992314338684:  63%|██████▎   | 4076/6470 [50:34<30:13,  1.32it/s]
3|4077|Loss: 0.8325905203819275:  63%|██████▎   | 4076/6470 [50:35<30:13,  1.32it/s]
3|4077|Loss: 0.8325905203819275:  63%|██████▎   | 4077/6470 [50:35<29:01,  1.37it/s]
3|4078|Loss: 0.8685243725776672:  63%|██████▎   | 4077/6470 [50:35<29:01,  1.37it/s]
3|4078|Loss: 0.8685243725776672:  63%|██████▎   | 4078/6470 [50:35<27:12,  1.47it/s]
3|4079|Loss: 0.8090113401412964:  63%|██████▎   | 4078/6470 [50:36<27:12,  1.47it/s]
3|4079|Loss: 0.8090113401412964:  63%|██████▎   | 4079/6470 [50:36<29:39,  1.34it/s]
3|4080|Loss: 0.8858899474143982:  63%|██████▎   | 4079/6470 [50:37<29:39,  1.34it/s]
3|4080|Loss: 0.8858899474143982:  63%|██████▎   | 4080/6470 [50:37<30:45,  1.29it/s]
3|4081|Loss: 0.9515889883041382:  63%|██████▎   | 4080/6470 [50:38<30:45,  1.29it/s]
3|4081|Loss: 0.9515889883041382:  63%|██████▎   | 4081/6470 [50:38<31:39,  1.26it/s]
3|4082|Loss: 0.7823651432991028:  63%|██████▎   | 4081/6470 [50:38<31:39,  1.26it/s]
3|4082|Loss: 0.7823651432991028:  63%|██████▎   | 4082/6470 [50:39<31:35,  1.26it/s]
3|4083|Loss: 0.6559417247772217:  63%|██████▎   | 4082/6470 [50:39<31:35,  1.26it/s]
3|4083|Loss: 0.6559417247772217:  63%|██████▎   | 4083/6470 [50:40<32:34,  1.22it/s]
3|4084|Loss: 0.8168032765388489:  63%|██████▎   | 4083/6470 [50:40<32:34,  1.22it/s]
3|4084|Loss: 0.8168032765388489:  63%|██████▎   | 4084/6470 [50:41<33:54,  1.17it/s]
3|4085|Loss: 0.8401291370391846:  63%|██████▎   | 4084/6470 [50:41<33:54,  1.17it/s]
3|4085|Loss: 0.8401291370391846:  63%|██████▎   | 4085/6470 [50:42<34:19,  1.16it/s]
3|4086|Loss: 0.7347312569618225:  63%|██████▎   | 4085/6470 [50:42<34:19,  1.16it/s]
3|4086|Loss: 0.7347312569618225:  63%|██████▎   | 4086/6470 [50:42<33:47,  1.18it/s]
3|4087|Loss: 0.6610637307167053:  63%|██████▎   | 4086/6470 [50:43<33:47,  1.18it/s]
3|4087|Loss: 0.6610637307167053:  63%|██████▎   | 4087/6470 [50:43<32:41,  1.21it/s]
3|4088|Loss: 0.7937071323394775:  63%|██████▎   | 4087/6470 [50:43<32:41,  1.21it/s]
3|4088|Loss: 0.7937071323394775:  63%|██████▎   | 4088/6470 [50:44<28:17,  1.40it/s]
3|4089|Loss: 0.7818856239318848:  63%|██████▎   | 4088/6470 [50:44<28:17,  1.40it/s]
3|4089|Loss: 0.7818856239318848:  63%|██████▎   | 4089/6470 [50:44<28:08,  1.41it/s]
3|4090|Loss: 0.8276272416114807:  63%|██████▎   | 4089/6470 [50:45<28:08,  1.41it/s]
3|4090|Loss: 0.8276272416114807:  63%|██████▎   | 4090/6470 [50:45<29:14,  1.36it/s]
3|4091|Loss: 0.7330633997917175:  63%|██████▎   | 4090/6470 [50:45<29:14,  1.36it/s]
3|4091|Loss: 0.7330633997917175:  63%|██████▎   | 4091/6470 [50:46<28:19,  1.40it/s]
3|4092|Loss: 0.8219743371009827:  63%|██████▎   | 4091/6470 [50:46<28:19,  1.40it/s]
3|4092|Loss: 0.8219743371009827:  63%|██████▎   | 4092/6470 [50:47<28:55,  1.37it/s]
3|4093|Loss: 0.9546705484390259:  63%|██████▎   | 4092/6470 [50:47<28:55,  1.37it/s]
3|4093|Loss: 0.9546705484390259:  63%|██████▎   | 4093/6470 [50:47<30:56,  1.28it/s]
3|4094|Loss: 0.8123400807380676:  63%|██████▎   | 4093/6470 [50:48<30:56,  1.28it/s]
3|4094|Loss: 0.8123400807380676:  63%|██████▎   | 4094/6470 [50:48<30:23,  1.30it/s]
3|4095|Loss: 0.802126407623291:  63%|██████▎   | 4094/6470 [50:49<30:23,  1.30it/s] 
3|4095|Loss: 0.802126407623291:  63%|██████▎   | 4095/6470 [50:49<30:23,  1.30it/s]
3|4096|Loss: 0.9405727982521057:  63%|██████▎   | 4095/6470 [50:49<30:23,  1.30it/s]
3|4096|Loss: 0.9405727982521057:  63%|██████▎   | 4096/6470 [50:50<28:24,  1.39it/s]
3|4097|Loss: 0.9323412179946899:  63%|██████▎   | 4096/6470 [50:50<28:24,  1.39it/s]
3|4097|Loss: 0.9323412179946899:  63%|██████▎   | 4097/6470 [50:50<28:47,  1.37it/s]
3|4098|Loss: 0.7854180932044983:  63%|██████▎   | 4097/6470 [50:51<28:47,  1.37it/s]
3|4098|Loss: 0.7854180932044983:  63%|██████▎   | 4098/6470 [50:51<25:59,  1.52it/s]
3|4099|Loss: 0.9169416427612305:  63%|██████▎   | 4098/6470 [50:51<25:59,  1.52it/s]
3|4099|Loss: 0.9169416427612305:  63%|██████▎   | 4099/6470 [50:51<26:46,  1.48it/s]
3|4100|Loss: 0.8225849270820618:  63%|██████▎   | 4099/6470 [50:52<26:46,  1.48it/s]
3|4100|Loss: 0.8225849270820618:  63%|██████▎   | 4100/6470 [50:52<30:09,  1.31it/s]
3|4101|Loss: 0.9382236003875732:  63%|██████▎   | 4100/6470 [50:53<30:09,  1.31it/s]
3|4101|Loss: 0.9382236003875732:  63%|██████▎   | 4101/6470 [50:53<28:06,  1.40it/s]
3|4102|Loss: 0.7058557271957397:  63%|██████▎   | 4101/6470 [50:53<28:06,  1.40it/s]
3|4102|Loss: 0.7058557271957397:  63%|██████▎   | 4102/6470 [50:54<28:43,  1.37it/s]
3|4103|Loss: 0.7159590721130371:  63%|██████▎   | 4102/6470 [50:54<28:43,  1.37it/s]
3|4103|Loss: 0.7159590721130371:  63%|██████▎   | 4103/6470 [50:55<29:40,  1.33it/s]
3|4104|Loss: 0.8264411091804504:  63%|██████▎   | 4103/6470 [50:55<29:40,  1.33it/s]
3|4104|Loss: 0.8264411091804504:  63%|██████▎   | 4104/6470 [50:55<31:10,  1.26it/s]
3|4105|Loss: 0.840567946434021:  63%|██████▎   | 4104/6470 [50:56<31:10,  1.26it/s] 
3|4105|Loss: 0.840567946434021:  63%|██████▎   | 4105/6470 [50:56<30:28,  1.29it/s]
3|4106|Loss: 0.6478231549263:  63%|██████▎   | 4105/6470 [50:57<30:28,  1.29it/s]  
3|4106|Loss: 0.6478231549263:  63%|██████▎   | 4106/6470 [50:57<29:19,  1.34it/s]
3|4107|Loss: 0.908332347869873:  63%|██████▎   | 4106/6470 [50:57<29:19,  1.34it/s]
3|4107|Loss: 0.908332347869873:  63%|██████▎   | 4107/6470 [50:58<29:31,  1.33it/s]
3|4108|Loss: 0.9252846240997314:  63%|██████▎   | 4107/6470 [50:58<29:31,  1.33it/s]
3|4108|Loss: 0.9252846240997314:  63%|██████▎   | 4108/6470 [50:58<28:51,  1.36it/s]
3|4109|Loss: 0.8287846446037292:  63%|██████▎   | 4108/6470 [50:59<28:51,  1.36it/s]
3|4109|Loss: 0.8287846446037292:  64%|██████▎   | 4109/6470 [50:59<29:22,  1.34it/s]
3|4110|Loss: 0.9380422830581665:  64%|██████▎   | 4109/6470 [51:00<29:22,  1.34it/s]
3|4110|Loss: 0.9380422830581665:  64%|██████▎   | 4110/6470 [51:00<29:34,  1.33it/s]
3|4111|Loss: 0.8749895095825195:  64%|██████▎   | 4110/6470 [51:00<29:34,  1.33it/s]
3|4111|Loss: 0.8749895095825195:  64%|██████▎   | 4111/6470 [51:01<29:35,  1.33it/s]
3|4112|Loss: 0.8426756262779236:  64%|██████▎   | 4111/6470 [51:01<29:35,  1.33it/s]
3|4112|Loss: 0.8426756262779236:  64%|██████▎   | 4112/6470 [51:01<28:12,  1.39it/s]
3|4113|Loss: 0.7549502849578857:  64%|██████▎   | 4112/6470 [51:02<28:12,  1.39it/s]
3|4113|Loss: 0.7549502849578857:  64%|██████▎   | 4113/6470 [51:02<26:55,  1.46it/s]
3|4114|Loss: 0.8134120106697083:  64%|██████▎   | 4113/6470 [51:02<26:55,  1.46it/s]
3|4114|Loss: 0.8134120106697083:  64%|██████▎   | 4114/6470 [51:03<26:19,  1.49it/s]
3|4115|Loss: 0.7354274988174438:  64%|██████▎   | 4114/6470 [51:03<26:19,  1.49it/s]
3|4115|Loss: 0.7354274988174438:  64%|██████▎   | 4115/6470 [51:03<26:30,  1.48it/s]
3|4116|Loss: 0.9305285215377808:  64%|██████▎   | 4115/6470 [51:04<26:30,  1.48it/s]
3|4116|Loss: 0.9305285215377808:  64%|██████▎   | 4116/6470 [51:04<26:27,  1.48it/s]
3|4117|Loss: 0.9797115921974182:  64%|██████▎   | 4116/6470 [51:04<26:27,  1.48it/s]
3|4117|Loss: 0.9797115921974182:  64%|██████▎   | 4117/6470 [51:05<29:07,  1.35it/s]
3|4118|Loss: 0.976620614528656:  64%|██████▎   | 4117/6470 [51:05<29:07,  1.35it/s] 
3|4118|Loss: 0.976620614528656:  64%|██████▎   | 4118/6470 [51:06<29:25,  1.33it/s]
3|4119|Loss: 0.7089918255805969:  64%|██████▎   | 4118/6470 [51:06<29:25,  1.33it/s]
3|4119|Loss: 0.7089918255805969:  64%|██████▎   | 4119/6470 [51:06<28:49,  1.36it/s]
3|4120|Loss: 0.8828436732292175:  64%|██████▎   | 4119/6470 [51:07<28:49,  1.36it/s]
3|4120|Loss: 0.8828436732292175:  64%|██████▎   | 4120/6470 [51:07<29:30,  1.33it/s]
3|4121|Loss: 0.8734707832336426:  64%|██████▎   | 4120/6470 [51:07<29:30,  1.33it/s]
3|4121|Loss: 0.8734707832336426:  64%|██████▎   | 4121/6470 [51:08<28:28,  1.38it/s]
3|4122|Loss: 0.811934769153595:  64%|██████▎   | 4121/6470 [51:08<28:28,  1.38it/s] 
3|4122|Loss: 0.811934769153595:  64%|██████▎   | 4122/6470 [51:08<25:08,  1.56it/s]
3|4123|Loss: 0.7508727312088013:  64%|██████▎   | 4122/6470 [51:09<25:08,  1.56it/s]
3|4123|Loss: 0.7508727312088013:  64%|██████▎   | 4123/6470 [51:09<26:17,  1.49it/s]
3|4124|Loss: 0.9470459818840027:  64%|██████▎   | 4123/6470 [51:09<26:17,  1.49it/s]
3|4124|Loss: 0.9470459818840027:  64%|██████▎   | 4124/6470 [51:10<26:28,  1.48it/s]
3|4125|Loss: 0.9212672114372253:  64%|██████▎   | 4124/6470 [51:10<26:28,  1.48it/s]
3|4125|Loss: 0.9212672114372253:  64%|██████▍   | 4125/6470 [51:10<29:02,  1.35it/s]
3|4126|Loss: 0.9468286633491516:  64%|██████▍   | 4125/6470 [51:11<29:02,  1.35it/s]
3|4126|Loss: 0.9468286633491516:  64%|██████▍   | 4126/6470 [51:11<30:05,  1.30it/s]
3|4127|Loss: 0.8810058832168579:  64%|██████▍   | 4126/6470 [51:12<30:05,  1.30it/s]
3|4127|Loss: 0.8810058832168579:  64%|██████▍   | 4127/6470 [51:12<30:03,  1.30it/s]
3|4128|Loss: 0.7526838779449463:  64%|██████▍   | 4127/6470 [51:13<30:03,  1.30it/s]
3|4128|Loss: 0.7526838779449463:  64%|██████▍   | 4128/6470 [51:13<31:18,  1.25it/s]
3|4129|Loss: 0.7801097631454468:  64%|██████▍   | 4128/6470 [51:13<31:18,  1.25it/s]
3|4129|Loss: 0.7801097631454468:  64%|██████▍   | 4129/6470 [51:14<33:14,  1.17it/s]
3|4130|Loss: 0.9346442222595215:  64%|██████▍   | 4129/6470 [51:14<33:14,  1.17it/s]
3|4130|Loss: 0.9346442222595215:  64%|██████▍   | 4130/6470 [51:15<33:14,  1.17it/s]
3|4131|Loss: 0.9230098724365234:  64%|██████▍   | 4130/6470 [51:15<33:14,  1.17it/s]
3|4131|Loss: 0.9230098724365234:  64%|██████▍   | 4131/6470 [51:16<31:31,  1.24it/s]
3|4132|Loss: 0.7843402028083801:  64%|██████▍   | 4131/6470 [51:16<31:31,  1.24it/s]
3|4132|Loss: 0.7843402028083801:  64%|██████▍   | 4132/6470 [51:16<29:21,  1.33it/s]
3|4133|Loss: 0.7075026631355286:  64%|██████▍   | 4132/6470 [51:17<29:21,  1.33it/s]
3|4133|Loss: 0.7075026631355286:  64%|██████▍   | 4133/6470 [51:17<29:26,  1.32it/s]
3|4134|Loss: 0.7399336099624634:  64%|██████▍   | 4133/6470 [51:17<29:26,  1.32it/s]
3|4134|Loss: 0.7399336099624634:  64%|██████▍   | 4134/6470 [51:18<28:44,  1.35it/s]
3|4135|Loss: 0.7891969084739685:  64%|██████▍   | 4134/6470 [51:18<28:44,  1.35it/s]
3|4135|Loss: 0.7891969084739685:  64%|██████▍   | 4135/6470 [51:18<30:05,  1.29it/s]
3|4136|Loss: 0.8516151309013367:  64%|██████▍   | 4135/6470 [51:19<30:05,  1.29it/s]
3|4136|Loss: 0.8516151309013367:  64%|██████▍   | 4136/6470 [51:19<29:41,  1.31it/s]
3|4137|Loss: 0.8695861101150513:  64%|██████▍   | 4136/6470 [51:20<29:41,  1.31it/s]
3|4137|Loss: 0.8695861101150513:  64%|██████▍   | 4137/6470 [51:20<31:15,  1.24it/s]
3|4138|Loss: 0.9134743809700012:  64%|██████▍   | 4137/6470 [51:21<31:15,  1.24it/s]
3|4138|Loss: 0.9134743809700012:  64%|██████▍   | 4138/6470 [51:21<32:48,  1.18it/s]
3|4139|Loss: 0.8419381976127625:  64%|██████▍   | 4138/6470 [51:21<32:48,  1.18it/s]
3|4139|Loss: 0.8419381976127625:  64%|██████▍   | 4139/6470 [51:22<30:13,  1.29it/s]
3|4140|Loss: 0.7253890037536621:  64%|██████▍   | 4139/6470 [51:22<30:13,  1.29it/s]
3|4140|Loss: 0.7253890037536621:  64%|██████▍   | 4140/6470 [51:22<28:34,  1.36it/s]
3|4141|Loss: 0.9354886412620544:  64%|██████▍   | 4140/6470 [51:23<28:34,  1.36it/s]
3|4141|Loss: 0.9354886412620544:  64%|██████▍   | 4141/6470 [51:23<30:45,  1.26it/s]
3|4142|Loss: 0.7493802905082703:  64%|██████▍   | 4141/6470 [51:24<30:45,  1.26it/s]
3|4142|Loss: 0.7493802905082703:  64%|██████▍   | 4142/6470 [51:24<32:14,  1.20it/s]
3|4143|Loss: 0.8582058548927307:  64%|██████▍   | 4142/6470 [51:25<32:14,  1.20it/s]
3|4143|Loss: 0.8582058548927307:  64%|██████▍   | 4143/6470 [51:25<31:09,  1.24it/s]
3|4144|Loss: 0.7918841242790222:  64%|██████▍   | 4143/6470 [51:25<31:09,  1.24it/s]
3|4144|Loss: 0.7918841242790222:  64%|██████▍   | 4144/6470 [51:26<32:11,  1.20it/s]
3|4145|Loss: 0.8633226752281189:  64%|██████▍   | 4144/6470 [51:26<32:11,  1.20it/s]
3|4145|Loss: 0.8633226752281189:  64%|██████▍   | 4145/6470 [51:26<30:39,  1.26it/s]
3|4146|Loss: 0.922447919845581:  64%|██████▍   | 4145/6470 [51:27<30:39,  1.26it/s] 
3|4146|Loss: 0.922447919845581:  64%|██████▍   | 4146/6470 [51:27<30:26,  1.27it/s]
3|4147|Loss: 0.7931223511695862:  64%|██████▍   | 4146/6470 [51:28<30:26,  1.27it/s]
3|4147|Loss: 0.7931223511695862:  64%|██████▍   | 4147/6470 [51:28<30:23,  1.27it/s]
3|4148|Loss: 0.8872494697570801:  64%|██████▍   | 4147/6470 [51:28<30:23,  1.27it/s]
3|4148|Loss: 0.8872494697570801:  64%|██████▍   | 4148/6470 [51:29<29:32,  1.31it/s]
3|4149|Loss: 0.7465706467628479:  64%|██████▍   | 4148/6470 [51:29<29:32,  1.31it/s]
3|4149|Loss: 0.7465706467628479:  64%|██████▍   | 4149/6470 [51:30<31:07,  1.24it/s]
3|4150|Loss: 0.7501210570335388:  64%|██████▍   | 4149/6470 [51:30<31:07,  1.24it/s]
3|4150|Loss: 0.7501210570335388:  64%|██████▍   | 4150/6470 [51:30<28:47,  1.34it/s]
3|4151|Loss: 0.843126118183136:  64%|██████▍   | 4150/6470 [51:31<28:47,  1.34it/s] 
3|4151|Loss: 0.843126118183136:  64%|██████▍   | 4151/6470 [51:31<26:54,  1.44it/s]
3|4152|Loss: 0.7811500430107117:  64%|██████▍   | 4151/6470 [51:31<26:54,  1.44it/s]
3|4152|Loss: 0.7811500430107117:  64%|██████▍   | 4152/6470 [51:32<28:53,  1.34it/s]
3|4153|Loss: 0.8081737160682678:  64%|██████▍   | 4152/6470 [51:32<28:53,  1.34it/s]
3|4153|Loss: 0.8081737160682678:  64%|██████▍   | 4153/6470 [51:32<29:23,  1.31it/s]
3|4154|Loss: 0.6606345176696777:  64%|██████▍   | 4153/6470 [51:33<29:23,  1.31it/s]
3|4154|Loss: 0.6606345176696777:  64%|██████▍   | 4154/6470 [51:33<27:32,  1.40it/s]
3|4155|Loss: 0.9202547073364258:  64%|██████▍   | 4154/6470 [51:33<27:32,  1.40it/s]
3|4155|Loss: 0.9202547073364258:  64%|██████▍   | 4155/6470 [51:34<27:08,  1.42it/s]
3|4156|Loss: 0.7928792238235474:  64%|██████▍   | 4155/6470 [51:34<27:08,  1.42it/s]
3|4156|Loss: 0.7928792238235474:  64%|██████▍   | 4156/6470 [51:35<27:57,  1.38it/s]
3|4157|Loss: 0.8514888286590576:  64%|██████▍   | 4156/6470 [51:35<27:57,  1.38it/s]
3|4157|Loss: 0.8514888286590576:  64%|██████▍   | 4157/6470 [51:35<28:19,  1.36it/s]
3|4158|Loss: 0.7987638711929321:  64%|██████▍   | 4157/6470 [51:36<28:19,  1.36it/s]
3|4158|Loss: 0.7987638711929321:  64%|██████▍   | 4158/6470 [51:36<28:46,  1.34it/s]
3|4159|Loss: 0.856772780418396:  64%|██████▍   | 4158/6470 [51:37<28:46,  1.34it/s] 
3|4159|Loss: 0.856772780418396:  64%|██████▍   | 4159/6470 [51:37<29:58,  1.29it/s]
3|4160|Loss: 0.7854404449462891:  64%|██████▍   | 4159/6470 [51:37<29:58,  1.29it/s]
3|4160|Loss: 0.7854404449462891:  64%|██████▍   | 4160/6470 [51:38<31:34,  1.22it/s]
3|4161|Loss: 0.7550283670425415:  64%|██████▍   | 4160/6470 [51:38<31:34,  1.22it/s]
3|4161|Loss: 0.7550283670425415:  64%|██████▍   | 4161/6470 [51:39<31:32,  1.22it/s]
3|4162|Loss: 1.0059254169464111:  64%|██████▍   | 4161/6470 [51:39<31:32,  1.22it/s]
3|4162|Loss: 1.0059254169464111:  64%|██████▍   | 4162/6470 [51:39<29:20,  1.31it/s]
3|4163|Loss: 0.8780240416526794:  64%|██████▍   | 4162/6470 [51:40<29:20,  1.31it/s]
3|4163|Loss: 0.8780240416526794:  64%|██████▍   | 4163/6470 [51:40<27:54,  1.38it/s]
3|4164|Loss: 0.8557539582252502:  64%|██████▍   | 4163/6470 [51:40<27:54,  1.38it/s]
3|4164|Loss: 0.8557539582252502:  64%|██████▍   | 4164/6470 [51:41<28:18,  1.36it/s]
3|4165|Loss: 1.0268877744674683:  64%|██████▍   | 4164/6470 [51:41<28:18,  1.36it/s]
3|4165|Loss: 1.0268877744674683:  64%|██████▍   | 4165/6470 [51:41<28:59,  1.33it/s]
3|4166|Loss: 0.6842631101608276:  64%|██████▍   | 4165/6470 [51:42<28:59,  1.33it/s]
3|4166|Loss: 0.6842631101608276:  64%|██████▍   | 4166/6470 [51:42<28:33,  1.34it/s]
3|4167|Loss: 0.8901560306549072:  64%|██████▍   | 4166/6470 [51:43<28:33,  1.34it/s]
3|4167|Loss: 0.8901560306549072:  64%|██████▍   | 4167/6470 [51:43<27:59,  1.37it/s]
3|4168|Loss: 0.8656092882156372:  64%|██████▍   | 4167/6470 [51:43<27:59,  1.37it/s]
3|4168|Loss: 0.8656092882156372:  64%|██████▍   | 4168/6470 [51:44<26:53,  1.43it/s]
3|4169|Loss: 0.8602405190467834:  64%|██████▍   | 4168/6470 [51:44<26:53,  1.43it/s]
3|4169|Loss: 0.8602405190467834:  64%|██████▍   | 4169/6470 [51:44<26:17,  1.46it/s]
3|4170|Loss: 0.8118183612823486:  64%|██████▍   | 4169/6470 [51:45<26:17,  1.46it/s]
3|4170|Loss: 0.8118183612823486:  64%|██████▍   | 4170/6470 [51:45<28:28,  1.35it/s]
3|4171|Loss: 0.8231378793716431:  64%|██████▍   | 4170/6470 [51:46<28:28,  1.35it/s]
3|4171|Loss: 0.8231378793716431:  64%|██████▍   | 4171/6470 [51:46<29:29,  1.30it/s]
3|4172|Loss: 0.9088265299797058:  64%|██████▍   | 4171/6470 [51:46<29:29,  1.30it/s]
3|4172|Loss: 0.9088265299797058:  64%|██████▍   | 4172/6470 [51:47<30:46,  1.24it/s]
3|4173|Loss: 0.7685987949371338:  64%|██████▍   | 4172/6470 [51:47<30:46,  1.24it/s]
3|4173|Loss: 0.7685987949371338:  64%|██████▍   | 4173/6470 [51:48<30:46,  1.24it/s]
3|4174|Loss: 0.8547128438949585:  64%|██████▍   | 4173/6470 [51:48<30:46,  1.24it/s]
3|4174|Loss: 0.8547128438949585:  65%|██████▍   | 4174/6470 [51:48<28:44,  1.33it/s]
3|4175|Loss: 0.8238993287086487:  65%|██████▍   | 4174/6470 [51:49<28:44,  1.33it/s]
3|4175|Loss: 0.8238993287086487:  65%|██████▍   | 4175/6470 [51:49<31:06,  1.23it/s]
3|4176|Loss: 0.7831799387931824:  65%|██████▍   | 4175/6470 [51:50<31:06,  1.23it/s]
3|4176|Loss: 0.7831799387931824:  65%|██████▍   | 4176/6470 [51:50<30:11,  1.27it/s]
3|4177|Loss: 0.8300450444221497:  65%|██████▍   | 4176/6470 [51:50<30:11,  1.27it/s]
3|4177|Loss: 0.8300450444221497:  65%|██████▍   | 4177/6470 [51:51<29:29,  1.30it/s]
3|4178|Loss: 0.8124368190765381:  65%|██████▍   | 4177/6470 [51:51<29:29,  1.30it/s]
3|4178|Loss: 0.8124368190765381:  65%|██████▍   | 4178/6470 [51:51<28:28,  1.34it/s]
3|4179|Loss: 0.6877195835113525:  65%|██████▍   | 4178/6470 [51:52<28:28,  1.34it/s]
3|4179|Loss: 0.6877195835113525:  65%|██████▍   | 4179/6470 [51:52<27:40,  1.38it/s]
3|4180|Loss: 0.8936144113540649:  65%|██████▍   | 4179/6470 [51:52<27:40,  1.38it/s]
3|4180|Loss: 0.8936144113540649:  65%|██████▍   | 4180/6470 [51:53<28:11,  1.35it/s]
3|4181|Loss: 0.9749361872673035:  65%|██████▍   | 4180/6470 [51:53<28:11,  1.35it/s]
3|4181|Loss: 0.9749361872673035:  65%|██████▍   | 4181/6470 [51:54<28:59,  1.32it/s]
3|4182|Loss: 0.8877867460250854:  65%|██████▍   | 4181/6470 [51:54<28:59,  1.32it/s]
3|4182|Loss: 0.8877867460250854:  65%|██████▍   | 4182/6470 [51:54<27:48,  1.37it/s]
3|4183|Loss: 0.6978935599327087:  65%|██████▍   | 4182/6470 [51:55<27:48,  1.37it/s]
3|4183|Loss: 0.6978935599327087:  65%|██████▍   | 4183/6470 [51:55<25:56,  1.47it/s]
3|4184|Loss: 0.7519678473472595:  65%|██████▍   | 4183/6470 [51:55<25:56,  1.47it/s]
3|4184|Loss: 0.7519678473472595:  65%|██████▍   | 4184/6470 [51:55<25:51,  1.47it/s]
3|4185|Loss: 0.7106320858001709:  65%|██████▍   | 4184/6470 [51:56<25:51,  1.47it/s]
3|4185|Loss: 0.7106320858001709:  65%|██████▍   | 4185/6470 [51:56<26:34,  1.43it/s]
3|4186|Loss: 1.0529584884643555:  65%|██████▍   | 4185/6470 [51:57<26:34,  1.43it/s]
3|4186|Loss: 1.0529584884643555:  65%|██████▍   | 4186/6470 [51:57<28:42,  1.33it/s]
3|4187|Loss: 0.9094015955924988:  65%|██████▍   | 4186/6470 [51:58<28:42,  1.33it/s]
3|4187|Loss: 0.9094015955924988:  65%|██████▍   | 4187/6470 [51:58<30:29,  1.25it/s]
3|4188|Loss: 0.8641542196273804:  65%|██████▍   | 4187/6470 [51:58<30:29,  1.25it/s]
3|4188|Loss: 0.8641542196273804:  65%|██████▍   | 4188/6470 [51:59<28:20,  1.34it/s]
3|4189|Loss: 0.9283444881439209:  65%|██████▍   | 4188/6470 [51:59<28:20,  1.34it/s]
3|4189|Loss: 0.9283444881439209:  65%|██████▍   | 4189/6470 [51:59<27:05,  1.40it/s]
3|4190|Loss: 1.0107765197753906:  65%|██████▍   | 4189/6470 [52:00<27:05,  1.40it/s]
3|4190|Loss: 1.0107765197753906:  65%|██████▍   | 4190/6470 [52:00<24:49,  1.53it/s]
3|4191|Loss: 0.8363504409790039:  65%|██████▍   | 4190/6470 [52:00<24:49,  1.53it/s]
3|4191|Loss: 0.8363504409790039:  65%|██████▍   | 4191/6470 [52:00<23:20,  1.63it/s]
3|4192|Loss: 0.7143334746360779:  65%|██████▍   | 4191/6470 [52:01<23:20,  1.63it/s]
3|4192|Loss: 0.7143334746360779:  65%|██████▍   | 4192/6470 [52:01<20:48,  1.82it/s]
3|4193|Loss: 0.6945091485977173:  65%|██████▍   | 4192/6470 [52:01<20:48,  1.82it/s]
3|4193|Loss: 0.6945091485977173:  65%|██████▍   | 4193/6470 [52:01<20:12,  1.88it/s]
3|4194|Loss: 0.7778251767158508:  65%|██████▍   | 4193/6470 [52:02<20:12,  1.88it/s]
3|4194|Loss: 0.7778251767158508:  65%|██████▍   | 4194/6470 [52:02<22:37,  1.68it/s]
3|4195|Loss: 0.758124053478241:  65%|██████▍   | 4194/6470 [52:02<22:37,  1.68it/s] 
3|4195|Loss: 0.758124053478241:  65%|██████▍   | 4195/6470 [52:03<22:45,  1.67it/s]
3|4196|Loss: 0.9777821898460388:  65%|██████▍   | 4195/6470 [52:03<22:45,  1.67it/s]
3|4196|Loss: 0.9777821898460388:  65%|██████▍   | 4196/6470 [52:03<22:31,  1.68it/s]
3|4197|Loss: 0.7766711115837097:  65%|██████▍   | 4196/6470 [52:03<22:31,  1.68it/s]
3|4197|Loss: 0.7766711115837097:  65%|██████▍   | 4197/6470 [52:04<21:27,  1.77it/s]
3|4198|Loss: 0.787391722202301:  65%|██████▍   | 4197/6470 [52:04<21:27,  1.77it/s] 
3|4198|Loss: 0.787391722202301:  65%|██████▍   | 4198/6470 [52:04<22:17,  1.70it/s]
3|4199|Loss: 0.7530282735824585:  65%|██████▍   | 4198/6470 [52:05<22:17,  1.70it/s]
3|4199|Loss: 0.7530282735824585:  65%|██████▍   | 4199/6470 [52:05<22:09,  1.71it/s]
3|4200|Loss: 0.8713567852973938:  65%|██████▍   | 4199/6470 [52:05<22:09,  1.71it/s]
3|4200|Loss: 0.8713567852973938:  65%|██████▍   | 4200/6470 [52:05<22:37,  1.67it/s]
3|4201|Loss: 0.7954085469245911:  65%|██████▍   | 4200/6470 [52:06<22:37,  1.67it/s]
3|4201|Loss: 0.7954085469245911:  65%|██████▍   | 4201/6470 [52:06<25:26,  1.49it/s]
3|4202|Loss: 0.9961409568786621:  65%|██████▍   | 4201/6470 [52:07<25:26,  1.49it/s]
3|4202|Loss: 0.9961409568786621:  65%|██████▍   | 4202/6470 [52:07<24:54,  1.52it/s]
3|4203|Loss: 0.921319842338562:  65%|██████▍   | 4202/6470 [52:07<24:54,  1.52it/s] 
3|4203|Loss: 0.921319842338562:  65%|██████▍   | 4203/6470 [52:08<25:05,  1.51it/s]
3|4204|Loss: 0.8262025117874146:  65%|██████▍   | 4203/6470 [52:08<25:05,  1.51it/s]
3|4204|Loss: 0.8262025117874146:  65%|██████▍   | 4204/6470 [52:08<24:56,  1.51it/s]
3|4205|Loss: 0.8903074860572815:  65%|██████▍   | 4204/6470 [52:09<24:56,  1.51it/s]
3|4205|Loss: 0.8903074860572815:  65%|██████▍   | 4205/6470 [52:09<24:09,  1.56it/s]
3|4206|Loss: 0.807344377040863:  65%|██████▍   | 4205/6470 [52:09<24:09,  1.56it/s] 
3|4206|Loss: 0.807344377040863:  65%|██████▌   | 4206/6470 [52:10<26:29,  1.42it/s]
3|4207|Loss: 0.805084764957428:  65%|██████▌   | 4206/6470 [52:10<26:29,  1.42it/s]
3|4207|Loss: 0.805084764957428:  65%|██████▌   | 4207/6470 [52:10<26:15,  1.44it/s]
3|4208|Loss: 0.7148758172988892:  65%|██████▌   | 4207/6470 [52:11<26:15,  1.44it/s]
3|4208|Loss: 0.7148758172988892:  65%|██████▌   | 4208/6470 [52:11<26:06,  1.44it/s]
3|4209|Loss: 0.7770450711250305:  65%|██████▌   | 4208/6470 [52:12<26:06,  1.44it/s]
3|4209|Loss: 0.7770450711250305:  65%|██████▌   | 4209/6470 [52:12<27:13,  1.38it/s]
3|4210|Loss: 0.7356924414634705:  65%|██████▌   | 4209/6470 [52:12<27:13,  1.38it/s]
3|4210|Loss: 0.7356924414634705:  65%|██████▌   | 4210/6470 [52:13<29:22,  1.28it/s]
3|4211|Loss: 0.9083935022354126:  65%|██████▌   | 4210/6470 [52:13<29:22,  1.28it/s]
3|4211|Loss: 0.9083935022354126:  65%|██████▌   | 4211/6470 [52:14<29:39,  1.27it/s]
3|4212|Loss: 0.8350263237953186:  65%|██████▌   | 4211/6470 [52:14<29:39,  1.27it/s]
3|4212|Loss: 0.8350263237953186:  65%|██████▌   | 4212/6470 [52:14<29:41,  1.27it/s]
3|4213|Loss: 0.7719026803970337:  65%|██████▌   | 4212/6470 [52:15<29:41,  1.27it/s]
3|4213|Loss: 0.7719026803970337:  65%|██████▌   | 4213/6470 [52:15<26:58,  1.39it/s]
3|4214|Loss: 0.7770158052444458:  65%|██████▌   | 4213/6470 [52:15<26:58,  1.39it/s]
3|4214|Loss: 0.7770158052444458:  65%|██████▌   | 4214/6470 [52:16<25:25,  1.48it/s]
3|4215|Loss: 0.787588357925415:  65%|██████▌   | 4214/6470 [52:16<25:25,  1.48it/s] 
3|4215|Loss: 0.787588357925415:  65%|██████▌   | 4215/6470 [52:16<24:42,  1.52it/s]
3|4216|Loss: 0.7072937488555908:  65%|██████▌   | 4215/6470 [52:16<24:42,  1.52it/s]
3|4216|Loss: 0.7072937488555908:  65%|██████▌   | 4216/6470 [52:17<24:41,  1.52it/s]
3|4217|Loss: 0.8441353440284729:  65%|██████▌   | 4216/6470 [52:17<24:41,  1.52it/s]
3|4217|Loss: 0.8441353440284729:  65%|██████▌   | 4217/6470 [52:18<25:50,  1.45it/s]
3|4218|Loss: 0.8257735371589661:  65%|██████▌   | 4217/6470 [52:18<25:50,  1.45it/s]
3|4218|Loss: 0.8257735371589661:  65%|██████▌   | 4218/6470 [52:18<27:54,  1.34it/s]
3|4219|Loss: 0.9388365745544434:  65%|██████▌   | 4218/6470 [52:19<27:54,  1.34it/s]
3|4219|Loss: 0.9388365745544434:  65%|██████▌   | 4219/6470 [52:19<27:27,  1.37it/s]
3|4220|Loss: 0.7549042105674744:  65%|██████▌   | 4219/6470 [52:20<27:27,  1.37it/s]
3|4220|Loss: 0.7549042105674744:  65%|██████▌   | 4220/6470 [52:20<27:48,  1.35it/s]
3|4221|Loss: 0.8107814788818359:  65%|██████▌   | 4220/6470 [52:20<27:48,  1.35it/s]
3|4221|Loss: 0.8107814788818359:  65%|██████▌   | 4221/6470 [52:21<29:25,  1.27it/s]
3|4222|Loss: 0.9205979108810425:  65%|██████▌   | 4221/6470 [52:21<29:25,  1.27it/s]
3|4222|Loss: 0.9205979108810425:  65%|██████▌   | 4222/6470 [52:21<28:31,  1.31it/s]
3|4223|Loss: 0.9674893021583557:  65%|██████▌   | 4222/6470 [52:22<28:31,  1.31it/s]
3|4223|Loss: 0.9674893021583557:  65%|██████▌   | 4223/6470 [52:22<27:50,  1.35it/s]
3|4224|Loss: 0.8314242959022522:  65%|██████▌   | 4223/6470 [52:23<27:50,  1.35it/s]
3|4224|Loss: 0.8314242959022522:  65%|██████▌   | 4224/6470 [52:23<26:57,  1.39it/s]
3|4225|Loss: 0.8845221996307373:  65%|██████▌   | 4224/6470 [52:23<26:57,  1.39it/s]
3|4225|Loss: 0.8845221996307373:  65%|██████▌   | 4225/6470 [52:24<27:11,  1.38it/s]
3|4226|Loss: 0.7661862373352051:  65%|██████▌   | 4225/6470 [52:24<27:11,  1.38it/s]
3|4226|Loss: 0.7661862373352051:  65%|██████▌   | 4226/6470 [52:24<27:42,  1.35it/s]
3|4227|Loss: 0.7874377965927124:  65%|██████▌   | 4226/6470 [52:25<27:42,  1.35it/s]
3|4227|Loss: 0.7874377965927124:  65%|██████▌   | 4227/6470 [52:25<26:14,  1.42it/s]
3|4228|Loss: 0.6608001589775085:  65%|██████▌   | 4227/6470 [52:25<26:14,  1.42it/s]
3|4228|Loss: 0.6608001589775085:  65%|██████▌   | 4228/6470 [52:26<26:16,  1.42it/s]
3|4229|Loss: 0.7865056395530701:  65%|██████▌   | 4228/6470 [52:26<26:16,  1.42it/s]
3|4229|Loss: 0.7865056395530701:  65%|██████▌   | 4229/6470 [52:26<25:37,  1.46it/s]
3|4230|Loss: 0.7470090389251709:  65%|██████▌   | 4229/6470 [52:27<25:37,  1.46it/s]
3|4230|Loss: 0.7470090389251709:  65%|██████▌   | 4230/6470 [52:27<26:28,  1.41it/s]
3|4231|Loss: 0.9213132262229919:  65%|██████▌   | 4230/6470 [52:28<26:28,  1.41it/s]
3|4231|Loss: 0.9213132262229919:  65%|██████▌   | 4231/6470 [52:28<26:38,  1.40it/s]
3|4232|Loss: 0.8978952765464783:  65%|██████▌   | 4231/6470 [52:28<26:38,  1.40it/s]
3|4232|Loss: 0.8978952765464783:  65%|██████▌   | 4232/6470 [52:29<29:56,  1.25it/s]
3|4233|Loss: 0.9228149056434631:  65%|██████▌   | 4232/6470 [52:29<29:56,  1.25it/s]
3|4233|Loss: 0.9228149056434631:  65%|██████▌   | 4233/6470 [52:30<30:16,  1.23it/s]
3|4234|Loss: 0.7556864619255066:  65%|██████▌   | 4233/6470 [52:30<30:16,  1.23it/s]
3|4234|Loss: 0.7556864619255066:  65%|██████▌   | 4234/6470 [52:30<27:03,  1.38it/s]
3|4235|Loss: 0.7622806429862976:  65%|██████▌   | 4234/6470 [52:31<27:03,  1.38it/s]
3|4235|Loss: 0.7622806429862976:  65%|██████▌   | 4235/6470 [52:31<25:27,  1.46it/s]
3|4236|Loss: 0.8045594096183777:  65%|██████▌   | 4235/6470 [52:31<25:27,  1.46it/s]
3|4236|Loss: 0.8045594096183777:  65%|██████▌   | 4236/6470 [52:31<23:53,  1.56it/s]
3|4237|Loss: 0.702546238899231:  65%|██████▌   | 4236/6470 [52:32<23:53,  1.56it/s] 
3|4237|Loss: 0.702546238899231:  65%|██████▌   | 4237/6470 [52:32<25:59,  1.43it/s]
3|4238|Loss: 0.9166305661201477:  65%|██████▌   | 4237/6470 [52:33<25:59,  1.43it/s]
3|4238|Loss: 0.9166305661201477:  66%|██████▌   | 4238/6470 [52:33<28:29,  1.31it/s]
3|4239|Loss: 0.7404485940933228:  66%|██████▌   | 4238/6470 [52:34<28:29,  1.31it/s]
3|4239|Loss: 0.7404485940933228:  66%|██████▌   | 4239/6470 [52:34<29:21,  1.27it/s]
3|4240|Loss: 0.7378448843955994:  66%|██████▌   | 4239/6470 [52:34<29:21,  1.27it/s]
3|4240|Loss: 0.7378448843955994:  66%|██████▌   | 4240/6470 [52:35<30:09,  1.23it/s]
3|4241|Loss: 0.7271413207054138:  66%|██████▌   | 4240/6470 [52:35<30:09,  1.23it/s]
3|4241|Loss: 0.7271413207054138:  66%|██████▌   | 4241/6470 [52:36<29:45,  1.25it/s]
3|4242|Loss: 0.834575355052948:  66%|██████▌   | 4241/6470 [52:36<29:45,  1.25it/s] 
3|4242|Loss: 0.834575355052948:  66%|██████▌   | 4242/6470 [52:36<30:10,  1.23it/s]
3|4243|Loss: 1.0005000829696655:  66%|██████▌   | 4242/6470 [52:37<30:10,  1.23it/s]
3|4243|Loss: 1.0005000829696655:  66%|██████▌   | 4243/6470 [52:37<28:23,  1.31it/s]
3|4244|Loss: 0.806452214717865:  66%|██████▌   | 4243/6470 [52:37<28:23,  1.31it/s] 
3|4244|Loss: 0.806452214717865:  66%|██████▌   | 4244/6470 [52:38<28:08,  1.32it/s]
3|4245|Loss: 0.9198179244995117:  66%|██████▌   | 4244/6470 [52:38<28:08,  1.32it/s]
3|4245|Loss: 0.9198179244995117:  66%|██████▌   | 4245/6470 [52:39<28:34,  1.30it/s]
3|4246|Loss: 0.8662176132202148:  66%|██████▌   | 4245/6470 [52:39<28:34,  1.30it/s]
3|4246|Loss: 0.8662176132202148:  66%|██████▌   | 4246/6470 [52:39<28:29,  1.30it/s]
3|4247|Loss: 0.7476785182952881:  66%|██████▌   | 4246/6470 [52:40<28:29,  1.30it/s]
3|4247|Loss: 0.7476785182952881:  66%|██████▌   | 4247/6470 [52:40<28:25,  1.30it/s]
3|4248|Loss: 0.8110148906707764:  66%|██████▌   | 4247/6470 [52:41<28:25,  1.30it/s]
3|4248|Loss: 0.8110148906707764:  66%|██████▌   | 4248/6470 [52:41<27:30,  1.35it/s]
3|4249|Loss: 0.8768687844276428:  66%|██████▌   | 4248/6470 [52:41<27:30,  1.35it/s]
3|4249|Loss: 0.8768687844276428:  66%|██████▌   | 4249/6470 [52:42<27:48,  1.33it/s]
3|4250|Loss: 0.8295689821243286:  66%|██████▌   | 4249/6470 [52:42<27:48,  1.33it/s]
3|4250|Loss: 0.8295689821243286:  66%|██████▌   | 4250/6470 [52:42<26:25,  1.40it/s]
3|4251|Loss: 0.6834525465965271:  66%|██████▌   | 4250/6470 [52:43<26:25,  1.40it/s]
3|4251|Loss: 0.6834525465965271:  66%|██████▌   | 4251/6470 [52:43<25:36,  1.44it/s]
3|4252|Loss: 0.7182753682136536:  66%|██████▌   | 4251/6470 [52:43<25:36,  1.44it/s]
3|4252|Loss: 0.7182753682136536:  66%|██████▌   | 4252/6470 [52:44<26:18,  1.41it/s]
3|4253|Loss: 0.6569547057151794:  66%|██████▌   | 4252/6470 [52:44<26:18,  1.41it/s]
3|4253|Loss: 0.6569547057151794:  66%|██████▌   | 4253/6470 [52:44<25:58,  1.42it/s]
3|4254|Loss: 0.8805201053619385:  66%|██████▌   | 4253/6470 [52:45<25:58,  1.42it/s]
3|4254|Loss: 0.8805201053619385:  66%|██████▌   | 4254/6470 [52:45<24:09,  1.53it/s]
3|4255|Loss: 0.8451939821243286:  66%|██████▌   | 4254/6470 [52:45<24:09,  1.53it/s]
3|4255|Loss: 0.8451939821243286:  66%|██████▌   | 4255/6470 [52:45<24:13,  1.52it/s]
3|4256|Loss: 0.7344207167625427:  66%|██████▌   | 4255/6470 [52:46<24:13,  1.52it/s]
3|4256|Loss: 0.7344207167625427:  66%|██████▌   | 4256/6470 [52:46<25:33,  1.44it/s]
3|4257|Loss: 0.9576334953308105:  66%|██████▌   | 4256/6470 [52:47<25:33,  1.44it/s]
3|4257|Loss: 0.9576334953308105:  66%|██████▌   | 4257/6470 [52:47<26:25,  1.40it/s]
3|4258|Loss: 0.8788341879844666:  66%|██████▌   | 4257/6470 [52:47<26:25,  1.40it/s]
3|4258|Loss: 0.8788341879844666:  66%|██████▌   | 4258/6470 [52:48<26:50,  1.37it/s]
3|4259|Loss: 0.8357949256896973:  66%|██████▌   | 4258/6470 [52:48<26:50,  1.37it/s]
3|4259|Loss: 0.8357949256896973:  66%|██████▌   | 4259/6470 [52:49<27:36,  1.33it/s]
3|4260|Loss: 0.7887200713157654:  66%|██████▌   | 4259/6470 [52:49<27:36,  1.33it/s]
3|4260|Loss: 0.7887200713157654:  66%|██████▌   | 4260/6470 [52:50<29:38,  1.24it/s]
3|4261|Loss: 0.7273669838905334:  66%|██████▌   | 4260/6470 [52:50<29:38,  1.24it/s]
3|4261|Loss: 0.7273669838905334:  66%|██████▌   | 4261/6470 [52:50<30:58,  1.19it/s]
3|4262|Loss: 0.7947373986244202:  66%|██████▌   | 4261/6470 [52:51<30:58,  1.19it/s]
3|4262|Loss: 0.7947373986244202:  66%|██████▌   | 4262/6470 [52:51<28:59,  1.27it/s]
3|4263|Loss: 0.8114107251167297:  66%|██████▌   | 4262/6470 [52:52<28:59,  1.27it/s]
3|4263|Loss: 0.8114107251167297:  66%|██████▌   | 4263/6470 [52:52<29:27,  1.25it/s]
3|4264|Loss: 0.8001047372817993:  66%|██████▌   | 4263/6470 [52:52<29:27,  1.25it/s]
3|4264|Loss: 0.8001047372817993:  66%|██████▌   | 4264/6470 [52:53<29:25,  1.25it/s]
3|4265|Loss: 0.8444022536277771:  66%|██████▌   | 4264/6470 [52:53<29:25,  1.25it/s]
3|4265|Loss: 0.8444022536277771:  66%|██████▌   | 4265/6470 [52:53<27:53,  1.32it/s]
3|4266|Loss: 0.9408429265022278:  66%|██████▌   | 4265/6470 [52:54<27:53,  1.32it/s]
3|4266|Loss: 0.9408429265022278:  66%|██████▌   | 4266/6470 [52:54<25:45,  1.43it/s]
3|4267|Loss: 0.7510000467300415:  66%|██████▌   | 4266/6470 [52:54<25:45,  1.43it/s]
3|4267|Loss: 0.7510000467300415:  66%|██████▌   | 4267/6470 [52:55<24:08,  1.52it/s]
3|4268|Loss: 0.7757611870765686:  66%|██████▌   | 4267/6470 [52:55<24:08,  1.52it/s]
3|4268|Loss: 0.7757611870765686:  66%|██████▌   | 4268/6470 [52:55<25:07,  1.46it/s]
3|4269|Loss: 0.9639374613761902:  66%|██████▌   | 4268/6470 [52:56<25:07,  1.46it/s]
3|4269|Loss: 0.9639374613761902:  66%|██████▌   | 4269/6470 [52:56<23:33,  1.56it/s]
3|4270|Loss: 0.8213967680931091:  66%|██████▌   | 4269/6470 [52:56<23:33,  1.56it/s]
3|4270|Loss: 0.8213967680931091:  66%|██████▌   | 4270/6470 [52:57<24:21,  1.50it/s]
3|4271|Loss: 0.9673834443092346:  66%|██████▌   | 4270/6470 [52:57<24:21,  1.50it/s]
3|4271|Loss: 0.9673834443092346:  66%|██████▌   | 4271/6470 [52:57<26:55,  1.36it/s]
3|4272|Loss: 0.8960056900978088:  66%|██████▌   | 4271/6470 [52:58<26:55,  1.36it/s]
3|4272|Loss: 0.8960056900978088:  66%|██████▌   | 4272/6470 [52:58<28:01,  1.31it/s]
3|4273|Loss: 0.8551785945892334:  66%|██████▌   | 4272/6470 [52:59<28:01,  1.31it/s]
3|4273|Loss: 0.8551785945892334:  66%|██████▌   | 4273/6470 [52:59<28:39,  1.28it/s]
3|4274|Loss: 0.7733688950538635:  66%|██████▌   | 4273/6470 [53:00<28:39,  1.28it/s]
3|4274|Loss: 0.7733688950538635:  66%|██████▌   | 4274/6470 [53:00<28:03,  1.30it/s]
3|4275|Loss: 0.8656767010688782:  66%|██████▌   | 4274/6470 [53:00<28:03,  1.30it/s]
3|4275|Loss: 0.8656767010688782:  66%|██████▌   | 4275/6470 [53:01<29:07,  1.26it/s]
3|4276|Loss: 0.835427463054657:  66%|██████▌   | 4275/6470 [53:01<29:07,  1.26it/s] 
3|4276|Loss: 0.835427463054657:  66%|██████▌   | 4276/6470 [53:01<29:21,  1.25it/s]
3|4277|Loss: 0.8398287296295166:  66%|██████▌   | 4276/6470 [53:02<29:21,  1.25it/s]
3|4277|Loss: 0.8398287296295166:  66%|██████▌   | 4277/6470 [53:02<29:16,  1.25it/s]
3|4278|Loss: 0.7050207853317261:  66%|██████▌   | 4277/6470 [53:03<29:16,  1.25it/s]
3|4278|Loss: 0.7050207853317261:  66%|██████▌   | 4278/6470 [53:03<29:59,  1.22it/s]
3|4279|Loss: 0.8043028116226196:  66%|██████▌   | 4278/6470 [53:04<29:59,  1.22it/s]
3|4279|Loss: 0.8043028116226196:  66%|██████▌   | 4279/6470 [53:04<29:52,  1.22it/s]
3|4280|Loss: 0.8826234340667725:  66%|██████▌   | 4279/6470 [53:04<29:52,  1.22it/s]
3|4280|Loss: 0.8826234340667725:  66%|██████▌   | 4280/6470 [53:05<29:45,  1.23it/s]
3|4281|Loss: 0.7629809379577637:  66%|██████▌   | 4280/6470 [53:05<29:45,  1.23it/s]
3|4281|Loss: 0.7629809379577637:  66%|██████▌   | 4281/6470 [53:06<29:16,  1.25it/s]
3|4282|Loss: 0.901196300983429:  66%|██████▌   | 4281/6470 [53:06<29:16,  1.25it/s] 
3|4282|Loss: 0.901196300983429:  66%|██████▌   | 4282/6470 [53:06<28:10,  1.29it/s]
3|4283|Loss: 1.1856067180633545:  66%|██████▌   | 4282/6470 [53:07<28:10,  1.29it/s]
3|4283|Loss: 1.1856067180633545:  66%|██████▌   | 4283/6470 [53:07<28:03,  1.30it/s]
3|4284|Loss: 0.905574381351471:  66%|██████▌   | 4283/6470 [53:07<28:03,  1.30it/s] 
3|4284|Loss: 0.905574381351471:  66%|██████▌   | 4284/6470 [53:08<28:15,  1.29it/s]
3|4285|Loss: 0.7616472840309143:  66%|██████▌   | 4284/6470 [53:08<28:15,  1.29it/s]
3|4285|Loss: 0.7616472840309143:  66%|██████▌   | 4285/6470 [53:08<27:01,  1.35it/s]
3|4286|Loss: 0.9911825060844421:  66%|██████▌   | 4285/6470 [53:09<27:01,  1.35it/s]
3|4286|Loss: 0.9911825060844421:  66%|██████▌   | 4286/6470 [53:09<26:00,  1.40it/s]
3|4287|Loss: 0.7947214245796204:  66%|██████▌   | 4286/6470 [53:10<26:00,  1.40it/s]
3|4287|Loss: 0.7947214245796204:  66%|██████▋   | 4287/6470 [53:10<25:32,  1.42it/s]
3|4288|Loss: 0.6759390830993652:  66%|██████▋   | 4287/6470 [53:10<25:32,  1.42it/s]
3|4288|Loss: 0.6759390830993652:  66%|██████▋   | 4288/6470 [53:11<27:57,  1.30it/s]
3|4289|Loss: 0.8403749465942383:  66%|██████▋   | 4288/6470 [53:11<27:57,  1.30it/s]
3|4289|Loss: 0.8403749465942383:  66%|██████▋   | 4289/6470 [53:11<27:43,  1.31it/s]
3|4290|Loss: 1.148569107055664:  66%|██████▋   | 4289/6470 [53:12<27:43,  1.31it/s] 
3|4290|Loss: 1.148569107055664:  66%|██████▋   | 4290/6470 [53:12<28:37,  1.27it/s]
3|4291|Loss: 0.7959762215614319:  66%|██████▋   | 4290/6470 [53:13<28:37,  1.27it/s]
3|4291|Loss: 0.7959762215614319:  66%|██████▋   | 4291/6470 [53:13<29:25,  1.23it/s]
3|4292|Loss: 0.8301969766616821:  66%|██████▋   | 4291/6470 [53:14<29:25,  1.23it/s]
3|4292|Loss: 0.8301969766616821:  66%|██████▋   | 4292/6470 [53:14<26:56,  1.35it/s]
3|4293|Loss: 0.7994281053543091:  66%|██████▋   | 4292/6470 [53:14<26:56,  1.35it/s]
3|4293|Loss: 0.7994281053543091:  66%|██████▋   | 4293/6470 [53:14<24:03,  1.51it/s]
3|4294|Loss: 0.6756725907325745:  66%|██████▋   | 4293/6470 [53:15<24:03,  1.51it/s]
3|4294|Loss: 0.6756725907325745:  66%|██████▋   | 4294/6470 [53:15<23:47,  1.52it/s]
3|4295|Loss: 0.852985143661499:  66%|██████▋   | 4294/6470 [53:15<23:47,  1.52it/s] 
3|4295|Loss: 0.852985143661499:  66%|██████▋   | 4295/6470 [53:16<24:46,  1.46it/s]
3|4296|Loss: 0.8644962906837463:  66%|██████▋   | 4295/6470 [53:16<24:46,  1.46it/s]
3|4296|Loss: 0.8644962906837463:  66%|██████▋   | 4296/6470 [53:16<25:55,  1.40it/s]
3|4297|Loss: 0.8839438557624817:  66%|██████▋   | 4296/6470 [53:17<25:55,  1.40it/s]
3|4297|Loss: 0.8839438557624817:  66%|██████▋   | 4297/6470 [53:17<26:24,  1.37it/s]
3|4298|Loss: 0.6619067788124084:  66%|██████▋   | 4297/6470 [53:18<26:24,  1.37it/s]
3|4298|Loss: 0.6619067788124084:  66%|██████▋   | 4298/6470 [53:18<24:28,  1.48it/s]
3|4299|Loss: 0.8713808655738831:  66%|██████▋   | 4298/6470 [53:18<24:28,  1.48it/s]
3|4299|Loss: 0.8713808655738831:  66%|██████▋   | 4299/6470 [53:18<23:14,  1.56it/s]
3|4300|Loss: 0.9069283604621887:  66%|██████▋   | 4299/6470 [53:19<23:14,  1.56it/s]
3|4300|Loss: 0.9069283604621887:  66%|██████▋   | 4300/6470 [53:19<22:11,  1.63it/s]
3|4301|Loss: 0.788955807685852:  66%|██████▋   | 4300/6470 [53:19<22:11,  1.63it/s] 
3|4301|Loss: 0.788955807685852:  66%|██████▋   | 4301/6470 [53:20<23:33,  1.54it/s]
3|4302|Loss: 0.9192711114883423:  66%|██████▋   | 4301/6470 [53:20<23:33,  1.54it/s]
3|4302|Loss: 0.9192711114883423:  66%|██████▋   | 4302/6470 [53:20<26:12,  1.38it/s]
3|4303|Loss: 0.8997167944908142:  66%|██████▋   | 4302/6470 [53:21<26:12,  1.38it/s]
3|4303|Loss: 0.8997167944908142:  67%|██████▋   | 4303/6470 [53:21<26:52,  1.34it/s]
3|4304|Loss: 0.6582171320915222:  67%|██████▋   | 4303/6470 [53:22<26:52,  1.34it/s]
3|4304|Loss: 0.6582171320915222:  67%|██████▋   | 4304/6470 [53:22<27:48,  1.30it/s]
3|4305|Loss: 0.8701748251914978:  67%|██████▋   | 4304/6470 [53:23<27:48,  1.30it/s]
3|4305|Loss: 0.8701748251914978:  67%|██████▋   | 4305/6470 [53:23<28:17,  1.28it/s]
3|4306|Loss: 0.8005536198616028:  67%|██████▋   | 4305/6470 [53:23<28:17,  1.28it/s]
3|4306|Loss: 0.8005536198616028:  67%|██████▋   | 4306/6470 [53:24<26:55,  1.34it/s]
3|4307|Loss: 0.7330528497695923:  67%|██████▋   | 4306/6470 [53:24<26:55,  1.34it/s]
3|4307|Loss: 0.7330528497695923:  67%|██████▋   | 4307/6470 [53:24<25:52,  1.39it/s]
3|4308|Loss: 0.8060241341590881:  67%|██████▋   | 4307/6470 [53:25<25:52,  1.39it/s]
3|4308|Loss: 0.8060241341590881:  67%|██████▋   | 4308/6470 [53:25<26:06,  1.38it/s]
3|4309|Loss: 0.892616331577301:  67%|██████▋   | 4308/6470 [53:25<26:06,  1.38it/s] 
3|4309|Loss: 0.892616331577301:  67%|██████▋   | 4309/6470 [53:26<24:17,  1.48it/s]
3|4310|Loss: 0.7385039329528809:  67%|██████▋   | 4309/6470 [53:26<24:17,  1.48it/s]
3|4310|Loss: 0.7385039329528809:  67%|██████▋   | 4310/6470 [53:26<26:06,  1.38it/s]
3|4311|Loss: 0.7652822732925415:  67%|██████▋   | 4310/6470 [53:27<26:06,  1.38it/s]
3|4311|Loss: 0.7652822732925415:  67%|██████▋   | 4311/6470 [53:27<27:16,  1.32it/s]
3|4312|Loss: 0.8073427677154541:  67%|██████▋   | 4311/6470 [53:28<27:16,  1.32it/s]
3|4312|Loss: 0.8073427677154541:  67%|██████▋   | 4312/6470 [53:28<25:33,  1.41it/s]
3|4313|Loss: 0.6646610498428345:  67%|██████▋   | 4312/6470 [53:28<25:33,  1.41it/s]
3|4313|Loss: 0.6646610498428345:  67%|██████▋   | 4313/6470 [53:29<26:17,  1.37it/s]
3|4314|Loss: 0.7093119621276855:  67%|██████▋   | 4313/6470 [53:29<26:17,  1.37it/s]
3|4314|Loss: 0.7093119621276855:  67%|██████▋   | 4314/6470 [53:29<28:02,  1.28it/s]
3|4315|Loss: 0.6936283111572266:  67%|██████▋   | 4314/6470 [53:30<28:02,  1.28it/s]
3|4315|Loss: 0.6936283111572266:  67%|██████▋   | 4315/6470 [53:30<26:55,  1.33it/s]
3|4316|Loss: 0.7227099537849426:  67%|██████▋   | 4315/6470 [53:31<26:55,  1.33it/s]
3|4316|Loss: 0.7227099537849426:  67%|██████▋   | 4316/6470 [53:31<26:36,  1.35it/s]
3|4317|Loss: 0.9983773231506348:  67%|██████▋   | 4316/6470 [53:31<26:36,  1.35it/s]
3|4317|Loss: 0.9983773231506348:  67%|██████▋   | 4317/6470 [53:32<27:41,  1.30it/s]
3|4318|Loss: 0.7933957576751709:  67%|██████▋   | 4317/6470 [53:32<27:41,  1.30it/s]
3|4318|Loss: 0.7933957576751709:  67%|██████▋   | 4318/6470 [53:33<28:07,  1.28it/s]
3|4319|Loss: 0.6867287755012512:  67%|██████▋   | 4318/6470 [53:33<28:07,  1.28it/s]
3|4319|Loss: 0.6867287755012512:  67%|██████▋   | 4319/6470 [53:33<26:15,  1.37it/s]
3|4320|Loss: 0.91849285364151:  67%|██████▋   | 4319/6470 [53:33<26:15,  1.37it/s]  
3|4320|Loss: 0.91849285364151:  67%|██████▋   | 4320/6470 [53:34<24:58,  1.43it/s]
3|4321|Loss: 0.8772147297859192:  67%|██████▋   | 4320/6470 [53:34<24:58,  1.43it/s]
3|4321|Loss: 0.8772147297859192:  67%|██████▋   | 4321/6470 [53:35<27:22,  1.31it/s]
3|4322|Loss: 0.604860782623291:  67%|██████▋   | 4321/6470 [53:35<27:22,  1.31it/s] 
3|4322|Loss: 0.604860782623291:  67%|██████▋   | 4322/6470 [53:36<28:45,  1.25it/s]
3|4323|Loss: 0.7888432145118713:  67%|██████▋   | 4322/6470 [53:36<28:45,  1.25it/s]
3|4323|Loss: 0.7888432145118713:  67%|██████▋   | 4323/6470 [53:36<28:05,  1.27it/s]
3|4324|Loss: 1.0277875661849976:  67%|██████▋   | 4323/6470 [53:37<28:05,  1.27it/s]
3|4324|Loss: 1.0277875661849976:  67%|██████▋   | 4324/6470 [53:37<28:56,  1.24it/s]
3|4325|Loss: 0.8799754977226257:  67%|██████▋   | 4324/6470 [53:38<28:56,  1.24it/s]
3|4325|Loss: 0.8799754977226257:  67%|██████▋   | 4325/6470 [53:38<26:47,  1.33it/s]
3|4326|Loss: 0.8210035562515259:  67%|██████▋   | 4325/6470 [53:38<26:47,  1.33it/s]
3|4326|Loss: 0.8210035562515259:  67%|██████▋   | 4326/6470 [53:39<27:06,  1.32it/s]
3|4327|Loss: 0.7836460471153259:  67%|██████▋   | 4326/6470 [53:39<27:06,  1.32it/s]
3|4327|Loss: 0.7836460471153259:  67%|██████▋   | 4327/6470 [53:39<26:27,  1.35it/s]
3|4328|Loss: 0.964905321598053:  67%|██████▋   | 4327/6470 [53:40<26:27,  1.35it/s] 
3|4328|Loss: 0.964905321598053:  67%|██████▋   | 4328/6470 [53:40<26:48,  1.33it/s]
3|4329|Loss: 0.9383434057235718:  67%|██████▋   | 4328/6470 [53:40<26:48,  1.33it/s]
3|4329|Loss: 0.9383434057235718:  67%|██████▋   | 4329/6470 [53:41<26:34,  1.34it/s]
3|4330|Loss: 0.8097254633903503:  67%|██████▋   | 4329/6470 [53:41<26:34,  1.34it/s]
3|4330|Loss: 0.8097254633903503:  67%|██████▋   | 4330/6470 [53:41<26:06,  1.37it/s]
3|4331|Loss: 0.8042511940002441:  67%|██████▋   | 4330/6470 [53:42<26:06,  1.37it/s]
3|4331|Loss: 0.8042511940002441:  67%|██████▋   | 4331/6470 [53:42<26:12,  1.36it/s]
3|4332|Loss: 0.7318121790885925:  67%|██████▋   | 4331/6470 [53:43<26:12,  1.36it/s]
3|4332|Loss: 0.7318121790885925:  67%|██████▋   | 4332/6470 [53:43<25:50,  1.38it/s]
3|4333|Loss: 0.8183568716049194:  67%|██████▋   | 4332/6470 [53:43<25:50,  1.38it/s]
3|4333|Loss: 0.8183568716049194:  67%|██████▋   | 4333/6470 [53:44<26:00,  1.37it/s]
3|4334|Loss: 0.6134060025215149:  67%|██████▋   | 4333/6470 [53:44<26:00,  1.37it/s]
3|4334|Loss: 0.6134060025215149:  67%|██████▋   | 4334/6470 [53:44<23:48,  1.50it/s]
3|4335|Loss: 0.8196572065353394:  67%|██████▋   | 4334/6470 [53:45<23:48,  1.50it/s]
3|4335|Loss: 0.8196572065353394:  67%|██████▋   | 4335/6470 [53:45<24:11,  1.47it/s]
3|4336|Loss: 0.8819024562835693:  67%|██████▋   | 4335/6470 [53:45<24:11,  1.47it/s]
3|4336|Loss: 0.8819024562835693:  67%|██████▋   | 4336/6470 [53:46<25:19,  1.40it/s]
3|4337|Loss: 0.7856069803237915:  67%|██████▋   | 4336/6470 [53:46<25:19,  1.40it/s]
3|4337|Loss: 0.7856069803237915:  67%|██████▋   | 4337/6470 [53:47<27:50,  1.28it/s]
3|4338|Loss: 0.7609784007072449:  67%|██████▋   | 4337/6470 [53:47<27:50,  1.28it/s]
3|4338|Loss: 0.7609784007072449:  67%|██████▋   | 4338/6470 [53:47<28:00,  1.27it/s]
3|4339|Loss: 0.8408002257347107:  67%|██████▋   | 4338/6470 [53:48<28:00,  1.27it/s]
3|4339|Loss: 0.8408002257347107:  67%|██████▋   | 4339/6470 [53:48<27:55,  1.27it/s]
3|4340|Loss: 0.8258501887321472:  67%|██████▋   | 4339/6470 [53:49<27:55,  1.27it/s]
3|4340|Loss: 0.8258501887321472:  67%|██████▋   | 4340/6470 [53:49<26:24,  1.34it/s]
3|4341|Loss: 0.8061988949775696:  67%|██████▋   | 4340/6470 [53:49<26:24,  1.34it/s]
3|4341|Loss: 0.8061988949775696:  67%|██████▋   | 4341/6470 [53:50<25:45,  1.38it/s]
3|4342|Loss: 0.8468079566955566:  67%|██████▋   | 4341/6470 [53:50<25:45,  1.38it/s]
3|4342|Loss: 0.8468079566955566:  67%|██████▋   | 4342/6470 [53:50<26:22,  1.34it/s]
3|4343|Loss: 0.8784745931625366:  67%|██████▋   | 4342/6470 [53:51<26:22,  1.34it/s]
3|4343|Loss: 0.8784745931625366:  67%|██████▋   | 4343/6470 [53:51<26:15,  1.35it/s]
3|4344|Loss: 0.9332164525985718:  67%|██████▋   | 4343/6470 [53:51<26:15,  1.35it/s]
3|4344|Loss: 0.9332164525985718:  67%|██████▋   | 4344/6470 [53:52<26:34,  1.33it/s]
3|4345|Loss: 0.772183895111084:  67%|██████▋   | 4344/6470 [53:52<26:34,  1.33it/s] 
3|4345|Loss: 0.772183895111084:  67%|██████▋   | 4345/6470 [53:53<26:17,  1.35it/s]
3|4346|Loss: 0.8382497429847717:  67%|██████▋   | 4345/6470 [53:53<26:17,  1.35it/s]
3|4346|Loss: 0.8382497429847717:  67%|██████▋   | 4346/6470 [53:53<24:58,  1.42it/s]
3|4347|Loss: 0.8021036982536316:  67%|██████▋   | 4346/6470 [53:54<24:58,  1.42it/s]
3|4347|Loss: 0.8021036982536316:  67%|██████▋   | 4347/6470 [53:54<25:31,  1.39it/s]
3|4348|Loss: 0.7902175784111023:  67%|██████▋   | 4347/6470 [53:54<25:31,  1.39it/s]
3|4348|Loss: 0.7902175784111023:  67%|██████▋   | 4348/6470 [53:55<25:17,  1.40it/s]
3|4349|Loss: 0.8582387566566467:  67%|██████▋   | 4348/6470 [53:55<25:17,  1.40it/s]
3|4349|Loss: 0.8582387566566467:  67%|██████▋   | 4349/6470 [53:56<26:59,  1.31it/s]
3|4350|Loss: 0.8878157734870911:  67%|██████▋   | 4349/6470 [53:56<26:59,  1.31it/s]
3|4350|Loss: 0.8878157734870911:  67%|██████▋   | 4350/6470 [53:56<28:26,  1.24it/s]
3|4351|Loss: 0.7882589101791382:  67%|██████▋   | 4350/6470 [53:57<28:26,  1.24it/s]
3|4351|Loss: 0.7882589101791382:  67%|██████▋   | 4351/6470 [53:57<29:00,  1.22it/s]
3|4352|Loss: 0.8436428904533386:  67%|██████▋   | 4351/6470 [53:58<29:00,  1.22it/s]
3|4352|Loss: 0.8436428904533386:  67%|██████▋   | 4352/6470 [53:58<29:17,  1.21it/s]
3|4353|Loss: 0.9172318577766418:  67%|██████▋   | 4352/6470 [53:59<29:17,  1.21it/s]
3|4353|Loss: 0.9172318577766418:  67%|██████▋   | 4353/6470 [53:59<28:43,  1.23it/s]
3|4354|Loss: 0.7926837205886841:  67%|██████▋   | 4353/6470 [53:59<28:43,  1.23it/s]
3|4354|Loss: 0.7926837205886841:  67%|██████▋   | 4354/6470 [54:00<27:23,  1.29it/s]
3|4355|Loss: 0.7717310786247253:  67%|██████▋   | 4354/6470 [54:00<27:23,  1.29it/s]
3|4355|Loss: 0.7717310786247253:  67%|██████▋   | 4355/6470 [54:00<26:02,  1.35it/s]
3|4356|Loss: 0.7970830798149109:  67%|██████▋   | 4355/6470 [54:01<26:02,  1.35it/s]
3|4356|Loss: 0.7970830798149109:  67%|██████▋   | 4356/6470 [54:01<24:43,  1.43it/s]
3|4357|Loss: 0.8608022332191467:  67%|██████▋   | 4356/6470 [54:01<24:43,  1.43it/s]
3|4357|Loss: 0.8608022332191467:  67%|██████▋   | 4357/6470 [54:02<25:28,  1.38it/s]
3|4358|Loss: 0.7089582085609436:  67%|██████▋   | 4357/6470 [54:02<25:28,  1.38it/s]
3|4358|Loss: 0.7089582085609436:  67%|██████▋   | 4358/6470 [54:02<25:09,  1.40it/s]
3|4359|Loss: 0.8116143941879272:  67%|██████▋   | 4358/6470 [54:03<25:09,  1.40it/s]
3|4359|Loss: 0.8116143941879272:  67%|██████▋   | 4359/6470 [54:03<23:30,  1.50it/s]
3|4360|Loss: 0.7827361226081848:  67%|██████▋   | 4359/6470 [54:03<23:30,  1.50it/s]
3|4360|Loss: 0.7827361226081848:  67%|██████▋   | 4360/6470 [54:03<22:08,  1.59it/s]
3|4361|Loss: 0.8513290286064148:  67%|██████▋   | 4360/6470 [54:04<22:08,  1.59it/s]
3|4361|Loss: 0.8513290286064148:  67%|██████▋   | 4361/6470 [54:04<21:33,  1.63it/s]
3|4362|Loss: 0.814679741859436:  67%|██████▋   | 4361/6470 [54:04<21:33,  1.63it/s] 
3|4362|Loss: 0.814679741859436:  67%|██████▋   | 4362/6470 [54:05<22:47,  1.54it/s]
3|4363|Loss: 0.9231721758842468:  67%|██████▋   | 4362/6470 [54:05<22:47,  1.54it/s]
3|4363|Loss: 0.9231721758842468:  67%|██████▋   | 4363/6470 [54:05<23:19,  1.51it/s]
3|4364|Loss: 0.7247759699821472:  67%|██████▋   | 4363/6470 [54:06<23:19,  1.51it/s]
3|4364|Loss: 0.7247759699821472:  67%|██████▋   | 4364/6470 [54:06<24:28,  1.43it/s]
3|4365|Loss: 0.7202935814857483:  67%|██████▋   | 4364/6470 [54:07<24:28,  1.43it/s]
3|4365|Loss: 0.7202935814857483:  67%|██████▋   | 4365/6470 [54:07<25:24,  1.38it/s]
3|4366|Loss: 0.853361189365387:  67%|██████▋   | 4365/6470 [54:07<25:24,  1.38it/s] 
3|4366|Loss: 0.853361189365387:  67%|██████▋   | 4366/6470 [54:08<26:43,  1.31it/s]
3|4367|Loss: 0.9584934711456299:  67%|██████▋   | 4366/6470 [54:08<26:43,  1.31it/s]
3|4367|Loss: 0.9584934711456299:  67%|██████▋   | 4367/6470 [54:08<25:20,  1.38it/s]
3|4368|Loss: 1.01560378074646:  67%|██████▋   | 4367/6470 [54:09<25:20,  1.38it/s]  
3|4368|Loss: 1.01560378074646:  68%|██████▊   | 4368/6470 [54:09<26:20,  1.33it/s]
3|4369|Loss: 0.8058363795280457:  68%|██████▊   | 4368/6470 [54:10<26:20,  1.33it/s]
3|4369|Loss: 0.8058363795280457:  68%|██████▊   | 4369/6470 [54:10<27:49,  1.26it/s]
3|4370|Loss: 0.738519012928009:  68%|██████▊   | 4369/6470 [54:11<27:49,  1.26it/s] 
3|4370|Loss: 0.738519012928009:  68%|██████▊   | 4370/6470 [54:11<27:41,  1.26it/s]
3|4371|Loss: 0.8347194790840149:  68%|██████▊   | 4370/6470 [54:11<27:41,  1.26it/s]
3|4371|Loss: 0.8347194790840149:  68%|██████▊   | 4371/6470 [54:12<28:12,  1.24it/s]
3|4372|Loss: 0.9149031639099121:  68%|██████▊   | 4371/6470 [54:12<28:12,  1.24it/s]
3|4372|Loss: 0.9149031639099121:  68%|██████▊   | 4372/6470 [54:13<28:09,  1.24it/s]
3|4373|Loss: 0.8261383175849915:  68%|██████▊   | 4372/6470 [54:13<28:09,  1.24it/s]
3|4373|Loss: 0.8261383175849915:  68%|██████▊   | 4373/6470 [54:13<28:02,  1.25it/s]
3|4374|Loss: 0.6444109082221985:  68%|██████▊   | 4373/6470 [54:14<28:02,  1.25it/s]
3|4374|Loss: 0.6444109082221985:  68%|██████▊   | 4374/6470 [54:14<27:43,  1.26it/s]
3|4375|Loss: 0.7195637226104736:  68%|██████▊   | 4374/6470 [54:15<27:43,  1.26it/s]
3|4375|Loss: 0.7195637226104736:  68%|██████▊   | 4375/6470 [54:15<24:55,  1.40it/s]
3|4376|Loss: 0.82057124376297:  68%|██████▊   | 4375/6470 [54:15<24:55,  1.40it/s]  
3|4376|Loss: 0.82057124376297:  68%|██████▊   | 4376/6470 [54:15<23:42,  1.47it/s]
3|4377|Loss: 0.7703348994255066:  68%|██████▊   | 4376/6470 [54:16<23:42,  1.47it/s]
3|4377|Loss: 0.7703348994255066:  68%|██████▊   | 4377/6470 [54:16<25:40,  1.36it/s]
3|4378|Loss: 0.8024597764015198:  68%|██████▊   | 4377/6470 [54:17<25:40,  1.36it/s]
3|4378|Loss: 0.8024597764015198:  68%|██████▊   | 4378/6470 [54:17<23:57,  1.46it/s]
3|4379|Loss: 0.8394523859024048:  68%|██████▊   | 4378/6470 [54:17<23:57,  1.46it/s]
3|4379|Loss: 0.8394523859024048:  68%|██████▊   | 4379/6470 [54:17<23:33,  1.48it/s]
3|4380|Loss: 0.8871894478797913:  68%|██████▊   | 4379/6470 [54:18<23:33,  1.48it/s]
3|4380|Loss: 0.8871894478797913:  68%|██████▊   | 4380/6470 [54:18<23:52,  1.46it/s]
3|4381|Loss: 0.7921629548072815:  68%|██████▊   | 4380/6470 [54:18<23:52,  1.46it/s]
3|4381|Loss: 0.7921629548072815:  68%|██████▊   | 4381/6470 [54:19<22:07,  1.57it/s]
3|4382|Loss: 0.8645320534706116:  68%|██████▊   | 4381/6470 [54:19<22:07,  1.57it/s]
3|4382|Loss: 0.8645320534706116:  68%|██████▊   | 4382/6470 [54:19<21:40,  1.61it/s]
3|4383|Loss: 0.7836115956306458:  68%|██████▊   | 4382/6470 [54:20<21:40,  1.61it/s]
3|4383|Loss: 0.7836115956306458:  68%|██████▊   | 4383/6470 [54:20<22:00,  1.58it/s]
3|4384|Loss: 0.8715144395828247:  68%|██████▊   | 4383/6470 [54:20<22:00,  1.58it/s]
3|4384|Loss: 0.8715144395828247:  68%|██████▊   | 4384/6470 [54:21<24:09,  1.44it/s]
3|4385|Loss: 0.7177437543869019:  68%|██████▊   | 4384/6470 [54:21<24:09,  1.44it/s]
3|4385|Loss: 0.7177437543869019:  68%|██████▊   | 4385/6470 [54:21<24:58,  1.39it/s]
3|4386|Loss: 0.8208408355712891:  68%|██████▊   | 4385/6470 [54:22<24:58,  1.39it/s]
3|4386|Loss: 0.8208408355712891:  68%|██████▊   | 4386/6470 [54:22<23:32,  1.48it/s]
3|4387|Loss: 0.855776846408844:  68%|██████▊   | 4386/6470 [54:22<23:32,  1.48it/s] 
3|4387|Loss: 0.855776846408844:  68%|██████▊   | 4387/6470 [54:23<24:49,  1.40it/s]
3|4388|Loss: 0.9774445295333862:  68%|██████▊   | 4387/6470 [54:23<24:49,  1.40it/s]
3|4388|Loss: 0.9774445295333862:  68%|██████▊   | 4388/6470 [54:24<25:17,  1.37it/s]
3|4389|Loss: 0.8369391560554504:  68%|██████▊   | 4388/6470 [54:24<25:17,  1.37it/s]
3|4389|Loss: 0.8369391560554504:  68%|██████▊   | 4389/6470 [54:24<25:07,  1.38it/s]
3|4390|Loss: 0.8290493488311768:  68%|██████▊   | 4389/6470 [54:25<25:07,  1.38it/s]
3|4390|Loss: 0.8290493488311768:  68%|██████▊   | 4390/6470 [54:25<24:52,  1.39it/s]
3|4391|Loss: 0.9028290510177612:  68%|██████▊   | 4390/6470 [54:25<24:52,  1.39it/s]
3|4391|Loss: 0.9028290510177612:  68%|██████▊   | 4391/6470 [54:26<25:23,  1.36it/s]
3|4392|Loss: 0.8519836664199829:  68%|██████▊   | 4391/6470 [54:26<25:23,  1.36it/s]
3|4392|Loss: 0.8519836664199829:  68%|██████▊   | 4392/6470 [54:27<25:41,  1.35it/s]
3|4393|Loss: 0.8120518326759338:  68%|██████▊   | 4392/6470 [54:27<25:41,  1.35it/s]
3|4393|Loss: 0.8120518326759338:  68%|██████▊   | 4393/6470 [54:27<26:50,  1.29it/s]
3|4394|Loss: 0.741943895816803:  68%|██████▊   | 4393/6470 [54:28<26:50,  1.29it/s] 
3|4394|Loss: 0.741943895816803:  68%|██████▊   | 4394/6470 [54:28<27:08,  1.28it/s]
3|4395|Loss: 0.8447325229644775:  68%|██████▊   | 4394/6470 [54:29<27:08,  1.28it/s]
3|4395|Loss: 0.8447325229644775:  68%|██████▊   | 4395/6470 [54:29<28:07,  1.23it/s]
3|4396|Loss: 0.857867419719696:  68%|██████▊   | 4395/6470 [54:30<28:07,  1.23it/s] 
3|4396|Loss: 0.857867419719696:  68%|██████▊   | 4396/6470 [54:30<27:46,  1.24it/s]
3|4397|Loss: 0.9860883355140686:  68%|██████▊   | 4396/6470 [54:30<27:46,  1.24it/s]
3|4397|Loss: 0.9860883355140686:  68%|██████▊   | 4397/6470 [54:31<26:15,  1.32it/s]
3|4398|Loss: 0.9059019088745117:  68%|██████▊   | 4397/6470 [54:31<26:15,  1.32it/s]
3|4398|Loss: 0.9059019088745117:  68%|██████▊   | 4398/6470 [54:31<23:32,  1.47it/s]
3|4399|Loss: 0.8471453189849854:  68%|██████▊   | 4398/6470 [54:31<23:32,  1.47it/s]
3|4399|Loss: 0.8471453189849854:  68%|██████▊   | 4399/6470 [54:32<22:50,  1.51it/s]
3|4400|Loss: 0.9265942573547363:  68%|██████▊   | 4399/6470 [54:32<22:50,  1.51it/s]
3|4400|Loss: 0.9265942573547363:  68%|██████▊   | 4400/6470 [54:32<24:07,  1.43it/s]
3|4401|Loss: 0.9032225012779236:  68%|██████▊   | 4400/6470 [54:33<24:07,  1.43it/s]
3|4401|Loss: 0.9032225012779236:  68%|██████▊   | 4401/6470 [54:33<25:11,  1.37it/s]
3|4402|Loss: 0.7645219564437866:  68%|██████▊   | 4401/6470 [54:34<25:11,  1.37it/s]
3|4402|Loss: 0.7645219564437866:  68%|██████▊   | 4402/6470 [54:34<24:07,  1.43it/s]
3|4403|Loss: 0.9917542338371277:  68%|██████▊   | 4402/6470 [54:34<24:07,  1.43it/s]
3|4403|Loss: 0.9917542338371277:  68%|██████▊   | 4403/6470 [54:35<24:15,  1.42it/s]
3|4404|Loss: 0.7702813148498535:  68%|██████▊   | 4403/6470 [54:35<24:15,  1.42it/s]
3|4404|Loss: 0.7702813148498535:  68%|██████▊   | 4404/6470 [54:35<25:07,  1.37it/s]
3|4405|Loss: 0.9220514297485352:  68%|██████▊   | 4404/6470 [54:36<25:07,  1.37it/s]
3|4405|Loss: 0.9220514297485352:  68%|██████▊   | 4405/6470 [54:36<25:39,  1.34it/s]
3|4406|Loss: 0.8407284617424011:  68%|██████▊   | 4405/6470 [54:37<25:39,  1.34it/s]
3|4406|Loss: 0.8407284617424011:  68%|██████▊   | 4406/6470 [54:37<24:57,  1.38it/s]
3|4407|Loss: 0.6916637420654297:  68%|██████▊   | 4406/6470 [54:37<24:57,  1.38it/s]
3|4407|Loss: 0.6916637420654297:  68%|██████▊   | 4407/6470 [54:38<25:21,  1.36it/s]
3|4408|Loss: 0.8800277709960938:  68%|██████▊   | 4407/6470 [54:38<25:21,  1.36it/s]
3|4408|Loss: 0.8800277709960938:  68%|██████▊   | 4408/6470 [54:39<27:04,  1.27it/s]
3|4409|Loss: 0.8025757670402527:  68%|██████▊   | 4408/6470 [54:39<27:04,  1.27it/s]
3|4409|Loss: 0.8025757670402527:  68%|██████▊   | 4409/6470 [54:39<28:29,  1.21it/s]
3|4410|Loss: 0.693366289138794:  68%|██████▊   | 4409/6470 [54:40<28:29,  1.21it/s] 
3|4410|Loss: 0.693366289138794:  68%|██████▊   | 4410/6470 [54:40<28:18,  1.21it/s]
3|4411|Loss: 0.7336448431015015:  68%|██████▊   | 4410/6470 [54:41<28:18,  1.21it/s]
3|4411|Loss: 0.7336448431015015:  68%|██████▊   | 4411/6470 [54:41<26:54,  1.28it/s]
3|4412|Loss: 0.9393327832221985:  68%|██████▊   | 4411/6470 [54:41<26:54,  1.28it/s]
3|4412|Loss: 0.9393327832221985:  68%|██████▊   | 4412/6470 [54:42<26:23,  1.30it/s]
3|4413|Loss: 0.7982645034790039:  68%|██████▊   | 4412/6470 [54:42<26:23,  1.30it/s]
3|4413|Loss: 0.7982645034790039:  68%|██████▊   | 4413/6470 [54:42<26:30,  1.29it/s]
3|4414|Loss: 0.6988852620124817:  68%|██████▊   | 4413/6470 [54:43<26:30,  1.29it/s]
3|4414|Loss: 0.6988852620124817:  68%|██████▊   | 4414/6470 [54:43<28:53,  1.19it/s]
3|4415|Loss: 0.772601306438446:  68%|██████▊   | 4414/6470 [54:44<28:53,  1.19it/s] 
3|4415|Loss: 0.772601306438446:  68%|██████▊   | 4415/6470 [54:44<30:18,  1.13it/s]
3|4416|Loss: 0.8865750432014465:  68%|██████▊   | 4415/6470 [54:45<30:18,  1.13it/s]
3|4416|Loss: 0.8865750432014465:  68%|██████▊   | 4416/6470 [54:45<28:53,  1.18it/s]
3|4417|Loss: 0.8079177141189575:  68%|██████▊   | 4416/6470 [54:46<28:53,  1.18it/s]
3|4417|Loss: 0.8079177141189575:  68%|██████▊   | 4417/6470 [54:46<27:22,  1.25it/s]
3|4418|Loss: 0.9039252400398254:  68%|██████▊   | 4417/6470 [54:46<27:22,  1.25it/s]
3|4418|Loss: 0.9039252400398254:  68%|██████▊   | 4418/6470 [54:47<26:08,  1.31it/s]
3|4419|Loss: 0.7744532227516174:  68%|██████▊   | 4418/6470 [54:47<26:08,  1.31it/s]
3|4419|Loss: 0.7744532227516174:  68%|██████▊   | 4419/6470 [54:47<25:20,  1.35it/s]
3|4420|Loss: 0.849899411201477:  68%|██████▊   | 4419/6470 [54:48<25:20,  1.35it/s] 
3|4420|Loss: 0.849899411201477:  68%|██████▊   | 4420/6470 [54:48<25:46,  1.33it/s]
3|4421|Loss: 0.8441106677055359:  68%|██████▊   | 4420/6470 [54:49<25:46,  1.33it/s]
3|4421|Loss: 0.8441106677055359:  68%|██████▊   | 4421/6470 [54:49<27:20,  1.25it/s]
3|4422|Loss: 0.990310549736023:  68%|██████▊   | 4421/6470 [54:49<27:20,  1.25it/s] 
3|4422|Loss: 0.990310549736023:  68%|██████▊   | 4422/6470 [54:50<27:16,  1.25it/s]
3|4423|Loss: 0.6995235681533813:  68%|██████▊   | 4422/6470 [54:50<27:16,  1.25it/s]
3|4423|Loss: 0.6995235681533813:  68%|██████▊   | 4423/6470 [54:50<26:24,  1.29it/s]
3|4424|Loss: 0.8296538591384888:  68%|██████▊   | 4423/6470 [54:51<26:24,  1.29it/s]
3|4424|Loss: 0.8296538591384888:  68%|██████▊   | 4424/6470 [54:51<26:30,  1.29it/s]
3|4425|Loss: 0.8382896780967712:  68%|██████▊   | 4424/6470 [54:52<26:30,  1.29it/s]
3|4425|Loss: 0.8382896780967712:  68%|██████▊   | 4425/6470 [54:52<25:42,  1.33it/s]
3|4426|Loss: 0.8688414096832275:  68%|██████▊   | 4425/6470 [54:52<25:42,  1.33it/s]
3|4426|Loss: 0.8688414096832275:  68%|██████▊   | 4426/6470 [54:53<25:51,  1.32it/s]
3|4427|Loss: 0.8912604451179504:  68%|██████▊   | 4426/6470 [54:53<25:51,  1.32it/s]
3|4427|Loss: 0.8912604451179504:  68%|██████▊   | 4427/6470 [54:54<25:53,  1.31it/s]
3|4428|Loss: 0.9188231229782104:  68%|██████▊   | 4427/6470 [54:54<25:53,  1.31it/s]
3|4428|Loss: 0.9188231229782104:  68%|██████▊   | 4428/6470 [54:54<26:19,  1.29it/s]
3|4429|Loss: 0.9628583192825317:  68%|██████▊   | 4428/6470 [54:55<26:19,  1.29it/s]
3|4429|Loss: 0.9628583192825317:  68%|██████▊   | 4429/6470 [54:55<27:29,  1.24it/s]
3|4430|Loss: 0.8460171818733215:  68%|██████▊   | 4429/6470 [54:56<27:29,  1.24it/s]
3|4430|Loss: 0.8460171818733215:  68%|██████▊   | 4430/6470 [54:56<26:44,  1.27it/s]
3|4431|Loss: 0.7484611868858337:  68%|██████▊   | 4430/6470 [54:56<26:44,  1.27it/s]
3|4431|Loss: 0.7484611868858337:  68%|██████▊   | 4431/6470 [54:57<25:52,  1.31it/s]
3|4432|Loss: 0.784615695476532:  68%|██████▊   | 4431/6470 [54:57<25:52,  1.31it/s] 
3|4432|Loss: 0.784615695476532:  69%|██████▊   | 4432/6470 [54:57<23:21,  1.45it/s]
3|4433|Loss: 0.7825278043746948:  69%|██████▊   | 4432/6470 [54:58<23:21,  1.45it/s]
3|4433|Loss: 0.7825278043746948:  69%|██████▊   | 4433/6470 [54:58<23:16,  1.46it/s]
3|4434|Loss: 0.9638509154319763:  69%|██████▊   | 4433/6470 [54:58<23:16,  1.46it/s]
3|4434|Loss: 0.9638509154319763:  69%|██████▊   | 4434/6470 [54:58<22:26,  1.51it/s]
3|4435|Loss: 0.7615621089935303:  69%|██████▊   | 4434/6470 [54:59<22:26,  1.51it/s]
3|4435|Loss: 0.7615621089935303:  69%|██████▊   | 4435/6470 [54:59<24:17,  1.40it/s]
3|4436|Loss: 0.8388803601264954:  69%|██████▊   | 4435/6470 [55:00<24:17,  1.40it/s]
3|4436|Loss: 0.8388803601264954:  69%|██████▊   | 4436/6470 [55:00<24:36,  1.38it/s]
3|4437|Loss: 0.7394353151321411:  69%|██████▊   | 4436/6470 [55:00<24:36,  1.38it/s]
3|4437|Loss: 0.7394353151321411:  69%|██████▊   | 4437/6470 [55:01<23:07,  1.47it/s]
3|4438|Loss: 0.8908112645149231:  69%|██████▊   | 4437/6470 [55:01<23:07,  1.47it/s]
3|4438|Loss: 0.8908112645149231:  69%|██████▊   | 4438/6470 [55:01<24:18,  1.39it/s]
3|4439|Loss: 0.875092625617981:  69%|██████▊   | 4438/6470 [55:02<24:18,  1.39it/s] 
3|4439|Loss: 0.875092625617981:  69%|██████▊   | 4439/6470 [55:02<25:15,  1.34it/s]
3|4440|Loss: 0.8994864225387573:  69%|██████▊   | 4439/6470 [55:03<25:15,  1.34it/s]
3|4440|Loss: 0.8994864225387573:  69%|██████▊   | 4440/6470 [55:03<23:15,  1.45it/s]
3|4441|Loss: 0.776409387588501:  69%|██████▊   | 4440/6470 [55:03<23:15,  1.45it/s] 
3|4441|Loss: 0.776409387588501:  69%|██████▊   | 4441/6470 [55:03<21:35,  1.57it/s]
3|4442|Loss: 0.9235905408859253:  69%|██████▊   | 4441/6470 [55:04<21:35,  1.57it/s]
3|4442|Loss: 0.9235905408859253:  69%|██████▊   | 4442/6470 [55:04<21:11,  1.59it/s]
3|4443|Loss: 0.8488768339157104:  69%|██████▊   | 4442/6470 [55:04<21:11,  1.59it/s]
3|4443|Loss: 0.8488768339157104:  69%|██████▊   | 4443/6470 [55:05<21:06,  1.60it/s]
3|4444|Loss: 0.8939328193664551:  69%|██████▊   | 4443/6470 [55:05<21:06,  1.60it/s]
3|4444|Loss: 0.8939328193664551:  69%|██████▊   | 4444/6470 [55:05<23:16,  1.45it/s]
3|4445|Loss: 0.8220164775848389:  69%|██████▊   | 4444/6470 [55:06<23:16,  1.45it/s]
3|4445|Loss: 0.8220164775848389:  69%|██████▊   | 4445/6470 [55:06<25:19,  1.33it/s]
3|4446|Loss: 0.7207024097442627:  69%|██████▊   | 4445/6470 [55:07<25:19,  1.33it/s]
3|4446|Loss: 0.7207024097442627:  69%|██████▊   | 4446/6470 [55:07<26:49,  1.26it/s]
3|4447|Loss: 0.71846604347229:  69%|██████▊   | 4446/6470 [55:08<26:49,  1.26it/s]  
3|4447|Loss: 0.71846604347229:  69%|██████▊   | 4447/6470 [55:08<25:22,  1.33it/s]
3|4448|Loss: 0.7862358689308167:  69%|██████▊   | 4447/6470 [55:08<25:22,  1.33it/s]
3|4448|Loss: 0.7862358689308167:  69%|██████▊   | 4448/6470 [55:08<24:35,  1.37it/s]
3|4449|Loss: 0.7981535196304321:  69%|██████▊   | 4448/6470 [55:09<24:35,  1.37it/s]
3|4449|Loss: 0.7981535196304321:  69%|██████▉   | 4449/6470 [55:09<23:01,  1.46it/s]
3|4450|Loss: 0.6984140872955322:  69%|██████▉   | 4449/6470 [55:09<23:01,  1.46it/s]
3|4450|Loss: 0.6984140872955322:  69%|██████▉   | 4450/6470 [55:10<21:30,  1.57it/s]
3|4451|Loss: 0.9668284058570862:  69%|██████▉   | 4450/6470 [55:10<21:30,  1.57it/s]
3|4451|Loss: 0.9668284058570862:  69%|██████▉   | 4451/6470 [55:10<22:25,  1.50it/s]
3|4452|Loss: 0.8276439905166626:  69%|██████▉   | 4451/6470 [55:11<22:25,  1.50it/s]
3|4452|Loss: 0.8276439905166626:  69%|██████▉   | 4452/6470 [55:11<22:51,  1.47it/s]
3|4453|Loss: 0.7654452323913574:  69%|██████▉   | 4452/6470 [55:11<22:51,  1.47it/s]
3|4453|Loss: 0.7654452323913574:  69%|██████▉   | 4453/6470 [55:12<21:43,  1.55it/s]
3|4454|Loss: 0.8934719562530518:  69%|██████▉   | 4453/6470 [55:12<21:43,  1.55it/s]
3|4454|Loss: 0.8934719562530518:  69%|██████▉   | 4454/6470 [55:12<22:27,  1.50it/s]
3|4455|Loss: 0.7967717051506042:  69%|██████▉   | 4454/6470 [55:13<22:27,  1.50it/s]
3|4455|Loss: 0.7967717051506042:  69%|██████▉   | 4455/6470 [55:13<21:48,  1.54it/s]
3|4456|Loss: 0.8276100754737854:  69%|██████▉   | 4455/6470 [55:13<21:48,  1.54it/s]
3|4456|Loss: 0.8276100754737854:  69%|██████▉   | 4456/6470 [55:14<21:19,  1.57it/s]
3|4457|Loss: 0.8195134401321411:  69%|██████▉   | 4456/6470 [55:14<21:19,  1.57it/s]
3|4457|Loss: 0.8195134401321411:  69%|██████▉   | 4457/6470 [55:14<20:21,  1.65it/s]
3|4458|Loss: 0.9253260493278503:  69%|██████▉   | 4457/6470 [55:14<20:21,  1.65it/s]
3|4458|Loss: 0.9253260493278503:  69%|██████▉   | 4458/6470 [55:15<19:54,  1.68it/s]
3|4459|Loss: 0.9847304821014404:  69%|██████▉   | 4458/6470 [55:15<19:54,  1.68it/s]
3|4459|Loss: 0.9847304821014404:  69%|██████▉   | 4459/6470 [55:15<21:28,  1.56it/s]
3|4460|Loss: 0.8553948402404785:  69%|██████▉   | 4459/6470 [55:16<21:28,  1.56it/s]
3|4460|Loss: 0.8553948402404785:  69%|██████▉   | 4460/6470 [55:16<21:38,  1.55it/s]
3|4461|Loss: 0.8805950284004211:  69%|██████▉   | 4460/6470 [55:16<21:38,  1.55it/s]
3|4461|Loss: 0.8805950284004211:  69%|██████▉   | 4461/6470 [55:17<20:09,  1.66it/s]
3|4462|Loss: 0.5861186385154724:  69%|██████▉   | 4461/6470 [55:17<20:09,  1.66it/s]
3|4462|Loss: 0.5861186385154724:  69%|██████▉   | 4462/6470 [55:17<20:28,  1.63it/s]
3|4463|Loss: 0.8059134483337402:  69%|██████▉   | 4462/6470 [55:18<20:28,  1.63it/s]
3|4463|Loss: 0.8059134483337402:  69%|██████▉   | 4463/6470 [55:18<20:29,  1.63it/s]
3|4464|Loss: 0.9680041670799255:  69%|██████▉   | 4463/6470 [55:18<20:29,  1.63it/s]
3|4464|Loss: 0.9680041670799255:  69%|██████▉   | 4464/6470 [55:19<21:51,  1.53it/s]
3|4465|Loss: 0.9250510931015015:  69%|██████▉   | 4464/6470 [55:19<21:51,  1.53it/s]
3|4465|Loss: 0.9250510931015015:  69%|██████▉   | 4465/6470 [55:19<23:47,  1.40it/s]
3|4466|Loss: 0.71805739402771:  69%|██████▉   | 4465/6470 [55:20<23:47,  1.40it/s]  
3|4466|Loss: 0.71805739402771:  69%|██████▉   | 4466/6470 [55:20<25:18,  1.32it/s]
3|4467|Loss: 0.8146000504493713:  69%|██████▉   | 4466/6470 [55:21<25:18,  1.32it/s]
3|4467|Loss: 0.8146000504493713:  69%|██████▉   | 4467/6470 [55:21<23:45,  1.41it/s]
3|4468|Loss: 0.7380940914154053:  69%|██████▉   | 4467/6470 [55:21<23:45,  1.41it/s]
3|4468|Loss: 0.7380940914154053:  69%|██████▉   | 4468/6470 [55:21<22:35,  1.48it/s]
3|4469|Loss: 0.79203861951828:  69%|██████▉   | 4468/6470 [55:22<22:35,  1.48it/s]  
3|4469|Loss: 0.79203861951828:  69%|██████▉   | 4469/6470 [55:22<23:30,  1.42it/s]
3|4470|Loss: 0.9255791902542114:  69%|██████▉   | 4469/6470 [55:23<23:30,  1.42it/s]
3|4470|Loss: 0.9255791902542114:  69%|██████▉   | 4470/6470 [55:23<23:13,  1.44it/s]
3|4471|Loss: 0.8272717595100403:  69%|██████▉   | 4470/6470 [55:23<23:13,  1.44it/s]
3|4471|Loss: 0.8272717595100403:  69%|██████▉   | 4471/6470 [55:24<24:20,  1.37it/s]
3|4472|Loss: 1.0252325534820557:  69%|██████▉   | 4471/6470 [55:24<24:20,  1.37it/s]
3|4472|Loss: 1.0252325534820557:  69%|██████▉   | 4472/6470 [55:25<26:07,  1.27it/s]
3|4473|Loss: 0.7636785507202148:  69%|██████▉   | 4472/6470 [55:25<26:07,  1.27it/s]
3|4473|Loss: 0.7636785507202148:  69%|██████▉   | 4473/6470 [55:25<25:25,  1.31it/s]
3|4474|Loss: 0.9402333498001099:  69%|██████▉   | 4473/6470 [55:26<25:25,  1.31it/s]
3|4474|Loss: 0.9402333498001099:  69%|██████▉   | 4474/6470 [55:26<25:11,  1.32it/s]
3|4475|Loss: 0.7240929007530212:  69%|██████▉   | 4474/6470 [55:26<25:11,  1.32it/s]
3|4475|Loss: 0.7240929007530212:  69%|██████▉   | 4475/6470 [55:27<24:05,  1.38it/s]
3|4476|Loss: 0.841688871383667:  69%|██████▉   | 4475/6470 [55:27<24:05,  1.38it/s] 
3|4476|Loss: 0.841688871383667:  69%|██████▉   | 4476/6470 [55:27<24:10,  1.38it/s]
3|4477|Loss: 0.7980780005455017:  69%|██████▉   | 4476/6470 [55:28<24:10,  1.38it/s]
3|4477|Loss: 0.7980780005455017:  69%|██████▉   | 4477/6470 [55:28<23:18,  1.42it/s]
3|4478|Loss: 0.7582772970199585:  69%|██████▉   | 4477/6470 [55:28<23:18,  1.42it/s]
3|4478|Loss: 0.7582772970199585:  69%|██████▉   | 4478/6470 [55:29<22:46,  1.46it/s]
3|4479|Loss: 0.6381636261940002:  69%|██████▉   | 4478/6470 [55:29<22:46,  1.46it/s]
3|4479|Loss: 0.6381636261940002:  69%|██████▉   | 4479/6470 [55:30<24:23,  1.36it/s]
3|4480|Loss: 0.9085986614227295:  69%|██████▉   | 4479/6470 [55:30<24:23,  1.36it/s]
3|4480|Loss: 0.9085986614227295:  69%|██████▉   | 4480/6470 [55:30<24:22,  1.36it/s]
3|4481|Loss: 0.8764732480049133:  69%|██████▉   | 4480/6470 [55:31<24:22,  1.36it/s]
3|4481|Loss: 0.8764732480049133:  69%|██████▉   | 4481/6470 [55:31<24:01,  1.38it/s]
3|4482|Loss: 0.9163922667503357:  69%|██████▉   | 4481/6470 [55:31<24:01,  1.38it/s]
3|4482|Loss: 0.9163922667503357:  69%|██████▉   | 4482/6470 [55:32<25:20,  1.31it/s]
3|4483|Loss: 0.9524282217025757:  69%|██████▉   | 4482/6470 [55:32<25:20,  1.31it/s]
3|4483|Loss: 0.9524282217025757:  69%|██████▉   | 4483/6470 [55:33<27:59,  1.18it/s]
3|4484|Loss: 0.8825548887252808:  69%|██████▉   | 4483/6470 [55:33<27:59,  1.18it/s]
3|4484|Loss: 0.8825548887252808:  69%|██████▉   | 4484/6470 [55:34<29:06,  1.14it/s]
3|4485|Loss: 0.8037319183349609:  69%|██████▉   | 4484/6470 [55:34<29:06,  1.14it/s]
3|4485|Loss: 0.8037319183349609:  69%|██████▉   | 4485/6470 [55:35<28:28,  1.16it/s]
3|4486|Loss: 0.8991484642028809:  69%|██████▉   | 4485/6470 [55:35<28:28,  1.16it/s]
3|4486|Loss: 0.8991484642028809:  69%|██████▉   | 4486/6470 [55:36<27:55,  1.18it/s]
3|4487|Loss: 0.8821693658828735:  69%|██████▉   | 4486/6470 [55:36<27:55,  1.18it/s]
3|4487|Loss: 0.8821693658828735:  69%|██████▉   | 4487/6470 [55:36<27:31,  1.20it/s]
3|4488|Loss: 0.8118115067481995:  69%|██████▉   | 4487/6470 [55:37<27:31,  1.20it/s]
3|4488|Loss: 0.8118115067481995:  69%|██████▉   | 4488/6470 [55:37<28:46,  1.15it/s]
3|4489|Loss: 0.8213014602661133:  69%|██████▉   | 4488/6470 [55:38<28:46,  1.15it/s]
3|4489|Loss: 0.8213014602661133:  69%|██████▉   | 4489/6470 [55:38<25:55,  1.27it/s]
3|4490|Loss: 0.81812584400177:  69%|██████▉   | 4489/6470 [55:38<25:55,  1.27it/s]  
3|4490|Loss: 0.81812584400177:  69%|██████▉   | 4490/6470 [55:39<26:34,  1.24it/s]
3|4491|Loss: 0.7922669649124146:  69%|██████▉   | 4490/6470 [55:39<26:34,  1.24it/s]
3|4491|Loss: 0.7922669649124146:  69%|██████▉   | 4491/6470 [55:39<26:04,  1.26it/s]
3|4492|Loss: 0.7752363681793213:  69%|██████▉   | 4491/6470 [55:40<26:04,  1.26it/s]
3|4492|Loss: 0.7752363681793213:  69%|██████▉   | 4492/6470 [55:40<25:48,  1.28it/s]
3|4493|Loss: 0.8154616951942444:  69%|██████▉   | 4492/6470 [55:41<25:48,  1.28it/s]
3|4493|Loss: 0.8154616951942444:  69%|██████▉   | 4493/6470 [55:41<25:31,  1.29it/s]
3|4494|Loss: 0.7840346693992615:  69%|██████▉   | 4493/6470 [55:41<25:31,  1.29it/s]
3|4494|Loss: 0.7840346693992615:  69%|██████▉   | 4494/6470 [55:42<25:34,  1.29it/s]
3|4495|Loss: 0.7763150334358215:  69%|██████▉   | 4494/6470 [55:42<25:34,  1.29it/s]
3|4495|Loss: 0.7763150334358215:  69%|██████▉   | 4495/6470 [55:42<23:55,  1.38it/s]
3|4496|Loss: 0.7992783784866333:  69%|██████▉   | 4495/6470 [55:43<23:55,  1.38it/s]
3|4496|Loss: 0.7992783784866333:  69%|██████▉   | 4496/6470 [55:43<25:11,  1.31it/s]
3|4497|Loss: 0.8689302802085876:  69%|██████▉   | 4496/6470 [55:44<25:11,  1.31it/s]
3|4497|Loss: 0.8689302802085876:  70%|██████▉   | 4497/6470 [55:44<25:43,  1.28it/s]
3|4498|Loss: 0.8435996174812317:  70%|██████▉   | 4497/6470 [55:45<25:43,  1.28it/s]
3|4498|Loss: 0.8435996174812317:  70%|██████▉   | 4498/6470 [55:45<26:48,  1.23it/s]
3|4499|Loss: 0.8063048124313354:  70%|██████▉   | 4498/6470 [55:45<26:48,  1.23it/s]
3|4499|Loss: 0.8063048124313354:  70%|██████▉   | 4499/6470 [55:46<26:36,  1.23it/s]
3|4500|Loss: 0.9295909404754639:  70%|██████▉   | 4499/6470 [55:46<26:36,  1.23it/s]
3|4500|Loss: 0.9295909404754639:  70%|██████▉   | 4500/6470 [55:46<25:21,  1.29it/s]
3|4501|Loss: 0.7567662000656128:  70%|██████▉   | 4500/6470 [55:47<25:21,  1.29it/s]
3|4501|Loss: 0.7567662000656128:  70%|██████▉   | 4501/6470 [55:47<24:29,  1.34it/s]
3|4502|Loss: 0.764183521270752:  70%|██████▉   | 4501/6470 [55:48<24:29,  1.34it/s] 
3|4502|Loss: 0.764183521270752:  70%|██████▉   | 4502/6470 [55:48<25:46,  1.27it/s]
3|4503|Loss: 0.7913780212402344:  70%|██████▉   | 4502/6470 [55:48<25:46,  1.27it/s]
3|4503|Loss: 0.7913780212402344:  70%|██████▉   | 4503/6470 [55:49<25:27,  1.29it/s]
3|4504|Loss: 0.778449296951294:  70%|██████▉   | 4503/6470 [55:49<25:27,  1.29it/s] 
3|4504|Loss: 0.778449296951294:  70%|██████▉   | 4504/6470 [55:49<23:26,  1.40it/s]
3|4505|Loss: 0.8714702725410461:  70%|██████▉   | 4504/6470 [55:50<23:26,  1.40it/s]
3|4505|Loss: 0.8714702725410461:  70%|██████▉   | 4505/6470 [55:50<24:44,  1.32it/s]
3|4506|Loss: 0.7857773303985596:  70%|██████▉   | 4505/6470 [55:51<24:44,  1.32it/s]
3|4506|Loss: 0.7857773303985596:  70%|██████▉   | 4506/6470 [55:51<24:33,  1.33it/s]
3|4507|Loss: 0.9049381613731384:  70%|██████▉   | 4506/6470 [55:51<24:33,  1.33it/s]
3|4507|Loss: 0.9049381613731384:  70%|██████▉   | 4507/6470 [55:52<25:51,  1.26it/s]
3|4508|Loss: 0.8196730017662048:  70%|██████▉   | 4507/6470 [55:52<25:51,  1.26it/s]
3|4508|Loss: 0.8196730017662048:  70%|██████▉   | 4508/6470 [55:53<25:49,  1.27it/s]
3|4509|Loss: 0.7812044620513916:  70%|██████▉   | 4508/6470 [55:53<25:49,  1.27it/s]
3|4509|Loss: 0.7812044620513916:  70%|██████▉   | 4509/6470 [55:53<24:53,  1.31it/s]
3|4510|Loss: 0.8946328163146973:  70%|██████▉   | 4509/6470 [55:54<24:53,  1.31it/s]
3|4510|Loss: 0.8946328163146973:  70%|██████▉   | 4510/6470 [55:54<23:09,  1.41it/s]
3|4511|Loss: 0.8232041001319885:  70%|██████▉   | 4510/6470 [55:54<23:09,  1.41it/s]
3|4511|Loss: 0.8232041001319885:  70%|██████▉   | 4511/6470 [55:54<22:13,  1.47it/s]
3|4512|Loss: 0.6354653835296631:  70%|██████▉   | 4511/6470 [55:55<22:13,  1.47it/s]
3|4512|Loss: 0.6354653835296631:  70%|██████▉   | 4512/6470 [55:55<20:28,  1.59it/s]
3|4513|Loss: 0.6874023079872131:  70%|██████▉   | 4512/6470 [55:55<20:28,  1.59it/s]
3|4513|Loss: 0.6874023079872131:  70%|██████▉   | 4513/6470 [55:56<21:24,  1.52it/s]
3|4514|Loss: 0.6920931339263916:  70%|██████▉   | 4513/6470 [55:56<21:24,  1.52it/s]
3|4514|Loss: 0.6920931339263916:  70%|██████▉   | 4514/6470 [55:56<21:46,  1.50it/s]
3|4515|Loss: 0.8235486745834351:  70%|██████▉   | 4514/6470 [55:57<21:46,  1.50it/s]
3|4515|Loss: 0.8235486745834351:  70%|██████▉   | 4515/6470 [55:57<20:42,  1.57it/s]
3|4516|Loss: 0.9488211870193481:  70%|██████▉   | 4515/6470 [55:57<20:42,  1.57it/s]
3|4516|Loss: 0.9488211870193481:  70%|██████▉   | 4516/6470 [55:58<21:20,  1.53it/s]
3|4517|Loss: 0.8937135338783264:  70%|██████▉   | 4516/6470 [55:58<21:20,  1.53it/s]
3|4517|Loss: 0.8937135338783264:  70%|██████▉   | 4517/6470 [55:58<21:54,  1.49it/s]
3|4518|Loss: 0.7311031818389893:  70%|██████▉   | 4517/6470 [55:59<21:54,  1.49it/s]
3|4518|Loss: 0.7311031818389893:  70%|██████▉   | 4518/6470 [55:59<20:24,  1.59it/s]
3|4519|Loss: 0.7821378111839294:  70%|██████▉   | 4518/6470 [55:59<20:24,  1.59it/s]
3|4519|Loss: 0.7821378111839294:  70%|██████▉   | 4519/6470 [56:00<20:20,  1.60it/s]
3|4520|Loss: 0.722403347492218:  70%|██████▉   | 4519/6470 [56:00<20:20,  1.60it/s] 
3|4520|Loss: 0.722403347492218:  70%|██████▉   | 4520/6470 [56:00<21:49,  1.49it/s]
3|4521|Loss: 0.9175148606300354:  70%|██████▉   | 4520/6470 [56:01<21:49,  1.49it/s]
3|4521|Loss: 0.9175148606300354:  70%|██████▉   | 4521/6470 [56:01<22:44,  1.43it/s]
3|4522|Loss: 0.8971506953239441:  70%|██████▉   | 4521/6470 [56:01<22:44,  1.43it/s]
3|4522|Loss: 0.8971506953239441:  70%|██████▉   | 4522/6470 [56:02<23:19,  1.39it/s]
3|4523|Loss: 0.7101826667785645:  70%|██████▉   | 4522/6470 [56:02<23:19,  1.39it/s]
3|4523|Loss: 0.7101826667785645:  70%|██████▉   | 4523/6470 [56:03<22:55,  1.42it/s]
3|4524|Loss: 0.8042523860931396:  70%|██████▉   | 4523/6470 [56:03<22:55,  1.42it/s]
3|4524|Loss: 0.8042523860931396:  70%|██████▉   | 4524/6470 [56:03<24:20,  1.33it/s]
3|4525|Loss: 0.7853997945785522:  70%|██████▉   | 4524/6470 [56:04<24:20,  1.33it/s]
3|4525|Loss: 0.7853997945785522:  70%|██████▉   | 4525/6470 [56:04<23:32,  1.38it/s]
3|4526|Loss: 0.9482110738754272:  70%|██████▉   | 4525/6470 [56:04<23:32,  1.38it/s]
3|4526|Loss: 0.9482110738754272:  70%|██████▉   | 4526/6470 [56:05<23:46,  1.36it/s]
3|4527|Loss: 0.6766927242279053:  70%|██████▉   | 4526/6470 [56:05<23:46,  1.36it/s]
3|4527|Loss: 0.6766927242279053:  70%|██████▉   | 4527/6470 [56:05<22:24,  1.44it/s]
3|4528|Loss: 0.763887345790863:  70%|██████▉   | 4527/6470 [56:06<22:24,  1.44it/s] 
3|4528|Loss: 0.763887345790863:  70%|██████▉   | 4528/6470 [56:06<23:00,  1.41it/s]
3|4529|Loss: 0.7599442005157471:  70%|██████▉   | 4528/6470 [56:07<23:00,  1.41it/s]
3|4529|Loss: 0.7599442005157471:  70%|███████   | 4529/6470 [56:07<22:06,  1.46it/s]
3|4530|Loss: 0.9719682931900024:  70%|███████   | 4529/6470 [56:07<22:06,  1.46it/s]
3|4530|Loss: 0.9719682931900024:  70%|███████   | 4530/6470 [56:07<21:50,  1.48it/s]
3|4531|Loss: 0.896146297454834:  70%|███████   | 4530/6470 [56:08<21:50,  1.48it/s] 
3|4531|Loss: 0.896146297454834:  70%|███████   | 4531/6470 [56:08<21:56,  1.47it/s]
3|4532|Loss: 0.9452816247940063:  70%|███████   | 4531/6470 [56:08<21:56,  1.47it/s]
3|4532|Loss: 0.9452816247940063:  70%|███████   | 4532/6470 [56:09<20:18,  1.59it/s]
3|4533|Loss: 0.7434520721435547:  70%|███████   | 4532/6470 [56:09<20:18,  1.59it/s]
3|4533|Loss: 0.7434520721435547:  70%|███████   | 4533/6470 [56:09<20:29,  1.58it/s]
3|4534|Loss: 0.9605450630187988:  70%|███████   | 4533/6470 [56:10<20:29,  1.58it/s]
3|4534|Loss: 0.9605450630187988:  70%|███████   | 4534/6470 [56:10<21:09,  1.52it/s]
3|4535|Loss: 0.8813344836235046:  70%|███████   | 4534/6470 [56:10<21:09,  1.52it/s]
3|4535|Loss: 0.8813344836235046:  70%|███████   | 4535/6470 [56:11<21:36,  1.49it/s]
3|4536|Loss: 0.928104817867279:  70%|███████   | 4535/6470 [56:11<21:36,  1.49it/s] 
3|4536|Loss: 0.928104817867279:  70%|███████   | 4536/6470 [56:11<21:14,  1.52it/s]
3|4537|Loss: 0.8333834409713745:  70%|███████   | 4536/6470 [56:12<21:14,  1.52it/s]
3|4537|Loss: 0.8333834409713745:  70%|███████   | 4537/6470 [56:12<19:46,  1.63it/s]
3|4538|Loss: 0.7425606846809387:  70%|███████   | 4537/6470 [56:12<19:46,  1.63it/s]
3|4538|Loss: 0.7425606846809387:  70%|███████   | 4538/6470 [56:12<20:27,  1.57it/s]
3|4539|Loss: 0.747677743434906:  70%|███████   | 4538/6470 [56:13<20:27,  1.57it/s] 
3|4539|Loss: 0.747677743434906:  70%|███████   | 4539/6470 [56:13<22:49,  1.41it/s]
3|4540|Loss: 0.8477145433425903:  70%|███████   | 4539/6470 [56:14<22:49,  1.41it/s]
3|4540|Loss: 0.8477145433425903:  70%|███████   | 4540/6470 [56:14<23:57,  1.34it/s]
3|4541|Loss: 0.9330933690071106:  70%|███████   | 4540/6470 [56:15<23:57,  1.34it/s]
3|4541|Loss: 0.9330933690071106:  70%|███████   | 4541/6470 [56:15<24:13,  1.33it/s]
3|4542|Loss: 0.765717089176178:  70%|███████   | 4541/6470 [56:15<24:13,  1.33it/s] 
3|4542|Loss: 0.765717089176178:  70%|███████   | 4542/6470 [56:16<22:46,  1.41it/s]
3|4543|Loss: 0.8631334900856018:  70%|███████   | 4542/6470 [56:16<22:46,  1.41it/s]
3|4543|Loss: 0.8631334900856018:  70%|███████   | 4543/6470 [56:16<22:02,  1.46it/s]
3|4544|Loss: 0.952898383140564:  70%|███████   | 4543/6470 [56:17<22:02,  1.46it/s] 
3|4544|Loss: 0.952898383140564:  70%|███████   | 4544/6470 [56:17<23:27,  1.37it/s]
3|4545|Loss: 0.9290615916252136:  70%|███████   | 4544/6470 [56:18<23:27,  1.37it/s]
3|4545|Loss: 0.9290615916252136:  70%|███████   | 4545/6470 [56:18<24:47,  1.29it/s]
3|4546|Loss: 0.8030290007591248:  70%|███████   | 4545/6470 [56:18<24:47,  1.29it/s]
3|4546|Loss: 0.8030290007591248:  70%|███████   | 4546/6470 [56:19<25:04,  1.28it/s]
3|4547|Loss: 0.7610153555870056:  70%|███████   | 4546/6470 [56:19<25:04,  1.28it/s]
3|4547|Loss: 0.7610153555870056:  70%|███████   | 4547/6470 [56:20<25:35,  1.25it/s]
3|4548|Loss: 0.8275483250617981:  70%|███████   | 4547/6470 [56:20<25:35,  1.25it/s]
3|4548|Loss: 0.8275483250617981:  70%|███████   | 4548/6470 [56:20<24:25,  1.31it/s]
3|4549|Loss: 0.7190173268318176:  70%|███████   | 4548/6470 [56:21<24:25,  1.31it/s]
3|4549|Loss: 0.7190173268318176:  70%|███████   | 4549/6470 [56:21<23:27,  1.37it/s]
3|4550|Loss: 0.8994437456130981:  70%|███████   | 4549/6470 [56:21<23:27,  1.37it/s]
3|4550|Loss: 0.8994437456130981:  70%|███████   | 4550/6470 [56:22<23:05,  1.39it/s]
3|4551|Loss: 0.5541867017745972:  70%|███████   | 4550/6470 [56:22<23:05,  1.39it/s]
3|4551|Loss: 0.5541867017745972:  70%|███████   | 4551/6470 [56:23<26:21,  1.21it/s]
3|4552|Loss: 0.7898988127708435:  70%|███████   | 4551/6470 [56:23<26:21,  1.21it/s]
3|4552|Loss: 0.7898988127708435:  70%|███████   | 4552/6470 [56:24<28:04,  1.14it/s]
3|4553|Loss: 0.7820512652397156:  70%|███████   | 4552/6470 [56:24<28:04,  1.14it/s]
3|4553|Loss: 0.7820512652397156:  70%|███████   | 4553/6470 [56:25<28:54,  1.10it/s]
3|4554|Loss: 0.714582622051239:  70%|███████   | 4553/6470 [56:25<28:54,  1.10it/s] 
3|4554|Loss: 0.714582622051239:  70%|███████   | 4554/6470 [56:25<28:21,  1.13it/s]
3|4555|Loss: 0.9181835055351257:  70%|███████   | 4554/6470 [56:26<28:21,  1.13it/s]
3|4555|Loss: 0.9181835055351257:  70%|███████   | 4555/6470 [56:26<26:32,  1.20it/s]
3|4556|Loss: 0.6976045966148376:  70%|███████   | 4555/6470 [56:27<26:32,  1.20it/s]
3|4556|Loss: 0.6976045966148376:  70%|███████   | 4556/6470 [56:27<26:57,  1.18it/s]
3|4557|Loss: 0.7768492102622986:  70%|███████   | 4556/6470 [56:27<26:57,  1.18it/s]
3|4557|Loss: 0.7768492102622986:  70%|███████   | 4557/6470 [56:28<24:23,  1.31it/s]
3|4558|Loss: 0.8947943449020386:  70%|███████   | 4557/6470 [56:28<24:23,  1.31it/s]
3|4558|Loss: 0.8947943449020386:  70%|███████   | 4558/6470 [56:28<24:49,  1.28it/s]
3|4559|Loss: 0.7745006680488586:  70%|███████   | 4558/6470 [56:29<24:49,  1.28it/s]
3|4559|Loss: 0.7745006680488586:  70%|███████   | 4559/6470 [56:29<23:17,  1.37it/s]
3|4560|Loss: 0.9107738137245178:  70%|███████   | 4559/6470 [56:29<23:17,  1.37it/s]
3|4560|Loss: 0.9107738137245178:  70%|███████   | 4560/6470 [56:30<22:19,  1.43it/s]
3|4561|Loss: 0.8068788647651672:  70%|███████   | 4560/6470 [56:30<22:19,  1.43it/s]
3|4561|Loss: 0.8068788647651672:  70%|███████   | 4561/6470 [56:30<22:40,  1.40it/s]
3|4562|Loss: 0.8065233826637268:  70%|███████   | 4561/6470 [56:31<22:40,  1.40it/s]
3|4562|Loss: 0.8065233826637268:  71%|███████   | 4562/6470 [56:31<21:30,  1.48it/s]
3|4563|Loss: 0.9769957661628723:  71%|███████   | 4562/6470 [56:31<21:30,  1.48it/s]
3|4563|Loss: 0.9769957661628723:  71%|███████   | 4563/6470 [56:32<21:43,  1.46it/s]
3|4564|Loss: 0.7099093198776245:  71%|███████   | 4563/6470 [56:32<21:43,  1.46it/s]
3|4564|Loss: 0.7099093198776245:  71%|███████   | 4564/6470 [56:32<20:45,  1.53it/s]
3|4565|Loss: 0.7839699983596802:  71%|███████   | 4564/6470 [56:33<20:45,  1.53it/s]
3|4565|Loss: 0.7839699983596802:  71%|███████   | 4565/6470 [56:33<22:31,  1.41it/s]
3|4566|Loss: 0.8729082345962524:  71%|███████   | 4565/6470 [56:34<22:31,  1.41it/s]
3|4566|Loss: 0.8729082345962524:  71%|███████   | 4566/6470 [56:34<22:34,  1.41it/s]
3|4567|Loss: 0.705968976020813:  71%|███████   | 4566/6470 [56:34<22:34,  1.41it/s] 
3|4567|Loss: 0.705968976020813:  71%|███████   | 4567/6470 [56:35<23:01,  1.38it/s]
3|4568|Loss: 0.9133822321891785:  71%|███████   | 4567/6470 [56:35<23:01,  1.38it/s]
3|4568|Loss: 0.9133822321891785:  71%|███████   | 4568/6470 [56:36<24:30,  1.29it/s]
3|4569|Loss: 0.9970746636390686:  71%|███████   | 4568/6470 [56:36<24:30,  1.29it/s]
3|4569|Loss: 0.9970746636390686:  71%|███████   | 4569/6470 [56:36<24:51,  1.27it/s]
3|4570|Loss: 0.7543320655822754:  71%|███████   | 4569/6470 [56:37<24:51,  1.27it/s]
3|4570|Loss: 0.7543320655822754:  71%|███████   | 4570/6470 [56:37<23:57,  1.32it/s]
3|4571|Loss: 0.6638250350952148:  71%|███████   | 4570/6470 [56:37<23:57,  1.32it/s]
3|4571|Loss: 0.6638250350952148:  71%|███████   | 4571/6470 [56:38<23:58,  1.32it/s]
3|4572|Loss: 0.7992199659347534:  71%|███████   | 4571/6470 [56:38<23:58,  1.32it/s]
3|4572|Loss: 0.7992199659347534:  71%|███████   | 4572/6470 [56:39<25:10,  1.26it/s]
3|4573|Loss: 0.7099530100822449:  71%|███████   | 4572/6470 [56:39<25:10,  1.26it/s]
3|4573|Loss: 0.7099530100822449:  71%|███████   | 4573/6470 [56:39<24:45,  1.28it/s]
3|4574|Loss: 0.7686744928359985:  71%|███████   | 4573/6470 [56:40<24:45,  1.28it/s]
3|4574|Loss: 0.7686744928359985:  71%|███████   | 4574/6470 [56:40<24:49,  1.27it/s]
3|4575|Loss: 0.7362058162689209:  71%|███████   | 4574/6470 [56:41<24:49,  1.27it/s]
3|4575|Loss: 0.7362058162689209:  71%|███████   | 4575/6470 [56:41<24:39,  1.28it/s]
3|4576|Loss: 0.8832865357398987:  71%|███████   | 4575/6470 [56:41<24:39,  1.28it/s]
3|4576|Loss: 0.8832865357398987:  71%|███████   | 4576/6470 [56:42<22:46,  1.39it/s]
3|4577|Loss: 0.88336580991745:  71%|███████   | 4576/6470 [56:42<22:46,  1.39it/s]  
3|4577|Loss: 0.88336580991745:  71%|███████   | 4577/6470 [56:42<22:12,  1.42it/s]
3|4578|Loss: 0.8309282660484314:  71%|███████   | 4577/6470 [56:43<22:12,  1.42it/s]
3|4578|Loss: 0.8309282660484314:  71%|███████   | 4578/6470 [56:43<22:59,  1.37it/s]
3|4579|Loss: 0.7128467559814453:  71%|███████   | 4578/6470 [56:43<22:59,  1.37it/s]
3|4579|Loss: 0.7128467559814453:  71%|███████   | 4579/6470 [56:44<22:32,  1.40it/s]
3|4580|Loss: 0.7864481210708618:  71%|███████   | 4579/6470 [56:44<22:32,  1.40it/s]
3|4580|Loss: 0.7864481210708618:  71%|███████   | 4580/6470 [56:44<22:40,  1.39it/s]
3|4581|Loss: 0.7418013215065002:  71%|███████   | 4580/6470 [56:45<22:40,  1.39it/s]
3|4581|Loss: 0.7418013215065002:  71%|███████   | 4581/6470 [56:45<22:23,  1.41it/s]
3|4582|Loss: 0.7595168352127075:  71%|███████   | 4581/6470 [56:45<22:23,  1.41it/s]
3|4582|Loss: 0.7595168352127075:  71%|███████   | 4582/6470 [56:46<22:39,  1.39it/s]
3|4583|Loss: 0.7856764793395996:  71%|███████   | 4582/6470 [56:46<22:39,  1.39it/s]
3|4583|Loss: 0.7856764793395996:  71%|███████   | 4583/6470 [56:47<22:55,  1.37it/s]
3|4584|Loss: 0.8935895562171936:  71%|███████   | 4583/6470 [56:47<22:55,  1.37it/s]
3|4584|Loss: 0.8935895562171936:  71%|███████   | 4584/6470 [56:47<23:16,  1.35it/s]
3|4585|Loss: 0.8279220461845398:  71%|███████   | 4584/6470 [56:48<23:16,  1.35it/s]
3|4585|Loss: 0.8279220461845398:  71%|███████   | 4585/6470 [56:48<22:49,  1.38it/s]
3|4586|Loss: 0.7693068981170654:  71%|███████   | 4585/6470 [56:48<22:49,  1.38it/s]
3|4586|Loss: 0.7693068981170654:  71%|███████   | 4586/6470 [56:49<22:56,  1.37it/s]
3|4587|Loss: 0.8474830389022827:  71%|███████   | 4586/6470 [56:49<22:56,  1.37it/s]
3|4587|Loss: 0.8474830389022827:  71%|███████   | 4587/6470 [56:50<23:14,  1.35it/s]
3|4588|Loss: 0.9167181849479675:  71%|███████   | 4587/6470 [56:50<23:14,  1.35it/s]
3|4588|Loss: 0.9167181849479675:  71%|███████   | 4588/6470 [56:50<22:52,  1.37it/s]
3|4589|Loss: 0.9676624536514282:  71%|███████   | 4588/6470 [56:51<22:52,  1.37it/s]
3|4589|Loss: 0.9676624536514282:  71%|███████   | 4589/6470 [56:51<22:57,  1.37it/s]
3|4590|Loss: 0.9539126753807068:  71%|███████   | 4589/6470 [56:51<22:57,  1.37it/s]
3|4590|Loss: 0.9539126753807068:  71%|███████   | 4590/6470 [56:52<23:27,  1.34it/s]
3|4591|Loss: 0.7963203191757202:  71%|███████   | 4590/6470 [56:52<23:27,  1.34it/s]
3|4591|Loss: 0.7963203191757202:  71%|███████   | 4591/6470 [56:53<24:38,  1.27it/s]
3|4592|Loss: 0.8830628991127014:  71%|███████   | 4591/6470 [56:53<24:38,  1.27it/s]
3|4592|Loss: 0.8830628991127014:  71%|███████   | 4592/6470 [56:53<24:29,  1.28it/s]
3|4593|Loss: 0.8490807414054871:  71%|███████   | 4592/6470 [56:54<24:29,  1.28it/s]
3|4593|Loss: 0.8490807414054871:  71%|███████   | 4593/6470 [56:54<24:24,  1.28it/s]
3|4594|Loss: 0.7364248037338257:  71%|███████   | 4593/6470 [56:55<24:24,  1.28it/s]
3|4594|Loss: 0.7364248037338257:  71%|███████   | 4594/6470 [56:55<25:35,  1.22it/s]
3|4595|Loss: 0.8926328420639038:  71%|███████   | 4594/6470 [56:56<25:35,  1.22it/s]
3|4595|Loss: 0.8926328420639038:  71%|███████   | 4595/6470 [56:56<25:18,  1.23it/s]
3|4596|Loss: 0.669734001159668:  71%|███████   | 4595/6470 [56:56<25:18,  1.23it/s] 
3|4596|Loss: 0.669734001159668:  71%|███████   | 4596/6470 [56:57<26:59,  1.16it/s]
3|4597|Loss: 0.7729151248931885:  71%|███████   | 4596/6470 [56:57<26:59,  1.16it/s]
3|4597|Loss: 0.7729151248931885:  71%|███████   | 4597/6470 [56:58<28:05,  1.11it/s]
3|4598|Loss: 0.7449554204940796:  71%|███████   | 4597/6470 [56:58<28:05,  1.11it/s]
3|4598|Loss: 0.7449554204940796:  71%|███████   | 4598/6470 [56:59<27:14,  1.15it/s]
3|4599|Loss: 0.7474552989006042:  71%|███████   | 4598/6470 [56:59<27:14,  1.15it/s]
3|4599|Loss: 0.7474552989006042:  71%|███████   | 4599/6470 [57:00<26:52,  1.16it/s]
3|4600|Loss: 0.7981056571006775:  71%|███████   | 4599/6470 [57:00<26:52,  1.16it/s]
3|4600|Loss: 0.7981056571006775:  71%|███████   | 4600/6470 [57:00<25:14,  1.23it/s]
3|4601|Loss: 0.8272683024406433:  71%|███████   | 4600/6470 [57:01<25:14,  1.23it/s]
3|4601|Loss: 0.8272683024406433:  71%|███████   | 4601/6470 [57:01<25:19,  1.23it/s]
3|4602|Loss: 0.8475756049156189:  71%|███████   | 4601/6470 [57:01<25:19,  1.23it/s]
3|4602|Loss: 0.8475756049156189:  71%|███████   | 4602/6470 [57:02<25:24,  1.23it/s]
3|4603|Loss: 0.7254598140716553:  71%|███████   | 4602/6470 [57:02<25:24,  1.23it/s]
3|4603|Loss: 0.7254598140716553:  71%|███████   | 4603/6470 [57:03<23:44,  1.31it/s]
3|4604|Loss: 0.9451048374176025:  71%|███████   | 4603/6470 [57:03<23:44,  1.31it/s]
3|4604|Loss: 0.9451048374176025:  71%|███████   | 4604/6470 [57:03<23:14,  1.34it/s]
3|4605|Loss: 0.8467417359352112:  71%|███████   | 4604/6470 [57:04<23:14,  1.34it/s]
3|4605|Loss: 0.8467417359352112:  71%|███████   | 4605/6470 [57:04<22:47,  1.36it/s]
3|4606|Loss: 0.8065773844718933:  71%|███████   | 4605/6470 [57:04<22:47,  1.36it/s]
3|4606|Loss: 0.8065773844718933:  71%|███████   | 4606/6470 [57:05<22:13,  1.40it/s]
3|4607|Loss: 0.932440996170044:  71%|███████   | 4606/6470 [57:05<22:13,  1.40it/s] 
3|4607|Loss: 0.932440996170044:  71%|███████   | 4607/6470 [57:05<22:46,  1.36it/s]
3|4608|Loss: 0.8562484979629517:  71%|███████   | 4607/6470 [57:06<22:46,  1.36it/s]
3|4608|Loss: 0.8562484979629517:  71%|███████   | 4608/6470 [57:06<22:08,  1.40it/s]
3|4609|Loss: 0.8206007480621338:  71%|███████   | 4608/6470 [57:06<22:08,  1.40it/s]
3|4609|Loss: 0.8206007480621338:  71%|███████   | 4609/6470 [57:07<22:12,  1.40it/s]
3|4610|Loss: 0.9179422855377197:  71%|███████   | 4609/6470 [57:07<22:12,  1.40it/s]
3|4610|Loss: 0.9179422855377197:  71%|███████▏  | 4610/6470 [57:07<22:24,  1.38it/s]
3|4611|Loss: 0.8736552596092224:  71%|███████▏  | 4610/6470 [57:08<22:24,  1.38it/s]
3|4611|Loss: 0.8736552596092224:  71%|███████▏  | 4611/6470 [57:08<23:25,  1.32it/s]
3|4612|Loss: 0.706189751625061:  71%|███████▏  | 4611/6470 [57:09<23:25,  1.32it/s] 
3|4612|Loss: 0.706189751625061:  71%|███████▏  | 4612/6470 [57:09<23:51,  1.30it/s]
3|4613|Loss: 0.834422767162323:  71%|███████▏  | 4612/6470 [57:10<23:51,  1.30it/s]
3|4613|Loss: 0.834422767162323:  71%|███████▏  | 4613/6470 [57:10<24:00,  1.29it/s]
3|4614|Loss: 0.9038893580436707:  71%|███████▏  | 4613/6470 [57:10<24:00,  1.29it/s]
3|4614|Loss: 0.9038893580436707:  71%|███████▏  | 4614/6470 [57:11<23:31,  1.31it/s]
3|4615|Loss: 0.7440869212150574:  71%|███████▏  | 4614/6470 [57:11<23:31,  1.31it/s]
3|4615|Loss: 0.7440869212150574:  71%|███████▏  | 4615/6470 [57:11<22:46,  1.36it/s]
3|4616|Loss: 0.9056755900382996:  71%|███████▏  | 4615/6470 [57:12<22:46,  1.36it/s]
3|4616|Loss: 0.9056755900382996:  71%|███████▏  | 4616/6470 [57:12<22:05,  1.40it/s]
3|4617|Loss: 0.8079790472984314:  71%|███████▏  | 4616/6470 [57:12<22:05,  1.40it/s]
3|4617|Loss: 0.8079790472984314:  71%|███████▏  | 4617/6470 [57:13<22:13,  1.39it/s]
3|4618|Loss: 0.9064719080924988:  71%|███████▏  | 4617/6470 [57:13<22:13,  1.39it/s]
3|4618|Loss: 0.9064719080924988:  71%|███████▏  | 4618/6470 [57:14<23:17,  1.33it/s]
3|4619|Loss: 0.8855610489845276:  71%|███████▏  | 4618/6470 [57:14<23:17,  1.33it/s]
3|4619|Loss: 0.8855610489845276:  71%|███████▏  | 4619/6470 [57:14<23:59,  1.29it/s]
3|4620|Loss: 0.6697248816490173:  71%|███████▏  | 4619/6470 [57:15<23:59,  1.29it/s]
3|4620|Loss: 0.6697248816490173:  71%|███████▏  | 4620/6470 [57:15<23:19,  1.32it/s]
3|4621|Loss: 0.7754702568054199:  71%|███████▏  | 4620/6470 [57:15<23:19,  1.32it/s]
3|4621|Loss: 0.7754702568054199:  71%|███████▏  | 4621/6470 [57:16<22:29,  1.37it/s]
3|4622|Loss: 0.8238638043403625:  71%|███████▏  | 4621/6470 [57:16<22:29,  1.37it/s]
3|4622|Loss: 0.8238638043403625:  71%|███████▏  | 4622/6470 [57:16<22:16,  1.38it/s]
3|4623|Loss: 0.7824026346206665:  71%|███████▏  | 4622/6470 [57:17<22:16,  1.38it/s]
3|4623|Loss: 0.7824026346206665:  71%|███████▏  | 4623/6470 [57:17<21:38,  1.42it/s]
3|4624|Loss: 0.9137139916419983:  71%|███████▏  | 4623/6470 [57:18<21:38,  1.42it/s]
3|4624|Loss: 0.9137139916419983:  71%|███████▏  | 4624/6470 [57:18<22:34,  1.36it/s]
3|4625|Loss: 1.0170434713363647:  71%|███████▏  | 4624/6470 [57:18<22:34,  1.36it/s]
3|4625|Loss: 1.0170434713363647:  71%|███████▏  | 4625/6470 [57:19<24:08,  1.27it/s]
3|4626|Loss: 0.9710149168968201:  71%|███████▏  | 4625/6470 [57:19<24:08,  1.27it/s]
3|4626|Loss: 0.9710149168968201:  71%|███████▏  | 4626/6470 [57:20<25:33,  1.20it/s]
3|4627|Loss: 0.7714354395866394:  71%|███████▏  | 4626/6470 [57:20<25:33,  1.20it/s]
3|4627|Loss: 0.7714354395866394:  72%|███████▏  | 4627/6470 [57:21<25:29,  1.20it/s]
3|4628|Loss: 0.8302497863769531:  72%|███████▏  | 4627/6470 [57:21<25:29,  1.20it/s]
3|4628|Loss: 0.8302497863769531:  72%|███████▏  | 4628/6470 [57:21<25:59,  1.18it/s]
3|4629|Loss: 1.0606191158294678:  72%|███████▏  | 4628/6470 [57:22<25:59,  1.18it/s]
3|4629|Loss: 1.0606191158294678:  72%|███████▏  | 4629/6470 [57:22<26:40,  1.15it/s]
3|4630|Loss: 0.6997777223587036:  72%|███████▏  | 4629/6470 [57:23<26:40,  1.15it/s]
3|4630|Loss: 0.6997777223587036:  72%|███████▏  | 4630/6470 [57:23<26:10,  1.17it/s]
3|4631|Loss: 0.7668484449386597:  72%|███████▏  | 4630/6470 [57:24<26:10,  1.17it/s]
3|4631|Loss: 0.7668484449386597:  72%|███████▏  | 4631/6470 [57:24<27:29,  1.11it/s]
3|4632|Loss: 0.8817110061645508:  72%|███████▏  | 4631/6470 [57:25<27:29,  1.11it/s]
3|4632|Loss: 0.8817110061645508:  72%|███████▏  | 4632/6470 [57:25<25:25,  1.20it/s]
3|4633|Loss: 0.9127832055091858:  72%|███████▏  | 4632/6470 [57:25<25:25,  1.20it/s]
3|4633|Loss: 0.9127832055091858:  72%|███████▏  | 4633/6470 [57:26<25:45,  1.19it/s]
3|4634|Loss: 0.8181567192077637:  72%|███████▏  | 4633/6470 [57:26<25:45,  1.19it/s]
3|4634|Loss: 0.8181567192077637:  72%|███████▏  | 4634/6470 [57:26<23:18,  1.31it/s]
3|4635|Loss: 0.835638165473938:  72%|███████▏  | 4634/6470 [57:27<23:18,  1.31it/s] 
3|4635|Loss: 0.835638165473938:  72%|███████▏  | 4635/6470 [57:27<22:10,  1.38it/s]
3|4636|Loss: 0.7421661615371704:  72%|███████▏  | 4635/6470 [57:27<22:10,  1.38it/s]
3|4636|Loss: 0.7421661615371704:  72%|███████▏  | 4636/6470 [57:28<22:03,  1.39it/s]
3|4637|Loss: 0.8310979008674622:  72%|███████▏  | 4636/6470 [57:28<22:03,  1.39it/s]
3|4637|Loss: 0.8310979008674622:  72%|███████▏  | 4637/6470 [57:28<22:14,  1.37it/s]
3|4638|Loss: 0.7967869639396667:  72%|███████▏  | 4637/6470 [57:29<22:14,  1.37it/s]
3|4638|Loss: 0.7967869639396667:  72%|███████▏  | 4638/6470 [57:29<23:57,  1.27it/s]
3|4639|Loss: 0.9001918435096741:  72%|███████▏  | 4638/6470 [57:30<23:57,  1.27it/s]
3|4639|Loss: 0.9001918435096741:  72%|███████▏  | 4639/6470 [57:30<24:33,  1.24it/s]
3|4640|Loss: 0.8499993681907654:  72%|███████▏  | 4639/6470 [57:31<24:33,  1.24it/s]
3|4640|Loss: 0.8499993681907654:  72%|███████▏  | 4640/6470 [57:31<22:50,  1.34it/s]
3|4641|Loss: 0.6370528936386108:  72%|███████▏  | 4640/6470 [57:31<22:50,  1.34it/s]
3|4641|Loss: 0.6370528936386108:  72%|███████▏  | 4641/6470 [57:31<21:05,  1.44it/s]
3|4642|Loss: 0.8567695617675781:  72%|███████▏  | 4641/6470 [57:32<21:05,  1.44it/s]
3|4642|Loss: 0.8567695617675781:  72%|███████▏  | 4642/6470 [57:32<21:33,  1.41it/s]
3|4643|Loss: 0.8519627451896667:  72%|███████▏  | 4642/6470 [57:33<21:33,  1.41it/s]
3|4643|Loss: 0.8519627451896667:  72%|███████▏  | 4643/6470 [57:33<22:23,  1.36it/s]
3|4644|Loss: 0.8505415320396423:  72%|███████▏  | 4643/6470 [57:33<22:23,  1.36it/s]
3|4644|Loss: 0.8505415320396423:  72%|███████▏  | 4644/6470 [57:34<22:09,  1.37it/s]
3|4645|Loss: 0.8379330039024353:  72%|███████▏  | 4644/6470 [57:34<22:09,  1.37it/s]
3|4645|Loss: 0.8379330039024353:  72%|███████▏  | 4645/6470 [57:34<20:11,  1.51it/s]
3|4646|Loss: 0.9718591570854187:  72%|███████▏  | 4645/6470 [57:34<20:11,  1.51it/s]
3|4646|Loss: 0.9718591570854187:  72%|███████▏  | 4646/6470 [57:35<20:04,  1.51it/s]
3|4647|Loss: 0.8772919774055481:  72%|███████▏  | 4646/6470 [57:35<20:04,  1.51it/s]
3|4647|Loss: 0.8772919774055481:  72%|███████▏  | 4647/6470 [57:35<19:55,  1.53it/s]
3|4648|Loss: 0.8346678018569946:  72%|███████▏  | 4647/6470 [57:36<19:55,  1.53it/s]
3|4648|Loss: 0.8346678018569946:  72%|███████▏  | 4648/6470 [57:36<19:52,  1.53it/s]
3|4649|Loss: 0.8655102849006653:  72%|███████▏  | 4648/6470 [57:37<19:52,  1.53it/s]
3|4649|Loss: 0.8655102849006653:  72%|███████▏  | 4649/6470 [57:37<20:54,  1.45it/s]
3|4650|Loss: 0.7351565361022949:  72%|███████▏  | 4649/6470 [57:37<20:54,  1.45it/s]
3|4650|Loss: 0.7351565361022949:  72%|███████▏  | 4650/6470 [57:38<22:53,  1.33it/s]
3|4651|Loss: 0.8643674254417419:  72%|███████▏  | 4650/6470 [57:38<22:53,  1.33it/s]
3|4651|Loss: 0.8643674254417419:  72%|███████▏  | 4651/6470 [57:39<22:40,  1.34it/s]
3|4652|Loss: 0.7060472369194031:  72%|███████▏  | 4651/6470 [57:39<22:40,  1.34it/s]
3|4652|Loss: 0.7060472369194031:  72%|███████▏  | 4652/6470 [57:39<23:08,  1.31it/s]
3|4653|Loss: 0.7184140682220459:  72%|███████▏  | 4652/6470 [57:40<23:08,  1.31it/s]
3|4653|Loss: 0.7184140682220459:  72%|███████▏  | 4653/6470 [57:40<22:13,  1.36it/s]
3|4654|Loss: 0.7342109084129333:  72%|███████▏  | 4653/6470 [57:40<22:13,  1.36it/s]
3|4654|Loss: 0.7342109084129333:  72%|███████▏  | 4654/6470 [57:41<22:20,  1.35it/s]
3|4655|Loss: 0.7933399081230164:  72%|███████▏  | 4654/6470 [57:41<22:20,  1.35it/s]
3|4655|Loss: 0.7933399081230164:  72%|███████▏  | 4655/6470 [57:41<21:50,  1.38it/s]
3|4656|Loss: 0.8986928462982178:  72%|███████▏  | 4655/6470 [57:42<21:50,  1.38it/s]
3|4656|Loss: 0.8986928462982178:  72%|███████▏  | 4656/6470 [57:42<22:12,  1.36it/s]
3|4657|Loss: 0.785262405872345:  72%|███████▏  | 4656/6470 [57:43<22:12,  1.36it/s] 
3|4657|Loss: 0.785262405872345:  72%|███████▏  | 4657/6470 [57:43<22:40,  1.33it/s]
3|4658|Loss: 1.0013426542282104:  72%|███████▏  | 4657/6470 [57:43<22:40,  1.33it/s]
3|4658|Loss: 1.0013426542282104:  72%|███████▏  | 4658/6470 [57:44<23:43,  1.27it/s]
3|4659|Loss: 0.9040639400482178:  72%|███████▏  | 4658/6470 [57:44<23:43,  1.27it/s]
3|4659|Loss: 0.9040639400482178:  72%|███████▏  | 4659/6470 [57:45<24:28,  1.23it/s]
3|4660|Loss: 0.9469994306564331:  72%|███████▏  | 4659/6470 [57:45<24:28,  1.23it/s]
3|4660|Loss: 0.9469994306564331:  72%|███████▏  | 4660/6470 [57:46<25:43,  1.17it/s]
3|4661|Loss: 0.7917205095291138:  72%|███████▏  | 4660/6470 [57:46<25:43,  1.17it/s]
3|4661|Loss: 0.7917205095291138:  72%|███████▏  | 4661/6470 [57:46<25:39,  1.17it/s]
3|4662|Loss: 0.7029942274093628:  72%|███████▏  | 4661/6470 [57:47<25:39,  1.17it/s]
3|4662|Loss: 0.7029942274093628:  72%|███████▏  | 4662/6470 [57:47<23:26,  1.29it/s]
3|4663|Loss: 0.770408570766449:  72%|███████▏  | 4662/6470 [57:47<23:26,  1.29it/s] 
3|4663|Loss: 0.770408570766449:  72%|███████▏  | 4663/6470 [57:48<22:30,  1.34it/s]
3|4664|Loss: 0.721345067024231:  72%|███████▏  | 4663/6470 [57:48<22:30,  1.34it/s]
3|4664|Loss: 0.721345067024231:  72%|███████▏  | 4664/6470 [57:49<22:38,  1.33it/s]
3|4665|Loss: 0.9119390249252319:  72%|███████▏  | 4664/6470 [57:49<22:38,  1.33it/s]
3|4665|Loss: 0.9119390249252319:  72%|███████▏  | 4665/6470 [57:49<22:13,  1.35it/s]
3|4666|Loss: 0.8477479815483093:  72%|███████▏  | 4665/6470 [57:50<22:13,  1.35it/s]
3|4666|Loss: 0.8477479815483093:  72%|███████▏  | 4666/6470 [57:50<22:16,  1.35it/s]
3|4667|Loss: 0.7307791709899902:  72%|███████▏  | 4666/6470 [57:50<22:16,  1.35it/s]
3|4667|Loss: 0.7307791709899902:  72%|███████▏  | 4667/6470 [57:51<22:39,  1.33it/s]
3|4668|Loss: 0.7263737916946411:  72%|███████▏  | 4667/6470 [57:51<22:39,  1.33it/s]
3|4668|Loss: 0.7263737916946411:  72%|███████▏  | 4668/6470 [57:51<22:14,  1.35it/s]
3|4669|Loss: 0.7034395337104797:  72%|███████▏  | 4668/6470 [57:52<22:14,  1.35it/s]
3|4669|Loss: 0.7034395337104797:  72%|███████▏  | 4669/6470 [57:52<21:34,  1.39it/s]
3|4670|Loss: 0.8430225849151611:  72%|███████▏  | 4669/6470 [57:53<21:34,  1.39it/s]
3|4670|Loss: 0.8430225849151611:  72%|███████▏  | 4670/6470 [57:53<23:10,  1.29it/s]
3|4671|Loss: 0.7519770264625549:  72%|███████▏  | 4670/6470 [57:54<23:10,  1.29it/s]
3|4671|Loss: 0.7519770264625549:  72%|███████▏  | 4671/6470 [57:54<25:20,  1.18it/s]
3|4672|Loss: 0.7750234007835388:  72%|███████▏  | 4671/6470 [57:55<25:20,  1.18it/s]
3|4672|Loss: 0.7750234007835388:  72%|███████▏  | 4672/6470 [57:55<25:20,  1.18it/s]
3|4673|Loss: 0.8415212035179138:  72%|███████▏  | 4672/6470 [57:55<25:20,  1.18it/s]
3|4673|Loss: 0.8415212035179138:  72%|███████▏  | 4673/6470 [57:56<25:47,  1.16it/s]
3|4674|Loss: 0.8881101012229919:  72%|███████▏  | 4673/6470 [57:56<25:47,  1.16it/s]
3|4674|Loss: 0.8881101012229919:  72%|███████▏  | 4674/6470 [57:57<25:33,  1.17it/s]
3|4675|Loss: 0.9526190161705017:  72%|███████▏  | 4674/6470 [57:57<25:33,  1.17it/s]
3|4675|Loss: 0.9526190161705017:  72%|███████▏  | 4675/6470 [57:58<25:39,  1.17it/s]
3|4676|Loss: 0.8314779996871948:  72%|███████▏  | 4675/6470 [57:58<25:39,  1.17it/s]
3|4676|Loss: 0.8314779996871948:  72%|███████▏  | 4676/6470 [57:58<24:48,  1.21it/s]
3|4677|Loss: 0.8527228832244873:  72%|███████▏  | 4676/6470 [57:59<24:48,  1.21it/s]
3|4677|Loss: 0.8527228832244873:  72%|███████▏  | 4677/6470 [57:59<22:31,  1.33it/s]
3|4678|Loss: 0.8298143744468689:  72%|███████▏  | 4677/6470 [57:59<22:31,  1.33it/s]
3|4678|Loss: 0.8298143744468689:  72%|███████▏  | 4678/6470 [57:59<21:03,  1.42it/s]
3|4679|Loss: 0.6576195359230042:  72%|███████▏  | 4678/6470 [58:00<21:03,  1.42it/s]
3|4679|Loss: 0.6576195359230042:  72%|███████▏  | 4679/6470 [58:00<22:18,  1.34it/s]
3|4680|Loss: 0.9298780560493469:  72%|███████▏  | 4679/6470 [58:01<22:18,  1.34it/s]
3|4680|Loss: 0.9298780560493469:  72%|███████▏  | 4680/6470 [58:01<23:45,  1.26it/s]
3|4681|Loss: 0.8750561475753784:  72%|███████▏  | 4680/6470 [58:02<23:45,  1.26it/s]
3|4681|Loss: 0.8750561475753784:  72%|███████▏  | 4681/6470 [58:02<22:17,  1.34it/s]
3|4682|Loss: 0.8184239268302917:  72%|███████▏  | 4681/6470 [58:02<22:17,  1.34it/s]
3|4682|Loss: 0.8184239268302917:  72%|███████▏  | 4682/6470 [58:03<21:30,  1.39it/s]
3|4683|Loss: 0.863960862159729:  72%|███████▏  | 4682/6470 [58:03<21:30,  1.39it/s] 
3|4683|Loss: 0.863960862159729:  72%|███████▏  | 4683/6470 [58:03<21:59,  1.35it/s]
3|4684|Loss: 0.9116033315658569:  72%|███████▏  | 4683/6470 [58:04<21:59,  1.35it/s]
3|4684|Loss: 0.9116033315658569:  72%|███████▏  | 4684/6470 [58:04<21:48,  1.36it/s]
3|4685|Loss: 0.9239845871925354:  72%|███████▏  | 4684/6470 [58:04<21:48,  1.36it/s]
3|4685|Loss: 0.9239845871925354:  72%|███████▏  | 4685/6470 [58:05<21:29,  1.38it/s]
3|4686|Loss: 0.8222253322601318:  72%|███████▏  | 4685/6470 [58:05<21:29,  1.38it/s]
3|4686|Loss: 0.8222253322601318:  72%|███████▏  | 4686/6470 [58:05<22:02,  1.35it/s]
3|4687|Loss: 0.9194631576538086:  72%|███████▏  | 4686/6470 [58:06<22:02,  1.35it/s]
3|4687|Loss: 0.9194631576538086:  72%|███████▏  | 4687/6470 [58:06<23:22,  1.27it/s]
3|4688|Loss: 0.8639240264892578:  72%|███████▏  | 4687/6470 [58:07<23:22,  1.27it/s]
3|4688|Loss: 0.8639240264892578:  72%|███████▏  | 4688/6470 [58:07<24:27,  1.21it/s]
3|4689|Loss: 1.0013372898101807:  72%|███████▏  | 4688/6470 [58:08<24:27,  1.21it/s]
3|4689|Loss: 1.0013372898101807:  72%|███████▏  | 4689/6470 [58:08<24:09,  1.23it/s]
3|4690|Loss: 0.8489972352981567:  72%|███████▏  | 4689/6470 [58:08<24:09,  1.23it/s]
3|4690|Loss: 0.8489972352981567:  72%|███████▏  | 4690/6470 [58:09<22:31,  1.32it/s]
3|4691|Loss: 0.908186674118042:  72%|███████▏  | 4690/6470 [58:09<22:31,  1.32it/s] 
3|4691|Loss: 0.908186674118042:  73%|███████▎  | 4691/6470 [58:10<23:08,  1.28it/s]
3|4692|Loss: 0.777560293674469:  73%|███████▎  | 4691/6470 [58:10<23:08,  1.28it/s]
3|4692|Loss: 0.777560293674469:  73%|███████▎  | 4692/6470 [58:10<22:38,  1.31it/s]
3|4693|Loss: 0.7581313252449036:  73%|███████▎  | 4692/6470 [58:11<22:38,  1.31it/s]
3|4693|Loss: 0.7581313252449036:  73%|███████▎  | 4693/6470 [58:11<21:06,  1.40it/s]
3|4694|Loss: 0.8975030183792114:  73%|███████▎  | 4693/6470 [58:11<21:06,  1.40it/s]
3|4694|Loss: 0.8975030183792114:  73%|███████▎  | 4694/6470 [58:12<20:42,  1.43it/s]
3|4695|Loss: 0.7568855285644531:  73%|███████▎  | 4694/6470 [58:12<20:42,  1.43it/s]
3|4695|Loss: 0.7568855285644531:  73%|███████▎  | 4695/6470 [58:12<19:22,  1.53it/s]
3|4696|Loss: 0.8194880485534668:  73%|███████▎  | 4695/6470 [58:12<19:22,  1.53it/s]
3|4696|Loss: 0.8194880485534668:  73%|███████▎  | 4696/6470 [58:13<20:49,  1.42it/s]
3|4697|Loss: 0.8215668797492981:  73%|███████▎  | 4696/6470 [58:13<20:49,  1.42it/s]
3|4697|Loss: 0.8215668797492981:  73%|███████▎  | 4697/6470 [58:14<21:25,  1.38it/s]
3|4698|Loss: 0.8875948786735535:  73%|███████▎  | 4697/6470 [58:14<21:25,  1.38it/s]
3|4698|Loss: 0.8875948786735535:  73%|███████▎  | 4698/6470 [58:14<21:35,  1.37it/s]
3|4699|Loss: 0.7894717454910278:  73%|███████▎  | 4698/6470 [58:15<21:35,  1.37it/s]
3|4699|Loss: 0.7894717454910278:  73%|███████▎  | 4699/6470 [58:15<22:56,  1.29it/s]
3|4700|Loss: 0.7999436259269714:  73%|███████▎  | 4699/6470 [58:16<22:56,  1.29it/s]
3|4700|Loss: 0.7999436259269714:  73%|███████▎  | 4700/6470 [58:16<23:23,  1.26it/s]
3|4701|Loss: 0.7571582794189453:  73%|███████▎  | 4700/6470 [58:17<23:23,  1.26it/s]
3|4701|Loss: 0.7571582794189453:  73%|███████▎  | 4701/6470 [58:17<24:04,  1.22it/s]
3|4702|Loss: 0.8229376077651978:  73%|███████▎  | 4701/6470 [58:17<24:04,  1.22it/s]
3|4702|Loss: 0.8229376077651978:  73%|███████▎  | 4702/6470 [58:18<23:58,  1.23it/s]
3|4703|Loss: 0.9160692095756531:  73%|███████▎  | 4702/6470 [58:18<23:58,  1.23it/s]
3|4703|Loss: 0.9160692095756531:  73%|███████▎  | 4703/6470 [58:19<22:56,  1.28it/s]
3|4704|Loss: 0.8663207292556763:  73%|███████▎  | 4703/6470 [58:19<22:56,  1.28it/s]
3|4704|Loss: 0.8663207292556763:  73%|███████▎  | 4704/6470 [58:19<20:50,  1.41it/s]
3|4705|Loss: 0.8361754417419434:  73%|███████▎  | 4704/6470 [58:19<20:50,  1.41it/s]
3|4705|Loss: 0.8361754417419434:  73%|███████▎  | 4705/6470 [58:20<19:16,  1.53it/s]
3|4706|Loss: 0.8161665797233582:  73%|███████▎  | 4705/6470 [58:20<19:16,  1.53it/s]
3|4706|Loss: 0.8161665797233582:  73%|███████▎  | 4706/6470 [58:20<20:04,  1.46it/s]
3|4707|Loss: 0.9489923715591431:  73%|███████▎  | 4706/6470 [58:21<20:04,  1.46it/s]
3|4707|Loss: 0.9489923715591431:  73%|███████▎  | 4707/6470 [58:21<20:52,  1.41it/s]
3|4708|Loss: 0.869597315788269:  73%|███████▎  | 4707/6470 [58:21<20:52,  1.41it/s] 
3|4708|Loss: 0.869597315788269:  73%|███████▎  | 4708/6470 [58:22<20:37,  1.42it/s]
3|4709|Loss: 0.8079859018325806:  73%|███████▎  | 4708/6470 [58:22<20:37,  1.42it/s]
3|4709|Loss: 0.8079859018325806:  73%|███████▎  | 4709/6470 [58:23<21:10,  1.39it/s]
3|4710|Loss: 0.7487388253211975:  73%|███████▎  | 4709/6470 [58:23<21:10,  1.39it/s]
3|4710|Loss: 0.7487388253211975:  73%|███████▎  | 4710/6470 [58:23<21:31,  1.36it/s]
3|4711|Loss: 0.842374861240387:  73%|███████▎  | 4710/6470 [58:24<21:31,  1.36it/s] 
3|4711|Loss: 0.842374861240387:  73%|███████▎  | 4711/6470 [58:24<21:23,  1.37it/s]
3|4712|Loss: 0.8474577069282532:  73%|███████▎  | 4711/6470 [58:24<21:23,  1.37it/s]
3|4712|Loss: 0.8474577069282532:  73%|███████▎  | 4712/6470 [58:25<20:53,  1.40it/s]
3|4713|Loss: 0.8675378561019897:  73%|███████▎  | 4712/6470 [58:25<20:53,  1.40it/s]
3|4713|Loss: 0.8675378561019897:  73%|███████▎  | 4713/6470 [58:25<19:04,  1.54it/s]
3|4714|Loss: 0.8682688474655151:  73%|███████▎  | 4713/6470 [58:26<19:04,  1.54it/s]
3|4714|Loss: 0.8682688474655151:  73%|███████▎  | 4714/6470 [58:26<18:35,  1.57it/s]
3|4715|Loss: 0.9770579934120178:  73%|███████▎  | 4714/6470 [58:26<18:35,  1.57it/s]
3|4715|Loss: 0.9770579934120178:  73%|███████▎  | 4715/6470 [58:27<20:26,  1.43it/s]
3|4716|Loss: 0.865267276763916:  73%|███████▎  | 4715/6470 [58:27<20:26,  1.43it/s] 
3|4716|Loss: 0.865267276763916:  73%|███████▎  | 4716/6470 [58:28<22:16,  1.31it/s]
3|4717|Loss: 0.8326385617256165:  73%|███████▎  | 4716/6470 [58:28<22:16,  1.31it/s]
3|4717|Loss: 0.8326385617256165:  73%|███████▎  | 4717/6470 [58:28<22:43,  1.29it/s]
3|4718|Loss: 0.8627620339393616:  73%|███████▎  | 4717/6470 [58:29<22:43,  1.29it/s]
3|4718|Loss: 0.8627620339393616:  73%|███████▎  | 4718/6470 [58:29<22:20,  1.31it/s]
3|4719|Loss: 0.8368387818336487:  73%|███████▎  | 4718/6470 [58:30<22:20,  1.31it/s]
3|4719|Loss: 0.8368387818336487:  73%|███████▎  | 4719/6470 [58:30<22:21,  1.31it/s]
3|4720|Loss: 0.7255232930183411:  73%|███████▎  | 4719/6470 [58:30<22:21,  1.31it/s]
3|4720|Loss: 0.7255232930183411:  73%|███████▎  | 4720/6470 [58:31<21:46,  1.34it/s]
3|4721|Loss: 0.8107850551605225:  73%|███████▎  | 4720/6470 [58:31<21:46,  1.34it/s]
3|4721|Loss: 0.8107850551605225:  73%|███████▎  | 4721/6470 [58:31<20:53,  1.40it/s]
3|4722|Loss: 0.856322169303894:  73%|███████▎  | 4721/6470 [58:32<20:53,  1.40it/s] 
3|4722|Loss: 0.856322169303894:  73%|███████▎  | 4722/6470 [58:32<20:18,  1.43it/s]
3|4723|Loss: 0.7593442797660828:  73%|███████▎  | 4722/6470 [58:32<20:18,  1.43it/s]
3|4723|Loss: 0.7593442797660828:  73%|███████▎  | 4723/6470 [58:33<20:09,  1.44it/s]
3|4724|Loss: 0.9226787090301514:  73%|███████▎  | 4723/6470 [58:33<20:09,  1.44it/s]
3|4724|Loss: 0.9226787090301514:  73%|███████▎  | 4724/6470 [58:33<20:14,  1.44it/s]
3|4725|Loss: 0.8385482430458069:  73%|███████▎  | 4724/6470 [58:34<20:14,  1.44it/s]
3|4725|Loss: 0.8385482430458069:  73%|███████▎  | 4725/6470 [58:34<21:26,  1.36it/s]
3|4726|Loss: 0.7573462128639221:  73%|███████▎  | 4725/6470 [58:35<21:26,  1.36it/s]
3|4726|Loss: 0.7573462128639221:  73%|███████▎  | 4726/6470 [58:35<21:45,  1.34it/s]
3|4727|Loss: 0.8367086052894592:  73%|███████▎  | 4726/6470 [58:35<21:45,  1.34it/s]
3|4727|Loss: 0.8367086052894592:  73%|███████▎  | 4727/6470 [58:36<21:59,  1.32it/s]
3|4728|Loss: 0.9874333739280701:  73%|███████▎  | 4727/6470 [58:36<21:59,  1.32it/s]
3|4728|Loss: 0.9874333739280701:  73%|███████▎  | 4728/6470 [58:37<23:06,  1.26it/s]
3|4729|Loss: 0.935046911239624:  73%|███████▎  | 4728/6470 [58:37<23:06,  1.26it/s] 
3|4729|Loss: 0.935046911239624:  73%|███████▎  | 4729/6470 [58:37<22:21,  1.30it/s]
3|4730|Loss: 0.8772590160369873:  73%|███████▎  | 4729/6470 [58:38<22:21,  1.30it/s]
3|4730|Loss: 0.8772590160369873:  73%|███████▎  | 4730/6470 [58:38<22:04,  1.31it/s]
3|4731|Loss: 0.7301701307296753:  73%|███████▎  | 4730/6470 [58:38<22:04,  1.31it/s]
3|4731|Loss: 0.7301701307296753:  73%|███████▎  | 4731/6470 [58:39<22:14,  1.30it/s]
3|4732|Loss: 1.0356647968292236:  73%|███████▎  | 4731/6470 [58:39<22:14,  1.30it/s]
3|4732|Loss: 1.0356647968292236:  73%|███████▎  | 4732/6470 [58:40<22:28,  1.29it/s]
3|4733|Loss: 0.7615347504615784:  73%|███████▎  | 4732/6470 [58:40<22:28,  1.29it/s]
3|4733|Loss: 0.7615347504615784:  73%|███████▎  | 4733/6470 [58:40<22:24,  1.29it/s]
3|4734|Loss: 0.7569650411605835:  73%|███████▎  | 4733/6470 [58:41<22:24,  1.29it/s]
3|4734|Loss: 0.7569650411605835:  73%|███████▎  | 4734/6470 [58:41<22:30,  1.29it/s]
3|4735|Loss: 0.8807353973388672:  73%|███████▎  | 4734/6470 [58:42<22:30,  1.29it/s]
3|4735|Loss: 0.8807353973388672:  73%|███████▎  | 4735/6470 [58:42<21:56,  1.32it/s]
3|4736|Loss: 0.7715557217597961:  73%|███████▎  | 4735/6470 [58:42<21:56,  1.32it/s]
3|4736|Loss: 0.7715557217597961:  73%|███████▎  | 4736/6470 [58:43<21:07,  1.37it/s]
3|4737|Loss: 0.7023119926452637:  73%|███████▎  | 4736/6470 [58:43<21:07,  1.37it/s]
3|4737|Loss: 0.7023119926452637:  73%|███████▎  | 4737/6470 [58:43<20:45,  1.39it/s]
3|4738|Loss: 0.7683531641960144:  73%|███████▎  | 4737/6470 [58:44<20:45,  1.39it/s]
3|4738|Loss: 0.7683531641960144:  73%|███████▎  | 4738/6470 [58:44<20:41,  1.40it/s]
3|4739|Loss: 0.9214838743209839:  73%|███████▎  | 4738/6470 [58:44<20:41,  1.40it/s]
3|4739|Loss: 0.9214838743209839:  73%|███████▎  | 4739/6470 [58:45<21:01,  1.37it/s]
3|4740|Loss: 0.6554399132728577:  73%|███████▎  | 4739/6470 [58:45<21:01,  1.37it/s]
3|4740|Loss: 0.6554399132728577:  73%|███████▎  | 4740/6470 [58:46<21:53,  1.32it/s]
3|4741|Loss: 0.8299382328987122:  73%|███████▎  | 4740/6470 [58:46<21:53,  1.32it/s]
3|4741|Loss: 0.8299382328987122:  73%|███████▎  | 4741/6470 [58:46<20:08,  1.43it/s]
3|4742|Loss: 0.8588670492172241:  73%|███████▎  | 4741/6470 [58:46<20:08,  1.43it/s]
3|4742|Loss: 0.8588670492172241:  73%|███████▎  | 4742/6470 [58:47<19:26,  1.48it/s]
3|4743|Loss: 0.662437915802002:  73%|███████▎  | 4742/6470 [58:47<19:26,  1.48it/s] 
3|4743|Loss: 0.662437915802002:  73%|███████▎  | 4743/6470 [58:47<18:49,  1.53it/s]
3|4744|Loss: 0.760066032409668:  73%|███████▎  | 4743/6470 [58:48<18:49,  1.53it/s]
3|4744|Loss: 0.760066032409668:  73%|███████▎  | 4744/6470 [58:48<17:42,  1.62it/s]
3|4745|Loss: 0.8303321599960327:  73%|███████▎  | 4744/6470 [58:48<17:42,  1.62it/s]
3|4745|Loss: 0.8303321599960327:  73%|███████▎  | 4745/6470 [58:49<19:38,  1.46it/s]
3|4746|Loss: 0.7959373593330383:  73%|███████▎  | 4745/6470 [58:49<19:38,  1.46it/s]
3|4746|Loss: 0.7959373593330383:  73%|███████▎  | 4746/6470 [58:49<20:52,  1.38it/s]
3|4747|Loss: 0.8608574271202087:  73%|███████▎  | 4746/6470 [58:50<20:52,  1.38it/s]
3|4747|Loss: 0.8608574271202087:  73%|███████▎  | 4747/6470 [58:50<20:49,  1.38it/s]
3|4748|Loss: 0.7840784192085266:  73%|███████▎  | 4747/6470 [58:51<20:49,  1.38it/s]
3|4748|Loss: 0.7840784192085266:  73%|███████▎  | 4748/6470 [58:51<20:23,  1.41it/s]
3|4749|Loss: 0.8350683450698853:  73%|███████▎  | 4748/6470 [58:51<20:23,  1.41it/s]
3|4749|Loss: 0.8350683450698853:  73%|███████▎  | 4749/6470 [58:52<20:08,  1.42it/s]
3|4750|Loss: 0.8347342610359192:  73%|███████▎  | 4749/6470 [58:52<20:08,  1.42it/s]
3|4750|Loss: 0.8347342610359192:  73%|███████▎  | 4750/6470 [58:52<21:36,  1.33it/s]
3|4751|Loss: 0.8621341586112976:  73%|███████▎  | 4750/6470 [58:53<21:36,  1.33it/s]
3|4751|Loss: 0.8621341586112976:  73%|███████▎  | 4751/6470 [58:53<21:27,  1.34it/s]
3|4752|Loss: 0.8279690146446228:  73%|███████▎  | 4751/6470 [58:54<21:27,  1.34it/s]
3|4752|Loss: 0.8279690146446228:  73%|███████▎  | 4752/6470 [58:54<21:41,  1.32it/s]
3|4753|Loss: 0.9233366847038269:  73%|███████▎  | 4752/6470 [58:54<21:41,  1.32it/s]
3|4753|Loss: 0.9233366847038269:  73%|███████▎  | 4753/6470 [58:55<22:01,  1.30it/s]
3|4754|Loss: 0.8160166144371033:  73%|███████▎  | 4753/6470 [58:55<22:01,  1.30it/s]
3|4754|Loss: 0.8160166144371033:  73%|███████▎  | 4754/6470 [58:55<21:38,  1.32it/s]
3|4755|Loss: 0.7394033670425415:  73%|███████▎  | 4754/6470 [58:56<21:38,  1.32it/s]
3|4755|Loss: 0.7394033670425415:  73%|███████▎  | 4755/6470 [58:56<22:19,  1.28it/s]
3|4756|Loss: 0.9399880766868591:  73%|███████▎  | 4755/6470 [58:57<22:19,  1.28it/s]
3|4756|Loss: 0.9399880766868591:  74%|███████▎  | 4756/6470 [58:57<22:23,  1.28it/s]
3|4757|Loss: 0.8085432052612305:  74%|███████▎  | 4756/6470 [58:58<22:23,  1.28it/s]
3|4757|Loss: 0.8085432052612305:  74%|███████▎  | 4757/6470 [58:58<21:50,  1.31it/s]
3|4758|Loss: 0.9695202708244324:  74%|███████▎  | 4757/6470 [58:58<21:50,  1.31it/s]
3|4758|Loss: 0.9695202708244324:  74%|███████▎  | 4758/6470 [58:59<21:39,  1.32it/s]
3|4759|Loss: 0.9052133560180664:  74%|███████▎  | 4758/6470 [58:59<21:39,  1.32it/s]
3|4759|Loss: 0.9052133560180664:  74%|███████▎  | 4759/6470 [58:59<19:38,  1.45it/s]
3|4760|Loss: 0.7045618891716003:  74%|███████▎  | 4759/6470 [58:59<19:38,  1.45it/s]
3|4760|Loss: 0.7045618891716003:  74%|███████▎  | 4760/6470 [59:00<19:02,  1.50it/s]
3|4761|Loss: 0.7902297377586365:  74%|███████▎  | 4760/6470 [59:00<19:02,  1.50it/s]
3|4761|Loss: 0.7902297377586365:  74%|███████▎  | 4761/6470 [59:01<20:52,  1.36it/s]
3|4762|Loss: 0.7568074464797974:  74%|███████▎  | 4761/6470 [59:01<20:52,  1.36it/s]
3|4762|Loss: 0.7568074464797974:  74%|███████▎  | 4762/6470 [59:02<22:33,  1.26it/s]
3|4763|Loss: 0.8991551399230957:  74%|███████▎  | 4762/6470 [59:02<22:33,  1.26it/s]
3|4763|Loss: 0.8991551399230957:  74%|███████▎  | 4763/6470 [59:02<23:45,  1.20it/s]
3|4764|Loss: 0.7203948497772217:  74%|███████▎  | 4763/6470 [59:03<23:45,  1.20it/s]
3|4764|Loss: 0.7203948497772217:  74%|███████▎  | 4764/6470 [59:03<23:05,  1.23it/s]
3|4765|Loss: 0.9585244655609131:  74%|███████▎  | 4764/6470 [59:04<23:05,  1.23it/s]
3|4765|Loss: 0.9585244655609131:  74%|███████▎  | 4765/6470 [59:04<22:32,  1.26it/s]
3|4766|Loss: 0.7475852966308594:  74%|███████▎  | 4765/6470 [59:04<22:32,  1.26it/s]
3|4766|Loss: 0.7475852966308594:  74%|███████▎  | 4766/6470 [59:05<21:49,  1.30it/s]
3|4767|Loss: 0.8567349910736084:  74%|███████▎  | 4766/6470 [59:05<21:49,  1.30it/s]
3|4767|Loss: 0.8567349910736084:  74%|███████▎  | 4767/6470 [59:05<21:48,  1.30it/s]
3|4768|Loss: 0.654294490814209:  74%|███████▎  | 4767/6470 [59:06<21:48,  1.30it/s] 
3|4768|Loss: 0.654294490814209:  74%|███████▎  | 4768/6470 [59:06<21:09,  1.34it/s]
3|4769|Loss: 0.8098558783531189:  74%|███████▎  | 4768/6470 [59:07<21:09,  1.34it/s]
3|4769|Loss: 0.8098558783531189:  74%|███████▎  | 4769/6470 [59:07<20:15,  1.40it/s]
3|4770|Loss: 0.8236626982688904:  74%|███████▎  | 4769/6470 [59:07<20:15,  1.40it/s]
3|4770|Loss: 0.8236626982688904:  74%|███████▎  | 4770/6470 [59:08<21:20,  1.33it/s]
3|4771|Loss: 0.8370973467826843:  74%|███████▎  | 4770/6470 [59:08<21:20,  1.33it/s]
3|4771|Loss: 0.8370973467826843:  74%|███████▎  | 4771/6470 [59:08<20:54,  1.35it/s]
3|4772|Loss: 1.0157784223556519:  74%|███████▎  | 4771/6470 [59:09<20:54,  1.35it/s]
3|4772|Loss: 1.0157784223556519:  74%|███████▍  | 4772/6470 [59:09<19:46,  1.43it/s]
3|4773|Loss: 0.7170299887657166:  74%|███████▍  | 4772/6470 [59:09<19:46,  1.43it/s]
3|4773|Loss: 0.7170299887657166:  74%|███████▍  | 4773/6470 [59:10<20:06,  1.41it/s]
3|4774|Loss: 0.9078963994979858:  74%|███████▍  | 4773/6470 [59:10<20:06,  1.41it/s]
3|4774|Loss: 0.9078963994979858:  74%|███████▍  | 4774/6470 [59:10<19:32,  1.45it/s]
3|4775|Loss: 0.9969034194946289:  74%|███████▍  | 4774/6470 [59:11<19:32,  1.45it/s]
3|4775|Loss: 0.9969034194946289:  74%|███████▍  | 4775/6470 [59:11<19:27,  1.45it/s]
3|4776|Loss: 0.8944389224052429:  74%|███████▍  | 4775/6470 [59:11<19:27,  1.45it/s]
3|4776|Loss: 0.8944389224052429:  74%|███████▍  | 4776/6470 [59:12<19:31,  1.45it/s]
3|4777|Loss: 0.7889674305915833:  74%|███████▍  | 4776/6470 [59:12<19:31,  1.45it/s]
3|4777|Loss: 0.7889674305915833:  74%|███████▍  | 4777/6470 [59:12<19:25,  1.45it/s]
3|4778|Loss: 0.7759350538253784:  74%|███████▍  | 4777/6470 [59:13<19:25,  1.45it/s]
3|4778|Loss: 0.7759350538253784:  74%|███████▍  | 4778/6470 [59:13<19:15,  1.46it/s]
3|4779|Loss: 0.7290279269218445:  74%|███████▍  | 4778/6470 [59:14<19:15,  1.46it/s]
3|4779|Loss: 0.7290279269218445:  74%|███████▍  | 4779/6470 [59:14<21:42,  1.30it/s]
3|4780|Loss: 0.8899943828582764:  74%|███████▍  | 4779/6470 [59:15<21:42,  1.30it/s]
3|4780|Loss: 0.8899943828582764:  74%|███████▍  | 4780/6470 [59:15<22:03,  1.28it/s]
3|4781|Loss: 0.8219268918037415:  74%|███████▍  | 4780/6470 [59:15<22:03,  1.28it/s]
3|4781|Loss: 0.8219268918037415:  74%|███████▍  | 4781/6470 [59:16<23:44,  1.19it/s]
3|4782|Loss: 0.8734778165817261:  74%|███████▍  | 4781/6470 [59:16<23:44,  1.19it/s]
3|4782|Loss: 0.8734778165817261:  74%|███████▍  | 4782/6470 [59:17<23:41,  1.19it/s]
3|4783|Loss: 0.7184233069419861:  74%|███████▍  | 4782/6470 [59:17<23:41,  1.19it/s]
3|4783|Loss: 0.7184233069419861:  74%|███████▍  | 4783/6470 [59:18<24:24,  1.15it/s]
3|4784|Loss: 0.7622318863868713:  74%|███████▍  | 4783/6470 [59:18<24:24,  1.15it/s]
3|4784|Loss: 0.7622318863868713:  74%|███████▍  | 4784/6470 [59:18<24:02,  1.17it/s]
3|4785|Loss: 0.8294090032577515:  74%|███████▍  | 4784/6470 [59:19<24:02,  1.17it/s]
3|4785|Loss: 0.8294090032577515:  74%|███████▍  | 4785/6470 [59:19<22:43,  1.24it/s]
3|4786|Loss: 0.9787877202033997:  74%|███████▍  | 4785/6470 [59:19<22:43,  1.24it/s]
3|4786|Loss: 0.9787877202033997:  74%|███████▍  | 4786/6470 [59:20<20:38,  1.36it/s]
3|4787|Loss: 0.8261973857879639:  74%|███████▍  | 4786/6470 [59:20<20:38,  1.36it/s]
3|4787|Loss: 0.8261973857879639:  74%|███████▍  | 4787/6470 [59:20<20:43,  1.35it/s]
3|4788|Loss: 0.790867269039154:  74%|███████▍  | 4787/6470 [59:21<20:43,  1.35it/s] 
3|4788|Loss: 0.790867269039154:  74%|███████▍  | 4788/6470 [59:21<20:55,  1.34it/s]
3|4789|Loss: 0.8090776205062866:  74%|███████▍  | 4788/6470 [59:22<20:55,  1.34it/s]
3|4789|Loss: 0.8090776205062866:  74%|███████▍  | 4789/6470 [59:22<21:00,  1.33it/s]
3|4790|Loss: 0.9315263032913208:  74%|███████▍  | 4789/6470 [59:22<21:00,  1.33it/s]
3|4790|Loss: 0.9315263032913208:  74%|███████▍  | 4790/6470 [59:23<21:19,  1.31it/s]
3|4791|Loss: 0.7981347441673279:  74%|███████▍  | 4790/6470 [59:23<21:19,  1.31it/s]
3|4791|Loss: 0.7981347441673279:  74%|███████▍  | 4791/6470 [59:23<20:16,  1.38it/s]
3|4792|Loss: 1.1107213497161865:  74%|███████▍  | 4791/6470 [59:24<20:16,  1.38it/s]
3|4792|Loss: 1.1107213497161865:  74%|███████▍  | 4792/6470 [59:24<20:23,  1.37it/s]
3|4793|Loss: 0.8545129895210266:  74%|███████▍  | 4792/6470 [59:25<20:23,  1.37it/s]
3|4793|Loss: 0.8545129895210266:  74%|███████▍  | 4793/6470 [59:25<19:42,  1.42it/s]
3|4794|Loss: 0.914447009563446:  74%|███████▍  | 4793/6470 [59:25<19:42,  1.42it/s] 
3|4794|Loss: 0.914447009563446:  74%|███████▍  | 4794/6470 [59:26<20:14,  1.38it/s]
3|4795|Loss: 0.8367061614990234:  74%|███████▍  | 4794/6470 [59:26<20:14,  1.38it/s]
3|4795|Loss: 0.8367061614990234:  74%|███████▍  | 4795/6470 [59:26<20:48,  1.34it/s]
3|4796|Loss: 0.8538472652435303:  74%|███████▍  | 4795/6470 [59:27<20:48,  1.34it/s]
3|4796|Loss: 0.8538472652435303:  74%|███████▍  | 4796/6470 [59:27<21:44,  1.28it/s]
3|4797|Loss: 0.8657180666923523:  74%|███████▍  | 4796/6470 [59:28<21:44,  1.28it/s]
3|4797|Loss: 0.8657180666923523:  74%|███████▍  | 4797/6470 [59:28<22:41,  1.23it/s]
3|4798|Loss: 0.8617190718650818:  74%|███████▍  | 4797/6470 [59:28<22:41,  1.23it/s]
3|4798|Loss: 0.8617190718650818:  74%|███████▍  | 4798/6470 [59:29<21:02,  1.32it/s]
3|4799|Loss: 1.0136574506759644:  74%|███████▍  | 4798/6470 [59:29<21:02,  1.32it/s]
3|4799|Loss: 1.0136574506759644:  74%|███████▍  | 4799/6470 [59:29<20:47,  1.34it/s]
3|4800|Loss: 0.8425106406211853:  74%|███████▍  | 4799/6470 [59:30<20:47,  1.34it/s]
3|4800|Loss: 0.8425106406211853:  74%|███████▍  | 4800/6470 [59:30<20:11,  1.38it/s]
3|4801|Loss: 0.7348291277885437:  74%|███████▍  | 4800/6470 [59:31<20:11,  1.38it/s]
3|4801|Loss: 0.7348291277885437:  74%|███████▍  | 4801/6470 [59:31<20:54,  1.33it/s]
3|4802|Loss: 0.7723482847213745:  74%|███████▍  | 4801/6470 [59:31<20:54,  1.33it/s]
3|4802|Loss: 0.7723482847213745:  74%|███████▍  | 4802/6470 [59:31<19:11,  1.45it/s]
3|4803|Loss: 0.8204768300056458:  74%|███████▍  | 4802/6470 [59:32<19:11,  1.45it/s]
3|4803|Loss: 0.8204768300056458:  74%|███████▍  | 4803/6470 [59:32<20:44,  1.34it/s]
3|4804|Loss: 0.8917028307914734:  74%|███████▍  | 4803/6470 [59:33<20:44,  1.34it/s]
3|4804|Loss: 0.8917028307914734:  74%|███████▍  | 4804/6470 [59:33<20:50,  1.33it/s]
3|4805|Loss: 0.8733187317848206:  74%|███████▍  | 4804/6470 [59:34<20:50,  1.33it/s]
3|4805|Loss: 0.8733187317848206:  74%|███████▍  | 4805/6470 [59:34<20:29,  1.35it/s]
3|4806|Loss: 0.9012233018875122:  74%|███████▍  | 4805/6470 [59:34<20:29,  1.35it/s]
3|4806|Loss: 0.9012233018875122:  74%|███████▍  | 4806/6470 [59:35<20:39,  1.34it/s]
3|4807|Loss: 0.8693943619728088:  74%|███████▍  | 4806/6470 [59:35<20:39,  1.34it/s]
3|4807|Loss: 0.8693943619728088:  74%|███████▍  | 4807/6470 [59:35<20:14,  1.37it/s]
3|4808|Loss: 0.8232657313346863:  74%|███████▍  | 4807/6470 [59:36<20:14,  1.37it/s]
3|4808|Loss: 0.8232657313346863:  74%|███████▍  | 4808/6470 [59:36<18:50,  1.47it/s]
3|4809|Loss: 0.7568039298057556:  74%|███████▍  | 4808/6470 [59:36<18:50,  1.47it/s]
3|4809|Loss: 0.7568039298057556:  74%|███████▍  | 4809/6470 [59:37<19:07,  1.45it/s]
3|4810|Loss: 0.6787762641906738:  74%|███████▍  | 4809/6470 [59:37<19:07,  1.45it/s]
3|4810|Loss: 0.6787762641906738:  74%|███████▍  | 4810/6470 [59:37<18:20,  1.51it/s]
3|4811|Loss: 0.8811497092247009:  74%|███████▍  | 4810/6470 [59:37<18:20,  1.51it/s]
3|4811|Loss: 0.8811497092247009:  74%|███████▍  | 4811/6470 [59:38<17:37,  1.57it/s]
3|4812|Loss: 0.7341183423995972:  74%|███████▍  | 4811/6470 [59:38<17:37,  1.57it/s]
3|4812|Loss: 0.7341183423995972:  74%|███████▍  | 4812/6470 [59:38<18:05,  1.53it/s]
3|4813|Loss: 0.8049237132072449:  74%|███████▍  | 4812/6470 [59:39<18:05,  1.53it/s]
3|4813|Loss: 0.8049237132072449:  74%|███████▍  | 4813/6470 [59:39<19:52,  1.39it/s]
3|4814|Loss: 0.7282190322875977:  74%|███████▍  | 4813/6470 [59:40<19:52,  1.39it/s]
3|4814|Loss: 0.7282190322875977:  74%|███████▍  | 4814/6470 [59:40<20:03,  1.38it/s]
3|4815|Loss: 0.7113909125328064:  74%|███████▍  | 4814/6470 [59:40<20:03,  1.38it/s]
3|4815|Loss: 0.7113909125328064:  74%|███████▍  | 4815/6470 [59:41<19:47,  1.39it/s]
3|4816|Loss: 0.7105551958084106:  74%|███████▍  | 4815/6470 [59:41<19:47,  1.39it/s]
3|4816|Loss: 0.7105551958084106:  74%|███████▍  | 4816/6470 [59:42<20:41,  1.33it/s]
3|4817|Loss: 0.6235085725784302:  74%|███████▍  | 4816/6470 [59:42<20:41,  1.33it/s]
3|4817|Loss: 0.6235085725784302:  74%|███████▍  | 4817/6470 [59:42<19:38,  1.40it/s]
3|4818|Loss: 0.812441349029541:  74%|███████▍  | 4817/6470 [59:43<19:38,  1.40it/s] 
3|4818|Loss: 0.812441349029541:  74%|███████▍  | 4818/6470 [59:43<19:38,  1.40it/s]
3|4819|Loss: 0.8885992765426636:  74%|███████▍  | 4818/6470 [59:43<19:38,  1.40it/s]
3|4819|Loss: 0.8885992765426636:  74%|███████▍  | 4819/6470 [59:44<19:27,  1.41it/s]
3|4820|Loss: 0.8898947834968567:  74%|███████▍  | 4819/6470 [59:44<19:27,  1.41it/s]
3|4820|Loss: 0.8898947834968567:  74%|███████▍  | 4820/6470 [59:44<19:53,  1.38it/s]
3|4821|Loss: 0.8337987065315247:  74%|███████▍  | 4820/6470 [59:45<19:53,  1.38it/s]
3|4821|Loss: 0.8337987065315247:  75%|███████▍  | 4821/6470 [59:45<20:57,  1.31it/s]
3|4822|Loss: 0.9046474695205688:  75%|███████▍  | 4821/6470 [59:46<20:57,  1.31it/s]
3|4822|Loss: 0.9046474695205688:  75%|███████▍  | 4822/6470 [59:46<20:40,  1.33it/s]
3|4823|Loss: 0.7918227314949036:  75%|███████▍  | 4822/6470 [59:46<20:40,  1.33it/s]
3|4823|Loss: 0.7918227314949036:  75%|███████▍  | 4823/6470 [59:47<21:26,  1.28it/s]
3|4824|Loss: 0.8000497221946716:  75%|███████▍  | 4823/6470 [59:47<21:26,  1.28it/s]
3|4824|Loss: 0.8000497221946716:  75%|███████▍  | 4824/6470 [59:47<20:16,  1.35it/s]
3|4825|Loss: 0.7582100033760071:  75%|███████▍  | 4824/6470 [59:48<20:16,  1.35it/s]
3|4825|Loss: 0.7582100033760071:  75%|███████▍  | 4825/6470 [59:48<20:07,  1.36it/s]
3|4826|Loss: 0.943890392780304:  75%|███████▍  | 4825/6470 [59:49<20:07,  1.36it/s] 
3|4826|Loss: 0.943890392780304:  75%|███████▍  | 4826/6470 [59:49<18:54,  1.45it/s]
3|4827|Loss: 0.9186844229698181:  75%|███████▍  | 4826/6470 [59:49<18:54,  1.45it/s]
3|4827|Loss: 0.9186844229698181:  75%|███████▍  | 4827/6470 [59:49<18:38,  1.47it/s]
3|4828|Loss: 0.7810115814208984:  75%|███████▍  | 4827/6470 [59:50<18:38,  1.47it/s]
3|4828|Loss: 0.7810115814208984:  75%|███████▍  | 4828/6470 [59:50<19:24,  1.41it/s]
3|4829|Loss: 0.7149850130081177:  75%|███████▍  | 4828/6470 [59:51<19:24,  1.41it/s]
3|4829|Loss: 0.7149850130081177:  75%|███████▍  | 4829/6470 [59:51<18:34,  1.47it/s]
3|4830|Loss: 0.8596713542938232:  75%|███████▍  | 4829/6470 [59:51<18:34,  1.47it/s]
3|4830|Loss: 0.8596713542938232:  75%|███████▍  | 4830/6470 [59:51<17:53,  1.53it/s]
3|4831|Loss: 0.8612310886383057:  75%|███████▍  | 4830/6470 [59:52<17:53,  1.53it/s]
3|4831|Loss: 0.8612310886383057:  75%|███████▍  | 4831/6470 [59:52<18:42,  1.46it/s]
3|4832|Loss: 0.8531708717346191:  75%|███████▍  | 4831/6470 [59:53<18:42,  1.46it/s]
3|4832|Loss: 0.8531708717346191:  75%|███████▍  | 4832/6470 [59:53<18:54,  1.44it/s]
3|4833|Loss: 0.8551406860351562:  75%|███████▍  | 4832/6470 [59:53<18:54,  1.44it/s]
3|4833|Loss: 0.8551406860351562:  75%|███████▍  | 4833/6470 [59:54<19:18,  1.41it/s]
3|4834|Loss: 0.8399814963340759:  75%|███████▍  | 4833/6470 [59:54<19:18,  1.41it/s]
3|4834|Loss: 0.8399814963340759:  75%|███████▍  | 4834/6470 [59:54<18:45,  1.45it/s]
3|4835|Loss: 0.9718992710113525:  75%|███████▍  | 4834/6470 [59:55<18:45,  1.45it/s]
3|4835|Loss: 0.9718992710113525:  75%|███████▍  | 4835/6470 [59:55<18:27,  1.48it/s]
3|4836|Loss: 0.6366015076637268:  75%|███████▍  | 4835/6470 [59:55<18:27,  1.48it/s]
3|4836|Loss: 0.6366015076637268:  75%|███████▍  | 4836/6470 [59:56<19:41,  1.38it/s]
3|4837|Loss: 0.8321846127510071:  75%|███████▍  | 4836/6470 [59:56<19:41,  1.38it/s]
3|4837|Loss: 0.8321846127510071:  75%|███████▍  | 4837/6470 [59:57<20:20,  1.34it/s]
3|4838|Loss: 0.8321084380149841:  75%|███████▍  | 4837/6470 [59:57<20:20,  1.34it/s]
3|4838|Loss: 0.8321084380149841:  75%|███████▍  | 4838/6470 [59:57<20:39,  1.32it/s]
3|4839|Loss: 0.7524502277374268:  75%|███████▍  | 4838/6470 [59:58<20:39,  1.32it/s]
3|4839|Loss: 0.7524502277374268:  75%|███████▍  | 4839/6470 [59:58<20:51,  1.30it/s]
3|4840|Loss: 0.915005624294281:  75%|███████▍  | 4839/6470 [59:58<20:51,  1.30it/s] 
3|4840|Loss: 0.915005624294281:  75%|███████▍  | 4840/6470 [59:59<20:22,  1.33it/s]
3|4841|Loss: 0.7501797676086426:  75%|███████▍  | 4840/6470 [59:59<20:22,  1.33it/s]
3|4841|Loss: 0.7501797676086426:  75%|███████▍  | 4841/6470 [59:59<19:31,  1.39it/s]
3|4842|Loss: 0.6985105872154236:  75%|███████▍  | 4841/6470 [1:00:00<19:31,  1.39it/s]
3|4842|Loss: 0.6985105872154236:  75%|███████▍  | 4842/6470 [1:00:00<20:28,  1.33it/s]
3|4843|Loss: 0.8584591746330261:  75%|███████▍  | 4842/6470 [1:00:01<20:28,  1.33it/s]
3|4843|Loss: 0.8584591746330261:  75%|███████▍  | 4843/6470 [1:00:01<20:50,  1.30it/s]
3|4844|Loss: 0.885128378868103:  75%|███████▍  | 4843/6470 [1:00:01<20:50,  1.30it/s] 
3|4844|Loss: 0.885128378868103:  75%|███████▍  | 4844/6470 [1:00:02<20:09,  1.34it/s]
3|4845|Loss: 0.8889257907867432:  75%|███████▍  | 4844/6470 [1:00:02<20:09,  1.34it/s]
3|4845|Loss: 0.8889257907867432:  75%|███████▍  | 4845/6470 [1:00:02<19:42,  1.37it/s]
3|4846|Loss: 0.8621156811714172:  75%|███████▍  | 4845/6470 [1:00:03<19:42,  1.37it/s]
3|4846|Loss: 0.8621156811714172:  75%|███████▍  | 4846/6470 [1:00:03<18:58,  1.43it/s]
3|4847|Loss: 0.8040099143981934:  75%|███████▍  | 4846/6470 [1:00:03<18:58,  1.43it/s]
3|4847|Loss: 0.8040099143981934:  75%|███████▍  | 4847/6470 [1:00:04<19:25,  1.39it/s]
3|4848|Loss: 0.9658283591270447:  75%|███████▍  | 4847/6470 [1:00:04<19:25,  1.39it/s]
3|4848|Loss: 0.9658283591270447:  75%|███████▍  | 4848/6470 [1:00:05<20:00,  1.35it/s]
3|4849|Loss: 0.9471237659454346:  75%|███████▍  | 4848/6470 [1:00:05<20:00,  1.35it/s]
3|4849|Loss: 0.9471237659454346:  75%|███████▍  | 4849/6470 [1:00:06<21:01,  1.29it/s]
3|4850|Loss: 0.871513307094574:  75%|███████▍  | 4849/6470 [1:00:06<21:01,  1.29it/s] 
3|4850|Loss: 0.871513307094574:  75%|███████▍  | 4850/6470 [1:00:06<21:42,  1.24it/s]
3|4851|Loss: 0.9313231706619263:  75%|███████▍  | 4850/6470 [1:00:07<21:42,  1.24it/s]
3|4851|Loss: 0.9313231706619263:  75%|███████▍  | 4851/6470 [1:00:07<21:09,  1.28it/s]
3|4852|Loss: 0.8097827434539795:  75%|███████▍  | 4851/6470 [1:00:07<21:09,  1.28it/s]
3|4852|Loss: 0.8097827434539795:  75%|███████▍  | 4852/6470 [1:00:08<19:24,  1.39it/s]
3|4853|Loss: 0.714183509349823:  75%|███████▍  | 4852/6470 [1:00:08<19:24,  1.39it/s] 
3|4853|Loss: 0.714183509349823:  75%|███████▌  | 4853/6470 [1:00:08<17:56,  1.50it/s]
3|4854|Loss: 0.7982890009880066:  75%|███████▌  | 4853/6470 [1:00:09<17:56,  1.50it/s]
3|4854|Loss: 0.7982890009880066:  75%|███████▌  | 4854/6470 [1:00:09<17:50,  1.51it/s]
3|4855|Loss: 0.7628328204154968:  75%|███████▌  | 4854/6470 [1:00:09<17:50,  1.51it/s]
3|4855|Loss: 0.7628328204154968:  75%|███████▌  | 4855/6470 [1:00:10<19:30,  1.38it/s]
3|4856|Loss: 0.7336681485176086:  75%|███████▌  | 4855/6470 [1:00:10<19:30,  1.38it/s]
3|4856|Loss: 0.7336681485176086:  75%|███████▌  | 4856/6470 [1:00:10<17:36,  1.53it/s]
3|4857|Loss: 0.7318973541259766:  75%|███████▌  | 4856/6470 [1:00:11<17:36,  1.53it/s]
3|4857|Loss: 0.7318973541259766:  75%|███████▌  | 4857/6470 [1:00:11<17:15,  1.56it/s]
3|4858|Loss: 0.8912484049797058:  75%|███████▌  | 4857/6470 [1:00:11<17:15,  1.56it/s]
3|4858|Loss: 0.8912484049797058:  75%|███████▌  | 4858/6470 [1:00:12<18:18,  1.47it/s]
3|4859|Loss: 0.8229032754898071:  75%|███████▌  | 4858/6470 [1:00:12<18:18,  1.47it/s]
3|4859|Loss: 0.8229032754898071:  75%|███████▌  | 4859/6470 [1:00:13<20:07,  1.33it/s]
3|4860|Loss: 0.9647963643074036:  75%|███████▌  | 4859/6470 [1:00:13<20:07,  1.33it/s]
3|4860|Loss: 0.9647963643074036:  75%|███████▌  | 4860/6470 [1:00:13<19:54,  1.35it/s]
3|4861|Loss: 0.8965250253677368:  75%|███████▌  | 4860/6470 [1:00:14<19:54,  1.35it/s]
3|4861|Loss: 0.8965250253677368:  75%|███████▌  | 4861/6470 [1:00:14<19:50,  1.35it/s]
3|4862|Loss: 0.8236119747161865:  75%|███████▌  | 4861/6470 [1:00:14<19:50,  1.35it/s]
3|4862|Loss: 0.8236119747161865:  75%|███████▌  | 4862/6470 [1:00:15<21:19,  1.26it/s]
3|4863|Loss: 0.971270740032196:  75%|███████▌  | 4862/6470 [1:00:15<21:19,  1.26it/s] 
3|4863|Loss: 0.971270740032196:  75%|███████▌  | 4863/6470 [1:00:16<21:48,  1.23it/s]
3|4864|Loss: 0.8198451995849609:  75%|███████▌  | 4863/6470 [1:00:16<21:48,  1.23it/s]
3|4864|Loss: 0.8198451995849609:  75%|███████▌  | 4864/6470 [1:00:16<20:07,  1.33it/s]
3|4865|Loss: 0.9208698272705078:  75%|███████▌  | 4864/6470 [1:00:17<20:07,  1.33it/s]
3|4865|Loss: 0.9208698272705078:  75%|███████▌  | 4865/6470 [1:00:17<18:55,  1.41it/s]
3|4866|Loss: 0.8448465466499329:  75%|███████▌  | 4865/6470 [1:00:17<18:55,  1.41it/s]
3|4866|Loss: 0.8448465466499329:  75%|███████▌  | 4866/6470 [1:00:18<19:04,  1.40it/s]
3|4867|Loss: 0.8695331811904907:  75%|███████▌  | 4866/6470 [1:00:18<19:04,  1.40it/s]
3|4867|Loss: 0.8695331811904907:  75%|███████▌  | 4867/6470 [1:00:18<19:24,  1.38it/s]
3|4868|Loss: 0.7372944951057434:  75%|███████▌  | 4867/6470 [1:00:19<19:24,  1.38it/s]
3|4868|Loss: 0.7372944951057434:  75%|███████▌  | 4868/6470 [1:00:19<20:29,  1.30it/s]
3|4869|Loss: 0.7502005696296692:  75%|███████▌  | 4868/6470 [1:00:20<20:29,  1.30it/s]
3|4869|Loss: 0.7502005696296692:  75%|███████▌  | 4869/6470 [1:00:20<20:32,  1.30it/s]
3|4870|Loss: 0.8089627623558044:  75%|███████▌  | 4869/6470 [1:00:20<20:32,  1.30it/s]
3|4870|Loss: 0.8089627623558044:  75%|███████▌  | 4870/6470 [1:00:21<19:46,  1.35it/s]
3|4871|Loss: 0.866582453250885:  75%|███████▌  | 4870/6470 [1:00:21<19:46,  1.35it/s] 
3|4871|Loss: 0.866582453250885:  75%|███████▌  | 4871/6470 [1:00:21<18:17,  1.46it/s]
3|4872|Loss: 0.7650401592254639:  75%|███████▌  | 4871/6470 [1:00:22<18:17,  1.46it/s]
3|4872|Loss: 0.7650401592254639:  75%|███████▌  | 4872/6470 [1:00:22<18:30,  1.44it/s]
3|4873|Loss: 0.9953864812850952:  75%|███████▌  | 4872/6470 [1:00:23<18:30,  1.44it/s]
3|4873|Loss: 0.9953864812850952:  75%|███████▌  | 4873/6470 [1:00:23<19:42,  1.35it/s]
3|4874|Loss: 0.7122951745986938:  75%|███████▌  | 4873/6470 [1:00:23<19:42,  1.35it/s]
3|4874|Loss: 0.7122951745986938:  75%|███████▌  | 4874/6470 [1:00:24<19:22,  1.37it/s]
3|4875|Loss: 0.8437629342079163:  75%|███████▌  | 4874/6470 [1:00:24<19:22,  1.37it/s]
3|4875|Loss: 0.8437629342079163:  75%|███████▌  | 4875/6470 [1:00:24<19:02,  1.40it/s]
3|4876|Loss: 0.8041045069694519:  75%|███████▌  | 4875/6470 [1:00:25<19:02,  1.40it/s]
3|4876|Loss: 0.8041045069694519:  75%|███████▌  | 4876/6470 [1:00:25<19:24,  1.37it/s]
3|4877|Loss: 0.8030003309249878:  75%|███████▌  | 4876/6470 [1:00:26<19:24,  1.37it/s]
3|4877|Loss: 0.8030003309249878:  75%|███████▌  | 4877/6470 [1:00:26<20:15,  1.31it/s]
3|4878|Loss: 0.8512574434280396:  75%|███████▌  | 4877/6470 [1:00:26<20:15,  1.31it/s]
3|4878|Loss: 0.8512574434280396:  75%|███████▌  | 4878/6470 [1:00:27<20:37,  1.29it/s]
3|4879|Loss: 0.7706475853919983:  75%|███████▌  | 4878/6470 [1:00:27<20:37,  1.29it/s]
3|4879|Loss: 0.7706475853919983:  75%|███████▌  | 4879/6470 [1:00:27<18:43,  1.42it/s]
3|4880|Loss: 0.8461968302726746:  75%|███████▌  | 4879/6470 [1:00:28<18:43,  1.42it/s]
3|4880|Loss: 0.8461968302726746:  75%|███████▌  | 4880/6470 [1:00:28<17:22,  1.53it/s]
3|4881|Loss: 0.8428146839141846:  75%|███████▌  | 4880/6470 [1:00:28<17:22,  1.53it/s]
3|4881|Loss: 0.8428146839141846:  75%|███████▌  | 4881/6470 [1:00:28<17:47,  1.49it/s]
3|4882|Loss: 0.76678067445755:  75%|███████▌  | 4881/6470 [1:00:29<17:47,  1.49it/s]  
3|4882|Loss: 0.76678067445755:  75%|███████▌  | 4882/6470 [1:00:29<17:53,  1.48it/s]
3|4883|Loss: 0.6982433795928955:  75%|███████▌  | 4882/6470 [1:00:30<17:53,  1.48it/s]
3|4883|Loss: 0.6982433795928955:  75%|███████▌  | 4883/6470 [1:00:30<16:58,  1.56it/s]
3|4884|Loss: 0.796984076499939:  75%|███████▌  | 4883/6470 [1:00:30<16:58,  1.56it/s] 
3|4884|Loss: 0.796984076499939:  75%|███████▌  | 4884/6470 [1:00:31<18:35,  1.42it/s]
3|4885|Loss: 0.9440643191337585:  75%|███████▌  | 4884/6470 [1:00:31<18:35,  1.42it/s]
3|4885|Loss: 0.9440643191337585:  76%|███████▌  | 4885/6470 [1:00:31<19:09,  1.38it/s]
3|4886|Loss: 0.8652456402778625:  76%|███████▌  | 4885/6470 [1:00:32<19:09,  1.38it/s]
3|4886|Loss: 0.8652456402778625:  76%|███████▌  | 4886/6470 [1:00:32<19:03,  1.38it/s]
3|4887|Loss: 0.9869419932365417:  76%|███████▌  | 4886/6470 [1:00:32<19:03,  1.38it/s]
3|4887|Loss: 0.9869419932365417:  76%|███████▌  | 4887/6470 [1:00:33<18:38,  1.42it/s]
3|4888|Loss: 0.7391244769096375:  76%|███████▌  | 4887/6470 [1:00:33<18:38,  1.42it/s]
3|4888|Loss: 0.7391244769096375:  76%|███████▌  | 4888/6470 [1:00:33<18:30,  1.42it/s]
3|4889|Loss: 0.6609165668487549:  76%|███████▌  | 4888/6470 [1:00:34<18:30,  1.42it/s]
3|4889|Loss: 0.6609165668487549:  76%|███████▌  | 4889/6470 [1:00:34<19:32,  1.35it/s]
3|4890|Loss: 0.8233440518379211:  76%|███████▌  | 4889/6470 [1:00:35<19:32,  1.35it/s]
3|4890|Loss: 0.8233440518379211:  76%|███████▌  | 4890/6470 [1:00:35<18:35,  1.42it/s]
3|4891|Loss: 0.9039095044136047:  76%|███████▌  | 4890/6470 [1:00:35<18:35,  1.42it/s]
3|4891|Loss: 0.9039095044136047:  76%|███████▌  | 4891/6470 [1:00:36<18:19,  1.44it/s]
3|4892|Loss: 0.8165345191955566:  76%|███████▌  | 4891/6470 [1:00:36<18:19,  1.44it/s]
3|4892|Loss: 0.8165345191955566:  76%|███████▌  | 4892/6470 [1:00:36<18:24,  1.43it/s]
3|4893|Loss: 0.8472984433174133:  76%|███████▌  | 4892/6470 [1:00:37<18:24,  1.43it/s]
3|4893|Loss: 0.8472984433174133:  76%|███████▌  | 4893/6470 [1:00:37<18:14,  1.44it/s]
3|4894|Loss: 0.8793202638626099:  76%|███████▌  | 4893/6470 [1:00:37<18:14,  1.44it/s]
3|4894|Loss: 0.8793202638626099:  76%|███████▌  | 4894/6470 [1:00:38<18:06,  1.45it/s]
3|4895|Loss: 0.8828848600387573:  76%|███████▌  | 4894/6470 [1:00:38<18:06,  1.45it/s]
3|4895|Loss: 0.8828848600387573:  76%|███████▌  | 4895/6470 [1:00:39<20:20,  1.29it/s]
3|4896|Loss: 0.8495537638664246:  76%|███████▌  | 4895/6470 [1:00:39<20:20,  1.29it/s]
3|4896|Loss: 0.8495537638664246:  76%|███████▌  | 4896/6470 [1:00:39<20:16,  1.29it/s]
3|4897|Loss: 0.9662324786186218:  76%|███████▌  | 4896/6470 [1:00:40<20:16,  1.29it/s]
3|4897|Loss: 0.9662324786186218:  76%|███████▌  | 4897/6470 [1:00:40<20:44,  1.26it/s]
3|4898|Loss: 0.7959550023078918:  76%|███████▌  | 4897/6470 [1:00:41<20:44,  1.26it/s]
3|4898|Loss: 0.7959550023078918:  76%|███████▌  | 4898/6470 [1:00:41<18:15,  1.44it/s]
3|4899|Loss: 0.8005461096763611:  76%|███████▌  | 4898/6470 [1:00:41<18:15,  1.44it/s]
3|4899|Loss: 0.8005461096763611:  76%|███████▌  | 4899/6470 [1:00:41<17:50,  1.47it/s]
3|4900|Loss: 0.8802922368049622:  76%|███████▌  | 4899/6470 [1:00:42<17:50,  1.47it/s]
3|4900|Loss: 0.8802922368049622:  76%|███████▌  | 4900/6470 [1:00:42<19:32,  1.34it/s]
3|4901|Loss: 0.9130176901817322:  76%|███████▌  | 4900/6470 [1:00:43<19:32,  1.34it/s]
3|4901|Loss: 0.9130176901817322:  76%|███████▌  | 4901/6470 [1:00:43<20:52,  1.25it/s]
3|4902|Loss: 1.0323495864868164:  76%|███████▌  | 4901/6470 [1:00:44<20:52,  1.25it/s]
3|4902|Loss: 1.0323495864868164:  76%|███████▌  | 4902/6470 [1:00:44<18:59,  1.38it/s]
3|4903|Loss: 0.8832563757896423:  76%|███████▌  | 4902/6470 [1:00:44<18:59,  1.38it/s]
3|4903|Loss: 0.8832563757896423:  76%|███████▌  | 4903/6470 [1:00:44<17:48,  1.47it/s]
3|4904|Loss: 0.8076378107070923:  76%|███████▌  | 4903/6470 [1:00:45<17:48,  1.47it/s]
3|4904|Loss: 0.8076378107070923:  76%|███████▌  | 4904/6470 [1:00:45<18:26,  1.42it/s]
3|4905|Loss: 0.875634491443634:  76%|███████▌  | 4904/6470 [1:00:45<18:26,  1.42it/s] 
3|4905|Loss: 0.875634491443634:  76%|███████▌  | 4905/6470 [1:00:46<18:17,  1.43it/s]
3|4906|Loss: 0.7858631610870361:  76%|███████▌  | 4905/6470 [1:00:46<18:17,  1.43it/s]
3|4906|Loss: 0.7858631610870361:  76%|███████▌  | 4906/6470 [1:00:46<17:58,  1.45it/s]
3|4907|Loss: 0.7285030484199524:  76%|███████▌  | 4906/6470 [1:00:47<17:58,  1.45it/s]
3|4907|Loss: 0.7285030484199524:  76%|███████▌  | 4907/6470 [1:00:47<17:28,  1.49it/s]
3|4908|Loss: 0.8182156085968018:  76%|███████▌  | 4907/6470 [1:00:47<17:28,  1.49it/s]
3|4908|Loss: 0.8182156085968018:  76%|███████▌  | 4908/6470 [1:00:48<17:16,  1.51it/s]
3|4909|Loss: 0.7818955779075623:  76%|███████▌  | 4908/6470 [1:00:48<17:16,  1.51it/s]
3|4909|Loss: 0.7818955779075623:  76%|███████▌  | 4909/6470 [1:00:48<18:02,  1.44it/s]
3|4910|Loss: 0.9521711468696594:  76%|███████▌  | 4909/6470 [1:00:49<18:02,  1.44it/s]
3|4910|Loss: 0.9521711468696594:  76%|███████▌  | 4910/6470 [1:00:49<17:23,  1.49it/s]
3|4911|Loss: 0.8003957271575928:  76%|███████▌  | 4910/6470 [1:00:49<17:23,  1.49it/s]
3|4911|Loss: 0.8003957271575928:  76%|███████▌  | 4911/6470 [1:00:50<18:08,  1.43it/s]
3|4912|Loss: 0.7021927237510681:  76%|███████▌  | 4911/6470 [1:00:50<18:08,  1.43it/s]
3|4912|Loss: 0.7021927237510681:  76%|███████▌  | 4912/6470 [1:00:51<17:57,  1.45it/s]
3|4913|Loss: 0.6873462200164795:  76%|███████▌  | 4912/6470 [1:00:51<17:57,  1.45it/s]
3|4913|Loss: 0.6873462200164795:  76%|███████▌  | 4913/6470 [1:00:51<18:40,  1.39it/s]
3|4914|Loss: 0.8045195937156677:  76%|███████▌  | 4913/6470 [1:00:52<18:40,  1.39it/s]
3|4914|Loss: 0.8045195937156677:  76%|███████▌  | 4914/6470 [1:00:52<20:09,  1.29it/s]
3|4915|Loss: 0.6597917675971985:  76%|███████▌  | 4914/6470 [1:00:53<20:09,  1.29it/s]
3|4915|Loss: 0.6597917675971985:  76%|███████▌  | 4915/6470 [1:00:53<19:39,  1.32it/s]
3|4916|Loss: 0.7913355231285095:  76%|███████▌  | 4915/6470 [1:00:53<19:39,  1.32it/s]
3|4916|Loss: 0.7913355231285095:  76%|███████▌  | 4916/6470 [1:00:54<18:24,  1.41it/s]
3|4917|Loss: 0.9021652340888977:  76%|███████▌  | 4916/6470 [1:00:54<18:24,  1.41it/s]
3|4917|Loss: 0.9021652340888977:  76%|███████▌  | 4917/6470 [1:00:54<17:10,  1.51it/s]
3|4918|Loss: 0.8439438939094543:  76%|███████▌  | 4917/6470 [1:00:54<17:10,  1.51it/s]
3|4918|Loss: 0.8439438939094543:  76%|███████▌  | 4918/6470 [1:00:55<18:12,  1.42it/s]
3|4919|Loss: 0.815605103969574:  76%|███████▌  | 4918/6470 [1:00:55<18:12,  1.42it/s] 
3|4919|Loss: 0.815605103969574:  76%|███████▌  | 4919/6470 [1:00:56<17:55,  1.44it/s]
3|4920|Loss: 0.8359621167182922:  76%|███████▌  | 4919/6470 [1:00:56<17:55,  1.44it/s]
3|4920|Loss: 0.8359621167182922:  76%|███████▌  | 4920/6470 [1:00:56<18:11,  1.42it/s]
3|4921|Loss: 0.9205386638641357:  76%|███████▌  | 4920/6470 [1:00:57<18:11,  1.42it/s]
3|4921|Loss: 0.9205386638641357:  76%|███████▌  | 4921/6470 [1:00:57<17:09,  1.50it/s]
3|4922|Loss: 0.7222816348075867:  76%|███████▌  | 4921/6470 [1:00:57<17:09,  1.50it/s]
3|4922|Loss: 0.7222816348075867:  76%|███████▌  | 4922/6470 [1:00:58<17:47,  1.45it/s]
3|4923|Loss: 0.9726329445838928:  76%|███████▌  | 4922/6470 [1:00:58<17:47,  1.45it/s]
3|4923|Loss: 0.9726329445838928:  76%|███████▌  | 4923/6470 [1:00:58<18:02,  1.43it/s]
3|4924|Loss: 0.8395279049873352:  76%|███████▌  | 4923/6470 [1:00:59<18:02,  1.43it/s]
3|4924|Loss: 0.8395279049873352:  76%|███████▌  | 4924/6470 [1:00:59<18:42,  1.38it/s]
3|4925|Loss: 0.8505263328552246:  76%|███████▌  | 4924/6470 [1:01:00<18:42,  1.38it/s]
3|4925|Loss: 0.8505263328552246:  76%|███████▌  | 4925/6470 [1:01:00<18:34,  1.39it/s]
3|4926|Loss: 0.887738049030304:  76%|███████▌  | 4925/6470 [1:01:00<18:34,  1.39it/s] 
3|4926|Loss: 0.887738049030304:  76%|███████▌  | 4926/6470 [1:01:01<18:57,  1.36it/s]
3|4927|Loss: 0.778435468673706:  76%|███████▌  | 4926/6470 [1:01:01<18:57,  1.36it/s]
3|4927|Loss: 0.778435468673706:  76%|███████▌  | 4927/6470 [1:01:01<19:28,  1.32it/s]
3|4928|Loss: 0.9589358568191528:  76%|███████▌  | 4927/6470 [1:01:02<19:28,  1.32it/s]
3|4928|Loss: 0.9589358568191528:  76%|███████▌  | 4928/6470 [1:01:02<20:43,  1.24it/s]
3|4929|Loss: 0.8314003348350525:  76%|███████▌  | 4928/6470 [1:01:03<20:43,  1.24it/s]
3|4929|Loss: 0.8314003348350525:  76%|███████▌  | 4929/6470 [1:01:03<20:40,  1.24it/s]
3|4930|Loss: 0.7472063302993774:  76%|███████▌  | 4929/6470 [1:01:04<20:40,  1.24it/s]
3|4930|Loss: 0.7472063302993774:  76%|███████▌  | 4930/6470 [1:01:04<19:45,  1.30it/s]
3|4931|Loss: 0.6891616582870483:  76%|███████▌  | 4930/6470 [1:01:04<19:45,  1.30it/s]
3|4931|Loss: 0.6891616582870483:  76%|███████▌  | 4931/6470 [1:01:05<20:10,  1.27it/s]
3|4932|Loss: 0.8590701222419739:  76%|███████▌  | 4931/6470 [1:01:05<20:10,  1.27it/s]
3|4932|Loss: 0.8590701222419739:  76%|███████▌  | 4932/6470 [1:01:06<21:09,  1.21it/s]
3|4933|Loss: 0.8061633110046387:  76%|███████▌  | 4932/6470 [1:01:06<21:09,  1.21it/s]
3|4933|Loss: 0.8061633110046387:  76%|███████▌  | 4933/6470 [1:01:06<20:53,  1.23it/s]
3|4934|Loss: 0.7333589792251587:  76%|███████▌  | 4933/6470 [1:01:07<20:53,  1.23it/s]
3|4934|Loss: 0.7333589792251587:  76%|███████▋  | 4934/6470 [1:01:07<20:34,  1.24it/s]
3|4935|Loss: 0.9208199977874756:  76%|███████▋  | 4934/6470 [1:01:07<20:34,  1.24it/s]
3|4935|Loss: 0.9208199977874756:  76%|███████▋  | 4935/6470 [1:01:08<18:53,  1.35it/s]
3|4936|Loss: 0.8184888958930969:  76%|███████▋  | 4935/6470 [1:01:08<18:53,  1.35it/s]
3|4936|Loss: 0.8184888958930969:  76%|███████▋  | 4936/6470 [1:01:08<18:39,  1.37it/s]
3|4937|Loss: 0.757254958152771:  76%|███████▋  | 4936/6470 [1:01:09<18:39,  1.37it/s] 
3|4937|Loss: 0.757254958152771:  76%|███████▋  | 4937/6470 [1:01:09<18:19,  1.39it/s]
3|4938|Loss: 0.7053201198577881:  76%|███████▋  | 4937/6470 [1:01:09<18:19,  1.39it/s]
3|4938|Loss: 0.7053201198577881:  76%|███████▋  | 4938/6470 [1:01:10<16:43,  1.53it/s]
3|4939|Loss: 0.809824526309967:  76%|███████▋  | 4938/6470 [1:01:10<16:43,  1.53it/s] 
3|4939|Loss: 0.809824526309967:  76%|███████▋  | 4939/6470 [1:01:10<17:12,  1.48it/s]
3|4940|Loss: 0.9033624529838562:  76%|███████▋  | 4939/6470 [1:01:11<17:12,  1.48it/s]
3|4940|Loss: 0.9033624529838562:  76%|███████▋  | 4940/6470 [1:01:11<18:46,  1.36it/s]
3|4941|Loss: 0.9480109810829163:  76%|███████▋  | 4940/6470 [1:01:12<18:46,  1.36it/s]
3|4941|Loss: 0.9480109810829163:  76%|███████▋  | 4941/6470 [1:01:12<19:25,  1.31it/s]
3|4942|Loss: 0.7578059434890747:  76%|███████▋  | 4941/6470 [1:01:12<19:25,  1.31it/s]
3|4942|Loss: 0.7578059434890747:  76%|███████▋  | 4942/6470 [1:01:13<20:28,  1.24it/s]
3|4943|Loss: 0.9076833128929138:  76%|███████▋  | 4942/6470 [1:01:13<20:28,  1.24it/s]
3|4943|Loss: 0.9076833128929138:  76%|███████▋  | 4943/6470 [1:01:14<21:07,  1.20it/s]
3|4944|Loss: 0.9300745129585266:  76%|███████▋  | 4943/6470 [1:01:14<21:07,  1.20it/s]
3|4944|Loss: 0.9300745129585266:  76%|███████▋  | 4944/6470 [1:01:14<19:33,  1.30it/s]
3|4945|Loss: 0.8338807821273804:  76%|███████▋  | 4944/6470 [1:01:15<19:33,  1.30it/s]
3|4945|Loss: 0.8338807821273804:  76%|███████▋  | 4945/6470 [1:01:15<18:14,  1.39it/s]
3|4946|Loss: 0.9125658869743347:  76%|███████▋  | 4945/6470 [1:01:15<18:14,  1.39it/s]
3|4946|Loss: 0.9125658869743347:  76%|███████▋  | 4946/6470 [1:01:16<17:46,  1.43it/s]
3|4947|Loss: 0.7658064961433411:  76%|███████▋  | 4946/6470 [1:01:16<17:46,  1.43it/s]
3|4947|Loss: 0.7658064961433411:  76%|███████▋  | 4947/6470 [1:01:16<16:43,  1.52it/s]
3|4948|Loss: 0.7066605687141418:  76%|███████▋  | 4947/6470 [1:01:17<16:43,  1.52it/s]
3|4948|Loss: 0.7066605687141418:  76%|███████▋  | 4948/6470 [1:01:17<16:43,  1.52it/s]
3|4949|Loss: 0.7053583860397339:  76%|███████▋  | 4948/6470 [1:01:17<16:43,  1.52it/s]
3|4949|Loss: 0.7053583860397339:  76%|███████▋  | 4949/6470 [1:01:18<16:29,  1.54it/s]
3|4950|Loss: 0.8020024299621582:  76%|███████▋  | 4949/6470 [1:01:18<16:29,  1.54it/s]
3|4950|Loss: 0.8020024299621582:  77%|███████▋  | 4950/6470 [1:01:18<17:54,  1.41it/s]
3|4951|Loss: 0.9262269139289856:  77%|███████▋  | 4950/6470 [1:01:19<17:54,  1.41it/s]
3|4951|Loss: 0.9262269139289856:  77%|███████▋  | 4951/6470 [1:01:19<18:39,  1.36it/s]
3|4952|Loss: 0.8303079605102539:  77%|███████▋  | 4951/6470 [1:01:20<18:39,  1.36it/s]
3|4952|Loss: 0.8303079605102539:  77%|███████▋  | 4952/6470 [1:01:20<18:22,  1.38it/s]
3|4953|Loss: 0.7313457131385803:  77%|███████▋  | 4952/6470 [1:01:20<18:22,  1.38it/s]
3|4953|Loss: 0.7313457131385803:  77%|███████▋  | 4953/6470 [1:01:20<17:09,  1.47it/s]
3|4954|Loss: 0.8641994595527649:  77%|███████▋  | 4953/6470 [1:01:21<17:09,  1.47it/s]
3|4954|Loss: 0.8641994595527649:  77%|███████▋  | 4954/6470 [1:01:21<17:59,  1.40it/s]
3|4955|Loss: 1.003259301185608:  77%|███████▋  | 4954/6470 [1:01:22<17:59,  1.40it/s] 
3|4955|Loss: 1.003259301185608:  77%|███████▋  | 4955/6470 [1:01:22<18:35,  1.36it/s]
3|4956|Loss: 0.8428483605384827:  77%|███████▋  | 4955/6470 [1:01:22<18:35,  1.36it/s]
3|4956|Loss: 0.8428483605384827:  77%|███████▋  | 4956/6470 [1:01:23<18:53,  1.34it/s]
3|4957|Loss: 0.9244130253791809:  77%|███████▋  | 4956/6470 [1:01:23<18:53,  1.34it/s]
3|4957|Loss: 0.9244130253791809:  77%|███████▋  | 4957/6470 [1:01:24<19:45,  1.28it/s]
3|4958|Loss: 0.9899126291275024:  77%|███████▋  | 4957/6470 [1:01:24<19:45,  1.28it/s]
3|4958|Loss: 0.9899126291275024:  77%|███████▋  | 4958/6470 [1:01:25<20:50,  1.21it/s]
3|4959|Loss: 0.8916161060333252:  77%|███████▋  | 4958/6470 [1:01:25<20:50,  1.21it/s]
3|4959|Loss: 0.8916161060333252:  77%|███████▋  | 4959/6470 [1:01:25<19:37,  1.28it/s]
3|4960|Loss: 0.9092479944229126:  77%|███████▋  | 4959/6470 [1:01:26<19:37,  1.28it/s]
3|4960|Loss: 0.9092479944229126:  77%|███████▋  | 4960/6470 [1:01:26<19:20,  1.30it/s]
3|4961|Loss: 0.7792670130729675:  77%|███████▋  | 4960/6470 [1:01:26<19:20,  1.30it/s]
3|4961|Loss: 0.7792670130729675:  77%|███████▋  | 4961/6470 [1:01:27<19:24,  1.30it/s]
3|4962|Loss: 0.8628126978874207:  77%|███████▋  | 4961/6470 [1:01:27<19:24,  1.30it/s]
3|4962|Loss: 0.8628126978874207:  77%|███████▋  | 4962/6470 [1:01:27<17:42,  1.42it/s]
3|4963|Loss: 0.8180738091468811:  77%|███████▋  | 4962/6470 [1:01:28<17:42,  1.42it/s]
3|4963|Loss: 0.8180738091468811:  77%|███████▋  | 4963/6470 [1:01:28<18:36,  1.35it/s]
3|4964|Loss: 0.7556892037391663:  77%|███████▋  | 4963/6470 [1:01:29<18:36,  1.35it/s]
3|4964|Loss: 0.7556892037391663:  77%|███████▋  | 4964/6470 [1:01:29<19:33,  1.28it/s]
3|4965|Loss: 0.8536376953125:  77%|███████▋  | 4964/6470 [1:01:29<19:33,  1.28it/s]   
3|4965|Loss: 0.8536376953125:  77%|███████▋  | 4965/6470 [1:01:30<18:36,  1.35it/s]
3|4966|Loss: 0.7688703536987305:  77%|███████▋  | 4965/6470 [1:01:30<18:36,  1.35it/s]
3|4966|Loss: 0.7688703536987305:  77%|███████▋  | 4966/6470 [1:01:31<19:35,  1.28it/s]
3|4967|Loss: 0.8666771054267883:  77%|███████▋  | 4966/6470 [1:01:31<19:35,  1.28it/s]
3|4967|Loss: 0.8666771054267883:  77%|███████▋  | 4967/6470 [1:01:31<20:03,  1.25it/s]
3|4968|Loss: 0.7927407026290894:  77%|███████▋  | 4967/6470 [1:01:32<20:03,  1.25it/s]
3|4968|Loss: 0.7927407026290894:  77%|███████▋  | 4968/6470 [1:01:32<19:29,  1.28it/s]
3|4969|Loss: 0.8004407286643982:  77%|███████▋  | 4968/6470 [1:01:32<19:29,  1.28it/s]
3|4969|Loss: 0.8004407286643982:  77%|███████▋  | 4969/6470 [1:01:33<17:44,  1.41it/s]
3|4970|Loss: 0.7557364106178284:  77%|███████▋  | 4969/6470 [1:01:33<17:44,  1.41it/s]
3|4970|Loss: 0.7557364106178284:  77%|███████▋  | 4970/6470 [1:01:33<16:01,  1.56it/s]
3|4971|Loss: 0.9246277809143066:  77%|███████▋  | 4970/6470 [1:01:34<16:01,  1.56it/s]
3|4971|Loss: 0.9246277809143066:  77%|███████▋  | 4971/6470 [1:01:34<16:23,  1.52it/s]
3|4972|Loss: 0.7833227515220642:  77%|███████▋  | 4971/6470 [1:01:34<16:23,  1.52it/s]
3|4972|Loss: 0.7833227515220642:  77%|███████▋  | 4972/6470 [1:01:35<17:23,  1.43it/s]
3|4973|Loss: 0.6316429376602173:  77%|███████▋  | 4972/6470 [1:01:35<17:23,  1.43it/s]
3|4973|Loss: 0.6316429376602173:  77%|███████▋  | 4973/6470 [1:01:36<18:33,  1.34it/s]
3|4974|Loss: 0.8241785764694214:  77%|███████▋  | 4973/6470 [1:01:36<18:33,  1.34it/s]
3|4974|Loss: 0.8241785764694214:  77%|███████▋  | 4974/6470 [1:01:36<19:03,  1.31it/s]
3|4975|Loss: 0.8056584000587463:  77%|███████▋  | 4974/6470 [1:01:37<19:03,  1.31it/s]
3|4975|Loss: 0.8056584000587463:  77%|███████▋  | 4975/6470 [1:01:37<20:28,  1.22it/s]
3|4976|Loss: 0.7936295866966248:  77%|███████▋  | 4975/6470 [1:01:38<20:28,  1.22it/s]
3|4976|Loss: 0.7936295866966248:  77%|███████▋  | 4976/6470 [1:01:38<20:59,  1.19it/s]
3|4977|Loss: 0.8824145793914795:  77%|███████▋  | 4976/6470 [1:01:39<20:59,  1.19it/s]
3|4977|Loss: 0.8824145793914795:  77%|███████▋  | 4977/6470 [1:01:39<23:58,  1.04it/s]
3|4978|Loss: 0.768092691898346:  77%|███████▋  | 4977/6470 [1:01:40<23:58,  1.04it/s] 
3|4978|Loss: 0.768092691898346:  77%|███████▋  | 4978/6470 [1:01:40<22:58,  1.08it/s]
3|4979|Loss: 0.8929199576377869:  77%|███████▋  | 4978/6470 [1:01:41<22:58,  1.08it/s]
3|4979|Loss: 0.8929199576377869:  77%|███████▋  | 4979/6470 [1:01:41<21:38,  1.15it/s]
3|4980|Loss: 0.7189871668815613:  77%|███████▋  | 4979/6470 [1:01:41<21:38,  1.15it/s]
3|4980|Loss: 0.7189871668815613:  77%|███████▋  | 4980/6470 [1:01:42<21:37,  1.15it/s]
3|4981|Loss: 0.8813014030456543:  77%|███████▋  | 4980/6470 [1:01:42<21:37,  1.15it/s]
3|4981|Loss: 0.8813014030456543:  77%|███████▋  | 4981/6470 [1:01:43<21:09,  1.17it/s]
3|4982|Loss: 0.7584362030029297:  77%|███████▋  | 4981/6470 [1:01:43<21:09,  1.17it/s]
3|4982|Loss: 0.7584362030029297:  77%|███████▋  | 4982/6470 [1:01:43<20:36,  1.20it/s]
3|4983|Loss: 0.9167375564575195:  77%|███████▋  | 4982/6470 [1:01:44<20:36,  1.20it/s]
3|4983|Loss: 0.9167375564575195:  77%|███████▋  | 4983/6470 [1:01:44<19:11,  1.29it/s]
3|4984|Loss: 0.7704063057899475:  77%|███████▋  | 4983/6470 [1:01:44<19:11,  1.29it/s]
3|4984|Loss: 0.7704063057899475:  77%|███████▋  | 4984/6470 [1:01:45<16:33,  1.50it/s]
3|4985|Loss: 0.8635309934616089:  77%|███████▋  | 4984/6470 [1:01:45<16:33,  1.50it/s]
3|4985|Loss: 0.8635309934616089:  77%|███████▋  | 4985/6470 [1:01:45<16:43,  1.48it/s]
3|4986|Loss: 0.9248243570327759:  77%|███████▋  | 4985/6470 [1:01:46<16:43,  1.48it/s]
3|4986|Loss: 0.9248243570327759:  77%|███████▋  | 4986/6470 [1:01:46<16:41,  1.48it/s]
3|4987|Loss: 0.8527777791023254:  77%|███████▋  | 4986/6470 [1:01:46<16:41,  1.48it/s]
3|4987|Loss: 0.8527777791023254:  77%|███████▋  | 4987/6470 [1:01:47<18:23,  1.34it/s]
3|4988|Loss: 0.8771868348121643:  77%|███████▋  | 4987/6470 [1:01:47<18:23,  1.34it/s]
3|4988|Loss: 0.8771868348121643:  77%|███████▋  | 4988/6470 [1:01:48<19:40,  1.26it/s]
3|4989|Loss: 0.8214733004570007:  77%|███████▋  | 4988/6470 [1:01:48<19:40,  1.26it/s]
3|4989|Loss: 0.8214733004570007:  77%|███████▋  | 4989/6470 [1:01:49<19:46,  1.25it/s]
3|4990|Loss: 0.805493950843811:  77%|███████▋  | 4989/6470 [1:01:49<19:46,  1.25it/s] 
3|4990|Loss: 0.805493950843811:  77%|███████▋  | 4990/6470 [1:01:49<20:28,  1.20it/s]
3|4991|Loss: 0.826143741607666:  77%|███████▋  | 4990/6470 [1:01:50<20:28,  1.20it/s]
3|4991|Loss: 0.826143741607666:  77%|███████▋  | 4991/6470 [1:01:50<19:55,  1.24it/s]
3|4992|Loss: 0.7686324119567871:  77%|███████▋  | 4991/6470 [1:01:51<19:55,  1.24it/s]
3|4992|Loss: 0.7686324119567871:  77%|███████▋  | 4992/6470 [1:01:51<19:38,  1.25it/s]
3|4993|Loss: 0.8488655686378479:  77%|███████▋  | 4992/6470 [1:01:51<19:38,  1.25it/s]
3|4993|Loss: 0.8488655686378479:  77%|███████▋  | 4993/6470 [1:01:52<19:20,  1.27it/s]
3|4994|Loss: 0.9283349514007568:  77%|███████▋  | 4993/6470 [1:01:52<19:20,  1.27it/s]
3|4994|Loss: 0.9283349514007568:  77%|███████▋  | 4994/6470 [1:01:52<18:16,  1.35it/s]
3|4995|Loss: 0.7323651909828186:  77%|███████▋  | 4994/6470 [1:01:53<18:16,  1.35it/s]
3|4995|Loss: 0.7323651909828186:  77%|███████▋  | 4995/6470 [1:01:53<18:22,  1.34it/s]
3|4996|Loss: 0.7829907536506653:  77%|███████▋  | 4995/6470 [1:01:54<18:22,  1.34it/s]
3|4996|Loss: 0.7829907536506653:  77%|███████▋  | 4996/6470 [1:01:54<17:41,  1.39it/s]
3|4997|Loss: 1.0249121189117432:  77%|███████▋  | 4996/6470 [1:01:54<17:41,  1.39it/s]
3|4997|Loss: 1.0249121189117432:  77%|███████▋  | 4997/6470 [1:01:54<16:23,  1.50it/s]
3|4998|Loss: 0.7818512320518494:  77%|███████▋  | 4997/6470 [1:01:55<16:23,  1.50it/s]
3|4998|Loss: 0.7818512320518494:  77%|███████▋  | 4998/6470 [1:01:55<16:06,  1.52it/s]
3|4999|Loss: 0.6769594550132751:  77%|███████▋  | 4998/6470 [1:01:55<16:06,  1.52it/s]
3|4999|Loss: 0.6769594550132751:  77%|███████▋  | 4999/6470 [1:01:56<15:51,  1.55it/s]
3|5000|Loss: 0.8188980221748352:  77%|███████▋  | 4999/6470 [1:01:56<15:51,  1.55it/s]
3|5000|Loss: 0.8188980221748352:  77%|███████▋  | 5000/6470 [1:01:56<16:34,  1.48it/s]
3|5001|Loss: 0.6894150972366333:  77%|███████▋  | 5000/6470 [1:01:57<16:34,  1.48it/s]
3|5001|Loss: 0.6894150972366333:  77%|███████▋  | 5001/6470 [1:01:57<17:43,  1.38it/s]
3|5002|Loss: 0.7067097425460815:  77%|███████▋  | 5001/6470 [1:01:58<17:43,  1.38it/s]
3|5002|Loss: 0.7067097425460815:  77%|███████▋  | 5002/6470 [1:01:58<19:00,  1.29it/s]
3|5003|Loss: 0.8307633399963379:  77%|███████▋  | 5002/6470 [1:01:59<19:00,  1.29it/s]
3|5003|Loss: 0.8307633399963379:  77%|███████▋  | 5003/6470 [1:01:59<19:19,  1.27it/s]
3|5004|Loss: 0.9926633834838867:  77%|███████▋  | 5003/6470 [1:01:59<19:19,  1.27it/s]
3|5004|Loss: 0.9926633834838867:  77%|███████▋  | 5004/6470 [1:02:00<19:26,  1.26it/s]
3|5005|Loss: 1.0182247161865234:  77%|███████▋  | 5004/6470 [1:02:00<19:26,  1.26it/s]
3|5005|Loss: 1.0182247161865234:  77%|███████▋  | 5005/6470 [1:02:00<18:48,  1.30it/s]
3|5006|Loss: 0.661919355392456:  77%|███████▋  | 5005/6470 [1:02:01<18:48,  1.30it/s] 
3|5006|Loss: 0.661919355392456:  77%|███████▋  | 5006/6470 [1:02:01<18:01,  1.35it/s]
3|5007|Loss: 1.0101431608200073:  77%|███████▋  | 5006/6470 [1:02:01<18:01,  1.35it/s]
3|5007|Loss: 1.0101431608200073:  77%|███████▋  | 5007/6470 [1:02:02<17:30,  1.39it/s]
3|5008|Loss: 0.7455930113792419:  77%|███████▋  | 5007/6470 [1:02:02<17:30,  1.39it/s]
3|5008|Loss: 0.7455930113792419:  77%|███████▋  | 5008/6470 [1:02:03<17:56,  1.36it/s]
3|5009|Loss: 0.7078783512115479:  77%|███████▋  | 5008/6470 [1:02:03<17:56,  1.36it/s]
3|5009|Loss: 0.7078783512115479:  77%|███████▋  | 5009/6470 [1:02:03<17:39,  1.38it/s]
3|5010|Loss: 0.7089433073997498:  77%|███████▋  | 5009/6470 [1:02:04<17:39,  1.38it/s]
3|5010|Loss: 0.7089433073997498:  77%|███████▋  | 5010/6470 [1:02:04<17:44,  1.37it/s]
3|5011|Loss: 0.8478016257286072:  77%|███████▋  | 5010/6470 [1:02:04<17:44,  1.37it/s]
3|5011|Loss: 0.8478016257286072:  77%|███████▋  | 5011/6470 [1:02:05<17:05,  1.42it/s]
3|5012|Loss: 0.7390252351760864:  77%|███████▋  | 5011/6470 [1:02:05<17:05,  1.42it/s]
3|5012|Loss: 0.7390252351760864:  77%|███████▋  | 5012/6470 [1:02:06<18:40,  1.30it/s]
3|5013|Loss: 0.8324087262153625:  77%|███████▋  | 5012/6470 [1:02:06<18:40,  1.30it/s]
3|5013|Loss: 0.8324087262153625:  77%|███████▋  | 5013/6470 [1:02:06<19:08,  1.27it/s]
3|5014|Loss: 0.8274766802787781:  77%|███████▋  | 5013/6470 [1:02:07<19:08,  1.27it/s]
3|5014|Loss: 0.8274766802787781:  77%|███████▋  | 5014/6470 [1:02:07<19:34,  1.24it/s]
3|5015|Loss: 0.7501256465911865:  77%|███████▋  | 5014/6470 [1:02:08<19:34,  1.24it/s]
3|5015|Loss: 0.7501256465911865:  78%|███████▊  | 5015/6470 [1:02:08<18:22,  1.32it/s]
3|5016|Loss: 0.8227977156639099:  78%|███████▊  | 5015/6470 [1:02:08<18:22,  1.32it/s]
3|5016|Loss: 0.8227977156639099:  78%|███████▊  | 5016/6470 [1:02:09<18:46,  1.29it/s]
3|5017|Loss: 0.9723594188690186:  78%|███████▊  | 5016/6470 [1:02:09<18:46,  1.29it/s]
3|5017|Loss: 0.9723594188690186:  78%|███████▊  | 5017/6470 [1:02:10<19:33,  1.24it/s]
3|5018|Loss: 0.8192500472068787:  78%|███████▊  | 5017/6470 [1:02:10<19:33,  1.24it/s]
3|5018|Loss: 0.8192500472068787:  78%|███████▊  | 5018/6470 [1:02:10<19:38,  1.23it/s]
3|5019|Loss: 0.9553017616271973:  78%|███████▊  | 5018/6470 [1:02:11<19:38,  1.23it/s]
3|5019|Loss: 0.9553017616271973:  78%|███████▊  | 5019/6470 [1:02:11<19:37,  1.23it/s]
3|5020|Loss: 0.7486857175827026:  78%|███████▊  | 5019/6470 [1:02:12<19:37,  1.23it/s]
3|5020|Loss: 0.7486857175827026:  78%|███████▊  | 5020/6470 [1:02:12<19:36,  1.23it/s]
3|5021|Loss: 0.7856007218360901:  78%|███████▊  | 5020/6470 [1:02:12<19:36,  1.23it/s]
3|5021|Loss: 0.7856007218360901:  78%|███████▊  | 5021/6470 [1:02:13<19:30,  1.24it/s]
3|5022|Loss: 0.9125959873199463:  78%|███████▊  | 5021/6470 [1:02:13<19:30,  1.24it/s]
3|5022|Loss: 0.9125959873199463:  78%|███████▊  | 5022/6470 [1:02:13<18:37,  1.30it/s]
3|5023|Loss: 0.804898202419281:  78%|███████▊  | 5022/6470 [1:02:14<18:37,  1.30it/s] 
3|5023|Loss: 0.804898202419281:  78%|███████▊  | 5023/6470 [1:02:14<16:43,  1.44it/s]
3|5024|Loss: 0.8288283348083496:  78%|███████▊  | 5023/6470 [1:02:14<16:43,  1.44it/s]
3|5024|Loss: 0.8288283348083496:  78%|███████▊  | 5024/6470 [1:02:15<17:20,  1.39it/s]
3|5025|Loss: 0.777556300163269:  78%|███████▊  | 5024/6470 [1:02:15<17:20,  1.39it/s] 
3|5025|Loss: 0.777556300163269:  78%|███████▊  | 5025/6470 [1:02:15<17:15,  1.40it/s]
3|5026|Loss: 0.7440181374549866:  78%|███████▊  | 5025/6470 [1:02:16<17:15,  1.40it/s]
3|5026|Loss: 0.7440181374549866:  78%|███████▊  | 5026/6470 [1:02:16<16:23,  1.47it/s]
3|5027|Loss: 0.7149481773376465:  78%|███████▊  | 5026/6470 [1:02:16<16:23,  1.47it/s]
3|5027|Loss: 0.7149481773376465:  78%|███████▊  | 5027/6470 [1:02:17<16:09,  1.49it/s]
3|5028|Loss: 0.8253598809242249:  78%|███████▊  | 5027/6470 [1:02:17<16:09,  1.49it/s]
3|5028|Loss: 0.8253598809242249:  78%|███████▊  | 5028/6470 [1:02:18<17:16,  1.39it/s]
3|5029|Loss: 0.9341634511947632:  78%|███████▊  | 5028/6470 [1:02:18<17:16,  1.39it/s]
3|5029|Loss: 0.9341634511947632:  78%|███████▊  | 5029/6470 [1:02:18<17:20,  1.38it/s]
3|5030|Loss: 0.9192762970924377:  78%|███████▊  | 5029/6470 [1:02:19<17:20,  1.38it/s]
3|5030|Loss: 0.9192762970924377:  78%|███████▊  | 5030/6470 [1:02:19<17:02,  1.41it/s]
3|5031|Loss: 0.6877588033676147:  78%|███████▊  | 5030/6470 [1:02:19<17:02,  1.41it/s]
3|5031|Loss: 0.6877588033676147:  78%|███████▊  | 5031/6470 [1:02:20<18:26,  1.30it/s]
3|5032|Loss: 0.6966611742973328:  78%|███████▊  | 5031/6470 [1:02:20<18:26,  1.30it/s]
3|5032|Loss: 0.6966611742973328:  78%|███████▊  | 5032/6470 [1:02:21<19:49,  1.21it/s]
3|5033|Loss: 1.0058492422103882:  78%|███████▊  | 5032/6470 [1:02:21<19:49,  1.21it/s]
3|5033|Loss: 1.0058492422103882:  78%|███████▊  | 5033/6470 [1:02:22<19:18,  1.24it/s]
3|5034|Loss: 0.7640805840492249:  78%|███████▊  | 5033/6470 [1:02:22<19:18,  1.24it/s]
3|5034|Loss: 0.7640805840492249:  78%|███████▊  | 5034/6470 [1:02:22<19:03,  1.26it/s]
3|5035|Loss: 0.9208997488021851:  78%|███████▊  | 5034/6470 [1:02:23<19:03,  1.26it/s]
3|5035|Loss: 0.9208997488021851:  78%|███████▊  | 5035/6470 [1:02:23<18:24,  1.30it/s]
3|5036|Loss: 0.7579172849655151:  78%|███████▊  | 5035/6470 [1:02:23<18:24,  1.30it/s]
3|5036|Loss: 0.7579172849655151:  78%|███████▊  | 5036/6470 [1:02:24<18:15,  1.31it/s]
3|5037|Loss: 0.8416388034820557:  78%|███████▊  | 5036/6470 [1:02:24<18:15,  1.31it/s]
3|5037|Loss: 0.8416388034820557:  78%|███████▊  | 5037/6470 [1:02:25<19:04,  1.25it/s]
3|5038|Loss: 0.8425869941711426:  78%|███████▊  | 5037/6470 [1:02:25<19:04,  1.25it/s]
3|5038|Loss: 0.8425869941711426:  78%|███████▊  | 5038/6470 [1:02:26<19:53,  1.20it/s]
3|5039|Loss: 0.9150201082229614:  78%|███████▊  | 5038/6470 [1:02:26<19:53,  1.20it/s]
3|5039|Loss: 0.9150201082229614:  78%|███████▊  | 5039/6470 [1:02:26<19:48,  1.20it/s]
3|5040|Loss: 0.9019094109535217:  78%|███████▊  | 5039/6470 [1:02:27<19:48,  1.20it/s]
3|5040|Loss: 0.9019094109535217:  78%|███████▊  | 5040/6470 [1:02:27<19:06,  1.25it/s]
3|5041|Loss: 1.0058740377426147:  78%|███████▊  | 5040/6470 [1:02:28<19:06,  1.25it/s]
3|5041|Loss: 1.0058740377426147:  78%|███████▊  | 5041/6470 [1:02:28<18:09,  1.31it/s]
3|5042|Loss: 0.7523833513259888:  78%|███████▊  | 5041/6470 [1:02:28<18:09,  1.31it/s]
3|5042|Loss: 0.7523833513259888:  78%|███████▊  | 5042/6470 [1:02:29<18:46,  1.27it/s]
3|5043|Loss: 0.9196231961250305:  78%|███████▊  | 5042/6470 [1:02:29<18:46,  1.27it/s]
3|5043|Loss: 0.9196231961250305:  78%|███████▊  | 5043/6470 [1:02:30<19:02,  1.25it/s]
3|5044|Loss: 0.8263981342315674:  78%|███████▊  | 5043/6470 [1:02:30<19:02,  1.25it/s]
3|5044|Loss: 0.8263981342315674:  78%|███████▊  | 5044/6470 [1:02:30<19:24,  1.22it/s]
3|5045|Loss: 0.9292700290679932:  78%|███████▊  | 5044/6470 [1:02:31<19:24,  1.22it/s]
3|5045|Loss: 0.9292700290679932:  78%|███████▊  | 5045/6470 [1:02:31<20:07,  1.18it/s]
3|5046|Loss: 0.8560088276863098:  78%|███████▊  | 5045/6470 [1:02:32<20:07,  1.18it/s]
3|5046|Loss: 0.8560088276863098:  78%|███████▊  | 5046/6470 [1:02:32<19:15,  1.23it/s]
3|5047|Loss: 0.7928767204284668:  78%|███████▊  | 5046/6470 [1:02:32<19:15,  1.23it/s]
3|5047|Loss: 0.7928767204284668:  78%|███████▊  | 5047/6470 [1:02:33<19:26,  1.22it/s]
3|5048|Loss: 0.7758028507232666:  78%|███████▊  | 5047/6470 [1:02:33<19:26,  1.22it/s]
3|5048|Loss: 0.7758028507232666:  78%|███████▊  | 5048/6470 [1:02:34<19:55,  1.19it/s]
3|5049|Loss: 0.8795776963233948:  78%|███████▊  | 5048/6470 [1:02:34<19:55,  1.19it/s]
3|5049|Loss: 0.8795776963233948:  78%|███████▊  | 5049/6470 [1:02:34<19:07,  1.24it/s]
3|5050|Loss: 0.7132933139801025:  78%|███████▊  | 5049/6470 [1:02:35<19:07,  1.24it/s]
3|5050|Loss: 0.7132933139801025:  78%|███████▊  | 5050/6470 [1:02:35<18:38,  1.27it/s]
3|5051|Loss: 0.9050649404525757:  78%|███████▊  | 5050/6470 [1:02:36<18:38,  1.27it/s]
3|5051|Loss: 0.9050649404525757:  78%|███████▊  | 5051/6470 [1:02:36<19:16,  1.23it/s]
3|5052|Loss: 0.8708305954933167:  78%|███████▊  | 5051/6470 [1:02:37<19:16,  1.23it/s]
3|5052|Loss: 0.8708305954933167:  78%|███████▊  | 5052/6470 [1:02:37<19:40,  1.20it/s]
3|5053|Loss: 0.7066490054130554:  78%|███████▊  | 5052/6470 [1:02:37<19:40,  1.20it/s]
3|5053|Loss: 0.7066490054130554:  78%|███████▊  | 5053/6470 [1:02:38<18:32,  1.27it/s]
3|5054|Loss: 0.8042071461677551:  78%|███████▊  | 5053/6470 [1:02:38<18:32,  1.27it/s]
3|5054|Loss: 0.8042071461677551:  78%|███████▊  | 5054/6470 [1:02:38<17:23,  1.36it/s]
3|5055|Loss: 0.8055264353752136:  78%|███████▊  | 5054/6470 [1:02:39<17:23,  1.36it/s]
3|5055|Loss: 0.8055264353752136:  78%|███████▊  | 5055/6470 [1:02:39<16:32,  1.43it/s]
3|5056|Loss: 0.8299119472503662:  78%|███████▊  | 5055/6470 [1:02:39<16:32,  1.43it/s]
3|5056|Loss: 0.8299119472503662:  78%|███████▊  | 5056/6470 [1:02:40<16:02,  1.47it/s]
3|5057|Loss: 0.7926108241081238:  78%|███████▊  | 5056/6470 [1:02:40<16:02,  1.47it/s]
3|5057|Loss: 0.7926108241081238:  78%|███████▊  | 5057/6470 [1:02:40<16:39,  1.41it/s]
3|5058|Loss: 0.9521664977073669:  78%|███████▊  | 5057/6470 [1:02:41<16:39,  1.41it/s]
3|5058|Loss: 0.9521664977073669:  78%|███████▊  | 5058/6470 [1:02:41<18:04,  1.30it/s]
3|5059|Loss: 0.8766909837722778:  78%|███████▊  | 5058/6470 [1:02:42<18:04,  1.30it/s]
3|5059|Loss: 0.8766909837722778:  78%|███████▊  | 5059/6470 [1:02:42<17:55,  1.31it/s]
3|5060|Loss: 0.7858072519302368:  78%|███████▊  | 5059/6470 [1:02:42<17:55,  1.31it/s]
3|5060|Loss: 0.7858072519302368:  78%|███████▊  | 5060/6470 [1:02:42<16:09,  1.45it/s]
3|5061|Loss: 0.9053983092308044:  78%|███████▊  | 5060/6470 [1:02:43<16:09,  1.45it/s]
3|5061|Loss: 0.9053983092308044:  78%|███████▊  | 5061/6470 [1:02:43<17:12,  1.37it/s]
3|5062|Loss: 0.7839328050613403:  78%|███████▊  | 5061/6470 [1:02:44<17:12,  1.37it/s]
3|5062|Loss: 0.7839328050613403:  78%|███████▊  | 5062/6470 [1:02:44<17:49,  1.32it/s]
3|5063|Loss: 0.8219906687736511:  78%|███████▊  | 5062/6470 [1:02:44<17:49,  1.32it/s]
3|5063|Loss: 0.8219906687736511:  78%|███████▊  | 5063/6470 [1:02:45<16:20,  1.43it/s]
3|5064|Loss: 0.8965123295783997:  78%|███████▊  | 5063/6470 [1:02:45<16:20,  1.43it/s]
3|5064|Loss: 0.8965123295783997:  78%|███████▊  | 5064/6470 [1:02:46<17:24,  1.35it/s]
3|5065|Loss: 0.7720385193824768:  78%|███████▊  | 5064/6470 [1:02:46<17:24,  1.35it/s]
3|5065|Loss: 0.7720385193824768:  78%|███████▊  | 5065/6470 [1:02:46<16:35,  1.41it/s]
3|5066|Loss: 0.7287085652351379:  78%|███████▊  | 5065/6470 [1:02:47<16:35,  1.41it/s]
3|5066|Loss: 0.7287085652351379:  78%|███████▊  | 5066/6470 [1:02:47<17:25,  1.34it/s]
3|5067|Loss: 0.8106421828269958:  78%|███████▊  | 5066/6470 [1:02:47<17:25,  1.34it/s]
3|5067|Loss: 0.8106421828269958:  78%|███████▊  | 5067/6470 [1:02:48<16:42,  1.40it/s]
3|5068|Loss: 0.930854856967926:  78%|███████▊  | 5067/6470 [1:02:48<16:42,  1.40it/s] 
3|5068|Loss: 0.930854856967926:  78%|███████▊  | 5068/6470 [1:02:48<16:55,  1.38it/s]
3|5069|Loss: 0.7183954119682312:  78%|███████▊  | 5068/6470 [1:02:49<16:55,  1.38it/s]
3|5069|Loss: 0.7183954119682312:  78%|███████▊  | 5069/6470 [1:02:49<16:36,  1.41it/s]
3|5070|Loss: 0.9732784628868103:  78%|███████▊  | 5069/6470 [1:02:49<16:36,  1.41it/s]
3|5070|Loss: 0.9732784628868103:  78%|███████▊  | 5070/6470 [1:02:50<16:58,  1.37it/s]
3|5071|Loss: 0.8479735851287842:  78%|███████▊  | 5070/6470 [1:02:50<16:58,  1.37it/s]
3|5071|Loss: 0.8479735851287842:  78%|███████▊  | 5071/6470 [1:02:51<17:11,  1.36it/s]
3|5072|Loss: 0.7582388520240784:  78%|███████▊  | 5071/6470 [1:02:51<17:11,  1.36it/s]
3|5072|Loss: 0.7582388520240784:  78%|███████▊  | 5072/6470 [1:02:51<16:56,  1.38it/s]
3|5073|Loss: 0.7200000286102295:  78%|███████▊  | 5072/6470 [1:02:52<16:56,  1.38it/s]
3|5073|Loss: 0.7200000286102295:  78%|███████▊  | 5073/6470 [1:02:52<15:27,  1.51it/s]
3|5074|Loss: 0.796873927116394:  78%|███████▊  | 5073/6470 [1:02:52<15:27,  1.51it/s] 
3|5074|Loss: 0.796873927116394:  78%|███████▊  | 5074/6470 [1:02:52<15:34,  1.49it/s]
3|5075|Loss: 0.9727379679679871:  78%|███████▊  | 5074/6470 [1:02:53<15:34,  1.49it/s]
3|5075|Loss: 0.9727379679679871:  78%|███████▊  | 5075/6470 [1:02:53<16:21,  1.42it/s]
3|5076|Loss: 0.7653349041938782:  78%|███████▊  | 5075/6470 [1:02:54<16:21,  1.42it/s]
3|5076|Loss: 0.7653349041938782:  78%|███████▊  | 5076/6470 [1:02:54<16:58,  1.37it/s]
3|5077|Loss: 0.8332539200782776:  78%|███████▊  | 5076/6470 [1:02:54<16:58,  1.37it/s]
3|5077|Loss: 0.8332539200782776:  78%|███████▊  | 5077/6470 [1:02:55<17:22,  1.34it/s]
3|5078|Loss: 0.9079357385635376:  78%|███████▊  | 5077/6470 [1:02:55<17:22,  1.34it/s]
3|5078|Loss: 0.9079357385635376:  78%|███████▊  | 5078/6470 [1:02:56<17:41,  1.31it/s]
3|5079|Loss: 0.9069003462791443:  78%|███████▊  | 5078/6470 [1:02:56<17:41,  1.31it/s]
3|5079|Loss: 0.9069003462791443:  79%|███████▊  | 5079/6470 [1:02:57<18:29,  1.25it/s]
3|5080|Loss: 0.7542151212692261:  79%|███████▊  | 5079/6470 [1:02:57<18:29,  1.25it/s]
3|5080|Loss: 0.7542151212692261:  79%|███████▊  | 5080/6470 [1:02:57<17:36,  1.32it/s]
3|5081|Loss: 0.7132385969161987:  79%|███████▊  | 5080/6470 [1:02:58<17:36,  1.32it/s]
3|5081|Loss: 0.7132385969161987:  79%|███████▊  | 5081/6470 [1:02:58<16:11,  1.43it/s]
3|5082|Loss: 0.881699800491333:  79%|███████▊  | 5081/6470 [1:02:58<16:11,  1.43it/s] 
3|5082|Loss: 0.881699800491333:  79%|███████▊  | 5082/6470 [1:02:58<16:13,  1.43it/s]
3|5083|Loss: 0.7744100093841553:  79%|███████▊  | 5082/6470 [1:02:59<16:13,  1.43it/s]
3|5083|Loss: 0.7744100093841553:  79%|███████▊  | 5083/6470 [1:02:59<16:11,  1.43it/s]
3|5084|Loss: 0.7410758137702942:  79%|███████▊  | 5083/6470 [1:03:00<16:11,  1.43it/s]
3|5084|Loss: 0.7410758137702942:  79%|███████▊  | 5084/6470 [1:03:00<17:16,  1.34it/s]
3|5085|Loss: 0.9176437258720398:  79%|███████▊  | 5084/6470 [1:03:00<17:16,  1.34it/s]
3|5085|Loss: 0.9176437258720398:  79%|███████▊  | 5085/6470 [1:03:01<17:01,  1.36it/s]
3|5086|Loss: 0.8143522143363953:  79%|███████▊  | 5085/6470 [1:03:01<17:01,  1.36it/s]
3|5086|Loss: 0.8143522143363953:  79%|███████▊  | 5086/6470 [1:03:02<17:45,  1.30it/s]
3|5087|Loss: 0.9339561462402344:  79%|███████▊  | 5086/6470 [1:03:02<17:45,  1.30it/s]
3|5087|Loss: 0.9339561462402344:  79%|███████▊  | 5087/6470 [1:03:02<17:45,  1.30it/s]
3|5088|Loss: 0.7750357389450073:  79%|███████▊  | 5087/6470 [1:03:03<17:45,  1.30it/s]
3|5088|Loss: 0.7750357389450073:  79%|███████▊  | 5088/6470 [1:03:03<18:22,  1.25it/s]
3|5089|Loss: 1.0637602806091309:  79%|███████▊  | 5088/6470 [1:03:04<18:22,  1.25it/s]
3|5089|Loss: 1.0637602806091309:  79%|███████▊  | 5089/6470 [1:03:04<19:05,  1.21it/s]
3|5090|Loss: 0.7332724928855896:  79%|███████▊  | 5089/6470 [1:03:05<19:05,  1.21it/s]
3|5090|Loss: 0.7332724928855896:  79%|███████▊  | 5090/6470 [1:03:05<18:22,  1.25it/s]
3|5091|Loss: 0.9004297256469727:  79%|███████▊  | 5090/6470 [1:03:05<18:22,  1.25it/s]
3|5091|Loss: 0.9004297256469727:  79%|███████▊  | 5091/6470 [1:03:06<18:54,  1.22it/s]
3|5092|Loss: 0.9314936399459839:  79%|███████▊  | 5091/6470 [1:03:06<18:54,  1.22it/s]
3|5092|Loss: 0.9314936399459839:  79%|███████▊  | 5092/6470 [1:03:06<18:23,  1.25it/s]
3|5093|Loss: 0.8659141063690186:  79%|███████▊  | 5092/6470 [1:03:07<18:23,  1.25it/s]
3|5093|Loss: 0.8659141063690186:  79%|███████▊  | 5093/6470 [1:03:07<17:40,  1.30it/s]
3|5094|Loss: 0.7863880395889282:  79%|███████▊  | 5093/6470 [1:03:07<17:40,  1.30it/s]
3|5094|Loss: 0.7863880395889282:  79%|███████▊  | 5094/6470 [1:03:08<15:33,  1.47it/s]
3|5095|Loss: 0.9136302471160889:  79%|███████▊  | 5094/6470 [1:03:08<15:33,  1.47it/s]
3|5095|Loss: 0.9136302471160889:  79%|███████▊  | 5095/6470 [1:03:08<15:19,  1.50it/s]
3|5096|Loss: 1.0298502445220947:  79%|███████▊  | 5095/6470 [1:03:09<15:19,  1.50it/s]
3|5096|Loss: 1.0298502445220947:  79%|███████▉  | 5096/6470 [1:03:09<16:46,  1.36it/s]
3|5097|Loss: 0.7037244439125061:  79%|███████▉  | 5096/6470 [1:03:10<16:46,  1.36it/s]
3|5097|Loss: 0.7037244439125061:  79%|███████▉  | 5097/6470 [1:03:10<18:16,  1.25it/s]
3|5098|Loss: 0.7180365920066833:  79%|███████▉  | 5097/6470 [1:03:11<18:16,  1.25it/s]
3|5098|Loss: 0.7180365920066833:  79%|███████▉  | 5098/6470 [1:03:11<19:16,  1.19it/s]
3|5099|Loss: 0.9634770750999451:  79%|███████▉  | 5098/6470 [1:03:12<19:16,  1.19it/s]
3|5099|Loss: 0.9634770750999451:  79%|███████▉  | 5099/6470 [1:03:12<19:50,  1.15it/s]
3|5100|Loss: 0.794955849647522:  79%|███████▉  | 5099/6470 [1:03:12<19:50,  1.15it/s] 
3|5100|Loss: 0.794955849647522:  79%|███████▉  | 5100/6470 [1:03:13<18:47,  1.22it/s]
3|5101|Loss: 1.0032031536102295:  79%|███████▉  | 5100/6470 [1:03:13<18:47,  1.22it/s]
3|5101|Loss: 1.0032031536102295:  79%|███████▉  | 5101/6470 [1:03:14<19:13,  1.19it/s]
3|5102|Loss: 0.8093323111534119:  79%|███████▉  | 5101/6470 [1:03:14<19:13,  1.19it/s]
3|5102|Loss: 0.8093323111534119:  79%|███████▉  | 5102/6470 [1:03:15<20:19,  1.12it/s]
3|5103|Loss: 0.7657323479652405:  79%|███████▉  | 5102/6470 [1:03:15<20:19,  1.12it/s]
3|5103|Loss: 0.7657323479652405:  79%|███████▉  | 5103/6470 [1:03:15<17:54,  1.27it/s]
3|5104|Loss: 0.8972272276878357:  79%|███████▉  | 5103/6470 [1:03:15<17:54,  1.27it/s]
3|5104|Loss: 0.8972272276878357:  79%|███████▉  | 5104/6470 [1:03:16<17:26,  1.31it/s]
3|5105|Loss: 0.9276973605155945:  79%|███████▉  | 5104/6470 [1:03:16<17:26,  1.31it/s]
3|5105|Loss: 0.9276973605155945:  79%|███████▉  | 5105/6470 [1:03:16<16:28,  1.38it/s]
3|5106|Loss: 0.8731986880302429:  79%|███████▉  | 5105/6470 [1:03:17<16:28,  1.38it/s]
3|5106|Loss: 0.8731986880302429:  79%|███████▉  | 5106/6470 [1:03:17<16:23,  1.39it/s]
3|5107|Loss: 0.85595703125:  79%|███████▉  | 5106/6470 [1:03:18<16:23,  1.39it/s]     
3|5107|Loss: 0.85595703125:  79%|███████▉  | 5107/6470 [1:03:18<15:58,  1.42it/s]
3|5108|Loss: 0.8203070759773254:  79%|███████▉  | 5107/6470 [1:03:18<15:58,  1.42it/s]
3|5108|Loss: 0.8203070759773254:  79%|███████▉  | 5108/6470 [1:03:19<15:38,  1.45it/s]
3|5109|Loss: 0.7572630643844604:  79%|███████▉  | 5108/6470 [1:03:19<15:38,  1.45it/s]
3|5109|Loss: 0.7572630643844604:  79%|███████▉  | 5109/6470 [1:03:19<16:06,  1.41it/s]
3|5110|Loss: 0.8497959971427917:  79%|███████▉  | 5109/6470 [1:03:20<16:06,  1.41it/s]
3|5110|Loss: 0.8497959971427917:  79%|███████▉  | 5110/6470 [1:03:20<17:22,  1.30it/s]
3|5111|Loss: 0.7183672189712524:  79%|███████▉  | 5110/6470 [1:03:21<17:22,  1.30it/s]
3|5111|Loss: 0.7183672189712524:  79%|███████▉  | 5111/6470 [1:03:21<17:36,  1.29it/s]
3|5112|Loss: 0.7948609590530396:  79%|███████▉  | 5111/6470 [1:03:21<17:36,  1.29it/s]
3|5112|Loss: 0.7948609590530396:  79%|███████▉  | 5112/6470 [1:03:22<17:00,  1.33it/s]
3|5113|Loss: 0.7621690034866333:  79%|███████▉  | 5112/6470 [1:03:22<17:00,  1.33it/s]
3|5113|Loss: 0.7621690034866333:  79%|███████▉  | 5113/6470 [1:03:22<16:27,  1.37it/s]
3|5114|Loss: 0.7040143013000488:  79%|███████▉  | 5113/6470 [1:03:23<16:27,  1.37it/s]
3|5114|Loss: 0.7040143013000488:  79%|███████▉  | 5114/6470 [1:03:23<16:39,  1.36it/s]
3|5115|Loss: 0.7924277186393738:  79%|███████▉  | 5114/6470 [1:03:23<16:39,  1.36it/s]
3|5115|Loss: 0.7924277186393738:  79%|███████▉  | 5115/6470 [1:03:24<15:19,  1.47it/s]
3|5116|Loss: 0.9132769703865051:  79%|███████▉  | 5115/6470 [1:03:24<15:19,  1.47it/s]
3|5116|Loss: 0.9132769703865051:  79%|███████▉  | 5116/6470 [1:03:24<15:06,  1.49it/s]
3|5117|Loss: 0.8635663986206055:  79%|███████▉  | 5116/6470 [1:03:25<15:06,  1.49it/s]
3|5117|Loss: 0.8635663986206055:  79%|███████▉  | 5117/6470 [1:03:25<16:24,  1.37it/s]
3|5118|Loss: 0.7618062496185303:  79%|███████▉  | 5117/6470 [1:03:26<16:24,  1.37it/s]
3|5118|Loss: 0.7618062496185303:  79%|███████▉  | 5118/6470 [1:03:26<16:42,  1.35it/s]
3|5119|Loss: 0.8973748087882996:  79%|███████▉  | 5118/6470 [1:03:26<16:42,  1.35it/s]
3|5119|Loss: 0.8973748087882996:  79%|███████▉  | 5119/6470 [1:03:26<15:35,  1.44it/s]
3|5120|Loss: 0.9234194755554199:  79%|███████▉  | 5119/6470 [1:03:27<15:35,  1.44it/s]
3|5120|Loss: 0.9234194755554199:  79%|███████▉  | 5120/6470 [1:03:27<16:36,  1.35it/s]
3|5121|Loss: 0.9013481736183167:  79%|███████▉  | 5120/6470 [1:03:28<16:36,  1.35it/s]
3|5121|Loss: 0.9013481736183167:  79%|███████▉  | 5121/6470 [1:03:28<17:17,  1.30it/s]
3|5122|Loss: 0.7539317607879639:  79%|███████▉  | 5121/6470 [1:03:29<17:17,  1.30it/s]
3|5122|Loss: 0.7539317607879639:  79%|███████▉  | 5122/6470 [1:03:29<17:20,  1.30it/s]
3|5123|Loss: 0.8309102058410645:  79%|███████▉  | 5122/6470 [1:03:29<17:20,  1.30it/s]
3|5123|Loss: 0.8309102058410645:  79%|███████▉  | 5123/6470 [1:03:30<16:34,  1.35it/s]
3|5124|Loss: 0.9709982872009277:  79%|███████▉  | 5123/6470 [1:03:30<16:34,  1.35it/s]
3|5124|Loss: 0.9709982872009277:  79%|███████▉  | 5124/6470 [1:03:30<17:27,  1.28it/s]
3|5125|Loss: 0.8207618594169617:  79%|███████▉  | 5124/6470 [1:03:31<17:27,  1.28it/s]
3|5125|Loss: 0.8207618594169617:  79%|███████▉  | 5125/6470 [1:03:31<18:21,  1.22it/s]
3|5126|Loss: 0.9805579781532288:  79%|███████▉  | 5125/6470 [1:03:32<18:21,  1.22it/s]
3|5126|Loss: 0.9805579781532288:  79%|███████▉  | 5126/6470 [1:03:32<17:35,  1.27it/s]
3|5127|Loss: 0.8011376261711121:  79%|███████▉  | 5126/6470 [1:03:33<17:35,  1.27it/s]
3|5127|Loss: 0.8011376261711121:  79%|███████▉  | 5127/6470 [1:03:33<18:25,  1.22it/s]
3|5128|Loss: 0.8803907036781311:  79%|███████▉  | 5127/6470 [1:03:34<18:25,  1.22it/s]
3|5128|Loss: 0.8803907036781311:  79%|███████▉  | 5128/6470 [1:03:34<19:40,  1.14it/s]
3|5129|Loss: 0.9369216561317444:  79%|███████▉  | 5128/6470 [1:03:35<19:40,  1.14it/s]
3|5129|Loss: 0.9369216561317444:  79%|███████▉  | 5129/6470 [1:03:35<20:18,  1.10it/s]
3|5130|Loss: 0.8364725708961487:  79%|███████▉  | 5129/6470 [1:03:36<20:18,  1.10it/s]
3|5130|Loss: 0.8364725708961487:  79%|███████▉  | 5130/6470 [1:03:36<20:13,  1.10it/s]
3|5131|Loss: 0.7643022537231445:  79%|███████▉  | 5130/6470 [1:03:36<20:13,  1.10it/s]
3|5131|Loss: 0.7643022537231445:  79%|███████▉  | 5131/6470 [1:03:37<18:29,  1.21it/s]
3|5132|Loss: 0.683670699596405:  79%|███████▉  | 5131/6470 [1:03:37<18:29,  1.21it/s] 
3|5132|Loss: 0.683670699596405:  79%|███████▉  | 5132/6470 [1:03:37<16:40,  1.34it/s]
3|5133|Loss: 0.6555424928665161:  79%|███████▉  | 5132/6470 [1:03:37<16:40,  1.34it/s]
3|5133|Loss: 0.6555424928665161:  79%|███████▉  | 5133/6470 [1:03:38<14:22,  1.55it/s]
3|5134|Loss: 0.939378023147583:  79%|███████▉  | 5133/6470 [1:03:38<14:22,  1.55it/s] 
3|5134|Loss: 0.939378023147583:  79%|███████▉  | 5134/6470 [1:03:38<15:33,  1.43it/s]
3|5135|Loss: 0.7005364894866943:  79%|███████▉  | 5134/6470 [1:03:39<15:33,  1.43it/s]
3|5135|Loss: 0.7005364894866943:  79%|███████▉  | 5135/6470 [1:03:39<15:28,  1.44it/s]
3|5136|Loss: 0.6385101079940796:  79%|███████▉  | 5135/6470 [1:03:40<15:28,  1.44it/s]
3|5136|Loss: 0.6385101079940796:  79%|███████▉  | 5136/6470 [1:03:40<17:13,  1.29it/s]
3|5137|Loss: 0.7930976152420044:  79%|███████▉  | 5136/6470 [1:03:40<17:13,  1.29it/s]
3|5137|Loss: 0.7930976152420044:  79%|███████▉  | 5137/6470 [1:03:41<16:44,  1.33it/s]
3|5138|Loss: 0.8875045776367188:  79%|███████▉  | 5137/6470 [1:03:41<16:44,  1.33it/s]
3|5138|Loss: 0.8875045776367188:  79%|███████▉  | 5138/6470 [1:03:42<17:53,  1.24it/s]
3|5139|Loss: 0.7462906241416931:  79%|███████▉  | 5138/6470 [1:03:42<17:53,  1.24it/s]
3|5139|Loss: 0.7462906241416931:  79%|███████▉  | 5139/6470 [1:03:42<15:57,  1.39it/s]
3|5140|Loss: 0.7204989790916443:  79%|███████▉  | 5139/6470 [1:03:43<15:57,  1.39it/s]
3|5140|Loss: 0.7204989790916443:  79%|███████▉  | 5140/6470 [1:03:43<15:50,  1.40it/s]
3|5141|Loss: 0.8358637690544128:  79%|███████▉  | 5140/6470 [1:03:43<15:50,  1.40it/s]
3|5141|Loss: 0.8358637690544128:  79%|███████▉  | 5141/6470 [1:03:44<15:46,  1.40it/s]
3|5142|Loss: 0.7551313042640686:  79%|███████▉  | 5141/6470 [1:03:44<15:46,  1.40it/s]
3|5142|Loss: 0.7551313042640686:  79%|███████▉  | 5142/6470 [1:03:44<15:13,  1.45it/s]
3|5143|Loss: 0.8421469926834106:  79%|███████▉  | 5142/6470 [1:03:45<15:13,  1.45it/s]
3|5143|Loss: 0.8421469926834106:  79%|███████▉  | 5143/6470 [1:03:45<15:35,  1.42it/s]
3|5144|Loss: 0.8552562594413757:  79%|███████▉  | 5143/6470 [1:03:45<15:35,  1.42it/s]
3|5144|Loss: 0.8552562594413757:  80%|███████▉  | 5144/6470 [1:03:46<16:31,  1.34it/s]
3|5145|Loss: 0.6279763579368591:  80%|███████▉  | 5144/6470 [1:03:46<16:31,  1.34it/s]
3|5145|Loss: 0.6279763579368591:  80%|███████▉  | 5145/6470 [1:03:46<16:05,  1.37it/s]
3|5146|Loss: 0.6378387808799744:  80%|███████▉  | 5145/6470 [1:03:47<16:05,  1.37it/s]
3|5146|Loss: 0.6378387808799744:  80%|███████▉  | 5146/6470 [1:03:47<15:23,  1.43it/s]
3|5147|Loss: 0.8166822791099548:  80%|███████▉  | 5146/6470 [1:03:47<15:23,  1.43it/s]
3|5147|Loss: 0.8166822791099548:  80%|███████▉  | 5147/6470 [1:03:48<14:49,  1.49it/s]
3|5148|Loss: 0.7471917271614075:  80%|███████▉  | 5147/6470 [1:03:48<14:49,  1.49it/s]
3|5148|Loss: 0.7471917271614075:  80%|███████▉  | 5148/6470 [1:03:49<15:45,  1.40it/s]
3|5149|Loss: 0.8503138422966003:  80%|███████▉  | 5148/6470 [1:03:49<15:45,  1.40it/s]
3|5149|Loss: 0.8503138422966003:  80%|███████▉  | 5149/6470 [1:03:49<16:07,  1.37it/s]
3|5150|Loss: 0.8259687423706055:  80%|███████▉  | 5149/6470 [1:03:50<16:07,  1.37it/s]
3|5150|Loss: 0.8259687423706055:  80%|███████▉  | 5150/6470 [1:03:50<16:34,  1.33it/s]
3|5151|Loss: 0.8233529925346375:  80%|███████▉  | 5150/6470 [1:03:50<16:34,  1.33it/s]
3|5151|Loss: 0.8233529925346375:  80%|███████▉  | 5151/6470 [1:03:51<16:15,  1.35it/s]
3|5152|Loss: 0.7808110117912292:  80%|███████▉  | 5151/6470 [1:03:51<16:15,  1.35it/s]
3|5152|Loss: 0.7808110117912292:  80%|███████▉  | 5152/6470 [1:03:52<17:16,  1.27it/s]
3|5153|Loss: 0.8600693941116333:  80%|███████▉  | 5152/6470 [1:03:52<17:16,  1.27it/s]
3|5153|Loss: 0.8600693941116333:  80%|███████▉  | 5153/6470 [1:03:52<16:58,  1.29it/s]
3|5154|Loss: 0.7017915844917297:  80%|███████▉  | 5153/6470 [1:03:53<16:58,  1.29it/s]
3|5154|Loss: 0.7017915844917297:  80%|███████▉  | 5154/6470 [1:03:53<16:53,  1.30it/s]
3|5155|Loss: 0.8672304749488831:  80%|███████▉  | 5154/6470 [1:03:54<16:53,  1.30it/s]
3|5155|Loss: 0.8672304749488831:  80%|███████▉  | 5155/6470 [1:03:54<17:00,  1.29it/s]
3|5156|Loss: 0.8609228134155273:  80%|███████▉  | 5155/6470 [1:03:54<17:00,  1.29it/s]
3|5156|Loss: 0.8609228134155273:  80%|███████▉  | 5156/6470 [1:03:55<17:08,  1.28it/s]
3|5157|Loss: 0.7390333414077759:  80%|███████▉  | 5156/6470 [1:03:55<17:08,  1.28it/s]
3|5157|Loss: 0.7390333414077759:  80%|███████▉  | 5157/6470 [1:03:55<16:17,  1.34it/s]
3|5158|Loss: 0.8013985753059387:  80%|███████▉  | 5157/6470 [1:03:56<16:17,  1.34it/s]
3|5158|Loss: 0.8013985753059387:  80%|███████▉  | 5158/6470 [1:03:56<17:05,  1.28it/s]
3|5159|Loss: 0.9300932288169861:  80%|███████▉  | 5158/6470 [1:03:57<17:05,  1.28it/s]
3|5159|Loss: 0.9300932288169861:  80%|███████▉  | 5159/6470 [1:03:57<16:41,  1.31it/s]
3|5160|Loss: 0.9276220202445984:  80%|███████▉  | 5159/6470 [1:03:57<16:41,  1.31it/s]
3|5160|Loss: 0.9276220202445984:  80%|███████▉  | 5160/6470 [1:03:58<15:37,  1.40it/s]
3|5161|Loss: 0.9226872324943542:  80%|███████▉  | 5160/6470 [1:03:58<15:37,  1.40it/s]
3|5161|Loss: 0.9226872324943542:  80%|███████▉  | 5161/6470 [1:03:58<15:41,  1.39it/s]
3|5162|Loss: 1.0014054775238037:  80%|███████▉  | 5161/6470 [1:03:59<15:41,  1.39it/s]
3|5162|Loss: 1.0014054775238037:  80%|███████▉  | 5162/6470 [1:03:59<15:32,  1.40it/s]
3|5163|Loss: 0.7838465571403503:  80%|███████▉  | 5162/6470 [1:03:59<15:32,  1.40it/s]
3|5163|Loss: 0.7838465571403503:  80%|███████▉  | 5163/6470 [1:04:00<14:33,  1.50it/s]
3|5164|Loss: 0.759190022945404:  80%|███████▉  | 5163/6470 [1:04:00<14:33,  1.50it/s] 
3|5164|Loss: 0.759190022945404:  80%|███████▉  | 5164/6470 [1:04:00<13:57,  1.56it/s]
3|5165|Loss: 0.7124514579772949:  80%|███████▉  | 5164/6470 [1:04:01<13:57,  1.56it/s]
3|5165|Loss: 0.7124514579772949:  80%|███████▉  | 5165/6470 [1:04:01<15:46,  1.38it/s]
3|5166|Loss: 1.0252550840377808:  80%|███████▉  | 5165/6470 [1:04:02<15:46,  1.38it/s]
3|5166|Loss: 1.0252550840377808:  80%|███████▉  | 5166/6470 [1:04:02<16:33,  1.31it/s]
3|5167|Loss: 0.8095020055770874:  80%|███████▉  | 5166/6470 [1:04:02<16:33,  1.31it/s]
3|5167|Loss: 0.8095020055770874:  80%|███████▉  | 5167/6470 [1:04:03<16:20,  1.33it/s]
3|5168|Loss: 0.7387167811393738:  80%|███████▉  | 5167/6470 [1:04:03<16:20,  1.33it/s]
3|5168|Loss: 0.7387167811393738:  80%|███████▉  | 5168/6470 [1:04:04<17:04,  1.27it/s]
3|5169|Loss: 0.8786789774894714:  80%|███████▉  | 5168/6470 [1:04:04<17:04,  1.27it/s]
3|5169|Loss: 0.8786789774894714:  80%|███████▉  | 5169/6470 [1:04:04<16:45,  1.29it/s]
3|5170|Loss: 0.8046268820762634:  80%|███████▉  | 5169/6470 [1:04:05<16:45,  1.29it/s]
3|5170|Loss: 0.8046268820762634:  80%|███████▉  | 5170/6470 [1:04:05<16:35,  1.31it/s]
3|5171|Loss: 0.6720376014709473:  80%|███████▉  | 5170/6470 [1:04:05<16:35,  1.31it/s]
3|5171|Loss: 0.6720376014709473:  80%|███████▉  | 5171/6470 [1:04:06<16:31,  1.31it/s]
3|5172|Loss: 0.9317800402641296:  80%|███████▉  | 5171/6470 [1:04:06<16:31,  1.31it/s]
3|5172|Loss: 0.9317800402641296:  80%|███████▉  | 5172/6470 [1:04:07<16:11,  1.34it/s]
3|5173|Loss: 0.8481850028038025:  80%|███████▉  | 5172/6470 [1:04:07<16:11,  1.34it/s]
3|5173|Loss: 0.8481850028038025:  80%|███████▉  | 5173/6470 [1:04:07<15:57,  1.36it/s]
3|5174|Loss: 0.9276114106178284:  80%|███████▉  | 5173/6470 [1:04:08<15:57,  1.36it/s]
3|5174|Loss: 0.9276114106178284:  80%|███████▉  | 5174/6470 [1:04:08<15:17,  1.41it/s]
3|5175|Loss: 0.8534371256828308:  80%|███████▉  | 5174/6470 [1:04:08<15:17,  1.41it/s]
3|5175|Loss: 0.8534371256828308:  80%|███████▉  | 5175/6470 [1:04:09<15:58,  1.35it/s]
3|5176|Loss: 0.8737507462501526:  80%|███████▉  | 5175/6470 [1:04:09<15:58,  1.35it/s]
3|5176|Loss: 0.8737507462501526:  80%|████████  | 5176/6470 [1:04:09<14:56,  1.44it/s]
3|5177|Loss: 0.9141121506690979:  80%|████████  | 5176/6470 [1:04:10<14:56,  1.44it/s]
3|5177|Loss: 0.9141121506690979:  80%|████████  | 5177/6470 [1:04:10<14:40,  1.47it/s]
3|5178|Loss: 0.825873851776123:  80%|████████  | 5177/6470 [1:04:10<14:40,  1.47it/s] 
3|5178|Loss: 0.825873851776123:  80%|████████  | 5178/6470 [1:04:11<14:39,  1.47it/s]
3|5179|Loss: 0.778798520565033:  80%|████████  | 5178/6470 [1:04:11<14:39,  1.47it/s]
3|5179|Loss: 0.778798520565033:  80%|████████  | 5179/6470 [1:04:11<13:59,  1.54it/s]
3|5180|Loss: 0.818530797958374:  80%|████████  | 5179/6470 [1:04:12<13:59,  1.54it/s]
3|5180|Loss: 0.818530797958374:  80%|████████  | 5180/6470 [1:04:12<15:15,  1.41it/s]
3|5181|Loss: 0.8564828634262085:  80%|████████  | 5180/6470 [1:04:12<15:15,  1.41it/s]
3|5181|Loss: 0.8564828634262085:  80%|████████  | 5181/6470 [1:04:13<15:30,  1.39it/s]
3|5182|Loss: 0.8051408529281616:  80%|████████  | 5181/6470 [1:04:13<15:30,  1.39it/s]
3|5182|Loss: 0.8051408529281616:  80%|████████  | 5182/6470 [1:04:14<16:11,  1.33it/s]
3|5183|Loss: 0.7616411447525024:  80%|████████  | 5182/6470 [1:04:14<16:11,  1.33it/s]
3|5183|Loss: 0.7616411447525024:  80%|████████  | 5183/6470 [1:04:14<15:37,  1.37it/s]
3|5184|Loss: 0.8544345498085022:  80%|████████  | 5183/6470 [1:04:15<15:37,  1.37it/s]
3|5184|Loss: 0.8544345498085022:  80%|████████  | 5184/6470 [1:04:15<14:57,  1.43it/s]
3|5185|Loss: 0.7104573845863342:  80%|████████  | 5184/6470 [1:04:15<14:57,  1.43it/s]
3|5185|Loss: 0.7104573845863342:  80%|████████  | 5185/6470 [1:04:16<14:39,  1.46it/s]
3|5186|Loss: 0.7892960906028748:  80%|████████  | 5185/6470 [1:04:16<14:39,  1.46it/s]
3|5186|Loss: 0.7892960906028748:  80%|████████  | 5186/6470 [1:04:16<15:42,  1.36it/s]
3|5187|Loss: 0.8509138822555542:  80%|████████  | 5186/6470 [1:04:17<15:42,  1.36it/s]
3|5187|Loss: 0.8509138822555542:  80%|████████  | 5187/6470 [1:04:17<16:14,  1.32it/s]
3|5188|Loss: 0.814486563205719:  80%|████████  | 5187/6470 [1:04:18<16:14,  1.32it/s] 
3|5188|Loss: 0.814486563205719:  80%|████████  | 5188/6470 [1:04:18<16:57,  1.26it/s]
3|5189|Loss: 0.8567777872085571:  80%|████████  | 5188/6470 [1:04:19<16:57,  1.26it/s]
3|5189|Loss: 0.8567777872085571:  80%|████████  | 5189/6470 [1:04:19<17:07,  1.25it/s]
3|5190|Loss: 0.746760368347168:  80%|████████  | 5189/6470 [1:04:19<17:07,  1.25it/s] 
3|5190|Loss: 0.746760368347168:  80%|████████  | 5190/6470 [1:04:20<15:53,  1.34it/s]
3|5191|Loss: 0.9444187879562378:  80%|████████  | 5190/6470 [1:04:20<15:53,  1.34it/s]
3|5191|Loss: 0.9444187879562378:  80%|████████  | 5191/6470 [1:04:20<15:22,  1.39it/s]
3|5192|Loss: 0.8938124179840088:  80%|████████  | 5191/6470 [1:04:21<15:22,  1.39it/s]
3|5192|Loss: 0.8938124179840088:  80%|████████  | 5192/6470 [1:04:21<14:52,  1.43it/s]
3|5193|Loss: 0.9180957078933716:  80%|████████  | 5192/6470 [1:04:21<14:52,  1.43it/s]
3|5193|Loss: 0.9180957078933716:  80%|████████  | 5193/6470 [1:04:22<14:41,  1.45it/s]
3|5194|Loss: 0.912478506565094:  80%|████████  | 5193/6470 [1:04:22<14:41,  1.45it/s] 
3|5194|Loss: 0.912478506565094:  80%|████████  | 5194/6470 [1:04:22<14:38,  1.45it/s]
3|5195|Loss: 0.8911075592041016:  80%|████████  | 5194/6470 [1:04:23<14:38,  1.45it/s]
3|5195|Loss: 0.8911075592041016:  80%|████████  | 5195/6470 [1:04:23<14:41,  1.45it/s]
3|5196|Loss: 0.6261012554168701:  80%|████████  | 5195/6470 [1:04:23<14:41,  1.45it/s]
3|5196|Loss: 0.6261012554168701:  80%|████████  | 5196/6470 [1:04:24<15:17,  1.39it/s]
3|5197|Loss: 0.8507899641990662:  80%|████████  | 5196/6470 [1:04:24<15:17,  1.39it/s]
3|5197|Loss: 0.8507899641990662:  80%|████████  | 5197/6470 [1:04:24<15:23,  1.38it/s]
3|5198|Loss: 1.002607822418213:  80%|████████  | 5197/6470 [1:04:25<15:23,  1.38it/s] 
3|5198|Loss: 1.002607822418213:  80%|████████  | 5198/6470 [1:04:25<16:22,  1.30it/s]
3|5199|Loss: 0.9741684198379517:  80%|████████  | 5198/6470 [1:04:26<16:22,  1.30it/s]
3|5199|Loss: 0.9741684198379517:  80%|████████  | 5199/6470 [1:04:26<17:19,  1.22it/s]
3|5200|Loss: 0.8488466739654541:  80%|████████  | 5199/6470 [1:04:27<17:19,  1.22it/s]
3|5200|Loss: 0.8488466739654541:  80%|████████  | 5200/6470 [1:04:27<16:21,  1.29it/s]
3|5201|Loss: 0.7778286337852478:  80%|████████  | 5200/6470 [1:04:27<16:21,  1.29it/s]
3|5201|Loss: 0.7778286337852478:  80%|████████  | 5201/6470 [1:04:28<15:16,  1.38it/s]
3|5202|Loss: 0.8840672969818115:  80%|████████  | 5201/6470 [1:04:28<15:16,  1.38it/s]
3|5202|Loss: 0.8840672969818115:  80%|████████  | 5202/6470 [1:04:28<14:20,  1.47it/s]
3|5203|Loss: 0.6988169550895691:  80%|████████  | 5202/6470 [1:04:28<14:20,  1.47it/s]
3|5203|Loss: 0.6988169550895691:  80%|████████  | 5203/6470 [1:04:29<14:49,  1.42it/s]
3|5204|Loss: 0.7657420039176941:  80%|████████  | 5203/6470 [1:04:29<14:49,  1.42it/s]
3|5204|Loss: 0.7657420039176941:  80%|████████  | 5204/6470 [1:04:29<14:11,  1.49it/s]
3|5205|Loss: 0.9388301372528076:  80%|████████  | 5204/6470 [1:04:30<14:11,  1.49it/s]
3|5205|Loss: 0.9388301372528076:  80%|████████  | 5205/6470 [1:04:30<13:58,  1.51it/s]
3|5206|Loss: 0.7787985801696777:  80%|████████  | 5205/6470 [1:04:31<13:58,  1.51it/s]
3|5206|Loss: 0.7787985801696777:  80%|████████  | 5206/6470 [1:04:31<14:59,  1.41it/s]
3|5207|Loss: 0.8335462212562561:  80%|████████  | 5206/6470 [1:04:31<14:59,  1.41it/s]
3|5207|Loss: 0.8335462212562561:  80%|████████  | 5207/6470 [1:04:32<15:59,  1.32it/s]
3|5208|Loss: 0.8436339497566223:  80%|████████  | 5207/6470 [1:04:32<15:59,  1.32it/s]
3|5208|Loss: 0.8436339497566223:  80%|████████  | 5208/6470 [1:04:33<16:14,  1.29it/s]
3|5209|Loss: 0.7881878614425659:  80%|████████  | 5208/6470 [1:04:33<16:14,  1.29it/s]
3|5209|Loss: 0.7881878614425659:  81%|████████  | 5209/6470 [1:04:33<15:03,  1.40it/s]
3|5210|Loss: 0.664243221282959:  81%|████████  | 5209/6470 [1:04:34<15:03,  1.40it/s] 
3|5210|Loss: 0.664243221282959:  81%|████████  | 5210/6470 [1:04:34<15:12,  1.38it/s]
3|5211|Loss: 0.8500972986221313:  81%|████████  | 5210/6470 [1:04:34<15:12,  1.38it/s]
3|5211|Loss: 0.8500972986221313:  81%|████████  | 5211/6470 [1:04:35<15:26,  1.36it/s]
3|5212|Loss: 0.7097822427749634:  81%|████████  | 5211/6470 [1:04:35<15:26,  1.36it/s]
3|5212|Loss: 0.7097822427749634:  81%|████████  | 5212/6470 [1:04:35<15:45,  1.33it/s]
3|5213|Loss: 0.8718501329421997:  81%|████████  | 5212/6470 [1:04:36<15:45,  1.33it/s]
3|5213|Loss: 0.8718501329421997:  81%|████████  | 5213/6470 [1:04:36<14:43,  1.42it/s]
3|5214|Loss: 0.7384893298149109:  81%|████████  | 5213/6470 [1:04:36<14:43,  1.42it/s]
3|5214|Loss: 0.7384893298149109:  81%|████████  | 5214/6470 [1:04:37<15:37,  1.34it/s]
3|5215|Loss: 0.8498055338859558:  81%|████████  | 5214/6470 [1:04:37<15:37,  1.34it/s]
3|5215|Loss: 0.8498055338859558:  81%|████████  | 5215/6470 [1:04:38<14:54,  1.40it/s]
3|5216|Loss: 0.9068508148193359:  81%|████████  | 5215/6470 [1:04:38<14:54,  1.40it/s]
3|5216|Loss: 0.9068508148193359:  81%|████████  | 5216/6470 [1:04:38<14:50,  1.41it/s]
3|5217|Loss: 0.8367226123809814:  81%|████████  | 5216/6470 [1:04:39<14:50,  1.41it/s]
3|5217|Loss: 0.8367226123809814:  81%|████████  | 5217/6470 [1:04:39<15:38,  1.33it/s]
3|5218|Loss: 0.8824684619903564:  81%|████████  | 5217/6470 [1:04:40<15:38,  1.33it/s]
3|5218|Loss: 0.8824684619903564:  81%|████████  | 5218/6470 [1:04:40<16:48,  1.24it/s]
3|5219|Loss: 0.845597505569458:  81%|████████  | 5218/6470 [1:04:40<16:48,  1.24it/s] 
3|5219|Loss: 0.845597505569458:  81%|████████  | 5219/6470 [1:04:41<16:32,  1.26it/s]
3|5220|Loss: 0.7125495672225952:  81%|████████  | 5219/6470 [1:04:41<16:32,  1.26it/s]
3|5220|Loss: 0.7125495672225952:  81%|████████  | 5220/6470 [1:04:42<16:30,  1.26it/s]
3|5221|Loss: 0.9432005286216736:  81%|████████  | 5220/6470 [1:04:42<16:30,  1.26it/s]
3|5221|Loss: 0.9432005286216736:  81%|████████  | 5221/6470 [1:04:42<16:58,  1.23it/s]
3|5222|Loss: 0.805290162563324:  81%|████████  | 5221/6470 [1:04:43<16:58,  1.23it/s] 
3|5222|Loss: 0.805290162563324:  81%|████████  | 5222/6470 [1:04:43<16:15,  1.28it/s]
3|5223|Loss: 0.9690608978271484:  81%|████████  | 5222/6470 [1:04:43<16:15,  1.28it/s]
3|5223|Loss: 0.9690608978271484:  81%|████████  | 5223/6470 [1:04:44<15:00,  1.38it/s]
3|5224|Loss: 0.9386522769927979:  81%|████████  | 5223/6470 [1:04:44<15:00,  1.38it/s]
3|5224|Loss: 0.9386522769927979:  81%|████████  | 5224/6470 [1:04:45<15:43,  1.32it/s]
3|5225|Loss: 0.7565877437591553:  81%|████████  | 5224/6470 [1:04:45<15:43,  1.32it/s]
3|5225|Loss: 0.7565877437591553:  81%|████████  | 5225/6470 [1:04:45<14:41,  1.41it/s]
3|5226|Loss: 0.7969173789024353:  81%|████████  | 5225/6470 [1:04:46<14:41,  1.41it/s]
3|5226|Loss: 0.7969173789024353:  81%|████████  | 5226/6470 [1:04:46<15:15,  1.36it/s]
3|5227|Loss: 0.8002612590789795:  81%|████████  | 5226/6470 [1:04:46<15:15,  1.36it/s]
3|5227|Loss: 0.8002612590789795:  81%|████████  | 5227/6470 [1:04:47<15:25,  1.34it/s]
3|5228|Loss: 0.8831697106361389:  81%|████████  | 5227/6470 [1:04:47<15:25,  1.34it/s]
3|5228|Loss: 0.8831697106361389:  81%|████████  | 5228/6470 [1:04:47<15:02,  1.38it/s]
3|5229|Loss: 0.8773984909057617:  81%|████████  | 5228/6470 [1:04:48<15:02,  1.38it/s]
3|5229|Loss: 0.8773984909057617:  81%|████████  | 5229/6470 [1:04:48<15:40,  1.32it/s]
3|5230|Loss: 0.8833009600639343:  81%|████████  | 5229/6470 [1:04:49<15:40,  1.32it/s]
3|5230|Loss: 0.8833009600639343:  81%|████████  | 5230/6470 [1:04:49<15:44,  1.31it/s]
3|5231|Loss: 0.7246798276901245:  81%|████████  | 5230/6470 [1:04:49<15:44,  1.31it/s]
3|5231|Loss: 0.7246798276901245:  81%|████████  | 5231/6470 [1:04:50<15:53,  1.30it/s]
3|5232|Loss: 0.6709966659545898:  81%|████████  | 5231/6470 [1:04:50<15:53,  1.30it/s]
3|5232|Loss: 0.6709966659545898:  81%|████████  | 5232/6470 [1:04:50<15:11,  1.36it/s]
3|5233|Loss: 0.6788718104362488:  81%|████████  | 5232/6470 [1:04:51<15:11,  1.36it/s]
3|5233|Loss: 0.6788718104362488:  81%|████████  | 5233/6470 [1:04:51<14:23,  1.43it/s]
3|5234|Loss: 0.8084595203399658:  81%|████████  | 5233/6470 [1:04:52<14:23,  1.43it/s]
3|5234|Loss: 0.8084595203399658:  81%|████████  | 5234/6470 [1:04:52<15:19,  1.34it/s]
3|5235|Loss: 0.6442399024963379:  81%|████████  | 5234/6470 [1:04:52<15:19,  1.34it/s]
3|5235|Loss: 0.6442399024963379:  81%|████████  | 5235/6470 [1:04:53<14:52,  1.38it/s]
3|5236|Loss: 0.8463315367698669:  81%|████████  | 5235/6470 [1:04:53<14:52,  1.38it/s]
3|5236|Loss: 0.8463315367698669:  81%|████████  | 5236/6470 [1:04:53<14:56,  1.38it/s]
3|5237|Loss: 0.9402838945388794:  81%|████████  | 5236/6470 [1:04:54<14:56,  1.38it/s]
3|5237|Loss: 0.9402838945388794:  81%|████████  | 5237/6470 [1:04:54<15:41,  1.31it/s]
3|5238|Loss: 0.656773567199707:  81%|████████  | 5237/6470 [1:04:55<15:41,  1.31it/s] 
3|5238|Loss: 0.656773567199707:  81%|████████  | 5238/6470 [1:04:55<16:57,  1.21it/s]
3|5239|Loss: 0.8064113259315491:  81%|████████  | 5238/6470 [1:04:56<16:57,  1.21it/s]
3|5239|Loss: 0.8064113259315491:  81%|████████  | 5239/6470 [1:04:56<17:39,  1.16it/s]
3|5240|Loss: 0.8821264505386353:  81%|████████  | 5239/6470 [1:04:56<17:39,  1.16it/s]
3|5240|Loss: 0.8821264505386353:  81%|████████  | 5240/6470 [1:04:57<16:16,  1.26it/s]
3|5241|Loss: 0.8378453850746155:  81%|████████  | 5240/6470 [1:04:57<16:16,  1.26it/s]
3|5241|Loss: 0.8378453850746155:  81%|████████  | 5241/6470 [1:04:57<15:03,  1.36it/s]
3|5242|Loss: 0.8186317682266235:  81%|████████  | 5241/6470 [1:04:58<15:03,  1.36it/s]
3|5242|Loss: 0.8186317682266235:  81%|████████  | 5242/6470 [1:04:58<15:02,  1.36it/s]
3|5243|Loss: 0.8129729628562927:  81%|████████  | 5242/6470 [1:04:58<15:02,  1.36it/s]
3|5243|Loss: 0.8129729628562927:  81%|████████  | 5243/6470 [1:04:59<15:14,  1.34it/s]
3|5244|Loss: 0.9661208391189575:  81%|████████  | 5243/6470 [1:04:59<15:14,  1.34it/s]
3|5244|Loss: 0.9661208391189575:  81%|████████  | 5244/6470 [1:05:00<16:08,  1.27it/s]
3|5245|Loss: 0.7961707711219788:  81%|████████  | 5244/6470 [1:05:00<16:08,  1.27it/s]
3|5245|Loss: 0.7961707711219788:  81%|████████  | 5245/6470 [1:05:00<15:31,  1.31it/s]
3|5246|Loss: 0.9104562401771545:  81%|████████  | 5245/6470 [1:05:01<15:31,  1.31it/s]
3|5246|Loss: 0.9104562401771545:  81%|████████  | 5246/6470 [1:05:01<15:18,  1.33it/s]
3|5247|Loss: 0.8641855716705322:  81%|████████  | 5246/6470 [1:05:02<15:18,  1.33it/s]
3|5247|Loss: 0.8641855716705322:  81%|████████  | 5247/6470 [1:05:02<14:15,  1.43it/s]
3|5248|Loss: 0.9220377802848816:  81%|████████  | 5247/6470 [1:05:02<14:15,  1.43it/s]
3|5248|Loss: 0.9220377802848816:  81%|████████  | 5248/6470 [1:05:02<14:13,  1.43it/s]
3|5249|Loss: 0.7679262757301331:  81%|████████  | 5248/6470 [1:05:03<14:13,  1.43it/s]
3|5249|Loss: 0.7679262757301331:  81%|████████  | 5249/6470 [1:05:03<15:26,  1.32it/s]
3|5250|Loss: 0.9253636002540588:  81%|████████  | 5249/6470 [1:05:04<15:26,  1.32it/s]
3|5250|Loss: 0.9253636002540588:  81%|████████  | 5250/6470 [1:05:04<14:22,  1.42it/s]
3|5251|Loss: 0.8005139827728271:  81%|████████  | 5250/6470 [1:05:04<14:22,  1.42it/s]
3|5251|Loss: 0.8005139827728271:  81%|████████  | 5251/6470 [1:05:05<15:43,  1.29it/s]
3|5252|Loss: 0.7676162123680115:  81%|████████  | 5251/6470 [1:05:05<15:43,  1.29it/s]
3|5252|Loss: 0.7676162123680115:  81%|████████  | 5252/6470 [1:05:06<15:39,  1.30it/s]
3|5253|Loss: 0.7079911231994629:  81%|████████  | 5252/6470 [1:05:06<15:39,  1.30it/s]
3|5253|Loss: 0.7079911231994629:  81%|████████  | 5253/6470 [1:05:06<15:27,  1.31it/s]
3|5254|Loss: 0.7428053617477417:  81%|████████  | 5253/6470 [1:05:07<15:27,  1.31it/s]
3|5254|Loss: 0.7428053617477417:  81%|████████  | 5254/6470 [1:05:07<14:51,  1.36it/s]
3|5255|Loss: 0.6912590265274048:  81%|████████  | 5254/6470 [1:05:07<14:51,  1.36it/s]
3|5255|Loss: 0.6912590265274048:  81%|████████  | 5255/6470 [1:05:07<13:07,  1.54it/s]
3|5256|Loss: 0.967645525932312:  81%|████████  | 5255/6470 [1:05:08<13:07,  1.54it/s] 
3|5256|Loss: 0.967645525932312:  81%|████████  | 5256/6470 [1:05:08<13:21,  1.51it/s]
3|5257|Loss: 0.7199257612228394:  81%|████████  | 5256/6470 [1:05:09<13:21,  1.51it/s]
3|5257|Loss: 0.7199257612228394:  81%|████████▏ | 5257/6470 [1:05:09<14:11,  1.42it/s]
3|5258|Loss: 0.6997701525688171:  81%|████████▏ | 5257/6470 [1:05:09<14:11,  1.42it/s]
3|5258|Loss: 0.6997701525688171:  81%|████████▏ | 5258/6470 [1:05:10<13:45,  1.47it/s]
3|5259|Loss: 0.9248919486999512:  81%|████████▏ | 5258/6470 [1:05:10<13:45,  1.47it/s]
3|5259|Loss: 0.9248919486999512:  81%|████████▏ | 5259/6470 [1:05:10<13:21,  1.51it/s]
3|5260|Loss: 0.8723900318145752:  81%|████████▏ | 5259/6470 [1:05:11<13:21,  1.51it/s]
3|5260|Loss: 0.8723900318145752:  81%|████████▏ | 5260/6470 [1:05:11<13:53,  1.45it/s]
3|5261|Loss: 0.9331570267677307:  81%|████████▏ | 5260/6470 [1:05:11<13:53,  1.45it/s]
3|5261|Loss: 0.9331570267677307:  81%|████████▏ | 5261/6470 [1:05:12<13:39,  1.48it/s]
3|5262|Loss: 0.8044077754020691:  81%|████████▏ | 5261/6470 [1:05:12<13:39,  1.48it/s]
3|5262|Loss: 0.8044077754020691:  81%|████████▏ | 5262/6470 [1:05:12<13:55,  1.45it/s]
3|5263|Loss: 0.7334720492362976:  81%|████████▏ | 5262/6470 [1:05:13<13:55,  1.45it/s]
3|5263|Loss: 0.7334720492362976:  81%|████████▏ | 5263/6470 [1:05:13<14:20,  1.40it/s]
3|5264|Loss: 0.7341842651367188:  81%|████████▏ | 5263/6470 [1:05:13<14:20,  1.40it/s]
3|5264|Loss: 0.7341842651367188:  81%|████████▏ | 5264/6470 [1:05:14<14:08,  1.42it/s]
3|5265|Loss: 0.8479853868484497:  81%|████████▏ | 5264/6470 [1:05:14<14:08,  1.42it/s]
3|5265|Loss: 0.8479853868484497:  81%|████████▏ | 5265/6470 [1:05:14<12:58,  1.55it/s]
3|5266|Loss: 0.8418324589729309:  81%|████████▏ | 5265/6470 [1:05:15<12:58,  1.55it/s]
3|5266|Loss: 0.8418324589729309:  81%|████████▏ | 5266/6470 [1:05:15<13:29,  1.49it/s]
3|5267|Loss: 0.9619603157043457:  81%|████████▏ | 5266/6470 [1:05:15<13:29,  1.49it/s]
3|5267|Loss: 0.9619603157043457:  81%|████████▏ | 5267/6470 [1:05:16<13:10,  1.52it/s]
3|5268|Loss: 0.9733056426048279:  81%|████████▏ | 5267/6470 [1:05:16<13:10,  1.52it/s]
3|5268|Loss: 0.9733056426048279:  81%|████████▏ | 5268/6470 [1:05:16<12:50,  1.56it/s]
3|5269|Loss: 0.8936528563499451:  81%|████████▏ | 5268/6470 [1:05:17<12:50,  1.56it/s]
3|5269|Loss: 0.8936528563499451:  81%|████████▏ | 5269/6470 [1:05:17<14:05,  1.42it/s]
3|5270|Loss: 0.9604292511940002:  81%|████████▏ | 5269/6470 [1:05:18<14:05,  1.42it/s]
3|5270|Loss: 0.9604292511940002:  81%|████████▏ | 5270/6470 [1:05:18<14:37,  1.37it/s]
3|5271|Loss: 0.8784246444702148:  81%|████████▏ | 5270/6470 [1:05:18<14:37,  1.37it/s]
3|5271|Loss: 0.8784246444702148:  81%|████████▏ | 5271/6470 [1:05:19<14:23,  1.39it/s]
3|5272|Loss: 0.7784258127212524:  81%|████████▏ | 5271/6470 [1:05:19<14:23,  1.39it/s]
3|5272|Loss: 0.7784258127212524:  81%|████████▏ | 5272/6470 [1:05:19<14:05,  1.42it/s]
3|5273|Loss: 0.9329591393470764:  81%|████████▏ | 5272/6470 [1:05:20<14:05,  1.42it/s]
3|5273|Loss: 0.9329591393470764:  81%|████████▏ | 5273/6470 [1:05:20<14:59,  1.33it/s]
3|5274|Loss: 0.8551158905029297:  81%|████████▏ | 5273/6470 [1:05:21<14:59,  1.33it/s]
3|5274|Loss: 0.8551158905029297:  82%|████████▏ | 5274/6470 [1:05:21<15:49,  1.26it/s]
3|5275|Loss: 0.9048115015029907:  82%|████████▏ | 5274/6470 [1:05:21<15:49,  1.26it/s]
3|5275|Loss: 0.9048115015029907:  82%|████████▏ | 5275/6470 [1:05:22<15:58,  1.25it/s]
3|5276|Loss: 0.783303439617157:  82%|████████▏ | 5275/6470 [1:05:22<15:58,  1.25it/s] 
3|5276|Loss: 0.783303439617157:  82%|████████▏ | 5276/6470 [1:05:23<16:38,  1.20it/s]
3|5277|Loss: 0.7752092480659485:  82%|████████▏ | 5276/6470 [1:05:23<16:38,  1.20it/s]
3|5277|Loss: 0.7752092480659485:  82%|████████▏ | 5277/6470 [1:05:24<16:29,  1.21it/s]
3|5278|Loss: 0.9355632662773132:  82%|████████▏ | 5277/6470 [1:05:24<16:29,  1.21it/s]
3|5278|Loss: 0.9355632662773132:  82%|████████▏ | 5278/6470 [1:05:24<15:16,  1.30it/s]
3|5279|Loss: 0.7520627379417419:  82%|████████▏ | 5278/6470 [1:05:24<15:16,  1.30it/s]
3|5279|Loss: 0.7520627379417419:  82%|████████▏ | 5279/6470 [1:05:25<13:13,  1.50it/s]
3|5280|Loss: 0.8080450296401978:  82%|████████▏ | 5279/6470 [1:05:25<13:13,  1.50it/s]
3|5280|Loss: 0.8080450296401978:  82%|████████▏ | 5280/6470 [1:05:25<12:58,  1.53it/s]
3|5281|Loss: 0.8880870938301086:  82%|████████▏ | 5280/6470 [1:05:26<12:58,  1.53it/s]
3|5281|Loss: 0.8880870938301086:  82%|████████▏ | 5281/6470 [1:05:26<13:49,  1.43it/s]
3|5282|Loss: 0.7554791569709778:  82%|████████▏ | 5281/6470 [1:05:26<13:49,  1.43it/s]
3|5282|Loss: 0.7554791569709778:  82%|████████▏ | 5282/6470 [1:05:27<14:17,  1.38it/s]
3|5283|Loss: 0.8710727095603943:  82%|████████▏ | 5282/6470 [1:05:27<14:17,  1.38it/s]
3|5283|Loss: 0.8710727095603943:  82%|████████▏ | 5283/6470 [1:05:28<14:45,  1.34it/s]
3|5284|Loss: 0.6882278323173523:  82%|████████▏ | 5283/6470 [1:05:28<14:45,  1.34it/s]
3|5284|Loss: 0.6882278323173523:  82%|████████▏ | 5284/6470 [1:05:28<15:03,  1.31it/s]
3|5285|Loss: 0.6773858666419983:  82%|████████▏ | 5284/6470 [1:05:29<15:03,  1.31it/s]
3|5285|Loss: 0.6773858666419983:  82%|████████▏ | 5285/6470 [1:05:29<14:36,  1.35it/s]
3|5286|Loss: 0.9201071262359619:  82%|████████▏ | 5285/6470 [1:05:30<14:36,  1.35it/s]
3|5286|Loss: 0.9201071262359619:  82%|████████▏ | 5286/6470 [1:05:30<14:58,  1.32it/s]
3|5287|Loss: 0.8245822787284851:  82%|████████▏ | 5286/6470 [1:05:30<14:58,  1.32it/s]
3|5287|Loss: 0.8245822787284851:  82%|████████▏ | 5287/6470 [1:05:31<15:58,  1.23it/s]
3|5288|Loss: 0.9170995354652405:  82%|████████▏ | 5287/6470 [1:05:31<15:58,  1.23it/s]
3|5288|Loss: 0.9170995354652405:  82%|████████▏ | 5288/6470 [1:05:32<16:03,  1.23it/s]
3|5289|Loss: 0.707161545753479:  82%|████████▏ | 5288/6470 [1:05:32<16:03,  1.23it/s] 
3|5289|Loss: 0.707161545753479:  82%|████████▏ | 5289/6470 [1:05:32<15:53,  1.24it/s]
3|5290|Loss: 0.9526903033256531:  82%|████████▏ | 5289/6470 [1:05:33<15:53,  1.24it/s]
3|5290|Loss: 0.9526903033256531:  82%|████████▏ | 5290/6470 [1:05:33<14:55,  1.32it/s]
3|5291|Loss: 0.8194284439086914:  82%|████████▏ | 5290/6470 [1:05:33<14:55,  1.32it/s]
3|5291|Loss: 0.8194284439086914:  82%|████████▏ | 5291/6470 [1:05:34<13:46,  1.43it/s]
3|5292|Loss: 0.7656391859054565:  82%|████████▏ | 5291/6470 [1:05:34<13:46,  1.43it/s]
3|5292|Loss: 0.7656391859054565:  82%|████████▏ | 5292/6470 [1:05:34<13:34,  1.45it/s]
3|5293|Loss: 0.8429602980613708:  82%|████████▏ | 5292/6470 [1:05:35<13:34,  1.45it/s]
3|5293|Loss: 0.8429602980613708:  82%|████████▏ | 5293/6470 [1:05:35<13:29,  1.45it/s]
3|5294|Loss: 0.7674117088317871:  82%|████████▏ | 5293/6470 [1:05:35<13:29,  1.45it/s]
3|5294|Loss: 0.7674117088317871:  82%|████████▏ | 5294/6470 [1:05:36<13:30,  1.45it/s]
3|5295|Loss: 0.8466396331787109:  82%|████████▏ | 5294/6470 [1:05:36<13:30,  1.45it/s]
3|5295|Loss: 0.8466396331787109:  82%|████████▏ | 5295/6470 [1:05:36<13:23,  1.46it/s]
3|5296|Loss: 0.6670699715614319:  82%|████████▏ | 5295/6470 [1:05:37<13:23,  1.46it/s]
3|5296|Loss: 0.6670699715614319:  82%|████████▏ | 5296/6470 [1:05:37<14:29,  1.35it/s]
3|5297|Loss: 1.0154114961624146:  82%|████████▏ | 5296/6470 [1:05:38<14:29,  1.35it/s]
3|5297|Loss: 1.0154114961624146:  82%|████████▏ | 5297/6470 [1:05:38<13:46,  1.42it/s]
3|5298|Loss: 0.7552002668380737:  82%|████████▏ | 5297/6470 [1:05:38<13:46,  1.42it/s]
3|5298|Loss: 0.7552002668380737:  82%|████████▏ | 5298/6470 [1:05:39<13:45,  1.42it/s]
3|5299|Loss: 0.811595618724823:  82%|████████▏ | 5298/6470 [1:05:39<13:45,  1.42it/s] 
3|5299|Loss: 0.811595618724823:  82%|████████▏ | 5299/6470 [1:05:39<13:45,  1.42it/s]
3|5300|Loss: 0.7890613079071045:  82%|████████▏ | 5299/6470 [1:05:40<13:45,  1.42it/s]
3|5300|Loss: 0.7890613079071045:  82%|████████▏ | 5300/6470 [1:05:40<14:14,  1.37it/s]
3|5301|Loss: 0.8510634899139404:  82%|████████▏ | 5300/6470 [1:05:40<14:14,  1.37it/s]
3|5301|Loss: 0.8510634899139404:  82%|████████▏ | 5301/6470 [1:05:41<14:05,  1.38it/s]
3|5302|Loss: 0.9301762580871582:  82%|████████▏ | 5301/6470 [1:05:41<14:05,  1.38it/s]
3|5302|Loss: 0.9301762580871582:  82%|████████▏ | 5302/6470 [1:05:42<14:24,  1.35it/s]
3|5303|Loss: 0.8047099709510803:  82%|████████▏ | 5302/6470 [1:05:42<14:24,  1.35it/s]
3|5303|Loss: 0.8047099709510803:  82%|████████▏ | 5303/6470 [1:05:42<14:19,  1.36it/s]
3|5304|Loss: 0.9350370168685913:  82%|████████▏ | 5303/6470 [1:05:43<14:19,  1.36it/s]
3|5304|Loss: 0.9350370168685913:  82%|████████▏ | 5304/6470 [1:05:43<13:42,  1.42it/s]
3|5305|Loss: 0.8912501931190491:  82%|████████▏ | 5304/6470 [1:05:43<13:42,  1.42it/s]
3|5305|Loss: 0.8912501931190491:  82%|████████▏ | 5305/6470 [1:05:44<13:31,  1.44it/s]
3|5306|Loss: 0.725151002407074:  82%|████████▏ | 5305/6470 [1:05:44<13:31,  1.44it/s] 
3|5306|Loss: 0.725151002407074:  82%|████████▏ | 5306/6470 [1:05:44<14:20,  1.35it/s]
3|5307|Loss: 0.8469855189323425:  82%|████████▏ | 5306/6470 [1:05:45<14:20,  1.35it/s]
3|5307|Loss: 0.8469855189323425:  82%|████████▏ | 5307/6470 [1:05:45<13:53,  1.39it/s]
3|5308|Loss: 0.8497262597084045:  82%|████████▏ | 5307/6470 [1:05:46<13:53,  1.39it/s]
3|5308|Loss: 0.8497262597084045:  82%|████████▏ | 5308/6470 [1:05:46<14:34,  1.33it/s]
3|5309|Loss: 0.7738870978355408:  82%|████████▏ | 5308/6470 [1:05:46<14:34,  1.33it/s]
3|5309|Loss: 0.7738870978355408:  82%|████████▏ | 5309/6470 [1:05:47<14:16,  1.36it/s]
3|5310|Loss: 0.9073427319526672:  82%|████████▏ | 5309/6470 [1:05:47<14:16,  1.36it/s]
3|5310|Loss: 0.9073427319526672:  82%|████████▏ | 5310/6470 [1:05:47<14:17,  1.35it/s]
3|5311|Loss: 0.9668207764625549:  82%|████████▏ | 5310/6470 [1:05:48<14:17,  1.35it/s]
3|5311|Loss: 0.9668207764625549:  82%|████████▏ | 5311/6470 [1:05:48<15:07,  1.28it/s]
3|5312|Loss: 0.8059151768684387:  82%|████████▏ | 5311/6470 [1:05:49<15:07,  1.28it/s]
3|5312|Loss: 0.8059151768684387:  82%|████████▏ | 5312/6470 [1:05:49<15:58,  1.21it/s]
3|5313|Loss: 0.6958270072937012:  82%|████████▏ | 5312/6470 [1:05:50<15:58,  1.21it/s]
3|5313|Loss: 0.6958270072937012:  82%|████████▏ | 5313/6470 [1:05:50<15:12,  1.27it/s]
3|5314|Loss: 0.8731695413589478:  82%|████████▏ | 5313/6470 [1:05:50<15:12,  1.27it/s]
3|5314|Loss: 0.8731695413589478:  82%|████████▏ | 5314/6470 [1:05:51<14:24,  1.34it/s]
3|5315|Loss: 0.724503219127655:  82%|████████▏ | 5314/6470 [1:05:51<14:24,  1.34it/s] 
3|5315|Loss: 0.724503219127655:  82%|████████▏ | 5315/6470 [1:05:51<15:14,  1.26it/s]
3|5316|Loss: 0.824156641960144:  82%|████████▏ | 5315/6470 [1:05:52<15:14,  1.26it/s]
3|5316|Loss: 0.824156641960144:  82%|████████▏ | 5316/6470 [1:05:52<16:02,  1.20it/s]
3|5317|Loss: 0.9379978775978088:  82%|████████▏ | 5316/6470 [1:05:53<16:02,  1.20it/s]
3|5317|Loss: 0.9379978775978088:  82%|████████▏ | 5317/6470 [1:05:53<15:08,  1.27it/s]
3|5318|Loss: 0.7730957269668579:  82%|████████▏ | 5317/6470 [1:05:53<15:08,  1.27it/s]
3|5318|Loss: 0.7730957269668579:  82%|████████▏ | 5318/6470 [1:05:54<14:11,  1.35it/s]
3|5319|Loss: 0.8446365594863892:  82%|████████▏ | 5318/6470 [1:05:54<14:11,  1.35it/s]
3|5319|Loss: 0.8446365594863892:  82%|████████▏ | 5319/6470 [1:05:54<14:07,  1.36it/s]
3|5320|Loss: 0.9886873364448547:  82%|████████▏ | 5319/6470 [1:05:55<14:07,  1.36it/s]
3|5320|Loss: 0.9886873364448547:  82%|████████▏ | 5320/6470 [1:05:55<13:13,  1.45it/s]
3|5321|Loss: 0.7206435203552246:  82%|████████▏ | 5320/6470 [1:05:55<13:13,  1.45it/s]
3|5321|Loss: 0.7206435203552246:  82%|████████▏ | 5321/6470 [1:05:56<14:10,  1.35it/s]
3|5322|Loss: 0.8545502424240112:  82%|████████▏ | 5321/6470 [1:05:56<14:10,  1.35it/s]
3|5322|Loss: 0.8545502424240112:  82%|████████▏ | 5322/6470 [1:05:57<13:54,  1.38it/s]
3|5323|Loss: 0.8175556063652039:  82%|████████▏ | 5322/6470 [1:05:57<13:54,  1.38it/s]
3|5323|Loss: 0.8175556063652039:  82%|████████▏ | 5323/6470 [1:05:57<14:09,  1.35it/s]
3|5324|Loss: 0.8606876730918884:  82%|████████▏ | 5323/6470 [1:05:58<14:09,  1.35it/s]
3|5324|Loss: 0.8606876730918884:  82%|████████▏ | 5324/6470 [1:05:58<14:31,  1.31it/s]
3|5325|Loss: 0.8970537185668945:  82%|████████▏ | 5324/6470 [1:05:59<14:31,  1.31it/s]
3|5325|Loss: 0.8970537185668945:  82%|████████▏ | 5325/6470 [1:05:59<14:34,  1.31it/s]
3|5326|Loss: 1.0267678499221802:  82%|████████▏ | 5325/6470 [1:05:59<14:34,  1.31it/s]
3|5326|Loss: 1.0267678499221802:  82%|████████▏ | 5326/6470 [1:06:00<14:43,  1.29it/s]
3|5327|Loss: 0.9147738218307495:  82%|████████▏ | 5326/6470 [1:06:00<14:43,  1.29it/s]
3|5327|Loss: 0.9147738218307495:  82%|████████▏ | 5327/6470 [1:06:00<14:44,  1.29it/s]
3|5328|Loss: 0.7422453165054321:  82%|████████▏ | 5327/6470 [1:06:01<14:44,  1.29it/s]
3|5328|Loss: 0.7422453165054321:  82%|████████▏ | 5328/6470 [1:06:01<14:03,  1.35it/s]
3|5329|Loss: 0.7282743453979492:  82%|████████▏ | 5328/6470 [1:06:01<14:03,  1.35it/s]
3|5329|Loss: 0.7282743453979492:  82%|████████▏ | 5329/6470 [1:06:02<13:45,  1.38it/s]
3|5330|Loss: 0.7983113527297974:  82%|████████▏ | 5329/6470 [1:06:02<13:45,  1.38it/s]
3|5330|Loss: 0.7983113527297974:  82%|████████▏ | 5330/6470 [1:06:03<13:35,  1.40it/s]
3|5331|Loss: 0.8184317946434021:  82%|████████▏ | 5330/6470 [1:06:03<13:35,  1.40it/s]
3|5331|Loss: 0.8184317946434021:  82%|████████▏ | 5331/6470 [1:06:03<13:52,  1.37it/s]
3|5332|Loss: 0.824140191078186:  82%|████████▏ | 5331/6470 [1:06:04<13:52,  1.37it/s] 
3|5332|Loss: 0.824140191078186:  82%|████████▏ | 5332/6470 [1:06:04<13:15,  1.43it/s]
3|5333|Loss: 0.7410068511962891:  82%|████████▏ | 5332/6470 [1:06:04<13:15,  1.43it/s]
3|5333|Loss: 0.7410068511962891:  82%|████████▏ | 5333/6470 [1:06:05<14:01,  1.35it/s]
3|5334|Loss: 0.752761960029602:  82%|████████▏ | 5333/6470 [1:06:05<14:01,  1.35it/s] 
3|5334|Loss: 0.752761960029602:  82%|████████▏ | 5334/6470 [1:06:05<13:57,  1.36it/s]
3|5335|Loss: 0.8176666498184204:  82%|████████▏ | 5334/6470 [1:06:06<13:57,  1.36it/s]
3|5335|Loss: 0.8176666498184204:  82%|████████▏ | 5335/6470 [1:06:06<13:44,  1.38it/s]
3|5336|Loss: 0.834991991519928:  82%|████████▏ | 5335/6470 [1:06:07<13:44,  1.38it/s] 
3|5336|Loss: 0.834991991519928:  82%|████████▏ | 5336/6470 [1:06:07<14:02,  1.35it/s]
3|5337|Loss: 0.8365873694419861:  82%|████████▏ | 5336/6470 [1:06:07<14:02,  1.35it/s]
3|5337|Loss: 0.8365873694419861:  82%|████████▏ | 5337/6470 [1:06:08<13:49,  1.37it/s]
3|5338|Loss: 0.7472478151321411:  82%|████████▏ | 5337/6470 [1:06:08<13:49,  1.37it/s]
3|5338|Loss: 0.7472478151321411:  83%|████████▎ | 5338/6470 [1:06:08<13:04,  1.44it/s]
3|5339|Loss: 0.59238600730896:  83%|████████▎ | 5338/6470 [1:06:09<13:04,  1.44it/s]  
3|5339|Loss: 0.59238600730896:  83%|████████▎ | 5339/6470 [1:06:09<13:43,  1.37it/s]
3|5340|Loss: 0.8234944939613342:  83%|████████▎ | 5339/6470 [1:06:10<13:43,  1.37it/s]
3|5340|Loss: 0.8234944939613342:  83%|████████▎ | 5340/6470 [1:06:10<14:35,  1.29it/s]
3|5341|Loss: 0.8618959784507751:  83%|████████▎ | 5340/6470 [1:06:10<14:35,  1.29it/s]
3|5341|Loss: 0.8618959784507751:  83%|████████▎ | 5341/6470 [1:06:11<14:48,  1.27it/s]
3|5342|Loss: 0.8753512501716614:  83%|████████▎ | 5341/6470 [1:06:11<14:48,  1.27it/s]
3|5342|Loss: 0.8753512501716614:  83%|████████▎ | 5342/6470 [1:06:11<13:29,  1.39it/s]
3|5343|Loss: 0.8282482624053955:  83%|████████▎ | 5342/6470 [1:06:12<13:29,  1.39it/s]
3|5343|Loss: 0.8282482624053955:  83%|████████▎ | 5343/6470 [1:06:12<12:54,  1.45it/s]
3|5344|Loss: 0.7596009373664856:  83%|████████▎ | 5343/6470 [1:06:12<12:54,  1.45it/s]
3|5344|Loss: 0.7596009373664856:  83%|████████▎ | 5344/6470 [1:06:13<12:17,  1.53it/s]
3|5345|Loss: 0.8229859471321106:  83%|████████▎ | 5344/6470 [1:06:13<12:17,  1.53it/s]
3|5345|Loss: 0.8229859471321106:  83%|████████▎ | 5345/6470 [1:06:13<11:45,  1.59it/s]
3|5346|Loss: 0.8905788064002991:  83%|████████▎ | 5345/6470 [1:06:13<11:45,  1.59it/s]
3|5346|Loss: 0.8905788064002991:  83%|████████▎ | 5346/6470 [1:06:14<11:43,  1.60it/s]
3|5347|Loss: 0.8215233683586121:  83%|████████▎ | 5346/6470 [1:06:14<11:43,  1.60it/s]
3|5347|Loss: 0.8215233683586121:  83%|████████▎ | 5347/6470 [1:06:14<11:47,  1.59it/s]
3|5348|Loss: 0.9344456791877747:  83%|████████▎ | 5347/6470 [1:06:15<11:47,  1.59it/s]
3|5348|Loss: 0.9344456791877747:  83%|████████▎ | 5348/6470 [1:06:15<12:29,  1.50it/s]
3|5349|Loss: 0.8598271012306213:  83%|████████▎ | 5348/6470 [1:06:16<12:29,  1.50it/s]
3|5349|Loss: 0.8598271012306213:  83%|████████▎ | 5349/6470 [1:06:16<13:50,  1.35it/s]
3|5350|Loss: 0.965023398399353:  83%|████████▎ | 5349/6470 [1:06:16<13:50,  1.35it/s] 
3|5350|Loss: 0.965023398399353:  83%|████████▎ | 5350/6470 [1:06:17<13:51,  1.35it/s]
3|5351|Loss: 0.8486446142196655:  83%|████████▎ | 5350/6470 [1:06:17<13:51,  1.35it/s]
3|5351|Loss: 0.8486446142196655:  83%|████████▎ | 5351/6470 [1:06:18<14:40,  1.27it/s]
3|5352|Loss: 0.9947872757911682:  83%|████████▎ | 5351/6470 [1:06:18<14:40,  1.27it/s]
3|5352|Loss: 0.9947872757911682:  83%|████████▎ | 5352/6470 [1:06:18<14:53,  1.25it/s]
3|5353|Loss: 0.813710629940033:  83%|████████▎ | 5352/6470 [1:06:19<14:53,  1.25it/s] 
3|5353|Loss: 0.813710629940033:  83%|████████▎ | 5353/6470 [1:06:19<14:52,  1.25it/s]
3|5354|Loss: 0.8614480495452881:  83%|████████▎ | 5353/6470 [1:06:20<14:52,  1.25it/s]
3|5354|Loss: 0.8614480495452881:  83%|████████▎ | 5354/6470 [1:06:20<14:53,  1.25it/s]
3|5355|Loss: 0.7391077280044556:  83%|████████▎ | 5354/6470 [1:06:21<14:53,  1.25it/s]
3|5355|Loss: 0.7391077280044556:  83%|████████▎ | 5355/6470 [1:06:21<14:45,  1.26it/s]
3|5356|Loss: 0.8264404535293579:  83%|████████▎ | 5355/6470 [1:06:21<14:45,  1.26it/s]
3|5356|Loss: 0.8264404535293579:  83%|████████▎ | 5356/6470 [1:06:21<13:55,  1.33it/s]
3|5357|Loss: 0.9217202067375183:  83%|████████▎ | 5356/6470 [1:06:22<13:55,  1.33it/s]
3|5357|Loss: 0.9217202067375183:  83%|████████▎ | 5357/6470 [1:06:22<12:55,  1.44it/s]
3|5358|Loss: 0.8545845746994019:  83%|████████▎ | 5357/6470 [1:06:22<12:55,  1.44it/s]
3|5358|Loss: 0.8545845746994019:  83%|████████▎ | 5358/6470 [1:06:23<13:28,  1.38it/s]
3|5359|Loss: 0.9108707904815674:  83%|████████▎ | 5358/6470 [1:06:23<13:28,  1.38it/s]
3|5359|Loss: 0.9108707904815674:  83%|████████▎ | 5359/6470 [1:06:24<14:10,  1.31it/s]
3|5360|Loss: 0.9184772968292236:  83%|████████▎ | 5359/6470 [1:06:24<14:10,  1.31it/s]
3|5360|Loss: 0.9184772968292236:  83%|████████▎ | 5360/6470 [1:06:24<13:38,  1.36it/s]
3|5361|Loss: 0.8364615440368652:  83%|████████▎ | 5360/6470 [1:06:25<13:38,  1.36it/s]
3|5361|Loss: 0.8364615440368652:  83%|████████▎ | 5361/6470 [1:06:25<13:10,  1.40it/s]
3|5362|Loss: 0.6939889192581177:  83%|████████▎ | 5361/6470 [1:06:26<13:10,  1.40it/s]
3|5362|Loss: 0.6939889192581177:  83%|████████▎ | 5362/6470 [1:06:26<14:09,  1.30it/s]
3|5363|Loss: 0.7799688577651978:  83%|████████▎ | 5362/6470 [1:06:26<14:09,  1.30it/s]
3|5363|Loss: 0.7799688577651978:  83%|████████▎ | 5363/6470 [1:06:27<13:16,  1.39it/s]
3|5364|Loss: 0.8096786141395569:  83%|████████▎ | 5363/6470 [1:06:27<13:16,  1.39it/s]
3|5364|Loss: 0.8096786141395569:  83%|████████▎ | 5364/6470 [1:06:27<12:25,  1.48it/s]
3|5365|Loss: 0.9037095904350281:  83%|████████▎ | 5364/6470 [1:06:28<12:25,  1.48it/s]
3|5365|Loss: 0.9037095904350281:  83%|████████▎ | 5365/6470 [1:06:28<12:49,  1.44it/s]
3|5366|Loss: 0.8044143319129944:  83%|████████▎ | 5365/6470 [1:06:28<12:49,  1.44it/s]
3|5366|Loss: 0.8044143319129944:  83%|████████▎ | 5366/6470 [1:06:29<12:47,  1.44it/s]
3|5367|Loss: 0.9225932359695435:  83%|████████▎ | 5366/6470 [1:06:29<12:47,  1.44it/s]
3|5367|Loss: 0.9225932359695435:  83%|████████▎ | 5367/6470 [1:06:29<13:42,  1.34it/s]
3|5368|Loss: 0.7818939685821533:  83%|████████▎ | 5367/6470 [1:06:30<13:42,  1.34it/s]
3|5368|Loss: 0.7818939685821533:  83%|████████▎ | 5368/6470 [1:06:30<14:24,  1.27it/s]
3|5369|Loss: 0.853182315826416:  83%|████████▎ | 5368/6470 [1:06:31<14:24,  1.27it/s] 
3|5369|Loss: 0.853182315826416:  83%|████████▎ | 5369/6470 [1:06:31<14:21,  1.28it/s]
3|5370|Loss: 0.9482536315917969:  83%|████████▎ | 5369/6470 [1:06:31<14:21,  1.28it/s]
3|5370|Loss: 0.9482536315917969:  83%|████████▎ | 5370/6470 [1:06:32<13:57,  1.31it/s]
3|5371|Loss: 0.8010188341140747:  83%|████████▎ | 5370/6470 [1:06:32<13:57,  1.31it/s]
3|5371|Loss: 0.8010188341140747:  83%|████████▎ | 5371/6470 [1:06:33<14:37,  1.25it/s]
3|5372|Loss: 0.8515748977661133:  83%|████████▎ | 5371/6470 [1:06:33<14:37,  1.25it/s]
3|5372|Loss: 0.8515748977661133:  83%|████████▎ | 5372/6470 [1:06:34<15:29,  1.18it/s]
3|5373|Loss: 0.7547627687454224:  83%|████████▎ | 5372/6470 [1:06:34<15:29,  1.18it/s]
3|5373|Loss: 0.7547627687454224:  83%|████████▎ | 5373/6470 [1:06:35<15:53,  1.15it/s]
3|5374|Loss: 0.8036437630653381:  83%|████████▎ | 5373/6470 [1:06:35<15:53,  1.15it/s]
3|5374|Loss: 0.8036437630653381:  83%|████████▎ | 5374/6470 [1:06:35<15:55,  1.15it/s]
3|5375|Loss: 0.6786487698554993:  83%|████████▎ | 5374/6470 [1:06:36<15:55,  1.15it/s]
3|5375|Loss: 0.6786487698554993:  83%|████████▎ | 5375/6470 [1:06:36<13:41,  1.33it/s]
3|5376|Loss: 0.8576653599739075:  83%|████████▎ | 5375/6470 [1:06:36<13:41,  1.33it/s]
3|5376|Loss: 0.8576653599739075:  83%|████████▎ | 5376/6470 [1:06:36<12:28,  1.46it/s]
3|5377|Loss: 0.8987839818000793:  83%|████████▎ | 5376/6470 [1:06:37<12:28,  1.46it/s]
3|5377|Loss: 0.8987839818000793:  83%|████████▎ | 5377/6470 [1:06:37<12:14,  1.49it/s]
3|5378|Loss: 0.8886383771896362:  83%|████████▎ | 5377/6470 [1:06:37<12:14,  1.49it/s]
3|5378|Loss: 0.8886383771896362:  83%|████████▎ | 5378/6470 [1:06:38<12:23,  1.47it/s]
3|5379|Loss: 0.9682915806770325:  83%|████████▎ | 5378/6470 [1:06:38<12:23,  1.47it/s]
3|5379|Loss: 0.9682915806770325:  83%|████████▎ | 5379/6470 [1:06:39<12:41,  1.43it/s]
3|5380|Loss: 0.8937026262283325:  83%|████████▎ | 5379/6470 [1:06:39<12:41,  1.43it/s]
3|5380|Loss: 0.8937026262283325:  83%|████████▎ | 5380/6470 [1:06:39<12:34,  1.44it/s]
3|5381|Loss: 0.6911972165107727:  83%|████████▎ | 5380/6470 [1:06:40<12:34,  1.44it/s]
3|5381|Loss: 0.6911972165107727:  83%|████████▎ | 5381/6470 [1:06:40<12:24,  1.46it/s]
3|5382|Loss: 0.6693982481956482:  83%|████████▎ | 5381/6470 [1:06:40<12:24,  1.46it/s]
3|5382|Loss: 0.6693982481956482:  83%|████████▎ | 5382/6470 [1:06:40<11:49,  1.53it/s]
3|5383|Loss: 0.8819716572761536:  83%|████████▎ | 5382/6470 [1:06:41<11:49,  1.53it/s]
3|5383|Loss: 0.8819716572761536:  83%|████████▎ | 5383/6470 [1:06:41<12:16,  1.48it/s]
3|5384|Loss: 0.8897491097450256:  83%|████████▎ | 5383/6470 [1:06:42<12:16,  1.48it/s]
3|5384|Loss: 0.8897491097450256:  83%|████████▎ | 5384/6470 [1:06:42<12:41,  1.43it/s]
3|5385|Loss: 0.8414007425308228:  83%|████████▎ | 5384/6470 [1:06:42<12:41,  1.43it/s]
3|5385|Loss: 0.8414007425308228:  83%|████████▎ | 5385/6470 [1:06:43<13:33,  1.33it/s]
3|5386|Loss: 0.8707305788993835:  83%|████████▎ | 5385/6470 [1:06:43<13:33,  1.33it/s]
3|5386|Loss: 0.8707305788993835:  83%|████████▎ | 5386/6470 [1:06:44<14:26,  1.25it/s]
3|5387|Loss: 0.7741478681564331:  83%|████████▎ | 5386/6470 [1:06:44<14:26,  1.25it/s]
3|5387|Loss: 0.7741478681564331:  83%|████████▎ | 5387/6470 [1:06:44<14:01,  1.29it/s]
3|5388|Loss: 0.8161092400550842:  83%|████████▎ | 5387/6470 [1:06:45<14:01,  1.29it/s]
3|5388|Loss: 0.8161092400550842:  83%|████████▎ | 5388/6470 [1:06:45<14:33,  1.24it/s]
3|5389|Loss: 0.8644073605537415:  83%|████████▎ | 5388/6470 [1:06:46<14:33,  1.24it/s]
3|5389|Loss: 0.8644073605537415:  83%|████████▎ | 5389/6470 [1:06:46<13:54,  1.30it/s]
3|5390|Loss: 0.7743234038352966:  83%|████████▎ | 5389/6470 [1:06:46<13:54,  1.30it/s]
3|5390|Loss: 0.7743234038352966:  83%|████████▎ | 5390/6470 [1:06:47<13:38,  1.32it/s]
3|5391|Loss: 0.9252583980560303:  83%|████████▎ | 5390/6470 [1:06:47<13:38,  1.32it/s]
3|5391|Loss: 0.9252583980560303:  83%|████████▎ | 5391/6470 [1:06:48<13:44,  1.31it/s]
3|5392|Loss: 0.9043605923652649:  83%|████████▎ | 5391/6470 [1:06:48<13:44,  1.31it/s]
3|5392|Loss: 0.9043605923652649:  83%|████████▎ | 5392/6470 [1:06:48<14:17,  1.26it/s]
3|5393|Loss: 0.839024007320404:  83%|████████▎ | 5392/6470 [1:06:49<14:17,  1.26it/s] 
3|5393|Loss: 0.839024007320404:  83%|████████▎ | 5393/6470 [1:06:49<15:01,  1.19it/s]
3|5394|Loss: 0.8995675444602966:  83%|████████▎ | 5393/6470 [1:06:50<15:01,  1.19it/s]
3|5394|Loss: 0.8995675444602966:  83%|████████▎ | 5394/6470 [1:06:50<14:29,  1.24it/s]
3|5395|Loss: 0.784112811088562:  83%|████████▎ | 5394/6470 [1:06:50<14:29,  1.24it/s] 
3|5395|Loss: 0.784112811088562:  83%|████████▎ | 5395/6470 [1:06:51<13:55,  1.29it/s]
3|5396|Loss: 0.8295581340789795:  83%|████████▎ | 5395/6470 [1:06:51<13:55,  1.29it/s]
3|5396|Loss: 0.8295581340789795:  83%|████████▎ | 5396/6470 [1:06:51<13:31,  1.32it/s]
3|5397|Loss: 0.7870311141014099:  83%|████████▎ | 5396/6470 [1:06:52<13:31,  1.32it/s]
3|5397|Loss: 0.7870311141014099:  83%|████████▎ | 5397/6470 [1:06:52<12:51,  1.39it/s]
3|5398|Loss: 0.8084588646888733:  83%|████████▎ | 5397/6470 [1:06:53<12:51,  1.39it/s]
3|5398|Loss: 0.8084588646888733:  83%|████████▎ | 5398/6470 [1:06:53<13:20,  1.34it/s]
3|5399|Loss: 0.8958286046981812:  83%|████████▎ | 5398/6470 [1:06:53<13:20,  1.34it/s]
3|5399|Loss: 0.8958286046981812:  83%|████████▎ | 5399/6470 [1:06:54<14:17,  1.25it/s]
3|5400|Loss: 0.7782487273216248:  83%|████████▎ | 5399/6470 [1:06:54<14:17,  1.25it/s]
3|5400|Loss: 0.7782487273216248:  83%|████████▎ | 5400/6470 [1:06:55<15:05,  1.18it/s]
3|5401|Loss: 0.9459885954856873:  83%|████████▎ | 5400/6470 [1:06:55<15:05,  1.18it/s]
3|5401|Loss: 0.9459885954856873:  83%|████████▎ | 5401/6470 [1:06:56<15:24,  1.16it/s]
3|5402|Loss: 0.8613228797912598:  83%|████████▎ | 5401/6470 [1:06:56<15:24,  1.16it/s]
3|5402|Loss: 0.8613228797912598:  83%|████████▎ | 5402/6470 [1:06:56<14:55,  1.19it/s]
3|5403|Loss: 0.8384080529212952:  83%|████████▎ | 5402/6470 [1:06:57<14:55,  1.19it/s]
3|5403|Loss: 0.8384080529212952:  84%|████████▎ | 5403/6470 [1:06:57<14:38,  1.21it/s]
3|5404|Loss: 0.8753528594970703:  84%|████████▎ | 5403/6470 [1:06:58<14:38,  1.21it/s]
3|5404|Loss: 0.8753528594970703:  84%|████████▎ | 5404/6470 [1:06:58<14:28,  1.23it/s]
3|5405|Loss: 0.8006099462509155:  84%|████████▎ | 5404/6470 [1:06:58<14:28,  1.23it/s]
3|5405|Loss: 0.8006099462509155:  84%|████████▎ | 5405/6470 [1:06:59<14:12,  1.25it/s]
3|5406|Loss: 0.8837271332740784:  84%|████████▎ | 5405/6470 [1:06:59<14:12,  1.25it/s]
3|5406|Loss: 0.8837271332740784:  84%|████████▎ | 5406/6470 [1:06:59<12:55,  1.37it/s]
3|5407|Loss: 0.7736307382583618:  84%|████████▎ | 5406/6470 [1:07:00<12:55,  1.37it/s]
3|5407|Loss: 0.7736307382583618:  84%|████████▎ | 5407/6470 [1:07:00<12:28,  1.42it/s]
3|5408|Loss: 0.7154070138931274:  84%|████████▎ | 5407/6470 [1:07:00<12:28,  1.42it/s]
3|5408|Loss: 0.7154070138931274:  84%|████████▎ | 5408/6470 [1:07:01<11:31,  1.54it/s]
3|5409|Loss: 0.8015841245651245:  84%|████████▎ | 5408/6470 [1:07:01<11:31,  1.54it/s]
3|5409|Loss: 0.8015841245651245:  84%|████████▎ | 5409/6470 [1:07:01<11:25,  1.55it/s]
3|5410|Loss: 0.7526813745498657:  84%|████████▎ | 5409/6470 [1:07:02<11:25,  1.55it/s]
3|5410|Loss: 0.7526813745498657:  84%|████████▎ | 5410/6470 [1:07:02<11:39,  1.51it/s]
3|5411|Loss: 0.7826030850410461:  84%|████████▎ | 5410/6470 [1:07:02<11:39,  1.51it/s]
3|5411|Loss: 0.7826030850410461:  84%|████████▎ | 5411/6470 [1:07:03<12:15,  1.44it/s]
3|5412|Loss: 0.7573608756065369:  84%|████████▎ | 5411/6470 [1:07:03<12:15,  1.44it/s]
3|5412|Loss: 0.7573608756065369:  84%|████████▎ | 5412/6470 [1:07:03<12:11,  1.45it/s]
3|5413|Loss: 0.868803858757019:  84%|████████▎ | 5412/6470 [1:07:04<12:11,  1.45it/s] 
3|5413|Loss: 0.868803858757019:  84%|████████▎ | 5413/6470 [1:07:04<12:04,  1.46it/s]
3|5414|Loss: 0.8576127886772156:  84%|████████▎ | 5413/6470 [1:07:04<12:04,  1.46it/s]
3|5414|Loss: 0.8576127886772156:  84%|████████▎ | 5414/6470 [1:07:05<12:20,  1.43it/s]
3|5415|Loss: 0.9253201484680176:  84%|████████▎ | 5414/6470 [1:07:05<12:20,  1.43it/s]
3|5415|Loss: 0.9253201484680176:  84%|████████▎ | 5415/6470 [1:07:05<12:24,  1.42it/s]
3|5416|Loss: 0.807654082775116:  84%|████████▎ | 5415/6470 [1:07:06<12:24,  1.42it/s] 
3|5416|Loss: 0.807654082775116:  84%|████████▎ | 5416/6470 [1:07:06<12:02,  1.46it/s]
3|5417|Loss: 0.8751102685928345:  84%|████████▎ | 5416/6470 [1:07:06<12:02,  1.46it/s]
3|5417|Loss: 0.8751102685928345:  84%|████████▎ | 5417/6470 [1:07:07<12:01,  1.46it/s]
3|5418|Loss: 0.8854581117630005:  84%|████████▎ | 5417/6470 [1:07:07<12:01,  1.46it/s]
3|5418|Loss: 0.8854581117630005:  84%|████████▎ | 5418/6470 [1:07:07<12:06,  1.45it/s]
3|5419|Loss: 0.9130351543426514:  84%|████████▎ | 5418/6470 [1:07:08<12:06,  1.45it/s]
3|5419|Loss: 0.9130351543426514:  84%|████████▍ | 5419/6470 [1:07:08<12:09,  1.44it/s]
3|5420|Loss: 0.9089891910552979:  84%|████████▍ | 5419/6470 [1:07:09<12:09,  1.44it/s]
3|5420|Loss: 0.9089891910552979:  84%|████████▍ | 5420/6470 [1:07:09<11:14,  1.56it/s]
3|5421|Loss: 0.821791410446167:  84%|████████▍ | 5420/6470 [1:07:09<11:14,  1.56it/s] 
3|5421|Loss: 0.821791410446167:  84%|████████▍ | 5421/6470 [1:07:09<10:42,  1.63it/s]
3|5422|Loss: 0.9603813290596008:  84%|████████▍ | 5421/6470 [1:07:10<10:42,  1.63it/s]
3|5422|Loss: 0.9603813290596008:  84%|████████▍ | 5422/6470 [1:07:10<10:45,  1.62it/s]
3|5423|Loss: 0.8447709679603577:  84%|████████▍ | 5422/6470 [1:07:10<10:45,  1.62it/s]
3|5423|Loss: 0.8447709679603577:  84%|████████▍ | 5423/6470 [1:07:11<11:21,  1.54it/s]
3|5424|Loss: 0.8307161927223206:  84%|████████▍ | 5423/6470 [1:07:11<11:21,  1.54it/s]
3|5424|Loss: 0.8307161927223206:  84%|████████▍ | 5424/6470 [1:07:11<11:17,  1.54it/s]
3|5425|Loss: 0.6996974945068359:  84%|████████▍ | 5424/6470 [1:07:12<11:17,  1.54it/s]
3|5425|Loss: 0.6996974945068359:  84%|████████▍ | 5425/6470 [1:07:12<11:44,  1.48it/s]
3|5426|Loss: 0.7052565217018127:  84%|████████▍ | 5425/6470 [1:07:12<11:44,  1.48it/s]
3|5426|Loss: 0.7052565217018127:  84%|████████▍ | 5426/6470 [1:07:13<11:40,  1.49it/s]
3|5427|Loss: 0.8157587051391602:  84%|████████▍ | 5426/6470 [1:07:13<11:40,  1.49it/s]
3|5427|Loss: 0.8157587051391602:  84%|████████▍ | 5427/6470 [1:07:13<12:10,  1.43it/s]
3|5428|Loss: 0.8173435926437378:  84%|████████▍ | 5427/6470 [1:07:14<12:10,  1.43it/s]
3|5428|Loss: 0.8173435926437378:  84%|████████▍ | 5428/6470 [1:07:14<12:43,  1.36it/s]
3|5429|Loss: 0.8372814655303955:  84%|████████▍ | 5428/6470 [1:07:15<12:43,  1.36it/s]
3|5429|Loss: 0.8372814655303955:  84%|████████▍ | 5429/6470 [1:07:15<13:35,  1.28it/s]
3|5430|Loss: 0.8664599657058716:  84%|████████▍ | 5429/6470 [1:07:16<13:35,  1.28it/s]
3|5430|Loss: 0.8664599657058716:  84%|████████▍ | 5430/6470 [1:07:16<13:17,  1.30it/s]
3|5431|Loss: 0.7196650505065918:  84%|████████▍ | 5430/6470 [1:07:16<13:17,  1.30it/s]
3|5431|Loss: 0.7196650505065918:  84%|████████▍ | 5431/6470 [1:07:16<12:32,  1.38it/s]
3|5432|Loss: 0.9598133563995361:  84%|████████▍ | 5431/6470 [1:07:17<12:32,  1.38it/s]
3|5432|Loss: 0.9598133563995361:  84%|████████▍ | 5432/6470 [1:07:17<12:41,  1.36it/s]
3|5433|Loss: 0.8170943260192871:  84%|████████▍ | 5432/6470 [1:07:18<12:41,  1.36it/s]
3|5433|Loss: 0.8170943260192871:  84%|████████▍ | 5433/6470 [1:07:18<12:34,  1.37it/s]
3|5434|Loss: 0.8652237057685852:  84%|████████▍ | 5433/6470 [1:07:18<12:34,  1.37it/s]
3|5434|Loss: 0.8652237057685852:  84%|████████▍ | 5434/6470 [1:07:19<12:37,  1.37it/s]
3|5435|Loss: 0.8155378103256226:  84%|████████▍ | 5434/6470 [1:07:19<12:37,  1.37it/s]
3|5435|Loss: 0.8155378103256226:  84%|████████▍ | 5435/6470 [1:07:19<12:21,  1.40it/s]
3|5436|Loss: 0.801730215549469:  84%|████████▍ | 5435/6470 [1:07:20<12:21,  1.40it/s] 
3|5436|Loss: 0.801730215549469:  84%|████████▍ | 5436/6470 [1:07:20<12:30,  1.38it/s]
3|5437|Loss: 0.8274211883544922:  84%|████████▍ | 5436/6470 [1:07:21<12:30,  1.38it/s]
3|5437|Loss: 0.8274211883544922:  84%|████████▍ | 5437/6470 [1:07:21<12:40,  1.36it/s]
3|5438|Loss: 0.8165016770362854:  84%|████████▍ | 5437/6470 [1:07:21<12:40,  1.36it/s]
3|5438|Loss: 0.8165016770362854:  84%|████████▍ | 5438/6470 [1:07:22<12:23,  1.39it/s]
3|5439|Loss: 0.6984634399414062:  84%|████████▍ | 5438/6470 [1:07:22<12:23,  1.39it/s]
3|5439|Loss: 0.6984634399414062:  84%|████████▍ | 5439/6470 [1:07:22<12:04,  1.42it/s]
3|5440|Loss: 0.8253887891769409:  84%|████████▍ | 5439/6470 [1:07:23<12:04,  1.42it/s]
3|5440|Loss: 0.8253887891769409:  84%|████████▍ | 5440/6470 [1:07:23<11:41,  1.47it/s]
3|5441|Loss: 0.7056848406791687:  84%|████████▍ | 5440/6470 [1:07:23<11:41,  1.47it/s]
3|5441|Loss: 0.7056848406791687:  84%|████████▍ | 5441/6470 [1:07:24<11:43,  1.46it/s]
3|5442|Loss: 0.9333658218383789:  84%|████████▍ | 5441/6470 [1:07:24<11:43,  1.46it/s]
3|5442|Loss: 0.9333658218383789:  84%|████████▍ | 5442/6470 [1:07:24<12:09,  1.41it/s]
3|5443|Loss: 0.8759000301361084:  84%|████████▍ | 5442/6470 [1:07:25<12:09,  1.41it/s]
3|5443|Loss: 0.8759000301361084:  84%|████████▍ | 5443/6470 [1:07:25<12:13,  1.40it/s]
3|5444|Loss: 1.0928269624710083:  84%|████████▍ | 5443/6470 [1:07:25<12:13,  1.40it/s]
3|5444|Loss: 1.0928269624710083:  84%|████████▍ | 5444/6470 [1:07:26<12:29,  1.37it/s]
3|5445|Loss: 0.8260892033576965:  84%|████████▍ | 5444/6470 [1:07:26<12:29,  1.37it/s]
3|5445|Loss: 0.8260892033576965:  84%|████████▍ | 5445/6470 [1:07:27<13:49,  1.24it/s]
3|5446|Loss: 0.7508469223976135:  84%|████████▍ | 5445/6470 [1:07:27<13:49,  1.24it/s]
3|5446|Loss: 0.7508469223976135:  84%|████████▍ | 5446/6470 [1:07:28<13:29,  1.26it/s]
3|5447|Loss: 0.7020463347434998:  84%|████████▍ | 5446/6470 [1:07:28<13:29,  1.26it/s]
3|5447|Loss: 0.7020463347434998:  84%|████████▍ | 5447/6470 [1:07:28<14:11,  1.20it/s]
3|5448|Loss: 0.829025387763977:  84%|████████▍ | 5447/6470 [1:07:29<14:11,  1.20it/s] 
3|5448|Loss: 0.829025387763977:  84%|████████▍ | 5448/6470 [1:07:29<14:45,  1.15it/s]
3|5449|Loss: 0.8233425617218018:  84%|████████▍ | 5448/6470 [1:07:30<14:45,  1.15it/s]
3|5449|Loss: 0.8233425617218018:  84%|████████▍ | 5449/6470 [1:07:30<15:13,  1.12it/s]
3|5450|Loss: 0.7031571269035339:  84%|████████▍ | 5449/6470 [1:07:31<15:13,  1.12it/s]
3|5450|Loss: 0.7031571269035339:  84%|████████▍ | 5450/6470 [1:07:31<15:26,  1.10it/s]
3|5451|Loss: 0.7417884469032288:  84%|████████▍ | 5450/6470 [1:07:32<15:26,  1.10it/s]
3|5451|Loss: 0.7417884469032288:  84%|████████▍ | 5451/6470 [1:07:32<14:00,  1.21it/s]
3|5452|Loss: 0.8189781904220581:  84%|████████▍ | 5451/6470 [1:07:32<14:00,  1.21it/s]
3|5452|Loss: 0.8189781904220581:  84%|████████▍ | 5452/6470 [1:07:33<13:31,  1.26it/s]
3|5453|Loss: 0.8107760548591614:  84%|████████▍ | 5452/6470 [1:07:33<13:31,  1.26it/s]
3|5453|Loss: 0.8107760548591614:  84%|████████▍ | 5453/6470 [1:07:33<13:21,  1.27it/s]
3|5454|Loss: 0.6765733957290649:  84%|████████▍ | 5453/6470 [1:07:34<13:21,  1.27it/s]
3|5454|Loss: 0.6765733957290649:  84%|████████▍ | 5454/6470 [1:07:34<12:38,  1.34it/s]
3|5455|Loss: 0.7856425642967224:  84%|████████▍ | 5454/6470 [1:07:34<12:38,  1.34it/s]
3|5455|Loss: 0.7856425642967224:  84%|████████▍ | 5455/6470 [1:07:35<12:30,  1.35it/s]
3|5456|Loss: 0.782050371170044:  84%|████████▍ | 5455/6470 [1:07:35<12:30,  1.35it/s] 
3|5456|Loss: 0.782050371170044:  84%|████████▍ | 5456/6470 [1:07:35<12:00,  1.41it/s]
3|5457|Loss: 0.7654432654380798:  84%|████████▍ | 5456/6470 [1:07:36<12:00,  1.41it/s]
3|5457|Loss: 0.7654432654380798:  84%|████████▍ | 5457/6470 [1:07:36<12:11,  1.39it/s]
3|5458|Loss: 0.8268731236457825:  84%|████████▍ | 5457/6470 [1:07:37<12:11,  1.39it/s]
3|5458|Loss: 0.8268731236457825:  84%|████████▍ | 5458/6470 [1:07:37<12:30,  1.35it/s]
3|5459|Loss: 0.9511795043945312:  84%|████████▍ | 5458/6470 [1:07:37<12:30,  1.35it/s]
3|5459|Loss: 0.9511795043945312:  84%|████████▍ | 5459/6470 [1:07:38<12:47,  1.32it/s]
3|5460|Loss: 1.0238430500030518:  84%|████████▍ | 5459/6470 [1:07:38<12:47,  1.32it/s]
3|5460|Loss: 1.0238430500030518:  84%|████████▍ | 5460/6470 [1:07:39<13:24,  1.26it/s]
3|5461|Loss: 0.7179597020149231:  84%|████████▍ | 5460/6470 [1:07:39<13:24,  1.26it/s]
3|5461|Loss: 0.7179597020149231:  84%|████████▍ | 5461/6470 [1:07:40<13:53,  1.21it/s]
3|5462|Loss: 0.8436183333396912:  84%|████████▍ | 5461/6470 [1:07:40<13:53,  1.21it/s]
3|5462|Loss: 0.8436183333396912:  84%|████████▍ | 5462/6470 [1:07:40<13:51,  1.21it/s]
3|5463|Loss: 0.8134759664535522:  84%|████████▍ | 5462/6470 [1:07:41<13:51,  1.21it/s]
3|5463|Loss: 0.8134759664535522:  84%|████████▍ | 5463/6470 [1:07:41<13:19,  1.26it/s]
3|5464|Loss: 0.7151497006416321:  84%|████████▍ | 5463/6470 [1:07:42<13:19,  1.26it/s]
3|5464|Loss: 0.7151497006416321:  84%|████████▍ | 5464/6470 [1:07:42<13:46,  1.22it/s]
3|5465|Loss: 0.7768887877464294:  84%|████████▍ | 5464/6470 [1:07:42<13:46,  1.22it/s]
3|5465|Loss: 0.7768887877464294:  84%|████████▍ | 5465/6470 [1:07:43<13:36,  1.23it/s]
3|5466|Loss: 0.8043411374092102:  84%|████████▍ | 5465/6470 [1:07:43<13:36,  1.23it/s]
3|5466|Loss: 0.8043411374092102:  84%|████████▍ | 5466/6470 [1:07:44<13:05,  1.28it/s]
3|5467|Loss: 0.9837685823440552:  84%|████████▍ | 5466/6470 [1:07:44<13:05,  1.28it/s]
3|5467|Loss: 0.9837685823440552:  84%|████████▍ | 5467/6470 [1:07:44<13:00,  1.28it/s]
3|5468|Loss: 0.8968878984451294:  84%|████████▍ | 5467/6470 [1:07:45<13:00,  1.28it/s]
3|5468|Loss: 0.8968878984451294:  85%|████████▍ | 5468/6470 [1:07:45<13:08,  1.27it/s]
3|5469|Loss: 0.7907426357269287:  85%|████████▍ | 5468/6470 [1:07:46<13:08,  1.27it/s]
3|5469|Loss: 0.7907426357269287:  85%|████████▍ | 5469/6470 [1:07:46<13:51,  1.20it/s]
3|5470|Loss: 0.8030675649642944:  85%|████████▍ | 5469/6470 [1:07:46<13:51,  1.20it/s]
3|5470|Loss: 0.8030675649642944:  85%|████████▍ | 5470/6470 [1:07:47<13:32,  1.23it/s]
3|5471|Loss: 0.8886171579360962:  85%|████████▍ | 5470/6470 [1:07:47<13:32,  1.23it/s]
3|5471|Loss: 0.8886171579360962:  85%|████████▍ | 5471/6470 [1:07:48<13:16,  1.25it/s]
3|5472|Loss: 0.8315118551254272:  85%|████████▍ | 5471/6470 [1:07:48<13:16,  1.25it/s]
3|5472|Loss: 0.8315118551254272:  85%|████████▍ | 5472/6470 [1:07:48<12:59,  1.28it/s]
3|5473|Loss: 0.6925633549690247:  85%|████████▍ | 5472/6470 [1:07:49<12:59,  1.28it/s]
3|5473|Loss: 0.6925633549690247:  85%|████████▍ | 5473/6470 [1:07:49<13:02,  1.27it/s]
3|5474|Loss: 0.9692617058753967:  85%|████████▍ | 5473/6470 [1:07:50<13:02,  1.27it/s]
3|5474|Loss: 0.9692617058753967:  85%|████████▍ | 5474/6470 [1:07:50<13:02,  1.27it/s]
3|5475|Loss: 0.9938751459121704:  85%|████████▍ | 5474/6470 [1:07:50<13:02,  1.27it/s]
3|5475|Loss: 0.9938751459121704:  85%|████████▍ | 5475/6470 [1:07:51<13:03,  1.27it/s]
3|5476|Loss: 0.8796696066856384:  85%|████████▍ | 5475/6470 [1:07:51<13:03,  1.27it/s]
3|5476|Loss: 0.8796696066856384:  85%|████████▍ | 5476/6470 [1:07:51<12:11,  1.36it/s]
3|5477|Loss: 0.8832072615623474:  85%|████████▍ | 5476/6470 [1:07:52<12:11,  1.36it/s]
3|5477|Loss: 0.8832072615623474:  85%|████████▍ | 5477/6470 [1:07:52<11:32,  1.43it/s]
3|5478|Loss: 0.7300158143043518:  85%|████████▍ | 5477/6470 [1:07:52<11:32,  1.43it/s]
3|5478|Loss: 0.7300158143043518:  85%|████████▍ | 5478/6470 [1:07:53<11:23,  1.45it/s]
3|5479|Loss: 0.8302967548370361:  85%|████████▍ | 5478/6470 [1:07:53<11:23,  1.45it/s]
3|5479|Loss: 0.8302967548370361:  85%|████████▍ | 5479/6470 [1:07:53<10:45,  1.54it/s]
3|5480|Loss: 0.8273422718048096:  85%|████████▍ | 5479/6470 [1:07:53<10:45,  1.54it/s]
3|5480|Loss: 0.8273422718048096:  85%|████████▍ | 5480/6470 [1:07:54<10:46,  1.53it/s]
3|5481|Loss: 0.902436375617981:  85%|████████▍ | 5480/6470 [1:07:54<10:46,  1.53it/s] 
3|5481|Loss: 0.902436375617981:  85%|████████▍ | 5481/6470 [1:07:55<11:57,  1.38it/s]
3|5482|Loss: 0.9828901290893555:  85%|████████▍ | 5481/6470 [1:07:55<11:57,  1.38it/s]
3|5482|Loss: 0.9828901290893555:  85%|████████▍ | 5482/6470 [1:07:55<12:01,  1.37it/s]
3|5483|Loss: 0.8919073343276978:  85%|████████▍ | 5482/6470 [1:07:56<12:01,  1.37it/s]
3|5483|Loss: 0.8919073343276978:  85%|████████▍ | 5483/6470 [1:07:56<11:36,  1.42it/s]
3|5484|Loss: 0.8072071671485901:  85%|████████▍ | 5483/6470 [1:07:56<11:36,  1.42it/s]
3|5484|Loss: 0.8072071671485901:  85%|████████▍ | 5484/6470 [1:07:57<11:20,  1.45it/s]
3|5485|Loss: 0.7712575793266296:  85%|████████▍ | 5484/6470 [1:07:57<11:20,  1.45it/s]
3|5485|Loss: 0.7712575793266296:  85%|████████▍ | 5485/6470 [1:07:57<11:39,  1.41it/s]
3|5486|Loss: 0.9243016242980957:  85%|████████▍ | 5485/6470 [1:07:58<11:39,  1.41it/s]
3|5486|Loss: 0.9243016242980957:  85%|████████▍ | 5486/6470 [1:07:58<11:53,  1.38it/s]
3|5487|Loss: 0.8505595922470093:  85%|████████▍ | 5486/6470 [1:07:59<11:53,  1.38it/s]
3|5487|Loss: 0.8505595922470093:  85%|████████▍ | 5487/6470 [1:07:59<12:09,  1.35it/s]
3|5488|Loss: 0.7481319308280945:  85%|████████▍ | 5487/6470 [1:07:59<12:09,  1.35it/s]
3|5488|Loss: 0.7481319308280945:  85%|████████▍ | 5488/6470 [1:08:00<11:29,  1.42it/s]
3|5489|Loss: 0.8717647790908813:  85%|████████▍ | 5488/6470 [1:08:00<11:29,  1.42it/s]
3|5489|Loss: 0.8717647790908813:  85%|████████▍ | 5489/6470 [1:08:00<11:30,  1.42it/s]
3|5490|Loss: 0.6267800331115723:  85%|████████▍ | 5489/6470 [1:08:01<11:30,  1.42it/s]
3|5490|Loss: 0.6267800331115723:  85%|████████▍ | 5490/6470 [1:08:01<11:43,  1.39it/s]
3|5491|Loss: 0.6454738974571228:  85%|████████▍ | 5490/6470 [1:08:01<11:43,  1.39it/s]
3|5491|Loss: 0.6454738974571228:  85%|████████▍ | 5491/6470 [1:08:01<10:11,  1.60it/s]
3|5492|Loss: 0.8160387277603149:  85%|████████▍ | 5491/6470 [1:08:02<10:11,  1.60it/s]
3|5492|Loss: 0.8160387277603149:  85%|████████▍ | 5492/6470 [1:08:02<10:10,  1.60it/s]
3|5493|Loss: 0.9011040925979614:  85%|████████▍ | 5492/6470 [1:08:03<10:10,  1.60it/s]
3|5493|Loss: 0.9011040925979614:  85%|████████▍ | 5493/6470 [1:08:03<10:24,  1.57it/s]
3|5494|Loss: 0.69877028465271:  85%|████████▍ | 5493/6470 [1:08:03<10:24,  1.57it/s]  
3|5494|Loss: 0.69877028465271:  85%|████████▍ | 5494/6470 [1:08:04<11:00,  1.48it/s]
3|5495|Loss: 0.8565325736999512:  85%|████████▍ | 5494/6470 [1:08:04<11:00,  1.48it/s]
3|5495|Loss: 0.8565325736999512:  85%|████████▍ | 5495/6470 [1:08:04<10:58,  1.48it/s]
3|5496|Loss: 0.7471067309379578:  85%|████████▍ | 5495/6470 [1:08:05<10:58,  1.48it/s]
3|5496|Loss: 0.7471067309379578:  85%|████████▍ | 5496/6470 [1:08:05<10:51,  1.50it/s]
3|5497|Loss: 0.9015852808952332:  85%|████████▍ | 5496/6470 [1:08:05<10:51,  1.50it/s]
3|5497|Loss: 0.9015852808952332:  85%|████████▍ | 5497/6470 [1:08:06<11:22,  1.43it/s]
3|5498|Loss: 0.8916326761245728:  85%|████████▍ | 5497/6470 [1:08:06<11:22,  1.43it/s]
3|5498|Loss: 0.8916326761245728:  85%|████████▍ | 5498/6470 [1:08:06<11:47,  1.37it/s]
3|5499|Loss: 0.7654622793197632:  85%|████████▍ | 5498/6470 [1:08:07<11:47,  1.37it/s]
3|5499|Loss: 0.7654622793197632:  85%|████████▍ | 5499/6470 [1:08:07<11:00,  1.47it/s]
3|5500|Loss: 0.932168185710907:  85%|████████▍ | 5499/6470 [1:08:07<11:00,  1.47it/s] 
3|5500|Loss: 0.932168185710907:  85%|████████▌ | 5500/6470 [1:08:08<11:41,  1.38it/s]
3|5501|Loss: 0.8627001643180847:  85%|████████▌ | 5500/6470 [1:08:08<11:41,  1.38it/s]
3|5501|Loss: 0.8627001643180847:  85%|████████▌ | 5501/6470 [1:08:09<12:09,  1.33it/s]
3|5502|Loss: 0.8271709084510803:  85%|████████▌ | 5501/6470 [1:08:09<12:09,  1.33it/s]
3|5502|Loss: 0.8271709084510803:  85%|████████▌ | 5502/6470 [1:08:09<11:48,  1.37it/s]
3|5503|Loss: 0.8974304795265198:  85%|████████▌ | 5502/6470 [1:08:10<11:48,  1.37it/s]
3|5503|Loss: 0.8974304795265198:  85%|████████▌ | 5503/6470 [1:08:10<11:44,  1.37it/s]
3|5504|Loss: 0.8767049908638:  85%|████████▌ | 5503/6470 [1:08:10<11:44,  1.37it/s]   
3|5504|Loss: 0.8767049908638:  85%|████████▌ | 5504/6470 [1:08:11<11:30,  1.40it/s]
3|5505|Loss: 1.0216025114059448:  85%|████████▌ | 5504/6470 [1:08:11<11:30,  1.40it/s]
3|5505|Loss: 1.0216025114059448:  85%|████████▌ | 5505/6470 [1:08:11<10:50,  1.48it/s]
3|5506|Loss: 0.7311517000198364:  85%|████████▌ | 5505/6470 [1:08:12<10:50,  1.48it/s]
3|5506|Loss: 0.7311517000198364:  85%|████████▌ | 5506/6470 [1:08:12<11:11,  1.43it/s]
3|5507|Loss: 0.9494199156761169:  85%|████████▌ | 5506/6470 [1:08:13<11:11,  1.43it/s]
3|5507|Loss: 0.9494199156761169:  85%|████████▌ | 5507/6470 [1:08:13<12:07,  1.32it/s]
3|5508|Loss: 0.907580554485321:  85%|████████▌ | 5507/6470 [1:08:13<12:07,  1.32it/s] 
3|5508|Loss: 0.907580554485321:  85%|████████▌ | 5508/6470 [1:08:14<12:48,  1.25it/s]
3|5509|Loss: 0.8264387249946594:  85%|████████▌ | 5508/6470 [1:08:14<12:48,  1.25it/s]
3|5509|Loss: 0.8264387249946594:  85%|████████▌ | 5509/6470 [1:08:15<12:29,  1.28it/s]
3|5510|Loss: 0.6956616044044495:  85%|████████▌ | 5509/6470 [1:08:15<12:29,  1.28it/s]
3|5510|Loss: 0.6956616044044495:  85%|████████▌ | 5510/6470 [1:08:15<12:26,  1.29it/s]
3|5511|Loss: 0.8228907585144043:  85%|████████▌ | 5510/6470 [1:08:16<12:26,  1.29it/s]
3|5511|Loss: 0.8228907585144043:  85%|████████▌ | 5511/6470 [1:08:16<12:46,  1.25it/s]
3|5512|Loss: 0.9468395113945007:  85%|████████▌ | 5511/6470 [1:08:17<12:46,  1.25it/s]
3|5512|Loss: 0.9468395113945007:  85%|████████▌ | 5512/6470 [1:08:17<12:28,  1.28it/s]
3|5513|Loss: 0.7374452352523804:  85%|████████▌ | 5512/6470 [1:08:17<12:28,  1.28it/s]
3|5513|Loss: 0.7374452352523804:  85%|████████▌ | 5513/6470 [1:08:18<12:47,  1.25it/s]
3|5514|Loss: 1.006069540977478:  85%|████████▌ | 5513/6470 [1:08:18<12:47,  1.25it/s] 
3|5514|Loss: 1.006069540977478:  85%|████████▌ | 5514/6470 [1:08:19<13:00,  1.22it/s]
3|5515|Loss: 0.7696642875671387:  85%|████████▌ | 5514/6470 [1:08:19<13:00,  1.22it/s]
3|5515|Loss: 0.7696642875671387:  85%|████████▌ | 5515/6470 [1:08:19<12:27,  1.28it/s]
3|5516|Loss: 0.9068235754966736:  85%|████████▌ | 5515/6470 [1:08:20<12:27,  1.28it/s]
3|5516|Loss: 0.9068235754966736:  85%|████████▌ | 5516/6470 [1:08:20<11:27,  1.39it/s]
3|5517|Loss: 0.782492458820343:  85%|████████▌ | 5516/6470 [1:08:20<11:27,  1.39it/s] 
3|5517|Loss: 0.782492458820343:  85%|████████▌ | 5517/6470 [1:08:21<11:04,  1.43it/s]
3|5518|Loss: 0.7831218242645264:  85%|████████▌ | 5517/6470 [1:08:21<11:04,  1.43it/s]
3|5518|Loss: 0.7831218242645264:  85%|████████▌ | 5518/6470 [1:08:21<11:15,  1.41it/s]
3|5519|Loss: 0.7674909830093384:  85%|████████▌ | 5518/6470 [1:08:22<11:15,  1.41it/s]
3|5519|Loss: 0.7674909830093384:  85%|████████▌ | 5519/6470 [1:08:22<10:41,  1.48it/s]
3|5520|Loss: 0.9809336066246033:  85%|████████▌ | 5519/6470 [1:08:22<10:41,  1.48it/s]
3|5520|Loss: 0.9809336066246033:  85%|████████▌ | 5520/6470 [1:08:23<10:40,  1.48it/s]
3|5521|Loss: 0.7813571691513062:  85%|████████▌ | 5520/6470 [1:08:23<10:40,  1.48it/s]
3|5521|Loss: 0.7813571691513062:  85%|████████▌ | 5521/6470 [1:08:23<11:44,  1.35it/s]
3|5522|Loss: 0.9974268078804016:  85%|████████▌ | 5521/6470 [1:08:24<11:44,  1.35it/s]
3|5522|Loss: 0.9974268078804016:  85%|████████▌ | 5522/6470 [1:08:24<11:29,  1.37it/s]
3|5523|Loss: 0.8107946515083313:  85%|████████▌ | 5522/6470 [1:08:25<11:29,  1.37it/s]
3|5523|Loss: 0.8107946515083313:  85%|████████▌ | 5523/6470 [1:08:25<11:28,  1.37it/s]
3|5524|Loss: 1.0159553289413452:  85%|████████▌ | 5523/6470 [1:08:25<11:28,  1.37it/s]
3|5524|Loss: 1.0159553289413452:  85%|████████▌ | 5524/6470 [1:08:26<11:16,  1.40it/s]
3|5525|Loss: 0.770085871219635:  85%|████████▌ | 5524/6470 [1:08:26<11:16,  1.40it/s] 
3|5525|Loss: 0.770085871219635:  85%|████████▌ | 5525/6470 [1:08:26<10:38,  1.48it/s]
3|5526|Loss: 1.0074909925460815:  85%|████████▌ | 5525/6470 [1:08:27<10:38,  1.48it/s]
3|5526|Loss: 1.0074909925460815:  85%|████████▌ | 5526/6470 [1:08:27<09:54,  1.59it/s]
3|5527|Loss: 0.8632051348686218:  85%|████████▌ | 5526/6470 [1:08:27<09:54,  1.59it/s]
3|5527|Loss: 0.8632051348686218:  85%|████████▌ | 5527/6470 [1:08:27<09:57,  1.58it/s]
3|5528|Loss: 0.9348007440567017:  85%|████████▌ | 5527/6470 [1:08:28<09:57,  1.58it/s]
3|5528|Loss: 0.9348007440567017:  85%|████████▌ | 5528/6470 [1:08:28<10:16,  1.53it/s]
3|5529|Loss: 0.889089047908783:  85%|████████▌ | 5528/6470 [1:08:29<10:16,  1.53it/s] 
3|5529|Loss: 0.889089047908783:  85%|████████▌ | 5529/6470 [1:08:29<11:12,  1.40it/s]
3|5530|Loss: 0.8294410109519958:  85%|████████▌ | 5529/6470 [1:08:29<11:12,  1.40it/s]
3|5530|Loss: 0.8294410109519958:  85%|████████▌ | 5530/6470 [1:08:30<10:55,  1.43it/s]
3|5531|Loss: 0.747484028339386:  85%|████████▌ | 5530/6470 [1:08:30<10:55,  1.43it/s] 
3|5531|Loss: 0.747484028339386:  85%|████████▌ | 5531/6470 [1:08:30<10:33,  1.48it/s]
3|5532|Loss: 0.7926545143127441:  85%|████████▌ | 5531/6470 [1:08:31<10:33,  1.48it/s]
3|5532|Loss: 0.7926545143127441:  86%|████████▌ | 5532/6470 [1:08:31<10:55,  1.43it/s]
3|5533|Loss: 0.7575901746749878:  86%|████████▌ | 5532/6470 [1:08:31<10:55,  1.43it/s]
3|5533|Loss: 0.7575901746749878:  86%|████████▌ | 5533/6470 [1:08:32<11:23,  1.37it/s]
3|5534|Loss: 0.7941823601722717:  86%|████████▌ | 5533/6470 [1:08:32<11:23,  1.37it/s]
3|5534|Loss: 0.7941823601722717:  86%|████████▌ | 5534/6470 [1:08:33<11:42,  1.33it/s]
3|5535|Loss: 0.751979410648346:  86%|████████▌ | 5534/6470 [1:08:33<11:42,  1.33it/s] 
3|5535|Loss: 0.751979410648346:  86%|████████▌ | 5535/6470 [1:08:33<11:56,  1.30it/s]
3|5536|Loss: 0.9533544778823853:  86%|████████▌ | 5535/6470 [1:08:34<11:56,  1.30it/s]
3|5536|Loss: 0.9533544778823853:  86%|████████▌ | 5536/6470 [1:08:34<11:56,  1.30it/s]
3|5537|Loss: 0.9718931317329407:  86%|████████▌ | 5536/6470 [1:08:35<11:56,  1.30it/s]
3|5537|Loss: 0.9718931317329407:  86%|████████▌ | 5537/6470 [1:08:35<12:03,  1.29it/s]
3|5538|Loss: 0.666501522064209:  86%|████████▌ | 5537/6470 [1:08:35<12:03,  1.29it/s] 
3|5538|Loss: 0.666501522064209:  86%|████████▌ | 5538/6470 [1:08:36<12:37,  1.23it/s]
3|5539|Loss: 0.7771748900413513:  86%|████████▌ | 5538/6470 [1:08:36<12:37,  1.23it/s]
3|5539|Loss: 0.7771748900413513:  86%|████████▌ | 5539/6470 [1:08:36<11:42,  1.33it/s]
3|5540|Loss: 0.813023567199707:  86%|████████▌ | 5539/6470 [1:08:37<11:42,  1.33it/s] 
3|5540|Loss: 0.813023567199707:  86%|████████▌ | 5540/6470 [1:08:37<11:26,  1.35it/s]
3|5541|Loss: 0.6987498998641968:  86%|████████▌ | 5540/6470 [1:08:38<11:26,  1.35it/s]
3|5541|Loss: 0.6987498998641968:  86%|████████▌ | 5541/6470 [1:08:38<12:05,  1.28it/s]
3|5542|Loss: 0.8327832221984863:  86%|████████▌ | 5541/6470 [1:08:39<12:05,  1.28it/s]
3|5542|Loss: 0.8327832221984863:  86%|████████▌ | 5542/6470 [1:08:39<12:53,  1.20it/s]
3|5543|Loss: 0.864177405834198:  86%|████████▌ | 5542/6470 [1:08:39<12:53,  1.20it/s] 
3|5543|Loss: 0.864177405834198:  86%|████████▌ | 5543/6470 [1:08:40<12:25,  1.24it/s]
3|5544|Loss: 0.8965653777122498:  86%|████████▌ | 5543/6470 [1:08:40<12:25,  1.24it/s]
3|5544|Loss: 0.8965653777122498:  86%|████████▌ | 5544/6470 [1:08:40<11:55,  1.29it/s]
3|5545|Loss: 0.7400754690170288:  86%|████████▌ | 5544/6470 [1:08:41<11:55,  1.29it/s]
3|5545|Loss: 0.7400754690170288:  86%|████████▌ | 5545/6470 [1:08:41<11:47,  1.31it/s]
3|5546|Loss: 0.7324946522712708:  86%|████████▌ | 5545/6470 [1:08:42<11:47,  1.31it/s]
3|5546|Loss: 0.7324946522712708:  86%|████████▌ | 5546/6470 [1:08:42<11:24,  1.35it/s]
3|5547|Loss: 0.8322948217391968:  86%|████████▌ | 5546/6470 [1:08:42<11:24,  1.35it/s]
3|5547|Loss: 0.8322948217391968:  86%|████████▌ | 5547/6470 [1:08:43<11:31,  1.33it/s]
3|5548|Loss: 0.8416375517845154:  86%|████████▌ | 5547/6470 [1:08:43<11:31,  1.33it/s]
3|5548|Loss: 0.8416375517845154:  86%|████████▌ | 5548/6470 [1:08:44<12:20,  1.25it/s]
3|5549|Loss: 0.7930915951728821:  86%|████████▌ | 5548/6470 [1:08:44<12:20,  1.25it/s]
3|5549|Loss: 0.7930915951728821:  86%|████████▌ | 5549/6470 [1:08:44<11:28,  1.34it/s]
3|5550|Loss: 0.7725328207015991:  86%|████████▌ | 5549/6470 [1:08:45<11:28,  1.34it/s]
3|5550|Loss: 0.7725328207015991:  86%|████████▌ | 5550/6470 [1:08:45<12:08,  1.26it/s]
3|5551|Loss: 0.6248995065689087:  86%|████████▌ | 5550/6470 [1:08:45<12:08,  1.26it/s]
3|5551|Loss: 0.6248995065689087:  86%|████████▌ | 5551/6470 [1:08:46<11:51,  1.29it/s]
3|5552|Loss: 1.0127525329589844:  86%|████████▌ | 5551/6470 [1:08:46<11:51,  1.29it/s]
3|5552|Loss: 1.0127525329589844:  86%|████████▌ | 5552/6470 [1:08:47<11:55,  1.28it/s]
3|5553|Loss: 0.8450098037719727:  86%|████████▌ | 5552/6470 [1:08:47<11:55,  1.28it/s]
3|5553|Loss: 0.8450098037719727:  86%|████████▌ | 5553/6470 [1:08:47<12:28,  1.23it/s]
3|5554|Loss: 0.8174941539764404:  86%|████████▌ | 5553/6470 [1:08:48<12:28,  1.23it/s]
3|5554|Loss: 0.8174941539764404:  86%|████████▌ | 5554/6470 [1:08:48<12:18,  1.24it/s]
3|5555|Loss: 0.861238420009613:  86%|████████▌ | 5554/6470 [1:08:49<12:18,  1.24it/s] 
3|5555|Loss: 0.861238420009613:  86%|████████▌ | 5555/6470 [1:08:49<12:06,  1.26it/s]
3|5556|Loss: 0.9256229996681213:  86%|████████▌ | 5555/6470 [1:08:49<12:06,  1.26it/s]
3|5556|Loss: 0.9256229996681213:  86%|████████▌ | 5556/6470 [1:08:50<10:56,  1.39it/s]
3|5557|Loss: 0.9375742673873901:  86%|████████▌ | 5556/6470 [1:08:50<10:56,  1.39it/s]
3|5557|Loss: 0.9375742673873901:  86%|████████▌ | 5557/6470 [1:08:50<10:22,  1.47it/s]
3|5558|Loss: 0.9326092600822449:  86%|████████▌ | 5557/6470 [1:08:51<10:22,  1.47it/s]
3|5558|Loss: 0.9326092600822449:  86%|████████▌ | 5558/6470 [1:08:51<09:58,  1.52it/s]
3|5559|Loss: 0.8350028991699219:  86%|████████▌ | 5558/6470 [1:08:51<09:58,  1.52it/s]
3|5559|Loss: 0.8350028991699219:  86%|████████▌ | 5559/6470 [1:08:51<09:58,  1.52it/s]
3|5560|Loss: 0.8121045231819153:  86%|████████▌ | 5559/6470 [1:08:52<09:58,  1.52it/s]
3|5560|Loss: 0.8121045231819153:  86%|████████▌ | 5560/6470 [1:08:52<09:45,  1.55it/s]
3|5561|Loss: 0.9264277219772339:  86%|████████▌ | 5560/6470 [1:08:52<09:45,  1.55it/s]
3|5561|Loss: 0.9264277219772339:  86%|████████▌ | 5561/6470 [1:08:53<10:14,  1.48it/s]
3|5562|Loss: 0.7005389332771301:  86%|████████▌ | 5561/6470 [1:08:53<10:14,  1.48it/s]
3|5562|Loss: 0.7005389332771301:  86%|████████▌ | 5562/6470 [1:08:54<11:15,  1.34it/s]
3|5563|Loss: 0.7863988876342773:  86%|████████▌ | 5562/6470 [1:08:54<11:15,  1.34it/s]
3|5563|Loss: 0.7863988876342773:  86%|████████▌ | 5563/6470 [1:08:54<11:14,  1.34it/s]
3|5564|Loss: 0.8023611307144165:  86%|████████▌ | 5563/6470 [1:08:55<11:14,  1.34it/s]
3|5564|Loss: 0.8023611307144165:  86%|████████▌ | 5564/6470 [1:08:55<11:23,  1.33it/s]
3|5565|Loss: 0.7702003717422485:  86%|████████▌ | 5564/6470 [1:08:56<11:23,  1.33it/s]
3|5565|Loss: 0.7702003717422485:  86%|████████▌ | 5565/6470 [1:08:56<11:23,  1.32it/s]
3|5566|Loss: 0.9675431251525879:  86%|████████▌ | 5565/6470 [1:08:56<11:23,  1.32it/s]
3|5566|Loss: 0.9675431251525879:  86%|████████▌ | 5566/6470 [1:08:57<11:30,  1.31it/s]
3|5567|Loss: 0.7996438145637512:  86%|████████▌ | 5566/6470 [1:08:57<11:30,  1.31it/s]
3|5567|Loss: 0.7996438145637512:  86%|████████▌ | 5567/6470 [1:08:58<12:22,  1.22it/s]
3|5568|Loss: 0.749657154083252:  86%|████████▌ | 5567/6470 [1:08:58<12:22,  1.22it/s] 
3|5568|Loss: 0.749657154083252:  86%|████████▌ | 5568/6470 [1:08:58<11:58,  1.26it/s]
3|5569|Loss: 0.6959528923034668:  86%|████████▌ | 5568/6470 [1:08:59<11:58,  1.26it/s]
3|5569|Loss: 0.6959528923034668:  86%|████████▌ | 5569/6470 [1:08:59<11:51,  1.27it/s]
3|5570|Loss: 0.6729341149330139:  86%|████████▌ | 5569/6470 [1:09:00<11:51,  1.27it/s]
3|5570|Loss: 0.6729341149330139:  86%|████████▌ | 5570/6470 [1:09:00<11:42,  1.28it/s]
3|5571|Loss: 0.8965227007865906:  86%|████████▌ | 5570/6470 [1:09:00<11:42,  1.28it/s]
3|5571|Loss: 0.8965227007865906:  86%|████████▌ | 5571/6470 [1:09:01<11:45,  1.27it/s]
3|5572|Loss: 0.8554496169090271:  86%|████████▌ | 5571/6470 [1:09:01<11:45,  1.27it/s]
3|5572|Loss: 0.8554496169090271:  86%|████████▌ | 5572/6470 [1:09:01<11:28,  1.30it/s]
3|5573|Loss: 0.8116257786750793:  86%|████████▌ | 5572/6470 [1:09:02<11:28,  1.30it/s]
3|5573|Loss: 0.8116257786750793:  86%|████████▌ | 5573/6470 [1:09:02<11:24,  1.31it/s]
3|5574|Loss: 0.8422272801399231:  86%|████████▌ | 5573/6470 [1:09:03<11:24,  1.31it/s]
3|5574|Loss: 0.8422272801399231:  86%|████████▌ | 5574/6470 [1:09:03<11:28,  1.30it/s]
3|5575|Loss: 0.9738143086433411:  86%|████████▌ | 5574/6470 [1:09:03<11:28,  1.30it/s]
3|5575|Loss: 0.9738143086433411:  86%|████████▌ | 5575/6470 [1:09:04<11:56,  1.25it/s]
3|5576|Loss: 0.8404185771942139:  86%|████████▌ | 5575/6470 [1:09:04<11:56,  1.25it/s]
3|5576|Loss: 0.8404185771942139:  86%|████████▌ | 5576/6470 [1:09:05<11:58,  1.24it/s]
3|5577|Loss: 0.9630873203277588:  86%|████████▌ | 5576/6470 [1:09:05<11:58,  1.24it/s]
3|5577|Loss: 0.9630873203277588:  86%|████████▌ | 5577/6470 [1:09:06<12:09,  1.22it/s]
3|5578|Loss: 0.8980468511581421:  86%|████████▌ | 5577/6470 [1:09:06<12:09,  1.22it/s]
3|5578|Loss: 0.8980468511581421:  86%|████████▌ | 5578/6470 [1:09:06<12:04,  1.23it/s]
3|5579|Loss: 0.7673581838607788:  86%|████████▌ | 5578/6470 [1:09:07<12:04,  1.23it/s]
3|5579|Loss: 0.7673581838607788:  86%|████████▌ | 5579/6470 [1:09:07<11:20,  1.31it/s]
3|5580|Loss: 0.7592079043388367:  86%|████████▌ | 5579/6470 [1:09:07<11:20,  1.31it/s]
3|5580|Loss: 0.7592079043388367:  86%|████████▌ | 5580/6470 [1:09:08<11:29,  1.29it/s]
3|5581|Loss: 0.908006489276886:  86%|████████▌ | 5580/6470 [1:09:08<11:29,  1.29it/s] 
3|5581|Loss: 0.908006489276886:  86%|████████▋ | 5581/6470 [1:09:08<11:00,  1.35it/s]
3|5582|Loss: 0.7399378418922424:  86%|████████▋ | 5581/6470 [1:09:09<11:00,  1.35it/s]
3|5582|Loss: 0.7399378418922424:  86%|████████▋ | 5582/6470 [1:09:09<11:05,  1.34it/s]
3|5583|Loss: 0.844731330871582:  86%|████████▋ | 5582/6470 [1:09:10<11:05,  1.34it/s] 
3|5583|Loss: 0.844731330871582:  86%|████████▋ | 5583/6470 [1:09:10<11:08,  1.33it/s]
3|5584|Loss: 0.7580645680427551:  86%|████████▋ | 5583/6470 [1:09:10<11:08,  1.33it/s]
3|5584|Loss: 0.7580645680427551:  86%|████████▋ | 5584/6470 [1:09:11<11:15,  1.31it/s]
3|5585|Loss: 0.8336628675460815:  86%|████████▋ | 5584/6470 [1:09:11<11:15,  1.31it/s]
3|5585|Loss: 0.8336628675460815:  86%|████████▋ | 5585/6470 [1:09:12<11:19,  1.30it/s]
3|5586|Loss: 0.8451324105262756:  86%|████████▋ | 5585/6470 [1:09:12<11:19,  1.30it/s]
3|5586|Loss: 0.8451324105262756:  86%|████████▋ | 5586/6470 [1:09:12<11:25,  1.29it/s]
3|5587|Loss: 0.8780200481414795:  86%|████████▋ | 5586/6470 [1:09:13<11:25,  1.29it/s]
3|5587|Loss: 0.8780200481414795:  86%|████████▋ | 5587/6470 [1:09:13<11:06,  1.33it/s]
3|5588|Loss: 0.7003939151763916:  86%|████████▋ | 5587/6470 [1:09:13<11:06,  1.33it/s]
3|5588|Loss: 0.7003939151763916:  86%|████████▋ | 5588/6470 [1:09:14<11:00,  1.34it/s]
3|5589|Loss: 0.7324291467666626:  86%|████████▋ | 5588/6470 [1:09:14<11:00,  1.34it/s]
3|5589|Loss: 0.7324291467666626:  86%|████████▋ | 5589/6470 [1:09:15<10:49,  1.36it/s]
3|5590|Loss: 0.7923386096954346:  86%|████████▋ | 5589/6470 [1:09:15<10:49,  1.36it/s]
3|5590|Loss: 0.7923386096954346:  86%|████████▋ | 5590/6470 [1:09:15<11:00,  1.33it/s]
3|5591|Loss: 0.7661603689193726:  86%|████████▋ | 5590/6470 [1:09:16<11:00,  1.33it/s]
3|5591|Loss: 0.7661603689193726:  86%|████████▋ | 5591/6470 [1:09:16<11:36,  1.26it/s]
3|5592|Loss: 0.9481463432312012:  86%|████████▋ | 5591/6470 [1:09:17<11:36,  1.26it/s]
3|5592|Loss: 0.9481463432312012:  86%|████████▋ | 5592/6470 [1:09:17<11:46,  1.24it/s]
3|5593|Loss: 0.7942606210708618:  86%|████████▋ | 5592/6470 [1:09:17<11:46,  1.24it/s]
3|5593|Loss: 0.7942606210708618:  86%|████████▋ | 5593/6470 [1:09:18<11:18,  1.29it/s]
3|5594|Loss: 0.9123871326446533:  86%|████████▋ | 5593/6470 [1:09:18<11:18,  1.29it/s]
3|5594|Loss: 0.9123871326446533:  86%|████████▋ | 5594/6470 [1:09:19<11:18,  1.29it/s]
3|5595|Loss: 0.9113479852676392:  86%|████████▋ | 5594/6470 [1:09:19<11:18,  1.29it/s]
3|5595|Loss: 0.9113479852676392:  86%|████████▋ | 5595/6470 [1:09:19<11:36,  1.26it/s]
3|5596|Loss: 0.8537185788154602:  86%|████████▋ | 5595/6470 [1:09:20<11:36,  1.26it/s]
3|5596|Loss: 0.8537185788154602:  86%|████████▋ | 5596/6470 [1:09:20<11:21,  1.28it/s]
3|5597|Loss: 0.8077622652053833:  86%|████████▋ | 5596/6470 [1:09:20<11:21,  1.28it/s]
3|5597|Loss: 0.8077622652053833:  87%|████████▋ | 5597/6470 [1:09:21<10:43,  1.36it/s]
3|5598|Loss: 0.5777038335800171:  87%|████████▋ | 5597/6470 [1:09:21<10:43,  1.36it/s]
3|5598|Loss: 0.5777038335800171:  87%|████████▋ | 5598/6470 [1:09:22<11:21,  1.28it/s]
3|5599|Loss: 0.8686818480491638:  87%|████████▋ | 5598/6470 [1:09:22<11:21,  1.28it/s]
3|5599|Loss: 0.8686818480491638:  87%|████████▋ | 5599/6470 [1:09:22<11:14,  1.29it/s]
3|5600|Loss: 0.9645112156867981:  87%|████████▋ | 5599/6470 [1:09:23<11:14,  1.29it/s]
3|5600|Loss: 0.9645112156867981:  87%|████████▋ | 5600/6470 [1:09:23<11:51,  1.22it/s]
3|5601|Loss: 0.9083114266395569:  87%|████████▋ | 5600/6470 [1:09:24<11:51,  1.22it/s]
3|5601|Loss: 0.9083114266395569:  87%|████████▋ | 5601/6470 [1:09:24<11:54,  1.22it/s]
3|5602|Loss: 0.7619534730911255:  87%|████████▋ | 5601/6470 [1:09:25<11:54,  1.22it/s]
3|5602|Loss: 0.7619534730911255:  87%|████████▋ | 5602/6470 [1:09:25<11:40,  1.24it/s]
3|5603|Loss: 0.7880566716194153:  87%|████████▋ | 5602/6470 [1:09:25<11:40,  1.24it/s]
3|5603|Loss: 0.7880566716194153:  87%|████████▋ | 5603/6470 [1:09:26<10:50,  1.33it/s]
3|5604|Loss: 0.8324109315872192:  87%|████████▋ | 5603/6470 [1:09:26<10:50,  1.33it/s]
3|5604|Loss: 0.8324109315872192:  87%|████████▋ | 5604/6470 [1:09:26<10:48,  1.33it/s]
3|5605|Loss: 0.9118067026138306:  87%|████████▋ | 5604/6470 [1:09:27<10:48,  1.33it/s]
3|5605|Loss: 0.9118067026138306:  87%|████████▋ | 5605/6470 [1:09:27<10:38,  1.35it/s]
3|5606|Loss: 0.9652546644210815:  87%|████████▋ | 5605/6470 [1:09:27<10:38,  1.35it/s]
3|5606|Loss: 0.9652546644210815:  87%|████████▋ | 5606/6470 [1:09:28<10:50,  1.33it/s]
3|5607|Loss: 0.8150027990341187:  87%|████████▋ | 5606/6470 [1:09:28<10:50,  1.33it/s]
3|5607|Loss: 0.8150027990341187:  87%|████████▋ | 5607/6470 [1:09:28<10:33,  1.36it/s]
3|5608|Loss: 0.8440036177635193:  87%|████████▋ | 5607/6470 [1:09:29<10:33,  1.36it/s]
3|5608|Loss: 0.8440036177635193:  87%|████████▋ | 5608/6470 [1:09:29<11:04,  1.30it/s]
3|5609|Loss: 0.8937707543373108:  87%|████████▋ | 5608/6470 [1:09:30<11:04,  1.30it/s]
3|5609|Loss: 0.8937707543373108:  87%|████████▋ | 5609/6470 [1:09:30<11:12,  1.28it/s]
3|5610|Loss: 0.8170245885848999:  87%|████████▋ | 5609/6470 [1:09:31<11:12,  1.28it/s]
3|5610|Loss: 0.8170245885848999:  87%|████████▋ | 5610/6470 [1:09:31<11:16,  1.27it/s]
3|5611|Loss: 0.7845504283905029:  87%|████████▋ | 5610/6470 [1:09:31<11:16,  1.27it/s]
3|5611|Loss: 0.7845504283905029:  87%|████████▋ | 5611/6470 [1:09:32<11:23,  1.26it/s]
3|5612|Loss: 0.8913268446922302:  87%|████████▋ | 5611/6470 [1:09:32<11:23,  1.26it/s]
3|5612|Loss: 0.8913268446922302:  87%|████████▋ | 5612/6470 [1:09:33<11:39,  1.23it/s]
3|5613|Loss: 0.8080453872680664:  87%|████████▋ | 5612/6470 [1:09:33<11:39,  1.23it/s]
3|5613|Loss: 0.8080453872680664:  87%|████████▋ | 5613/6470 [1:09:33<12:01,  1.19it/s]
3|5614|Loss: 0.7949906587600708:  87%|████████▋ | 5613/6470 [1:09:34<12:01,  1.19it/s]
3|5614|Loss: 0.7949906587600708:  87%|████████▋ | 5614/6470 [1:09:34<11:17,  1.26it/s]
3|5615|Loss: 0.8111718893051147:  87%|████████▋ | 5614/6470 [1:09:35<11:17,  1.26it/s]
3|5615|Loss: 0.8111718893051147:  87%|████████▋ | 5615/6470 [1:09:35<11:05,  1.29it/s]
3|5616|Loss: 0.764428436756134:  87%|████████▋ | 5615/6470 [1:09:35<11:05,  1.29it/s] 
3|5616|Loss: 0.764428436756134:  87%|████████▋ | 5616/6470 [1:09:36<10:32,  1.35it/s]
3|5617|Loss: 0.7975082397460938:  87%|████████▋ | 5616/6470 [1:09:36<10:32,  1.35it/s]
3|5617|Loss: 0.7975082397460938:  87%|████████▋ | 5617/6470 [1:09:36<09:36,  1.48it/s]
3|5618|Loss: 0.9610067009925842:  87%|████████▋ | 5617/6470 [1:09:37<09:36,  1.48it/s]
3|5618|Loss: 0.9610067009925842:  87%|████████▋ | 5618/6470 [1:09:37<10:19,  1.38it/s]
3|5619|Loss: 0.7979729771614075:  87%|████████▋ | 5618/6470 [1:09:37<10:19,  1.38it/s]
3|5619|Loss: 0.7979729771614075:  87%|████████▋ | 5619/6470 [1:09:38<10:23,  1.36it/s]
3|5620|Loss: 0.7867333292961121:  87%|████████▋ | 5619/6470 [1:09:38<10:23,  1.36it/s]
3|5620|Loss: 0.7867333292961121:  87%|████████▋ | 5620/6470 [1:09:39<10:54,  1.30it/s]
3|5621|Loss: 0.899324893951416:  87%|████████▋ | 5620/6470 [1:09:39<10:54,  1.30it/s] 
3|5621|Loss: 0.899324893951416:  87%|████████▋ | 5621/6470 [1:09:39<10:55,  1.30it/s]
3|5622|Loss: 0.9112223386764526:  87%|████████▋ | 5621/6470 [1:09:40<10:55,  1.30it/s]
3|5622|Loss: 0.9112223386764526:  87%|████████▋ | 5622/6470 [1:09:40<10:39,  1.33it/s]
3|5623|Loss: 0.9310588836669922:  87%|████████▋ | 5622/6470 [1:09:40<10:39,  1.33it/s]
3|5623|Loss: 0.9310588836669922:  87%|████████▋ | 5623/6470 [1:09:41<10:26,  1.35it/s]
3|5624|Loss: 0.8255906701087952:  87%|████████▋ | 5623/6470 [1:09:41<10:26,  1.35it/s]
3|5624|Loss: 0.8255906701087952:  87%|████████▋ | 5624/6470 [1:09:41<10:15,  1.37it/s]
3|5625|Loss: 0.8907720446586609:  87%|████████▋ | 5624/6470 [1:09:42<10:15,  1.37it/s]
3|5625|Loss: 0.8907720446586609:  87%|████████▋ | 5625/6470 [1:09:42<10:08,  1.39it/s]
3|5626|Loss: 0.7821717858314514:  87%|████████▋ | 5625/6470 [1:09:43<10:08,  1.39it/s]
3|5626|Loss: 0.7821717858314514:  87%|████████▋ | 5626/6470 [1:09:43<10:23,  1.35it/s]
3|5627|Loss: 0.9493449330329895:  87%|████████▋ | 5626/6470 [1:09:43<10:23,  1.35it/s]
3|5627|Loss: 0.9493449330329895:  87%|████████▋ | 5627/6470 [1:09:44<10:54,  1.29it/s]
3|5628|Loss: 0.6104283332824707:  87%|████████▋ | 5627/6470 [1:09:44<10:54,  1.29it/s]
3|5628|Loss: 0.6104283332824707:  87%|████████▋ | 5628/6470 [1:09:45<11:58,  1.17it/s]
3|5629|Loss: 0.8858354091644287:  87%|████████▋ | 5628/6470 [1:09:45<11:58,  1.17it/s]
3|5629|Loss: 0.8858354091644287:  87%|████████▋ | 5629/6470 [1:09:46<11:32,  1.21it/s]
3|5630|Loss: 0.9023616313934326:  87%|████████▋ | 5629/6470 [1:09:46<11:32,  1.21it/s]
3|5630|Loss: 0.9023616313934326:  87%|████████▋ | 5630/6470 [1:09:46<11:14,  1.25it/s]
3|5631|Loss: 0.7538372278213501:  87%|████████▋ | 5630/6470 [1:09:47<11:14,  1.25it/s]
3|5631|Loss: 0.7538372278213501:  87%|████████▋ | 5631/6470 [1:09:47<11:22,  1.23it/s]
3|5632|Loss: 0.8106086850166321:  87%|████████▋ | 5631/6470 [1:09:48<11:22,  1.23it/s]
3|5632|Loss: 0.8106086850166321:  87%|████████▋ | 5632/6470 [1:09:48<11:23,  1.23it/s]
3|5633|Loss: 0.8106414079666138:  87%|████████▋ | 5632/6470 [1:09:48<11:23,  1.23it/s]
3|5633|Loss: 0.8106414079666138:  87%|████████▋ | 5633/6470 [1:09:49<11:11,  1.25it/s]
3|5634|Loss: 0.7684277892112732:  87%|████████▋ | 5633/6470 [1:09:49<11:11,  1.25it/s]
3|5634|Loss: 0.7684277892112732:  87%|████████▋ | 5634/6470 [1:09:49<10:34,  1.32it/s]
3|5635|Loss: 0.8812546133995056:  87%|████████▋ | 5634/6470 [1:09:50<10:34,  1.32it/s]
3|5635|Loss: 0.8812546133995056:  87%|████████▋ | 5635/6470 [1:09:50<10:24,  1.34it/s]
3|5636|Loss: 0.9321708083152771:  87%|████████▋ | 5635/6470 [1:09:50<10:24,  1.34it/s]
3|5636|Loss: 0.9321708083152771:  87%|████████▋ | 5636/6470 [1:09:51<09:28,  1.47it/s]
3|5637|Loss: 0.8822738528251648:  87%|████████▋ | 5636/6470 [1:09:51<09:28,  1.47it/s]
3|5637|Loss: 0.8822738528251648:  87%|████████▋ | 5637/6470 [1:09:51<09:11,  1.51it/s]
3|5638|Loss: 0.7693132162094116:  87%|████████▋ | 5637/6470 [1:09:52<09:11,  1.51it/s]
3|5638|Loss: 0.7693132162094116:  87%|████████▋ | 5638/6470 [1:09:52<09:03,  1.53it/s]
3|5639|Loss: 0.7262718081474304:  87%|████████▋ | 5638/6470 [1:09:52<09:03,  1.53it/s]
3|5639|Loss: 0.7262718081474304:  87%|████████▋ | 5639/6470 [1:09:53<09:30,  1.46it/s]
3|5640|Loss: 0.87932288646698:  87%|████████▋ | 5639/6470 [1:09:53<09:30,  1.46it/s]  
3|5640|Loss: 0.87932288646698:  87%|████████▋ | 5640/6470 [1:09:54<10:11,  1.36it/s]
3|5641|Loss: 0.7311558723449707:  87%|████████▋ | 5640/6470 [1:09:54<10:11,  1.36it/s]
3|5641|Loss: 0.7311558723449707:  87%|████████▋ | 5641/6470 [1:09:54<10:09,  1.36it/s]
3|5642|Loss: 0.9779868125915527:  87%|████████▋ | 5641/6470 [1:09:55<10:09,  1.36it/s]
3|5642|Loss: 0.9779868125915527:  87%|████████▋ | 5642/6470 [1:09:55<10:39,  1.29it/s]
3|5643|Loss: 0.6664830446243286:  87%|████████▋ | 5642/6470 [1:09:56<10:39,  1.29it/s]
3|5643|Loss: 0.6664830446243286:  87%|████████▋ | 5643/6470 [1:09:56<11:15,  1.22it/s]
3|5644|Loss: 0.7918806076049805:  87%|████████▋ | 5643/6470 [1:09:57<11:15,  1.22it/s]
3|5644|Loss: 0.7918806076049805:  87%|████████▋ | 5644/6470 [1:09:57<11:14,  1.22it/s]
3|5645|Loss: 0.8073073625564575:  87%|████████▋ | 5644/6470 [1:09:57<11:14,  1.22it/s]
3|5645|Loss: 0.8073073625564575:  87%|████████▋ | 5645/6470 [1:09:58<11:00,  1.25it/s]
3|5646|Loss: 0.8493523597717285:  87%|████████▋ | 5645/6470 [1:09:58<11:00,  1.25it/s]
3|5646|Loss: 0.8493523597717285:  87%|████████▋ | 5646/6470 [1:09:58<10:14,  1.34it/s]
3|5647|Loss: 0.752878725528717:  87%|████████▋ | 5646/6470 [1:09:59<10:14,  1.34it/s] 
3|5647|Loss: 0.752878725528717:  87%|████████▋ | 5647/6470 [1:09:59<09:53,  1.39it/s]
3|5648|Loss: 0.8665387630462646:  87%|████████▋ | 5647/6470 [1:09:59<09:53,  1.39it/s]
3|5648|Loss: 0.8665387630462646:  87%|████████▋ | 5648/6470 [1:10:00<10:07,  1.35it/s]
3|5649|Loss: 0.8409281969070435:  87%|████████▋ | 5648/6470 [1:10:00<10:07,  1.35it/s]
3|5649|Loss: 0.8409281969070435:  87%|████████▋ | 5649/6470 [1:10:00<09:35,  1.43it/s]
3|5650|Loss: 0.8551235198974609:  87%|████████▋ | 5649/6470 [1:10:01<09:35,  1.43it/s]
3|5650|Loss: 0.8551235198974609:  87%|████████▋ | 5650/6470 [1:10:01<09:42,  1.41it/s]
3|5651|Loss: 0.8148152828216553:  87%|████████▋ | 5650/6470 [1:10:01<09:42,  1.41it/s]
3|5651|Loss: 0.8148152828216553:  87%|████████▋ | 5651/6470 [1:10:02<10:01,  1.36it/s]
3|5652|Loss: 1.0155476331710815:  87%|████████▋ | 5651/6470 [1:10:02<10:01,  1.36it/s]
3|5652|Loss: 1.0155476331710815:  87%|████████▋ | 5652/6470 [1:10:03<10:17,  1.33it/s]
3|5653|Loss: 0.8138912916183472:  87%|████████▋ | 5652/6470 [1:10:03<10:17,  1.33it/s]
3|5653|Loss: 0.8138912916183472:  87%|████████▋ | 5653/6470 [1:10:03<10:30,  1.30it/s]
3|5654|Loss: 0.9160040020942688:  87%|████████▋ | 5653/6470 [1:10:04<10:30,  1.30it/s]
3|5654|Loss: 0.9160040020942688:  87%|████████▋ | 5654/6470 [1:10:04<10:46,  1.26it/s]
3|5655|Loss: 0.8550726175308228:  87%|████████▋ | 5654/6470 [1:10:05<10:46,  1.26it/s]
3|5655|Loss: 0.8550726175308228:  87%|████████▋ | 5655/6470 [1:10:05<10:41,  1.27it/s]
3|5656|Loss: 0.8620397448539734:  87%|████████▋ | 5655/6470 [1:10:05<10:41,  1.27it/s]
3|5656|Loss: 0.8620397448539734:  87%|████████▋ | 5656/6470 [1:10:06<10:31,  1.29it/s]
3|5657|Loss: 0.6802298426628113:  87%|████████▋ | 5656/6470 [1:10:06<10:31,  1.29it/s]
3|5657|Loss: 0.6802298426628113:  87%|████████▋ | 5657/6470 [1:10:07<10:29,  1.29it/s]
3|5658|Loss: 0.8889875411987305:  87%|████████▋ | 5657/6470 [1:10:07<10:29,  1.29it/s]
3|5658|Loss: 0.8889875411987305:  87%|████████▋ | 5658/6470 [1:10:07<11:02,  1.23it/s]
3|5659|Loss: 0.881071150302887:  87%|████████▋ | 5658/6470 [1:10:08<11:02,  1.23it/s] 
3|5659|Loss: 0.881071150302887:  87%|████████▋ | 5659/6470 [1:10:08<11:26,  1.18it/s]
3|5660|Loss: 0.9947627782821655:  87%|████████▋ | 5659/6470 [1:10:09<11:26,  1.18it/s]
3|5660|Loss: 0.9947627782821655:  87%|████████▋ | 5660/6470 [1:10:09<11:15,  1.20it/s]
3|5661|Loss: 0.7425482869148254:  87%|████████▋ | 5660/6470 [1:10:10<11:15,  1.20it/s]
3|5661|Loss: 0.7425482869148254:  87%|████████▋ | 5661/6470 [1:10:10<10:38,  1.27it/s]
3|5662|Loss: 0.9277318120002747:  87%|████████▋ | 5661/6470 [1:10:10<10:38,  1.27it/s]
3|5662|Loss: 0.9277318120002747:  88%|████████▊ | 5662/6470 [1:10:11<10:24,  1.29it/s]
3|5663|Loss: 0.8237372636795044:  88%|████████▊ | 5662/6470 [1:10:11<10:24,  1.29it/s]
3|5663|Loss: 0.8237372636795044:  88%|████████▊ | 5663/6470 [1:10:11<10:09,  1.32it/s]
3|5664|Loss: 0.9603572487831116:  88%|████████▊ | 5663/6470 [1:10:12<10:09,  1.32it/s]
3|5664|Loss: 0.9603572487831116:  88%|████████▊ | 5664/6470 [1:10:12<10:05,  1.33it/s]
3|5665|Loss: 0.828245997428894:  88%|████████▊ | 5664/6470 [1:10:13<10:05,  1.33it/s] 
3|5665|Loss: 0.828245997428894:  88%|████████▊ | 5665/6470 [1:10:13<10:12,  1.31it/s]
3|5666|Loss: 0.9057058095932007:  88%|████████▊ | 5665/6470 [1:10:13<10:12,  1.31it/s]
3|5666|Loss: 0.9057058095932007:  88%|████████▊ | 5666/6470 [1:10:13<09:28,  1.41it/s]
3|5667|Loss: 0.8140195608139038:  88%|████████▊ | 5666/6470 [1:10:14<09:28,  1.41it/s]
3|5667|Loss: 0.8140195608139038:  88%|████████▊ | 5667/6470 [1:10:14<09:10,  1.46it/s]
3|5668|Loss: 0.7647539973258972:  88%|████████▊ | 5667/6470 [1:10:14<09:10,  1.46it/s]
3|5668|Loss: 0.7647539973258972:  88%|████████▊ | 5668/6470 [1:10:15<09:14,  1.45it/s]
3|5669|Loss: 0.9104233384132385:  88%|████████▊ | 5668/6470 [1:10:15<09:14,  1.45it/s]
3|5669|Loss: 0.9104233384132385:  88%|████████▊ | 5669/6470 [1:10:16<09:33,  1.40it/s]
3|5670|Loss: 1.0172288417816162:  88%|████████▊ | 5669/6470 [1:10:16<09:33,  1.40it/s]
3|5670|Loss: 1.0172288417816162:  88%|████████▊ | 5670/6470 [1:10:16<09:26,  1.41it/s]
3|5671|Loss: 0.7331544756889343:  88%|████████▊ | 5670/6470 [1:10:17<09:26,  1.41it/s]
3|5671|Loss: 0.7331544756889343:  88%|████████▊ | 5671/6470 [1:10:17<09:14,  1.44it/s]
3|5672|Loss: 0.8155228495597839:  88%|████████▊ | 5671/6470 [1:10:17<09:14,  1.44it/s]
3|5672|Loss: 0.8155228495597839:  88%|████████▊ | 5672/6470 [1:10:18<09:06,  1.46it/s]
3|5673|Loss: 0.8619222640991211:  88%|████████▊ | 5672/6470 [1:10:18<09:06,  1.46it/s]
3|5673|Loss: 0.8619222640991211:  88%|████████▊ | 5673/6470 [1:10:18<09:25,  1.41it/s]
3|5674|Loss: 0.7673032879829407:  88%|████████▊ | 5673/6470 [1:10:19<09:25,  1.41it/s]
3|5674|Loss: 0.7673032879829407:  88%|████████▊ | 5674/6470 [1:10:19<09:57,  1.33it/s]
3|5675|Loss: 0.7554075717926025:  88%|████████▊ | 5674/6470 [1:10:20<09:57,  1.33it/s]
3|5675|Loss: 0.7554075717926025:  88%|████████▊ | 5675/6470 [1:10:20<10:04,  1.31it/s]
3|5676|Loss: 0.8067120313644409:  88%|████████▊ | 5675/6470 [1:10:20<10:04,  1.31it/s]
3|5676|Loss: 0.8067120313644409:  88%|████████▊ | 5676/6470 [1:10:21<10:02,  1.32it/s]
3|5677|Loss: 0.8333507180213928:  88%|████████▊ | 5676/6470 [1:10:21<10:02,  1.32it/s]
3|5677|Loss: 0.8333507180213928:  88%|████████▊ | 5677/6470 [1:10:22<10:19,  1.28it/s]
3|5678|Loss: 0.9147619009017944:  88%|████████▊ | 5677/6470 [1:10:22<10:19,  1.28it/s]
3|5678|Loss: 0.9147619009017944:  88%|████████▊ | 5678/6470 [1:10:22<10:25,  1.27it/s]
3|5679|Loss: 0.83924800157547:  88%|████████▊ | 5678/6470 [1:10:23<10:25,  1.27it/s]  
3|5679|Loss: 0.83924800157547:  88%|████████▊ | 5679/6470 [1:10:23<10:43,  1.23it/s]
3|5680|Loss: 0.7514971494674683:  88%|████████▊ | 5679/6470 [1:10:24<10:43,  1.23it/s]
3|5680|Loss: 0.7514971494674683:  88%|████████▊ | 5680/6470 [1:10:24<10:18,  1.28it/s]
3|5681|Loss: 0.7291784882545471:  88%|████████▊ | 5680/6470 [1:10:24<10:18,  1.28it/s]
3|5681|Loss: 0.7291784882545471:  88%|████████▊ | 5681/6470 [1:10:25<10:14,  1.28it/s]
3|5682|Loss: 0.7245628833770752:  88%|████████▊ | 5681/6470 [1:10:25<10:14,  1.28it/s]
3|5682|Loss: 0.7245628833770752:  88%|████████▊ | 5682/6470 [1:10:26<10:46,  1.22it/s]
3|5683|Loss: 0.9480162262916565:  88%|████████▊ | 5682/6470 [1:10:26<10:46,  1.22it/s]
3|5683|Loss: 0.9480162262916565:  88%|████████▊ | 5683/6470 [1:10:26<10:24,  1.26it/s]
3|5684|Loss: 0.866645872592926:  88%|████████▊ | 5683/6470 [1:10:27<10:24,  1.26it/s] 
3|5684|Loss: 0.866645872592926:  88%|████████▊ | 5684/6470 [1:10:27<09:56,  1.32it/s]
3|5685|Loss: 0.8833159804344177:  88%|████████▊ | 5684/6470 [1:10:27<09:56,  1.32it/s]
3|5685|Loss: 0.8833159804344177:  88%|████████▊ | 5685/6470 [1:10:28<09:56,  1.32it/s]
3|5686|Loss: 0.7249528765678406:  88%|████████▊ | 5685/6470 [1:10:28<09:56,  1.32it/s]
3|5686|Loss: 0.7249528765678406:  88%|████████▊ | 5686/6470 [1:10:29<10:33,  1.24it/s]
3|5687|Loss: 0.735031247138977:  88%|████████▊ | 5686/6470 [1:10:29<10:33,  1.24it/s] 
3|5687|Loss: 0.735031247138977:  88%|████████▊ | 5687/6470 [1:10:30<10:53,  1.20it/s]
3|5688|Loss: 0.9787347316741943:  88%|████████▊ | 5687/6470 [1:10:30<10:53,  1.20it/s]
3|5688|Loss: 0.9787347316741943:  88%|████████▊ | 5688/6470 [1:10:30<09:32,  1.37it/s]
3|5689|Loss: 0.7903070449829102:  88%|████████▊ | 5688/6470 [1:10:30<09:32,  1.37it/s]
3|5689|Loss: 0.7903070449829102:  88%|████████▊ | 5689/6470 [1:10:31<09:28,  1.37it/s]
3|5690|Loss: 0.9684779047966003:  88%|████████▊ | 5689/6470 [1:10:31<09:28,  1.37it/s]
3|5690|Loss: 0.9684779047966003:  88%|████████▊ | 5690/6470 [1:10:32<09:44,  1.33it/s]
3|5691|Loss: 0.822095513343811:  88%|████████▊ | 5690/6470 [1:10:32<09:44,  1.33it/s] 
3|5691|Loss: 0.822095513343811:  88%|████████▊ | 5691/6470 [1:10:32<09:47,  1.33it/s]
3|5692|Loss: 0.6419202089309692:  88%|████████▊ | 5691/6470 [1:10:33<09:47,  1.33it/s]
3|5692|Loss: 0.6419202089309692:  88%|████████▊ | 5692/6470 [1:10:33<09:45,  1.33it/s]
3|5693|Loss: 0.8478471040725708:  88%|████████▊ | 5692/6470 [1:10:34<09:45,  1.33it/s]
3|5693|Loss: 0.8478471040725708:  88%|████████▊ | 5693/6470 [1:10:34<10:11,  1.27it/s]
3|5694|Loss: 0.8450415134429932:  88%|████████▊ | 5693/6470 [1:10:34<10:11,  1.27it/s]
3|5694|Loss: 0.8450415134429932:  88%|████████▊ | 5694/6470 [1:10:35<09:18,  1.39it/s]
3|5695|Loss: 0.8527026176452637:  88%|████████▊ | 5694/6470 [1:10:35<09:18,  1.39it/s]
3|5695|Loss: 0.8527026176452637:  88%|████████▊ | 5695/6470 [1:10:35<09:07,  1.42it/s]
3|5696|Loss: 0.7908025979995728:  88%|████████▊ | 5695/6470 [1:10:36<09:07,  1.42it/s]
3|5696|Loss: 0.7908025979995728:  88%|████████▊ | 5696/6470 [1:10:36<08:52,  1.45it/s]
3|5697|Loss: 0.8507299423217773:  88%|████████▊ | 5696/6470 [1:10:36<08:52,  1.45it/s]
3|5697|Loss: 0.8507299423217773:  88%|████████▊ | 5697/6470 [1:10:37<09:20,  1.38it/s]
3|5698|Loss: 0.8643012642860413:  88%|████████▊ | 5697/6470 [1:10:37<09:20,  1.38it/s]
3|5698|Loss: 0.8643012642860413:  88%|████████▊ | 5698/6470 [1:10:37<09:15,  1.39it/s]
3|5699|Loss: 0.9566051363945007:  88%|████████▊ | 5698/6470 [1:10:38<09:15,  1.39it/s]
3|5699|Loss: 0.9566051363945007:  88%|████████▊ | 5699/6470 [1:10:38<09:21,  1.37it/s]
3|5700|Loss: 0.88996821641922:  88%|████████▊ | 5699/6470 [1:10:39<09:21,  1.37it/s]  
3|5700|Loss: 0.88996821641922:  88%|████████▊ | 5700/6470 [1:10:39<09:32,  1.34it/s]
3|5701|Loss: 0.8498596549034119:  88%|████████▊ | 5700/6470 [1:10:39<09:32,  1.34it/s]
3|5701|Loss: 0.8498596549034119:  88%|████████▊ | 5701/6470 [1:10:40<09:18,  1.38it/s]
3|5702|Loss: 0.7529922127723694:  88%|████████▊ | 5701/6470 [1:10:40<09:18,  1.38it/s]
3|5702|Loss: 0.7529922127723694:  88%|████████▊ | 5702/6470 [1:10:41<09:55,  1.29it/s]
3|5703|Loss: 0.8431762456893921:  88%|████████▊ | 5702/6470 [1:10:41<09:55,  1.29it/s]
3|5703|Loss: 0.8431762456893921:  88%|████████▊ | 5703/6470 [1:10:41<10:10,  1.26it/s]
3|5704|Loss: 0.7436293363571167:  88%|████████▊ | 5703/6470 [1:10:42<10:10,  1.26it/s]
3|5704|Loss: 0.7436293363571167:  88%|████████▊ | 5704/6470 [1:10:42<10:40,  1.20it/s]
3|5705|Loss: 0.7220104336738586:  88%|████████▊ | 5704/6470 [1:10:43<10:40,  1.20it/s]
3|5705|Loss: 0.7220104336738586:  88%|████████▊ | 5705/6470 [1:10:43<10:49,  1.18it/s]
3|5706|Loss: 0.7962945103645325:  88%|████████▊ | 5705/6470 [1:10:44<10:49,  1.18it/s]
3|5706|Loss: 0.7962945103645325:  88%|████████▊ | 5706/6470 [1:10:44<10:21,  1.23it/s]
3|5707|Loss: 1.0228683948516846:  88%|████████▊ | 5706/6470 [1:10:44<10:21,  1.23it/s]
3|5707|Loss: 1.0228683948516846:  88%|████████▊ | 5707/6470 [1:10:45<10:12,  1.25it/s]
3|5708|Loss: 0.8533008098602295:  88%|████████▊ | 5707/6470 [1:10:45<10:12,  1.25it/s]
3|5708|Loss: 0.8533008098602295:  88%|████████▊ | 5708/6470 [1:10:45<09:55,  1.28it/s]
3|5709|Loss: 0.8772225975990295:  88%|████████▊ | 5708/6470 [1:10:46<09:55,  1.28it/s]
3|5709|Loss: 0.8772225975990295:  88%|████████▊ | 5709/6470 [1:10:46<09:06,  1.39it/s]
3|5710|Loss: 0.7610923647880554:  88%|████████▊ | 5709/6470 [1:10:46<09:06,  1.39it/s]
3|5710|Loss: 0.7610923647880554:  88%|████████▊ | 5710/6470 [1:10:47<09:51,  1.28it/s]
3|5711|Loss: 0.7935148477554321:  88%|████████▊ | 5710/6470 [1:10:47<09:51,  1.28it/s]
3|5711|Loss: 0.7935148477554321:  88%|████████▊ | 5711/6470 [1:10:48<10:03,  1.26it/s]
3|5712|Loss: 0.8216103911399841:  88%|████████▊ | 5711/6470 [1:10:48<10:03,  1.26it/s]
3|5712|Loss: 0.8216103911399841:  88%|████████▊ | 5712/6470 [1:10:48<09:44,  1.30it/s]
3|5713|Loss: 0.8548629283905029:  88%|████████▊ | 5712/6470 [1:10:49<09:44,  1.30it/s]
3|5713|Loss: 0.8548629283905029:  88%|████████▊ | 5713/6470 [1:10:49<09:26,  1.34it/s]
3|5714|Loss: 0.7809113264083862:  88%|████████▊ | 5713/6470 [1:10:50<09:26,  1.34it/s]
3|5714|Loss: 0.7809113264083862:  88%|████████▊ | 5714/6470 [1:10:50<09:12,  1.37it/s]
3|5715|Loss: 0.7938545346260071:  88%|████████▊ | 5714/6470 [1:10:50<09:12,  1.37it/s]
3|5715|Loss: 0.7938545346260071:  88%|████████▊ | 5715/6470 [1:10:51<09:15,  1.36it/s]
3|5716|Loss: 0.9632914662361145:  88%|████████▊ | 5715/6470 [1:10:51<09:15,  1.36it/s]
3|5716|Loss: 0.9632914662361145:  88%|████████▊ | 5716/6470 [1:10:51<09:27,  1.33it/s]
3|5717|Loss: 0.9200531244277954:  88%|████████▊ | 5716/6470 [1:10:52<09:27,  1.33it/s]
3|5717|Loss: 0.9200531244277954:  88%|████████▊ | 5717/6470 [1:10:52<09:36,  1.31it/s]
3|5718|Loss: 0.9209177494049072:  88%|████████▊ | 5717/6470 [1:10:53<09:36,  1.31it/s]
3|5718|Loss: 0.9209177494049072:  88%|████████▊ | 5718/6470 [1:10:53<09:34,  1.31it/s]
3|5719|Loss: 0.7917988896369934:  88%|████████▊ | 5718/6470 [1:10:53<09:34,  1.31it/s]
3|5719|Loss: 0.7917988896369934:  88%|████████▊ | 5719/6470 [1:10:54<10:01,  1.25it/s]
3|5720|Loss: 0.6976907849311829:  88%|████████▊ | 5719/6470 [1:10:54<10:01,  1.25it/s]
3|5720|Loss: 0.6976907849311829:  88%|████████▊ | 5720/6470 [1:10:55<10:28,  1.19it/s]
3|5721|Loss: 0.7518146634101868:  88%|████████▊ | 5720/6470 [1:10:55<10:28,  1.19it/s]
3|5721|Loss: 0.7518146634101868:  88%|████████▊ | 5721/6470 [1:10:56<10:48,  1.16it/s]
3|5722|Loss: 0.8523223996162415:  88%|████████▊ | 5721/6470 [1:10:56<10:48,  1.16it/s]
3|5722|Loss: 0.8523223996162415:  88%|████████▊ | 5722/6470 [1:10:56<10:20,  1.21it/s]
3|5723|Loss: 0.9532648324966431:  88%|████████▊ | 5722/6470 [1:10:57<10:20,  1.21it/s]
3|5723|Loss: 0.9532648324966431:  88%|████████▊ | 5723/6470 [1:10:57<09:48,  1.27it/s]
3|5724|Loss: 0.819512665271759:  88%|████████▊ | 5723/6470 [1:10:58<09:48,  1.27it/s] 
3|5724|Loss: 0.819512665271759:  88%|████████▊ | 5724/6470 [1:10:58<09:42,  1.28it/s]
3|5725|Loss: 0.8330316543579102:  88%|████████▊ | 5724/6470 [1:10:58<09:42,  1.28it/s]
3|5725|Loss: 0.8330316543579102:  88%|████████▊ | 5725/6470 [1:10:59<09:07,  1.36it/s]
3|5726|Loss: 0.8120405077934265:  88%|████████▊ | 5725/6470 [1:10:59<09:07,  1.36it/s]
3|5726|Loss: 0.8120405077934265:  89%|████████▊ | 5726/6470 [1:10:59<08:51,  1.40it/s]
3|5727|Loss: 1.0630896091461182:  89%|████████▊ | 5726/6470 [1:11:00<08:51,  1.40it/s]
3|5727|Loss: 1.0630896091461182:  89%|████████▊ | 5727/6470 [1:11:00<08:33,  1.45it/s]
3|5728|Loss: 0.8706616163253784:  89%|████████▊ | 5727/6470 [1:11:00<08:33,  1.45it/s]
3|5728|Loss: 0.8706616163253784:  89%|████████▊ | 5728/6470 [1:11:00<08:17,  1.49it/s]
3|5729|Loss: 0.8648091554641724:  89%|████████▊ | 5728/6470 [1:11:01<08:17,  1.49it/s]
3|5729|Loss: 0.8648091554641724:  89%|████████▊ | 5729/6470 [1:11:01<07:42,  1.60it/s]
3|5730|Loss: 0.7751093506813049:  89%|████████▊ | 5729/6470 [1:11:01<07:42,  1.60it/s]
3|5730|Loss: 0.7751093506813049:  89%|████████▊ | 5730/6470 [1:11:02<08:13,  1.50it/s]
3|5731|Loss: 0.8152678608894348:  89%|████████▊ | 5730/6470 [1:11:02<08:13,  1.50it/s]
3|5731|Loss: 0.8152678608894348:  89%|████████▊ | 5731/6470 [1:11:03<08:43,  1.41it/s]
3|5732|Loss: 0.9233446717262268:  89%|████████▊ | 5731/6470 [1:11:03<08:43,  1.41it/s]
3|5732|Loss: 0.9233446717262268:  89%|████████▊ | 5732/6470 [1:11:03<08:57,  1.37it/s]
3|5733|Loss: 0.6959646940231323:  89%|████████▊ | 5732/6470 [1:11:04<08:57,  1.37it/s]
3|5733|Loss: 0.6959646940231323:  89%|████████▊ | 5733/6470 [1:11:04<09:01,  1.36it/s]
3|5734|Loss: 0.8923594951629639:  89%|████████▊ | 5733/6470 [1:11:04<09:01,  1.36it/s]
3|5734|Loss: 0.8923594951629639:  89%|████████▊ | 5734/6470 [1:11:05<08:16,  1.48it/s]
3|5735|Loss: 0.7982501983642578:  89%|████████▊ | 5734/6470 [1:11:05<08:16,  1.48it/s]
3|5735|Loss: 0.7982501983642578:  89%|████████▊ | 5735/6470 [1:11:05<08:09,  1.50it/s]
3|5736|Loss: 0.7679159045219421:  89%|████████▊ | 5735/6470 [1:11:06<08:09,  1.50it/s]
3|5736|Loss: 0.7679159045219421:  89%|████████▊ | 5736/6470 [1:11:06<08:18,  1.47it/s]
3|5737|Loss: 0.8521478176116943:  89%|████████▊ | 5736/6470 [1:11:06<08:18,  1.47it/s]
3|5737|Loss: 0.8521478176116943:  89%|████████▊ | 5737/6470 [1:11:07<08:40,  1.41it/s]
3|5738|Loss: 0.8227498531341553:  89%|████████▊ | 5737/6470 [1:11:07<08:40,  1.41it/s]
3|5738|Loss: 0.8227498531341553:  89%|████████▊ | 5738/6470 [1:11:07<08:50,  1.38it/s]
3|5739|Loss: 0.8226947784423828:  89%|████████▊ | 5738/6470 [1:11:08<08:50,  1.38it/s]
3|5739|Loss: 0.8226947784423828:  89%|████████▊ | 5739/6470 [1:11:08<08:45,  1.39it/s]
3|5740|Loss: 0.9772764444351196:  89%|████████▊ | 5739/6470 [1:11:09<08:45,  1.39it/s]
3|5740|Loss: 0.9772764444351196:  89%|████████▊ | 5740/6470 [1:11:09<09:10,  1.33it/s]
3|5741|Loss: 0.7429656982421875:  89%|████████▊ | 5740/6470 [1:11:09<09:10,  1.33it/s]
3|5741|Loss: 0.7429656982421875:  89%|████████▊ | 5741/6470 [1:11:10<09:01,  1.35it/s]
3|5742|Loss: 0.8547167778015137:  89%|████████▊ | 5741/6470 [1:11:10<09:01,  1.35it/s]
3|5742|Loss: 0.8547167778015137:  89%|████████▊ | 5742/6470 [1:11:10<08:59,  1.35it/s]
3|5743|Loss: 0.9577895402908325:  89%|████████▊ | 5742/6470 [1:11:11<08:59,  1.35it/s]
3|5743|Loss: 0.9577895402908325:  89%|████████▉ | 5743/6470 [1:11:11<09:07,  1.33it/s]
3|5744|Loss: 0.8116477131843567:  89%|████████▉ | 5743/6470 [1:11:12<09:07,  1.33it/s]
3|5744|Loss: 0.8116477131843567:  89%|████████▉ | 5744/6470 [1:11:12<09:43,  1.24it/s]
3|5745|Loss: 0.7215553522109985:  89%|████████▉ | 5744/6470 [1:11:13<09:43,  1.24it/s]
3|5745|Loss: 0.7215553522109985:  89%|████████▉ | 5745/6470 [1:11:13<09:51,  1.22it/s]
3|5746|Loss: 0.9918602108955383:  89%|████████▉ | 5745/6470 [1:11:13<09:51,  1.22it/s]
3|5746|Loss: 0.9918602108955383:  89%|████████▉ | 5746/6470 [1:11:14<09:28,  1.27it/s]
3|5747|Loss: 0.8875714540481567:  89%|████████▉ | 5746/6470 [1:11:14<09:28,  1.27it/s]
3|5747|Loss: 0.8875714540481567:  89%|████████▉ | 5747/6470 [1:11:15<09:25,  1.28it/s]
3|5748|Loss: 0.7649555802345276:  89%|████████▉ | 5747/6470 [1:11:15<09:25,  1.28it/s]
3|5748|Loss: 0.7649555802345276:  89%|████████▉ | 5748/6470 [1:11:15<08:54,  1.35it/s]
3|5749|Loss: 0.7707714438438416:  89%|████████▉ | 5748/6470 [1:11:16<08:54,  1.35it/s]
3|5749|Loss: 0.7707714438438416:  89%|████████▉ | 5749/6470 [1:11:16<08:46,  1.37it/s]
3|5750|Loss: 1.033145785331726:  89%|████████▉ | 5749/6470 [1:11:16<08:46,  1.37it/s] 
3|5750|Loss: 1.033145785331726:  89%|████████▉ | 5750/6470 [1:11:17<08:55,  1.35it/s]
3|5751|Loss: 0.8039509057998657:  89%|████████▉ | 5750/6470 [1:11:17<08:55,  1.35it/s]
3|5751|Loss: 0.8039509057998657:  89%|████████▉ | 5751/6470 [1:11:17<08:43,  1.37it/s]
3|5752|Loss: 0.8345794081687927:  89%|████████▉ | 5751/6470 [1:11:18<08:43,  1.37it/s]
3|5752|Loss: 0.8345794081687927:  89%|████████▉ | 5752/6470 [1:11:18<08:43,  1.37it/s]
3|5753|Loss: 0.8084807991981506:  89%|████████▉ | 5752/6470 [1:11:18<08:43,  1.37it/s]
3|5753|Loss: 0.8084807991981506:  89%|████████▉ | 5753/6470 [1:11:19<08:36,  1.39it/s]
3|5754|Loss: 0.7562466263771057:  89%|████████▉ | 5753/6470 [1:11:19<08:36,  1.39it/s]
3|5754|Loss: 0.7562466263771057:  89%|████████▉ | 5754/6470 [1:11:20<08:46,  1.36it/s]
3|5755|Loss: 0.742577314376831:  89%|████████▉ | 5754/6470 [1:11:20<08:46,  1.36it/s] 
3|5755|Loss: 0.742577314376831:  89%|████████▉ | 5755/6470 [1:11:20<08:41,  1.37it/s]
3|5756|Loss: 0.8693917393684387:  89%|████████▉ | 5755/6470 [1:11:21<08:41,  1.37it/s]
3|5756|Loss: 0.8693917393684387:  89%|████████▉ | 5756/6470 [1:11:21<08:28,  1.40it/s]
3|5757|Loss: 0.828120768070221:  89%|████████▉ | 5756/6470 [1:11:21<08:28,  1.40it/s] 
3|5757|Loss: 0.828120768070221:  89%|████████▉ | 5757/6470 [1:11:22<09:00,  1.32it/s]
3|5758|Loss: 0.8415476679801941:  89%|████████▉ | 5757/6470 [1:11:22<09:00,  1.32it/s]
3|5758|Loss: 0.8415476679801941:  89%|████████▉ | 5758/6470 [1:11:23<09:08,  1.30it/s]
3|5759|Loss: 0.8107328414916992:  89%|████████▉ | 5758/6470 [1:11:23<09:08,  1.30it/s]
3|5759|Loss: 0.8107328414916992:  89%|████████▉ | 5759/6470 [1:11:23<09:05,  1.30it/s]
3|5760|Loss: 0.7268959879875183:  89%|████████▉ | 5759/6470 [1:11:24<09:05,  1.30it/s]
3|5760|Loss: 0.7268959879875183:  89%|████████▉ | 5760/6470 [1:11:24<09:17,  1.27it/s]
3|5761|Loss: 0.841080367565155:  89%|████████▉ | 5760/6470 [1:11:25<09:17,  1.27it/s] 
3|5761|Loss: 0.841080367565155:  89%|████████▉ | 5761/6470 [1:11:25<09:04,  1.30it/s]
3|5762|Loss: 0.7928211092948914:  89%|████████▉ | 5761/6470 [1:11:25<09:04,  1.30it/s]
3|5762|Loss: 0.7928211092948914:  89%|████████▉ | 5762/6470 [1:11:26<09:02,  1.31it/s]
3|5763|Loss: 0.8986294865608215:  89%|████████▉ | 5762/6470 [1:11:26<09:02,  1.31it/s]
3|5763|Loss: 0.8986294865608215:  89%|████████▉ | 5763/6470 [1:11:26<08:23,  1.41it/s]
3|5764|Loss: 0.7005493640899658:  89%|████████▉ | 5763/6470 [1:11:27<08:23,  1.41it/s]
3|5764|Loss: 0.7005493640899658:  89%|████████▉ | 5764/6470 [1:11:27<07:34,  1.55it/s]
3|5765|Loss: 0.7423097491264343:  89%|████████▉ | 5764/6470 [1:11:27<07:34,  1.55it/s]
3|5765|Loss: 0.7423097491264343:  89%|████████▉ | 5765/6470 [1:11:27<07:49,  1.50it/s]
3|5766|Loss: 0.7181217670440674:  89%|████████▉ | 5765/6470 [1:11:28<07:49,  1.50it/s]
3|5766|Loss: 0.7181217670440674:  89%|████████▉ | 5766/6470 [1:11:28<08:14,  1.42it/s]
3|5767|Loss: 0.7510905861854553:  89%|████████▉ | 5766/6470 [1:11:29<08:14,  1.42it/s]
3|5767|Loss: 0.7510905861854553:  89%|████████▉ | 5767/6470 [1:11:29<08:03,  1.46it/s]
3|5768|Loss: 0.7008189558982849:  89%|████████▉ | 5767/6470 [1:11:29<08:03,  1.46it/s]
3|5768|Loss: 0.7008189558982849:  89%|████████▉ | 5768/6470 [1:11:30<08:14,  1.42it/s]
3|5769|Loss: 0.859708845615387:  89%|████████▉ | 5768/6470 [1:11:30<08:14,  1.42it/s] 
3|5769|Loss: 0.859708845615387:  89%|████████▉ | 5769/6470 [1:11:30<07:45,  1.51it/s]
3|5770|Loss: 0.7679169774055481:  89%|████████▉ | 5769/6470 [1:11:31<07:45,  1.51it/s]
3|5770|Loss: 0.7679169774055481:  89%|████████▉ | 5770/6470 [1:11:31<08:12,  1.42it/s]
3|5771|Loss: 0.8944680690765381:  89%|████████▉ | 5770/6470 [1:11:31<08:12,  1.42it/s]
3|5771|Loss: 0.8944680690765381:  89%|████████▉ | 5771/6470 [1:11:32<07:54,  1.47it/s]
3|5772|Loss: 0.7577436566352844:  89%|████████▉ | 5771/6470 [1:11:32<07:54,  1.47it/s]
3|5772|Loss: 0.7577436566352844:  89%|████████▉ | 5772/6470 [1:11:32<07:45,  1.50it/s]
3|5773|Loss: 0.9072701334953308:  89%|████████▉ | 5772/6470 [1:11:33<07:45,  1.50it/s]
3|5773|Loss: 0.9072701334953308:  89%|████████▉ | 5773/6470 [1:11:33<07:50,  1.48it/s]
3|5774|Loss: 0.9145084023475647:  89%|████████▉ | 5773/6470 [1:11:33<07:50,  1.48it/s]
3|5774|Loss: 0.9145084023475647:  89%|████████▉ | 5774/6470 [1:11:34<07:52,  1.47it/s]
3|5775|Loss: 0.8375527262687683:  89%|████████▉ | 5774/6470 [1:11:34<07:52,  1.47it/s]
3|5775|Loss: 0.8375527262687683:  89%|████████▉ | 5775/6470 [1:11:34<08:05,  1.43it/s]
3|5776|Loss: 0.9125488996505737:  89%|████████▉ | 5775/6470 [1:11:35<08:05,  1.43it/s]
3|5776|Loss: 0.9125488996505737:  89%|████████▉ | 5776/6470 [1:11:35<08:01,  1.44it/s]
3|5777|Loss: 0.7472397685050964:  89%|████████▉ | 5776/6470 [1:11:35<08:01,  1.44it/s]
3|5777|Loss: 0.7472397685050964:  89%|████████▉ | 5777/6470 [1:11:36<07:22,  1.57it/s]
3|5778|Loss: 0.8073397278785706:  89%|████████▉ | 5777/6470 [1:11:36<07:22,  1.57it/s]
3|5778|Loss: 0.8073397278785706:  89%|████████▉ | 5778/6470 [1:11:36<07:04,  1.63it/s]
3|5779|Loss: 0.7799079418182373:  89%|████████▉ | 5778/6470 [1:11:36<07:04,  1.63it/s]
3|5779|Loss: 0.7799079418182373:  89%|████████▉ | 5779/6470 [1:11:37<07:03,  1.63it/s]
3|5780|Loss: 0.8042462468147278:  89%|████████▉ | 5779/6470 [1:11:37<07:03,  1.63it/s]
3|5780|Loss: 0.8042462468147278:  89%|████████▉ | 5780/6470 [1:11:38<07:53,  1.46it/s]
3|5781|Loss: 0.8859915137290955:  89%|████████▉ | 5780/6470 [1:11:38<07:53,  1.46it/s]
3|5781|Loss: 0.8859915137290955:  89%|████████▉ | 5781/6470 [1:11:38<07:53,  1.45it/s]
3|5782|Loss: 0.9918494820594788:  89%|████████▉ | 5781/6470 [1:11:39<07:53,  1.45it/s]
3|5782|Loss: 0.9918494820594788:  89%|████████▉ | 5782/6470 [1:11:39<08:03,  1.42it/s]
3|5783|Loss: 0.9729407429695129:  89%|████████▉ | 5782/6470 [1:11:39<08:03,  1.42it/s]
3|5783|Loss: 0.9729407429695129:  89%|████████▉ | 5783/6470 [1:11:40<07:59,  1.43it/s]
3|5784|Loss: 0.9635602235794067:  89%|████████▉ | 5783/6470 [1:11:40<07:59,  1.43it/s]
3|5784|Loss: 0.9635602235794067:  89%|████████▉ | 5784/6470 [1:11:40<07:58,  1.43it/s]
3|5785|Loss: 0.8085364699363708:  89%|████████▉ | 5784/6470 [1:11:41<07:58,  1.43it/s]
3|5785|Loss: 0.8085364699363708:  89%|████████▉ | 5785/6470 [1:11:41<08:14,  1.39it/s]
3|5786|Loss: 0.7911731004714966:  89%|████████▉ | 5785/6470 [1:11:42<08:14,  1.39it/s]
3|5786|Loss: 0.7911731004714966:  89%|████████▉ | 5786/6470 [1:11:42<08:22,  1.36it/s]
3|5787|Loss: 0.9164556860923767:  89%|████████▉ | 5786/6470 [1:11:42<08:22,  1.36it/s]
3|5787|Loss: 0.9164556860923767:  89%|████████▉ | 5787/6470 [1:11:43<08:47,  1.30it/s]
3|5788|Loss: 0.7831988334655762:  89%|████████▉ | 5787/6470 [1:11:43<08:47,  1.30it/s]
3|5788|Loss: 0.7831988334655762:  89%|████████▉ | 5788/6470 [1:11:44<08:54,  1.28it/s]
3|5789|Loss: 0.8066316843032837:  89%|████████▉ | 5788/6470 [1:11:44<08:54,  1.28it/s]
3|5789|Loss: 0.8066316843032837:  89%|████████▉ | 5789/6470 [1:11:44<08:53,  1.28it/s]
3|5790|Loss: 0.9530896544456482:  89%|████████▉ | 5789/6470 [1:11:45<08:53,  1.28it/s]
3|5790|Loss: 0.9530896544456482:  89%|████████▉ | 5790/6470 [1:11:45<08:53,  1.28it/s]
3|5791|Loss: 1.0367114543914795:  89%|████████▉ | 5790/6470 [1:11:46<08:53,  1.28it/s]
3|5791|Loss: 1.0367114543914795:  90%|████████▉ | 5791/6470 [1:11:46<09:10,  1.23it/s]
3|5792|Loss: 0.8850688934326172:  90%|████████▉ | 5791/6470 [1:11:47<09:10,  1.23it/s]
3|5792|Loss: 0.8850688934326172:  90%|████████▉ | 5792/6470 [1:11:47<09:25,  1.20it/s]
3|5793|Loss: 0.8740575909614563:  90%|████████▉ | 5792/6470 [1:11:47<09:25,  1.20it/s]
3|5793|Loss: 0.8740575909614563:  90%|████████▉ | 5793/6470 [1:11:48<09:32,  1.18it/s]
3|5794|Loss: 0.9237810969352722:  90%|████████▉ | 5793/6470 [1:11:48<09:32,  1.18it/s]
3|5794|Loss: 0.9237810969352722:  90%|████████▉ | 5794/6470 [1:11:49<09:17,  1.21it/s]
3|5795|Loss: 0.7695230841636658:  90%|████████▉ | 5794/6470 [1:11:49<09:17,  1.21it/s]
3|5795|Loss: 0.7695230841636658:  90%|████████▉ | 5795/6470 [1:11:49<08:27,  1.33it/s]
3|5796|Loss: 0.8525329232215881:  90%|████████▉ | 5795/6470 [1:11:50<08:27,  1.33it/s]
3|5796|Loss: 0.8525329232215881:  90%|████████▉ | 5796/6470 [1:11:50<08:08,  1.38it/s]
3|5797|Loss: 0.7950047850608826:  90%|████████▉ | 5796/6470 [1:11:50<08:08,  1.38it/s]
3|5797|Loss: 0.7950047850608826:  90%|████████▉ | 5797/6470 [1:11:50<07:40,  1.46it/s]
3|5798|Loss: 0.7082326412200928:  90%|████████▉ | 5797/6470 [1:11:51<07:40,  1.46it/s]
3|5798|Loss: 0.7082326412200928:  90%|████████▉ | 5798/6470 [1:11:51<08:00,  1.40it/s]
3|5799|Loss: 1.011446237564087:  90%|████████▉ | 5798/6470 [1:11:52<08:00,  1.40it/s] 
3|5799|Loss: 1.011446237564087:  90%|████████▉ | 5799/6470 [1:11:52<08:36,  1.30it/s]
3|5800|Loss: 0.7576379776000977:  90%|████████▉ | 5799/6470 [1:11:53<08:36,  1.30it/s]
3|5800|Loss: 0.7576379776000977:  90%|████████▉ | 5800/6470 [1:11:53<09:12,  1.21it/s]
3|5801|Loss: 0.8337126970291138:  90%|████████▉ | 5800/6470 [1:11:54<09:12,  1.21it/s]
3|5801|Loss: 0.8337126970291138:  90%|████████▉ | 5801/6470 [1:11:54<09:31,  1.17it/s]
3|5802|Loss: 0.8531938791275024:  90%|████████▉ | 5801/6470 [1:11:54<09:31,  1.17it/s]
3|5802|Loss: 0.8531938791275024:  90%|████████▉ | 5802/6470 [1:11:55<09:03,  1.23it/s]
3|5803|Loss: 0.8109933137893677:  90%|████████▉ | 5802/6470 [1:11:55<09:03,  1.23it/s]
3|5803|Loss: 0.8109933137893677:  90%|████████▉ | 5803/6470 [1:11:55<08:32,  1.30it/s]
3|5804|Loss: 0.7253739833831787:  90%|████████▉ | 5803/6470 [1:11:56<08:32,  1.30it/s]
3|5804|Loss: 0.7253739833831787:  90%|████████▉ | 5804/6470 [1:11:56<08:16,  1.34it/s]
3|5805|Loss: 0.6948313117027283:  90%|████████▉ | 5804/6470 [1:11:56<08:16,  1.34it/s]
3|5805|Loss: 0.6948313117027283:  90%|████████▉ | 5805/6470 [1:11:57<08:22,  1.32it/s]
3|5806|Loss: 0.7879831790924072:  90%|████████▉ | 5805/6470 [1:11:57<08:22,  1.32it/s]
3|5806|Loss: 0.7879831790924072:  90%|████████▉ | 5806/6470 [1:11:58<08:23,  1.32it/s]
3|5807|Loss: 0.5858201384544373:  90%|████████▉ | 5806/6470 [1:11:58<08:23,  1.32it/s]
3|5807|Loss: 0.5858201384544373:  90%|████████▉ | 5807/6470 [1:11:59<08:58,  1.23it/s]
3|5808|Loss: 0.9148142337799072:  90%|████████▉ | 5807/6470 [1:11:59<08:58,  1.23it/s]
3|5808|Loss: 0.9148142337799072:  90%|████████▉ | 5808/6470 [1:11:59<09:22,  1.18it/s]
3|5809|Loss: 0.7932530045509338:  90%|████████▉ | 5808/6470 [1:12:00<09:22,  1.18it/s]
3|5809|Loss: 0.7932530045509338:  90%|████████▉ | 5809/6470 [1:12:00<09:27,  1.16it/s]
3|5810|Loss: 0.8963524103164673:  90%|████████▉ | 5809/6470 [1:12:01<09:27,  1.16it/s]
3|5810|Loss: 0.8963524103164673:  90%|████████▉ | 5810/6470 [1:12:01<09:18,  1.18it/s]
3|5811|Loss: 0.9729039072990417:  90%|████████▉ | 5810/6470 [1:12:02<09:18,  1.18it/s]
3|5811|Loss: 0.9729039072990417:  90%|████████▉ | 5811/6470 [1:12:02<08:59,  1.22it/s]
3|5812|Loss: 0.9351885914802551:  90%|████████▉ | 5811/6470 [1:12:02<08:59,  1.22it/s]
3|5812|Loss: 0.9351885914802551:  90%|████████▉ | 5812/6470 [1:12:03<09:07,  1.20it/s]
3|5813|Loss: 0.7793121337890625:  90%|████████▉ | 5812/6470 [1:12:03<09:07,  1.20it/s]
3|5813|Loss: 0.7793121337890625:  90%|████████▉ | 5813/6470 [1:12:04<09:01,  1.21it/s]
3|5814|Loss: 0.8038116693496704:  90%|████████▉ | 5813/6470 [1:12:04<09:01,  1.21it/s]
3|5814|Loss: 0.8038116693496704:  90%|████████▉ | 5814/6470 [1:12:04<08:48,  1.24it/s]
3|5815|Loss: 0.9520813226699829:  90%|████████▉ | 5814/6470 [1:12:05<08:48,  1.24it/s]
3|5815|Loss: 0.9520813226699829:  90%|████████▉ | 5815/6470 [1:12:05<08:45,  1.25it/s]
3|5816|Loss: 0.9378223419189453:  90%|████████▉ | 5815/6470 [1:12:06<08:45,  1.25it/s]
3|5816|Loss: 0.9378223419189453:  90%|████████▉ | 5816/6470 [1:12:06<08:04,  1.35it/s]
3|5817|Loss: 0.8110546469688416:  90%|████████▉ | 5816/6470 [1:12:06<08:04,  1.35it/s]
3|5817|Loss: 0.8110546469688416:  90%|████████▉ | 5817/6470 [1:12:06<07:30,  1.45it/s]
3|5818|Loss: 0.8906425833702087:  90%|████████▉ | 5817/6470 [1:12:07<07:30,  1.45it/s]
3|5818|Loss: 0.8906425833702087:  90%|████████▉ | 5818/6470 [1:12:07<07:52,  1.38it/s]
3|5819|Loss: 0.8594077229499817:  90%|████████▉ | 5818/6470 [1:12:08<07:52,  1.38it/s]
3|5819|Loss: 0.8594077229499817:  90%|████████▉ | 5819/6470 [1:12:08<07:28,  1.45it/s]
3|5820|Loss: 1.0298815965652466:  90%|████████▉ | 5819/6470 [1:12:08<07:28,  1.45it/s]
3|5820|Loss: 1.0298815965652466:  90%|████████▉ | 5820/6470 [1:12:08<07:24,  1.46it/s]
3|5821|Loss: 0.9965189099311829:  90%|████████▉ | 5820/6470 [1:12:09<07:24,  1.46it/s]
3|5821|Loss: 0.9965189099311829:  90%|████████▉ | 5821/6470 [1:12:09<07:25,  1.46it/s]
3|5822|Loss: 0.8528356552124023:  90%|████████▉ | 5821/6470 [1:12:10<07:25,  1.46it/s]
3|5822|Loss: 0.8528356552124023:  90%|████████▉ | 5822/6470 [1:12:10<08:06,  1.33it/s]
3|5823|Loss: 0.8285015821456909:  90%|████████▉ | 5822/6470 [1:12:10<08:06,  1.33it/s]
3|5823|Loss: 0.8285015821456909:  90%|█████████ | 5823/6470 [1:12:11<08:09,  1.32it/s]
3|5824|Loss: 0.7655979990959167:  90%|█████████ | 5823/6470 [1:12:11<08:09,  1.32it/s]
3|5824|Loss: 0.7655979990959167:  90%|█████████ | 5824/6470 [1:12:12<08:26,  1.28it/s]
3|5825|Loss: 0.8514800667762756:  90%|█████████ | 5824/6470 [1:12:12<08:26,  1.28it/s]
3|5825|Loss: 0.8514800667762756:  90%|█████████ | 5825/6470 [1:12:12<07:59,  1.34it/s]
3|5826|Loss: 0.7935008406639099:  90%|█████████ | 5825/6470 [1:12:13<07:59,  1.34it/s]
3|5826|Loss: 0.7935008406639099:  90%|█████████ | 5826/6470 [1:12:13<07:32,  1.42it/s]
3|5827|Loss: 1.0774255990982056:  90%|█████████ | 5826/6470 [1:12:13<07:32,  1.42it/s]
3|5827|Loss: 1.0774255990982056:  90%|█████████ | 5827/6470 [1:12:14<07:36,  1.41it/s]
3|5828|Loss: 0.8354189395904541:  90%|█████████ | 5827/6470 [1:12:14<07:36,  1.41it/s]
3|5828|Loss: 0.8354189395904541:  90%|█████████ | 5828/6470 [1:12:14<07:56,  1.35it/s]
3|5829|Loss: 0.6981430053710938:  90%|█████████ | 5828/6470 [1:12:15<07:56,  1.35it/s]
3|5829|Loss: 0.6981430053710938:  90%|█████████ | 5829/6470 [1:12:15<08:08,  1.31it/s]
3|5830|Loss: 0.7732294797897339:  90%|█████████ | 5829/6470 [1:12:16<08:08,  1.31it/s]
3|5830|Loss: 0.7732294797897339:  90%|█████████ | 5830/6470 [1:12:16<08:38,  1.23it/s]
3|5831|Loss: 0.7998847961425781:  90%|█████████ | 5830/6470 [1:12:17<08:38,  1.23it/s]
3|5831|Loss: 0.7998847961425781:  90%|█████████ | 5831/6470 [1:12:17<08:53,  1.20it/s]
3|5832|Loss: 0.7504435181617737:  90%|█████████ | 5831/6470 [1:12:17<08:53,  1.20it/s]
3|5832|Loss: 0.7504435181617737:  90%|█████████ | 5832/6470 [1:12:18<08:28,  1.25it/s]
3|5833|Loss: 0.9213463664054871:  90%|█████████ | 5832/6470 [1:12:18<08:28,  1.25it/s]
3|5833|Loss: 0.9213463664054871:  90%|█████████ | 5833/6470 [1:12:19<08:23,  1.26it/s]
3|5834|Loss: 0.873414933681488:  90%|█████████ | 5833/6470 [1:12:19<08:23,  1.26it/s] 
3|5834|Loss: 0.873414933681488:  90%|█████████ | 5834/6470 [1:12:19<08:44,  1.21it/s]
3|5835|Loss: 0.7850428819656372:  90%|█████████ | 5834/6470 [1:12:20<08:44,  1.21it/s]
3|5835|Loss: 0.7850428819656372:  90%|█████████ | 5835/6470 [1:12:20<09:00,  1.17it/s]
3|5836|Loss: 0.6283880472183228:  90%|█████████ | 5835/6470 [1:12:21<09:00,  1.17it/s]
3|5836|Loss: 0.6283880472183228:  90%|█████████ | 5836/6470 [1:12:21<08:33,  1.23it/s]
3|5837|Loss: 0.9111761450767517:  90%|█████████ | 5836/6470 [1:12:22<08:33,  1.23it/s]
3|5837|Loss: 0.9111761450767517:  90%|█████████ | 5837/6470 [1:12:22<08:48,  1.20it/s]
3|5838|Loss: 0.8417518734931946:  90%|█████████ | 5837/6470 [1:12:22<08:48,  1.20it/s]
3|5838|Loss: 0.8417518734931946:  90%|█████████ | 5838/6470 [1:12:23<08:30,  1.24it/s]
3|5839|Loss: 0.8030993938446045:  90%|█████████ | 5838/6470 [1:12:23<08:30,  1.24it/s]
3|5839|Loss: 0.8030993938446045:  90%|█████████ | 5839/6470 [1:12:23<07:40,  1.37it/s]
3|5840|Loss: 0.8974173665046692:  90%|█████████ | 5839/6470 [1:12:24<07:40,  1.37it/s]
3|5840|Loss: 0.8974173665046692:  90%|█████████ | 5840/6470 [1:12:24<07:17,  1.44it/s]
3|5841|Loss: 0.8209989666938782:  90%|█████████ | 5840/6470 [1:12:24<07:17,  1.44it/s]
3|5841|Loss: 0.8209989666938782:  90%|█████████ | 5841/6470 [1:12:25<08:09,  1.29it/s]
3|5842|Loss: 0.8716456294059753:  90%|█████████ | 5841/6470 [1:12:25<08:09,  1.29it/s]
3|5842|Loss: 0.8716456294059753:  90%|█████████ | 5842/6470 [1:12:26<08:45,  1.20it/s]
3|5843|Loss: 0.8553441166877747:  90%|█████████ | 5842/6470 [1:12:26<08:45,  1.20it/s]
3|5843|Loss: 0.8553441166877747:  90%|█████████ | 5843/6470 [1:12:26<08:14,  1.27it/s]
3|5844|Loss: 0.9506756663322449:  90%|█████████ | 5843/6470 [1:12:27<08:14,  1.27it/s]
3|5844|Loss: 0.9506756663322449:  90%|█████████ | 5844/6470 [1:12:27<08:15,  1.26it/s]
3|5845|Loss: 0.7741891145706177:  90%|█████████ | 5844/6470 [1:12:28<08:15,  1.26it/s]
3|5845|Loss: 0.7741891145706177:  90%|█████████ | 5845/6470 [1:12:28<08:17,  1.26it/s]
3|5846|Loss: 0.7987983226776123:  90%|█████████ | 5845/6470 [1:12:28<08:17,  1.26it/s]
3|5846|Loss: 0.7987983226776123:  90%|█████████ | 5846/6470 [1:12:29<07:56,  1.31it/s]
3|5847|Loss: 0.801928699016571:  90%|█████████ | 5846/6470 [1:12:29<07:56,  1.31it/s] 
3|5847|Loss: 0.801928699016571:  90%|█████████ | 5847/6470 [1:12:30<07:55,  1.31it/s]
3|5848|Loss: 0.7897279262542725:  90%|█████████ | 5847/6470 [1:12:30<07:55,  1.31it/s]
3|5848|Loss: 0.7897279262542725:  90%|█████████ | 5848/6470 [1:12:30<08:01,  1.29it/s]
3|5849|Loss: 0.9046733975410461:  90%|█████████ | 5848/6470 [1:12:31<08:01,  1.29it/s]
3|5849|Loss: 0.9046733975410461:  90%|█████████ | 5849/6470 [1:12:31<08:24,  1.23it/s]
3|5850|Loss: 0.8387572765350342:  90%|█████████ | 5849/6470 [1:12:32<08:24,  1.23it/s]
3|5850|Loss: 0.8387572765350342:  90%|█████████ | 5850/6470 [1:12:32<08:26,  1.22it/s]
3|5851|Loss: 0.9190947413444519:  90%|█████████ | 5850/6470 [1:12:33<08:26,  1.22it/s]
3|5851|Loss: 0.9190947413444519:  90%|█████████ | 5851/6470 [1:12:33<08:23,  1.23it/s]
3|5852|Loss: 1.0262904167175293:  90%|█████████ | 5851/6470 [1:12:33<08:23,  1.23it/s]
3|5852|Loss: 1.0262904167175293:  90%|█████████ | 5852/6470 [1:12:33<07:31,  1.37it/s]
3|5853|Loss: 0.7341198325157166:  90%|█████████ | 5852/6470 [1:12:34<07:31,  1.37it/s]
3|5853|Loss: 0.7341198325157166:  90%|█████████ | 5853/6470 [1:12:34<06:52,  1.50it/s]
3|5854|Loss: 0.8069021701812744:  90%|█████████ | 5853/6470 [1:12:34<06:52,  1.50it/s]
3|5854|Loss: 0.8069021701812744:  90%|█████████ | 5854/6470 [1:12:35<07:27,  1.38it/s]
3|5855|Loss: 0.9288179874420166:  90%|█████████ | 5854/6470 [1:12:35<07:27,  1.38it/s]
3|5855|Loss: 0.9288179874420166:  90%|█████████ | 5855/6470 [1:12:36<08:04,  1.27it/s]
3|5856|Loss: 0.7859477400779724:  90%|█████████ | 5855/6470 [1:12:36<08:04,  1.27it/s]
3|5856|Loss: 0.7859477400779724:  91%|█████████ | 5856/6470 [1:12:37<08:09,  1.25it/s]
3|5857|Loss: 0.6783398985862732:  91%|█████████ | 5856/6470 [1:12:37<08:09,  1.25it/s]
3|5857|Loss: 0.6783398985862732:  91%|█████████ | 5857/6470 [1:12:37<08:21,  1.22it/s]
3|5858|Loss: 0.7775464057922363:  91%|█████████ | 5857/6470 [1:12:38<08:21,  1.22it/s]
3|5858|Loss: 0.7775464057922363:  91%|█████████ | 5858/6470 [1:12:38<08:12,  1.24it/s]
3|5859|Loss: 0.6086541414260864:  91%|█████████ | 5858/6470 [1:12:39<08:12,  1.24it/s]
3|5859|Loss: 0.6086541414260864:  91%|█████████ | 5859/6470 [1:12:39<07:52,  1.29it/s]
3|5860|Loss: 0.8776385188102722:  91%|█████████ | 5859/6470 [1:12:39<07:52,  1.29it/s]
3|5860|Loss: 0.8776385188102722:  91%|█████████ | 5860/6470 [1:12:40<07:50,  1.30it/s]
3|5861|Loss: 0.8409877419471741:  91%|█████████ | 5860/6470 [1:12:40<07:50,  1.30it/s]
3|5861|Loss: 0.8409877419471741:  91%|█████████ | 5861/6470 [1:12:40<07:39,  1.33it/s]
3|5862|Loss: 0.6979538202285767:  91%|█████████ | 5861/6470 [1:12:41<07:39,  1.33it/s]
3|5862|Loss: 0.6979538202285767:  91%|█████████ | 5862/6470 [1:12:41<08:00,  1.27it/s]
3|5863|Loss: 0.7381755709648132:  91%|█████████ | 5862/6470 [1:12:42<08:00,  1.27it/s]
3|5863|Loss: 0.7381755709648132:  91%|█████████ | 5863/6470 [1:12:42<08:35,  1.18it/s]
3|5864|Loss: 0.7962449193000793:  91%|█████████ | 5863/6470 [1:12:43<08:35,  1.18it/s]
3|5864|Loss: 0.7962449193000793:  91%|█████████ | 5864/6470 [1:12:43<08:43,  1.16it/s]
3|5865|Loss: 1.0425870418548584:  91%|█████████ | 5864/6470 [1:12:44<08:43,  1.16it/s]
3|5865|Loss: 1.0425870418548584:  91%|█████████ | 5865/6470 [1:12:44<07:59,  1.26it/s]
3|5866|Loss: 0.8894771933555603:  91%|█████████ | 5865/6470 [1:12:44<07:59,  1.26it/s]
3|5866|Loss: 0.8894771933555603:  91%|█████████ | 5866/6470 [1:12:44<07:30,  1.34it/s]
3|5867|Loss: 0.664189338684082:  91%|█████████ | 5866/6470 [1:12:45<07:30,  1.34it/s] 
3|5867|Loss: 0.664189338684082:  91%|█████████ | 5867/6470 [1:12:45<07:48,  1.29it/s]
3|5868|Loss: 0.7349386811256409:  91%|█████████ | 5867/6470 [1:12:46<07:48,  1.29it/s]
3|5868|Loss: 0.7349386811256409:  91%|█████████ | 5868/6470 [1:12:46<07:19,  1.37it/s]
3|5869|Loss: 0.9021337032318115:  91%|█████████ | 5868/6470 [1:12:46<07:19,  1.37it/s]
3|5869|Loss: 0.9021337032318115:  91%|█████████ | 5869/6470 [1:12:47<07:39,  1.31it/s]
3|5870|Loss: 0.9563955664634705:  91%|█████████ | 5869/6470 [1:12:47<07:39,  1.31it/s]
3|5870|Loss: 0.9563955664634705:  91%|█████████ | 5870/6470 [1:12:48<08:18,  1.20it/s]
3|5871|Loss: 0.8174066543579102:  91%|█████████ | 5870/6470 [1:12:48<08:18,  1.20it/s]
3|5871|Loss: 0.8174066543579102:  91%|█████████ | 5871/6470 [1:12:49<08:29,  1.18it/s]
3|5872|Loss: 0.9707087874412537:  91%|█████████ | 5871/6470 [1:12:49<08:29,  1.18it/s]
3|5872|Loss: 0.9707087874412537:  91%|█████████ | 5872/6470 [1:12:49<08:05,  1.23it/s]
3|5873|Loss: 0.786143958568573:  91%|█████████ | 5872/6470 [1:12:50<08:05,  1.23it/s] 
3|5873|Loss: 0.786143958568573:  91%|█████████ | 5873/6470 [1:12:50<07:14,  1.37it/s]
3|5874|Loss: 0.8655983209609985:  91%|█████████ | 5873/6470 [1:12:50<07:14,  1.37it/s]
3|5874|Loss: 0.8655983209609985:  91%|█████████ | 5874/6470 [1:12:51<07:37,  1.30it/s]
3|5875|Loss: 0.7602217793464661:  91%|█████████ | 5874/6470 [1:12:51<07:37,  1.30it/s]
3|5875|Loss: 0.7602217793464661:  91%|█████████ | 5875/6470 [1:12:52<08:05,  1.22it/s]
3|5876|Loss: 0.7368988990783691:  91%|█████████ | 5875/6470 [1:12:52<08:05,  1.22it/s]
3|5876|Loss: 0.7368988990783691:  91%|█████████ | 5876/6470 [1:12:52<08:07,  1.22it/s]
3|5877|Loss: 0.8731046319007874:  91%|█████████ | 5876/6470 [1:12:53<08:07,  1.22it/s]
3|5877|Loss: 0.8731046319007874:  91%|█████████ | 5877/6470 [1:12:53<08:25,  1.17it/s]
3|5878|Loss: 0.8158621788024902:  91%|█████████ | 5877/6470 [1:12:54<08:25,  1.17it/s]
3|5878|Loss: 0.8158621788024902:  91%|█████████ | 5878/6470 [1:12:54<08:08,  1.21it/s]
3|5879|Loss: 0.9985612034797668:  91%|█████████ | 5878/6470 [1:12:55<08:08,  1.21it/s]
3|5879|Loss: 0.9985612034797668:  91%|█████████ | 5879/6470 [1:12:55<07:46,  1.27it/s]
3|5880|Loss: 0.8328089118003845:  91%|█████████ | 5879/6470 [1:12:55<07:46,  1.27it/s]
3|5880|Loss: 0.8328089118003845:  91%|█████████ | 5880/6470 [1:12:56<07:38,  1.29it/s]
3|5881|Loss: 0.9942139983177185:  91%|█████████ | 5880/6470 [1:12:56<07:38,  1.29it/s]
3|5881|Loss: 0.9942139983177185:  91%|█████████ | 5881/6470 [1:12:57<08:01,  1.22it/s]
3|5882|Loss: 0.9030465483665466:  91%|█████████ | 5881/6470 [1:12:57<08:01,  1.22it/s]
3|5882|Loss: 0.9030465483665466:  91%|█████████ | 5882/6470 [1:12:57<07:57,  1.23it/s]
3|5883|Loss: 0.7713192701339722:  91%|█████████ | 5882/6470 [1:12:58<07:57,  1.23it/s]
3|5883|Loss: 0.7713192701339722:  91%|█████████ | 5883/6470 [1:12:58<07:32,  1.30it/s]
3|5884|Loss: 0.7340019941329956:  91%|█████████ | 5883/6470 [1:12:58<07:32,  1.30it/s]
3|5884|Loss: 0.7340019941329956:  91%|█████████ | 5884/6470 [1:12:59<06:57,  1.40it/s]
3|5885|Loss: 0.8956362009048462:  91%|█████████ | 5884/6470 [1:12:59<06:57,  1.40it/s]
3|5885|Loss: 0.8956362009048462:  91%|█████████ | 5885/6470 [1:12:59<07:24,  1.32it/s]
3|5886|Loss: 0.9185360074043274:  91%|█████████ | 5885/6470 [1:13:00<07:24,  1.32it/s]
3|5886|Loss: 0.9185360074043274:  91%|█████████ | 5886/6470 [1:13:00<07:02,  1.38it/s]
3|5887|Loss: 0.7323311567306519:  91%|█████████ | 5886/6470 [1:13:00<07:02,  1.38it/s]
3|5887|Loss: 0.7323311567306519:  91%|█████████ | 5887/6470 [1:13:01<07:03,  1.38it/s]
3|5888|Loss: 0.9195958375930786:  91%|█████████ | 5887/6470 [1:13:01<07:03,  1.38it/s]
3|5888|Loss: 0.9195958375930786:  91%|█████████ | 5888/6470 [1:13:02<07:39,  1.27it/s]
3|5889|Loss: 0.7067422270774841:  91%|█████████ | 5888/6470 [1:13:02<07:39,  1.27it/s]
3|5889|Loss: 0.7067422270774841:  91%|█████████ | 5889/6470 [1:13:03<08:05,  1.20it/s]
3|5890|Loss: 0.8254576325416565:  91%|█████████ | 5889/6470 [1:13:03<08:05,  1.20it/s]
3|5890|Loss: 0.8254576325416565:  91%|█████████ | 5890/6470 [1:13:03<07:46,  1.24it/s]
3|5891|Loss: 0.8353463411331177:  91%|█████████ | 5890/6470 [1:13:04<07:46,  1.24it/s]
3|5891|Loss: 0.8353463411331177:  91%|█████████ | 5891/6470 [1:13:04<07:22,  1.31it/s]
3|5892|Loss: 0.8581531047821045:  91%|█████████ | 5891/6470 [1:13:04<07:22,  1.31it/s]
3|5892|Loss: 0.8581531047821045:  91%|█████████ | 5892/6470 [1:13:05<07:01,  1.37it/s]
3|5893|Loss: 0.7737228870391846:  91%|█████████ | 5892/6470 [1:13:05<07:01,  1.37it/s]
3|5893|Loss: 0.7737228870391846:  91%|█████████ | 5893/6470 [1:13:05<06:42,  1.43it/s]
3|5894|Loss: 0.810708224773407:  91%|█████████ | 5893/6470 [1:13:06<06:42,  1.43it/s] 
3|5894|Loss: 0.810708224773407:  91%|█████████ | 5894/6470 [1:13:06<06:47,  1.41it/s]
3|5895|Loss: 0.6715529561042786:  91%|█████████ | 5894/6470 [1:13:06<06:47,  1.41it/s]
3|5895|Loss: 0.6715529561042786:  91%|█████████ | 5895/6470 [1:13:07<06:48,  1.41it/s]
3|5896|Loss: 0.8023192882537842:  91%|█████████ | 5895/6470 [1:13:07<06:48,  1.41it/s]
3|5896|Loss: 0.8023192882537842:  91%|█████████ | 5896/6470 [1:13:08<07:00,  1.36it/s]
3|5897|Loss: 0.7850145101547241:  91%|█████████ | 5896/6470 [1:13:08<07:00,  1.36it/s]
3|5897|Loss: 0.7850145101547241:  91%|█████████ | 5897/6470 [1:13:09<07:32,  1.27it/s]
3|5898|Loss: 0.7473331689834595:  91%|█████████ | 5897/6470 [1:13:09<07:32,  1.27it/s]
3|5898|Loss: 0.7473331689834595:  91%|█████████ | 5898/6470 [1:13:09<07:20,  1.30it/s]
3|5899|Loss: 0.7958013415336609:  91%|█████████ | 5898/6470 [1:13:10<07:20,  1.30it/s]
3|5899|Loss: 0.7958013415336609:  91%|█████████ | 5899/6470 [1:13:10<06:56,  1.37it/s]
3|5900|Loss: 0.929227352142334:  91%|█████████ | 5899/6470 [1:13:10<06:56,  1.37it/s] 
3|5900|Loss: 0.929227352142334:  91%|█████████ | 5900/6470 [1:13:11<06:48,  1.39it/s]
3|5901|Loss: 0.7313438653945923:  91%|█████████ | 5900/6470 [1:13:11<06:48,  1.39it/s]
3|5901|Loss: 0.7313438653945923:  91%|█████████ | 5901/6470 [1:13:11<07:18,  1.30it/s]
3|5902|Loss: 0.9225342273712158:  91%|█████████ | 5901/6470 [1:13:12<07:18,  1.30it/s]
3|5902|Loss: 0.9225342273712158:  91%|█████████ | 5902/6470 [1:13:12<07:01,  1.35it/s]
3|5903|Loss: 0.7831311821937561:  91%|█████████ | 5902/6470 [1:13:12<07:01,  1.35it/s]
3|5903|Loss: 0.7831311821937561:  91%|█████████ | 5903/6470 [1:13:13<06:44,  1.40it/s]
3|5904|Loss: 0.7528095841407776:  91%|█████████ | 5903/6470 [1:13:13<06:44,  1.40it/s]
3|5904|Loss: 0.7528095841407776:  91%|█████████▏| 5904/6470 [1:13:13<06:40,  1.41it/s]
3|5905|Loss: 0.8626280426979065:  91%|█████████▏| 5904/6470 [1:13:14<06:40,  1.41it/s]
3|5905|Loss: 0.8626280426979065:  91%|█████████▏| 5905/6470 [1:13:14<07:20,  1.28it/s]
3|5906|Loss: 0.8610818982124329:  91%|█████████▏| 5905/6470 [1:13:15<07:20,  1.28it/s]
3|5906|Loss: 0.8610818982124329:  91%|█████████▏| 5906/6470 [1:13:15<07:24,  1.27it/s]
3|5907|Loss: 0.8367207050323486:  91%|█████████▏| 5906/6470 [1:13:16<07:24,  1.27it/s]
3|5907|Loss: 0.8367207050323486:  91%|█████████▏| 5907/6470 [1:13:16<07:31,  1.25it/s]
3|5908|Loss: 0.7587157487869263:  91%|█████████▏| 5907/6470 [1:13:16<07:31,  1.25it/s]
3|5908|Loss: 0.7587157487869263:  91%|█████████▏| 5908/6470 [1:13:17<07:17,  1.28it/s]
3|5909|Loss: 0.8570103645324707:  91%|█████████▏| 5908/6470 [1:13:17<07:17,  1.28it/s]
3|5909|Loss: 0.8570103645324707:  91%|█████████▏| 5909/6470 [1:13:18<07:12,  1.30it/s]
3|5910|Loss: 0.7612318396568298:  91%|█████████▏| 5909/6470 [1:13:18<07:12,  1.30it/s]
3|5910|Loss: 0.7612318396568298:  91%|█████████▏| 5910/6470 [1:13:18<07:02,  1.33it/s]
3|5911|Loss: 0.9441678524017334:  91%|█████████▏| 5910/6470 [1:13:19<07:02,  1.33it/s]
3|5911|Loss: 0.9441678524017334:  91%|█████████▏| 5911/6470 [1:13:19<07:05,  1.31it/s]
3|5912|Loss: 0.9499909281730652:  91%|█████████▏| 5911/6470 [1:13:19<07:05,  1.31it/s]
3|5912|Loss: 0.9499909281730652:  91%|█████████▏| 5912/6470 [1:13:20<07:25,  1.25it/s]
3|5913|Loss: 0.7438064813613892:  91%|█████████▏| 5912/6470 [1:13:20<07:25,  1.25it/s]
3|5913|Loss: 0.7438064813613892:  91%|█████████▏| 5913/6470 [1:13:21<07:27,  1.24it/s]
3|5914|Loss: 0.5439541339874268:  91%|█████████▏| 5913/6470 [1:13:21<07:27,  1.24it/s]
3|5914|Loss: 0.5439541339874268:  91%|█████████▏| 5914/6470 [1:13:22<07:36,  1.22it/s]
3|5915|Loss: 0.709828794002533:  91%|█████████▏| 5914/6470 [1:13:22<07:36,  1.22it/s] 
3|5915|Loss: 0.709828794002533:  91%|█████████▏| 5915/6470 [1:13:22<07:16,  1.27it/s]
3|5916|Loss: 0.7661300301551819:  91%|█████████▏| 5915/6470 [1:13:23<07:16,  1.27it/s]
3|5916|Loss: 0.7661300301551819:  91%|█████████▏| 5916/6470 [1:13:23<07:05,  1.30it/s]
3|5917|Loss: 0.8086720705032349:  91%|█████████▏| 5916/6470 [1:13:23<07:05,  1.30it/s]
3|5917|Loss: 0.8086720705032349:  91%|█████████▏| 5917/6470 [1:13:24<07:05,  1.30it/s]
3|5918|Loss: 0.751068115234375:  91%|█████████▏| 5917/6470 [1:13:24<07:05,  1.30it/s] 
3|5918|Loss: 0.751068115234375:  91%|█████████▏| 5918/6470 [1:13:25<07:21,  1.25it/s]
3|5919|Loss: 0.8839020729064941:  91%|█████████▏| 5918/6470 [1:13:25<07:21,  1.25it/s]
3|5919|Loss: 0.8839020729064941:  91%|█████████▏| 5919/6470 [1:13:25<07:10,  1.28it/s]
3|5920|Loss: 1.0142111778259277:  91%|█████████▏| 5919/6470 [1:13:26<07:10,  1.28it/s]
3|5920|Loss: 1.0142111778259277:  91%|█████████▏| 5920/6470 [1:13:26<07:23,  1.24it/s]
3|5921|Loss: 0.7430843710899353:  91%|█████████▏| 5920/6470 [1:13:27<07:23,  1.24it/s]
3|5921|Loss: 0.7430843710899353:  92%|█████████▏| 5921/6470 [1:13:27<07:11,  1.27it/s]
3|5922|Loss: 0.825874924659729:  92%|█████████▏| 5921/6470 [1:13:27<07:11,  1.27it/s] 
3|5922|Loss: 0.825874924659729:  92%|█████████▏| 5922/6470 [1:13:28<06:54,  1.32it/s]
3|5923|Loss: 0.9697992205619812:  92%|█████████▏| 5922/6470 [1:13:28<06:54,  1.32it/s]
3|5923|Loss: 0.9697992205619812:  92%|█████████▏| 5923/6470 [1:13:29<07:12,  1.26it/s]
3|5924|Loss: 0.8077102899551392:  92%|█████████▏| 5923/6470 [1:13:29<07:12,  1.26it/s]
3|5924|Loss: 0.8077102899551392:  92%|█████████▏| 5924/6470 [1:13:29<07:18,  1.24it/s]
3|5925|Loss: 0.700076162815094:  92%|█████████▏| 5924/6470 [1:13:30<07:18,  1.24it/s] 
3|5925|Loss: 0.700076162815094:  92%|█████████▏| 5925/6470 [1:13:30<07:12,  1.26it/s]
3|5926|Loss: 0.8474718332290649:  92%|█████████▏| 5925/6470 [1:13:31<07:12,  1.26it/s]
3|5926|Loss: 0.8474718332290649:  92%|█████████▏| 5926/6470 [1:13:31<06:57,  1.30it/s]
3|5927|Loss: 0.8613121509552002:  92%|█████████▏| 5926/6470 [1:13:31<06:57,  1.30it/s]
3|5927|Loss: 0.8613121509552002:  92%|█████████▏| 5927/6470 [1:13:31<06:24,  1.41it/s]
3|5928|Loss: 0.8343435525894165:  92%|█████████▏| 5927/6470 [1:13:32<06:24,  1.41it/s]
3|5928|Loss: 0.8343435525894165:  92%|█████████▏| 5928/6470 [1:13:32<05:36,  1.61it/s]
3|5929|Loss: 0.8199377655982971:  92%|█████████▏| 5928/6470 [1:13:32<05:36,  1.61it/s]
3|5929|Loss: 0.8199377655982971:  92%|█████████▏| 5929/6470 [1:13:32<05:27,  1.65it/s]
3|5930|Loss: 0.618736207485199:  92%|█████████▏| 5929/6470 [1:13:33<05:27,  1.65it/s] 
3|5930|Loss: 0.618736207485199:  92%|█████████▏| 5930/6470 [1:13:33<05:09,  1.74it/s]
3|5931|Loss: 0.5989338755607605:  92%|█████████▏| 5930/6470 [1:13:33<05:09,  1.74it/s]
3|5931|Loss: 0.5989338755607605:  92%|█████████▏| 5931/6470 [1:13:34<05:10,  1.74it/s]
3|5932|Loss: 0.8842235803604126:  92%|█████████▏| 5931/6470 [1:13:34<05:10,  1.74it/s]
3|5932|Loss: 0.8842235803604126:  92%|█████████▏| 5932/6470 [1:13:34<05:49,  1.54it/s]
3|5933|Loss: 0.9503204822540283:  92%|█████████▏| 5932/6470 [1:13:35<05:49,  1.54it/s]
3|5933|Loss: 0.9503204822540283:  92%|█████████▏| 5933/6470 [1:13:35<05:56,  1.50it/s]
3|5934|Loss: 0.8709156513214111:  92%|█████████▏| 5933/6470 [1:13:35<05:56,  1.50it/s]
3|5934|Loss: 0.8709156513214111:  92%|█████████▏| 5934/6470 [1:13:36<05:53,  1.52it/s]
3|5935|Loss: 0.9285938143730164:  92%|█████████▏| 5934/6470 [1:13:36<05:53,  1.52it/s]
3|5935|Loss: 0.9285938143730164:  92%|█████████▏| 5935/6470 [1:13:36<05:35,  1.59it/s]
3|5936|Loss: 0.9396060705184937:  92%|█████████▏| 5935/6470 [1:13:37<05:35,  1.59it/s]
3|5936|Loss: 0.9396060705184937:  92%|█████████▏| 5936/6470 [1:13:37<05:52,  1.52it/s]
3|5937|Loss: 0.9840991497039795:  92%|█████████▏| 5936/6470 [1:13:37<05:52,  1.52it/s]
3|5937|Loss: 0.9840991497039795:  92%|█████████▏| 5937/6470 [1:13:38<06:08,  1.45it/s]
3|5938|Loss: 0.8521227240562439:  92%|█████████▏| 5937/6470 [1:13:38<06:08,  1.45it/s]
3|5938|Loss: 0.8521227240562439:  92%|█████████▏| 5938/6470 [1:13:39<06:32,  1.36it/s]
3|5939|Loss: 0.874155580997467:  92%|█████████▏| 5938/6470 [1:13:39<06:32,  1.36it/s] 
3|5939|Loss: 0.874155580997467:  92%|█████████▏| 5939/6470 [1:13:39<06:22,  1.39it/s]
3|5940|Loss: 0.9439360499382019:  92%|█████████▏| 5939/6470 [1:13:40<06:22,  1.39it/s]
3|5940|Loss: 0.9439360499382019:  92%|█████████▏| 5940/6470 [1:13:40<06:28,  1.36it/s]
3|5941|Loss: 0.9379543662071228:  92%|█████████▏| 5940/6470 [1:13:41<06:28,  1.36it/s]
3|5941|Loss: 0.9379543662071228:  92%|█████████▏| 5941/6470 [1:13:41<06:52,  1.28it/s]
3|5942|Loss: 0.7525803446769714:  92%|█████████▏| 5941/6470 [1:13:41<06:52,  1.28it/s]
3|5942|Loss: 0.7525803446769714:  92%|█████████▏| 5942/6470 [1:13:41<06:18,  1.40it/s]
3|5943|Loss: 0.7336878776550293:  92%|█████████▏| 5942/6470 [1:13:42<06:18,  1.40it/s]
3|5943|Loss: 0.7336878776550293:  92%|█████████▏| 5943/6470 [1:13:42<06:31,  1.35it/s]
3|5944|Loss: 1.0183956623077393:  92%|█████████▏| 5943/6470 [1:13:43<06:31,  1.35it/s]
3|5944|Loss: 1.0183956623077393:  92%|█████████▏| 5944/6470 [1:13:43<06:56,  1.26it/s]
3|5945|Loss: 0.9363198280334473:  92%|█████████▏| 5944/6470 [1:13:44<06:56,  1.26it/s]
3|5945|Loss: 0.9363198280334473:  92%|█████████▏| 5945/6470 [1:13:44<06:58,  1.26it/s]
3|5946|Loss: 1.1227867603302002:  92%|█████████▏| 5945/6470 [1:13:44<06:58,  1.26it/s]
3|5946|Loss: 1.1227867603302002:  92%|█████████▏| 5946/6470 [1:13:45<07:14,  1.21it/s]
3|5947|Loss: 0.9550360441207886:  92%|█████████▏| 5946/6470 [1:13:45<07:14,  1.21it/s]
3|5947|Loss: 0.9550360441207886:  92%|█████████▏| 5947/6470 [1:13:46<07:01,  1.24it/s]
3|5948|Loss: 0.9369093775749207:  92%|█████████▏| 5947/6470 [1:13:46<07:01,  1.24it/s]
3|5948|Loss: 0.9369093775749207:  92%|█████████▏| 5948/6470 [1:13:46<06:16,  1.39it/s]
3|5949|Loss: 0.7865836024284363:  92%|█████████▏| 5948/6470 [1:13:47<06:16,  1.39it/s]
3|5949|Loss: 0.7865836024284363:  92%|█████████▏| 5949/6470 [1:13:47<06:33,  1.32it/s]
3|5950|Loss: 0.8801799416542053:  92%|█████████▏| 5949/6470 [1:13:48<06:33,  1.32it/s]
3|5950|Loss: 0.8801799416542053:  92%|█████████▏| 5950/6470 [1:13:48<07:02,  1.23it/s]
3|5951|Loss: 0.8083406090736389:  92%|█████████▏| 5950/6470 [1:13:48<07:02,  1.23it/s]
3|5951|Loss: 0.8083406090736389:  92%|█████████▏| 5951/6470 [1:13:49<07:01,  1.23it/s]
3|5952|Loss: 0.8860188126564026:  92%|█████████▏| 5951/6470 [1:13:49<07:01,  1.23it/s]
3|5952|Loss: 0.8860188126564026:  92%|█████████▏| 5952/6470 [1:13:50<07:14,  1.19it/s]
3|5953|Loss: 0.8979423642158508:  92%|█████████▏| 5952/6470 [1:13:50<07:14,  1.19it/s]
3|5953|Loss: 0.8979423642158508:  92%|█████████▏| 5953/6470 [1:13:50<06:57,  1.24it/s]
3|5954|Loss: 0.8385480642318726:  92%|█████████▏| 5953/6470 [1:13:51<06:57,  1.24it/s]
3|5954|Loss: 0.8385480642318726:  92%|█████████▏| 5954/6470 [1:13:51<07:03,  1.22it/s]
3|5955|Loss: 0.820151686668396:  92%|█████████▏| 5954/6470 [1:13:52<07:03,  1.22it/s] 
3|5955|Loss: 0.820151686668396:  92%|█████████▏| 5955/6470 [1:13:52<06:49,  1.26it/s]
3|5956|Loss: 0.9537264108657837:  92%|█████████▏| 5955/6470 [1:13:52<06:49,  1.26it/s]
3|5956|Loss: 0.9537264108657837:  92%|█████████▏| 5956/6470 [1:13:53<06:59,  1.23it/s]
3|5957|Loss: 0.8408511877059937:  92%|█████████▏| 5956/6470 [1:13:53<06:59,  1.23it/s]
3|5957|Loss: 0.8408511877059937:  92%|█████████▏| 5957/6470 [1:13:54<06:45,  1.27it/s]
3|5958|Loss: 0.8557997941970825:  92%|█████████▏| 5957/6470 [1:13:54<06:45,  1.27it/s]
3|5958|Loss: 0.8557997941970825:  92%|█████████▏| 5958/6470 [1:13:54<06:42,  1.27it/s]
3|5959|Loss: 0.7166316509246826:  92%|█████████▏| 5958/6470 [1:13:55<06:42,  1.27it/s]
3|5959|Loss: 0.7166316509246826:  92%|█████████▏| 5959/6470 [1:13:55<05:54,  1.44it/s]
3|5960|Loss: 0.7353150844573975:  92%|█████████▏| 5959/6470 [1:13:55<05:54,  1.44it/s]
3|5960|Loss: 0.7353150844573975:  92%|█████████▏| 5960/6470 [1:13:56<06:10,  1.38it/s]
3|5961|Loss: 0.7971903085708618:  92%|█████████▏| 5960/6470 [1:13:56<06:10,  1.38it/s]
3|5961|Loss: 0.7971903085708618:  92%|█████████▏| 5961/6470 [1:13:56<06:07,  1.39it/s]
3|5962|Loss: 0.847691535949707:  92%|█████████▏| 5961/6470 [1:13:57<06:07,  1.39it/s] 
3|5962|Loss: 0.847691535949707:  92%|█████████▏| 5962/6470 [1:13:57<06:16,  1.35it/s]
3|5963|Loss: 0.7132619619369507:  92%|█████████▏| 5962/6470 [1:13:58<06:16,  1.35it/s]
3|5963|Loss: 0.7132619619369507:  92%|█████████▏| 5963/6470 [1:13:58<06:13,  1.36it/s]
3|5964|Loss: 0.9086480140686035:  92%|█████████▏| 5963/6470 [1:13:58<06:13,  1.36it/s]
3|5964|Loss: 0.9086480140686035:  92%|█████████▏| 5964/6470 [1:13:59<06:19,  1.33it/s]
3|5965|Loss: 0.8381961584091187:  92%|█████████▏| 5964/6470 [1:13:59<06:19,  1.33it/s]
3|5965|Loss: 0.8381961584091187:  92%|█████████▏| 5965/6470 [1:14:00<06:33,  1.28it/s]
3|5966|Loss: 0.755813479423523:  92%|█████████▏| 5965/6470 [1:14:00<06:33,  1.28it/s] 
3|5966|Loss: 0.755813479423523:  92%|█████████▏| 5966/6470 [1:14:00<06:22,  1.32it/s]
3|5967|Loss: 0.7983126640319824:  92%|█████████▏| 5966/6470 [1:14:01<06:22,  1.32it/s]
3|5967|Loss: 0.7983126640319824:  92%|█████████▏| 5967/6470 [1:14:01<06:24,  1.31it/s]
3|5968|Loss: 0.8452876806259155:  92%|█████████▏| 5967/6470 [1:14:01<06:24,  1.31it/s]
3|5968|Loss: 0.8452876806259155:  92%|█████████▏| 5968/6470 [1:14:02<06:17,  1.33it/s]
3|5969|Loss: 0.9396517872810364:  92%|█████████▏| 5968/6470 [1:14:02<06:17,  1.33it/s]
3|5969|Loss: 0.9396517872810364:  92%|█████████▏| 5969/6470 [1:14:03<06:23,  1.31it/s]
3|5970|Loss: 0.7879828214645386:  92%|█████████▏| 5969/6470 [1:14:03<06:23,  1.31it/s]
3|5970|Loss: 0.7879828214645386:  92%|█████████▏| 5970/6470 [1:14:03<06:15,  1.33it/s]
3|5971|Loss: 0.8778800368309021:  92%|█████████▏| 5970/6470 [1:14:04<06:15,  1.33it/s]
3|5971|Loss: 0.8778800368309021:  92%|█████████▏| 5971/6470 [1:14:04<06:33,  1.27it/s]
3|5972|Loss: 0.8499189019203186:  92%|█████████▏| 5971/6470 [1:14:05<06:33,  1.27it/s]
3|5972|Loss: 0.8499189019203186:  92%|█████████▏| 5972/6470 [1:14:05<06:48,  1.22it/s]
3|5973|Loss: 0.8987411260604858:  92%|█████████▏| 5972/6470 [1:14:05<06:48,  1.22it/s]
3|5973|Loss: 0.8987411260604858:  92%|█████████▏| 5973/6470 [1:14:06<06:45,  1.22it/s]
3|5974|Loss: 0.7052947878837585:  92%|█████████▏| 5973/6470 [1:14:06<06:45,  1.22it/s]
3|5974|Loss: 0.7052947878837585:  92%|█████████▏| 5974/6470 [1:14:06<06:25,  1.29it/s]
3|5975|Loss: 0.7510343194007874:  92%|█████████▏| 5974/6470 [1:14:07<06:25,  1.29it/s]
3|5975|Loss: 0.7510343194007874:  92%|█████████▏| 5975/6470 [1:14:07<06:11,  1.33it/s]
3|5976|Loss: 0.9482426047325134:  92%|█████████▏| 5975/6470 [1:14:08<06:11,  1.33it/s]
3|5976|Loss: 0.9482426047325134:  92%|█████████▏| 5976/6470 [1:14:08<06:21,  1.30it/s]
3|5977|Loss: 0.916642427444458:  92%|█████████▏| 5976/6470 [1:14:08<06:21,  1.30it/s] 
3|5977|Loss: 0.916642427444458:  92%|█████████▏| 5977/6470 [1:14:09<06:26,  1.28it/s]
3|5978|Loss: 0.8815235495567322:  92%|█████████▏| 5977/6470 [1:14:09<06:26,  1.28it/s]
3|5978|Loss: 0.8815235495567322:  92%|█████████▏| 5978/6470 [1:14:10<06:27,  1.27it/s]
3|5979|Loss: 0.796867847442627:  92%|█████████▏| 5978/6470 [1:14:10<06:27,  1.27it/s] 
3|5979|Loss: 0.796867847442627:  92%|█████████▏| 5979/6470 [1:14:11<06:43,  1.22it/s]
3|5980|Loss: 0.7592272758483887:  92%|█████████▏| 5979/6470 [1:14:11<06:43,  1.22it/s]
3|5980|Loss: 0.7592272758483887:  92%|█████████▏| 5980/6470 [1:14:11<06:52,  1.19it/s]
3|5981|Loss: 0.9022868275642395:  92%|█████████▏| 5980/6470 [1:14:12<06:52,  1.19it/s]
3|5981|Loss: 0.9022868275642395:  92%|█████████▏| 5981/6470 [1:14:12<06:46,  1.20it/s]
3|5982|Loss: 0.7243643999099731:  92%|█████████▏| 5981/6470 [1:14:13<06:46,  1.20it/s]
3|5982|Loss: 0.7243643999099731:  92%|█████████▏| 5982/6470 [1:14:13<06:48,  1.19it/s]
3|5983|Loss: 0.9796156883239746:  92%|█████████▏| 5982/6470 [1:14:14<06:48,  1.19it/s]
3|5983|Loss: 0.9796156883239746:  92%|█████████▏| 5983/6470 [1:14:14<06:43,  1.21it/s]
3|5984|Loss: 0.7565498352050781:  92%|█████████▏| 5983/6470 [1:14:14<06:43,  1.21it/s]
3|5984|Loss: 0.7565498352050781:  92%|█████████▏| 5984/6470 [1:14:15<06:39,  1.22it/s]
3|5985|Loss: 0.8602508902549744:  92%|█████████▏| 5984/6470 [1:14:15<06:39,  1.22it/s]
3|5985|Loss: 0.8602508902549744:  93%|█████████▎| 5985/6470 [1:14:15<06:16,  1.29it/s]
3|5986|Loss: 0.8296234011650085:  93%|█████████▎| 5985/6470 [1:14:16<06:16,  1.29it/s]
3|5986|Loss: 0.8296234011650085:  93%|█████████▎| 5986/6470 [1:14:16<06:09,  1.31it/s]
3|5987|Loss: 0.778701663017273:  93%|█████████▎| 5986/6470 [1:14:17<06:09,  1.31it/s] 
3|5987|Loss: 0.778701663017273:  93%|█████████▎| 5987/6470 [1:14:17<06:20,  1.27it/s]
3|5988|Loss: 0.8735426664352417:  93%|█████████▎| 5987/6470 [1:14:17<06:20,  1.27it/s]
3|5988|Loss: 0.8735426664352417:  93%|█████████▎| 5988/6470 [1:14:18<06:09,  1.31it/s]
3|5989|Loss: 0.8299071788787842:  93%|█████████▎| 5988/6470 [1:14:18<06:09,  1.31it/s]
3|5989|Loss: 0.8299071788787842:  93%|█████████▎| 5989/6470 [1:14:18<05:41,  1.41it/s]
3|5990|Loss: 0.8472321629524231:  93%|█████████▎| 5989/6470 [1:14:19<05:41,  1.41it/s]
3|5990|Loss: 0.8472321629524231:  93%|█████████▎| 5990/6470 [1:14:19<05:32,  1.44it/s]
3|5991|Loss: 0.7299951314926147:  93%|█████████▎| 5990/6470 [1:14:19<05:32,  1.44it/s]
3|5991|Loss: 0.7299951314926147:  93%|█████████▎| 5991/6470 [1:14:19<05:20,  1.50it/s]
3|5992|Loss: 0.792996883392334:  93%|█████████▎| 5991/6470 [1:14:20<05:20,  1.50it/s] 
3|5992|Loss: 0.792996883392334:  93%|█████████▎| 5992/6470 [1:14:20<05:30,  1.45it/s]
3|5993|Loss: 0.9565919637680054:  93%|█████████▎| 5992/6470 [1:14:21<05:30,  1.45it/s]
3|5993|Loss: 0.9565919637680054:  93%|█████████▎| 5993/6470 [1:14:21<05:45,  1.38it/s]
3|5994|Loss: 0.7662104368209839:  93%|█████████▎| 5993/6470 [1:14:21<05:45,  1.38it/s]
3|5994|Loss: 0.7662104368209839:  93%|█████████▎| 5994/6470 [1:14:22<05:41,  1.39it/s]
3|5995|Loss: 0.6991941928863525:  93%|█████████▎| 5994/6470 [1:14:22<05:41,  1.39it/s]
3|5995|Loss: 0.6991941928863525:  93%|█████████▎| 5995/6470 [1:14:23<05:51,  1.35it/s]
3|5996|Loss: 0.9702303409576416:  93%|█████████▎| 5995/6470 [1:14:23<05:51,  1.35it/s]
3|5996|Loss: 0.9702303409576416:  93%|█████████▎| 5996/6470 [1:14:23<05:45,  1.37it/s]
3|5997|Loss: 0.9754344820976257:  93%|█████████▎| 5996/6470 [1:14:24<05:45,  1.37it/s]
3|5997|Loss: 0.9754344820976257:  93%|█████████▎| 5997/6470 [1:14:24<05:57,  1.32it/s]
3|5998|Loss: 0.8025592565536499:  93%|█████████▎| 5997/6470 [1:14:24<05:57,  1.32it/s]
3|5998|Loss: 0.8025592565536499:  93%|█████████▎| 5998/6470 [1:14:25<05:54,  1.33it/s]
3|5999|Loss: 0.7632710933685303:  93%|█████████▎| 5998/6470 [1:14:25<05:54,  1.33it/s]
3|5999|Loss: 0.7632710933685303:  93%|█████████▎| 5999/6470 [1:14:26<05:54,  1.33it/s]
3|6000|Loss: 0.9357814192771912:  93%|█████████▎| 5999/6470 [1:14:26<05:54,  1.33it/s]
3|6000|Loss: 0.9357814192771912:  93%|█████████▎| 6000/6470 [1:14:26<05:50,  1.34it/s]
3|6001|Loss: 0.7247530221939087:  93%|█████████▎| 6000/6470 [1:14:27<05:50,  1.34it/s]
3|6001|Loss: 0.7247530221939087:  93%|█████████▎| 6001/6470 [1:14:27<05:35,  1.40it/s]
3|6002|Loss: 0.8210291266441345:  93%|█████████▎| 6001/6470 [1:14:27<05:35,  1.40it/s]
3|6002|Loss: 0.8210291266441345:  93%|█████████▎| 6002/6470 [1:14:28<05:35,  1.40it/s]
3|6003|Loss: 0.9395669102668762:  93%|█████████▎| 6002/6470 [1:14:28<05:35,  1.40it/s]
3|6003|Loss: 0.9395669102668762:  93%|█████████▎| 6003/6470 [1:14:28<05:43,  1.36it/s]
3|6004|Loss: 0.8510849475860596:  93%|█████████▎| 6003/6470 [1:14:29<05:43,  1.36it/s]
3|6004|Loss: 0.8510849475860596:  93%|█████████▎| 6004/6470 [1:14:29<05:14,  1.48it/s]
3|6005|Loss: 0.840552568435669:  93%|█████████▎| 6004/6470 [1:14:29<05:14,  1.48it/s] 
3|6005|Loss: 0.840552568435669:  93%|█████████▎| 6005/6470 [1:14:30<05:14,  1.48it/s]
3|6006|Loss: 0.9081441760063171:  93%|█████████▎| 6005/6470 [1:14:30<05:14,  1.48it/s]
3|6006|Loss: 0.9081441760063171:  93%|█████████▎| 6006/6470 [1:14:30<05:17,  1.46it/s]
3|6007|Loss: 0.9029161930084229:  93%|█████████▎| 6006/6470 [1:14:31<05:17,  1.46it/s]
3|6007|Loss: 0.9029161930084229:  93%|█████████▎| 6007/6470 [1:14:31<05:24,  1.42it/s]
3|6008|Loss: 0.7637508511543274:  93%|█████████▎| 6007/6470 [1:14:31<05:24,  1.42it/s]
3|6008|Loss: 0.7637508511543274:  93%|█████████▎| 6008/6470 [1:14:32<05:30,  1.40it/s]
3|6009|Loss: 1.110769271850586:  93%|█████████▎| 6008/6470 [1:14:32<05:30,  1.40it/s] 
3|6009|Loss: 1.110769271850586:  93%|█████████▎| 6009/6470 [1:14:32<05:17,  1.45it/s]
3|6010|Loss: 0.8836557269096375:  93%|█████████▎| 6009/6470 [1:14:33<05:17,  1.45it/s]
3|6010|Loss: 0.8836557269096375:  93%|█████████▎| 6010/6470 [1:14:33<05:36,  1.37it/s]
3|6011|Loss: 0.8126826286315918:  93%|█████████▎| 6010/6470 [1:14:34<05:36,  1.37it/s]
3|6011|Loss: 0.8126826286315918:  93%|█████████▎| 6011/6470 [1:14:34<05:31,  1.38it/s]
3|6012|Loss: 0.9172922968864441:  93%|█████████▎| 6011/6470 [1:14:34<05:31,  1.38it/s]
3|6012|Loss: 0.9172922968864441:  93%|█████████▎| 6012/6470 [1:14:35<05:39,  1.35it/s]
3|6013|Loss: 0.8653992414474487:  93%|█████████▎| 6012/6470 [1:14:35<05:39,  1.35it/s]
3|6013|Loss: 0.8653992414474487:  93%|█████████▎| 6013/6470 [1:14:35<05:24,  1.41it/s]
3|6014|Loss: 0.8711362481117249:  93%|█████████▎| 6013/6470 [1:14:36<05:24,  1.41it/s]
3|6014|Loss: 0.8711362481117249:  93%|█████████▎| 6014/6470 [1:14:36<05:39,  1.34it/s]
3|6015|Loss: 0.8672980070114136:  93%|█████████▎| 6014/6470 [1:14:37<05:39,  1.34it/s]
3|6015|Loss: 0.8672980070114136:  93%|█████████▎| 6015/6470 [1:14:37<05:17,  1.43it/s]
3|6016|Loss: 0.8353922963142395:  93%|█████████▎| 6015/6470 [1:14:37<05:17,  1.43it/s]
3|6016|Loss: 0.8353922963142395:  93%|█████████▎| 6016/6470 [1:14:37<05:06,  1.48it/s]
3|6017|Loss: 0.7014333605766296:  93%|█████████▎| 6016/6470 [1:14:38<05:06,  1.48it/s]
3|6017|Loss: 0.7014333605766296:  93%|█████████▎| 6017/6470 [1:14:38<04:53,  1.54it/s]
3|6018|Loss: 0.8133923411369324:  93%|█████████▎| 6017/6470 [1:14:38<04:53,  1.54it/s]
3|6018|Loss: 0.8133923411369324:  93%|█████████▎| 6018/6470 [1:14:39<04:51,  1.55it/s]
3|6019|Loss: 0.8532634377479553:  93%|█████████▎| 6018/6470 [1:14:39<04:51,  1.55it/s]
3|6019|Loss: 0.8532634377479553:  93%|█████████▎| 6019/6470 [1:14:39<04:53,  1.54it/s]
3|6020|Loss: 0.8184354305267334:  93%|█████████▎| 6019/6470 [1:14:40<04:53,  1.54it/s]
3|6020|Loss: 0.8184354305267334:  93%|█████████▎| 6020/6470 [1:14:40<04:33,  1.65it/s]
3|6021|Loss: 0.8799704313278198:  93%|█████████▎| 6020/6470 [1:14:40<04:33,  1.65it/s]
3|6021|Loss: 0.8799704313278198:  93%|█████████▎| 6021/6470 [1:14:40<04:28,  1.67it/s]
3|6022|Loss: 0.944510817527771:  93%|█████████▎| 6021/6470 [1:14:41<04:28,  1.67it/s] 
3|6022|Loss: 0.944510817527771:  93%|█████████▎| 6022/6470 [1:14:41<04:24,  1.69it/s]
3|6023|Loss: 0.7883275151252747:  93%|█████████▎| 6022/6470 [1:14:41<04:24,  1.69it/s]
3|6023|Loss: 0.7883275151252747:  93%|█████████▎| 6023/6470 [1:14:42<04:34,  1.63it/s]
3|6024|Loss: 0.8171310424804688:  93%|█████████▎| 6023/6470 [1:14:42<04:34,  1.63it/s]
3|6024|Loss: 0.8171310424804688:  93%|█████████▎| 6024/6470 [1:14:42<04:27,  1.67it/s]
3|6025|Loss: 0.8201472163200378:  93%|█████████▎| 6024/6470 [1:14:43<04:27,  1.67it/s]
3|6025|Loss: 0.8201472163200378:  93%|█████████▎| 6025/6470 [1:14:43<04:41,  1.58it/s]
3|6026|Loss: 1.0085164308547974:  93%|█████████▎| 6025/6470 [1:14:43<04:41,  1.58it/s]
3|6026|Loss: 1.0085164308547974:  93%|█████████▎| 6026/6470 [1:14:44<04:49,  1.53it/s]
3|6027|Loss: 0.7329570651054382:  93%|█████████▎| 6026/6470 [1:14:44<04:49,  1.53it/s]
3|6027|Loss: 0.7329570651054382:  93%|█████████▎| 6027/6470 [1:14:44<04:45,  1.55it/s]
3|6028|Loss: 0.688395082950592:  93%|█████████▎| 6027/6470 [1:14:45<04:45,  1.55it/s] 
3|6028|Loss: 0.688395082950592:  93%|█████████▎| 6028/6470 [1:14:45<04:48,  1.53it/s]
3|6029|Loss: 0.857337474822998:  93%|█████████▎| 6028/6470 [1:14:45<04:48,  1.53it/s]
3|6029|Loss: 0.857337474822998:  93%|█████████▎| 6029/6470 [1:14:46<04:47,  1.53it/s]
3|6030|Loss: 0.76491379737854:  93%|█████████▎| 6029/6470 [1:14:46<04:47,  1.53it/s] 
3|6030|Loss: 0.76491379737854:  93%|█████████▎| 6030/6470 [1:14:46<04:30,  1.63it/s]
3|6031|Loss: 0.6983230710029602:  93%|█████████▎| 6030/6470 [1:14:46<04:30,  1.63it/s]
3|6031|Loss: 0.6983230710029602:  93%|█████████▎| 6031/6470 [1:14:47<04:46,  1.53it/s]
3|6032|Loss: 0.9266498684883118:  93%|█████████▎| 6031/6470 [1:14:47<04:46,  1.53it/s]
3|6032|Loss: 0.9266498684883118:  93%|█████████▎| 6032/6470 [1:14:48<05:05,  1.43it/s]
3|6033|Loss: 0.8584389686584473:  93%|█████████▎| 6032/6470 [1:14:48<05:05,  1.43it/s]
3|6033|Loss: 0.8584389686584473:  93%|█████████▎| 6033/6470 [1:14:48<05:14,  1.39it/s]
3|6034|Loss: 0.8155845999717712:  93%|█████████▎| 6033/6470 [1:14:49<05:14,  1.39it/s]
3|6034|Loss: 0.8155845999717712:  93%|█████████▎| 6034/6470 [1:14:49<05:19,  1.37it/s]
3|6035|Loss: 0.6575255393981934:  93%|█████████▎| 6034/6470 [1:14:50<05:19,  1.37it/s]
3|6035|Loss: 0.6575255393981934:  93%|█████████▎| 6035/6470 [1:14:50<05:06,  1.42it/s]
3|6036|Loss: 0.7536972165107727:  93%|█████████▎| 6035/6470 [1:14:50<05:06,  1.42it/s]
3|6036|Loss: 0.7536972165107727:  93%|█████████▎| 6036/6470 [1:14:51<05:12,  1.39it/s]
3|6037|Loss: 0.8414390087127686:  93%|█████████▎| 6036/6470 [1:14:51<05:12,  1.39it/s]
3|6037|Loss: 0.8414390087127686:  93%|█████████▎| 6037/6470 [1:14:51<05:23,  1.34it/s]
3|6038|Loss: 0.9314035177230835:  93%|█████████▎| 6037/6470 [1:14:52<05:23,  1.34it/s]
3|6038|Loss: 0.9314035177230835:  93%|█████████▎| 6038/6470 [1:14:52<05:41,  1.26it/s]
3|6039|Loss: 0.7819435596466064:  93%|█████████▎| 6038/6470 [1:14:53<05:41,  1.26it/s]
3|6039|Loss: 0.7819435596466064:  93%|█████████▎| 6039/6470 [1:14:53<05:30,  1.30it/s]
3|6040|Loss: 0.7910643815994263:  93%|█████████▎| 6039/6470 [1:14:53<05:30,  1.30it/s]
3|6040|Loss: 0.7910643815994263:  93%|█████████▎| 6040/6470 [1:14:54<05:19,  1.35it/s]
3|6041|Loss: 0.8375482559204102:  93%|█████████▎| 6040/6470 [1:14:54<05:19,  1.35it/s]
3|6041|Loss: 0.8375482559204102:  93%|█████████▎| 6041/6470 [1:14:54<04:57,  1.44it/s]
3|6042|Loss: 0.7099588513374329:  93%|█████████▎| 6041/6470 [1:14:55<04:57,  1.44it/s]
3|6042|Loss: 0.7099588513374329:  93%|█████████▎| 6042/6470 [1:14:55<04:33,  1.56it/s]
3|6043|Loss: 0.9004348516464233:  93%|█████████▎| 6042/6470 [1:14:55<04:33,  1.56it/s]
3|6043|Loss: 0.9004348516464233:  93%|█████████▎| 6043/6470 [1:14:55<04:44,  1.50it/s]
3|6044|Loss: 0.8244480490684509:  93%|█████████▎| 6043/6470 [1:14:56<04:44,  1.50it/s]
3|6044|Loss: 0.8244480490684509:  93%|█████████▎| 6044/6470 [1:14:56<04:41,  1.51it/s]
3|6045|Loss: 0.7033108472824097:  93%|█████████▎| 6044/6470 [1:14:57<04:41,  1.51it/s]
3|6045|Loss: 0.7033108472824097:  93%|█████████▎| 6045/6470 [1:14:57<04:52,  1.45it/s]
3|6046|Loss: 0.9121137261390686:  93%|█████████▎| 6045/6470 [1:14:57<04:52,  1.45it/s]
3|6046|Loss: 0.9121137261390686:  93%|█████████▎| 6046/6470 [1:14:58<05:02,  1.40it/s]
3|6047|Loss: 0.7217071056365967:  93%|█████████▎| 6046/6470 [1:14:58<05:02,  1.40it/s]
3|6047|Loss: 0.7217071056365967:  93%|█████████▎| 6047/6470 [1:14:58<05:08,  1.37it/s]
3|6048|Loss: 0.8864755630493164:  93%|█████████▎| 6047/6470 [1:14:59<05:08,  1.37it/s]
3|6048|Loss: 0.8864755630493164:  93%|█████████▎| 6048/6470 [1:14:59<05:21,  1.31it/s]
3|6049|Loss: 0.7204059958457947:  93%|█████████▎| 6048/6470 [1:15:00<05:21,  1.31it/s]
3|6049|Loss: 0.7204059958457947:  93%|█████████▎| 6049/6470 [1:15:00<05:37,  1.25it/s]
3|6050|Loss: 0.9668067097663879:  93%|█████████▎| 6049/6470 [1:15:01<05:37,  1.25it/s]
3|6050|Loss: 0.9668067097663879:  94%|█████████▎| 6050/6470 [1:15:01<05:28,  1.28it/s]
3|6051|Loss: 0.8582132458686829:  94%|█████████▎| 6050/6470 [1:15:01<05:28,  1.28it/s]
3|6051|Loss: 0.8582132458686829:  94%|█████████▎| 6051/6470 [1:15:01<05:04,  1.38it/s]
3|6052|Loss: 0.8415165543556213:  94%|█████████▎| 6051/6470 [1:15:02<05:04,  1.38it/s]
3|6052|Loss: 0.8415165543556213:  94%|█████████▎| 6052/6470 [1:15:02<05:15,  1.33it/s]
3|6053|Loss: 0.9483081102371216:  94%|█████████▎| 6052/6470 [1:15:03<05:15,  1.33it/s]
3|6053|Loss: 0.9483081102371216:  94%|█████████▎| 6053/6470 [1:15:03<05:05,  1.37it/s]
3|6054|Loss: 0.707996129989624:  94%|█████████▎| 6053/6470 [1:15:03<05:05,  1.37it/s] 
3|6054|Loss: 0.707996129989624:  94%|█████████▎| 6054/6470 [1:15:04<05:03,  1.37it/s]
3|6055|Loss: 0.7965827584266663:  94%|█████████▎| 6054/6470 [1:15:04<05:03,  1.37it/s]
3|6055|Loss: 0.7965827584266663:  94%|█████████▎| 6055/6470 [1:15:05<05:10,  1.34it/s]
3|6056|Loss: 0.7746049761772156:  94%|█████████▎| 6055/6470 [1:15:05<05:10,  1.34it/s]
3|6056|Loss: 0.7746049761772156:  94%|█████████▎| 6056/6470 [1:15:05<05:14,  1.32it/s]
3|6057|Loss: 0.9950109720230103:  94%|█████████▎| 6056/6470 [1:15:06<05:14,  1.32it/s]
3|6057|Loss: 0.9950109720230103:  94%|█████████▎| 6057/6470 [1:15:06<05:15,  1.31it/s]
3|6058|Loss: 0.7681675553321838:  94%|█████████▎| 6057/6470 [1:15:06<05:15,  1.31it/s]
3|6058|Loss: 0.7681675553321838:  94%|█████████▎| 6058/6470 [1:15:07<05:05,  1.35it/s]
3|6059|Loss: 0.8217713236808777:  94%|█████████▎| 6058/6470 [1:15:07<05:05,  1.35it/s]
3|6059|Loss: 0.8217713236808777:  94%|█████████▎| 6059/6470 [1:15:08<05:08,  1.33it/s]
3|6060|Loss: 0.981338620185852:  94%|█████████▎| 6059/6470 [1:15:08<05:08,  1.33it/s] 
3|6060|Loss: 0.981338620185852:  94%|█████████▎| 6060/6470 [1:15:08<05:00,  1.37it/s]
3|6061|Loss: 0.8801876306533813:  94%|█████████▎| 6060/6470 [1:15:09<05:00,  1.37it/s]
3|6061|Loss: 0.8801876306533813:  94%|█████████▎| 6061/6470 [1:15:09<04:39,  1.46it/s]
3|6062|Loss: 0.7063713669776917:  94%|█████████▎| 6061/6470 [1:15:09<04:39,  1.46it/s]
3|6062|Loss: 0.7063713669776917:  94%|█████████▎| 6062/6470 [1:15:10<05:11,  1.31it/s]
3|6063|Loss: 0.8307026028633118:  94%|█████████▎| 6062/6470 [1:15:10<05:11,  1.31it/s]
3|6063|Loss: 0.8307026028633118:  94%|█████████▎| 6063/6470 [1:15:11<05:30,  1.23it/s]
3|6064|Loss: 0.7348141074180603:  94%|█████████▎| 6063/6470 [1:15:11<05:30,  1.23it/s]
3|6064|Loss: 0.7348141074180603:  94%|█████████▎| 6064/6470 [1:15:11<04:59,  1.36it/s]
3|6065|Loss: 0.6976163983345032:  94%|█████████▎| 6064/6470 [1:15:12<04:59,  1.36it/s]
3|6065|Loss: 0.6976163983345032:  94%|█████████▎| 6065/6470 [1:15:12<05:05,  1.32it/s]
3|6066|Loss: 0.8788062930107117:  94%|█████████▎| 6065/6470 [1:15:12<05:05,  1.32it/s]
3|6066|Loss: 0.8788062930107117:  94%|█████████▍| 6066/6470 [1:15:13<05:11,  1.30it/s]
3|6067|Loss: 0.8471601605415344:  94%|█████████▍| 6066/6470 [1:15:13<05:11,  1.30it/s]
3|6067|Loss: 0.8471601605415344:  94%|█████████▍| 6067/6470 [1:15:14<05:28,  1.23it/s]
3|6068|Loss: 0.7811906337738037:  94%|█████████▍| 6067/6470 [1:15:14<05:28,  1.23it/s]
3|6068|Loss: 0.7811906337738037:  94%|█████████▍| 6068/6470 [1:15:15<05:36,  1.19it/s]
3|6069|Loss: 0.8614159226417542:  94%|█████████▍| 6068/6470 [1:15:15<05:36,  1.19it/s]
3|6069|Loss: 0.8614159226417542:  94%|█████████▍| 6069/6470 [1:15:15<05:22,  1.24it/s]
3|6070|Loss: 0.8290243148803711:  94%|█████████▍| 6069/6470 [1:15:16<05:22,  1.24it/s]
3|6070|Loss: 0.8290243148803711:  94%|█████████▍| 6070/6470 [1:15:16<05:15,  1.27it/s]
3|6071|Loss: 0.7527323365211487:  94%|█████████▍| 6070/6470 [1:15:17<05:15,  1.27it/s]
3|6071|Loss: 0.7527323365211487:  94%|█████████▍| 6071/6470 [1:15:17<05:04,  1.31it/s]
3|6072|Loss: 0.8241149187088013:  94%|█████████▍| 6071/6470 [1:15:17<05:04,  1.31it/s]
3|6072|Loss: 0.8241149187088013:  94%|█████████▍| 6072/6470 [1:15:17<04:47,  1.38it/s]
3|6073|Loss: 0.8941270112991333:  94%|█████████▍| 6072/6470 [1:15:18<04:47,  1.38it/s]
3|6073|Loss: 0.8941270112991333:  94%|█████████▍| 6073/6470 [1:15:18<04:51,  1.36it/s]
3|6074|Loss: 0.9293429255485535:  94%|█████████▍| 6073/6470 [1:15:19<04:51,  1.36it/s]
3|6074|Loss: 0.9293429255485535:  94%|█████████▍| 6074/6470 [1:15:19<04:40,  1.41it/s]
3|6075|Loss: 0.8106223344802856:  94%|█████████▍| 6074/6470 [1:15:19<04:40,  1.41it/s]
3|6075|Loss: 0.8106223344802856:  94%|█████████▍| 6075/6470 [1:15:20<04:43,  1.39it/s]
3|6076|Loss: 0.9477708339691162:  94%|█████████▍| 6075/6470 [1:15:20<04:43,  1.39it/s]
3|6076|Loss: 0.9477708339691162:  94%|█████████▍| 6076/6470 [1:15:20<04:49,  1.36it/s]
3|6077|Loss: 0.9349666833877563:  94%|█████████▍| 6076/6470 [1:15:21<04:49,  1.36it/s]
3|6077|Loss: 0.9349666833877563:  94%|█████████▍| 6077/6470 [1:15:21<04:56,  1.33it/s]
3|6078|Loss: 0.8564044833183289:  94%|█████████▍| 6077/6470 [1:15:22<04:56,  1.33it/s]
3|6078|Loss: 0.8564044833183289:  94%|█████████▍| 6078/6470 [1:15:22<05:00,  1.31it/s]
3|6079|Loss: 0.8988661766052246:  94%|█████████▍| 6078/6470 [1:15:22<05:00,  1.31it/s]
3|6079|Loss: 0.8988661766052246:  94%|█████████▍| 6079/6470 [1:15:23<05:00,  1.30it/s]
3|6080|Loss: 0.8800293803215027:  94%|█████████▍| 6079/6470 [1:15:23<05:00,  1.30it/s]
3|6080|Loss: 0.8800293803215027:  94%|█████████▍| 6080/6470 [1:15:23<04:58,  1.31it/s]
3|6081|Loss: 0.7533167004585266:  94%|█████████▍| 6080/6470 [1:15:24<04:58,  1.31it/s]
3|6081|Loss: 0.7533167004585266:  94%|█████████▍| 6081/6470 [1:15:24<05:12,  1.24it/s]
3|6082|Loss: 0.8416813611984253:  94%|█████████▍| 6081/6470 [1:15:25<05:12,  1.24it/s]
3|6082|Loss: 0.8416813611984253:  94%|█████████▍| 6082/6470 [1:15:25<05:05,  1.27it/s]
3|6083|Loss: 0.7926732301712036:  94%|█████████▍| 6082/6470 [1:15:25<05:05,  1.27it/s]
3|6083|Loss: 0.7926732301712036:  94%|█████████▍| 6083/6470 [1:15:26<04:33,  1.41it/s]
3|6084|Loss: 0.8193917870521545:  94%|█████████▍| 6083/6470 [1:15:26<04:33,  1.41it/s]
3|6084|Loss: 0.8193917870521545:  94%|█████████▍| 6084/6470 [1:15:26<04:36,  1.40it/s]
3|6085|Loss: 0.7266474366188049:  94%|█████████▍| 6084/6470 [1:15:27<04:36,  1.40it/s]
3|6085|Loss: 0.7266474366188049:  94%|█████████▍| 6085/6470 [1:15:27<04:55,  1.30it/s]
3|6086|Loss: 0.7082870006561279:  94%|█████████▍| 6085/6470 [1:15:28<04:55,  1.30it/s]
3|6086|Loss: 0.7082870006561279:  94%|█████████▍| 6086/6470 [1:15:28<05:15,  1.22it/s]
3|6087|Loss: 0.7582029104232788:  94%|█████████▍| 6086/6470 [1:15:29<05:15,  1.22it/s]
3|6087|Loss: 0.7582029104232788:  94%|█████████▍| 6087/6470 [1:15:29<05:02,  1.27it/s]
3|6088|Loss: 0.775563657283783:  94%|█████████▍| 6087/6470 [1:15:29<05:02,  1.27it/s] 
3|6088|Loss: 0.775563657283783:  94%|█████████▍| 6088/6470 [1:15:30<04:48,  1.32it/s]
3|6089|Loss: 0.8366857767105103:  94%|█████████▍| 6088/6470 [1:15:30<04:48,  1.32it/s]
3|6089|Loss: 0.8366857767105103:  94%|█████████▍| 6089/6470 [1:15:31<05:03,  1.25it/s]
3|6090|Loss: 0.835274875164032:  94%|█████████▍| 6089/6470 [1:15:31<05:03,  1.25it/s] 
3|6090|Loss: 0.835274875164032:  94%|█████████▍| 6090/6470 [1:15:31<05:06,  1.24it/s]
3|6091|Loss: 0.9347852468490601:  94%|█████████▍| 6090/6470 [1:15:32<05:06,  1.24it/s]
3|6091|Loss: 0.9347852468490601:  94%|█████████▍| 6091/6470 [1:15:32<04:41,  1.35it/s]
3|6092|Loss: 0.8490029573440552:  94%|█████████▍| 6091/6470 [1:15:32<04:41,  1.35it/s]
3|6092|Loss: 0.8490029573440552:  94%|█████████▍| 6092/6470 [1:15:33<04:40,  1.35it/s]
3|6093|Loss: 0.8324983716011047:  94%|█████████▍| 6092/6470 [1:15:33<04:40,  1.35it/s]
3|6093|Loss: 0.8324983716011047:  94%|█████████▍| 6093/6470 [1:15:34<04:59,  1.26it/s]
3|6094|Loss: 0.7476198673248291:  94%|█████████▍| 6093/6470 [1:15:34<04:59,  1.26it/s]
3|6094|Loss: 0.7476198673248291:  94%|█████████▍| 6094/6470 [1:15:34<04:40,  1.34it/s]
3|6095|Loss: 0.8321740627288818:  94%|█████████▍| 6094/6470 [1:15:35<04:40,  1.34it/s]
3|6095|Loss: 0.8321740627288818:  94%|█████████▍| 6095/6470 [1:15:35<04:34,  1.37it/s]
3|6096|Loss: 0.7080074548721313:  94%|█████████▍| 6095/6470 [1:15:35<04:34,  1.37it/s]
3|6096|Loss: 0.7080074548721313:  94%|█████████▍| 6096/6470 [1:15:36<04:28,  1.39it/s]
3|6097|Loss: 0.7923870086669922:  94%|█████████▍| 6096/6470 [1:15:36<04:28,  1.39it/s]
3|6097|Loss: 0.7923870086669922:  94%|█████████▍| 6097/6470 [1:15:36<04:24,  1.41it/s]
3|6098|Loss: 0.962428867816925:  94%|█████████▍| 6097/6470 [1:15:37<04:24,  1.41it/s] 
3|6098|Loss: 0.962428867816925:  94%|█████████▍| 6098/6470 [1:15:37<04:28,  1.39it/s]
3|6099|Loss: 1.0277897119522095:  94%|█████████▍| 6098/6470 [1:15:38<04:28,  1.39it/s]
3|6099|Loss: 1.0277897119522095:  94%|█████████▍| 6099/6470 [1:15:38<04:45,  1.30it/s]
3|6100|Loss: 0.7348686456680298:  94%|█████████▍| 6099/6470 [1:15:38<04:45,  1.30it/s]
3|6100|Loss: 0.7348686456680298:  94%|█████████▍| 6100/6470 [1:15:39<04:33,  1.35it/s]
3|6101|Loss: 0.9033551216125488:  94%|█████████▍| 6100/6470 [1:15:39<04:33,  1.35it/s]
3|6101|Loss: 0.9033551216125488:  94%|█████████▍| 6101/6470 [1:15:39<04:19,  1.42it/s]
3|6102|Loss: 0.7380725741386414:  94%|█████████▍| 6101/6470 [1:15:40<04:19,  1.42it/s]
3|6102|Loss: 0.7380725741386414:  94%|█████████▍| 6102/6470 [1:15:40<04:09,  1.47it/s]
3|6103|Loss: 0.9696693420410156:  94%|█████████▍| 6102/6470 [1:15:40<04:09,  1.47it/s]
3|6103|Loss: 0.9696693420410156:  94%|█████████▍| 6103/6470 [1:15:41<04:15,  1.44it/s]
3|6104|Loss: 0.8340520262718201:  94%|█████████▍| 6103/6470 [1:15:41<04:15,  1.44it/s]
3|6104|Loss: 0.8340520262718201:  94%|█████████▍| 6104/6470 [1:15:41<04:23,  1.39it/s]
3|6105|Loss: 0.8242858648300171:  94%|█████████▍| 6104/6470 [1:15:42<04:23,  1.39it/s]
3|6105|Loss: 0.8242858648300171:  94%|█████████▍| 6105/6470 [1:15:42<04:27,  1.36it/s]
3|6106|Loss: 0.764035165309906:  94%|█████████▍| 6105/6470 [1:15:43<04:27,  1.36it/s] 
3|6106|Loss: 0.764035165309906:  94%|█████████▍| 6106/6470 [1:15:43<04:44,  1.28it/s]
3|6107|Loss: 0.7945029139518738:  94%|█████████▍| 6106/6470 [1:15:43<04:44,  1.28it/s]
3|6107|Loss: 0.7945029139518738:  94%|█████████▍| 6107/6470 [1:15:44<04:41,  1.29it/s]
3|6108|Loss: 0.8180508613586426:  94%|█████████▍| 6107/6470 [1:15:44<04:41,  1.29it/s]
3|6108|Loss: 0.8180508613586426:  94%|█████████▍| 6108/6470 [1:15:45<04:37,  1.30it/s]
3|6109|Loss: 0.7568216323852539:  94%|█████████▍| 6108/6470 [1:15:45<04:37,  1.30it/s]
3|6109|Loss: 0.7568216323852539:  94%|█████████▍| 6109/6470 [1:15:45<04:26,  1.35it/s]
3|6110|Loss: 0.9016898274421692:  94%|█████████▍| 6109/6470 [1:15:46<04:26,  1.35it/s]
3|6110|Loss: 0.9016898274421692:  94%|█████████▍| 6110/6470 [1:15:46<04:18,  1.39it/s]
3|6111|Loss: 0.7647790908813477:  94%|█████████▍| 6110/6470 [1:15:46<04:18,  1.39it/s]
3|6111|Loss: 0.7647790908813477:  94%|█████████▍| 6111/6470 [1:15:46<04:08,  1.45it/s]
3|6112|Loss: 0.8798702359199524:  94%|█████████▍| 6111/6470 [1:15:47<04:08,  1.45it/s]
3|6112|Loss: 0.8798702359199524:  94%|█████████▍| 6112/6470 [1:15:47<04:11,  1.43it/s]
3|6113|Loss: 0.9321616291999817:  94%|█████████▍| 6112/6470 [1:15:48<04:11,  1.43it/s]
3|6113|Loss: 0.9321616291999817:  94%|█████████▍| 6113/6470 [1:15:48<04:16,  1.39it/s]
3|6114|Loss: 0.7698287963867188:  94%|█████████▍| 6113/6470 [1:15:48<04:16,  1.39it/s]
3|6114|Loss: 0.7698287963867188:  94%|█████████▍| 6114/6470 [1:15:49<04:23,  1.35it/s]
3|6115|Loss: 0.8207132816314697:  94%|█████████▍| 6114/6470 [1:15:49<04:23,  1.35it/s]
3|6115|Loss: 0.8207132816314697:  95%|█████████▍| 6115/6470 [1:15:49<04:20,  1.36it/s]
3|6116|Loss: 0.8114362955093384:  95%|█████████▍| 6115/6470 [1:15:50<04:20,  1.36it/s]
3|6116|Loss: 0.8114362955093384:  95%|█████████▍| 6116/6470 [1:15:50<04:38,  1.27it/s]
3|6117|Loss: 0.7265165448188782:  95%|█████████▍| 6116/6470 [1:15:51<04:38,  1.27it/s]
3|6117|Loss: 0.7265165448188782:  95%|█████████▍| 6117/6470 [1:15:51<04:23,  1.34it/s]
3|6118|Loss: 0.9114077091217041:  95%|█████████▍| 6117/6470 [1:15:51<04:23,  1.34it/s]
3|6118|Loss: 0.9114077091217041:  95%|█████████▍| 6118/6470 [1:15:52<03:56,  1.49it/s]
3|6119|Loss: 0.8042188882827759:  95%|█████████▍| 6118/6470 [1:15:52<03:56,  1.49it/s]
3|6119|Loss: 0.8042188882827759:  95%|█████████▍| 6119/6470 [1:15:52<03:50,  1.53it/s]
3|6120|Loss: 0.8321557641029358:  95%|█████████▍| 6119/6470 [1:15:53<03:50,  1.53it/s]
3|6120|Loss: 0.8321557641029358:  95%|█████████▍| 6120/6470 [1:15:53<04:01,  1.45it/s]
3|6121|Loss: 0.6972617506980896:  95%|█████████▍| 6120/6470 [1:15:53<04:01,  1.45it/s]
3|6121|Loss: 0.6972617506980896:  95%|█████████▍| 6121/6470 [1:15:54<04:12,  1.38it/s]
3|6122|Loss: 0.7376722693443298:  95%|█████████▍| 6121/6470 [1:15:54<04:12,  1.38it/s]
3|6122|Loss: 0.7376722693443298:  95%|█████████▍| 6122/6470 [1:15:54<04:00,  1.45it/s]
3|6123|Loss: 0.8840739727020264:  95%|█████████▍| 6122/6470 [1:15:55<04:00,  1.45it/s]
3|6123|Loss: 0.8840739727020264:  95%|█████████▍| 6123/6470 [1:15:55<03:49,  1.51it/s]
3|6124|Loss: 0.8089577555656433:  95%|█████████▍| 6123/6470 [1:15:55<03:49,  1.51it/s]
3|6124|Loss: 0.8089577555656433:  95%|█████████▍| 6124/6470 [1:15:56<03:50,  1.50it/s]
3|6125|Loss: 0.8443422913551331:  95%|█████████▍| 6124/6470 [1:15:56<03:50,  1.50it/s]
3|6125|Loss: 0.8443422913551331:  95%|█████████▍| 6125/6470 [1:15:56<04:10,  1.38it/s]
3|6126|Loss: 0.7860425114631653:  95%|█████████▍| 6125/6470 [1:15:57<04:10,  1.38it/s]
3|6126|Loss: 0.7860425114631653:  95%|█████████▍| 6126/6470 [1:15:57<04:17,  1.33it/s]
3|6127|Loss: 0.8070524334907532:  95%|█████████▍| 6126/6470 [1:15:58<04:17,  1.33it/s]
3|6127|Loss: 0.8070524334907532:  95%|█████████▍| 6127/6470 [1:15:58<04:22,  1.31it/s]
3|6128|Loss: 0.7061126828193665:  95%|█████████▍| 6127/6470 [1:15:59<04:22,  1.31it/s]
3|6128|Loss: 0.7061126828193665:  95%|█████████▍| 6128/6470 [1:15:59<04:22,  1.30it/s]
3|6129|Loss: 0.9065833687782288:  95%|█████████▍| 6128/6470 [1:15:59<04:22,  1.30it/s]
3|6129|Loss: 0.9065833687782288:  95%|█████████▍| 6129/6470 [1:16:00<04:24,  1.29it/s]
3|6130|Loss: 0.9236711263656616:  95%|█████████▍| 6129/6470 [1:16:00<04:24,  1.29it/s]
3|6130|Loss: 0.9236711263656616:  95%|█████████▍| 6130/6470 [1:16:00<04:14,  1.34it/s]
3|6131|Loss: 0.7197189927101135:  95%|█████████▍| 6130/6470 [1:16:01<04:14,  1.34it/s]
3|6131|Loss: 0.7197189927101135:  95%|█████████▍| 6131/6470 [1:16:01<04:08,  1.37it/s]
3|6132|Loss: 0.8398220539093018:  95%|█████████▍| 6131/6470 [1:16:01<04:08,  1.37it/s]
3|6132|Loss: 0.8398220539093018:  95%|█████████▍| 6132/6470 [1:16:02<03:58,  1.42it/s]
3|6133|Loss: 0.9879048466682434:  95%|█████████▍| 6132/6470 [1:16:02<03:58,  1.42it/s]
3|6133|Loss: 0.9879048466682434:  95%|█████████▍| 6133/6470 [1:16:03<04:10,  1.35it/s]
3|6134|Loss: 0.7102266550064087:  95%|█████████▍| 6133/6470 [1:16:03<04:10,  1.35it/s]
3|6134|Loss: 0.7102266550064087:  95%|█████████▍| 6134/6470 [1:16:03<04:17,  1.30it/s]
3|6135|Loss: 0.8982587456703186:  95%|█████████▍| 6134/6470 [1:16:04<04:17,  1.30it/s]
3|6135|Loss: 0.8982587456703186:  95%|█████████▍| 6135/6470 [1:16:04<04:18,  1.30it/s]
3|6136|Loss: 0.9043304324150085:  95%|█████████▍| 6135/6470 [1:16:05<04:18,  1.30it/s]
3|6136|Loss: 0.9043304324150085:  95%|█████████▍| 6136/6470 [1:16:05<04:19,  1.28it/s]
3|6137|Loss: 1.030272364616394:  95%|█████████▍| 6136/6470 [1:16:05<04:19,  1.28it/s] 
3|6137|Loss: 1.030272364616394:  95%|█████████▍| 6137/6470 [1:16:06<04:18,  1.29it/s]
3|6138|Loss: 0.7754945158958435:  95%|█████████▍| 6137/6470 [1:16:06<04:18,  1.29it/s]
3|6138|Loss: 0.7754945158958435:  95%|█████████▍| 6138/6470 [1:16:07<04:32,  1.22it/s]
3|6139|Loss: 0.8052585124969482:  95%|█████████▍| 6138/6470 [1:16:07<04:32,  1.22it/s]
3|6139|Loss: 0.8052585124969482:  95%|█████████▍| 6139/6470 [1:16:07<04:16,  1.29it/s]
3|6140|Loss: 0.8465267419815063:  95%|█████████▍| 6139/6470 [1:16:08<04:16,  1.29it/s]
3|6140|Loss: 0.8465267419815063:  95%|█████████▍| 6140/6470 [1:16:08<03:59,  1.38it/s]
3|6141|Loss: 0.8603917360305786:  95%|█████████▍| 6140/6470 [1:16:08<03:59,  1.38it/s]
3|6141|Loss: 0.8603917360305786:  95%|█████████▍| 6141/6470 [1:16:09<04:03,  1.35it/s]
3|6142|Loss: 0.9906212687492371:  95%|█████████▍| 6141/6470 [1:16:09<04:03,  1.35it/s]
3|6142|Loss: 0.9906212687492371:  95%|█████████▍| 6142/6470 [1:16:09<04:00,  1.36it/s]
3|6143|Loss: 0.8981996774673462:  95%|█████████▍| 6142/6470 [1:16:10<04:00,  1.36it/s]
3|6143|Loss: 0.8981996774673462:  95%|█████████▍| 6143/6470 [1:16:10<04:01,  1.35it/s]
3|6144|Loss: 0.7996787428855896:  95%|█████████▍| 6143/6470 [1:16:11<04:01,  1.35it/s]
3|6144|Loss: 0.7996787428855896:  95%|█████████▍| 6144/6470 [1:16:11<04:02,  1.34it/s]
3|6145|Loss: 1.0209239721298218:  95%|█████████▍| 6144/6470 [1:16:11<04:02,  1.34it/s]
3|6145|Loss: 1.0209239721298218:  95%|█████████▍| 6145/6470 [1:16:12<03:56,  1.37it/s]
3|6146|Loss: 0.8276997208595276:  95%|█████████▍| 6145/6470 [1:16:12<03:56,  1.37it/s]
3|6146|Loss: 0.8276997208595276:  95%|█████████▍| 6146/6470 [1:16:12<03:51,  1.40it/s]
3|6147|Loss: 0.8731858134269714:  95%|█████████▍| 6146/6470 [1:16:13<03:51,  1.40it/s]
3|6147|Loss: 0.8731858134269714:  95%|█████████▌| 6147/6470 [1:16:13<03:40,  1.46it/s]
3|6148|Loss: 0.7567921876907349:  95%|█████████▌| 6147/6470 [1:16:13<03:40,  1.46it/s]
3|6148|Loss: 0.7567921876907349:  95%|█████████▌| 6148/6470 [1:16:14<03:47,  1.42it/s]
3|6149|Loss: 0.8936415314674377:  95%|█████████▌| 6148/6470 [1:16:14<03:47,  1.42it/s]
3|6149|Loss: 0.8936415314674377:  95%|█████████▌| 6149/6470 [1:16:14<03:55,  1.36it/s]
3|6150|Loss: 0.7832834124565125:  95%|█████████▌| 6149/6470 [1:16:15<03:55,  1.36it/s]
3|6150|Loss: 0.7832834124565125:  95%|█████████▌| 6150/6470 [1:16:15<03:45,  1.42it/s]
3|6151|Loss: 0.8872931003570557:  95%|█████████▌| 6150/6470 [1:16:15<03:45,  1.42it/s]
3|6151|Loss: 0.8872931003570557:  95%|█████████▌| 6151/6470 [1:16:16<03:46,  1.41it/s]
3|6152|Loss: 0.7790007591247559:  95%|█████████▌| 6151/6470 [1:16:16<03:46,  1.41it/s]
3|6152|Loss: 0.7790007591247559:  95%|█████████▌| 6152/6470 [1:16:17<03:52,  1.37it/s]
3|6153|Loss: 0.8912178874015808:  95%|█████████▌| 6152/6470 [1:16:17<03:52,  1.37it/s]
3|6153|Loss: 0.8912178874015808:  95%|█████████▌| 6153/6470 [1:16:17<03:26,  1.54it/s]
3|6154|Loss: 0.8586210608482361:  95%|█████████▌| 6153/6470 [1:16:17<03:26,  1.54it/s]
3|6154|Loss: 0.8586210608482361:  95%|█████████▌| 6154/6470 [1:16:18<03:25,  1.54it/s]
3|6155|Loss: 0.7109106183052063:  95%|█████████▌| 6154/6470 [1:16:18<03:25,  1.54it/s]
3|6155|Loss: 0.7109106183052063:  95%|█████████▌| 6155/6470 [1:16:18<03:34,  1.47it/s]
3|6156|Loss: 0.7667303085327148:  95%|█████████▌| 6155/6470 [1:16:19<03:34,  1.47it/s]
3|6156|Loss: 0.7667303085327148:  95%|█████████▌| 6156/6470 [1:16:19<03:33,  1.47it/s]
3|6157|Loss: 0.7823277115821838:  95%|█████████▌| 6156/6470 [1:16:20<03:33,  1.47it/s]
3|6157|Loss: 0.7823277115821838:  95%|█████████▌| 6157/6470 [1:16:20<03:47,  1.38it/s]
3|6158|Loss: 0.8940879106521606:  95%|█████████▌| 6157/6470 [1:16:20<03:47,  1.38it/s]
3|6158|Loss: 0.8940879106521606:  95%|█████████▌| 6158/6470 [1:16:21<03:45,  1.38it/s]
3|6159|Loss: 0.9580450654029846:  95%|█████████▌| 6158/6470 [1:16:21<03:45,  1.38it/s]
3|6159|Loss: 0.9580450654029846:  95%|█████████▌| 6159/6470 [1:16:21<03:46,  1.37it/s]
3|6160|Loss: 0.8220499753952026:  95%|█████████▌| 6159/6470 [1:16:22<03:46,  1.37it/s]
3|6160|Loss: 0.8220499753952026:  95%|█████████▌| 6160/6470 [1:16:22<03:40,  1.40it/s]
3|6161|Loss: 0.7976408004760742:  95%|█████████▌| 6160/6470 [1:16:22<03:40,  1.40it/s]
3|6161|Loss: 0.7976408004760742:  95%|█████████▌| 6161/6470 [1:16:23<03:21,  1.53it/s]
3|6162|Loss: 0.7787017226219177:  95%|█████████▌| 6161/6470 [1:16:23<03:21,  1.53it/s]
3|6162|Loss: 0.7787017226219177:  95%|█████████▌| 6162/6470 [1:16:23<03:20,  1.54it/s]
3|6163|Loss: 1.025238037109375:  95%|█████████▌| 6162/6470 [1:16:24<03:20,  1.54it/s] 
3|6163|Loss: 1.025238037109375:  95%|█████████▌| 6163/6470 [1:16:24<03:40,  1.39it/s]
3|6164|Loss: 0.8753684759140015:  95%|█████████▌| 6163/6470 [1:16:25<03:40,  1.39it/s]
3|6164|Loss: 0.8753684759140015:  95%|█████████▌| 6164/6470 [1:16:25<03:51,  1.32it/s]
3|6165|Loss: 1.001491665840149:  95%|█████████▌| 6164/6470 [1:16:25<03:51,  1.32it/s] 
3|6165|Loss: 1.001491665840149:  95%|█████████▌| 6165/6470 [1:16:26<04:02,  1.26it/s]
3|6166|Loss: 0.7324936985969543:  95%|█████████▌| 6165/6470 [1:16:26<04:02,  1.26it/s]
3|6166|Loss: 0.7324936985969543:  95%|█████████▌| 6166/6470 [1:16:27<03:50,  1.32it/s]
3|6167|Loss: 0.9588361978530884:  95%|█████████▌| 6166/6470 [1:16:27<03:50,  1.32it/s]
3|6167|Loss: 0.9588361978530884:  95%|█████████▌| 6167/6470 [1:16:27<03:42,  1.36it/s]
3|6168|Loss: 0.8720443248748779:  95%|█████████▌| 6167/6470 [1:16:28<03:42,  1.36it/s]
3|6168|Loss: 0.8720443248748779:  95%|█████████▌| 6168/6470 [1:16:28<03:29,  1.44it/s]
3|6169|Loss: 0.9143460392951965:  95%|█████████▌| 6168/6470 [1:16:28<03:29,  1.44it/s]
3|6169|Loss: 0.9143460392951965:  95%|█████████▌| 6169/6470 [1:16:28<03:25,  1.46it/s]
3|6170|Loss: 0.8699284195899963:  95%|█████████▌| 6169/6470 [1:16:29<03:25,  1.46it/s]
3|6170|Loss: 0.8699284195899963:  95%|█████████▌| 6170/6470 [1:16:29<03:21,  1.49it/s]
3|6171|Loss: 0.8080406188964844:  95%|█████████▌| 6170/6470 [1:16:29<03:21,  1.49it/s]
3|6171|Loss: 0.8080406188964844:  95%|█████████▌| 6171/6470 [1:16:30<03:16,  1.52it/s]
3|6172|Loss: 0.85161292552948:  95%|█████████▌| 6171/6470 [1:16:30<03:16,  1.52it/s]  
3|6172|Loss: 0.85161292552948:  95%|█████████▌| 6172/6470 [1:16:30<03:09,  1.57it/s]
3|6173|Loss: 1.0345827341079712:  95%|█████████▌| 6172/6470 [1:16:31<03:09,  1.57it/s]
3|6173|Loss: 1.0345827341079712:  95%|█████████▌| 6173/6470 [1:16:31<03:27,  1.43it/s]
3|6174|Loss: 0.8215373158454895:  95%|█████████▌| 6173/6470 [1:16:32<03:27,  1.43it/s]
3|6174|Loss: 0.8215373158454895:  95%|█████████▌| 6174/6470 [1:16:32<03:46,  1.31it/s]
3|6175|Loss: 0.8119472861289978:  95%|█████████▌| 6174/6470 [1:16:32<03:46,  1.31it/s]
3|6175|Loss: 0.8119472861289978:  95%|█████████▌| 6175/6470 [1:16:33<03:31,  1.40it/s]
3|6176|Loss: 0.8175150156021118:  95%|█████████▌| 6175/6470 [1:16:33<03:31,  1.40it/s]
3|6176|Loss: 0.8175150156021118:  95%|█████████▌| 6176/6470 [1:16:34<03:47,  1.29it/s]
3|6177|Loss: 0.8362721800804138:  95%|█████████▌| 6176/6470 [1:16:34<03:47,  1.29it/s]
3|6177|Loss: 0.8362721800804138:  95%|█████████▌| 6177/6470 [1:16:34<03:34,  1.37it/s]
3|6178|Loss: 0.817119836807251:  95%|█████████▌| 6177/6470 [1:16:35<03:34,  1.37it/s] 
3|6178|Loss: 0.817119836807251:  95%|█████████▌| 6178/6470 [1:16:35<03:44,  1.30it/s]
3|6179|Loss: 0.7587616443634033:  95%|█████████▌| 6178/6470 [1:16:36<03:44,  1.30it/s]
3|6179|Loss: 0.7587616443634033:  96%|█████████▌| 6179/6470 [1:16:36<03:45,  1.29it/s]
3|6180|Loss: 0.8036651611328125:  96%|█████████▌| 6179/6470 [1:16:36<03:45,  1.29it/s]
3|6180|Loss: 0.8036651611328125:  96%|█████████▌| 6180/6470 [1:16:37<03:36,  1.34it/s]
3|6181|Loss: 0.8500107526779175:  96%|█████████▌| 6180/6470 [1:16:37<03:36,  1.34it/s]
3|6181|Loss: 0.8500107526779175:  96%|█████████▌| 6181/6470 [1:16:37<03:29,  1.38it/s]
3|6182|Loss: 0.8454494476318359:  96%|█████████▌| 6181/6470 [1:16:38<03:29,  1.38it/s]
3|6182|Loss: 0.8454494476318359:  96%|█████████▌| 6182/6470 [1:16:38<03:38,  1.32it/s]
3|6183|Loss: 0.8313892483711243:  96%|█████████▌| 6182/6470 [1:16:39<03:38,  1.32it/s]
3|6183|Loss: 0.8313892483711243:  96%|█████████▌| 6183/6470 [1:16:39<03:46,  1.27it/s]
3|6184|Loss: 0.870658814907074:  96%|█████████▌| 6183/6470 [1:16:39<03:46,  1.27it/s] 
3|6184|Loss: 0.870658814907074:  96%|█████████▌| 6184/6470 [1:16:40<03:39,  1.30it/s]
3|6185|Loss: 0.8338949084281921:  96%|█████████▌| 6184/6470 [1:16:40<03:39,  1.30it/s]
3|6185|Loss: 0.8338949084281921:  96%|█████████▌| 6185/6470 [1:16:40<03:39,  1.30it/s]
3|6186|Loss: 0.7668250799179077:  96%|█████████▌| 6185/6470 [1:16:41<03:39,  1.30it/s]
3|6186|Loss: 0.7668250799179077:  96%|█████████▌| 6186/6470 [1:16:41<03:25,  1.38it/s]
3|6187|Loss: 0.8571613430976868:  96%|█████████▌| 6186/6470 [1:16:41<03:25,  1.38it/s]
3|6187|Loss: 0.8571613430976868:  96%|█████████▌| 6187/6470 [1:16:42<03:20,  1.41it/s]
3|6188|Loss: 0.8655218482017517:  96%|█████████▌| 6187/6470 [1:16:42<03:20,  1.41it/s]
3|6188|Loss: 0.8655218482017517:  96%|█████████▌| 6188/6470 [1:16:42<03:20,  1.41it/s]
3|6189|Loss: 0.8479992151260376:  96%|█████████▌| 6188/6470 [1:16:43<03:20,  1.41it/s]
3|6189|Loss: 0.8479992151260376:  96%|█████████▌| 6189/6470 [1:16:43<03:18,  1.41it/s]
3|6190|Loss: 0.7494493126869202:  96%|█████████▌| 6189/6470 [1:16:44<03:18,  1.41it/s]
3|6190|Loss: 0.7494493126869202:  96%|█████████▌| 6190/6470 [1:16:45<04:16,  1.09it/s]
3|6191|Loss: 0.8852468729019165:  96%|█████████▌| 6190/6470 [1:16:45<04:16,  1.09it/s]
3|6191|Loss: 0.8852468729019165:  96%|█████████▌| 6191/6470 [1:16:45<04:18,  1.08it/s]
3|6192|Loss: 0.7891659736633301:  96%|█████████▌| 6191/6470 [1:16:46<04:18,  1.08it/s]
3|6192|Loss: 0.7891659736633301:  96%|█████████▌| 6192/6470 [1:16:46<04:06,  1.13it/s]
3|6193|Loss: 0.7156524658203125:  96%|█████████▌| 6192/6470 [1:16:47<04:06,  1.13it/s]
3|6193|Loss: 0.7156524658203125:  96%|█████████▌| 6193/6470 [1:16:47<03:57,  1.17it/s]
3|6194|Loss: 0.9307811856269836:  96%|█████████▌| 6193/6470 [1:16:47<03:57,  1.17it/s]
3|6194|Loss: 0.9307811856269836:  96%|█████████▌| 6194/6470 [1:16:48<03:37,  1.27it/s]
3|6195|Loss: 0.9437220692634583:  96%|█████████▌| 6194/6470 [1:16:48<03:37,  1.27it/s]
3|6195|Loss: 0.9437220692634583:  96%|█████████▌| 6195/6470 [1:16:48<03:34,  1.28it/s]
3|6196|Loss: 0.9404351115226746:  96%|█████████▌| 6195/6470 [1:16:49<03:34,  1.28it/s]
3|6196|Loss: 0.9404351115226746:  96%|█████████▌| 6196/6470 [1:16:49<03:27,  1.32it/s]
3|6197|Loss: 0.9628586173057556:  96%|█████████▌| 6196/6470 [1:16:50<03:27,  1.32it/s]
3|6197|Loss: 0.9628586173057556:  96%|█████████▌| 6197/6470 [1:16:50<03:29,  1.30it/s]
3|6198|Loss: 0.7870633602142334:  96%|█████████▌| 6197/6470 [1:16:50<03:29,  1.30it/s]
3|6198|Loss: 0.7870633602142334:  96%|█████████▌| 6198/6470 [1:16:50<03:06,  1.46it/s]
3|6199|Loss: 0.7933775186538696:  96%|█████████▌| 6198/6470 [1:16:51<03:06,  1.46it/s]
3|6199|Loss: 0.7933775186538696:  96%|█████████▌| 6199/6470 [1:16:51<02:52,  1.57it/s]
3|6200|Loss: 0.7027234435081482:  96%|█████████▌| 6199/6470 [1:16:51<02:52,  1.57it/s]
3|6200|Loss: 0.7027234435081482:  96%|█████████▌| 6200/6470 [1:16:52<02:58,  1.51it/s]
3|6201|Loss: 0.7613105773925781:  96%|█████████▌| 6200/6470 [1:16:52<02:58,  1.51it/s]
3|6201|Loss: 0.7613105773925781:  96%|█████████▌| 6201/6470 [1:16:52<03:11,  1.40it/s]
3|6202|Loss: 0.8036935329437256:  96%|█████████▌| 6201/6470 [1:16:53<03:11,  1.40it/s]
3|6202|Loss: 0.8036935329437256:  96%|█████████▌| 6202/6470 [1:16:53<03:16,  1.37it/s]
3|6203|Loss: 1.052991509437561:  96%|█████████▌| 6202/6470 [1:16:54<03:16,  1.37it/s] 
3|6203|Loss: 1.052991509437561:  96%|█████████▌| 6203/6470 [1:16:54<03:05,  1.44it/s]
3|6204|Loss: 0.7797332406044006:  96%|█████████▌| 6203/6470 [1:16:54<03:05,  1.44it/s]
3|6204|Loss: 0.7797332406044006:  96%|█████████▌| 6204/6470 [1:16:55<03:17,  1.35it/s]
3|6205|Loss: 0.8630300164222717:  96%|█████████▌| 6204/6470 [1:16:55<03:17,  1.35it/s]
3|6205|Loss: 0.8630300164222717:  96%|█████████▌| 6205/6470 [1:16:56<03:29,  1.27it/s]
3|6206|Loss: 0.9470214247703552:  96%|█████████▌| 6205/6470 [1:16:56<03:29,  1.27it/s]
3|6206|Loss: 0.9470214247703552:  96%|█████████▌| 6206/6470 [1:16:56<03:28,  1.27it/s]
3|6207|Loss: 0.8566325306892395:  96%|█████████▌| 6206/6470 [1:16:57<03:28,  1.27it/s]
3|6207|Loss: 0.8566325306892395:  96%|█████████▌| 6207/6470 [1:16:57<03:33,  1.23it/s]
3|6208|Loss: 0.7495280504226685:  96%|█████████▌| 6207/6470 [1:16:58<03:33,  1.23it/s]
3|6208|Loss: 0.7495280504226685:  96%|█████████▌| 6208/6470 [1:16:58<03:24,  1.28it/s]
3|6209|Loss: 0.8349142670631409:  96%|█████████▌| 6208/6470 [1:16:58<03:24,  1.28it/s]
3|6209|Loss: 0.8349142670631409:  96%|█████████▌| 6209/6470 [1:16:59<03:16,  1.33it/s]
3|6210|Loss: 0.7340561747550964:  96%|█████████▌| 6209/6470 [1:16:59<03:16,  1.33it/s]
3|6210|Loss: 0.7340561747550964:  96%|█████████▌| 6210/6470 [1:17:00<03:28,  1.25it/s]
3|6211|Loss: 0.8109234571456909:  96%|█████████▌| 6210/6470 [1:17:00<03:28,  1.25it/s]
3|6211|Loss: 0.8109234571456909:  96%|█████████▌| 6211/6470 [1:17:01<03:39,  1.18it/s]
3|6212|Loss: 0.9511744976043701:  96%|█████████▌| 6211/6470 [1:17:01<03:39,  1.18it/s]
3|6212|Loss: 0.9511744976043701:  96%|█████████▌| 6212/6470 [1:17:01<03:39,  1.18it/s]
3|6213|Loss: 0.8827500343322754:  96%|█████████▌| 6212/6470 [1:17:02<03:39,  1.18it/s]
3|6213|Loss: 0.8827500343322754:  96%|█████████▌| 6213/6470 [1:17:02<03:24,  1.26it/s]
3|6214|Loss: 0.7740763425827026:  96%|█████████▌| 6213/6470 [1:17:02<03:24,  1.26it/s]
3|6214|Loss: 0.7740763425827026:  96%|█████████▌| 6214/6470 [1:17:03<03:20,  1.28it/s]
3|6215|Loss: 0.8231150507926941:  96%|█████████▌| 6214/6470 [1:17:03<03:20,  1.28it/s]
3|6215|Loss: 0.8231150507926941:  96%|█████████▌| 6215/6470 [1:17:04<03:14,  1.31it/s]
3|6216|Loss: 0.9184345006942749:  96%|█████████▌| 6215/6470 [1:17:04<03:14,  1.31it/s]
3|6216|Loss: 0.9184345006942749:  96%|█████████▌| 6216/6470 [1:17:04<03:06,  1.36it/s]
3|6217|Loss: 0.8304826021194458:  96%|█████████▌| 6216/6470 [1:17:05<03:06,  1.36it/s]
3|6217|Loss: 0.8304826021194458:  96%|█████████▌| 6217/6470 [1:17:05<02:55,  1.45it/s]
3|6218|Loss: 0.9111770391464233:  96%|█████████▌| 6217/6470 [1:17:05<02:55,  1.45it/s]
3|6218|Loss: 0.9111770391464233:  96%|█████████▌| 6218/6470 [1:17:05<02:50,  1.48it/s]
3|6219|Loss: 0.8461947441101074:  96%|█████████▌| 6218/6470 [1:17:06<02:50,  1.48it/s]
3|6219|Loss: 0.8461947441101074:  96%|█████████▌| 6219/6470 [1:17:06<02:50,  1.47it/s]
3|6220|Loss: 0.7966457605361938:  96%|█████████▌| 6219/6470 [1:17:06<02:50,  1.47it/s]
3|6220|Loss: 0.7966457605361938:  96%|█████████▌| 6220/6470 [1:17:07<02:44,  1.52it/s]
3|6221|Loss: 0.7314884066581726:  96%|█████████▌| 6220/6470 [1:17:07<02:44,  1.52it/s]
3|6221|Loss: 0.7314884066581726:  96%|█████████▌| 6221/6470 [1:17:07<02:43,  1.53it/s]
3|6222|Loss: 0.9831311106681824:  96%|█████████▌| 6221/6470 [1:17:08<02:43,  1.53it/s]
3|6222|Loss: 0.9831311106681824:  96%|█████████▌| 6222/6470 [1:17:08<02:59,  1.38it/s]
3|6223|Loss: 0.8182983994483948:  96%|█████████▌| 6222/6470 [1:17:09<02:59,  1.38it/s]
3|6223|Loss: 0.8182983994483948:  96%|█████████▌| 6223/6470 [1:17:09<03:08,  1.31it/s]
3|6224|Loss: 0.8419833183288574:  96%|█████████▌| 6223/6470 [1:17:10<03:08,  1.31it/s]
3|6224|Loss: 0.8419833183288574:  96%|█████████▌| 6224/6470 [1:17:10<03:11,  1.28it/s]
3|6225|Loss: 0.772363543510437:  96%|█████████▌| 6224/6470 [1:17:10<03:11,  1.28it/s] 
3|6225|Loss: 0.772363543510437:  96%|█████████▌| 6225/6470 [1:17:11<03:16,  1.25it/s]
3|6226|Loss: 0.8801525831222534:  96%|█████████▌| 6225/6470 [1:17:11<03:16,  1.25it/s]
3|6226|Loss: 0.8801525831222534:  96%|█████████▌| 6226/6470 [1:17:12<03:13,  1.26it/s]
3|6227|Loss: 0.8381781578063965:  96%|█████████▌| 6226/6470 [1:17:12<03:13,  1.26it/s]
3|6227|Loss: 0.8381781578063965:  96%|█████████▌| 6227/6470 [1:17:12<03:07,  1.30it/s]
3|6228|Loss: 0.8804323077201843:  96%|█████████▌| 6227/6470 [1:17:13<03:07,  1.30it/s]
3|6228|Loss: 0.8804323077201843:  96%|█████████▋| 6228/6470 [1:17:13<03:06,  1.30it/s]
3|6229|Loss: 0.8358598351478577:  96%|█████████▋| 6228/6470 [1:17:13<03:06,  1.30it/s]
3|6229|Loss: 0.8358598351478577:  96%|█████████▋| 6229/6470 [1:17:14<02:57,  1.36it/s]
3|6230|Loss: 0.7358837723731995:  96%|█████████▋| 6229/6470 [1:17:14<02:57,  1.36it/s]
3|6230|Loss: 0.7358837723731995:  96%|█████████▋| 6230/6470 [1:17:14<02:47,  1.44it/s]
3|6231|Loss: 0.8518648147583008:  96%|█████████▋| 6230/6470 [1:17:15<02:47,  1.44it/s]
3|6231|Loss: 0.8518648147583008:  96%|█████████▋| 6231/6470 [1:17:15<02:36,  1.53it/s]
3|6232|Loss: 0.818221390247345:  96%|█████████▋| 6231/6470 [1:17:15<02:36,  1.53it/s] 
3|6232|Loss: 0.818221390247345:  96%|█████████▋| 6232/6470 [1:17:16<02:42,  1.47it/s]
3|6233|Loss: 0.8952383399009705:  96%|█████████▋| 6232/6470 [1:17:16<02:42,  1.47it/s]
3|6233|Loss: 0.8952383399009705:  96%|█████████▋| 6233/6470 [1:17:16<02:54,  1.36it/s]
3|6234|Loss: 0.9151570200920105:  96%|█████████▋| 6233/6470 [1:17:17<02:54,  1.36it/s]
3|6234|Loss: 0.9151570200920105:  96%|█████████▋| 6234/6470 [1:17:17<02:52,  1.37it/s]
3|6235|Loss: 0.840996265411377:  96%|█████████▋| 6234/6470 [1:17:18<02:52,  1.37it/s] 
3|6235|Loss: 0.840996265411377:  96%|█████████▋| 6235/6470 [1:17:18<02:59,  1.31it/s]
3|6236|Loss: 0.669367253780365:  96%|█████████▋| 6235/6470 [1:17:18<02:59,  1.31it/s]
3|6236|Loss: 0.669367253780365:  96%|█████████▋| 6236/6470 [1:17:19<02:56,  1.33it/s]
3|6237|Loss: 0.7666952013969421:  96%|█████████▋| 6236/6470 [1:17:19<02:56,  1.33it/s]
3|6237|Loss: 0.7666952013969421:  96%|█████████▋| 6237/6470 [1:17:20<02:56,  1.32it/s]
3|6238|Loss: 0.9753645062446594:  96%|█████████▋| 6237/6470 [1:17:20<02:56,  1.32it/s]
3|6238|Loss: 0.9753645062446594:  96%|█████████▋| 6238/6470 [1:17:20<02:50,  1.36it/s]
3|6239|Loss: 0.8401166796684265:  96%|█████████▋| 6238/6470 [1:17:21<02:50,  1.36it/s]
3|6239|Loss: 0.8401166796684265:  96%|█████████▋| 6239/6470 [1:17:21<02:45,  1.40it/s]
3|6240|Loss: 0.7223694324493408:  96%|█████████▋| 6239/6470 [1:17:21<02:45,  1.40it/s]
3|6240|Loss: 0.7223694324493408:  96%|█████████▋| 6240/6470 [1:17:22<02:43,  1.41it/s]
3|6241|Loss: 0.9666871428489685:  96%|█████████▋| 6240/6470 [1:17:22<02:43,  1.41it/s]
3|6241|Loss: 0.9666871428489685:  96%|█████████▋| 6241/6470 [1:17:22<02:45,  1.38it/s]
3|6242|Loss: 0.8753915429115295:  96%|█████████▋| 6241/6470 [1:17:23<02:45,  1.38it/s]
3|6242|Loss: 0.8753915429115295:  96%|█████████▋| 6242/6470 [1:17:23<02:51,  1.33it/s]
3|6243|Loss: 0.693060040473938:  96%|█████████▋| 6242/6470 [1:17:24<02:51,  1.33it/s] 
3|6243|Loss: 0.693060040473938:  96%|█████████▋| 6243/6470 [1:17:24<03:03,  1.24it/s]
3|6244|Loss: 0.6710636019706726:  96%|█████████▋| 6243/6470 [1:17:25<03:03,  1.24it/s]
3|6244|Loss: 0.6710636019706726:  97%|█████████▋| 6244/6470 [1:17:25<03:08,  1.20it/s]
3|6245|Loss: 0.805069625377655:  97%|█████████▋| 6244/6470 [1:17:25<03:08,  1.20it/s] 
3|6245|Loss: 0.805069625377655:  97%|█████████▋| 6245/6470 [1:17:26<03:06,  1.21it/s]
3|6246|Loss: 0.7746520042419434:  97%|█████████▋| 6245/6470 [1:17:26<03:06,  1.21it/s]
3|6246|Loss: 0.7746520042419434:  97%|█████████▋| 6246/6470 [1:17:27<03:14,  1.15it/s]
3|6247|Loss: 0.9701515436172485:  97%|█████████▋| 6246/6470 [1:17:27<03:14,  1.15it/s]
3|6247|Loss: 0.9701515436172485:  97%|█████████▋| 6247/6470 [1:17:28<03:06,  1.20it/s]
3|6248|Loss: 0.8550981283187866:  97%|█████████▋| 6247/6470 [1:17:28<03:06,  1.20it/s]
3|6248|Loss: 0.8550981283187866:  97%|█████████▋| 6248/6470 [1:17:28<03:00,  1.23it/s]
3|6249|Loss: 0.71314537525177:  97%|█████████▋| 6248/6470 [1:17:29<03:00,  1.23it/s]  
3|6249|Loss: 0.71314537525177:  97%|█████████▋| 6249/6470 [1:17:29<02:55,  1.26it/s]
3|6250|Loss: 0.8393748998641968:  97%|█████████▋| 6249/6470 [1:17:29<02:55,  1.26it/s]
3|6250|Loss: 0.8393748998641968:  97%|█████████▋| 6250/6470 [1:17:30<02:57,  1.24it/s]
3|6251|Loss: 0.7986707091331482:  97%|█████████▋| 6250/6470 [1:17:30<02:57,  1.24it/s]
3|6251|Loss: 0.7986707091331482:  97%|█████████▋| 6251/6470 [1:17:31<02:52,  1.27it/s]
3|6252|Loss: 0.8808308839797974:  97%|█████████▋| 6251/6470 [1:17:31<02:52,  1.27it/s]
3|6252|Loss: 0.8808308839797974:  97%|█████████▋| 6252/6470 [1:17:31<02:51,  1.27it/s]
3|6253|Loss: 0.859813392162323:  97%|█████████▋| 6252/6470 [1:17:32<02:51,  1.27it/s] 
3|6253|Loss: 0.859813392162323:  97%|█████████▋| 6253/6470 [1:17:32<03:01,  1.20it/s]
3|6254|Loss: 0.7664190530776978:  97%|█████████▋| 6253/6470 [1:17:33<03:01,  1.20it/s]
3|6254|Loss: 0.7664190530776978:  97%|█████████▋| 6254/6470 [1:17:33<02:39,  1.35it/s]
3|6255|Loss: 0.820831298828125:  97%|█████████▋| 6254/6470 [1:17:33<02:39,  1.35it/s] 
3|6255|Loss: 0.820831298828125:  97%|█████████▋| 6255/6470 [1:17:34<02:37,  1.37it/s]
3|6256|Loss: 0.8774942755699158:  97%|█████████▋| 6255/6470 [1:17:34<02:37,  1.37it/s]
3|6256|Loss: 0.8774942755699158:  97%|█████████▋| 6256/6470 [1:17:34<02:41,  1.33it/s]
3|6257|Loss: 0.593710720539093:  97%|█████████▋| 6256/6470 [1:17:35<02:41,  1.33it/s] 
3|6257|Loss: 0.593710720539093:  97%|█████████▋| 6257/6470 [1:17:35<02:42,  1.31it/s]
3|6258|Loss: 0.7611264586448669:  97%|█████████▋| 6257/6470 [1:17:36<02:42,  1.31it/s]
3|6258|Loss: 0.7611264586448669:  97%|█████████▋| 6258/6470 [1:17:36<02:38,  1.34it/s]
3|6259|Loss: 0.8826251029968262:  97%|█████████▋| 6258/6470 [1:17:36<02:38,  1.34it/s]
3|6259|Loss: 0.8826251029968262:  97%|█████████▋| 6259/6470 [1:17:37<02:33,  1.38it/s]
3|6260|Loss: 0.7337594032287598:  97%|█████████▋| 6259/6470 [1:17:37<02:33,  1.38it/s]
3|6260|Loss: 0.7337594032287598:  97%|█████████▋| 6260/6470 [1:17:37<02:41,  1.30it/s]
3|6261|Loss: 0.7461663484573364:  97%|█████████▋| 6260/6470 [1:17:38<02:41,  1.30it/s]
3|6261|Loss: 0.7461663484573364:  97%|█████████▋| 6261/6470 [1:17:38<02:44,  1.27it/s]
3|6262|Loss: 0.687372088432312:  97%|█████████▋| 6261/6470 [1:17:39<02:44,  1.27it/s] 
3|6262|Loss: 0.687372088432312:  97%|█████████▋| 6262/6470 [1:17:39<02:37,  1.32it/s]
3|6263|Loss: 0.7144798636436462:  97%|█████████▋| 6262/6470 [1:17:39<02:37,  1.32it/s]
3|6263|Loss: 0.7144798636436462:  97%|█████████▋| 6263/6470 [1:17:40<02:51,  1.21it/s]
3|6264|Loss: 0.8240734934806824:  97%|█████████▋| 6263/6470 [1:17:40<02:51,  1.21it/s]
3|6264|Loss: 0.8240734934806824:  97%|█████████▋| 6264/6470 [1:17:41<02:51,  1.20it/s]
3|6265|Loss: 0.8007570505142212:  97%|█████████▋| 6264/6470 [1:17:41<02:51,  1.20it/s]
3|6265|Loss: 0.8007570505142212:  97%|█████████▋| 6265/6470 [1:17:41<02:41,  1.27it/s]
3|6266|Loss: 0.7420248985290527:  97%|█████████▋| 6265/6470 [1:17:42<02:41,  1.27it/s]
3|6266|Loss: 0.7420248985290527:  97%|█████████▋| 6266/6470 [1:17:42<02:38,  1.29it/s]
3|6267|Loss: 0.720365583896637:  97%|█████████▋| 6266/6470 [1:17:43<02:38,  1.29it/s] 
3|6267|Loss: 0.720365583896637:  97%|█████████▋| 6267/6470 [1:17:43<02:28,  1.36it/s]
3|6268|Loss: 0.8275022506713867:  97%|█████████▋| 6267/6470 [1:17:43<02:28,  1.36it/s]
3|6268|Loss: 0.8275022506713867:  97%|█████████▋| 6268/6470 [1:17:44<02:26,  1.37it/s]
3|6269|Loss: 0.822533905506134:  97%|█████████▋| 6268/6470 [1:17:44<02:26,  1.37it/s] 
3|6269|Loss: 0.822533905506134:  97%|█████████▋| 6269/6470 [1:17:44<02:33,  1.31it/s]
3|6270|Loss: 0.7910513877868652:  97%|█████████▋| 6269/6470 [1:17:45<02:33,  1.31it/s]
3|6270|Loss: 0.7910513877868652:  97%|█████████▋| 6270/6470 [1:17:45<02:26,  1.36it/s]
3|6271|Loss: 0.7585228681564331:  97%|█████████▋| 6270/6470 [1:17:45<02:26,  1.36it/s]
3|6271|Loss: 0.7585228681564331:  97%|█████████▋| 6271/6470 [1:17:46<02:25,  1.36it/s]
3|6272|Loss: 0.7687916159629822:  97%|█████████▋| 6271/6470 [1:17:46<02:25,  1.36it/s]
3|6272|Loss: 0.7687916159629822:  97%|█████████▋| 6272/6470 [1:17:47<02:27,  1.34it/s]
3|6273|Loss: 0.8625451922416687:  97%|█████████▋| 6272/6470 [1:17:47<02:27,  1.34it/s]
3|6273|Loss: 0.8625451922416687:  97%|█████████▋| 6273/6470 [1:17:47<02:18,  1.43it/s]
3|6274|Loss: 0.7712373733520508:  97%|█████████▋| 6273/6470 [1:17:47<02:18,  1.43it/s]
3|6274|Loss: 0.7712373733520508:  97%|█████████▋| 6274/6470 [1:17:48<02:08,  1.53it/s]
3|6275|Loss: 1.0037078857421875:  97%|█████████▋| 6274/6470 [1:17:48<02:08,  1.53it/s]
3|6275|Loss: 1.0037078857421875:  97%|█████████▋| 6275/6470 [1:17:48<02:04,  1.56it/s]
3|6276|Loss: 0.855947732925415:  97%|█████████▋| 6275/6470 [1:17:49<02:04,  1.56it/s] 
3|6276|Loss: 0.855947732925415:  97%|█████████▋| 6276/6470 [1:17:49<02:09,  1.50it/s]
3|6277|Loss: 0.7986943125724792:  97%|█████████▋| 6276/6470 [1:17:50<02:09,  1.50it/s]
3|6277|Loss: 0.7986943125724792:  97%|█████████▋| 6277/6470 [1:17:50<02:20,  1.37it/s]
3|6278|Loss: 0.8643031120300293:  97%|█████████▋| 6277/6470 [1:17:50<02:20,  1.37it/s]
3|6278|Loss: 0.8643031120300293:  97%|█████████▋| 6278/6470 [1:17:51<02:23,  1.34it/s]
3|6279|Loss: 0.8333555459976196:  97%|█████████▋| 6278/6470 [1:17:51<02:23,  1.34it/s]
3|6279|Loss: 0.8333555459976196:  97%|█████████▋| 6279/6470 [1:17:51<02:15,  1.41it/s]
3|6280|Loss: 0.8449335098266602:  97%|█████████▋| 6279/6470 [1:17:52<02:15,  1.41it/s]
3|6280|Loss: 0.8449335098266602:  97%|█████████▋| 6280/6470 [1:17:52<02:17,  1.38it/s]
3|6281|Loss: 0.8921678066253662:  97%|█████████▋| 6280/6470 [1:17:53<02:17,  1.38it/s]
3|6281|Loss: 0.8921678066253662:  97%|█████████▋| 6281/6470 [1:17:53<02:20,  1.35it/s]
3|6282|Loss: 0.8852335214614868:  97%|█████████▋| 6281/6470 [1:17:53<02:20,  1.35it/s]
3|6282|Loss: 0.8852335214614868:  97%|█████████▋| 6282/6470 [1:17:54<02:27,  1.27it/s]
3|6283|Loss: 0.8675486445426941:  97%|█████████▋| 6282/6470 [1:17:54<02:27,  1.27it/s]
3|6283|Loss: 0.8675486445426941:  97%|█████████▋| 6283/6470 [1:17:55<02:24,  1.29it/s]
3|6284|Loss: 0.6933047771453857:  97%|█████████▋| 6283/6470 [1:17:55<02:24,  1.29it/s]
3|6284|Loss: 0.6933047771453857:  97%|█████████▋| 6284/6470 [1:17:55<02:23,  1.30it/s]
3|6285|Loss: 0.8976251482963562:  97%|█████████▋| 6284/6470 [1:17:56<02:23,  1.30it/s]
3|6285|Loss: 0.8976251482963562:  97%|█████████▋| 6285/6470 [1:17:56<02:19,  1.32it/s]
3|6286|Loss: 0.6978569030761719:  97%|█████████▋| 6285/6470 [1:17:56<02:19,  1.32it/s]
3|6286|Loss: 0.6978569030761719:  97%|█████████▋| 6286/6470 [1:17:57<02:18,  1.33it/s]
3|6287|Loss: 0.8445281386375427:  97%|█████████▋| 6286/6470 [1:17:57<02:18,  1.33it/s]
3|6287|Loss: 0.8445281386375427:  97%|█████████▋| 6287/6470 [1:17:57<02:13,  1.37it/s]
3|6288|Loss: 0.8838374614715576:  97%|█████████▋| 6287/6470 [1:17:58<02:13,  1.37it/s]
3|6288|Loss: 0.8838374614715576:  97%|█████████▋| 6288/6470 [1:17:58<02:05,  1.45it/s]
3|6289|Loss: 0.9006423354148865:  97%|█████████▋| 6288/6470 [1:17:58<02:05,  1.45it/s]
3|6289|Loss: 0.9006423354148865:  97%|█████████▋| 6289/6470 [1:17:59<02:12,  1.36it/s]
3|6290|Loss: 0.9548430442810059:  97%|█████████▋| 6289/6470 [1:17:59<02:12,  1.36it/s]
3|6290|Loss: 0.9548430442810059:  97%|█████████▋| 6290/6470 [1:18:00<02:15,  1.32it/s]
3|6291|Loss: 0.9012784361839294:  97%|█████████▋| 6290/6470 [1:18:00<02:15,  1.32it/s]
3|6291|Loss: 0.9012784361839294:  97%|█████████▋| 6291/6470 [1:18:01<02:20,  1.27it/s]
3|6292|Loss: 0.807744562625885:  97%|█████████▋| 6291/6470 [1:18:01<02:20,  1.27it/s] 
3|6292|Loss: 0.807744562625885:  97%|█████████▋| 6292/6470 [1:18:01<02:17,  1.29it/s]
3|6293|Loss: 0.7605127692222595:  97%|█████████▋| 6292/6470 [1:18:02<02:17,  1.29it/s]
3|6293|Loss: 0.7605127692222595:  97%|█████████▋| 6293/6470 [1:18:02<02:15,  1.30it/s]
3|6294|Loss: 0.8170774579048157:  97%|█████████▋| 6293/6470 [1:18:02<02:15,  1.30it/s]
3|6294|Loss: 0.8170774579048157:  97%|█████████▋| 6294/6470 [1:18:03<02:03,  1.42it/s]
3|6295|Loss: 0.7561591267585754:  97%|█████████▋| 6294/6470 [1:18:03<02:03,  1.42it/s]
3|6295|Loss: 0.7561591267585754:  97%|█████████▋| 6295/6470 [1:18:03<01:57,  1.49it/s]
3|6296|Loss: 0.8795058131217957:  97%|█████████▋| 6295/6470 [1:18:04<01:57,  1.49it/s]
3|6296|Loss: 0.8795058131217957:  97%|█████████▋| 6296/6470 [1:18:04<01:57,  1.48it/s]
3|6297|Loss: 0.8559014201164246:  97%|█████████▋| 6296/6470 [1:18:04<01:57,  1.48it/s]
3|6297|Loss: 0.8559014201164246:  97%|█████████▋| 6297/6470 [1:18:05<02:02,  1.42it/s]
3|6298|Loss: 0.7693382501602173:  97%|█████████▋| 6297/6470 [1:18:05<02:02,  1.42it/s]
3|6298|Loss: 0.7693382501602173:  97%|█████████▋| 6298/6470 [1:18:05<02:01,  1.42it/s]
3|6299|Loss: 0.8454713225364685:  97%|█████████▋| 6298/6470 [1:18:06<02:01,  1.42it/s]
3|6299|Loss: 0.8454713225364685:  97%|█████████▋| 6299/6470 [1:18:06<02:02,  1.40it/s]
3|6300|Loss: 0.9069523811340332:  97%|█████████▋| 6299/6470 [1:18:07<02:02,  1.40it/s]
3|6300|Loss: 0.9069523811340332:  97%|█████████▋| 6300/6470 [1:18:07<02:10,  1.31it/s]
3|6301|Loss: 0.8086504936218262:  97%|█████████▋| 6300/6470 [1:18:07<02:10,  1.31it/s]
3|6301|Loss: 0.8086504936218262:  97%|█████████▋| 6301/6470 [1:18:08<02:16,  1.24it/s]
3|6302|Loss: 0.8311924338340759:  97%|█████████▋| 6301/6470 [1:18:08<02:16,  1.24it/s]
3|6302|Loss: 0.8311924338340759:  97%|█████████▋| 6302/6470 [1:18:09<02:19,  1.20it/s]
3|6303|Loss: 1.2247596979141235:  97%|█████████▋| 6302/6470 [1:18:09<02:19,  1.20it/s]
3|6303|Loss: 1.2247596979141235:  97%|█████████▋| 6303/6470 [1:18:10<02:21,  1.18it/s]
3|6304|Loss: 0.820075273513794:  97%|█████████▋| 6303/6470 [1:18:10<02:21,  1.18it/s] 
3|6304|Loss: 0.820075273513794:  97%|█████████▋| 6304/6470 [1:18:10<02:18,  1.20it/s]
3|6305|Loss: 0.8439480662345886:  97%|█████████▋| 6304/6470 [1:18:11<02:18,  1.20it/s]
3|6305|Loss: 0.8439480662345886:  97%|█████████▋| 6305/6470 [1:18:11<02:16,  1.21it/s]
3|6306|Loss: 0.8304117321968079:  97%|█████████▋| 6305/6470 [1:18:12<02:16,  1.21it/s]
3|6306|Loss: 0.8304117321968079:  97%|█████████▋| 6306/6470 [1:18:12<02:04,  1.31it/s]
3|6307|Loss: 0.8020760416984558:  97%|█████████▋| 6306/6470 [1:18:12<02:04,  1.31it/s]
3|6307|Loss: 0.8020760416984558:  97%|█████████▋| 6307/6470 [1:18:13<02:08,  1.27it/s]
3|6308|Loss: 0.7431977391242981:  97%|█████████▋| 6307/6470 [1:18:13<02:08,  1.27it/s]
3|6308|Loss: 0.7431977391242981:  97%|█████████▋| 6308/6470 [1:18:13<02:04,  1.31it/s]
3|6309|Loss: 0.8974279761314392:  97%|█████████▋| 6308/6470 [1:18:14<02:04,  1.31it/s]
3|6309|Loss: 0.8974279761314392:  98%|█████████▊| 6309/6470 [1:18:14<02:03,  1.30it/s]
3|6310|Loss: 0.7503580451011658:  98%|█████████▊| 6309/6470 [1:18:15<02:03,  1.30it/s]
3|6310|Loss: 0.7503580451011658:  98%|█████████▊| 6310/6470 [1:18:15<01:48,  1.47it/s]
3|6311|Loss: 0.8838958144187927:  98%|█████████▊| 6310/6470 [1:18:15<01:48,  1.47it/s]
3|6311|Loss: 0.8838958144187927:  98%|█████████▊| 6311/6470 [1:18:15<01:45,  1.51it/s]
3|6312|Loss: 0.7230147123336792:  98%|█████████▊| 6311/6470 [1:18:16<01:45,  1.51it/s]
3|6312|Loss: 0.7230147123336792:  98%|█████████▊| 6312/6470 [1:18:16<01:45,  1.50it/s]
3|6313|Loss: 0.844065248966217:  98%|█████████▊| 6312/6470 [1:18:16<01:45,  1.50it/s] 
3|6313|Loss: 0.844065248966217:  98%|█████████▊| 6313/6470 [1:18:17<01:42,  1.53it/s]
3|6314|Loss: 0.7839136123657227:  98%|█████████▊| 6313/6470 [1:18:17<01:42,  1.53it/s]
3|6314|Loss: 0.7839136123657227:  98%|█████████▊| 6314/6470 [1:18:17<01:51,  1.40it/s]
3|6315|Loss: 0.8970894813537598:  98%|█████████▊| 6314/6470 [1:18:18<01:51,  1.40it/s]
3|6315|Loss: 0.8970894813537598:  98%|█████████▊| 6315/6470 [1:18:18<01:59,  1.29it/s]
3|6316|Loss: 0.6529139280319214:  98%|█████████▊| 6315/6470 [1:18:19<01:59,  1.29it/s]
3|6316|Loss: 0.6529139280319214:  98%|█████████▊| 6316/6470 [1:18:19<01:56,  1.32it/s]
3|6317|Loss: 0.776818573474884:  98%|█████████▊| 6316/6470 [1:18:19<01:56,  1.32it/s] 
3|6317|Loss: 0.776818573474884:  98%|█████████▊| 6317/6470 [1:18:20<01:51,  1.37it/s]
3|6318|Loss: 0.7571582794189453:  98%|█████████▊| 6317/6470 [1:18:20<01:51,  1.37it/s]
3|6318|Loss: 0.7571582794189453:  98%|█████████▊| 6318/6470 [1:18:21<01:52,  1.35it/s]
3|6319|Loss: 0.8783373832702637:  98%|█████████▊| 6318/6470 [1:18:21<01:52,  1.35it/s]
3|6319|Loss: 0.8783373832702637:  98%|█████████▊| 6319/6470 [1:18:21<01:54,  1.32it/s]
3|6320|Loss: 0.9465888738632202:  98%|█████████▊| 6319/6470 [1:18:22<01:54,  1.32it/s]
3|6320|Loss: 0.9465888738632202:  98%|█████████▊| 6320/6470 [1:18:22<01:52,  1.34it/s]
3|6321|Loss: 0.827709436416626:  98%|█████████▊| 6320/6470 [1:18:22<01:52,  1.34it/s] 
3|6321|Loss: 0.827709436416626:  98%|█████████▊| 6321/6470 [1:18:23<01:49,  1.37it/s]
3|6322|Loss: 0.9134779572486877:  98%|█████████▊| 6321/6470 [1:18:23<01:49,  1.37it/s]
3|6322|Loss: 0.9134779572486877:  98%|█████████▊| 6322/6470 [1:18:23<01:43,  1.44it/s]
3|6323|Loss: 1.051591157913208:  98%|█████████▊| 6322/6470 [1:18:24<01:43,  1.44it/s] 
3|6323|Loss: 1.051591157913208:  98%|█████████▊| 6323/6470 [1:18:24<01:40,  1.46it/s]
3|6324|Loss: 0.9845929741859436:  98%|█████████▊| 6323/6470 [1:18:24<01:40,  1.46it/s]
3|6324|Loss: 0.9845929741859436:  98%|█████████▊| 6324/6470 [1:18:25<01:39,  1.46it/s]
3|6325|Loss: 0.7225501537322998:  98%|█████████▊| 6324/6470 [1:18:25<01:39,  1.46it/s]
3|6325|Loss: 0.7225501537322998:  98%|█████████▊| 6325/6470 [1:18:26<01:47,  1.35it/s]
3|6326|Loss: 0.8041865229606628:  98%|█████████▊| 6325/6470 [1:18:26<01:47,  1.35it/s]
3|6326|Loss: 0.8041865229606628:  98%|█████████▊| 6326/6470 [1:18:26<01:42,  1.41it/s]
3|6327|Loss: 0.8129905462265015:  98%|█████████▊| 6326/6470 [1:18:27<01:42,  1.41it/s]
3|6327|Loss: 0.8129905462265015:  98%|█████████▊| 6327/6470 [1:18:27<01:41,  1.41it/s]
3|6328|Loss: 0.7811053395271301:  98%|█████████▊| 6327/6470 [1:18:27<01:41,  1.41it/s]
3|6328|Loss: 0.7811053395271301:  98%|█████████▊| 6328/6470 [1:18:28<01:48,  1.30it/s]
3|6329|Loss: 0.8777538537979126:  98%|█████████▊| 6328/6470 [1:18:28<01:48,  1.30it/s]
3|6329|Loss: 0.8777538537979126:  98%|█████████▊| 6329/6470 [1:18:29<01:55,  1.22it/s]
3|6330|Loss: 0.8047916293144226:  98%|█████████▊| 6329/6470 [1:18:29<01:55,  1.22it/s]
3|6330|Loss: 0.8047916293144226:  98%|█████████▊| 6330/6470 [1:18:30<01:55,  1.21it/s]
3|6331|Loss: 0.8042529225349426:  98%|█████████▊| 6330/6470 [1:18:30<01:55,  1.21it/s]
3|6331|Loss: 0.8042529225349426:  98%|█████████▊| 6331/6470 [1:18:30<01:46,  1.30it/s]
3|6332|Loss: 0.7076029777526855:  98%|█████████▊| 6331/6470 [1:18:31<01:46,  1.30it/s]
3|6332|Loss: 0.7076029777526855:  98%|█████████▊| 6332/6470 [1:18:31<01:42,  1.35it/s]
3|6333|Loss: 0.8077501058578491:  98%|█████████▊| 6332/6470 [1:18:31<01:42,  1.35it/s]
3|6333|Loss: 0.8077501058578491:  98%|█████████▊| 6333/6470 [1:18:32<01:41,  1.35it/s]
3|6334|Loss: 0.7150927186012268:  98%|█████████▊| 6333/6470 [1:18:32<01:41,  1.35it/s]
3|6334|Loss: 0.7150927186012268:  98%|█████████▊| 6334/6470 [1:18:32<01:39,  1.37it/s]
3|6335|Loss: 0.8597121238708496:  98%|█████████▊| 6334/6470 [1:18:33<01:39,  1.37it/s]
3|6335|Loss: 0.8597121238708496:  98%|█████████▊| 6335/6470 [1:18:33<01:34,  1.43it/s]
3|6336|Loss: 0.6491267681121826:  98%|█████████▊| 6335/6470 [1:18:33<01:34,  1.43it/s]
3|6336|Loss: 0.6491267681121826:  98%|█████████▊| 6336/6470 [1:18:34<01:38,  1.37it/s]
3|6337|Loss: 0.9938597679138184:  98%|█████████▊| 6336/6470 [1:18:34<01:38,  1.37it/s]
3|6337|Loss: 0.9938597679138184:  98%|█████████▊| 6337/6470 [1:18:35<01:42,  1.30it/s]
3|6338|Loss: 0.6975891590118408:  98%|█████████▊| 6337/6470 [1:18:35<01:42,  1.30it/s]
3|6338|Loss: 0.6975891590118408:  98%|█████████▊| 6338/6470 [1:18:36<01:48,  1.22it/s]
3|6339|Loss: 0.8678397536277771:  98%|█████████▊| 6338/6470 [1:18:36<01:48,  1.22it/s]
3|6339|Loss: 0.8678397536277771:  98%|█████████▊| 6339/6470 [1:18:36<01:45,  1.24it/s]
3|6340|Loss: 0.9093952178955078:  98%|█████████▊| 6339/6470 [1:18:37<01:45,  1.24it/s]
3|6340|Loss: 0.9093952178955078:  98%|█████████▊| 6340/6470 [1:18:37<01:35,  1.36it/s]
3|6341|Loss: 0.88566654920578:  98%|█████████▊| 6340/6470 [1:18:37<01:35,  1.36it/s]  
3|6341|Loss: 0.88566654920578:  98%|█████████▊| 6341/6470 [1:18:38<01:35,  1.35it/s]
3|6342|Loss: 0.9115022420883179:  98%|█████████▊| 6341/6470 [1:18:38<01:35,  1.35it/s]
3|6342|Loss: 0.9115022420883179:  98%|█████████▊| 6342/6470 [1:18:38<01:34,  1.36it/s]
3|6343|Loss: 0.9487019777297974:  98%|█████████▊| 6342/6470 [1:18:39<01:34,  1.36it/s]
3|6343|Loss: 0.9487019777297974:  98%|█████████▊| 6343/6470 [1:18:39<01:29,  1.42it/s]
3|6344|Loss: 0.8553563952445984:  98%|█████████▊| 6343/6470 [1:18:39<01:29,  1.42it/s]
3|6344|Loss: 0.8553563952445984:  98%|█████████▊| 6344/6470 [1:18:40<01:28,  1.43it/s]
3|6345|Loss: 0.7841922044754028:  98%|█████████▊| 6344/6470 [1:18:40<01:28,  1.43it/s]
3|6345|Loss: 0.7841922044754028:  98%|█████████▊| 6345/6470 [1:18:40<01:29,  1.40it/s]
3|6346|Loss: 0.7003523111343384:  98%|█████████▊| 6345/6470 [1:18:41<01:29,  1.40it/s]
3|6346|Loss: 0.7003523111343384:  98%|█████████▊| 6346/6470 [1:18:41<01:34,  1.31it/s]
3|6347|Loss: 0.8256577253341675:  98%|█████████▊| 6346/6470 [1:18:42<01:34,  1.31it/s]
3|6347|Loss: 0.8256577253341675:  98%|█████████▊| 6347/6470 [1:18:42<01:35,  1.29it/s]
3|6348|Loss: 0.9072362184524536:  98%|█████████▊| 6347/6470 [1:18:43<01:35,  1.29it/s]
3|6348|Loss: 0.9072362184524536:  98%|█████████▊| 6348/6470 [1:18:43<01:31,  1.33it/s]
3|6349|Loss: 0.7583504915237427:  98%|█████████▊| 6348/6470 [1:18:43<01:31,  1.33it/s]
3|6349|Loss: 0.7583504915237427:  98%|█████████▊| 6349/6470 [1:18:44<01:30,  1.34it/s]
3|6350|Loss: 0.8870083093643188:  98%|█████████▊| 6349/6470 [1:18:44<01:30,  1.34it/s]
3|6350|Loss: 0.8870083093643188:  98%|█████████▊| 6350/6470 [1:18:44<01:26,  1.39it/s]
3|6351|Loss: 0.8845965266227722:  98%|█████████▊| 6350/6470 [1:18:45<01:26,  1.39it/s]
3|6351|Loss: 0.8845965266227722:  98%|█████████▊| 6351/6470 [1:18:45<01:24,  1.41it/s]
3|6352|Loss: 0.834551215171814:  98%|█████████▊| 6351/6470 [1:18:45<01:24,  1.41it/s] 
3|6352|Loss: 0.834551215171814:  98%|█████████▊| 6352/6470 [1:18:46<01:22,  1.43it/s]
3|6353|Loss: 0.8981946706771851:  98%|█████████▊| 6352/6470 [1:18:46<01:22,  1.43it/s]
3|6353|Loss: 0.8981946706771851:  98%|█████████▊| 6353/6470 [1:18:46<01:24,  1.38it/s]
3|6354|Loss: 0.8454458117485046:  98%|█████████▊| 6353/6470 [1:18:47<01:24,  1.38it/s]
3|6354|Loss: 0.8454458117485046:  98%|█████████▊| 6354/6470 [1:18:47<01:25,  1.35it/s]
3|6355|Loss: 0.950816810131073:  98%|█████████▊| 6354/6470 [1:18:48<01:25,  1.35it/s] 
3|6355|Loss: 0.950816810131073:  98%|█████████▊| 6355/6470 [1:18:48<01:24,  1.37it/s]
3|6356|Loss: 0.6523802876472473:  98%|█████████▊| 6355/6470 [1:18:48<01:24,  1.37it/s]
3|6356|Loss: 0.6523802876472473:  98%|█████████▊| 6356/6470 [1:18:49<01:27,  1.30it/s]
3|6357|Loss: 0.7334956526756287:  98%|█████████▊| 6356/6470 [1:18:49<01:27,  1.30it/s]
3|6357|Loss: 0.7334956526756287:  98%|█████████▊| 6357/6470 [1:18:49<01:22,  1.37it/s]
3|6358|Loss: 0.764081597328186:  98%|█████████▊| 6357/6470 [1:18:50<01:22,  1.37it/s] 
3|6358|Loss: 0.764081597328186:  98%|█████████▊| 6358/6470 [1:18:50<01:18,  1.42it/s]
3|6359|Loss: 1.004915475845337:  98%|█████████▊| 6358/6470 [1:18:50<01:18,  1.42it/s]
3|6359|Loss: 1.004915475845337:  98%|█████████▊| 6359/6470 [1:18:51<01:20,  1.38it/s]
3|6360|Loss: 0.9195898175239563:  98%|█████████▊| 6359/6470 [1:18:51<01:20,  1.38it/s]
3|6360|Loss: 0.9195898175239563:  98%|█████████▊| 6360/6470 [1:18:51<01:16,  1.44it/s]
3|6361|Loss: 0.7442936897277832:  98%|█████████▊| 6360/6470 [1:18:52<01:16,  1.44it/s]
3|6361|Loss: 0.7442936897277832:  98%|█████████▊| 6361/6470 [1:18:52<01:10,  1.54it/s]
3|6362|Loss: 0.8997982740402222:  98%|█████████▊| 6361/6470 [1:18:52<01:10,  1.54it/s]
3|6362|Loss: 0.8997982740402222:  98%|█████████▊| 6362/6470 [1:18:52<01:06,  1.62it/s]
3|6363|Loss: 0.9903680682182312:  98%|█████████▊| 6362/6470 [1:18:53<01:06,  1.62it/s]
3|6363|Loss: 0.9903680682182312:  98%|█████████▊| 6363/6470 [1:18:53<01:05,  1.63it/s]
3|6364|Loss: 0.8306112885475159:  98%|█████████▊| 6363/6470 [1:18:54<01:05,  1.63it/s]
3|6364|Loss: 0.8306112885475159:  98%|█████████▊| 6364/6470 [1:18:54<01:12,  1.45it/s]
3|6365|Loss: 0.8510903120040894:  98%|█████████▊| 6364/6470 [1:18:54<01:12,  1.45it/s]
3|6365|Loss: 0.8510903120040894:  98%|█████████▊| 6365/6470 [1:18:55<01:10,  1.49it/s]
3|6366|Loss: 0.7854914665222168:  98%|█████████▊| 6365/6470 [1:18:55<01:10,  1.49it/s]
3|6366|Loss: 0.7854914665222168:  98%|█████████▊| 6366/6470 [1:18:55<01:16,  1.36it/s]
3|6367|Loss: 0.8327624201774597:  98%|█████████▊| 6366/6470 [1:18:56<01:16,  1.36it/s]
3|6367|Loss: 0.8327624201774597:  98%|█████████▊| 6367/6470 [1:18:56<01:16,  1.35it/s]
3|6368|Loss: 0.9206624627113342:  98%|█████████▊| 6367/6470 [1:18:57<01:16,  1.35it/s]
3|6368|Loss: 0.9206624627113342:  98%|█████████▊| 6368/6470 [1:18:57<01:17,  1.32it/s]
3|6369|Loss: 1.0256054401397705:  98%|█████████▊| 6368/6470 [1:18:57<01:17,  1.32it/s]
3|6369|Loss: 1.0256054401397705:  98%|█████████▊| 6369/6470 [1:18:58<01:17,  1.30it/s]
3|6370|Loss: 0.89131760597229:  98%|█████████▊| 6369/6470 [1:18:58<01:17,  1.30it/s]  
3|6370|Loss: 0.89131760597229:  98%|█████████▊| 6370/6470 [1:18:59<01:17,  1.29it/s]
3|6371|Loss: 0.9224257469177246:  98%|█████████▊| 6370/6470 [1:18:59<01:17,  1.29it/s]
3|6371|Loss: 0.9224257469177246:  98%|█████████▊| 6371/6470 [1:18:59<01:09,  1.42it/s]
3|6372|Loss: 0.8184877634048462:  98%|█████████▊| 6371/6470 [1:19:00<01:09,  1.42it/s]
3|6372|Loss: 0.8184877634048462:  98%|█████████▊| 6372/6470 [1:19:00<01:09,  1.40it/s]
3|6373|Loss: 0.7533649206161499:  98%|█████████▊| 6372/6470 [1:19:00<01:09,  1.40it/s]
3|6373|Loss: 0.7533649206161499:  99%|█████████▊| 6373/6470 [1:19:01<01:14,  1.30it/s]
3|6374|Loss: 0.8621412515640259:  99%|█████████▊| 6373/6470 [1:19:01<01:14,  1.30it/s]
3|6374|Loss: 0.8621412515640259:  99%|█████████▊| 6374/6470 [1:19:02<01:15,  1.27it/s]
3|6375|Loss: 0.8452997803688049:  99%|█████████▊| 6374/6470 [1:19:02<01:15,  1.27it/s]
3|6375|Loss: 0.8452997803688049:  99%|█████████▊| 6375/6470 [1:19:02<01:17,  1.22it/s]
3|6376|Loss: 0.7997004389762878:  99%|█████████▊| 6375/6470 [1:19:03<01:17,  1.22it/s]
3|6376|Loss: 0.7997004389762878:  99%|█████████▊| 6376/6470 [1:19:03<01:14,  1.27it/s]
3|6377|Loss: 0.8273801803588867:  99%|█████████▊| 6376/6470 [1:19:04<01:14,  1.27it/s]
3|6377|Loss: 0.8273801803588867:  99%|█████████▊| 6377/6470 [1:19:04<01:10,  1.31it/s]
3|6378|Loss: 0.8431352972984314:  99%|█████████▊| 6377/6470 [1:19:04<01:10,  1.31it/s]
3|6378|Loss: 0.8431352972984314:  99%|█████████▊| 6378/6470 [1:19:05<01:09,  1.32it/s]
3|6379|Loss: 0.765636146068573:  99%|█████████▊| 6378/6470 [1:19:05<01:09,  1.32it/s] 
3|6379|Loss: 0.765636146068573:  99%|█████████▊| 6379/6470 [1:19:05<01:10,  1.29it/s]
3|6380|Loss: 1.0271021127700806:  99%|█████████▊| 6379/6470 [1:19:06<01:10,  1.29it/s]
3|6380|Loss: 1.0271021127700806:  99%|█████████▊| 6380/6470 [1:19:06<01:08,  1.32it/s]
3|6381|Loss: 0.9150090217590332:  99%|█████████▊| 6380/6470 [1:19:07<01:08,  1.32it/s]
3|6381|Loss: 0.9150090217590332:  99%|█████████▊| 6381/6470 [1:19:07<01:07,  1.32it/s]
3|6382|Loss: 0.7146985530853271:  99%|█████████▊| 6381/6470 [1:19:07<01:07,  1.32it/s]
3|6382|Loss: 0.7146985530853271:  99%|█████████▊| 6382/6470 [1:19:08<01:07,  1.30it/s]
3|6383|Loss: 0.8497509956359863:  99%|█████████▊| 6382/6470 [1:19:08<01:07,  1.30it/s]
3|6383|Loss: 0.8497509956359863:  99%|█████████▊| 6383/6470 [1:19:08<01:00,  1.43it/s]
3|6384|Loss: 0.8081195950508118:  99%|█████████▊| 6383/6470 [1:19:09<01:00,  1.43it/s]
3|6384|Loss: 0.8081195950508118:  99%|█████████▊| 6384/6470 [1:19:09<01:03,  1.36it/s]
3|6385|Loss: 0.7836967706680298:  99%|█████████▊| 6384/6470 [1:19:10<01:03,  1.36it/s]
3|6385|Loss: 0.7836967706680298:  99%|█████████▊| 6385/6470 [1:19:10<01:06,  1.29it/s]
3|6386|Loss: 0.9836423993110657:  99%|█████████▊| 6385/6470 [1:19:10<01:06,  1.29it/s]
3|6386|Loss: 0.9836423993110657:  99%|█████████▊| 6386/6470 [1:19:11<01:03,  1.32it/s]
3|6387|Loss: 0.8935041427612305:  99%|█████████▊| 6386/6470 [1:19:11<01:03,  1.32it/s]
3|6387|Loss: 0.8935041427612305:  99%|█████████▊| 6387/6470 [1:19:11<01:00,  1.37it/s]
3|6388|Loss: 0.8706680536270142:  99%|█████████▊| 6387/6470 [1:19:12<01:00,  1.37it/s]
3|6388|Loss: 0.8706680536270142:  99%|█████████▊| 6388/6470 [1:19:12<01:03,  1.29it/s]
3|6389|Loss: 0.8083328604698181:  99%|█████████▊| 6388/6470 [1:19:13<01:03,  1.29it/s]
3|6389|Loss: 0.8083328604698181:  99%|█████████▊| 6389/6470 [1:19:13<01:05,  1.24it/s]
3|6390|Loss: 0.8111297488212585:  99%|█████████▊| 6389/6470 [1:19:14<01:05,  1.24it/s]
3|6390|Loss: 0.8111297488212585:  99%|█████████▉| 6390/6470 [1:19:14<01:01,  1.30it/s]
3|6391|Loss: 0.7882307767868042:  99%|█████████▉| 6390/6470 [1:19:14<01:01,  1.30it/s]
3|6391|Loss: 0.7882307767868042:  99%|█████████▉| 6391/6470 [1:19:15<01:02,  1.27it/s]
3|6392|Loss: 0.9015806913375854:  99%|█████████▉| 6391/6470 [1:19:15<01:02,  1.27it/s]
3|6392|Loss: 0.9015806913375854:  99%|█████████▉| 6392/6470 [1:19:16<01:04,  1.21it/s]
3|6393|Loss: 1.0140025615692139:  99%|█████████▉| 6392/6470 [1:19:16<01:04,  1.21it/s]
3|6393|Loss: 1.0140025615692139:  99%|█████████▉| 6393/6470 [1:19:16<01:05,  1.18it/s]
3|6394|Loss: 0.8249316215515137:  99%|█████████▉| 6393/6470 [1:19:17<01:05,  1.18it/s]
3|6394|Loss: 0.8249316215515137:  99%|█████████▉| 6394/6470 [1:19:17<01:03,  1.20it/s]
3|6395|Loss: 0.7221794724464417:  99%|█████████▉| 6394/6470 [1:19:18<01:03,  1.20it/s]
3|6395|Loss: 0.7221794724464417:  99%|█████████▉| 6395/6470 [1:19:18<00:58,  1.29it/s]
3|6396|Loss: 0.7886226177215576:  99%|█████████▉| 6395/6470 [1:19:18<00:58,  1.29it/s]
3|6396|Loss: 0.7886226177215576:  99%|█████████▉| 6396/6470 [1:19:19<00:56,  1.31it/s]
3|6397|Loss: 0.9159017205238342:  99%|█████████▉| 6396/6470 [1:19:19<00:56,  1.31it/s]
3|6397|Loss: 0.9159017205238342:  99%|█████████▉| 6397/6470 [1:19:19<00:52,  1.39it/s]
3|6398|Loss: 0.8045481443405151:  99%|█████████▉| 6397/6470 [1:19:20<00:52,  1.39it/s]
3|6398|Loss: 0.8045481443405151:  99%|█████████▉| 6398/6470 [1:19:20<00:52,  1.38it/s]
3|6399|Loss: 0.8427082300186157:  99%|█████████▉| 6398/6470 [1:19:20<00:52,  1.38it/s]
3|6399|Loss: 0.8427082300186157:  99%|█████████▉| 6399/6470 [1:19:21<00:50,  1.42it/s]
3|6400|Loss: 0.8234336972236633:  99%|█████████▉| 6399/6470 [1:19:21<00:50,  1.42it/s]
3|6400|Loss: 0.8234336972236633:  99%|█████████▉| 6400/6470 [1:19:21<00:46,  1.52it/s]
3|6401|Loss: 0.7749922871589661:  99%|█████████▉| 6400/6470 [1:19:21<00:46,  1.52it/s]
3|6401|Loss: 0.7749922871589661:  99%|█████████▉| 6401/6470 [1:19:22<00:40,  1.69it/s]
3|6402|Loss: 0.8701133131980896:  99%|█████████▉| 6401/6470 [1:19:22<00:40,  1.69it/s]
3|6402|Loss: 0.8701133131980896:  99%|█████████▉| 6402/6470 [1:19:22<00:45,  1.49it/s]
3|6403|Loss: 0.775407075881958:  99%|█████████▉| 6402/6470 [1:19:23<00:45,  1.49it/s] 
3|6403|Loss: 0.775407075881958:  99%|█████████▉| 6403/6470 [1:19:23<00:49,  1.36it/s]
3|6404|Loss: 0.9177274703979492:  99%|█████████▉| 6403/6470 [1:19:24<00:49,  1.36it/s]
3|6404|Loss: 0.9177274703979492:  99%|█████████▉| 6404/6470 [1:19:24<00:50,  1.32it/s]
3|6405|Loss: 0.6959466338157654:  99%|█████████▉| 6404/6470 [1:19:25<00:50,  1.32it/s]
3|6405|Loss: 0.6959466338157654:  99%|█████████▉| 6405/6470 [1:19:25<00:49,  1.30it/s]
3|6406|Loss: 0.7852081656455994:  99%|█████████▉| 6405/6470 [1:19:25<00:49,  1.30it/s]
3|6406|Loss: 0.7852081656455994:  99%|█████████▉| 6406/6470 [1:19:26<00:49,  1.30it/s]
3|6407|Loss: 0.9434548616409302:  99%|█████████▉| 6406/6470 [1:19:26<00:49,  1.30it/s]
3|6407|Loss: 0.9434548616409302:  99%|█████████▉| 6407/6470 [1:19:27<00:50,  1.24it/s]
3|6408|Loss: 0.8228106498718262:  99%|█████████▉| 6407/6470 [1:19:27<00:50,  1.24it/s]
3|6408|Loss: 0.8228106498718262:  99%|█████████▉| 6408/6470 [1:19:27<00:47,  1.32it/s]
3|6409|Loss: 1.0272588729858398:  99%|█████████▉| 6408/6470 [1:19:28<00:47,  1.32it/s]
3|6409|Loss: 1.0272588729858398:  99%|█████████▉| 6409/6470 [1:19:28<00:48,  1.26it/s]
3|6410|Loss: 0.919123113155365:  99%|█████████▉| 6409/6470 [1:19:29<00:48,  1.26it/s] 
3|6410|Loss: 0.919123113155365:  99%|█████████▉| 6410/6470 [1:19:29<00:48,  1.25it/s]
3|6411|Loss: 1.12442147731781:  99%|█████████▉| 6410/6470 [1:19:29<00:48,  1.25it/s] 
3|6411|Loss: 1.12442147731781:  99%|█████████▉| 6411/6470 [1:19:30<00:43,  1.34it/s]
3|6412|Loss: 0.8829026222229004:  99%|█████████▉| 6411/6470 [1:19:30<00:43,  1.34it/s]
3|6412|Loss: 0.8829026222229004:  99%|█████████▉| 6412/6470 [1:19:30<00:42,  1.35it/s]
3|6413|Loss: 0.7413283586502075:  99%|█████████▉| 6412/6470 [1:19:31<00:42,  1.35it/s]
3|6413|Loss: 0.7413283586502075:  99%|█████████▉| 6413/6470 [1:19:31<00:43,  1.32it/s]
3|6414|Loss: 0.9340435862541199:  99%|█████████▉| 6413/6470 [1:19:32<00:43,  1.32it/s]
3|6414|Loss: 0.9340435862541199:  99%|█████████▉| 6414/6470 [1:19:32<00:44,  1.26it/s]
3|6415|Loss: 0.9377898573875427:  99%|█████████▉| 6414/6470 [1:19:32<00:44,  1.26it/s]
3|6415|Loss: 0.9377898573875427:  99%|█████████▉| 6415/6470 [1:19:33<00:44,  1.23it/s]
3|6416|Loss: 0.8458808064460754:  99%|█████████▉| 6415/6470 [1:19:33<00:44,  1.23it/s]
3|6416|Loss: 0.8458808064460754:  99%|█████████▉| 6416/6470 [1:19:34<00:43,  1.24it/s]
3|6417|Loss: 0.8830068707466125:  99%|█████████▉| 6416/6470 [1:19:34<00:43,  1.24it/s]
3|6417|Loss: 0.8830068707466125:  99%|█████████▉| 6417/6470 [1:19:34<00:42,  1.25it/s]
3|6418|Loss: 0.7404317855834961:  99%|█████████▉| 6417/6470 [1:19:35<00:42,  1.25it/s]
3|6418|Loss: 0.7404317855834961:  99%|█████████▉| 6418/6470 [1:19:35<00:38,  1.34it/s]
3|6419|Loss: 0.8390496969223022:  99%|█████████▉| 6418/6470 [1:19:35<00:38,  1.34it/s]
3|6419|Loss: 0.8390496969223022:  99%|█████████▉| 6419/6470 [1:19:36<00:36,  1.38it/s]
3|6420|Loss: 0.8354586362838745:  99%|█████████▉| 6419/6470 [1:19:36<00:36,  1.38it/s]
3|6420|Loss: 0.8354586362838745:  99%|█████████▉| 6420/6470 [1:19:36<00:35,  1.39it/s]
3|6421|Loss: 0.9075716137886047:  99%|█████████▉| 6420/6470 [1:19:37<00:35,  1.39it/s]
3|6421|Loss: 0.9075716137886047:  99%|█████████▉| 6421/6470 [1:19:37<00:34,  1.40it/s]
3|6422|Loss: 0.8259652853012085:  99%|█████████▉| 6421/6470 [1:19:38<00:34,  1.40it/s]
3|6422|Loss: 0.8259652853012085:  99%|█████████▉| 6422/6470 [1:19:38<00:35,  1.36it/s]
3|6423|Loss: 0.758466362953186:  99%|█████████▉| 6422/6470 [1:19:38<00:35,  1.36it/s] 
3|6423|Loss: 0.758466362953186:  99%|█████████▉| 6423/6470 [1:19:39<00:34,  1.37it/s]
3|6424|Loss: 0.9744893908500671:  99%|█████████▉| 6423/6470 [1:19:39<00:34,  1.37it/s]
3|6424|Loss: 0.9744893908500671:  99%|█████████▉| 6424/6470 [1:19:39<00:32,  1.41it/s]
3|6425|Loss: 0.6891975402832031:  99%|█████████▉| 6424/6470 [1:19:40<00:32,  1.41it/s]
3|6425|Loss: 0.6891975402832031:  99%|█████████▉| 6425/6470 [1:19:40<00:31,  1.43it/s]
3|6426|Loss: 0.894138753414154:  99%|█████████▉| 6425/6470 [1:19:40<00:31,  1.43it/s] 
3|6426|Loss: 0.894138753414154:  99%|█████████▉| 6426/6470 [1:19:41<00:30,  1.43it/s]
3|6427|Loss: 0.6994127631187439:  99%|█████████▉| 6426/6470 [1:19:41<00:30,  1.43it/s]
3|6427|Loss: 0.6994127631187439:  99%|█████████▉| 6427/6470 [1:19:42<00:32,  1.32it/s]
3|6428|Loss: 0.8797146081924438:  99%|█████████▉| 6427/6470 [1:19:42<00:32,  1.32it/s]
3|6428|Loss: 0.8797146081924438:  99%|█████████▉| 6428/6470 [1:19:42<00:30,  1.36it/s]
3|6429|Loss: 0.8686522245407104:  99%|█████████▉| 6428/6470 [1:19:43<00:30,  1.36it/s]
3|6429|Loss: 0.8686522245407104:  99%|█████████▉| 6429/6470 [1:19:43<00:28,  1.43it/s]
3|6430|Loss: 0.9020623564720154:  99%|█████████▉| 6429/6470 [1:19:43<00:28,  1.43it/s]
3|6430|Loss: 0.9020623564720154:  99%|█████████▉| 6430/6470 [1:19:44<00:29,  1.36it/s]
3|6431|Loss: 0.7922778725624084:  99%|█████████▉| 6430/6470 [1:19:44<00:29,  1.36it/s]
3|6431|Loss: 0.7922778725624084:  99%|█████████▉| 6431/6470 [1:19:44<00:26,  1.46it/s]
3|6432|Loss: 0.6705042719841003:  99%|█████████▉| 6431/6470 [1:19:45<00:26,  1.46it/s]
3|6432|Loss: 0.6705042719841003:  99%|█████████▉| 6432/6470 [1:19:45<00:23,  1.61it/s]
3|6433|Loss: 0.6825463175773621:  99%|█████████▉| 6432/6470 [1:19:45<00:23,  1.61it/s]
3|6433|Loss: 0.6825463175773621:  99%|█████████▉| 6433/6470 [1:19:45<00:23,  1.60it/s]
3|6434|Loss: 0.8696817755699158:  99%|█████████▉| 6433/6470 [1:19:46<00:23,  1.60it/s]
3|6434|Loss: 0.8696817755699158:  99%|█████████▉| 6434/6470 [1:19:46<00:23,  1.55it/s]
3|6435|Loss: 0.8068485856056213:  99%|█████████▉| 6434/6470 [1:19:46<00:23,  1.55it/s]
3|6435|Loss: 0.8068485856056213:  99%|█████████▉| 6435/6470 [1:19:47<00:24,  1.41it/s]
3|6436|Loss: 0.7875931859016418:  99%|█████████▉| 6435/6470 [1:19:47<00:24,  1.41it/s]
3|6436|Loss: 0.7875931859016418:  99%|█████████▉| 6436/6470 [1:19:47<00:21,  1.57it/s]
3|6437|Loss: 0.80635005235672:  99%|█████████▉| 6436/6470 [1:19:48<00:21,  1.57it/s]  
3|6437|Loss: 0.80635005235672:  99%|█████████▉| 6437/6470 [1:19:48<00:21,  1.57it/s]
3|6438|Loss: 0.7777807712554932:  99%|█████████▉| 6437/6470 [1:19:48<00:21,  1.57it/s]
3|6438|Loss: 0.7777807712554932: 100%|█████████▉| 6438/6470 [1:19:49<00:21,  1.51it/s]
3|6439|Loss: 0.9034572243690491: 100%|█████████▉| 6438/6470 [1:19:49<00:21,  1.51it/s]
3|6439|Loss: 0.9034572243690491: 100%|█████████▉| 6439/6470 [1:19:49<00:21,  1.44it/s]
3|6440|Loss: 0.8414586186408997: 100%|█████████▉| 6439/6470 [1:19:50<00:21,  1.44it/s]
3|6440|Loss: 0.8414586186408997: 100%|█████████▉| 6440/6470 [1:19:50<00:20,  1.44it/s]
3|6441|Loss: 0.8423575162887573: 100%|█████████▉| 6440/6470 [1:19:51<00:20,  1.44it/s]
3|6441|Loss: 0.8423575162887573: 100%|█████████▉| 6441/6470 [1:19:51<00:21,  1.37it/s]
3|6442|Loss: 0.8253255486488342: 100%|█████████▉| 6441/6470 [1:19:51<00:21,  1.37it/s]
3|6442|Loss: 0.8253255486488342: 100%|█████████▉| 6442/6470 [1:19:52<00:21,  1.32it/s]
3|6443|Loss: 0.9141513705253601: 100%|█████████▉| 6442/6470 [1:19:52<00:21,  1.32it/s]
3|6443|Loss: 0.9141513705253601: 100%|█████████▉| 6443/6470 [1:19:53<00:20,  1.35it/s]
3|6444|Loss: 0.8443665504455566: 100%|█████████▉| 6443/6470 [1:19:53<00:20,  1.35it/s]
3|6444|Loss: 0.8443665504455566: 100%|█████████▉| 6444/6470 [1:19:53<00:20,  1.27it/s]
3|6445|Loss: 0.8561177253723145: 100%|█████████▉| 6444/6470 [1:19:54<00:20,  1.27it/s]
3|6445|Loss: 0.8561177253723145: 100%|█████████▉| 6445/6470 [1:19:54<00:20,  1.22it/s]
3|6446|Loss: 0.8637142181396484: 100%|█████████▉| 6445/6470 [1:19:55<00:20,  1.22it/s]
3|6446|Loss: 0.8637142181396484: 100%|█████████▉| 6446/6470 [1:19:55<00:19,  1.22it/s]
3|6447|Loss: 0.8765503764152527: 100%|█████████▉| 6446/6470 [1:19:56<00:19,  1.22it/s]
3|6447|Loss: 0.8765503764152527: 100%|█████████▉| 6447/6470 [1:19:56<00:19,  1.17it/s]
3|6448|Loss: 0.8375958800315857: 100%|█████████▉| 6447/6470 [1:19:56<00:19,  1.17it/s]
3|6448|Loss: 0.8375958800315857: 100%|█████████▉| 6448/6470 [1:19:57<00:17,  1.27it/s]
3|6449|Loss: 0.8865152597427368: 100%|█████████▉| 6448/6470 [1:19:57<00:17,  1.27it/s]
3|6449|Loss: 0.8865152597427368: 100%|█████████▉| 6449/6470 [1:19:58<00:16,  1.25it/s]
3|6450|Loss: 0.9255943298339844: 100%|█████████▉| 6449/6470 [1:19:58<00:16,  1.25it/s]
3|6450|Loss: 0.9255943298339844: 100%|█████████▉| 6450/6470 [1:19:58<00:15,  1.28it/s]
3|6451|Loss: 0.8038715124130249: 100%|█████████▉| 6450/6470 [1:19:59<00:15,  1.28it/s]
3|6451|Loss: 0.8038715124130249: 100%|█████████▉| 6451/6470 [1:19:59<00:14,  1.32it/s]
3|6452|Loss: 0.8473122715950012: 100%|█████████▉| 6451/6470 [1:19:59<00:14,  1.32it/s]
3|6452|Loss: 0.8473122715950012: 100%|█████████▉| 6452/6470 [1:20:00<00:14,  1.25it/s]
3|6453|Loss: 0.8635451197624207: 100%|█████████▉| 6452/6470 [1:20:00<00:14,  1.25it/s]
3|6453|Loss: 0.8635451197624207: 100%|█████████▉| 6453/6470 [1:20:01<00:14,  1.18it/s]
3|6454|Loss: 0.8852259516716003: 100%|█████████▉| 6453/6470 [1:20:01<00:14,  1.18it/s]
3|6454|Loss: 0.8852259516716003: 100%|█████████▉| 6454/6470 [1:20:02<00:13,  1.22it/s]
3|6455|Loss: 0.8435682654380798: 100%|█████████▉| 6454/6470 [1:20:02<00:13,  1.22it/s]
3|6455|Loss: 0.8435682654380798: 100%|█████████▉| 6455/6470 [1:20:02<00:12,  1.24it/s]
3|6456|Loss: 0.8171507120132446: 100%|█████████▉| 6455/6470 [1:20:03<00:12,  1.24it/s]
3|6456|Loss: 0.8171507120132446: 100%|█████████▉| 6456/6470 [1:20:03<00:11,  1.25it/s]
3|6457|Loss: 1.0226408243179321: 100%|█████████▉| 6456/6470 [1:20:04<00:11,  1.25it/s]
3|6457|Loss: 1.0226408243179321: 100%|█████████▉| 6457/6470 [1:20:04<00:09,  1.33it/s]
3|6458|Loss: 0.7621062397956848: 100%|█████████▉| 6457/6470 [1:20:04<00:09,  1.33it/s]
3|6458|Loss: 0.7621062397956848: 100%|█████████▉| 6458/6470 [1:20:05<00:09,  1.26it/s]
3|6459|Loss: 0.8602311015129089: 100%|█████████▉| 6458/6470 [1:20:05<00:09,  1.26it/s]
3|6459|Loss: 0.8602311015129089: 100%|█████████▉| 6459/6470 [1:20:05<00:08,  1.29it/s]
3|6460|Loss: 0.7559624910354614: 100%|█████████▉| 6459/6470 [1:20:06<00:08,  1.29it/s]
3|6460|Loss: 0.7559624910354614: 100%|█████████▉| 6460/6470 [1:20:06<00:07,  1.35it/s]
3|6461|Loss: 0.8586469292640686: 100%|█████████▉| 6460/6470 [1:20:06<00:07,  1.35it/s]
3|6461|Loss: 0.8586469292640686: 100%|█████████▉| 6461/6470 [1:20:07<00:06,  1.39it/s]
3|6462|Loss: 0.8396601676940918: 100%|█████████▉| 6461/6470 [1:20:07<00:06,  1.39it/s]
3|6462|Loss: 0.8396601676940918: 100%|█████████▉| 6462/6470 [1:20:07<00:05,  1.45it/s]
3|6463|Loss: 0.7992597222328186: 100%|█████████▉| 6462/6470 [1:20:08<00:05,  1.45it/s]
3|6463|Loss: 0.7992597222328186: 100%|█████████▉| 6463/6470 [1:20:08<00:04,  1.47it/s]
3|6464|Loss: 0.6403238773345947: 100%|█████████▉| 6463/6470 [1:20:08<00:04,  1.47it/s]
3|6464|Loss: 0.6403238773345947: 100%|█████████▉| 6464/6470 [1:20:09<00:04,  1.43it/s]
3|6465|Loss: 0.7655317783355713: 100%|█████████▉| 6464/6470 [1:20:09<00:04,  1.43it/s]
3|6465|Loss: 0.7655317783355713: 100%|█████████▉| 6465/6470 [1:20:10<00:03,  1.38it/s]
3|6466|Loss: 0.9168104529380798: 100%|█████████▉| 6465/6470 [1:20:10<00:03,  1.38it/s]
3|6466|Loss: 0.9168104529380798: 100%|█████████▉| 6466/6470 [1:20:10<00:02,  1.44it/s]
3|6467|Loss: 0.8049440979957581: 100%|█████████▉| 6466/6470 [1:20:11<00:02,  1.44it/s]
3|6467|Loss: 0.8049440979957581: 100%|█████████▉| 6467/6470 [1:20:11<00:02,  1.35it/s]
3|6468|Loss: 0.9802713990211487: 100%|█████████▉| 6467/6470 [1:20:11<00:02,  1.35it/s]
3|6468|Loss: 0.9802713990211487: 100%|█████████▉| 6468/6470 [1:20:12<00:01,  1.35it/s]
3|6469|Loss: 0.8404450416564941: 100%|█████████▉| 6468/6470 [1:20:12<00:01,  1.35it/s]
3|6469|Loss: 0.8404450416564941: 100%|█████████▉| 6469/6470 [1:20:13<00:00,  1.29it/s]
3|6470|Loss: 0.9612289071083069: 100%|█████████▉| 6469/6470 [1:20:13<00:00,  1.29it/s]
3|6470|Loss: 0.9612289071083069: 100%|██████████| 6470/6470 [1:20:14<00:00,  1.23it/s]
3|6470|Loss: 0.9612289071083069: 100%|██████████| 6470/6470 [1:20:14<00:00,  1.34it/s]
Model checkpoint of size 25961 MB saved to output/alpaca-colorful-llama2-finetune/model_2_19410.ckpt
Skipping uploading to HuggingFace Hub (no repo id specified)

  0%|          | 0/6470 [00:00<?, ?it/s]
4|1|Loss: 0.8087262511253357:   0%|          | 0/6470 [00:00<?, ?it/s]
4|1|Loss: 0.8087262511253357:   0%|          | 1/6470 [00:00<1:03:42,  1.69it/s]
4|2|Loss: 0.8620941042900085:   0%|          | 1/6470 [00:00<1:03:42,  1.69it/s]
4|2|Loss: 0.8620941042900085:   0%|          | 2/6470 [00:01<1:08:26,  1.58it/s]
4|3|Loss: 0.8538207411766052:   0%|          | 2/6470 [00:01<1:08:26,  1.58it/s]
4|3|Loss: 0.8538207411766052:   0%|          | 3/6470 [00:02<1:18:05,  1.38it/s]
4|4|Loss: 0.8656133413314819:   0%|          | 3/6470 [00:02<1:18:05,  1.38it/s]
4|4|Loss: 0.8656133413314819:   0%|          | 4/6470 [00:02<1:17:20,  1.39it/s]
4|5|Loss: 0.961466372013092:   0%|          | 4/6470 [00:03<1:17:20,  1.39it/s] 
4|5|Loss: 0.961466372013092:   0%|          | 5/6470 [00:03<1:18:34,  1.37it/s]
4|6|Loss: 0.9079666137695312:   0%|          | 5/6470 [00:04<1:18:34,  1.37it/s]
4|6|Loss: 0.9079666137695312:   0%|          | 6/6470 [00:04<1:22:42,  1.30it/s]
4|7|Loss: 0.8525126576423645:   0%|          | 6/6470 [00:04<1:22:42,  1.30it/s]
4|7|Loss: 0.8525126576423645:   0%|          | 7/6470 [00:05<1:25:02,  1.27it/s]
4|8|Loss: 0.8466148972511292:   0%|          | 7/6470 [00:05<1:25:02,  1.27it/s]
4|8|Loss: 0.8466148972511292:   0%|          | 8/6470 [00:06<1:27:09,  1.24it/s]
4|9|Loss: 0.9288161993026733:   0%|          | 8/6470 [00:06<1:27:09,  1.24it/s]
4|9|Loss: 0.9288161993026733:   0%|          | 9/6470 [00:06<1:27:58,  1.22it/s]
4|10|Loss: 0.7630316615104675:   0%|          | 9/6470 [00:07<1:27:58,  1.22it/s]
4|10|Loss: 0.7630316615104675:   0%|          | 10/6470 [00:07<1:28:13,  1.22it/s]
4|11|Loss: 0.8423150181770325:   0%|          | 10/6470 [00:08<1:28:13,  1.22it/s]
4|11|Loss: 0.8423150181770325:   0%|          | 11/6470 [00:08<1:25:55,  1.25it/s]
4|12|Loss: 0.965417742729187:   0%|          | 11/6470 [00:08<1:25:55,  1.25it/s] 
4|12|Loss: 0.965417742729187:   0%|          | 12/6470 [00:09<1:21:12,  1.33it/s]
4|13|Loss: 0.835565447807312:   0%|          | 12/6470 [00:09<1:21:12,  1.33it/s]
4|13|Loss: 0.835565447807312:   0%|          | 13/6470 [00:09<1:22:09,  1.31it/s]
4|14|Loss: 0.812498152256012:   0%|          | 13/6470 [00:10<1:22:09,  1.31it/s]
4|14|Loss: 0.812498152256012:   0%|          | 14/6470 [00:10<1:20:26,  1.34it/s]
4|15|Loss: 0.8504820466041565:   0%|          | 14/6470 [00:11<1:20:26,  1.34it/s]
4|15|Loss: 0.8504820466041565:   0%|          | 15/6470 [00:11<1:18:15,  1.37it/s]
4|16|Loss: 0.8686360716819763:   0%|          | 15/6470 [00:11<1:18:15,  1.37it/s]
4|16|Loss: 0.8686360716819763:   0%|          | 16/6470 [00:11<1:13:59,  1.45it/s]
4|17|Loss: 0.7763632535934448:   0%|          | 16/6470 [00:12<1:13:59,  1.45it/s]
4|17|Loss: 0.7763632535934448:   0%|          | 17/6470 [00:12<1:18:27,  1.37it/s]
4|18|Loss: 0.803090512752533:   0%|          | 17/6470 [00:13<1:18:27,  1.37it/s] 
4|18|Loss: 0.803090512752533:   0%|          | 18/6470 [00:13<1:20:16,  1.34it/s]
4|19|Loss: 0.9076728820800781:   0%|          | 18/6470 [00:13<1:20:16,  1.34it/s]
4|19|Loss: 0.9076728820800781:   0%|          | 19/6470 [00:14<1:19:53,  1.35it/s]
4|20|Loss: 0.7895148396492004:   0%|          | 19/6470 [00:14<1:19:53,  1.35it/s]
4|20|Loss: 0.7895148396492004:   0%|          | 20/6470 [00:14<1:13:15,  1.47it/s]
4|21|Loss: 0.7713487148284912:   0%|          | 20/6470 [00:15<1:13:15,  1.47it/s]
4|21|Loss: 0.7713487148284912:   0%|          | 21/6470 [00:15<1:09:09,  1.55it/s]
4|22|Loss: 0.6981667876243591:   0%|          | 21/6470 [00:15<1:09:09,  1.55it/s]
4|22|Loss: 0.6981667876243591:   0%|          | 22/6470 [00:15<1:06:12,  1.62it/s]
4|23|Loss: 0.9296265244483948:   0%|          | 22/6470 [00:16<1:06:12,  1.62it/s]
4|23|Loss: 0.9296265244483948:   0%|          | 23/6470 [00:16<1:08:41,  1.56it/s]
4|24|Loss: 0.9016733169555664:   0%|          | 23/6470 [00:16<1:08:41,  1.56it/s]
4|24|Loss: 0.9016733169555664:   0%|          | 24/6470 [00:17<1:07:41,  1.59it/s]
4|25|Loss: 0.7834835052490234:   0%|          | 24/6470 [00:17<1:07:41,  1.59it/s]
4|25|Loss: 0.7834835052490234:   0%|          | 25/6470 [00:17<1:07:23,  1.59it/s]
4|26|Loss: 0.8887400031089783:   0%|          | 25/6470 [00:18<1:07:23,  1.59it/s]
4|26|Loss: 0.8887400031089783:   0%|          | 26/6470 [00:18<1:12:45,  1.48it/s]
4|27|Loss: 0.7361526489257812:   0%|          | 26/6470 [00:19<1:12:45,  1.48it/s]
4|27|Loss: 0.7361526489257812:   0%|          | 27/6470 [00:19<1:11:55,  1.49it/s]
4|28|Loss: 0.9691658616065979:   0%|          | 27/6470 [00:19<1:11:55,  1.49it/s]
4|28|Loss: 0.9691658616065979:   0%|          | 28/6470 [00:20<1:17:21,  1.39it/s]
4|29|Loss: 0.792140007019043:   0%|          | 28/6470 [00:20<1:17:21,  1.39it/s] 
4|29|Loss: 0.792140007019043:   0%|          | 29/6470 [00:20<1:13:37,  1.46it/s]
4|30|Loss: 0.8586403131484985:   0%|          | 29/6470 [00:21<1:13:37,  1.46it/s]
4|30|Loss: 0.8586403131484985:   0%|          | 30/6470 [00:21<1:16:13,  1.41it/s]
4|31|Loss: 0.8322011232376099:   0%|          | 30/6470 [00:21<1:16:13,  1.41it/s]
4|31|Loss: 0.8322011232376099:   0%|          | 31/6470 [00:22<1:17:39,  1.38it/s]
4|32|Loss: 0.8696681261062622:   0%|          | 31/6470 [00:22<1:17:39,  1.38it/s]
4|32|Loss: 0.8696681261062622:   0%|          | 32/6470 [00:22<1:12:30,  1.48it/s]
4|33|Loss: 0.9498785734176636:   0%|          | 32/6470 [00:23<1:12:30,  1.48it/s]
4|33|Loss: 0.9498785734176636:   1%|          | 33/6470 [00:23<1:13:37,  1.46it/s]
4|34|Loss: 0.9019660353660583:   1%|          | 33/6470 [00:23<1:13:37,  1.46it/s]
4|34|Loss: 0.9019660353660583:   1%|          | 34/6470 [00:24<1:08:28,  1.57it/s]
4|35|Loss: 0.741453230381012:   1%|          | 34/6470 [00:24<1:08:28,  1.57it/s] 
4|35|Loss: 0.741453230381012:   1%|          | 35/6470 [00:24<1:04:51,  1.65it/s]
4|36|Loss: 0.744497537612915:   1%|          | 35/6470 [00:24<1:04:51,  1.65it/s]
4|36|Loss: 0.744497537612915:   1%|          | 36/6470 [00:25<1:02:41,  1.71it/s]
4|37|Loss: 0.8232962489128113:   1%|          | 36/6470 [00:25<1:02:41,  1.71it/s]
4|37|Loss: 0.8232962489128113:   1%|          | 37/6470 [00:25<1:06:50,  1.60it/s]
4|38|Loss: 0.6909404397010803:   1%|          | 37/6470 [00:26<1:06:50,  1.60it/s]
4|38|Loss: 0.6909404397010803:   1%|          | 38/6470 [00:26<1:06:40,  1.61it/s]
4|39|Loss: 0.8392218947410583:   1%|          | 38/6470 [00:26<1:06:40,  1.61it/s]
4|39|Loss: 0.8392218947410583:   1%|          | 39/6470 [00:27<1:05:43,  1.63it/s]
4|40|Loss: 0.8246480822563171:   1%|          | 39/6470 [00:27<1:05:43,  1.63it/s]
4|40|Loss: 0.8246480822563171:   1%|          | 40/6470 [00:27<1:12:04,  1.49it/s]
4|41|Loss: 0.9413834810256958:   1%|          | 40/6470 [00:28<1:12:04,  1.49it/s]
4|41|Loss: 0.9413834810256958:   1%|          | 41/6470 [00:28<1:15:49,  1.41it/s]
4|42|Loss: 0.9059838056564331:   1%|          | 41/6470 [00:29<1:15:49,  1.41it/s]
4|42|Loss: 0.9059838056564331:   1%|          | 42/6470 [00:29<1:16:54,  1.39it/s]
4|43|Loss: 0.8264539241790771:   1%|          | 42/6470 [00:29<1:16:54,  1.39it/s]
4|43|Loss: 0.8264539241790771:   1%|          | 43/6470 [00:29<1:11:18,  1.50it/s]
4|44|Loss: 0.7767269611358643:   1%|          | 43/6470 [00:30<1:11:18,  1.50it/s]
4|44|Loss: 0.7767269611358643:   1%|          | 44/6470 [00:30<1:08:31,  1.56it/s]
4|45|Loss: 0.8442281484603882:   1%|          | 44/6470 [00:30<1:08:31,  1.56it/s]
4|45|Loss: 0.8442281484603882:   1%|          | 45/6470 [00:31<1:08:27,  1.56it/s]
4|46|Loss: 0.8875051140785217:   1%|          | 45/6470 [00:31<1:08:27,  1.56it/s]
4|46|Loss: 0.8875051140785217:   1%|          | 46/6470 [00:31<1:10:53,  1.51it/s]
4|47|Loss: 0.7669388651847839:   1%|          | 46/6470 [00:32<1:10:53,  1.51it/s]
4|47|Loss: 0.7669388651847839:   1%|          | 47/6470 [00:32<1:12:50,  1.47it/s]
4|48|Loss: 0.7897695899009705:   1%|          | 47/6470 [00:32<1:12:50,  1.47it/s]
4|48|Loss: 0.7897695899009705:   1%|          | 48/6470 [00:33<1:17:23,  1.38it/s]
4|49|Loss: 0.8318302035331726:   1%|          | 48/6470 [00:33<1:17:23,  1.38it/s]
4|49|Loss: 0.8318302035331726:   1%|          | 49/6470 [00:34<1:21:12,  1.32it/s]
4|50|Loss: 0.8599721193313599:   1%|          | 49/6470 [00:34<1:21:12,  1.32it/s]
4|50|Loss: 0.8599721193313599:   1%|          | 50/6470 [00:34<1:19:54,  1.34it/s]
4|51|Loss: 0.9740381240844727:   1%|          | 50/6470 [00:35<1:19:54,  1.34it/s]
4|51|Loss: 0.9740381240844727:   1%|          | 51/6470 [00:35<1:20:36,  1.33it/s]
4|52|Loss: 0.7341579794883728:   1%|          | 51/6470 [00:36<1:20:36,  1.33it/s]
4|52|Loss: 0.7341579794883728:   1%|          | 52/6470 [00:36<1:25:25,  1.25it/s]
4|53|Loss: 0.8834982514381409:   1%|          | 52/6470 [00:37<1:25:25,  1.25it/s]
4|53|Loss: 0.8834982514381409:   1%|          | 53/6470 [00:37<1:28:22,  1.21it/s]
4|54|Loss: 0.8695076107978821:   1%|          | 53/6470 [00:38<1:28:22,  1.21it/s]
4|54|Loss: 0.8695076107978821:   1%|          | 54/6470 [00:38<1:30:52,  1.18it/s]
4|55|Loss: 0.842322826385498:   1%|          | 54/6470 [00:38<1:30:52,  1.18it/s] 
4|55|Loss: 0.842322826385498:   1%|          | 55/6470 [00:39<1:30:48,  1.18it/s]
4|56|Loss: 0.785210371017456:   1%|          | 55/6470 [00:39<1:30:48,  1.18it/s]
4|56|Loss: 0.785210371017456:   1%|          | 56/6470 [00:39<1:26:32,  1.24it/s]
4|57|Loss: 0.8923448920249939:   1%|          | 56/6470 [00:40<1:26:32,  1.24it/s]
4|57|Loss: 0.8923448920249939:   1%|          | 57/6470 [00:40<1:24:03,  1.27it/s]
4|58|Loss: 0.8444942831993103:   1%|          | 57/6470 [00:41<1:24:03,  1.27it/s]
4|58|Loss: 0.8444942831993103:   1%|          | 58/6470 [00:41<1:26:46,  1.23it/s]
4|59|Loss: 0.8731803297996521:   1%|          | 58/6470 [00:42<1:26:46,  1.23it/s]
4|59|Loss: 0.8731803297996521:   1%|          | 59/6470 [00:42<1:27:13,  1.23it/s]
4|60|Loss: 0.9657650589942932:   1%|          | 59/6470 [00:42<1:27:13,  1.23it/s]
4|60|Loss: 0.9657650589942932:   1%|          | 60/6470 [00:43<1:28:57,  1.20it/s]
4|61|Loss: 0.8966002464294434:   1%|          | 60/6470 [00:43<1:28:57,  1.20it/s]
4|61|Loss: 0.8966002464294434:   1%|          | 61/6470 [00:43<1:24:44,  1.26it/s]
4|62|Loss: 0.8497069478034973:   1%|          | 61/6470 [00:44<1:24:44,  1.26it/s]
4|62|Loss: 0.8497069478034973:   1%|          | 62/6470 [00:44<1:23:18,  1.28it/s]
4|63|Loss: 0.7922409772872925:   1%|          | 62/6470 [00:45<1:23:18,  1.28it/s]
4|63|Loss: 0.7922409772872925:   1%|          | 63/6470 [00:45<1:22:42,  1.29it/s]
4|64|Loss: 0.8295044302940369:   1%|          | 63/6470 [00:45<1:22:42,  1.29it/s]
4|64|Loss: 0.8295044302940369:   1%|          | 64/6470 [00:46<1:26:38,  1.23it/s]
4|65|Loss: 1.11533522605896:   1%|          | 64/6470 [00:46<1:26:38,  1.23it/s]  
4|65|Loss: 1.11533522605896:   1%|          | 65/6470 [00:46<1:18:54,  1.35it/s]
4|66|Loss: 0.9204558730125427:   1%|          | 65/6470 [00:47<1:18:54,  1.35it/s]
4|66|Loss: 0.9204558730125427:   1%|          | 66/6470 [00:47<1:17:37,  1.38it/s]
4|67|Loss: 0.8165478110313416:   1%|          | 66/6470 [00:48<1:17:37,  1.38it/s]
4|67|Loss: 0.8165478110313416:   1%|          | 67/6470 [00:48<1:18:25,  1.36it/s]
4|68|Loss: 0.6178979277610779:   1%|          | 67/6470 [00:48<1:18:25,  1.36it/s]
4|68|Loss: 0.6178979277610779:   1%|          | 68/6470 [00:49<1:16:11,  1.40it/s]
4|69|Loss: 0.8566948175430298:   1%|          | 68/6470 [00:49<1:16:11,  1.40it/s]
4|69|Loss: 0.8566948175430298:   1%|          | 69/6470 [00:49<1:17:32,  1.38it/s]
4|70|Loss: 0.7313188314437866:   1%|          | 69/6470 [00:50<1:17:32,  1.38it/s]
4|70|Loss: 0.7313188314437866:   1%|          | 70/6470 [00:50<1:13:46,  1.45it/s]
4|71|Loss: 0.8045871257781982:   1%|          | 70/6470 [00:50<1:13:46,  1.45it/s]
4|71|Loss: 0.8045871257781982:   1%|          | 71/6470 [00:51<1:18:32,  1.36it/s]
4|72|Loss: 0.7978875637054443:   1%|          | 71/6470 [00:51<1:18:32,  1.36it/s]
4|72|Loss: 0.7978875637054443:   1%|          | 72/6470 [00:52<1:18:42,  1.35it/s]
4|73|Loss: 0.7474265098571777:   1%|          | 72/6470 [00:52<1:18:42,  1.35it/s]
4|73|Loss: 0.7474265098571777:   1%|          | 73/6470 [00:52<1:23:37,  1.28it/s]
4|74|Loss: 0.9731360673904419:   1%|          | 73/6470 [00:53<1:23:37,  1.28it/s]
4|74|Loss: 0.9731360673904419:   1%|          | 74/6470 [00:53<1:24:01,  1.27it/s]
4|75|Loss: 0.7920554876327515:   1%|          | 74/6470 [00:54<1:24:01,  1.27it/s]
4|75|Loss: 0.7920554876327515:   1%|          | 75/6470 [00:54<1:19:36,  1.34it/s]
4|76|Loss: 0.8043888807296753:   1%|          | 75/6470 [00:54<1:19:36,  1.34it/s]
4|76|Loss: 0.8043888807296753:   1%|          | 76/6470 [00:55<1:18:11,  1.36it/s]
4|77|Loss: 0.8050891757011414:   1%|          | 76/6470 [00:55<1:18:11,  1.36it/s]
4|77|Loss: 0.8050891757011414:   1%|          | 77/6470 [00:56<1:25:59,  1.24it/s]
4|78|Loss: 0.9564774632453918:   1%|          | 77/6470 [00:56<1:25:59,  1.24it/s]
4|78|Loss: 0.9564774632453918:   1%|          | 78/6470 [00:56<1:25:06,  1.25it/s]
4|79|Loss: 0.7099647521972656:   1%|          | 78/6470 [00:57<1:25:06,  1.25it/s]
4|79|Loss: 0.7099647521972656:   1%|          | 79/6470 [00:57<1:21:41,  1.30it/s]
4|80|Loss: 0.7672188878059387:   1%|          | 79/6470 [00:57<1:21:41,  1.30it/s]
4|80|Loss: 0.7672188878059387:   1%|          | 80/6470 [00:58<1:16:11,  1.40it/s]
4|81|Loss: 0.7898602485656738:   1%|          | 80/6470 [00:58<1:16:11,  1.40it/s]
4|81|Loss: 0.7898602485656738:   1%|▏         | 81/6470 [00:58<1:09:02,  1.54it/s]
4|82|Loss: 0.943069338798523:   1%|▏         | 81/6470 [00:58<1:09:02,  1.54it/s] 
4|82|Loss: 0.943069338798523:   1%|▏         | 82/6470 [00:59<1:11:06,  1.50it/s]
4|83|Loss: 0.7223981618881226:   1%|▏         | 82/6470 [00:59<1:11:06,  1.50it/s]
4|83|Loss: 0.7223981618881226:   1%|▏         | 83/6470 [01:00<1:14:02,  1.44it/s]
4|84|Loss: 0.979361891746521:   1%|▏         | 83/6470 [01:00<1:14:02,  1.44it/s] 
4|84|Loss: 0.979361891746521:   1%|▏         | 84/6470 [01:00<1:14:12,  1.43it/s]
4|85|Loss: 0.7540538907051086:   1%|▏         | 84/6470 [01:01<1:14:12,  1.43it/s]
4|85|Loss: 0.7540538907051086:   1%|▏         | 85/6470 [01:01<1:13:02,  1.46it/s]
4|86|Loss: 0.9138777256011963:   1%|▏         | 85/6470 [01:01<1:13:02,  1.46it/s]
4|86|Loss: 0.9138777256011963:   1%|▏         | 86/6470 [01:02<1:11:53,  1.48it/s]
4|87|Loss: 0.7792785167694092:   1%|▏         | 86/6470 [01:02<1:11:53,  1.48it/s]
4|87|Loss: 0.7792785167694092:   1%|▏         | 87/6470 [01:02<1:13:44,  1.44it/s]
4|88|Loss: 0.8445520997047424:   1%|▏         | 87/6470 [01:03<1:13:44,  1.44it/s]
4|88|Loss: 0.8445520997047424:   1%|▏         | 88/6470 [01:03<1:15:30,  1.41it/s]
4|89|Loss: 0.6582477688789368:   1%|▏         | 88/6470 [01:03<1:15:30,  1.41it/s]
4|89|Loss: 0.6582477688789368:   1%|▏         | 89/6470 [01:04<1:10:15,  1.51it/s]
4|90|Loss: 0.8501895666122437:   1%|▏         | 89/6470 [01:04<1:10:15,  1.51it/s]
4|90|Loss: 0.8501895666122437:   1%|▏         | 90/6470 [01:04<1:12:39,  1.46it/s]
4|91|Loss: 0.8106422424316406:   1%|▏         | 90/6470 [01:05<1:12:39,  1.46it/s]
4|91|Loss: 0.8106422424316406:   1%|▏         | 91/6470 [01:05<1:05:37,  1.62it/s]
4|92|Loss: 0.8716586232185364:   1%|▏         | 91/6470 [01:05<1:05:37,  1.62it/s]
4|92|Loss: 0.8716586232185364:   1%|▏         | 92/6470 [01:05<1:07:15,  1.58it/s]
4|93|Loss: 0.6067103147506714:   1%|▏         | 92/6470 [01:06<1:07:15,  1.58it/s]
4|93|Loss: 0.6067103147506714:   1%|▏         | 93/6470 [01:06<1:10:39,  1.50it/s]
4|94|Loss: 0.8495889902114868:   1%|▏         | 93/6470 [01:07<1:10:39,  1.50it/s]
4|94|Loss: 0.8495889902114868:   1%|▏         | 94/6470 [01:07<1:13:54,  1.44it/s]
4|95|Loss: 0.9252462983131409:   1%|▏         | 94/6470 [01:07<1:13:54,  1.44it/s]
4|95|Loss: 0.9252462983131409:   1%|▏         | 95/6470 [01:08<1:17:33,  1.37it/s]
4|96|Loss: 0.8539279103279114:   1%|▏         | 95/6470 [01:08<1:17:33,  1.37it/s]
4|96|Loss: 0.8539279103279114:   1%|▏         | 96/6470 [01:09<1:18:33,  1.35it/s]
4|97|Loss: 0.890770435333252:   1%|▏         | 96/6470 [01:09<1:18:33,  1.35it/s] 
4|97|Loss: 0.890770435333252:   1%|▏         | 97/6470 [01:09<1:18:34,  1.35it/s]
4|98|Loss: 0.7864750027656555:   1%|▏         | 97/6470 [01:10<1:18:34,  1.35it/s]
4|98|Loss: 0.7864750027656555:   2%|▏         | 98/6470 [01:10<1:17:28,  1.37it/s]
4|99|Loss: 0.8924956917762756:   2%|▏         | 98/6470 [01:10<1:17:28,  1.37it/s]
4|99|Loss: 0.8924956917762756:   2%|▏         | 99/6470 [01:10<1:08:13,  1.56it/s]
4|100|Loss: 0.7351369857788086:   2%|▏         | 99/6470 [01:11<1:08:13,  1.56it/s]
4|100|Loss: 0.7351369857788086:   2%|▏         | 100/6470 [01:11<1:05:09,  1.63it/s]
4|101|Loss: 0.7626209259033203:   2%|▏         | 100/6470 [01:11<1:05:09,  1.63it/s]
4|101|Loss: 0.7626209259033203:   2%|▏         | 101/6470 [01:12<1:01:19,  1.73it/s]
4|102|Loss: 0.8748432397842407:   2%|▏         | 101/6470 [01:12<1:01:19,  1.73it/s]
4|102|Loss: 0.8748432397842407:   2%|▏         | 102/6470 [01:12<1:05:35,  1.62it/s]
4|103|Loss: 0.7408618927001953:   2%|▏         | 102/6470 [01:13<1:05:35,  1.62it/s]
4|103|Loss: 0.7408618927001953:   2%|▏         | 103/6470 [01:13<1:07:32,  1.57it/s]
4|104|Loss: 0.8511071801185608:   2%|▏         | 103/6470 [01:13<1:07:32,  1.57it/s]
4|104|Loss: 0.8511071801185608:   2%|▏         | 104/6470 [01:14<1:14:50,  1.42it/s]
4|105|Loss: 0.82851642370224:   2%|▏         | 104/6470 [01:14<1:14:50,  1.42it/s]  
4|105|Loss: 0.82851642370224:   2%|▏         | 105/6470 [01:14<1:12:41,  1.46it/s]
4|106|Loss: 0.6987683773040771:   2%|▏         | 105/6470 [01:15<1:12:41,  1.46it/s]
4|106|Loss: 0.6987683773040771:   2%|▏         | 106/6470 [01:15<1:11:52,  1.48it/s]
4|107|Loss: 0.7700509428977966:   2%|▏         | 106/6470 [01:15<1:11:52,  1.48it/s]
4|107|Loss: 0.7700509428977966:   2%|▏         | 107/6470 [01:16<1:09:02,  1.54it/s]
4|108|Loss: 0.7305814027786255:   2%|▏         | 107/6470 [01:16<1:09:02,  1.54it/s]
4|108|Loss: 0.7305814027786255:   2%|▏         | 108/6470 [01:16<1:14:11,  1.43it/s]
4|109|Loss: 0.6940514445304871:   2%|▏         | 108/6470 [01:17<1:14:11,  1.43it/s]
4|109|Loss: 0.6940514445304871:   2%|▏         | 109/6470 [01:17<1:09:21,  1.53it/s]
4|110|Loss: 0.837999165058136:   2%|▏         | 109/6470 [01:17<1:09:21,  1.53it/s] 
4|110|Loss: 0.837999165058136:   2%|▏         | 110/6470 [01:17<1:03:15,  1.68it/s]
4|111|Loss: 0.7205130457878113:   2%|▏         | 110/6470 [01:18<1:03:15,  1.68it/s]
4|111|Loss: 0.7205130457878113:   2%|▏         | 111/6470 [01:18<1:07:30,  1.57it/s]
4|112|Loss: 0.957163393497467:   2%|▏         | 111/6470 [01:19<1:07:30,  1.57it/s] 
4|112|Loss: 0.957163393497467:   2%|▏         | 112/6470 [01:19<1:11:42,  1.48it/s]
4|113|Loss: 0.8297178149223328:   2%|▏         | 112/6470 [01:19<1:11:42,  1.48it/s]
4|113|Loss: 0.8297178149223328:   2%|▏         | 113/6470 [01:20<1:12:16,  1.47it/s]
4|114|Loss: 0.8760853409767151:   2%|▏         | 113/6470 [01:20<1:12:16,  1.47it/s]
4|114|Loss: 0.8760853409767151:   2%|▏         | 114/6470 [01:20<1:15:33,  1.40it/s]
4|115|Loss: 0.7296317219734192:   2%|▏         | 114/6470 [01:21<1:15:33,  1.40it/s]
4|115|Loss: 0.7296317219734192:   2%|▏         | 115/6470 [01:21<1:10:25,  1.50it/s]
4|116|Loss: 0.8785421252250671:   2%|▏         | 115/6470 [01:21<1:10:25,  1.50it/s]
4|116|Loss: 0.8785421252250671:   2%|▏         | 116/6470 [01:22<1:08:01,  1.56it/s]
4|117|Loss: 0.7704213857650757:   2%|▏         | 116/6470 [01:22<1:08:01,  1.56it/s]
4|117|Loss: 0.7704213857650757:   2%|▏         | 117/6470 [01:22<1:06:59,  1.58it/s]
4|118|Loss: 0.9204672574996948:   2%|▏         | 117/6470 [01:23<1:06:59,  1.58it/s]
4|118|Loss: 0.9204672574996948:   2%|▏         | 118/6470 [01:23<1:08:10,  1.55it/s]
4|119|Loss: 0.888751745223999:   2%|▏         | 118/6470 [01:23<1:08:10,  1.55it/s] 
4|119|Loss: 0.888751745223999:   2%|▏         | 119/6470 [01:24<1:09:13,  1.53it/s]
4|120|Loss: 0.9136799573898315:   2%|▏         | 119/6470 [01:24<1:09:13,  1.53it/s]
4|120|Loss: 0.9136799573898315:   2%|▏         | 120/6470 [01:24<1:16:30,  1.38it/s]
4|121|Loss: 0.9120027422904968:   2%|▏         | 120/6470 [01:25<1:16:30,  1.38it/s]
4|121|Loss: 0.9120027422904968:   2%|▏         | 121/6470 [01:25<1:22:37,  1.28it/s]
4|122|Loss: 0.8343302011489868:   2%|▏         | 121/6470 [01:26<1:22:37,  1.28it/s]
4|122|Loss: 0.8343302011489868:   2%|▏         | 122/6470 [01:26<1:23:48,  1.26it/s]
4|123|Loss: 0.6540606617927551:   2%|▏         | 122/6470 [01:27<1:23:48,  1.26it/s]
4|123|Loss: 0.6540606617927551:   2%|▏         | 123/6470 [01:27<1:30:02,  1.17it/s]
4|124|Loss: 0.8970103859901428:   2%|▏         | 123/6470 [01:28<1:30:02,  1.17it/s]
4|124|Loss: 0.8970103859901428:   2%|▏         | 124/6470 [01:28<1:31:21,  1.16it/s]
4|125|Loss: 0.8196171522140503:   2%|▏         | 124/6470 [01:28<1:31:21,  1.16it/s]
4|125|Loss: 0.8196171522140503:   2%|▏         | 125/6470 [01:29<1:29:00,  1.19it/s]
4|126|Loss: 0.7259272336959839:   2%|▏         | 125/6470 [01:29<1:29:00,  1.19it/s]
4|126|Loss: 0.7259272336959839:   2%|▏         | 126/6470 [01:30<1:26:22,  1.22it/s]
4|127|Loss: 0.8551535606384277:   2%|▏         | 126/6470 [01:30<1:26:22,  1.22it/s]
4|127|Loss: 0.8551535606384277:   2%|▏         | 127/6470 [01:30<1:17:01,  1.37it/s]
4|128|Loss: 0.8610000610351562:   2%|▏         | 127/6470 [01:30<1:17:01,  1.37it/s]
4|128|Loss: 0.8610000610351562:   2%|▏         | 128/6470 [01:31<1:15:54,  1.39it/s]
4|129|Loss: 0.8634751439094543:   2%|▏         | 128/6470 [01:31<1:15:54,  1.39it/s]
4|129|Loss: 0.8634751439094543:   2%|▏         | 129/6470 [01:32<1:18:39,  1.34it/s]
4|130|Loss: 0.7511457204818726:   2%|▏         | 129/6470 [01:32<1:18:39,  1.34it/s]
4|130|Loss: 0.7511457204818726:   2%|▏         | 130/6470 [01:32<1:18:03,  1.35it/s]
4|131|Loss: 0.7613016366958618:   2%|▏         | 130/6470 [01:33<1:18:03,  1.35it/s]
4|131|Loss: 0.7613016366958618:   2%|▏         | 131/6470 [01:33<1:14:24,  1.42it/s]
4|132|Loss: 0.6671693325042725:   2%|▏         | 131/6470 [01:33<1:14:24,  1.42it/s]
4|132|Loss: 0.6671693325042725:   2%|▏         | 132/6470 [01:34<1:16:07,  1.39it/s]
4|133|Loss: 0.7442666292190552:   2%|▏         | 132/6470 [01:34<1:16:07,  1.39it/s]
4|133|Loss: 0.7442666292190552:   2%|▏         | 133/6470 [01:34<1:15:45,  1.39it/s]
4|134|Loss: 0.8929460644721985:   2%|▏         | 133/6470 [01:35<1:15:45,  1.39it/s]
4|134|Loss: 0.8929460644721985:   2%|▏         | 134/6470 [01:35<1:19:42,  1.32it/s]
4|135|Loss: 0.6802249550819397:   2%|▏         | 134/6470 [01:36<1:19:42,  1.32it/s]
4|135|Loss: 0.6802249550819397:   2%|▏         | 135/6470 [01:36<1:26:24,  1.22it/s]
4|136|Loss: 0.9009426236152649:   2%|▏         | 135/6470 [01:37<1:26:24,  1.22it/s]
4|136|Loss: 0.9009426236152649:   2%|▏         | 136/6470 [01:37<1:27:11,  1.21it/s]
4|137|Loss: 0.8468499779701233:   2%|▏         | 136/6470 [01:38<1:27:11,  1.21it/s]
4|137|Loss: 0.8468499779701233:   2%|▏         | 137/6470 [01:38<1:25:05,  1.24it/s]
4|138|Loss: 0.8238184452056885:   2%|▏         | 137/6470 [01:38<1:25:05,  1.24it/s]
4|138|Loss: 0.8238184452056885:   2%|▏         | 138/6470 [01:39<1:22:54,  1.27it/s]
4|139|Loss: 0.7897396087646484:   2%|▏         | 138/6470 [01:39<1:22:54,  1.27it/s]
4|139|Loss: 0.7897396087646484:   2%|▏         | 139/6470 [01:39<1:18:46,  1.34it/s]
4|140|Loss: 0.7079214453697205:   2%|▏         | 139/6470 [01:40<1:18:46,  1.34it/s]
4|140|Loss: 0.7079214453697205:   2%|▏         | 140/6470 [01:40<1:19:04,  1.33it/s]
4|141|Loss: 0.8011488318443298:   2%|▏         | 140/6470 [01:40<1:19:04,  1.33it/s]
4|141|Loss: 0.8011488318443298:   2%|▏         | 141/6470 [01:41<1:17:54,  1.35it/s]
4|142|Loss: 0.6802473664283752:   2%|▏         | 141/6470 [01:41<1:17:54,  1.35it/s]
4|142|Loss: 0.6802473664283752:   2%|▏         | 142/6470 [01:41<1:18:38,  1.34it/s]
4|143|Loss: 0.8294500112533569:   2%|▏         | 142/6470 [01:42<1:18:38,  1.34it/s]
4|143|Loss: 0.8294500112533569:   2%|▏         | 143/6470 [01:42<1:14:40,  1.41it/s]
4|144|Loss: 0.8624415993690491:   2%|▏         | 143/6470 [01:42<1:14:40,  1.41it/s]
4|144|Loss: 0.8624415993690491:   2%|▏         | 144/6470 [01:43<1:11:15,  1.48it/s]
4|145|Loss: 0.9510034918785095:   2%|▏         | 144/6470 [01:43<1:11:15,  1.48it/s]
4|145|Loss: 0.9510034918785095:   2%|▏         | 145/6470 [01:43<1:09:23,  1.52it/s]
4|146|Loss: 0.9656737446784973:   2%|▏         | 145/6470 [01:44<1:09:23,  1.52it/s]
4|146|Loss: 0.9656737446784973:   2%|▏         | 146/6470 [01:44<1:16:07,  1.38it/s]
4|147|Loss: 0.9013365507125854:   2%|▏         | 146/6470 [01:45<1:16:07,  1.38it/s]
4|147|Loss: 0.9013365507125854:   2%|▏         | 147/6470 [01:45<1:14:56,  1.41it/s]
4|148|Loss: 0.9371615648269653:   2%|▏         | 147/6470 [01:45<1:14:56,  1.41it/s]
4|148|Loss: 0.9371615648269653:   2%|▏         | 148/6470 [01:46<1:16:16,  1.38it/s]
4|149|Loss: 0.9081888794898987:   2%|▏         | 148/6470 [01:46<1:16:16,  1.38it/s]
4|149|Loss: 0.9081888794898987:   2%|▏         | 149/6470 [01:46<1:14:12,  1.42it/s]
4|150|Loss: 0.8588220477104187:   2%|▏         | 149/6470 [01:47<1:14:12,  1.42it/s]
4|150|Loss: 0.8588220477104187:   2%|▏         | 150/6470 [01:47<1:19:12,  1.33it/s]
4|151|Loss: 0.8410405516624451:   2%|▏         | 150/6470 [01:48<1:19:12,  1.33it/s]
4|151|Loss: 0.8410405516624451:   2%|▏         | 151/6470 [01:48<1:24:09,  1.25it/s]
4|152|Loss: 0.9221881031990051:   2%|▏         | 151/6470 [01:49<1:24:09,  1.25it/s]
4|152|Loss: 0.9221881031990051:   2%|▏         | 152/6470 [01:49<1:26:00,  1.22it/s]
4|153|Loss: 0.7811717391014099:   2%|▏         | 152/6470 [01:49<1:26:00,  1.22it/s]
4|153|Loss: 0.7811717391014099:   2%|▏         | 153/6470 [01:50<1:23:29,  1.26it/s]
4|154|Loss: 0.8232690095901489:   2%|▏         | 153/6470 [01:50<1:23:29,  1.26it/s]
4|154|Loss: 0.8232690095901489:   2%|▏         | 154/6470 [01:51<1:26:35,  1.22it/s]
4|155|Loss: 0.6020895838737488:   2%|▏         | 154/6470 [01:51<1:26:35,  1.22it/s]
4|155|Loss: 0.6020895838737488:   2%|▏         | 155/6470 [01:51<1:23:38,  1.26it/s]
4|156|Loss: 0.6642759442329407:   2%|▏         | 155/6470 [01:52<1:23:38,  1.26it/s]
4|156|Loss: 0.6642759442329407:   2%|▏         | 156/6470 [01:52<1:14:34,  1.41it/s]
4|157|Loss: 0.8098611831665039:   2%|▏         | 156/6470 [01:52<1:14:34,  1.41it/s]
4|157|Loss: 0.8098611831665039:   2%|▏         | 157/6470 [01:53<1:15:32,  1.39it/s]
4|158|Loss: 0.8783390522003174:   2%|▏         | 157/6470 [01:53<1:15:32,  1.39it/s]
4|158|Loss: 0.8783390522003174:   2%|▏         | 158/6470 [01:53<1:16:08,  1.38it/s]
4|159|Loss: 0.7726364731788635:   2%|▏         | 158/6470 [01:54<1:16:08,  1.38it/s]
4|159|Loss: 0.7726364731788635:   2%|▏         | 159/6470 [01:54<1:10:36,  1.49it/s]
4|160|Loss: 0.7941699028015137:   2%|▏         | 159/6470 [01:54<1:10:36,  1.49it/s]
4|160|Loss: 0.7941699028015137:   2%|▏         | 160/6470 [01:54<1:11:11,  1.48it/s]
4|161|Loss: 0.7265580296516418:   2%|▏         | 160/6470 [01:55<1:11:11,  1.48it/s]
4|161|Loss: 0.7265580296516418:   2%|▏         | 161/6470 [01:55<1:12:02,  1.46it/s]
4|162|Loss: 0.7786083221435547:   2%|▏         | 161/6470 [01:56<1:12:02,  1.46it/s]
4|162|Loss: 0.7786083221435547:   3%|▎         | 162/6470 [01:56<1:12:20,  1.45it/s]
4|163|Loss: 0.9958779215812683:   3%|▎         | 162/6470 [01:56<1:12:20,  1.45it/s]
4|163|Loss: 0.9958779215812683:   3%|▎         | 163/6470 [01:57<1:18:18,  1.34it/s]
4|164|Loss: 1.0880049467086792:   3%|▎         | 163/6470 [01:57<1:18:18,  1.34it/s]
4|164|Loss: 1.0880049467086792:   3%|▎         | 164/6470 [01:58<1:19:37,  1.32it/s]
4|165|Loss: 0.8924892544746399:   3%|▎         | 164/6470 [01:58<1:19:37,  1.32it/s]
4|165|Loss: 0.8924892544746399:   3%|▎         | 165/6470 [01:58<1:15:02,  1.40it/s]
4|166|Loss: 0.8562932014465332:   3%|▎         | 165/6470 [01:59<1:15:02,  1.40it/s]
4|166|Loss: 0.8562932014465332:   3%|▎         | 166/6470 [01:59<1:11:22,  1.47it/s]
4|167|Loss: 0.7980186343193054:   3%|▎         | 166/6470 [01:59<1:11:22,  1.47it/s]
4|167|Loss: 0.7980186343193054:   3%|▎         | 167/6470 [01:59<1:09:51,  1.50it/s]
4|168|Loss: 0.6364170908927917:   3%|▎         | 167/6470 [02:00<1:09:51,  1.50it/s]
4|168|Loss: 0.6364170908927917:   3%|▎         | 168/6470 [02:00<1:07:16,  1.56it/s]
4|169|Loss: 0.7815092206001282:   3%|▎         | 168/6470 [02:00<1:07:16,  1.56it/s]
4|169|Loss: 0.7815092206001282:   3%|▎         | 169/6470 [02:01<1:03:39,  1.65it/s]
4|170|Loss: 0.8200049996376038:   3%|▎         | 169/6470 [02:01<1:03:39,  1.65it/s]
4|170|Loss: 0.8200049996376038:   3%|▎         | 170/6470 [02:01<1:09:10,  1.52it/s]
4|171|Loss: 0.6939482688903809:   3%|▎         | 170/6470 [02:02<1:09:10,  1.52it/s]
4|171|Loss: 0.6939482688903809:   3%|▎         | 171/6470 [02:02<1:12:21,  1.45it/s]
4|172|Loss: 0.9267147183418274:   3%|▎         | 171/6470 [02:02<1:12:21,  1.45it/s]
4|172|Loss: 0.9267147183418274:   3%|▎         | 172/6470 [02:03<1:14:57,  1.40it/s]
4|173|Loss: 0.7804828882217407:   3%|▎         | 172/6470 [02:03<1:14:57,  1.40it/s]
4|173|Loss: 0.7804828882217407:   3%|▎         | 173/6470 [02:04<1:17:34,  1.35it/s]
4|174|Loss: 0.971136212348938:   3%|▎         | 173/6470 [02:04<1:17:34,  1.35it/s] 
4|174|Loss: 0.971136212348938:   3%|▎         | 174/6470 [02:04<1:16:19,  1.37it/s]
4|175|Loss: 0.7575564384460449:   3%|▎         | 174/6470 [02:05<1:16:19,  1.37it/s]
4|175|Loss: 0.7575564384460449:   3%|▎         | 175/6470 [02:05<1:14:17,  1.41it/s]
4|176|Loss: 0.8982201814651489:   3%|▎         | 175/6470 [02:05<1:14:17,  1.41it/s]
4|176|Loss: 0.8982201814651489:   3%|▎         | 176/6470 [02:06<1:15:55,  1.38it/s]
4|177|Loss: 0.8147109150886536:   3%|▎         | 176/6470 [02:06<1:15:55,  1.38it/s]
4|177|Loss: 0.8147109150886536:   3%|▎         | 177/6470 [02:06<1:14:41,  1.40it/s]
4|178|Loss: 0.7314189076423645:   3%|▎         | 177/6470 [02:07<1:14:41,  1.40it/s]
4|178|Loss: 0.7314189076423645:   3%|▎         | 178/6470 [02:07<1:19:51,  1.31it/s]
4|179|Loss: 0.7409878373146057:   3%|▎         | 178/6470 [02:08<1:19:51,  1.31it/s]
4|179|Loss: 0.7409878373146057:   3%|▎         | 179/6470 [02:08<1:18:09,  1.34it/s]
4|180|Loss: 0.8904088139533997:   3%|▎         | 179/6470 [02:08<1:18:09,  1.34it/s]
4|180|Loss: 0.8904088139533997:   3%|▎         | 180/6470 [02:09<1:15:10,  1.39it/s]
4|181|Loss: 1.0226951837539673:   3%|▎         | 180/6470 [02:09<1:15:10,  1.39it/s]
4|181|Loss: 1.0226951837539673:   3%|▎         | 181/6470 [02:09<1:10:46,  1.48it/s]
4|182|Loss: 0.7804369926452637:   3%|▎         | 181/6470 [02:10<1:10:46,  1.48it/s]
4|182|Loss: 0.7804369926452637:   3%|▎         | 182/6470 [02:10<1:09:49,  1.50it/s]
4|183|Loss: 0.8109270334243774:   3%|▎         | 182/6470 [02:10<1:09:49,  1.50it/s]
4|183|Loss: 0.8109270334243774:   3%|▎         | 183/6470 [02:10<1:06:00,  1.59it/s]
4|184|Loss: 0.8984643816947937:   3%|▎         | 183/6470 [02:11<1:06:00,  1.59it/s]
4|184|Loss: 0.8984643816947937:   3%|▎         | 184/6470 [02:11<1:06:42,  1.57it/s]
4|185|Loss: 0.8558416366577148:   3%|▎         | 184/6470 [02:12<1:06:42,  1.57it/s]
4|185|Loss: 0.8558416366577148:   3%|▎         | 185/6470 [02:12<1:10:51,  1.48it/s]
4|186|Loss: 0.6626747250556946:   3%|▎         | 185/6470 [02:12<1:10:51,  1.48it/s]
4|186|Loss: 0.6626747250556946:   3%|▎         | 186/6470 [02:13<1:14:09,  1.41it/s]
4|187|Loss: 0.7431519627571106:   3%|▎         | 186/6470 [02:13<1:14:09,  1.41it/s]
4|187|Loss: 0.7431519627571106:   3%|▎         | 187/6470 [02:13<1:13:13,  1.43it/s]
4|188|Loss: 0.8037452697753906:   3%|▎         | 187/6470 [02:14<1:13:13,  1.43it/s]
4|188|Loss: 0.8037452697753906:   3%|▎         | 188/6470 [02:14<1:12:41,  1.44it/s]
4|189|Loss: 0.8264983892440796:   3%|▎         | 188/6470 [02:14<1:12:41,  1.44it/s]
4|189|Loss: 0.8264983892440796:   3%|▎         | 189/6470 [02:15<1:07:10,  1.56it/s]
4|190|Loss: 0.8546608686447144:   3%|▎         | 189/6470 [02:15<1:07:10,  1.56it/s]
4|190|Loss: 0.8546608686447144:   3%|▎         | 190/6470 [02:15<1:03:49,  1.64it/s]
4|191|Loss: 0.8597177267074585:   3%|▎         | 190/6470 [02:15<1:03:49,  1.64it/s]
4|191|Loss: 0.8597177267074585:   3%|▎         | 191/6470 [02:16<1:07:05,  1.56it/s]
4|192|Loss: 0.7329462170600891:   3%|▎         | 191/6470 [02:16<1:07:05,  1.56it/s]
4|192|Loss: 0.7329462170600891:   3%|▎         | 192/6470 [02:16<1:04:36,  1.62it/s]
4|193|Loss: 0.6810375452041626:   3%|▎         | 192/6470 [02:17<1:04:36,  1.62it/s]
4|193|Loss: 0.6810375452041626:   3%|▎         | 193/6470 [02:17<1:06:05,  1.58it/s]
4|194|Loss: 0.7635172009468079:   3%|▎         | 193/6470 [02:17<1:06:05,  1.58it/s]
4|194|Loss: 0.7635172009468079:   3%|▎         | 194/6470 [02:18<1:12:09,  1.45it/s]
4|195|Loss: 0.8468844890594482:   3%|▎         | 194/6470 [02:18<1:12:09,  1.45it/s]
4|195|Loss: 0.8468844890594482:   3%|▎         | 195/6470 [02:18<1:05:52,  1.59it/s]
4|196|Loss: 0.9837267398834229:   3%|▎         | 195/6470 [02:19<1:05:52,  1.59it/s]
4|196|Loss: 0.9837267398834229:   3%|▎         | 196/6470 [02:19<1:00:54,  1.72it/s]
4|197|Loss: 0.8719424605369568:   3%|▎         | 196/6470 [02:19<1:00:54,  1.72it/s]
4|197|Loss: 0.8719424605369568:   3%|▎         | 197/6470 [02:19<1:04:18,  1.63it/s]
4|198|Loss: 0.7700315117835999:   3%|▎         | 197/6470 [02:20<1:04:18,  1.63it/s]
4|198|Loss: 0.7700315117835999:   3%|▎         | 198/6470 [02:20<1:11:35,  1.46it/s]
4|199|Loss: 0.9189761877059937:   3%|▎         | 198/6470 [02:21<1:11:35,  1.46it/s]
4|199|Loss: 0.9189761877059937:   3%|▎         | 199/6470 [02:21<1:18:56,  1.32it/s]
4|200|Loss: 0.7581546902656555:   3%|▎         | 199/6470 [02:22<1:18:56,  1.32it/s]
4|200|Loss: 0.7581546902656555:   3%|▎         | 200/6470 [02:22<1:18:05,  1.34it/s]
4|201|Loss: 0.7818050980567932:   3%|▎         | 200/6470 [02:22<1:18:05,  1.34it/s]
4|201|Loss: 0.7818050980567932:   3%|▎         | 201/6470 [02:23<1:24:04,  1.24it/s]
4|202|Loss: 0.7546420097351074:   3%|▎         | 201/6470 [02:23<1:24:04,  1.24it/s]
4|202|Loss: 0.7546420097351074:   3%|▎         | 202/6470 [02:24<1:20:38,  1.30it/s]
4|203|Loss: 0.6763289570808411:   3%|▎         | 202/6470 [02:24<1:20:38,  1.30it/s]
4|203|Loss: 0.6763289570808411:   3%|▎         | 203/6470 [02:24<1:20:30,  1.30it/s]
4|204|Loss: 0.7101610898971558:   3%|▎         | 203/6470 [02:25<1:20:30,  1.30it/s]
4|204|Loss: 0.7101610898971558:   3%|▎         | 204/6470 [02:25<1:20:13,  1.30it/s]
4|205|Loss: 0.9235609769821167:   3%|▎         | 204/6470 [02:26<1:20:13,  1.30it/s]
4|205|Loss: 0.9235609769821167:   3%|▎         | 205/6470 [02:26<1:21:13,  1.29it/s]
4|206|Loss: 0.7666192650794983:   3%|▎         | 205/6470 [02:26<1:21:13,  1.29it/s]
4|206|Loss: 0.7666192650794983:   3%|▎         | 206/6470 [02:27<1:20:50,  1.29it/s]
4|207|Loss: 0.8405595421791077:   3%|▎         | 206/6470 [02:27<1:20:50,  1.29it/s]
4|207|Loss: 0.8405595421791077:   3%|▎         | 207/6470 [02:27<1:17:04,  1.35it/s]
4|208|Loss: 0.8370091915130615:   3%|▎         | 207/6470 [02:28<1:17:04,  1.35it/s]
4|208|Loss: 0.8370091915130615:   3%|▎         | 208/6470 [02:28<1:16:29,  1.36it/s]
4|209|Loss: 0.8588727116584778:   3%|▎         | 208/6470 [02:28<1:16:29,  1.36it/s]
4|209|Loss: 0.8588727116584778:   3%|▎         | 209/6470 [02:29<1:14:45,  1.40it/s]
4|210|Loss: 0.9610006213188171:   3%|▎         | 209/6470 [02:29<1:14:45,  1.40it/s]
4|210|Loss: 0.9610006213188171:   3%|▎         | 210/6470 [02:29<1:10:22,  1.48it/s]
4|211|Loss: 0.7777842283248901:   3%|▎         | 210/6470 [02:30<1:10:22,  1.48it/s]
4|211|Loss: 0.7777842283248901:   3%|▎         | 211/6470 [02:30<1:11:32,  1.46it/s]
4|212|Loss: 0.8096250891685486:   3%|▎         | 211/6470 [02:30<1:11:32,  1.46it/s]
4|212|Loss: 0.8096250891685486:   3%|▎         | 212/6470 [02:31<1:11:28,  1.46it/s]
4|213|Loss: 0.8041238784790039:   3%|▎         | 212/6470 [02:31<1:11:28,  1.46it/s]
4|213|Loss: 0.8041238784790039:   3%|▎         | 213/6470 [02:31<1:09:26,  1.50it/s]
4|214|Loss: 0.9295218586921692:   3%|▎         | 213/6470 [02:32<1:09:26,  1.50it/s]
4|214|Loss: 0.9295218586921692:   3%|▎         | 214/6470 [02:32<1:11:09,  1.47it/s]
4|215|Loss: 0.7145827412605286:   3%|▎         | 214/6470 [02:33<1:11:09,  1.47it/s]
4|215|Loss: 0.7145827412605286:   3%|▎         | 215/6470 [02:33<1:14:45,  1.39it/s]
4|216|Loss: 0.9280744791030884:   3%|▎         | 215/6470 [02:33<1:14:45,  1.39it/s]
4|216|Loss: 0.9280744791030884:   3%|▎         | 216/6470 [02:34<1:16:21,  1.37it/s]
4|217|Loss: 0.7857933044433594:   3%|▎         | 216/6470 [02:34<1:16:21,  1.37it/s]
4|217|Loss: 0.7857933044433594:   3%|▎         | 217/6470 [02:34<1:18:26,  1.33it/s]
4|218|Loss: 0.8304332494735718:   3%|▎         | 217/6470 [02:35<1:18:26,  1.33it/s]
4|218|Loss: 0.8304332494735718:   3%|▎         | 218/6470 [02:35<1:19:05,  1.32it/s]
4|219|Loss: 0.6149933934211731:   3%|▎         | 218/6470 [02:36<1:19:05,  1.32it/s]
4|219|Loss: 0.6149933934211731:   3%|▎         | 219/6470 [02:36<1:20:18,  1.30it/s]
4|220|Loss: 0.9445862174034119:   3%|▎         | 219/6470 [02:36<1:20:18,  1.30it/s]
4|220|Loss: 0.9445862174034119:   3%|▎         | 220/6470 [02:37<1:18:26,  1.33it/s]
4|221|Loss: 0.7588666081428528:   3%|▎         | 220/6470 [02:37<1:18:26,  1.33it/s]
4|221|Loss: 0.7588666081428528:   3%|▎         | 221/6470 [02:37<1:13:02,  1.43it/s]
4|222|Loss: 0.8403328061103821:   3%|▎         | 221/6470 [02:38<1:13:02,  1.43it/s]
4|222|Loss: 0.8403328061103821:   3%|▎         | 222/6470 [02:38<1:13:21,  1.42it/s]
4|223|Loss: 0.7401425838470459:   3%|▎         | 222/6470 [02:38<1:13:21,  1.42it/s]
4|223|Loss: 0.7401425838470459:   3%|▎         | 223/6470 [02:39<1:17:29,  1.34it/s]
4|224|Loss: 0.7982325553894043:   3%|▎         | 223/6470 [02:39<1:17:29,  1.34it/s]
4|224|Loss: 0.7982325553894043:   3%|▎         | 224/6470 [02:40<1:22:40,  1.26it/s]
4|225|Loss: 0.8279997706413269:   3%|▎         | 224/6470 [02:40<1:22:40,  1.26it/s]
4|225|Loss: 0.8279997706413269:   3%|▎         | 225/6470 [02:41<1:22:06,  1.27it/s]
4|226|Loss: 0.7069533467292786:   3%|▎         | 225/6470 [02:41<1:22:06,  1.27it/s]
4|226|Loss: 0.7069533467292786:   3%|▎         | 226/6470 [02:41<1:21:27,  1.28it/s]
4|227|Loss: 0.7844511866569519:   3%|▎         | 226/6470 [02:42<1:21:27,  1.28it/s]
4|227|Loss: 0.7844511866569519:   4%|▎         | 227/6470 [02:42<1:19:05,  1.32it/s]
4|228|Loss: 0.838737964630127:   4%|▎         | 227/6470 [02:42<1:19:05,  1.32it/s] 
4|228|Loss: 0.838737964630127:   4%|▎         | 228/6470 [02:43<1:12:35,  1.43it/s]
4|229|Loss: 0.7916483283042908:   4%|▎         | 228/6470 [02:43<1:12:35,  1.43it/s]
4|229|Loss: 0.7916483283042908:   4%|▎         | 229/6470 [02:43<1:12:22,  1.44it/s]
4|230|Loss: 0.8755938410758972:   4%|▎         | 229/6470 [02:44<1:12:22,  1.44it/s]
4|230|Loss: 0.8755938410758972:   4%|▎         | 230/6470 [02:44<1:15:04,  1.39it/s]
4|231|Loss: 0.7956733107566833:   4%|▎         | 230/6470 [02:44<1:15:04,  1.39it/s]
4|231|Loss: 0.7956733107566833:   4%|▎         | 231/6470 [02:45<1:14:07,  1.40it/s]
4|232|Loss: 0.822686493396759:   4%|▎         | 231/6470 [02:45<1:14:07,  1.40it/s] 
4|232|Loss: 0.822686493396759:   4%|▎         | 232/6470 [02:45<1:12:17,  1.44it/s]
4|233|Loss: 0.8936145901679993:   4%|▎         | 232/6470 [02:46<1:12:17,  1.44it/s]
4|233|Loss: 0.8936145901679993:   4%|▎         | 233/6470 [02:46<1:06:31,  1.56it/s]
4|234|Loss: 0.8819471001625061:   4%|▎         | 233/6470 [02:46<1:06:31,  1.56it/s]
4|234|Loss: 0.8819471001625061:   4%|▎         | 234/6470 [02:47<1:07:30,  1.54it/s]
4|235|Loss: 0.7703092694282532:   4%|▎         | 234/6470 [02:47<1:07:30,  1.54it/s]
4|235|Loss: 0.7703092694282532:   4%|▎         | 235/6470 [02:47<1:04:49,  1.60it/s]
4|236|Loss: 0.8810383081436157:   4%|▎         | 235/6470 [02:47<1:04:49,  1.60it/s]
4|236|Loss: 0.8810383081436157:   4%|▎         | 236/6470 [02:48<1:07:18,  1.54it/s]
4|237|Loss: 0.7326985001564026:   4%|▎         | 236/6470 [02:48<1:07:18,  1.54it/s]
4|237|Loss: 0.7326985001564026:   4%|▎         | 237/6470 [02:49<1:09:04,  1.50it/s]
4|238|Loss: 0.8783130645751953:   4%|▎         | 237/6470 [02:49<1:09:04,  1.50it/s]
4|238|Loss: 0.8783130645751953:   4%|▎         | 238/6470 [02:49<1:06:16,  1.57it/s]
4|239|Loss: 0.6911543011665344:   4%|▎         | 238/6470 [02:50<1:06:16,  1.57it/s]
4|239|Loss: 0.6911543011665344:   4%|▎         | 239/6470 [02:50<1:12:37,  1.43it/s]
4|240|Loss: 0.6744344830513:   4%|▎         | 239/6470 [02:50<1:12:37,  1.43it/s]   
4|240|Loss: 0.6744344830513:   4%|▎         | 240/6470 [02:51<1:17:33,  1.34it/s]
4|241|Loss: 0.9802392721176147:   4%|▎         | 240/6470 [02:51<1:17:33,  1.34it/s]
4|241|Loss: 0.9802392721176147:   4%|▎         | 241/6470 [02:52<1:19:51,  1.30it/s]
4|242|Loss: 1.0244102478027344:   4%|▎         | 241/6470 [02:52<1:19:51,  1.30it/s]
4|242|Loss: 1.0244102478027344:   4%|▎         | 242/6470 [02:52<1:18:37,  1.32it/s]
4|243|Loss: 0.8313546776771545:   4%|▎         | 242/6470 [02:53<1:18:37,  1.32it/s]
4|243|Loss: 0.8313546776771545:   4%|▍         | 243/6470 [02:53<1:18:56,  1.31it/s]
4|244|Loss: 1.0346955060958862:   4%|▍         | 243/6470 [02:54<1:18:56,  1.31it/s]
4|244|Loss: 1.0346955060958862:   4%|▍         | 244/6470 [02:54<1:19:37,  1.30it/s]
4|245|Loss: 0.8736708760261536:   4%|▍         | 244/6470 [02:54<1:19:37,  1.30it/s]
4|245|Loss: 0.8736708760261536:   4%|▍         | 245/6470 [02:55<1:20:27,  1.29it/s]
4|246|Loss: 0.6603668332099915:   4%|▍         | 245/6470 [02:55<1:20:27,  1.29it/s]
4|246|Loss: 0.6603668332099915:   4%|▍         | 246/6470 [02:56<1:24:34,  1.23it/s]
4|247|Loss: 0.8561848402023315:   4%|▍         | 246/6470 [02:56<1:24:34,  1.23it/s]
4|247|Loss: 0.8561848402023315:   4%|▍         | 247/6470 [02:56<1:21:33,  1.27it/s]
4|248|Loss: 0.720827043056488:   4%|▍         | 247/6470 [02:57<1:21:33,  1.27it/s] 
4|248|Loss: 0.720827043056488:   4%|▍         | 248/6470 [02:57<1:22:45,  1.25it/s]
4|249|Loss: 0.7990253567695618:   4%|▍         | 248/6470 [02:58<1:22:45,  1.25it/s]
4|249|Loss: 0.7990253567695618:   4%|▍         | 249/6470 [02:58<1:22:54,  1.25it/s]
4|250|Loss: 0.8326981067657471:   4%|▍         | 249/6470 [02:58<1:22:54,  1.25it/s]
4|250|Loss: 0.8326981067657471:   4%|▍         | 250/6470 [02:59<1:16:30,  1.35it/s]
4|251|Loss: 0.8170386552810669:   4%|▍         | 250/6470 [02:59<1:16:30,  1.35it/s]
4|251|Loss: 0.8170386552810669:   4%|▍         | 251/6470 [02:59<1:13:15,  1.41it/s]
4|252|Loss: 0.7015657424926758:   4%|▍         | 251/6470 [03:00<1:13:15,  1.41it/s]
4|252|Loss: 0.7015657424926758:   4%|▍         | 252/6470 [03:00<1:10:29,  1.47it/s]
4|253|Loss: 0.7480570077896118:   4%|▍         | 252/6470 [03:00<1:10:29,  1.47it/s]
4|253|Loss: 0.7480570077896118:   4%|▍         | 253/6470 [03:01<1:14:12,  1.40it/s]
4|254|Loss: 0.9398956894874573:   4%|▍         | 253/6470 [03:01<1:14:12,  1.40it/s]
4|254|Loss: 0.9398956894874573:   4%|▍         | 254/6470 [03:02<1:19:39,  1.30it/s]
4|255|Loss: 0.7921562790870667:   4%|▍         | 254/6470 [03:02<1:19:39,  1.30it/s]
4|255|Loss: 0.7921562790870667:   4%|▍         | 255/6470 [03:02<1:24:06,  1.23it/s]
4|256|Loss: 0.937789261341095:   4%|▍         | 255/6470 [03:03<1:24:06,  1.23it/s] 
4|256|Loss: 0.937789261341095:   4%|▍         | 256/6470 [03:03<1:18:26,  1.32it/s]
4|257|Loss: 0.7886470556259155:   4%|▍         | 256/6470 [03:03<1:18:26,  1.32it/s]
4|257|Loss: 0.7886470556259155:   4%|▍         | 257/6470 [03:04<1:16:23,  1.36it/s]
4|258|Loss: 1.0352377891540527:   4%|▍         | 257/6470 [03:04<1:16:23,  1.36it/s]
4|258|Loss: 1.0352377891540527:   4%|▍         | 258/6470 [03:05<1:17:22,  1.34it/s]
4|259|Loss: 0.9062053561210632:   4%|▍         | 258/6470 [03:05<1:17:22,  1.34it/s]
4|259|Loss: 0.9062053561210632:   4%|▍         | 259/6470 [03:05<1:16:03,  1.36it/s]
4|260|Loss: 0.9117635488510132:   4%|▍         | 259/6470 [03:06<1:16:03,  1.36it/s]
4|260|Loss: 0.9117635488510132:   4%|▍         | 260/6470 [03:06<1:17:47,  1.33it/s]
4|261|Loss: 0.8833577036857605:   4%|▍         | 260/6470 [03:06<1:17:47,  1.33it/s]
4|261|Loss: 0.8833577036857605:   4%|▍         | 261/6470 [03:07<1:14:02,  1.40it/s]
4|262|Loss: 0.8634670376777649:   4%|▍         | 261/6470 [03:07<1:14:02,  1.40it/s]
4|262|Loss: 0.8634670376777649:   4%|▍         | 262/6470 [03:07<1:13:52,  1.40it/s]
4|263|Loss: 0.7496711611747742:   4%|▍         | 262/6470 [03:08<1:13:52,  1.40it/s]
4|263|Loss: 0.7496711611747742:   4%|▍         | 263/6470 [03:08<1:18:12,  1.32it/s]
4|264|Loss: 0.7662057280540466:   4%|▍         | 263/6470 [03:09<1:18:12,  1.32it/s]
4|264|Loss: 0.7662057280540466:   4%|▍         | 264/6470 [03:09<1:16:43,  1.35it/s]
4|265|Loss: 0.775120198726654:   4%|▍         | 264/6470 [03:09<1:16:43,  1.35it/s] 
4|265|Loss: 0.775120198726654:   4%|▍         | 265/6470 [03:10<1:14:05,  1.40it/s]
4|266|Loss: 0.8926130533218384:   4%|▍         | 265/6470 [03:10<1:14:05,  1.40it/s]
4|266|Loss: 0.8926130533218384:   4%|▍         | 266/6470 [03:10<1:15:24,  1.37it/s]
4|267|Loss: 0.8354920744895935:   4%|▍         | 266/6470 [03:11<1:15:24,  1.37it/s]
4|267|Loss: 0.8354920744895935:   4%|▍         | 267/6470 [03:11<1:19:35,  1.30it/s]
4|268|Loss: 0.5532041192054749:   4%|▍         | 267/6470 [03:12<1:19:35,  1.30it/s]
4|268|Loss: 0.5532041192054749:   4%|▍         | 268/6470 [03:12<1:25:36,  1.21it/s]
4|269|Loss: 1.0364470481872559:   4%|▍         | 268/6470 [03:13<1:25:36,  1.21it/s]
4|269|Loss: 1.0364470481872559:   4%|▍         | 269/6470 [03:13<1:23:31,  1.24it/s]
4|270|Loss: 0.6552581191062927:   4%|▍         | 269/6470 [03:13<1:23:31,  1.24it/s]
4|270|Loss: 0.6552581191062927:   4%|▍         | 270/6470 [03:14<1:21:41,  1.26it/s]
4|271|Loss: 0.8521113395690918:   4%|▍         | 270/6470 [03:14<1:21:41,  1.26it/s]
4|271|Loss: 0.8521113395690918:   4%|▍         | 271/6470 [03:14<1:20:15,  1.29it/s]
4|272|Loss: 0.7660637497901917:   4%|▍         | 271/6470 [03:15<1:20:15,  1.29it/s]
4|272|Loss: 0.7660637497901917:   4%|▍         | 272/6470 [03:15<1:23:45,  1.23it/s]
4|273|Loss: 0.8716219067573547:   4%|▍         | 272/6470 [03:16<1:23:45,  1.23it/s]
4|273|Loss: 0.8716219067573547:   4%|▍         | 273/6470 [03:16<1:24:29,  1.22it/s]
4|274|Loss: 0.7510287165641785:   4%|▍         | 273/6470 [03:17<1:24:29,  1.22it/s]
4|274|Loss: 0.7510287165641785:   4%|▍         | 274/6470 [03:17<1:20:05,  1.29it/s]
4|275|Loss: 0.8694741129875183:   4%|▍         | 274/6470 [03:17<1:20:05,  1.29it/s]
4|275|Loss: 0.8694741129875183:   4%|▍         | 275/6470 [03:18<1:22:18,  1.25it/s]
4|276|Loss: 0.824576735496521:   4%|▍         | 275/6470 [03:18<1:22:18,  1.25it/s] 
4|276|Loss: 0.824576735496521:   4%|▍         | 276/6470 [03:18<1:15:14,  1.37it/s]
4|277|Loss: 0.768799901008606:   4%|▍         | 276/6470 [03:19<1:15:14,  1.37it/s]
4|277|Loss: 0.768799901008606:   4%|▍         | 277/6470 [03:19<1:13:50,  1.40it/s]
4|278|Loss: 0.8298488855361938:   4%|▍         | 277/6470 [03:19<1:13:50,  1.40it/s]
4|278|Loss: 0.8298488855361938:   4%|▍         | 278/6470 [03:20<1:13:04,  1.41it/s]
4|279|Loss: 0.8876025080680847:   4%|▍         | 278/6470 [03:20<1:13:04,  1.41it/s]
4|279|Loss: 0.8876025080680847:   4%|▍         | 279/6470 [03:20<1:11:53,  1.44it/s]
4|280|Loss: 0.8186056613922119:   4%|▍         | 279/6470 [03:21<1:11:53,  1.44it/s]
4|280|Loss: 0.8186056613922119:   4%|▍         | 280/6470 [03:21<1:09:00,  1.49it/s]
4|281|Loss: 0.7598427534103394:   4%|▍         | 280/6470 [03:21<1:09:00,  1.49it/s]
4|281|Loss: 0.7598427534103394:   4%|▍         | 281/6470 [03:21<1:07:07,  1.54it/s]
4|282|Loss: 0.8850409984588623:   4%|▍         | 281/6470 [03:22<1:07:07,  1.54it/s]
4|282|Loss: 0.8850409984588623:   4%|▍         | 282/6470 [03:22<1:08:17,  1.51it/s]
4|283|Loss: 0.8741351962089539:   4%|▍         | 282/6470 [03:23<1:08:17,  1.51it/s]
4|283|Loss: 0.8741351962089539:   4%|▍         | 283/6470 [03:23<1:09:57,  1.47it/s]
4|284|Loss: 0.7484298348426819:   4%|▍         | 283/6470 [03:23<1:09:57,  1.47it/s]
4|284|Loss: 0.7484298348426819:   4%|▍         | 284/6470 [03:24<1:10:18,  1.47it/s]
4|285|Loss: 0.950695276260376:   4%|▍         | 284/6470 [03:24<1:10:18,  1.47it/s] 
4|285|Loss: 0.950695276260376:   4%|▍         | 285/6470 [03:24<1:08:48,  1.50it/s]
4|286|Loss: 0.8567452430725098:   4%|▍         | 285/6470 [03:25<1:08:48,  1.50it/s]
4|286|Loss: 0.8567452430725098:   4%|▍         | 286/6470 [03:25<1:10:48,  1.46it/s]
4|287|Loss: 0.6373888254165649:   4%|▍         | 286/6470 [03:25<1:10:48,  1.46it/s]
4|287|Loss: 0.6373888254165649:   4%|▍         | 287/6470 [03:26<1:22:25,  1.25it/s]
4|288|Loss: 0.900208055973053:   4%|▍         | 287/6470 [03:27<1:22:25,  1.25it/s] 
4|288|Loss: 0.900208055973053:   4%|▍         | 288/6470 [03:27<1:21:59,  1.26it/s]
4|289|Loss: 0.9075039625167847:   4%|▍         | 288/6470 [03:27<1:21:59,  1.26it/s]
4|289|Loss: 0.9075039625167847:   4%|▍         | 289/6470 [03:27<1:18:05,  1.32it/s]
4|290|Loss: 0.9626989960670471:   4%|▍         | 289/6470 [03:28<1:18:05,  1.32it/s]
4|290|Loss: 0.9626989960670471:   4%|▍         | 290/6470 [03:28<1:15:51,  1.36it/s]
4|291|Loss: 0.8103177547454834:   4%|▍         | 290/6470 [03:29<1:15:51,  1.36it/s]
4|291|Loss: 0.8103177547454834:   4%|▍         | 291/6470 [03:29<1:12:38,  1.42it/s]
4|292|Loss: 0.9982197880744934:   4%|▍         | 291/6470 [03:29<1:12:38,  1.42it/s]
4|292|Loss: 0.9982197880744934:   5%|▍         | 292/6470 [03:30<1:17:43,  1.32it/s]
4|293|Loss: 0.8531226515769958:   5%|▍         | 292/6470 [03:30<1:17:43,  1.32it/s]
4|293|Loss: 0.8531226515769958:   5%|▍         | 293/6470 [03:31<1:20:24,  1.28it/s]
4|294|Loss: 0.9110802412033081:   5%|▍         | 293/6470 [03:31<1:20:24,  1.28it/s]
4|294|Loss: 0.9110802412033081:   5%|▍         | 294/6470 [03:31<1:21:10,  1.27it/s]
4|295|Loss: 0.796401858329773:   5%|▍         | 294/6470 [03:32<1:21:10,  1.27it/s] 
4|295|Loss: 0.796401858329773:   5%|▍         | 295/6470 [03:32<1:18:31,  1.31it/s]
4|296|Loss: 0.7399086356163025:   5%|▍         | 295/6470 [03:32<1:18:31,  1.31it/s]
4|296|Loss: 0.7399086356163025:   5%|▍         | 296/6470 [03:33<1:10:49,  1.45it/s]
4|297|Loss: 0.829905092716217:   5%|▍         | 296/6470 [03:33<1:10:49,  1.45it/s] 
4|297|Loss: 0.829905092716217:   5%|▍         | 297/6470 [03:33<1:15:35,  1.36it/s]
4|298|Loss: 0.8784583210945129:   5%|▍         | 297/6470 [03:34<1:15:35,  1.36it/s]
4|298|Loss: 0.8784583210945129:   5%|▍         | 298/6470 [03:34<1:13:58,  1.39it/s]
4|299|Loss: 0.8818023800849915:   5%|▍         | 298/6470 [03:34<1:13:58,  1.39it/s]
4|299|Loss: 0.8818023800849915:   5%|▍         | 299/6470 [03:35<1:17:04,  1.33it/s]
4|300|Loss: 0.8434765934944153:   5%|▍         | 299/6470 [03:35<1:17:04,  1.33it/s]
4|300|Loss: 0.8434765934944153:   5%|▍         | 300/6470 [03:35<1:12:29,  1.42it/s]
4|301|Loss: 0.9658882021903992:   5%|▍         | 300/6470 [03:36<1:12:29,  1.42it/s]
4|301|Loss: 0.9658882021903992:   5%|▍         | 301/6470 [03:36<1:09:20,  1.48it/s]
4|302|Loss: 0.7350128889083862:   5%|▍         | 301/6470 [03:36<1:09:20,  1.48it/s]
4|302|Loss: 0.7350128889083862:   5%|▍         | 302/6470 [03:37<1:05:35,  1.57it/s]
4|303|Loss: 0.8248381614685059:   5%|▍         | 302/6470 [03:37<1:05:35,  1.57it/s]
4|303|Loss: 0.8248381614685059:   5%|▍         | 303/6470 [03:37<1:04:16,  1.60it/s]
4|304|Loss: 0.886500895023346:   5%|▍         | 303/6470 [03:38<1:04:16,  1.60it/s] 
4|304|Loss: 0.886500895023346:   5%|▍         | 304/6470 [03:38<1:08:31,  1.50it/s]
4|305|Loss: 0.7916440963745117:   5%|▍         | 304/6470 [03:38<1:08:31,  1.50it/s]
4|305|Loss: 0.7916440963745117:   5%|▍         | 305/6470 [03:39<1:11:27,  1.44it/s]
4|306|Loss: 0.8710598349571228:   5%|▍         | 305/6470 [03:39<1:11:27,  1.44it/s]
4|306|Loss: 0.8710598349571228:   5%|▍         | 306/6470 [03:39<1:10:59,  1.45it/s]
4|307|Loss: 0.7020972371101379:   5%|▍         | 306/6470 [03:40<1:10:59,  1.45it/s]
4|307|Loss: 0.7020972371101379:   5%|▍         | 307/6470 [03:40<1:10:06,  1.47it/s]
4|308|Loss: 0.7259268164634705:   5%|▍         | 307/6470 [03:40<1:10:06,  1.47it/s]
4|308|Loss: 0.7259268164634705:   5%|▍         | 308/6470 [03:41<1:05:44,  1.56it/s]
4|309|Loss: 0.9339123964309692:   5%|▍         | 308/6470 [03:41<1:05:44,  1.56it/s]
4|309|Loss: 0.9339123964309692:   5%|▍         | 309/6470 [03:41<1:04:10,  1.60it/s]
4|310|Loss: 0.8800286650657654:   5%|▍         | 309/6470 [03:42<1:04:10,  1.60it/s]
4|310|Loss: 0.8800286650657654:   5%|▍         | 310/6470 [03:42<1:04:50,  1.58it/s]
4|311|Loss: 0.6732369065284729:   5%|▍         | 310/6470 [03:42<1:04:50,  1.58it/s]
4|311|Loss: 0.6732369065284729:   5%|▍         | 311/6470 [03:43<1:08:14,  1.50it/s]
4|312|Loss: 0.8758792877197266:   5%|▍         | 311/6470 [03:43<1:08:14,  1.50it/s]
4|312|Loss: 0.8758792877197266:   5%|▍         | 312/6470 [03:43<1:10:43,  1.45it/s]
4|313|Loss: 0.8619300127029419:   5%|▍         | 312/6470 [03:44<1:10:43,  1.45it/s]
4|313|Loss: 0.8619300127029419:   5%|▍         | 313/6470 [03:44<1:11:28,  1.44it/s]
4|314|Loss: 0.8881439566612244:   5%|▍         | 313/6470 [03:44<1:11:28,  1.44it/s]
4|314|Loss: 0.8881439566612244:   5%|▍         | 314/6470 [03:45<1:13:51,  1.39it/s]
4|315|Loss: 0.784593939781189:   5%|▍         | 314/6470 [03:45<1:13:51,  1.39it/s] 
4|315|Loss: 0.784593939781189:   5%|▍         | 315/6470 [03:46<1:12:50,  1.41it/s]
4|316|Loss: 0.9971711039543152:   5%|▍         | 315/6470 [03:46<1:12:50,  1.41it/s]
4|316|Loss: 0.9971711039543152:   5%|▍         | 316/6470 [03:46<1:13:42,  1.39it/s]
4|317|Loss: 0.8658508658409119:   5%|▍         | 316/6470 [03:47<1:13:42,  1.39it/s]
4|317|Loss: 0.8658508658409119:   5%|▍         | 317/6470 [03:47<1:18:15,  1.31it/s]
4|318|Loss: 0.8795930743217468:   5%|▍         | 317/6470 [03:48<1:18:15,  1.31it/s]
4|318|Loss: 0.8795930743217468:   5%|▍         | 318/6470 [03:48<1:17:18,  1.33it/s]
4|319|Loss: 0.8817647695541382:   5%|▍         | 318/6470 [03:48<1:17:18,  1.33it/s]
4|319|Loss: 0.8817647695541382:   5%|▍         | 319/6470 [03:49<1:17:04,  1.33it/s]
4|320|Loss: 0.8913587331771851:   5%|▍         | 319/6470 [03:49<1:17:04,  1.33it/s]
4|320|Loss: 0.8913587331771851:   5%|▍         | 320/6470 [03:49<1:14:53,  1.37it/s]
4|321|Loss: 0.8629904389381409:   5%|▍         | 320/6470 [03:50<1:14:53,  1.37it/s]
4|321|Loss: 0.8629904389381409:   5%|▍         | 321/6470 [03:50<1:13:47,  1.39it/s]
4|322|Loss: 0.8431513905525208:   5%|▍         | 321/6470 [03:50<1:13:47,  1.39it/s]
4|322|Loss: 0.8431513905525208:   5%|▍         | 322/6470 [03:51<1:12:16,  1.42it/s]
4|323|Loss: 0.856758177280426:   5%|▍         | 322/6470 [03:51<1:12:16,  1.42it/s] 
4|323|Loss: 0.856758177280426:   5%|▍         | 323/6470 [03:51<1:12:15,  1.42it/s]
4|324|Loss: 0.8532323241233826:   5%|▍         | 323/6470 [03:52<1:12:15,  1.42it/s]
4|324|Loss: 0.8532323241233826:   5%|▌         | 324/6470 [03:52<1:08:06,  1.50it/s]
4|325|Loss: 0.7268282771110535:   5%|▌         | 324/6470 [03:52<1:08:06,  1.50it/s]
4|325|Loss: 0.7268282771110535:   5%|▌         | 325/6470 [03:53<1:09:48,  1.47it/s]
4|326|Loss: 0.8674860596656799:   5%|▌         | 325/6470 [03:53<1:09:48,  1.47it/s]
4|326|Loss: 0.8674860596656799:   5%|▌         | 326/6470 [03:53<1:12:53,  1.40it/s]
4|327|Loss: 0.7173610329627991:   5%|▌         | 326/6470 [03:54<1:12:53,  1.40it/s]
4|327|Loss: 0.7173610329627991:   5%|▌         | 327/6470 [03:54<1:13:18,  1.40it/s]
4|328|Loss: 0.7243071794509888:   5%|▌         | 327/6470 [03:55<1:13:18,  1.40it/s]
4|328|Loss: 0.7243071794509888:   5%|▌         | 328/6470 [03:55<1:11:15,  1.44it/s]
4|329|Loss: 0.9090154767036438:   5%|▌         | 328/6470 [03:55<1:11:15,  1.44it/s]
4|329|Loss: 0.9090154767036438:   5%|▌         | 329/6470 [03:55<1:10:34,  1.45it/s]
4|330|Loss: 0.8356194496154785:   5%|▌         | 329/6470 [03:56<1:10:34,  1.45it/s]
4|330|Loss: 0.8356194496154785:   5%|▌         | 330/6470 [03:56<1:12:36,  1.41it/s]
4|331|Loss: 0.8126760721206665:   5%|▌         | 330/6470 [03:57<1:12:36,  1.41it/s]
4|331|Loss: 0.8126760721206665:   5%|▌         | 331/6470 [03:57<1:09:48,  1.47it/s]
4|332|Loss: 0.7438232898712158:   5%|▌         | 331/6470 [03:57<1:09:48,  1.47it/s]
4|332|Loss: 0.7438232898712158:   5%|▌         | 332/6470 [03:58<1:13:27,  1.39it/s]
4|333|Loss: 0.7384383082389832:   5%|▌         | 332/6470 [03:58<1:13:27,  1.39it/s]
4|333|Loss: 0.7384383082389832:   5%|▌         | 333/6470 [03:58<1:15:47,  1.35it/s]
4|334|Loss: 0.7719569802284241:   5%|▌         | 333/6470 [03:59<1:15:47,  1.35it/s]
4|334|Loss: 0.7719569802284241:   5%|▌         | 334/6470 [03:59<1:12:21,  1.41it/s]
4|335|Loss: 0.7770245671272278:   5%|▌         | 334/6470 [03:59<1:12:21,  1.41it/s]
4|335|Loss: 0.7770245671272278:   5%|▌         | 335/6470 [04:00<1:09:25,  1.47it/s]
4|336|Loss: 1.0294578075408936:   5%|▌         | 335/6470 [04:00<1:09:25,  1.47it/s]
4|336|Loss: 1.0294578075408936:   5%|▌         | 336/6470 [04:00<1:10:00,  1.46it/s]
4|337|Loss: 0.7645596861839294:   5%|▌         | 336/6470 [04:01<1:10:00,  1.46it/s]
4|337|Loss: 0.7645596861839294:   5%|▌         | 337/6470 [04:01<1:16:13,  1.34it/s]
4|338|Loss: 0.7358471155166626:   5%|▌         | 337/6470 [04:02<1:16:13,  1.34it/s]
4|338|Loss: 0.7358471155166626:   5%|▌         | 338/6470 [04:02<1:22:04,  1.25it/s]
4|339|Loss: 1.2731730937957764:   5%|▌         | 338/6470 [04:03<1:22:04,  1.25it/s]
4|339|Loss: 1.2731730937957764:   5%|▌         | 339/6470 [04:03<1:30:08,  1.13it/s]
4|340|Loss: 0.917410135269165:   5%|▌         | 339/6470 [04:04<1:30:08,  1.13it/s] 
4|340|Loss: 0.917410135269165:   5%|▌         | 340/6470 [04:04<1:28:26,  1.16it/s]
4|341|Loss: 0.9201186895370483:   5%|▌         | 340/6470 [04:05<1:28:26,  1.16it/s]
4|341|Loss: 0.9201186895370483:   5%|▌         | 341/6470 [04:05<1:29:33,  1.14it/s]
4|342|Loss: 0.8570614457130432:   5%|▌         | 341/6470 [04:05<1:29:33,  1.14it/s]
4|342|Loss: 0.8570614457130432:   5%|▌         | 342/6470 [04:06<1:27:19,  1.17it/s]
4|343|Loss: 0.9187653660774231:   5%|▌         | 342/6470 [04:06<1:27:19,  1.17it/s]
4|343|Loss: 0.9187653660774231:   5%|▌         | 343/6470 [04:07<1:25:51,  1.19it/s]
4|344|Loss: 0.9037707448005676:   5%|▌         | 343/6470 [04:07<1:25:51,  1.19it/s]
4|344|Loss: 0.9037707448005676:   5%|▌         | 344/6470 [04:07<1:21:53,  1.25it/s]
4|345|Loss: 0.8133350610733032:   5%|▌         | 344/6470 [04:08<1:21:53,  1.25it/s]
4|345|Loss: 0.8133350610733032:   5%|▌         | 345/6470 [04:08<1:16:51,  1.33it/s]
4|346|Loss: 0.8509073853492737:   5%|▌         | 345/6470 [04:08<1:16:51,  1.33it/s]
4|346|Loss: 0.8509073853492737:   5%|▌         | 346/6470 [04:09<1:14:02,  1.38it/s]
4|347|Loss: 0.8214389085769653:   5%|▌         | 346/6470 [04:09<1:14:02,  1.38it/s]
4|347|Loss: 0.8214389085769653:   5%|▌         | 347/6470 [04:09<1:15:31,  1.35it/s]
4|348|Loss: 0.8782103061676025:   5%|▌         | 347/6470 [04:10<1:15:31,  1.35it/s]
4|348|Loss: 0.8782103061676025:   5%|▌         | 348/6470 [04:10<1:14:37,  1.37it/s]
4|349|Loss: 0.8856062889099121:   5%|▌         | 348/6470 [04:11<1:14:37,  1.37it/s]
4|349|Loss: 0.8856062889099121:   5%|▌         | 349/6470 [04:11<1:19:02,  1.29it/s]
4|350|Loss: 0.8540850281715393:   5%|▌         | 349/6470 [04:11<1:19:02,  1.29it/s]
4|350|Loss: 0.8540850281715393:   5%|▌         | 350/6470 [04:12<1:20:32,  1.27it/s]
4|351|Loss: 0.731577455997467:   5%|▌         | 350/6470 [04:12<1:20:32,  1.27it/s] 
4|351|Loss: 0.731577455997467:   5%|▌         | 351/6470 [04:13<1:22:18,  1.24it/s]
4|352|Loss: 0.8632545471191406:   5%|▌         | 351/6470 [04:13<1:22:18,  1.24it/s]
4|352|Loss: 0.8632545471191406:   5%|▌         | 352/6470 [04:13<1:21:14,  1.26it/s]
4|353|Loss: 0.837997317314148:   5%|▌         | 352/6470 [04:14<1:21:14,  1.26it/s] 
4|353|Loss: 0.837997317314148:   5%|▌         | 353/6470 [04:14<1:21:09,  1.26it/s]
4|354|Loss: 0.7889013290405273:   5%|▌         | 353/6470 [04:15<1:21:09,  1.26it/s]
4|354|Loss: 0.7889013290405273:   5%|▌         | 354/6470 [04:15<1:20:30,  1.27it/s]
4|355|Loss: 0.8401105403900146:   5%|▌         | 354/6470 [04:15<1:20:30,  1.27it/s]
4|355|Loss: 0.8401105403900146:   5%|▌         | 355/6470 [04:16<1:23:00,  1.23it/s]
4|356|Loss: 0.7050682306289673:   5%|▌         | 355/6470 [04:16<1:23:00,  1.23it/s]
4|356|Loss: 0.7050682306289673:   6%|▌         | 356/6470 [04:17<1:16:46,  1.33it/s]
4|357|Loss: 0.782357394695282:   6%|▌         | 356/6470 [04:17<1:16:46,  1.33it/s] 
4|357|Loss: 0.782357394695282:   6%|▌         | 357/6470 [04:17<1:13:04,  1.39it/s]
4|358|Loss: 0.8832423686981201:   6%|▌         | 357/6470 [04:18<1:13:04,  1.39it/s]
4|358|Loss: 0.8832423686981201:   6%|▌         | 358/6470 [04:18<1:13:45,  1.38it/s]
4|359|Loss: 0.7830119729042053:   6%|▌         | 358/6470 [04:18<1:13:45,  1.38it/s]
4|359|Loss: 0.7830119729042053:   6%|▌         | 359/6470 [04:19<1:20:04,  1.27it/s]
4|360|Loss: 0.7764135599136353:   6%|▌         | 359/6470 [04:19<1:20:04,  1.27it/s]
4|360|Loss: 0.7764135599136353:   6%|▌         | 360/6470 [04:20<1:19:04,  1.29it/s]
4|361|Loss: 0.8826438784599304:   6%|▌         | 360/6470 [04:20<1:19:04,  1.29it/s]
4|361|Loss: 0.8826438784599304:   6%|▌         | 361/6470 [04:20<1:18:52,  1.29it/s]
4|362|Loss: 0.7377400994300842:   6%|▌         | 361/6470 [04:21<1:18:52,  1.29it/s]
4|362|Loss: 0.7377400994300842:   6%|▌         | 362/6470 [04:21<1:18:03,  1.30it/s]
4|363|Loss: 0.8851523399353027:   6%|▌         | 362/6470 [04:21<1:18:03,  1.30it/s]
4|363|Loss: 0.8851523399353027:   6%|▌         | 363/6470 [04:22<1:14:11,  1.37it/s]
4|364|Loss: 0.7812185883522034:   6%|▌         | 363/6470 [04:22<1:14:11,  1.37it/s]
4|364|Loss: 0.7812185883522034:   6%|▌         | 364/6470 [04:22<1:11:12,  1.43it/s]
4|365|Loss: 0.9014503955841064:   6%|▌         | 364/6470 [04:23<1:11:12,  1.43it/s]
4|365|Loss: 0.9014503955841064:   6%|▌         | 365/6470 [04:23<1:13:23,  1.39it/s]
4|366|Loss: 0.9219799637794495:   6%|▌         | 365/6470 [04:24<1:13:23,  1.39it/s]
4|366|Loss: 0.9219799637794495:   6%|▌         | 366/6470 [04:24<1:11:58,  1.41it/s]
4|367|Loss: 0.9246518611907959:   6%|▌         | 366/6470 [04:24<1:11:58,  1.41it/s]
4|367|Loss: 0.9246518611907959:   6%|▌         | 367/6470 [04:25<1:12:31,  1.40it/s]
4|368|Loss: 0.8387815356254578:   6%|▌         | 367/6470 [04:25<1:12:31,  1.40it/s]
4|368|Loss: 0.8387815356254578:   6%|▌         | 368/6470 [04:25<1:14:51,  1.36it/s]
4|369|Loss: 0.8578734397888184:   6%|▌         | 368/6470 [04:26<1:14:51,  1.36it/s]
4|369|Loss: 0.8578734397888184:   6%|▌         | 369/6470 [04:26<1:14:32,  1.36it/s]
4|370|Loss: 0.9254553318023682:   6%|▌         | 369/6470 [04:26<1:14:32,  1.36it/s]
4|370|Loss: 0.9254553318023682:   6%|▌         | 370/6470 [04:27<1:16:27,  1.33it/s]
4|371|Loss: 0.5811880230903625:   6%|▌         | 370/6470 [04:27<1:16:27,  1.33it/s]
4|371|Loss: 0.5811880230903625:   6%|▌         | 371/6470 [04:28<1:21:35,  1.25it/s]
4|372|Loss: 0.8280573487281799:   6%|▌         | 371/6470 [04:28<1:21:35,  1.25it/s]
4|372|Loss: 0.8280573487281799:   6%|▌         | 372/6470 [04:28<1:18:38,  1.29it/s]
4|373|Loss: 0.6598327159881592:   6%|▌         | 372/6470 [04:29<1:18:38,  1.29it/s]
4|373|Loss: 0.6598327159881592:   6%|▌         | 373/6470 [04:29<1:17:25,  1.31it/s]
4|374|Loss: 0.8835561871528625:   6%|▌         | 373/6470 [04:30<1:17:25,  1.31it/s]
4|374|Loss: 0.8835561871528625:   6%|▌         | 374/6470 [04:30<1:17:42,  1.31it/s]
4|375|Loss: 0.7857524156570435:   6%|▌         | 374/6470 [04:30<1:17:42,  1.31it/s]
4|375|Loss: 0.7857524156570435:   6%|▌         | 375/6470 [04:31<1:21:20,  1.25it/s]
4|376|Loss: 0.8335139751434326:   6%|▌         | 375/6470 [04:31<1:21:20,  1.25it/s]
4|376|Loss: 0.8335139751434326:   6%|▌         | 376/6470 [04:32<1:16:47,  1.32it/s]
4|377|Loss: 0.8083401322364807:   6%|▌         | 376/6470 [04:32<1:16:47,  1.32it/s]
4|377|Loss: 0.8083401322364807:   6%|▌         | 377/6470 [04:32<1:20:14,  1.27it/s]
4|378|Loss: 0.7799650430679321:   6%|▌         | 377/6470 [04:33<1:20:14,  1.27it/s]
4|378|Loss: 0.7799650430679321:   6%|▌         | 378/6470 [04:33<1:18:28,  1.29it/s]
4|379|Loss: 0.799071729183197:   6%|▌         | 378/6470 [04:34<1:18:28,  1.29it/s] 
4|379|Loss: 0.799071729183197:   6%|▌         | 379/6470 [04:34<1:20:59,  1.25it/s]
4|380|Loss: 0.7967026233673096:   6%|▌         | 379/6470 [04:34<1:20:59,  1.25it/s]
4|380|Loss: 0.7967026233673096:   6%|▌         | 380/6470 [04:35<1:21:14,  1.25it/s]
4|381|Loss: 0.7472953796386719:   6%|▌         | 380/6470 [04:35<1:21:14,  1.25it/s]
4|381|Loss: 0.7472953796386719:   6%|▌         | 381/6470 [04:35<1:18:28,  1.29it/s]
4|382|Loss: 0.8092296719551086:   6%|▌         | 381/6470 [04:36<1:18:28,  1.29it/s]
4|382|Loss: 0.8092296719551086:   6%|▌         | 382/6470 [04:36<1:14:07,  1.37it/s]
4|383|Loss: 0.5991453528404236:   6%|▌         | 382/6470 [04:36<1:14:07,  1.37it/s]
4|383|Loss: 0.5991453528404236:   6%|▌         | 383/6470 [04:37<1:11:48,  1.41it/s]
4|384|Loss: 0.944920003414154:   6%|▌         | 383/6470 [04:37<1:11:48,  1.41it/s] 
4|384|Loss: 0.944920003414154:   6%|▌         | 384/6470 [04:38<1:15:23,  1.35it/s]
4|385|Loss: 0.8157945275306702:   6%|▌         | 384/6470 [04:38<1:15:23,  1.35it/s]
4|385|Loss: 0.8157945275306702:   6%|▌         | 385/6470 [04:38<1:17:21,  1.31it/s]
4|386|Loss: 0.9528496861457825:   6%|▌         | 385/6470 [04:39<1:17:21,  1.31it/s]
4|386|Loss: 0.9528496861457825:   6%|▌         | 386/6470 [04:39<1:20:27,  1.26it/s]
4|387|Loss: 0.8295774459838867:   6%|▌         | 386/6470 [04:40<1:20:27,  1.26it/s]
4|387|Loss: 0.8295774459838867:   6%|▌         | 387/6470 [04:40<1:15:21,  1.35it/s]
4|388|Loss: 0.7270637154579163:   6%|▌         | 387/6470 [04:40<1:15:21,  1.35it/s]
4|388|Loss: 0.7270637154579163:   6%|▌         | 388/6470 [04:41<1:14:19,  1.36it/s]
4|389|Loss: 0.7137355208396912:   6%|▌         | 388/6470 [04:41<1:14:19,  1.36it/s]
4|389|Loss: 0.7137355208396912:   6%|▌         | 389/6470 [04:41<1:12:44,  1.39it/s]
4|390|Loss: 0.8576153516769409:   6%|▌         | 389/6470 [04:42<1:12:44,  1.39it/s]
4|390|Loss: 0.8576153516769409:   6%|▌         | 390/6470 [04:42<1:13:30,  1.38it/s]
4|391|Loss: 0.9263573884963989:   6%|▌         | 390/6470 [04:42<1:13:30,  1.38it/s]
4|391|Loss: 0.9263573884963989:   6%|▌         | 391/6470 [04:43<1:14:18,  1.36it/s]
4|392|Loss: 0.7412658929824829:   6%|▌         | 391/6470 [04:43<1:14:18,  1.36it/s]
4|392|Loss: 0.7412658929824829:   6%|▌         | 392/6470 [04:43<1:10:26,  1.44it/s]
4|393|Loss: 0.8448875546455383:   6%|▌         | 392/6470 [04:44<1:10:26,  1.44it/s]
4|393|Loss: 0.8448875546455383:   6%|▌         | 393/6470 [04:44<1:05:37,  1.54it/s]
4|394|Loss: 0.7390130758285522:   6%|▌         | 393/6470 [04:44<1:05:37,  1.54it/s]
4|394|Loss: 0.7390130758285522:   6%|▌         | 394/6470 [04:45<1:11:05,  1.42it/s]
4|395|Loss: 0.7918900847434998:   6%|▌         | 394/6470 [04:45<1:11:05,  1.42it/s]
4|395|Loss: 0.7918900847434998:   6%|▌         | 395/6470 [04:46<1:16:55,  1.32it/s]
4|396|Loss: 0.8533847332000732:   6%|▌         | 395/6470 [04:46<1:16:55,  1.32it/s]
4|396|Loss: 0.8533847332000732:   6%|▌         | 396/6470 [04:46<1:18:19,  1.29it/s]
4|397|Loss: 0.8726099729537964:   6%|▌         | 396/6470 [04:47<1:18:19,  1.29it/s]
4|397|Loss: 0.8726099729537964:   6%|▌         | 397/6470 [04:47<1:14:38,  1.36it/s]
4|398|Loss: 0.7440090775489807:   6%|▌         | 397/6470 [04:48<1:14:38,  1.36it/s]
4|398|Loss: 0.7440090775489807:   6%|▌         | 398/6470 [04:48<1:16:51,  1.32it/s]
4|399|Loss: 0.7833471298217773:   6%|▌         | 398/6470 [04:48<1:16:51,  1.32it/s]
4|399|Loss: 0.7833471298217773:   6%|▌         | 399/6470 [04:49<1:17:08,  1.31it/s]
4|400|Loss: 0.7890766263008118:   6%|▌         | 399/6470 [04:49<1:17:08,  1.31it/s]
4|400|Loss: 0.7890766263008118:   6%|▌         | 400/6470 [04:50<1:20:06,  1.26it/s]
4|401|Loss: 0.8636958003044128:   6%|▌         | 400/6470 [04:50<1:20:06,  1.26it/s]
4|401|Loss: 0.8636958003044128:   6%|▌         | 401/6470 [04:50<1:23:29,  1.21it/s]
4|402|Loss: 1.0474553108215332:   6%|▌         | 401/6470 [04:51<1:23:29,  1.21it/s]
4|402|Loss: 1.0474553108215332:   6%|▌         | 402/6470 [04:51<1:19:19,  1.28it/s]
4|403|Loss: 0.7914741635322571:   6%|▌         | 402/6470 [04:52<1:19:19,  1.28it/s]
4|403|Loss: 0.7914741635322571:   6%|▌         | 403/6470 [04:52<1:16:12,  1.33it/s]
4|404|Loss: 0.7214005589485168:   6%|▌         | 403/6470 [04:52<1:16:12,  1.33it/s]
4|404|Loss: 0.7214005589485168:   6%|▌         | 404/6470 [04:53<1:14:40,  1.35it/s]
4|405|Loss: 0.9044781923294067:   6%|▌         | 404/6470 [04:53<1:14:40,  1.35it/s]
4|405|Loss: 0.9044781923294067:   6%|▋         | 405/6470 [04:53<1:13:18,  1.38it/s]
4|406|Loss: 0.8936153650283813:   6%|▋         | 405/6470 [04:54<1:13:18,  1.38it/s]
4|406|Loss: 0.8936153650283813:   6%|▋         | 406/6470 [04:54<1:09:47,  1.45it/s]
4|407|Loss: 0.8119456171989441:   6%|▋         | 406/6470 [04:54<1:09:47,  1.45it/s]
4|407|Loss: 0.8119456171989441:   6%|▋         | 407/6470 [04:55<1:14:53,  1.35it/s]
4|408|Loss: 0.7689610719680786:   6%|▋         | 407/6470 [04:55<1:14:53,  1.35it/s]
4|408|Loss: 0.7689610719680786:   6%|▋         | 408/6470 [04:55<1:13:24,  1.38it/s]
4|409|Loss: 0.7708481550216675:   6%|▋         | 408/6470 [04:56<1:13:24,  1.38it/s]
4|409|Loss: 0.7708481550216675:   6%|▋         | 409/6470 [04:56<1:17:20,  1.31it/s]
4|410|Loss: 0.8841624855995178:   6%|▋         | 409/6470 [04:57<1:17:20,  1.31it/s]
4|410|Loss: 0.8841624855995178:   6%|▋         | 410/6470 [04:57<1:14:22,  1.36it/s]
4|411|Loss: 0.7864594459533691:   6%|▋         | 410/6470 [04:57<1:14:22,  1.36it/s]
4|411|Loss: 0.7864594459533691:   6%|▋         | 411/6470 [04:58<1:12:41,  1.39it/s]
4|412|Loss: 0.8457425236701965:   6%|▋         | 411/6470 [04:58<1:12:41,  1.39it/s]
4|412|Loss: 0.8457425236701965:   6%|▋         | 412/6470 [04:58<1:17:29,  1.30it/s]
4|413|Loss: 0.8152291178703308:   6%|▋         | 412/6470 [04:59<1:17:29,  1.30it/s]
4|413|Loss: 0.8152291178703308:   6%|▋         | 413/6470 [04:59<1:13:46,  1.37it/s]
4|414|Loss: 0.864529013633728:   6%|▋         | 413/6470 [05:00<1:13:46,  1.37it/s] 
4|414|Loss: 0.864529013633728:   6%|▋         | 414/6470 [05:00<1:16:41,  1.32it/s]
4|415|Loss: 0.8879215717315674:   6%|▋         | 414/6470 [05:00<1:16:41,  1.32it/s]
4|415|Loss: 0.8879215717315674:   6%|▋         | 415/6470 [05:01<1:14:59,  1.35it/s]
4|416|Loss: 0.5646582841873169:   6%|▋         | 415/6470 [05:01<1:14:59,  1.35it/s]
4|416|Loss: 0.5646582841873169:   6%|▋         | 416/6470 [05:02<1:18:41,  1.28it/s]
4|417|Loss: 0.8498809933662415:   6%|▋         | 416/6470 [05:02<1:18:41,  1.28it/s]
4|417|Loss: 0.8498809933662415:   6%|▋         | 417/6470 [05:02<1:18:40,  1.28it/s]
4|418|Loss: 0.8524423837661743:   6%|▋         | 417/6470 [05:03<1:18:40,  1.28it/s]
4|418|Loss: 0.8524423837661743:   6%|▋         | 418/6470 [05:03<1:16:50,  1.31it/s]
4|419|Loss: 0.8533605337142944:   6%|▋         | 418/6470 [05:03<1:16:50,  1.31it/s]
4|419|Loss: 0.8533605337142944:   6%|▋         | 419/6470 [05:04<1:16:56,  1.31it/s]
4|420|Loss: 0.7338610291481018:   6%|▋         | 419/6470 [05:04<1:16:56,  1.31it/s]
4|420|Loss: 0.7338610291481018:   6%|▋         | 420/6470 [05:05<1:19:48,  1.26it/s]
4|421|Loss: 1.137494683265686:   6%|▋         | 420/6470 [05:05<1:19:48,  1.26it/s] 
4|421|Loss: 1.137494683265686:   7%|▋         | 421/6470 [05:05<1:16:17,  1.32it/s]
4|422|Loss: 0.7472543716430664:   7%|▋         | 421/6470 [05:06<1:16:17,  1.32it/s]
4|422|Loss: 0.7472543716430664:   7%|▋         | 422/6470 [05:06<1:15:28,  1.34it/s]
4|423|Loss: 0.7680560946464539:   7%|▋         | 422/6470 [05:06<1:15:28,  1.34it/s]
4|423|Loss: 0.7680560946464539:   7%|▋         | 423/6470 [05:07<1:15:43,  1.33it/s]
4|424|Loss: 0.8371431231498718:   7%|▋         | 423/6470 [05:07<1:15:43,  1.33it/s]
4|424|Loss: 0.8371431231498718:   7%|▋         | 424/6470 [05:07<1:11:20,  1.41it/s]
4|425|Loss: 0.7734920978546143:   7%|▋         | 424/6470 [05:08<1:11:20,  1.41it/s]
4|425|Loss: 0.7734920978546143:   7%|▋         | 425/6470 [05:08<1:08:10,  1.48it/s]
4|426|Loss: 0.7859728336334229:   7%|▋         | 425/6470 [05:08<1:08:10,  1.48it/s]
4|426|Loss: 0.7859728336334229:   7%|▋         | 426/6470 [05:09<1:13:18,  1.37it/s]
4|427|Loss: 0.8557201027870178:   7%|▋         | 426/6470 [05:09<1:13:18,  1.37it/s]
4|427|Loss: 0.8557201027870178:   7%|▋         | 427/6470 [05:09<1:05:49,  1.53it/s]
4|428|Loss: 0.6617242693901062:   7%|▋         | 427/6470 [05:10<1:05:49,  1.53it/s]
4|428|Loss: 0.6617242693901062:   7%|▋         | 428/6470 [05:10<1:04:11,  1.57it/s]
4|429|Loss: 0.7702537178993225:   7%|▋         | 428/6470 [05:10<1:04:11,  1.57it/s]
4|429|Loss: 0.7702537178993225:   7%|▋         | 429/6470 [05:11<1:09:27,  1.45it/s]
4|430|Loss: 0.9232531785964966:   7%|▋         | 429/6470 [05:11<1:09:27,  1.45it/s]
4|430|Loss: 0.9232531785964966:   7%|▋         | 430/6470 [05:12<1:12:35,  1.39it/s]
4|431|Loss: 0.8050158619880676:   7%|▋         | 430/6470 [05:12<1:12:35,  1.39it/s]
4|431|Loss: 0.8050158619880676:   7%|▋         | 431/6470 [05:12<1:12:32,  1.39it/s]
4|432|Loss: 0.7435961365699768:   7%|▋         | 431/6470 [05:13<1:12:32,  1.39it/s]
4|432|Loss: 0.7435961365699768:   7%|▋         | 432/6470 [05:13<1:10:04,  1.44it/s]
4|433|Loss: 0.8214409351348877:   7%|▋         | 432/6470 [05:13<1:10:04,  1.44it/s]
4|433|Loss: 0.8214409351348877:   7%|▋         | 433/6470 [05:14<1:11:50,  1.40it/s]
4|434|Loss: 0.92118901014328:   7%|▋         | 433/6470 [05:14<1:11:50,  1.40it/s]  
4|434|Loss: 0.92118901014328:   7%|▋         | 434/6470 [05:14<1:14:02,  1.36it/s]
4|435|Loss: 0.8046656250953674:   7%|▋         | 434/6470 [05:15<1:14:02,  1.36it/s]
4|435|Loss: 0.8046656250953674:   7%|▋         | 435/6470 [05:15<1:13:42,  1.36it/s]
4|436|Loss: 0.859127938747406:   7%|▋         | 435/6470 [05:16<1:13:42,  1.36it/s] 
4|436|Loss: 0.859127938747406:   7%|▋         | 436/6470 [05:16<1:17:14,  1.30it/s]
4|437|Loss: 0.8460326790809631:   7%|▋         | 436/6470 [05:16<1:17:14,  1.30it/s]
4|437|Loss: 0.8460326790809631:   7%|▋         | 437/6470 [05:17<1:17:26,  1.30it/s]
4|438|Loss: 1.063269853591919:   7%|▋         | 437/6470 [05:17<1:17:26,  1.30it/s] 
4|438|Loss: 1.063269853591919:   7%|▋         | 438/6470 [05:18<1:21:20,  1.24it/s]
4|439|Loss: 0.7829121351242065:   7%|▋         | 438/6470 [05:18<1:21:20,  1.24it/s]
4|439|Loss: 0.7829121351242065:   7%|▋         | 439/6470 [05:19<1:22:20,  1.22it/s]
4|440|Loss: 0.7225876450538635:   7%|▋         | 439/6470 [05:19<1:22:20,  1.22it/s]
4|440|Loss: 0.7225876450538635:   7%|▋         | 440/6470 [05:19<1:19:28,  1.26it/s]
4|441|Loss: 0.9418455958366394:   7%|▋         | 440/6470 [05:20<1:19:28,  1.26it/s]
4|441|Loss: 0.9418455958366394:   7%|▋         | 441/6470 [05:20<1:18:02,  1.29it/s]
4|442|Loss: 0.7688245177268982:   7%|▋         | 441/6470 [05:20<1:18:02,  1.29it/s]
4|442|Loss: 0.7688245177268982:   7%|▋         | 442/6470 [05:21<1:18:28,  1.28it/s]
4|443|Loss: 0.7748050093650818:   7%|▋         | 442/6470 [05:21<1:18:28,  1.28it/s]
4|443|Loss: 0.7748050093650818:   7%|▋         | 443/6470 [05:22<1:19:25,  1.26it/s]
4|444|Loss: 0.836982786655426:   7%|▋         | 443/6470 [05:22<1:19:25,  1.26it/s] 
4|444|Loss: 0.836982786655426:   7%|▋         | 444/6470 [05:22<1:17:42,  1.29it/s]
4|445|Loss: 0.8956642150878906:   7%|▋         | 444/6470 [05:23<1:17:42,  1.29it/s]
4|445|Loss: 0.8956642150878906:   7%|▋         | 445/6470 [05:23<1:17:37,  1.29it/s]
4|446|Loss: 0.8468020558357239:   7%|▋         | 445/6470 [05:24<1:17:37,  1.29it/s]
4|446|Loss: 0.8468020558357239:   7%|▋         | 446/6470 [05:24<1:15:14,  1.33it/s]
4|447|Loss: 0.853860080242157:   7%|▋         | 446/6470 [05:24<1:15:14,  1.33it/s] 
4|447|Loss: 0.853860080242157:   7%|▋         | 447/6470 [05:24<1:13:03,  1.37it/s]
4|448|Loss: 0.8413432836532593:   7%|▋         | 447/6470 [05:25<1:13:03,  1.37it/s]
4|448|Loss: 0.8413432836532593:   7%|▋         | 448/6470 [05:25<1:16:13,  1.32it/s]
4|449|Loss: 0.7980864644050598:   7%|▋         | 448/6470 [05:26<1:16:13,  1.32it/s]
4|449|Loss: 0.7980864644050598:   7%|▋         | 449/6470 [05:26<1:13:30,  1.37it/s]
4|450|Loss: 0.9134168028831482:   7%|▋         | 449/6470 [05:26<1:13:30,  1.37it/s]
4|450|Loss: 0.9134168028831482:   7%|▋         | 450/6470 [05:27<1:13:23,  1.37it/s]
4|451|Loss: 0.9170894026756287:   7%|▋         | 450/6470 [05:27<1:13:23,  1.37it/s]
4|451|Loss: 0.9170894026756287:   7%|▋         | 451/6470 [05:28<1:14:52,  1.34it/s]
4|452|Loss: 0.7722538709640503:   7%|▋         | 451/6470 [05:28<1:14:52,  1.34it/s]
4|452|Loss: 0.7722538709640503:   7%|▋         | 452/6470 [05:28<1:16:15,  1.32it/s]
4|453|Loss: 0.8515596389770508:   7%|▋         | 452/6470 [05:29<1:16:15,  1.32it/s]
4|453|Loss: 0.8515596389770508:   7%|▋         | 453/6470 [05:29<1:18:49,  1.27it/s]
4|454|Loss: 0.6696251034736633:   7%|▋         | 453/6470 [05:30<1:18:49,  1.27it/s]
4|454|Loss: 0.6696251034736633:   7%|▋         | 454/6470 [05:30<1:17:18,  1.30it/s]
4|455|Loss: 0.7908732295036316:   7%|▋         | 454/6470 [05:30<1:17:18,  1.30it/s]
4|455|Loss: 0.7908732295036316:   7%|▋         | 455/6470 [05:31<1:14:15,  1.35it/s]
4|456|Loss: 1.0933561325073242:   7%|▋         | 455/6470 [05:31<1:14:15,  1.35it/s]
4|456|Loss: 1.0933561325073242:   7%|▋         | 456/6470 [05:31<1:07:13,  1.49it/s]
4|457|Loss: 0.9922013878822327:   7%|▋         | 456/6470 [05:31<1:07:13,  1.49it/s]
4|457|Loss: 0.9922013878822327:   7%|▋         | 457/6470 [05:32<1:03:52,  1.57it/s]
4|458|Loss: 0.8568689227104187:   7%|▋         | 457/6470 [05:32<1:03:52,  1.57it/s]
4|458|Loss: 0.8568689227104187:   7%|▋         | 458/6470 [05:32<1:06:38,  1.50it/s]
4|459|Loss: 0.8716555833816528:   7%|▋         | 458/6470 [05:33<1:06:38,  1.50it/s]
4|459|Loss: 0.8716555833816528:   7%|▋         | 459/6470 [05:33<1:06:17,  1.51it/s]
4|460|Loss: 0.8013622164726257:   7%|▋         | 459/6470 [05:33<1:06:17,  1.51it/s]
4|460|Loss: 0.8013622164726257:   7%|▋         | 460/6470 [05:34<1:06:06,  1.52it/s]
4|461|Loss: 0.9548629522323608:   7%|▋         | 460/6470 [05:34<1:06:06,  1.52it/s]
4|461|Loss: 0.9548629522323608:   7%|▋         | 461/6470 [05:34<1:11:27,  1.40it/s]
4|462|Loss: 0.8748124241828918:   7%|▋         | 461/6470 [05:35<1:11:27,  1.40it/s]
4|462|Loss: 0.8748124241828918:   7%|▋         | 462/6470 [05:35<1:10:19,  1.42it/s]
4|463|Loss: 1.0314334630966187:   7%|▋         | 462/6470 [05:36<1:10:19,  1.42it/s]
4|463|Loss: 1.0314334630966187:   7%|▋         | 463/6470 [05:36<1:09:35,  1.44it/s]
4|464|Loss: 0.7816261649131775:   7%|▋         | 463/6470 [05:36<1:09:35,  1.44it/s]
4|464|Loss: 0.7816261649131775:   7%|▋         | 464/6470 [05:37<1:14:54,  1.34it/s]
4|465|Loss: 0.9254269003868103:   7%|▋         | 464/6470 [05:37<1:14:54,  1.34it/s]
4|465|Loss: 0.9254269003868103:   7%|▋         | 465/6470 [05:37<1:14:12,  1.35it/s]
4|466|Loss: 0.8245835900306702:   7%|▋         | 465/6470 [05:38<1:14:12,  1.35it/s]
4|466|Loss: 0.8245835900306702:   7%|▋         | 466/6470 [05:38<1:20:11,  1.25it/s]
4|467|Loss: 0.8089404106140137:   7%|▋         | 466/6470 [05:39<1:20:11,  1.25it/s]
4|467|Loss: 0.8089404106140137:   7%|▋         | 467/6470 [05:39<1:21:15,  1.23it/s]
4|468|Loss: 0.7430999875068665:   7%|▋         | 467/6470 [05:40<1:21:15,  1.23it/s]
4|468|Loss: 0.7430999875068665:   7%|▋         | 468/6470 [05:40<1:22:39,  1.21it/s]
4|469|Loss: 0.6936249136924744:   7%|▋         | 468/6470 [05:40<1:22:39,  1.21it/s]
4|469|Loss: 0.6936249136924744:   7%|▋         | 469/6470 [05:41<1:16:15,  1.31it/s]
4|470|Loss: 0.8272627592086792:   7%|▋         | 469/6470 [05:41<1:16:15,  1.31it/s]
4|470|Loss: 0.8272627592086792:   7%|▋         | 470/6470 [05:41<1:15:17,  1.33it/s]
4|471|Loss: 0.8208094835281372:   7%|▋         | 470/6470 [05:42<1:15:17,  1.33it/s]
4|471|Loss: 0.8208094835281372:   7%|▋         | 471/6470 [05:42<1:12:28,  1.38it/s]
4|472|Loss: 0.9567943811416626:   7%|▋         | 471/6470 [05:42<1:12:28,  1.38it/s]
4|472|Loss: 0.9567943811416626:   7%|▋         | 472/6470 [05:43<1:08:19,  1.46it/s]
4|473|Loss: 0.8345074653625488:   7%|▋         | 472/6470 [05:43<1:08:19,  1.46it/s]
4|473|Loss: 0.8345074653625488:   7%|▋         | 473/6470 [05:43<1:06:13,  1.51it/s]
4|474|Loss: 0.8127003908157349:   7%|▋         | 473/6470 [05:44<1:06:13,  1.51it/s]
4|474|Loss: 0.8127003908157349:   7%|▋         | 474/6470 [05:44<1:06:09,  1.51it/s]
4|475|Loss: 0.7156492471694946:   7%|▋         | 474/6470 [05:44<1:06:09,  1.51it/s]
4|475|Loss: 0.7156492471694946:   7%|▋         | 475/6470 [05:45<1:06:58,  1.49it/s]
4|476|Loss: 0.7719497084617615:   7%|▋         | 475/6470 [05:45<1:06:58,  1.49it/s]
4|476|Loss: 0.7719497084617615:   7%|▋         | 476/6470 [05:45<1:07:01,  1.49it/s]
4|477|Loss: 0.7326511144638062:   7%|▋         | 476/6470 [05:46<1:07:01,  1.49it/s]
4|477|Loss: 0.7326511144638062:   7%|▋         | 477/6470 [05:46<1:05:32,  1.52it/s]
4|478|Loss: 0.806850016117096:   7%|▋         | 477/6470 [05:46<1:05:32,  1.52it/s] 
4|478|Loss: 0.806850016117096:   7%|▋         | 478/6470 [05:47<1:12:33,  1.38it/s]
4|479|Loss: 0.7776129841804504:   7%|▋         | 478/6470 [05:47<1:12:33,  1.38it/s]
4|479|Loss: 0.7776129841804504:   7%|▋         | 479/6470 [05:48<1:15:01,  1.33it/s]
4|480|Loss: 0.8735658526420593:   7%|▋         | 479/6470 [05:48<1:15:01,  1.33it/s]
4|480|Loss: 0.8735658526420593:   7%|▋         | 480/6470 [05:48<1:16:20,  1.31it/s]
4|481|Loss: 0.8030706644058228:   7%|▋         | 480/6470 [05:49<1:16:20,  1.31it/s]
4|481|Loss: 0.8030706644058228:   7%|▋         | 481/6470 [05:49<1:16:21,  1.31it/s]
4|482|Loss: 0.8470517992973328:   7%|▋         | 481/6470 [05:50<1:16:21,  1.31it/s]
4|482|Loss: 0.8470517992973328:   7%|▋         | 482/6470 [05:50<1:17:23,  1.29it/s]
4|483|Loss: 0.7695631980895996:   7%|▋         | 482/6470 [05:50<1:17:23,  1.29it/s]
4|483|Loss: 0.7695631980895996:   7%|▋         | 483/6470 [05:51<1:17:57,  1.28it/s]
4|484|Loss: 0.9125238060951233:   7%|▋         | 483/6470 [05:51<1:17:57,  1.28it/s]
4|484|Loss: 0.9125238060951233:   7%|▋         | 484/6470 [05:52<1:18:13,  1.28it/s]
4|485|Loss: 0.785244345664978:   7%|▋         | 484/6470 [05:52<1:18:13,  1.28it/s] 
4|485|Loss: 0.785244345664978:   7%|▋         | 485/6470 [05:52<1:20:52,  1.23it/s]
4|486|Loss: 0.8624587059020996:   7%|▋         | 485/6470 [05:53<1:20:52,  1.23it/s]
4|486|Loss: 0.8624587059020996:   8%|▊         | 486/6470 [05:53<1:26:55,  1.15it/s]
4|487|Loss: 0.7095404863357544:   8%|▊         | 486/6470 [05:54<1:26:55,  1.15it/s]
4|487|Loss: 0.7095404863357544:   8%|▊         | 487/6470 [05:54<1:22:17,  1.21it/s]
4|488|Loss: 0.7766345143318176:   8%|▊         | 487/6470 [05:55<1:22:17,  1.21it/s]
4|488|Loss: 0.7766345143318176:   8%|▊         | 488/6470 [05:55<1:20:01,  1.25it/s]
4|489|Loss: 0.8230591416358948:   8%|▊         | 488/6470 [05:55<1:20:01,  1.25it/s]
4|489|Loss: 0.8230591416358948:   8%|▊         | 489/6470 [05:56<1:18:22,  1.27it/s]
4|490|Loss: 0.7942183613777161:   8%|▊         | 489/6470 [05:56<1:18:22,  1.27it/s]
4|490|Loss: 0.7942183613777161:   8%|▊         | 490/6470 [05:57<1:21:07,  1.23it/s]
4|491|Loss: 0.94063800573349:   8%|▊         | 490/6470 [05:57<1:21:07,  1.23it/s]  
4|491|Loss: 0.94063800573349:   8%|▊         | 491/6470 [05:57<1:17:05,  1.29it/s]
4|492|Loss: 0.7681202292442322:   8%|▊         | 491/6470 [05:58<1:17:05,  1.29it/s]
4|492|Loss: 0.7681202292442322:   8%|▊         | 492/6470 [05:58<1:20:40,  1.23it/s]
4|493|Loss: 0.903572678565979:   8%|▊         | 492/6470 [05:59<1:20:40,  1.23it/s] 
4|493|Loss: 0.903572678565979:   8%|▊         | 493/6470 [05:59<1:17:38,  1.28it/s]
4|494|Loss: 0.8427572846412659:   8%|▊         | 493/6470 [05:59<1:17:38,  1.28it/s]
4|494|Loss: 0.8427572846412659:   8%|▊         | 494/6470 [06:00<1:17:07,  1.29it/s]
4|495|Loss: 0.7221911549568176:   8%|▊         | 494/6470 [06:00<1:17:07,  1.29it/s]
4|495|Loss: 0.7221911549568176:   8%|▊         | 495/6470 [06:00<1:17:42,  1.28it/s]
4|496|Loss: 0.8194972276687622:   8%|▊         | 495/6470 [06:01<1:17:42,  1.28it/s]
4|496|Loss: 0.8194972276687622:   8%|▊         | 496/6470 [06:01<1:20:39,  1.23it/s]
4|497|Loss: 0.7315803170204163:   8%|▊         | 496/6470 [06:02<1:20:39,  1.23it/s]
4|497|Loss: 0.7315803170204163:   8%|▊         | 497/6470 [06:02<1:18:26,  1.27it/s]
4|498|Loss: 0.8001285791397095:   8%|▊         | 497/6470 [06:02<1:18:26,  1.27it/s]
4|498|Loss: 0.8001285791397095:   8%|▊         | 498/6470 [06:03<1:13:10,  1.36it/s]
4|499|Loss: 0.7801036238670349:   8%|▊         | 498/6470 [06:03<1:13:10,  1.36it/s]
4|499|Loss: 0.7801036238670349:   8%|▊         | 499/6470 [06:04<1:17:25,  1.29it/s]
4|500|Loss: 0.7959476113319397:   8%|▊         | 499/6470 [06:04<1:17:25,  1.29it/s]
4|500|Loss: 0.7959476113319397:   8%|▊         | 500/6470 [06:04<1:16:05,  1.31it/s]
4|501|Loss: 0.8399578332901001:   8%|▊         | 500/6470 [06:05<1:16:05,  1.31it/s]
4|501|Loss: 0.8399578332901001:   8%|▊         | 501/6470 [06:05<1:13:25,  1.35it/s]
4|502|Loss: 0.8599539399147034:   8%|▊         | 501/6470 [06:05<1:13:25,  1.35it/s]
4|502|Loss: 0.8599539399147034:   8%|▊         | 502/6470 [06:05<1:06:55,  1.49it/s]
4|503|Loss: 0.7199640870094299:   8%|▊         | 502/6470 [06:06<1:06:55,  1.49it/s]
4|503|Loss: 0.7199640870094299:   8%|▊         | 503/6470 [06:06<1:05:56,  1.51it/s]
4|504|Loss: 0.7412990927696228:   8%|▊         | 503/6470 [06:06<1:05:56,  1.51it/s]
4|504|Loss: 0.7412990927696228:   8%|▊         | 504/6470 [06:07<1:08:50,  1.44it/s]
4|505|Loss: 0.7230217456817627:   8%|▊         | 504/6470 [06:07<1:08:50,  1.44it/s]
4|505|Loss: 0.7230217456817627:   8%|▊         | 505/6470 [06:08<1:09:25,  1.43it/s]
4|506|Loss: 0.8591516613960266:   8%|▊         | 505/6470 [06:08<1:09:25,  1.43it/s]
4|506|Loss: 0.8591516613960266:   8%|▊         | 506/6470 [06:08<1:14:57,  1.33it/s]
4|507|Loss: 0.776539146900177:   8%|▊         | 506/6470 [06:09<1:14:57,  1.33it/s] 
4|507|Loss: 0.776539146900177:   8%|▊         | 507/6470 [06:09<1:11:07,  1.40it/s]
4|508|Loss: 0.8240678906440735:   8%|▊         | 507/6470 [06:09<1:11:07,  1.40it/s]
4|508|Loss: 0.8240678906440735:   8%|▊         | 508/6470 [06:10<1:07:23,  1.47it/s]
4|509|Loss: 0.8424241542816162:   8%|▊         | 508/6470 [06:10<1:07:23,  1.47it/s]
4|509|Loss: 0.8424241542816162:   8%|▊         | 509/6470 [06:10<1:07:49,  1.46it/s]
4|510|Loss: 0.8176003098487854:   8%|▊         | 509/6470 [06:11<1:07:49,  1.46it/s]
4|510|Loss: 0.8176003098487854:   8%|▊         | 510/6470 [06:11<1:08:17,  1.45it/s]
4|511|Loss: 0.8133257627487183:   8%|▊         | 510/6470 [06:12<1:08:17,  1.45it/s]
4|511|Loss: 0.8133257627487183:   8%|▊         | 511/6470 [06:12<1:14:13,  1.34it/s]
4|512|Loss: 0.9785301685333252:   8%|▊         | 511/6470 [06:12<1:14:13,  1.34it/s]
4|512|Loss: 0.9785301685333252:   8%|▊         | 512/6470 [06:13<1:19:53,  1.24it/s]
4|513|Loss: 0.7971285581588745:   8%|▊         | 512/6470 [06:13<1:19:53,  1.24it/s]
4|513|Loss: 0.7971285581588745:   8%|▊         | 513/6470 [06:14<1:20:18,  1.24it/s]
4|514|Loss: 0.902745246887207:   8%|▊         | 513/6470 [06:14<1:20:18,  1.24it/s] 
4|514|Loss: 0.902745246887207:   8%|▊         | 514/6470 [06:14<1:19:04,  1.26it/s]
4|515|Loss: 0.8958576917648315:   8%|▊         | 514/6470 [06:15<1:19:04,  1.26it/s]
4|515|Loss: 0.8958576917648315:   8%|▊         | 515/6470 [06:15<1:16:32,  1.30it/s]
4|516|Loss: 0.781894862651825:   8%|▊         | 515/6470 [06:16<1:16:32,  1.30it/s] 
4|516|Loss: 0.781894862651825:   8%|▊         | 516/6470 [06:16<1:20:58,  1.23it/s]
4|517|Loss: 0.9336006045341492:   8%|▊         | 516/6470 [06:17<1:20:58,  1.23it/s]
4|517|Loss: 0.9336006045341492:   8%|▊         | 517/6470 [06:17<1:25:21,  1.16it/s]
4|518|Loss: 0.734741747379303:   8%|▊         | 517/6470 [06:17<1:25:21,  1.16it/s] 
4|518|Loss: 0.734741747379303:   8%|▊         | 518/6470 [06:18<1:22:15,  1.21it/s]
4|519|Loss: 0.8255937099456787:   8%|▊         | 518/6470 [06:18<1:22:15,  1.21it/s]
4|519|Loss: 0.8255937099456787:   8%|▊         | 519/6470 [06:19<1:24:59,  1.17it/s]
4|520|Loss: 0.8542973399162292:   8%|▊         | 519/6470 [06:19<1:24:59,  1.17it/s]
4|520|Loss: 0.8542973399162292:   8%|▊         | 520/6470 [06:20<1:24:09,  1.18it/s]
4|521|Loss: 0.9440335631370544:   8%|▊         | 520/6470 [06:20<1:24:09,  1.18it/s]
4|521|Loss: 0.9440335631370544:   8%|▊         | 521/6470 [06:20<1:20:07,  1.24it/s]
4|522|Loss: 0.8007777333259583:   8%|▊         | 521/6470 [06:21<1:20:07,  1.24it/s]
4|522|Loss: 0.8007777333259583:   8%|▊         | 522/6470 [06:21<1:14:50,  1.32it/s]
4|523|Loss: 1.0580552816390991:   8%|▊         | 522/6470 [06:21<1:14:50,  1.32it/s]
4|523|Loss: 1.0580552816390991:   8%|▊         | 523/6470 [06:22<1:15:30,  1.31it/s]
4|524|Loss: 0.9247487783432007:   8%|▊         | 523/6470 [06:22<1:15:30,  1.31it/s]
4|524|Loss: 0.9247487783432007:   8%|▊         | 524/6470 [06:22<1:13:54,  1.34it/s]
4|525|Loss: 0.9406656622886658:   8%|▊         | 524/6470 [06:23<1:13:54,  1.34it/s]
4|525|Loss: 0.9406656622886658:   8%|▊         | 525/6470 [06:23<1:14:09,  1.34it/s]
4|526|Loss: 0.7942370772361755:   8%|▊         | 525/6470 [06:24<1:14:09,  1.34it/s]
4|526|Loss: 0.7942370772361755:   8%|▊         | 526/6470 [06:24<1:15:15,  1.32it/s]
4|527|Loss: 0.7783287167549133:   8%|▊         | 526/6470 [06:24<1:15:15,  1.32it/s]
4|527|Loss: 0.7783287167549133:   8%|▊         | 527/6470 [06:25<1:10:21,  1.41it/s]
4|528|Loss: 0.8024691343307495:   8%|▊         | 527/6470 [06:25<1:10:21,  1.41it/s]
4|528|Loss: 0.8024691343307495:   8%|▊         | 528/6470 [06:25<1:07:08,  1.47it/s]
4|529|Loss: 0.8000686168670654:   8%|▊         | 528/6470 [06:26<1:07:08,  1.47it/s]
4|529|Loss: 0.8000686168670654:   8%|▊         | 529/6470 [06:26<1:07:22,  1.47it/s]
4|530|Loss: 0.870288610458374:   8%|▊         | 529/6470 [06:26<1:07:22,  1.47it/s] 
4|530|Loss: 0.870288610458374:   8%|▊         | 530/6470 [06:27<1:10:00,  1.41it/s]
4|531|Loss: 0.7233029007911682:   8%|▊         | 530/6470 [06:27<1:10:00,  1.41it/s]
4|531|Loss: 0.7233029007911682:   8%|▊         | 531/6470 [06:27<1:06:37,  1.49it/s]
4|532|Loss: 0.7694476246833801:   8%|▊         | 531/6470 [06:28<1:06:37,  1.49it/s]
4|532|Loss: 0.7694476246833801:   8%|▊         | 532/6470 [06:28<1:07:40,  1.46it/s]
4|533|Loss: 0.8256872296333313:   8%|▊         | 532/6470 [06:28<1:07:40,  1.46it/s]
4|533|Loss: 0.8256872296333313:   8%|▊         | 533/6470 [06:29<1:08:26,  1.45it/s]
4|534|Loss: 0.7564436197280884:   8%|▊         | 533/6470 [06:29<1:08:26,  1.45it/s]
4|534|Loss: 0.7564436197280884:   8%|▊         | 534/6470 [06:30<1:15:29,  1.31it/s]
4|535|Loss: 0.793889582157135:   8%|▊         | 534/6470 [06:30<1:15:29,  1.31it/s] 
4|535|Loss: 0.793889582157135:   8%|▊         | 535/6470 [06:30<1:15:11,  1.32it/s]
4|536|Loss: 0.8543096780776978:   8%|▊         | 535/6470 [06:31<1:15:11,  1.32it/s]
4|536|Loss: 0.8543096780776978:   8%|▊         | 536/6470 [06:31<1:12:35,  1.36it/s]
4|537|Loss: 0.894440770149231:   8%|▊         | 536/6470 [06:31<1:12:35,  1.36it/s] 
4|537|Loss: 0.894440770149231:   8%|▊         | 537/6470 [06:32<1:17:35,  1.27it/s]
4|538|Loss: 0.9332941174507141:   8%|▊         | 537/6470 [06:32<1:17:35,  1.27it/s]
4|538|Loss: 0.9332941174507141:   8%|▊         | 538/6470 [06:32<1:11:48,  1.38it/s]
4|539|Loss: 0.832950234413147:   8%|▊         | 538/6470 [06:33<1:11:48,  1.38it/s] 
4|539|Loss: 0.832950234413147:   8%|▊         | 539/6470 [06:33<1:13:50,  1.34it/s]
4|540|Loss: 0.8220130801200867:   8%|▊         | 539/6470 [06:34<1:13:50,  1.34it/s]
4|540|Loss: 0.8220130801200867:   8%|▊         | 540/6470 [06:34<1:14:28,  1.33it/s]
4|541|Loss: 0.6760367155075073:   8%|▊         | 540/6470 [06:34<1:14:28,  1.33it/s]
4|541|Loss: 0.6760367155075073:   8%|▊         | 541/6470 [06:35<1:09:51,  1.41it/s]
4|542|Loss: 0.9540813565254211:   8%|▊         | 541/6470 [06:35<1:09:51,  1.41it/s]
4|542|Loss: 0.9540813565254211:   8%|▊         | 542/6470 [06:35<1:08:37,  1.44it/s]
4|543|Loss: 0.8166855573654175:   8%|▊         | 542/6470 [06:36<1:08:37,  1.44it/s]
4|543|Loss: 0.8166855573654175:   8%|▊         | 543/6470 [06:36<1:10:19,  1.40it/s]
4|544|Loss: 0.8056817054748535:   8%|▊         | 543/6470 [06:36<1:10:19,  1.40it/s]
4|544|Loss: 0.8056817054748535:   8%|▊         | 544/6470 [06:37<1:12:16,  1.37it/s]
4|545|Loss: 0.8265960812568665:   8%|▊         | 544/6470 [06:37<1:12:16,  1.37it/s]
4|545|Loss: 0.8265960812568665:   8%|▊         | 545/6470 [06:38<1:12:57,  1.35it/s]
4|546|Loss: 0.9378337860107422:   8%|▊         | 545/6470 [06:38<1:12:57,  1.35it/s]
4|546|Loss: 0.9378337860107422:   8%|▊         | 546/6470 [06:38<1:17:06,  1.28it/s]
4|547|Loss: 0.6500930190086365:   8%|▊         | 546/6470 [06:39<1:17:06,  1.28it/s]
4|547|Loss: 0.6500930190086365:   8%|▊         | 547/6470 [06:39<1:11:02,  1.39it/s]
4|548|Loss: 0.8749074339866638:   8%|▊         | 547/6470 [06:39<1:11:02,  1.39it/s]
4|548|Loss: 0.8749074339866638:   8%|▊         | 548/6470 [06:40<1:05:10,  1.51it/s]
4|549|Loss: 0.678042471408844:   8%|▊         | 548/6470 [06:40<1:05:10,  1.51it/s] 
4|549|Loss: 0.678042471408844:   8%|▊         | 549/6470 [06:40<1:05:16,  1.51it/s]
4|550|Loss: 0.8369355797767639:   8%|▊         | 549/6470 [06:41<1:05:16,  1.51it/s]
4|550|Loss: 0.8369355797767639:   9%|▊         | 550/6470 [06:41<1:06:31,  1.48it/s]
4|551|Loss: 0.8382163047790527:   9%|▊         | 550/6470 [06:41<1:06:31,  1.48it/s]
4|551|Loss: 0.8382163047790527:   9%|▊         | 551/6470 [06:42<1:12:17,  1.36it/s]
4|552|Loss: 0.7399095296859741:   9%|▊         | 551/6470 [06:42<1:12:17,  1.36it/s]
4|552|Loss: 0.7399095296859741:   9%|▊         | 552/6470 [06:43<1:14:16,  1.33it/s]
4|553|Loss: 0.8476613759994507:   9%|▊         | 552/6470 [06:43<1:14:16,  1.33it/s]
4|553|Loss: 0.8476613759994507:   9%|▊         | 553/6470 [06:43<1:07:54,  1.45it/s]
4|554|Loss: 1.0221126079559326:   9%|▊         | 553/6470 [06:43<1:07:54,  1.45it/s]
4|554|Loss: 1.0221126079559326:   9%|▊         | 554/6470 [06:44<1:06:04,  1.49it/s]
4|555|Loss: 0.7995966672897339:   9%|▊         | 554/6470 [06:44<1:06:04,  1.49it/s]
4|555|Loss: 0.7995966672897339:   9%|▊         | 555/6470 [06:45<1:14:24,  1.32it/s]
4|556|Loss: 0.7994120717048645:   9%|▊         | 555/6470 [06:45<1:14:24,  1.32it/s]
4|556|Loss: 0.7994120717048645:   9%|▊         | 556/6470 [06:45<1:15:59,  1.30it/s]
4|557|Loss: 0.7572888731956482:   9%|▊         | 556/6470 [06:46<1:15:59,  1.30it/s]
4|557|Loss: 0.7572888731956482:   9%|▊         | 557/6470 [06:46<1:17:59,  1.26it/s]
4|558|Loss: 0.8606394529342651:   9%|▊         | 557/6470 [06:47<1:17:59,  1.26it/s]
4|558|Loss: 0.8606394529342651:   9%|▊         | 558/6470 [06:47<1:18:26,  1.26it/s]
4|559|Loss: 0.7298116683959961:   9%|▊         | 558/6470 [06:48<1:18:26,  1.26it/s]
4|559|Loss: 0.7298116683959961:   9%|▊         | 559/6470 [06:48<1:14:24,  1.32it/s]
4|560|Loss: 0.9199391007423401:   9%|▊         | 559/6470 [06:48<1:14:24,  1.32it/s]
4|560|Loss: 0.9199391007423401:   9%|▊         | 560/6470 [06:48<1:11:35,  1.38it/s]
4|561|Loss: 0.6624983549118042:   9%|▊         | 560/6470 [06:49<1:11:35,  1.38it/s]
4|561|Loss: 0.6624983549118042:   9%|▊         | 561/6470 [06:49<1:05:21,  1.51it/s]
4|562|Loss: 0.6030788421630859:   9%|▊         | 561/6470 [06:50<1:05:21,  1.51it/s]
4|562|Loss: 0.6030788421630859:   9%|▊         | 562/6470 [06:50<1:27:44,  1.12it/s]
4|563|Loss: 0.8384138941764832:   9%|▊         | 562/6470 [06:51<1:27:44,  1.12it/s]
4|563|Loss: 0.8384138941764832:   9%|▊         | 563/6470 [06:51<1:28:03,  1.12it/s]
4|564|Loss: 0.8358235359191895:   9%|▊         | 563/6470 [06:52<1:28:03,  1.12it/s]
4|564|Loss: 0.8358235359191895:   9%|▊         | 564/6470 [06:52<1:29:04,  1.11it/s]
4|565|Loss: 0.72484290599823:   9%|▊         | 564/6470 [06:53<1:29:04,  1.11it/s]  
4|565|Loss: 0.72484290599823:   9%|▊         | 565/6470 [06:53<1:22:29,  1.19it/s]
4|566|Loss: 0.6972112655639648:   9%|▊         | 565/6470 [06:53<1:22:29,  1.19it/s]
4|566|Loss: 0.6972112655639648:   9%|▊         | 566/6470 [06:54<1:16:10,  1.29it/s]
4|567|Loss: 0.8307729363441467:   9%|▊         | 566/6470 [06:54<1:16:10,  1.29it/s]
4|567|Loss: 0.8307729363441467:   9%|▉         | 567/6470 [06:54<1:11:52,  1.37it/s]
4|568|Loss: 0.7215715050697327:   9%|▉         | 567/6470 [06:55<1:11:52,  1.37it/s]
4|568|Loss: 0.7215715050697327:   9%|▉         | 568/6470 [06:55<1:19:05,  1.24it/s]
4|569|Loss: 0.7967454195022583:   9%|▉         | 568/6470 [06:56<1:19:05,  1.24it/s]
4|569|Loss: 0.7967454195022583:   9%|▉         | 569/6470 [06:56<1:21:02,  1.21it/s]
4|570|Loss: 0.6419158577919006:   9%|▉         | 569/6470 [06:56<1:21:02,  1.21it/s]
4|570|Loss: 0.6419158577919006:   9%|▉         | 570/6470 [06:57<1:16:20,  1.29it/s]
4|571|Loss: 0.9136756658554077:   9%|▉         | 570/6470 [06:57<1:16:20,  1.29it/s]
4|571|Loss: 0.9136756658554077:   9%|▉         | 571/6470 [06:57<1:08:25,  1.44it/s]
4|572|Loss: 0.9232357144355774:   9%|▉         | 571/6470 [06:58<1:08:25,  1.44it/s]
4|572|Loss: 0.9232357144355774:   9%|▉         | 572/6470 [06:58<1:08:17,  1.44it/s]
4|573|Loss: 0.7442297339439392:   9%|▉         | 572/6470 [06:58<1:08:17,  1.44it/s]
4|573|Loss: 0.7442297339439392:   9%|▉         | 573/6470 [06:59<1:13:58,  1.33it/s]
4|574|Loss: 0.7660320401191711:   9%|▉         | 573/6470 [06:59<1:13:58,  1.33it/s]
4|574|Loss: 0.7660320401191711:   9%|▉         | 574/6470 [07:00<1:14:55,  1.31it/s]
4|575|Loss: 0.7635869383811951:   9%|▉         | 574/6470 [07:00<1:14:55,  1.31it/s]
4|575|Loss: 0.7635869383811951:   9%|▉         | 575/6470 [07:00<1:11:30,  1.37it/s]
4|576|Loss: 0.7692747116088867:   9%|▉         | 575/6470 [07:01<1:11:30,  1.37it/s]
4|576|Loss: 0.7692747116088867:   9%|▉         | 576/6470 [07:01<1:11:24,  1.38it/s]
4|577|Loss: 0.7573522329330444:   9%|▉         | 576/6470 [07:01<1:11:24,  1.38it/s]
4|577|Loss: 0.7573522329330444:   9%|▉         | 577/6470 [07:02<1:16:24,  1.29it/s]
4|578|Loss: 0.875152051448822:   9%|▉         | 577/6470 [07:02<1:16:24,  1.29it/s] 
4|578|Loss: 0.875152051448822:   9%|▉         | 578/6470 [07:03<1:15:43,  1.30it/s]
4|579|Loss: 0.7934293746948242:   9%|▉         | 578/6470 [07:03<1:15:43,  1.30it/s]
4|579|Loss: 0.7934293746948242:   9%|▉         | 579/6470 [07:03<1:19:17,  1.24it/s]
4|580|Loss: 0.7717674970626831:   9%|▉         | 579/6470 [07:04<1:19:17,  1.24it/s]
4|580|Loss: 0.7717674970626831:   9%|▉         | 580/6470 [07:04<1:22:39,  1.19it/s]
4|581|Loss: 0.6313348412513733:   9%|▉         | 580/6470 [07:05<1:22:39,  1.19it/s]
4|581|Loss: 0.6313348412513733:   9%|▉         | 581/6470 [07:05<1:28:35,  1.11it/s]
4|582|Loss: 0.7338296175003052:   9%|▉         | 581/6470 [07:06<1:28:35,  1.11it/s]
4|582|Loss: 0.7338296175003052:   9%|▉         | 582/6470 [07:06<1:23:26,  1.18it/s]
4|583|Loss: 0.8035343289375305:   9%|▉         | 582/6470 [07:07<1:23:26,  1.18it/s]
4|583|Loss: 0.8035343289375305:   9%|▉         | 583/6470 [07:07<1:18:28,  1.25it/s]
4|584|Loss: 0.8509606122970581:   9%|▉         | 583/6470 [07:07<1:18:28,  1.25it/s]
4|584|Loss: 0.8509606122970581:   9%|▉         | 584/6470 [07:07<1:13:49,  1.33it/s]
4|585|Loss: 0.8485448360443115:   9%|▉         | 584/6470 [07:08<1:13:49,  1.33it/s]
4|585|Loss: 0.8485448360443115:   9%|▉         | 585/6470 [07:08<1:13:29,  1.33it/s]
4|586|Loss: 0.9767181277275085:   9%|▉         | 585/6470 [07:09<1:13:29,  1.33it/s]
4|586|Loss: 0.9767181277275085:   9%|▉         | 586/6470 [07:09<1:14:28,  1.32it/s]
4|587|Loss: 0.7450774312019348:   9%|▉         | 586/6470 [07:09<1:14:28,  1.32it/s]
4|587|Loss: 0.7450774312019348:   9%|▉         | 587/6470 [07:10<1:15:20,  1.30it/s]
4|588|Loss: 0.7149876356124878:   9%|▉         | 587/6470 [07:10<1:15:20,  1.30it/s]
4|588|Loss: 0.7149876356124878:   9%|▉         | 588/6470 [07:10<1:09:56,  1.40it/s]
4|589|Loss: 0.7899731397628784:   9%|▉         | 588/6470 [07:11<1:09:56,  1.40it/s]
4|589|Loss: 0.7899731397628784:   9%|▉         | 589/6470 [07:11<1:08:38,  1.43it/s]
4|590|Loss: 0.7098061442375183:   9%|▉         | 589/6470 [07:11<1:08:38,  1.43it/s]
4|590|Loss: 0.7098061442375183:   9%|▉         | 590/6470 [07:12<1:08:58,  1.42it/s]
4|591|Loss: 0.7065982222557068:   9%|▉         | 590/6470 [07:12<1:08:58,  1.42it/s]
4|591|Loss: 0.7065982222557068:   9%|▉         | 591/6470 [07:13<1:14:42,  1.31it/s]
4|592|Loss: 0.8225581049919128:   9%|▉         | 591/6470 [07:13<1:14:42,  1.31it/s]
4|592|Loss: 0.8225581049919128:   9%|▉         | 592/6470 [07:14<1:20:43,  1.21it/s]
4|593|Loss: 0.8965731263160706:   9%|▉         | 592/6470 [07:14<1:20:43,  1.21it/s]
4|593|Loss: 0.8965731263160706:   9%|▉         | 593/6470 [07:14<1:21:42,  1.20it/s]
4|594|Loss: 0.7207799553871155:   9%|▉         | 593/6470 [07:15<1:21:42,  1.20it/s]
4|594|Loss: 0.7207799553871155:   9%|▉         | 594/6470 [07:16<1:27:11,  1.12it/s]
4|595|Loss: 0.8095791339874268:   9%|▉         | 594/6470 [07:16<1:27:11,  1.12it/s]
4|595|Loss: 0.8095791339874268:   9%|▉         | 595/6470 [07:16<1:25:41,  1.14it/s]
4|596|Loss: 0.9018182754516602:   9%|▉         | 595/6470 [07:17<1:25:41,  1.14it/s]
4|596|Loss: 0.9018182754516602:   9%|▉         | 596/6470 [07:17<1:18:19,  1.25it/s]
4|597|Loss: 0.6164109706878662:   9%|▉         | 596/6470 [07:17<1:18:19,  1.25it/s]
4|597|Loss: 0.6164109706878662:   9%|▉         | 597/6470 [07:18<1:19:31,  1.23it/s]
4|598|Loss: 0.8367857933044434:   9%|▉         | 597/6470 [07:18<1:19:31,  1.23it/s]
4|598|Loss: 0.8367857933044434:   9%|▉         | 598/6470 [07:19<1:19:49,  1.23it/s]
4|599|Loss: 0.8976912498474121:   9%|▉         | 598/6470 [07:19<1:19:49,  1.23it/s]
4|599|Loss: 0.8976912498474121:   9%|▉         | 599/6470 [07:19<1:17:18,  1.27it/s]
4|600|Loss: 0.7415202260017395:   9%|▉         | 599/6470 [07:20<1:17:18,  1.27it/s]
4|600|Loss: 0.7415202260017395:   9%|▉         | 600/6470 [07:20<1:21:34,  1.20it/s]
4|601|Loss: 0.8265523314476013:   9%|▉         | 600/6470 [07:21<1:21:34,  1.20it/s]
4|601|Loss: 0.8265523314476013:   9%|▉         | 601/6470 [07:21<1:18:39,  1.24it/s]
4|602|Loss: 0.7541465163230896:   9%|▉         | 601/6470 [07:21<1:18:39,  1.24it/s]
4|602|Loss: 0.7541465163230896:   9%|▉         | 602/6470 [07:22<1:17:01,  1.27it/s]
4|603|Loss: 0.8617604970932007:   9%|▉         | 602/6470 [07:22<1:17:01,  1.27it/s]
4|603|Loss: 0.8617604970932007:   9%|▉         | 603/6470 [07:23<1:18:23,  1.25it/s]
4|604|Loss: 0.7794787883758545:   9%|▉         | 603/6470 [07:23<1:18:23,  1.25it/s]
4|604|Loss: 0.7794787883758545:   9%|▉         | 604/6470 [07:24<1:22:26,  1.19it/s]
4|605|Loss: 0.8229305744171143:   9%|▉         | 604/6470 [07:24<1:22:26,  1.19it/s]
4|605|Loss: 0.8229305744171143:   9%|▉         | 605/6470 [07:24<1:22:28,  1.19it/s]
4|606|Loss: 0.8442875146865845:   9%|▉         | 605/6470 [07:25<1:22:28,  1.19it/s]
4|606|Loss: 0.8442875146865845:   9%|▉         | 606/6470 [07:25<1:21:16,  1.20it/s]
4|607|Loss: 0.849072277545929:   9%|▉         | 606/6470 [07:26<1:21:16,  1.20it/s] 
4|607|Loss: 0.849072277545929:   9%|▉         | 607/6470 [07:26<1:20:06,  1.22it/s]
4|608|Loss: 0.7875626087188721:   9%|▉         | 607/6470 [07:26<1:20:06,  1.22it/s]
4|608|Loss: 0.7875626087188721:   9%|▉         | 608/6470 [07:27<1:16:09,  1.28it/s]
4|609|Loss: 0.7014257311820984:   9%|▉         | 608/6470 [07:27<1:16:09,  1.28it/s]
4|609|Loss: 0.7014257311820984:   9%|▉         | 609/6470 [07:28<1:22:48,  1.18it/s]
4|610|Loss: 0.7880986332893372:   9%|▉         | 609/6470 [07:28<1:22:48,  1.18it/s]
4|610|Loss: 0.7880986332893372:   9%|▉         | 610/6470 [07:28<1:19:33,  1.23it/s]
4|611|Loss: 0.8730776309967041:   9%|▉         | 610/6470 [07:29<1:19:33,  1.23it/s]
4|611|Loss: 0.8730776309967041:   9%|▉         | 611/6470 [07:29<1:14:57,  1.30it/s]
4|612|Loss: 0.8877131342887878:   9%|▉         | 611/6470 [07:30<1:14:57,  1.30it/s]
4|612|Loss: 0.8877131342887878:   9%|▉         | 612/6470 [07:30<1:13:53,  1.32it/s]
4|613|Loss: 0.7848024964332581:   9%|▉         | 612/6470 [07:30<1:13:53,  1.32it/s]
4|613|Loss: 0.7848024964332581:   9%|▉         | 613/6470 [07:31<1:12:33,  1.35it/s]
4|614|Loss: 0.7845082879066467:   9%|▉         | 613/6470 [07:31<1:12:33,  1.35it/s]
4|614|Loss: 0.7845082879066467:   9%|▉         | 614/6470 [07:31<1:12:24,  1.35it/s]
4|615|Loss: 0.9399130940437317:   9%|▉         | 614/6470 [07:32<1:12:24,  1.35it/s]
4|615|Loss: 0.9399130940437317:  10%|▉         | 615/6470 [07:32<1:11:15,  1.37it/s]
4|616|Loss: 0.8377233743667603:  10%|▉         | 615/6470 [07:32<1:11:15,  1.37it/s]
4|616|Loss: 0.8377233743667603:  10%|▉         | 616/6470 [07:33<1:12:33,  1.34it/s]
4|617|Loss: 0.9360916018486023:  10%|▉         | 616/6470 [07:33<1:12:33,  1.34it/s]
4|617|Loss: 0.9360916018486023:  10%|▉         | 617/6470 [07:34<1:14:02,  1.32it/s]
4|618|Loss: 0.6398751139640808:  10%|▉         | 617/6470 [07:34<1:14:02,  1.32it/s]
4|618|Loss: 0.6398751139640808:  10%|▉         | 618/6470 [07:34<1:19:01,  1.23it/s]
4|619|Loss: 0.7422335743904114:  10%|▉         | 618/6470 [07:35<1:19:01,  1.23it/s]
4|619|Loss: 0.7422335743904114:  10%|▉         | 619/6470 [07:35<1:19:46,  1.22it/s]
4|620|Loss: 0.8606327772140503:  10%|▉         | 619/6470 [07:36<1:19:46,  1.22it/s]
4|620|Loss: 0.8606327772140503:  10%|▉         | 620/6470 [07:36<1:16:14,  1.28it/s]
4|621|Loss: 0.8862179517745972:  10%|▉         | 620/6470 [07:36<1:16:14,  1.28it/s]
4|621|Loss: 0.8862179517745972:  10%|▉         | 621/6470 [07:37<1:13:51,  1.32it/s]
4|622|Loss: 0.7704944014549255:  10%|▉         | 621/6470 [07:37<1:13:51,  1.32it/s]
4|622|Loss: 0.7704944014549255:  10%|▉         | 622/6470 [07:37<1:09:45,  1.40it/s]
4|623|Loss: 0.7412528395652771:  10%|▉         | 622/6470 [07:38<1:09:45,  1.40it/s]
4|623|Loss: 0.7412528395652771:  10%|▉         | 623/6470 [07:38<1:12:36,  1.34it/s]
4|624|Loss: 0.7357306480407715:  10%|▉         | 623/6470 [07:39<1:12:36,  1.34it/s]
4|624|Loss: 0.7357306480407715:  10%|▉         | 624/6470 [07:39<1:12:33,  1.34it/s]
4|625|Loss: 0.8052088618278503:  10%|▉         | 624/6470 [07:39<1:12:33,  1.34it/s]
4|625|Loss: 0.8052088618278503:  10%|▉         | 625/6470 [07:40<1:17:31,  1.26it/s]
4|626|Loss: 0.9013731479644775:  10%|▉         | 625/6470 [07:40<1:17:31,  1.26it/s]
4|626|Loss: 0.9013731479644775:  10%|▉         | 626/6470 [07:41<1:17:41,  1.25it/s]
4|627|Loss: 0.8523334264755249:  10%|▉         | 626/6470 [07:41<1:17:41,  1.25it/s]
4|627|Loss: 0.8523334264755249:  10%|▉         | 627/6470 [07:41<1:17:09,  1.26it/s]
4|628|Loss: 1.1265912055969238:  10%|▉         | 627/6470 [07:42<1:17:09,  1.26it/s]
4|628|Loss: 1.1265912055969238:  10%|▉         | 628/6470 [07:42<1:14:14,  1.31it/s]
4|629|Loss: 0.7887195348739624:  10%|▉         | 628/6470 [07:42<1:14:14,  1.31it/s]
4|629|Loss: 0.7887195348739624:  10%|▉         | 629/6470 [07:43<1:13:40,  1.32it/s]
4|630|Loss: 0.9025560021400452:  10%|▉         | 629/6470 [07:43<1:13:40,  1.32it/s]
4|630|Loss: 0.9025560021400452:  10%|▉         | 630/6470 [07:44<1:12:27,  1.34it/s]
4|631|Loss: 0.7701117992401123:  10%|▉         | 630/6470 [07:44<1:12:27,  1.34it/s]
4|631|Loss: 0.7701117992401123:  10%|▉         | 631/6470 [07:44<1:08:18,  1.42it/s]
4|632|Loss: 0.8891488909721375:  10%|▉         | 631/6470 [07:45<1:08:18,  1.42it/s]
4|632|Loss: 0.8891488909721375:  10%|▉         | 632/6470 [07:45<1:11:06,  1.37it/s]
4|633|Loss: 0.9030326008796692:  10%|▉         | 632/6470 [07:45<1:11:06,  1.37it/s]
4|633|Loss: 0.9030326008796692:  10%|▉         | 633/6470 [07:46<1:12:18,  1.35it/s]
4|634|Loss: 1.0031650066375732:  10%|▉         | 633/6470 [07:46<1:12:18,  1.35it/s]
4|634|Loss: 1.0031650066375732:  10%|▉         | 634/6470 [07:46<1:08:04,  1.43it/s]
4|635|Loss: 0.6647927165031433:  10%|▉         | 634/6470 [07:47<1:08:04,  1.43it/s]
4|635|Loss: 0.6647927165031433:  10%|▉         | 635/6470 [07:47<1:06:14,  1.47it/s]
4|636|Loss: 0.8517630100250244:  10%|▉         | 635/6470 [07:47<1:06:14,  1.47it/s]
4|636|Loss: 0.8517630100250244:  10%|▉         | 636/6470 [07:48<1:09:18,  1.40it/s]
4|637|Loss: 0.9365823864936829:  10%|▉         | 636/6470 [07:48<1:09:18,  1.40it/s]
4|637|Loss: 0.9365823864936829:  10%|▉         | 637/6470 [07:49<1:12:06,  1.35it/s]
4|638|Loss: 0.9051210880279541:  10%|▉         | 637/6470 [07:49<1:12:06,  1.35it/s]
4|638|Loss: 0.9051210880279541:  10%|▉         | 638/6470 [07:49<1:15:42,  1.28it/s]
4|639|Loss: 0.9199289679527283:  10%|▉         | 638/6470 [07:50<1:15:42,  1.28it/s]
4|639|Loss: 0.9199289679527283:  10%|▉         | 639/6470 [07:50<1:16:33,  1.27it/s]
4|640|Loss: 0.8618590831756592:  10%|▉         | 639/6470 [07:51<1:16:33,  1.27it/s]
4|640|Loss: 0.8618590831756592:  10%|▉         | 640/6470 [07:51<1:17:31,  1.25it/s]
4|641|Loss: 0.8573430776596069:  10%|▉         | 640/6470 [07:51<1:17:31,  1.25it/s]
4|641|Loss: 0.8573430776596069:  10%|▉         | 641/6470 [07:52<1:16:43,  1.27it/s]
4|642|Loss: 0.6406912207603455:  10%|▉         | 641/6470 [07:52<1:16:43,  1.27it/s]
4|642|Loss: 0.6406912207603455:  10%|▉         | 642/6470 [07:53<1:15:20,  1.29it/s]
4|643|Loss: 0.8017736673355103:  10%|▉         | 642/6470 [07:53<1:15:20,  1.29it/s]
4|643|Loss: 0.8017736673355103:  10%|▉         | 643/6470 [07:53<1:18:50,  1.23it/s]
4|644|Loss: 0.773128867149353:  10%|▉         | 643/6470 [07:54<1:18:50,  1.23it/s] 
4|644|Loss: 0.773128867149353:  10%|▉         | 644/6470 [07:54<1:25:23,  1.14it/s]
4|645|Loss: 0.7698575854301453:  10%|▉         | 644/6470 [07:55<1:25:23,  1.14it/s]
4|645|Loss: 0.7698575854301453:  10%|▉         | 645/6470 [07:55<1:19:10,  1.23it/s]
4|646|Loss: 0.7830331921577454:  10%|▉         | 645/6470 [07:55<1:19:10,  1.23it/s]
4|646|Loss: 0.7830331921577454:  10%|▉         | 646/6470 [07:56<1:12:58,  1.33it/s]
4|647|Loss: 0.858493983745575:  10%|▉         | 646/6470 [07:56<1:12:58,  1.33it/s] 
4|647|Loss: 0.858493983745575:  10%|█         | 647/6470 [07:56<1:10:53,  1.37it/s]
4|648|Loss: 0.7733811140060425:  10%|█         | 647/6470 [07:57<1:10:53,  1.37it/s]
4|648|Loss: 0.7733811140060425:  10%|█         | 648/6470 [07:57<1:11:58,  1.35it/s]
4|649|Loss: 0.8191729784011841:  10%|█         | 648/6470 [07:58<1:11:58,  1.35it/s]
4|649|Loss: 0.8191729784011841:  10%|█         | 649/6470 [07:58<1:13:18,  1.32it/s]
4|650|Loss: 0.778151273727417:  10%|█         | 649/6470 [07:58<1:13:18,  1.32it/s] 
4|650|Loss: 0.778151273727417:  10%|█         | 650/6470 [07:59<1:16:23,  1.27it/s]
4|651|Loss: 0.8559916615486145:  10%|█         | 650/6470 [07:59<1:16:23,  1.27it/s]
4|651|Loss: 0.8559916615486145:  10%|█         | 651/6470 [08:00<1:14:45,  1.30it/s]
4|652|Loss: 0.8025832772254944:  10%|█         | 651/6470 [08:00<1:14:45,  1.30it/s]
4|652|Loss: 0.8025832772254944:  10%|█         | 652/6470 [08:00<1:14:01,  1.31it/s]
4|653|Loss: 0.8702880144119263:  10%|█         | 652/6470 [08:01<1:14:01,  1.31it/s]
4|653|Loss: 0.8702880144119263:  10%|█         | 653/6470 [08:01<1:18:32,  1.23it/s]
4|654|Loss: 0.8705422878265381:  10%|█         | 653/6470 [08:02<1:18:32,  1.23it/s]
4|654|Loss: 0.8705422878265381:  10%|█         | 654/6470 [08:02<1:19:45,  1.22it/s]
4|655|Loss: 0.6502771973609924:  10%|█         | 654/6470 [08:02<1:19:45,  1.22it/s]
4|655|Loss: 0.6502771973609924:  10%|█         | 655/6470 [08:03<1:13:40,  1.32it/s]
4|656|Loss: 0.9970524907112122:  10%|█         | 655/6470 [08:03<1:13:40,  1.32it/s]
4|656|Loss: 0.9970524907112122:  10%|█         | 656/6470 [08:03<1:10:02,  1.38it/s]
4|657|Loss: 0.7770035266876221:  10%|█         | 656/6470 [08:04<1:10:02,  1.38it/s]
4|657|Loss: 0.7770035266876221:  10%|█         | 657/6470 [08:04<1:15:30,  1.28it/s]
4|658|Loss: 0.8473045229911804:  10%|█         | 657/6470 [08:05<1:15:30,  1.28it/s]
4|658|Loss: 0.8473045229911804:  10%|█         | 658/6470 [08:05<1:11:40,  1.35it/s]
4|659|Loss: 0.698764443397522:  10%|█         | 658/6470 [08:05<1:11:40,  1.35it/s] 
4|659|Loss: 0.698764443397522:  10%|█         | 659/6470 [08:06<1:10:38,  1.37it/s]
4|660|Loss: 0.9736298322677612:  10%|█         | 659/6470 [08:06<1:10:38,  1.37it/s]
4|660|Loss: 0.9736298322677612:  10%|█         | 660/6470 [08:07<1:15:40,  1.28it/s]
4|661|Loss: 0.7186640501022339:  10%|█         | 660/6470 [08:07<1:15:40,  1.28it/s]
4|661|Loss: 0.7186640501022339:  10%|█         | 661/6470 [08:07<1:17:09,  1.25it/s]
4|662|Loss: 0.6785342693328857:  10%|█         | 661/6470 [08:08<1:17:09,  1.25it/s]
4|662|Loss: 0.6785342693328857:  10%|█         | 662/6470 [08:08<1:16:16,  1.27it/s]
4|663|Loss: 0.8243581056594849:  10%|█         | 662/6470 [08:09<1:16:16,  1.27it/s]
4|663|Loss: 0.8243581056594849:  10%|█         | 663/6470 [08:09<1:15:00,  1.29it/s]
4|664|Loss: 0.8390544652938843:  10%|█         | 663/6470 [08:09<1:15:00,  1.29it/s]
4|664|Loss: 0.8390544652938843:  10%|█         | 664/6470 [08:10<1:14:21,  1.30it/s]
4|665|Loss: 0.9648758769035339:  10%|█         | 664/6470 [08:10<1:14:21,  1.30it/s]
4|665|Loss: 0.9648758769035339:  10%|█         | 665/6470 [08:10<1:09:04,  1.40it/s]
4|666|Loss: 0.8227190971374512:  10%|█         | 665/6470 [08:11<1:09:04,  1.40it/s]
4|666|Loss: 0.8227190971374512:  10%|█         | 666/6470 [08:11<1:05:25,  1.48it/s]
4|667|Loss: 1.0386830568313599:  10%|█         | 666/6470 [08:11<1:05:25,  1.48it/s]
4|667|Loss: 1.0386830568313599:  10%|█         | 667/6470 [08:12<1:06:59,  1.44it/s]
4|668|Loss: 0.8109737634658813:  10%|█         | 667/6470 [08:12<1:06:59,  1.44it/s]
4|668|Loss: 0.8109737634658813:  10%|█         | 668/6470 [08:12<1:02:15,  1.55it/s]
4|669|Loss: 0.745077908039093:  10%|█         | 668/6470 [08:12<1:02:15,  1.55it/s] 
4|669|Loss: 0.745077908039093:  10%|█         | 669/6470 [08:13<1:03:38,  1.52it/s]
4|670|Loss: 0.7730550765991211:  10%|█         | 669/6470 [08:13<1:03:38,  1.52it/s]
4|670|Loss: 0.7730550765991211:  10%|█         | 670/6470 [08:13<1:05:12,  1.48it/s]
4|671|Loss: 0.8638625741004944:  10%|█         | 670/6470 [08:14<1:05:12,  1.48it/s]
4|671|Loss: 0.8638625741004944:  10%|█         | 671/6470 [08:14<1:08:13,  1.42it/s]
4|672|Loss: 0.6967957615852356:  10%|█         | 671/6470 [08:15<1:08:13,  1.42it/s]
4|672|Loss: 0.6967957615852356:  10%|█         | 672/6470 [08:15<1:04:58,  1.49it/s]
4|673|Loss: 0.7989828586578369:  10%|█         | 672/6470 [08:15<1:04:58,  1.49it/s]
4|673|Loss: 0.7989828586578369:  10%|█         | 673/6470 [08:15<1:01:01,  1.58it/s]
4|674|Loss: 0.8715145587921143:  10%|█         | 673/6470 [08:16<1:01:01,  1.58it/s]
4|674|Loss: 0.8715145587921143:  10%|█         | 674/6470 [08:16<1:01:22,  1.57it/s]
4|675|Loss: 0.9008973836898804:  10%|█         | 674/6470 [08:16<1:01:22,  1.57it/s]
4|675|Loss: 0.9008973836898804:  10%|█         | 675/6470 [08:17<1:03:28,  1.52it/s]
4|676|Loss: 0.8646172285079956:  10%|█         | 675/6470 [08:17<1:03:28,  1.52it/s]
4|676|Loss: 0.8646172285079956:  10%|█         | 676/6470 [08:17<1:03:08,  1.53it/s]
4|677|Loss: 0.8977530598640442:  10%|█         | 676/6470 [08:18<1:03:08,  1.53it/s]
4|677|Loss: 0.8977530598640442:  10%|█         | 677/6470 [08:18<1:06:03,  1.46it/s]
4|678|Loss: 0.8286231756210327:  10%|█         | 677/6470 [08:19<1:06:03,  1.46it/s]
4|678|Loss: 0.8286231756210327:  10%|█         | 678/6470 [08:19<1:09:33,  1.39it/s]
4|679|Loss: 0.8540062308311462:  10%|█         | 678/6470 [08:19<1:09:33,  1.39it/s]
4|679|Loss: 0.8540062308311462:  10%|█         | 679/6470 [08:20<1:11:08,  1.36it/s]
4|680|Loss: 0.8759649395942688:  10%|█         | 679/6470 [08:20<1:11:08,  1.36it/s]
4|680|Loss: 0.8759649395942688:  11%|█         | 680/6470 [08:21<1:12:59,  1.32it/s]
4|681|Loss: 0.836959183216095:  11%|█         | 680/6470 [08:21<1:12:59,  1.32it/s] 
4|681|Loss: 0.836959183216095:  11%|█         | 681/6470 [08:21<1:16:03,  1.27it/s]
4|682|Loss: 0.8629355430603027:  11%|█         | 681/6470 [08:22<1:16:03,  1.27it/s]
4|682|Loss: 0.8629355430603027:  11%|█         | 682/6470 [08:22<1:09:17,  1.39it/s]
4|683|Loss: 0.830893874168396:  11%|█         | 682/6470 [08:22<1:09:17,  1.39it/s] 
4|683|Loss: 0.830893874168396:  11%|█         | 683/6470 [08:23<1:12:24,  1.33it/s]
4|684|Loss: 0.9914491176605225:  11%|█         | 683/6470 [08:23<1:12:24,  1.33it/s]
4|684|Loss: 0.9914491176605225:  11%|█         | 684/6470 [08:23<1:08:12,  1.41it/s]
4|685|Loss: 0.7902476191520691:  11%|█         | 684/6470 [08:24<1:08:12,  1.41it/s]
4|685|Loss: 0.7902476191520691:  11%|█         | 685/6470 [08:24<1:08:39,  1.40it/s]
4|686|Loss: 0.7150614261627197:  11%|█         | 685/6470 [08:24<1:08:39,  1.40it/s]
4|686|Loss: 0.7150614261627197:  11%|█         | 686/6470 [08:25<1:05:17,  1.48it/s]
4|687|Loss: 0.9797612428665161:  11%|█         | 686/6470 [08:25<1:05:17,  1.48it/s]
4|687|Loss: 0.9797612428665161:  11%|█         | 687/6470 [08:25<1:01:45,  1.56it/s]
4|688|Loss: 0.8673131465911865:  11%|█         | 687/6470 [08:26<1:01:45,  1.56it/s]
4|688|Loss: 0.8673131465911865:  11%|█         | 688/6470 [08:26<1:03:55,  1.51it/s]
4|689|Loss: 0.8046374320983887:  11%|█         | 688/6470 [08:26<1:03:55,  1.51it/s]
4|689|Loss: 0.8046374320983887:  11%|█         | 689/6470 [08:27<1:07:58,  1.42it/s]
4|690|Loss: 1.0126138925552368:  11%|█         | 689/6470 [08:27<1:07:58,  1.42it/s]
4|690|Loss: 1.0126138925552368:  11%|█         | 690/6470 [08:27<1:06:40,  1.44it/s]
4|691|Loss: 0.713657557964325:  11%|█         | 690/6470 [08:28<1:06:40,  1.44it/s] 
4|691|Loss: 0.713657557964325:  11%|█         | 691/6470 [08:28<1:08:40,  1.40it/s]
4|692|Loss: 0.7920252680778503:  11%|█         | 691/6470 [08:29<1:08:40,  1.40it/s]
4|692|Loss: 0.7920252680778503:  11%|█         | 692/6470 [08:29<1:05:52,  1.46it/s]
4|693|Loss: 0.8844205737113953:  11%|█         | 692/6470 [08:29<1:05:52,  1.46it/s]
4|693|Loss: 0.8844205737113953:  11%|█         | 693/6470 [08:30<1:06:53,  1.44it/s]
4|694|Loss: 1.0468329191207886:  11%|█         | 693/6470 [08:30<1:06:53,  1.44it/s]
4|694|Loss: 1.0468329191207886:  11%|█         | 694/6470 [08:30<1:09:12,  1.39it/s]
4|695|Loss: 0.8337588310241699:  11%|█         | 694/6470 [08:31<1:09:12,  1.39it/s]
4|695|Loss: 0.8337588310241699:  11%|█         | 695/6470 [08:31<1:11:12,  1.35it/s]
4|696|Loss: 0.7644081115722656:  11%|█         | 695/6470 [08:32<1:11:12,  1.35it/s]
4|696|Loss: 0.7644081115722656:  11%|█         | 696/6470 [08:32<1:12:38,  1.32it/s]
4|697|Loss: 0.8301928043365479:  11%|█         | 696/6470 [08:32<1:12:38,  1.32it/s]
4|697|Loss: 0.8301928043365479:  11%|█         | 697/6470 [08:33<1:15:57,  1.27it/s]
4|698|Loss: 0.8337834477424622:  11%|█         | 697/6470 [08:33<1:15:57,  1.27it/s]
4|698|Loss: 0.8337834477424622:  11%|█         | 698/6470 [08:34<1:19:49,  1.21it/s]
4|699|Loss: 0.8276897668838501:  11%|█         | 698/6470 [08:34<1:19:49,  1.21it/s]
4|699|Loss: 0.8276897668838501:  11%|█         | 699/6470 [08:34<1:18:46,  1.22it/s]
4|700|Loss: 0.8950464725494385:  11%|█         | 699/6470 [08:35<1:18:46,  1.22it/s]
4|700|Loss: 0.8950464725494385:  11%|█         | 700/6470 [08:35<1:13:41,  1.30it/s]
4|701|Loss: 0.5650308728218079:  11%|█         | 700/6470 [08:35<1:13:41,  1.30it/s]
4|701|Loss: 0.5650308728218079:  11%|█         | 701/6470 [08:36<1:12:56,  1.32it/s]
4|702|Loss: 0.8743166327476501:  11%|█         | 701/6470 [08:36<1:12:56,  1.32it/s]
4|702|Loss: 0.8743166327476501:  11%|█         | 702/6470 [08:37<1:10:54,  1.36it/s]
4|703|Loss: 0.9113234281539917:  11%|█         | 702/6470 [08:37<1:10:54,  1.36it/s]
4|703|Loss: 0.9113234281539917:  11%|█         | 703/6470 [08:37<1:14:29,  1.29it/s]
4|704|Loss: 0.6906707882881165:  11%|█         | 703/6470 [08:38<1:14:29,  1.29it/s]
4|704|Loss: 0.6906707882881165:  11%|█         | 704/6470 [08:38<1:17:08,  1.25it/s]
4|705|Loss: 0.7270458340644836:  11%|█         | 704/6470 [08:39<1:17:08,  1.25it/s]
4|705|Loss: 0.7270458340644836:  11%|█         | 705/6470 [08:39<1:17:12,  1.24it/s]
4|706|Loss: 0.8274015784263611:  11%|█         | 705/6470 [08:39<1:17:12,  1.24it/s]
4|706|Loss: 0.8274015784263611:  11%|█         | 706/6470 [08:40<1:16:07,  1.26it/s]
4|707|Loss: 0.8915519118309021:  11%|█         | 706/6470 [08:40<1:16:07,  1.26it/s]
4|707|Loss: 0.8915519118309021:  11%|█         | 707/6470 [08:41<1:15:02,  1.28it/s]
4|708|Loss: 0.8284364938735962:  11%|█         | 707/6470 [08:41<1:15:02,  1.28it/s]
4|708|Loss: 0.8284364938735962:  11%|█         | 708/6470 [08:41<1:15:01,  1.28it/s]
4|709|Loss: 0.7838967442512512:  11%|█         | 708/6470 [08:42<1:15:01,  1.28it/s]
4|709|Loss: 0.7838967442512512:  11%|█         | 709/6470 [08:42<1:12:15,  1.33it/s]
4|710|Loss: 0.7924432158470154:  11%|█         | 709/6470 [08:43<1:12:15,  1.33it/s]
4|710|Loss: 0.7924432158470154:  11%|█         | 710/6470 [08:43<1:16:03,  1.26it/s]
4|711|Loss: 0.8869604468345642:  11%|█         | 710/6470 [08:43<1:16:03,  1.26it/s]
4|711|Loss: 0.8869604468345642:  11%|█         | 711/6470 [08:44<1:14:37,  1.29it/s]
4|712|Loss: 0.8060886859893799:  11%|█         | 711/6470 [08:44<1:14:37,  1.29it/s]
4|712|Loss: 0.8060886859893799:  11%|█         | 712/6470 [08:44<1:14:31,  1.29it/s]
4|713|Loss: 0.8129241466522217:  11%|█         | 712/6470 [08:45<1:14:31,  1.29it/s]
4|713|Loss: 0.8129241466522217:  11%|█         | 713/6470 [08:45<1:09:40,  1.38it/s]
4|714|Loss: 0.9493485689163208:  11%|█         | 713/6470 [08:45<1:09:40,  1.38it/s]
4|714|Loss: 0.9493485689163208:  11%|█         | 714/6470 [08:46<1:13:36,  1.30it/s]
4|715|Loss: 0.7813378572463989:  11%|█         | 714/6470 [08:46<1:13:36,  1.30it/s]
4|715|Loss: 0.7813378572463989:  11%|█         | 715/6470 [08:47<1:12:24,  1.32it/s]
4|716|Loss: 0.7944775223731995:  11%|█         | 715/6470 [08:47<1:12:24,  1.32it/s]
4|716|Loss: 0.7944775223731995:  11%|█         | 716/6470 [08:47<1:11:53,  1.33it/s]
4|717|Loss: 0.8057087659835815:  11%|█         | 716/6470 [08:48<1:11:53,  1.33it/s]
4|717|Loss: 0.8057087659835815:  11%|█         | 717/6470 [08:48<1:10:29,  1.36it/s]
4|718|Loss: 0.8836030960083008:  11%|█         | 717/6470 [08:49<1:10:29,  1.36it/s]
4|718|Loss: 0.8836030960083008:  11%|█         | 718/6470 [08:49<1:12:01,  1.33it/s]
4|719|Loss: 0.7708624005317688:  11%|█         | 718/6470 [08:49<1:12:01,  1.33it/s]
4|719|Loss: 0.7708624005317688:  11%|█         | 719/6470 [08:49<1:07:49,  1.41it/s]
4|720|Loss: 0.8123181462287903:  11%|█         | 719/6470 [08:50<1:07:49,  1.41it/s]
4|720|Loss: 0.8123181462287903:  11%|█         | 720/6470 [08:50<1:06:17,  1.45it/s]
4|721|Loss: 0.8685588240623474:  11%|█         | 720/6470 [08:51<1:06:17,  1.45it/s]
4|721|Loss: 0.8685588240623474:  11%|█         | 721/6470 [08:51<1:05:55,  1.45it/s]
4|722|Loss: 0.7141537070274353:  11%|█         | 721/6470 [08:51<1:05:55,  1.45it/s]
4|722|Loss: 0.7141537070274353:  11%|█         | 722/6470 [08:52<1:06:05,  1.45it/s]
4|723|Loss: 0.7519059181213379:  11%|█         | 722/6470 [08:52<1:06:05,  1.45it/s]
4|723|Loss: 0.7519059181213379:  11%|█         | 723/6470 [08:52<1:05:43,  1.46it/s]
4|724|Loss: 0.5512606501579285:  11%|█         | 723/6470 [08:53<1:05:43,  1.46it/s]
4|724|Loss: 0.5512606501579285:  11%|█         | 724/6470 [08:53<1:07:58,  1.41it/s]
4|725|Loss: 0.831900417804718:  11%|█         | 724/6470 [08:53<1:07:58,  1.41it/s] 
4|725|Loss: 0.831900417804718:  11%|█         | 725/6470 [08:54<1:08:08,  1.41it/s]
4|726|Loss: 0.8883132934570312:  11%|█         | 725/6470 [08:54<1:08:08,  1.41it/s]
4|726|Loss: 0.8883132934570312:  11%|█         | 726/6470 [08:55<1:11:33,  1.34it/s]
4|727|Loss: 0.7562201619148254:  11%|█         | 726/6470 [08:55<1:11:33,  1.34it/s]
4|727|Loss: 0.7562201619148254:  11%|█         | 727/6470 [08:55<1:09:22,  1.38it/s]
4|728|Loss: 0.8108153343200684:  11%|█         | 727/6470 [08:56<1:09:22,  1.38it/s]
4|728|Loss: 0.8108153343200684:  11%|█▏        | 728/6470 [08:56<1:07:33,  1.42it/s]
4|729|Loss: 0.8011128902435303:  11%|█▏        | 728/6470 [08:56<1:07:33,  1.42it/s]
4|729|Loss: 0.8011128902435303:  11%|█▏        | 729/6470 [08:57<1:06:25,  1.44it/s]
4|730|Loss: 0.7921732664108276:  11%|█▏        | 729/6470 [08:57<1:06:25,  1.44it/s]
4|730|Loss: 0.7921732664108276:  11%|█▏        | 730/6470 [08:57<1:02:57,  1.52it/s]
4|731|Loss: 0.901670515537262:  11%|█▏        | 730/6470 [08:57<1:02:57,  1.52it/s] 
4|731|Loss: 0.901670515537262:  11%|█▏        | 731/6470 [08:58<1:01:45,  1.55it/s]
4|732|Loss: 0.8792492747306824:  11%|█▏        | 731/6470 [08:58<1:01:45,  1.55it/s]
4|732|Loss: 0.8792492747306824:  11%|█▏        | 732/6470 [08:59<1:07:29,  1.42it/s]
4|733|Loss: 0.8580673336982727:  11%|█▏        | 732/6470 [08:59<1:07:29,  1.42it/s]
4|733|Loss: 0.8580673336982727:  11%|█▏        | 733/6470 [08:59<1:04:59,  1.47it/s]
4|734|Loss: 0.8820086717605591:  11%|█▏        | 733/6470 [09:00<1:04:59,  1.47it/s]
4|734|Loss: 0.8820086717605591:  11%|█▏        | 734/6470 [09:00<1:04:05,  1.49it/s]
4|735|Loss: 0.9058458805084229:  11%|█▏        | 734/6470 [09:00<1:04:05,  1.49it/s]
4|735|Loss: 0.9058458805084229:  11%|█▏        | 735/6470 [09:00<1:04:25,  1.48it/s]
4|736|Loss: 0.7402970790863037:  11%|█▏        | 735/6470 [09:01<1:04:25,  1.48it/s]
4|736|Loss: 0.7402970790863037:  11%|█▏        | 736/6470 [09:01<1:03:07,  1.51it/s]
4|737|Loss: 0.7683690190315247:  11%|█▏        | 736/6470 [09:02<1:03:07,  1.51it/s]
4|737|Loss: 0.7683690190315247:  11%|█▏        | 737/6470 [09:02<1:05:56,  1.45it/s]
4|738|Loss: 0.6784674525260925:  11%|█▏        | 737/6470 [09:02<1:05:56,  1.45it/s]
4|738|Loss: 0.6784674525260925:  11%|█▏        | 738/6470 [09:03<1:05:06,  1.47it/s]
4|739|Loss: 0.8939290642738342:  11%|█▏        | 738/6470 [09:03<1:05:06,  1.47it/s]
4|739|Loss: 0.8939290642738342:  11%|█▏        | 739/6470 [09:03<1:05:11,  1.47it/s]
4|740|Loss: 0.7458939552307129:  11%|█▏        | 739/6470 [09:04<1:05:11,  1.47it/s]
4|740|Loss: 0.7458939552307129:  11%|█▏        | 740/6470 [09:04<1:05:58,  1.45it/s]
4|741|Loss: 0.8029277324676514:  11%|█▏        | 740/6470 [09:04<1:05:58,  1.45it/s]
4|741|Loss: 0.8029277324676514:  11%|█▏        | 741/6470 [09:05<1:08:55,  1.39it/s]
4|742|Loss: 0.9605239033699036:  11%|█▏        | 741/6470 [09:05<1:08:55,  1.39it/s]
4|742|Loss: 0.9605239033699036:  11%|█▏        | 742/6470 [09:06<1:13:36,  1.30it/s]
4|743|Loss: 0.9457800984382629:  11%|█▏        | 742/6470 [09:06<1:13:36,  1.30it/s]
4|743|Loss: 0.9457800984382629:  11%|█▏        | 743/6470 [09:06<1:07:37,  1.41it/s]
4|744|Loss: 0.8748059272766113:  11%|█▏        | 743/6470 [09:07<1:07:37,  1.41it/s]
4|744|Loss: 0.8748059272766113:  11%|█▏        | 744/6470 [09:07<1:07:05,  1.42it/s]
4|745|Loss: 0.7220034599304199:  11%|█▏        | 744/6470 [09:07<1:07:05,  1.42it/s]
4|745|Loss: 0.7220034599304199:  12%|█▏        | 745/6470 [09:07<1:04:36,  1.48it/s]
4|746|Loss: 0.8505347371101379:  12%|█▏        | 745/6470 [09:08<1:04:36,  1.48it/s]
4|746|Loss: 0.8505347371101379:  12%|█▏        | 746/6470 [09:08<1:09:21,  1.38it/s]
4|747|Loss: 1.0346406698226929:  12%|█▏        | 746/6470 [09:09<1:09:21,  1.38it/s]
4|747|Loss: 1.0346406698226929:  12%|█▏        | 747/6470 [09:09<1:14:56,  1.27it/s]
4|748|Loss: 0.8197357654571533:  12%|█▏        | 747/6470 [09:10<1:14:56,  1.27it/s]
4|748|Loss: 0.8197357654571533:  12%|█▏        | 748/6470 [09:10<1:10:29,  1.35it/s]
4|749|Loss: 0.9792847633361816:  12%|█▏        | 748/6470 [09:10<1:10:29,  1.35it/s]
4|749|Loss: 0.9792847633361816:  12%|█▏        | 749/6470 [09:10<1:04:39,  1.47it/s]
4|750|Loss: 0.7588614821434021:  12%|█▏        | 749/6470 [09:11<1:04:39,  1.47it/s]
4|750|Loss: 0.7588614821434021:  12%|█▏        | 750/6470 [09:11<1:04:17,  1.48it/s]
4|751|Loss: 0.8661360144615173:  12%|█▏        | 750/6470 [09:12<1:04:17,  1.48it/s]
4|751|Loss: 0.8661360144615173:  12%|█▏        | 751/6470 [09:12<1:07:58,  1.40it/s]
4|752|Loss: 0.9529563784599304:  12%|█▏        | 751/6470 [09:12<1:07:58,  1.40it/s]
4|752|Loss: 0.9529563784599304:  12%|█▏        | 752/6470 [09:12<1:04:55,  1.47it/s]
4|753|Loss: 0.7504535913467407:  12%|█▏        | 752/6470 [09:13<1:04:55,  1.47it/s]
4|753|Loss: 0.7504535913467407:  12%|█▏        | 753/6470 [09:13<1:05:43,  1.45it/s]
4|754|Loss: 0.9063297510147095:  12%|█▏        | 753/6470 [09:14<1:05:43,  1.45it/s]
4|754|Loss: 0.9063297510147095:  12%|█▏        | 754/6470 [09:14<1:12:28,  1.31it/s]
4|755|Loss: 0.8285056948661804:  12%|█▏        | 754/6470 [09:15<1:12:28,  1.31it/s]
4|755|Loss: 0.8285056948661804:  12%|█▏        | 755/6470 [09:15<1:12:41,  1.31it/s]
4|756|Loss: 0.8736581206321716:  12%|█▏        | 755/6470 [09:15<1:12:41,  1.31it/s]
4|756|Loss: 0.8736581206321716:  12%|█▏        | 756/6470 [09:16<1:15:43,  1.26it/s]
4|757|Loss: 0.7980642914772034:  12%|█▏        | 756/6470 [09:16<1:15:43,  1.26it/s]
4|757|Loss: 0.7980642914772034:  12%|█▏        | 757/6470 [09:17<1:15:25,  1.26it/s]
4|758|Loss: 0.9052713513374329:  12%|█▏        | 757/6470 [09:17<1:15:25,  1.26it/s]
4|758|Loss: 0.9052713513374329:  12%|█▏        | 758/6470 [09:17<1:15:33,  1.26it/s]
4|759|Loss: 0.7211118936538696:  12%|█▏        | 758/6470 [09:18<1:15:33,  1.26it/s]
4|759|Loss: 0.7211118936538696:  12%|█▏        | 759/6470 [09:18<1:15:09,  1.27it/s]
4|760|Loss: 0.8622424006462097:  12%|█▏        | 759/6470 [09:19<1:15:09,  1.27it/s]
4|760|Loss: 0.8622424006462097:  12%|█▏        | 760/6470 [09:19<1:17:50,  1.22it/s]
4|761|Loss: 0.75275719165802:  12%|█▏        | 760/6470 [09:20<1:17:50,  1.22it/s]  
4|761|Loss: 0.75275719165802:  12%|█▏        | 761/6470 [09:20<1:20:40,  1.18it/s]
4|762|Loss: 0.9464419484138489:  12%|█▏        | 761/6470 [09:20<1:20:40,  1.18it/s]
4|762|Loss: 0.9464419484138489:  12%|█▏        | 762/6470 [09:21<1:19:07,  1.20it/s]
4|763|Loss: 0.8677749037742615:  12%|█▏        | 762/6470 [09:21<1:19:07,  1.20it/s]
4|763|Loss: 0.8677749037742615:  12%|█▏        | 763/6470 [09:21<1:15:37,  1.26it/s]
4|764|Loss: 0.8606675863265991:  12%|█▏        | 763/6470 [09:22<1:15:37,  1.26it/s]
4|764|Loss: 0.8606675863265991:  12%|█▏        | 764/6470 [09:22<1:15:07,  1.27it/s]
4|765|Loss: 0.8550167083740234:  12%|█▏        | 764/6470 [09:23<1:15:07,  1.27it/s]
4|765|Loss: 0.8550167083740234:  12%|█▏        | 765/6470 [09:23<1:16:32,  1.24it/s]
4|766|Loss: 0.854374349117279:  12%|█▏        | 765/6470 [09:23<1:16:32,  1.24it/s] 
4|766|Loss: 0.854374349117279:  12%|█▏        | 766/6470 [09:24<1:14:31,  1.28it/s]
4|767|Loss: 0.8043537139892578:  12%|█▏        | 766/6470 [09:24<1:14:31,  1.28it/s]
4|767|Loss: 0.8043537139892578:  12%|█▏        | 767/6470 [09:24<1:06:58,  1.42it/s]
4|768|Loss: 0.7465065717697144:  12%|█▏        | 767/6470 [09:25<1:06:58,  1.42it/s]
4|768|Loss: 0.7465065717697144:  12%|█▏        | 768/6470 [09:25<1:10:18,  1.35it/s]
4|769|Loss: 0.7852040529251099:  12%|█▏        | 768/6470 [09:26<1:10:18,  1.35it/s]
4|769|Loss: 0.7852040529251099:  12%|█▏        | 769/6470 [09:26<1:12:10,  1.32it/s]
4|770|Loss: 0.9440348148345947:  12%|█▏        | 769/6470 [09:26<1:12:10,  1.32it/s]
4|770|Loss: 0.9440348148345947:  12%|█▏        | 770/6470 [09:27<1:16:19,  1.24it/s]
4|771|Loss: 0.8851273655891418:  12%|█▏        | 770/6470 [09:27<1:16:19,  1.24it/s]
4|771|Loss: 0.8851273655891418:  12%|█▏        | 771/6470 [09:28<1:16:07,  1.25it/s]
4|772|Loss: 0.8569079637527466:  12%|█▏        | 771/6470 [09:28<1:16:07,  1.25it/s]
4|772|Loss: 0.8569079637527466:  12%|█▏        | 772/6470 [09:29<1:17:59,  1.22it/s]
4|773|Loss: 0.976932942867279:  12%|█▏        | 772/6470 [09:29<1:17:59,  1.22it/s] 
4|773|Loss: 0.976932942867279:  12%|█▏        | 773/6470 [09:29<1:20:35,  1.18it/s]
4|774|Loss: 0.6686670780181885:  12%|█▏        | 773/6470 [09:30<1:20:35,  1.18it/s]
4|774|Loss: 0.6686670780181885:  12%|█▏        | 774/6470 [09:30<1:17:00,  1.23it/s]
4|775|Loss: 0.6619534492492676:  12%|█▏        | 774/6470 [09:31<1:17:00,  1.23it/s]
4|775|Loss: 0.6619534492492676:  12%|█▏        | 775/6470 [09:31<1:19:11,  1.20it/s]
4|776|Loss: 0.8126981854438782:  12%|█▏        | 775/6470 [09:31<1:19:11,  1.20it/s]
4|776|Loss: 0.8126981854438782:  12%|█▏        | 776/6470 [09:32<1:15:54,  1.25it/s]
4|777|Loss: 0.8861420750617981:  12%|█▏        | 776/6470 [09:32<1:15:54,  1.25it/s]
4|777|Loss: 0.8861420750617981:  12%|█▏        | 777/6470 [09:33<1:15:45,  1.25it/s]
4|778|Loss: 0.9489269852638245:  12%|█▏        | 777/6470 [09:33<1:15:45,  1.25it/s]
4|778|Loss: 0.9489269852638245:  12%|█▏        | 778/6470 [09:33<1:17:21,  1.23it/s]
4|779|Loss: 0.9990827441215515:  12%|█▏        | 778/6470 [09:34<1:17:21,  1.23it/s]
4|779|Loss: 0.9990827441215515:  12%|█▏        | 779/6470 [09:34<1:14:34,  1.27it/s]
4|780|Loss: 0.8022850155830383:  12%|█▏        | 779/6470 [09:35<1:14:34,  1.27it/s]
4|780|Loss: 0.8022850155830383:  12%|█▏        | 780/6470 [09:35<1:13:05,  1.30it/s]
4|781|Loss: 0.7677019238471985:  12%|█▏        | 780/6470 [09:35<1:13:05,  1.30it/s]
4|781|Loss: 0.7677019238471985:  12%|█▏        | 781/6470 [09:36<1:13:38,  1.29it/s]
4|782|Loss: 0.8884488940238953:  12%|█▏        | 781/6470 [09:36<1:13:38,  1.29it/s]
4|782|Loss: 0.8884488940238953:  12%|█▏        | 782/6470 [09:36<1:07:59,  1.39it/s]
4|783|Loss: 0.9961504936218262:  12%|█▏        | 782/6470 [09:37<1:07:59,  1.39it/s]
4|783|Loss: 0.9961504936218262:  12%|█▏        | 783/6470 [09:37<1:13:10,  1.30it/s]
4|784|Loss: 0.7700212597846985:  12%|█▏        | 783/6470 [09:38<1:13:10,  1.30it/s]
4|784|Loss: 0.7700212597846985:  12%|█▏        | 784/6470 [09:38<1:16:32,  1.24it/s]
4|785|Loss: 0.7358580827713013:  12%|█▏        | 784/6470 [09:38<1:16:32,  1.24it/s]
4|785|Loss: 0.7358580827713013:  12%|█▏        | 785/6470 [09:39<1:14:18,  1.27it/s]
4|786|Loss: 0.7999972701072693:  12%|█▏        | 785/6470 [09:39<1:14:18,  1.27it/s]
4|786|Loss: 0.7999972701072693:  12%|█▏        | 786/6470 [09:39<1:09:15,  1.37it/s]
4|787|Loss: 0.9262619018554688:  12%|█▏        | 786/6470 [09:40<1:09:15,  1.37it/s]
4|787|Loss: 0.9262619018554688:  12%|█▏        | 787/6470 [09:40<1:11:42,  1.32it/s]
4|788|Loss: 0.9270510673522949:  12%|█▏        | 787/6470 [09:41<1:11:42,  1.32it/s]
4|788|Loss: 0.9270510673522949:  12%|█▏        | 788/6470 [09:41<1:10:20,  1.35it/s]
4|789|Loss: 0.8176925778388977:  12%|█▏        | 788/6470 [09:41<1:10:20,  1.35it/s]
4|789|Loss: 0.8176925778388977:  12%|█▏        | 789/6470 [09:42<1:11:21,  1.33it/s]
4|790|Loss: 0.845201849937439:  12%|█▏        | 789/6470 [09:42<1:11:21,  1.33it/s] 
4|790|Loss: 0.845201849937439:  12%|█▏        | 790/6470 [09:42<1:11:44,  1.32it/s]
4|791|Loss: 0.8280882239341736:  12%|█▏        | 790/6470 [09:43<1:11:44,  1.32it/s]
4|791|Loss: 0.8280882239341736:  12%|█▏        | 791/6470 [09:43<1:11:51,  1.32it/s]
4|792|Loss: 0.738591730594635:  12%|█▏        | 791/6470 [09:44<1:11:51,  1.32it/s] 
4|792|Loss: 0.738591730594635:  12%|█▏        | 792/6470 [09:44<1:15:11,  1.26it/s]
4|793|Loss: 0.9355767369270325:  12%|█▏        | 792/6470 [09:44<1:15:11,  1.26it/s]
4|793|Loss: 0.9355767369270325:  12%|█▏        | 793/6470 [09:45<1:13:41,  1.28it/s]
4|794|Loss: 0.8587045073509216:  12%|█▏        | 793/6470 [09:45<1:13:41,  1.28it/s]
4|794|Loss: 0.8587045073509216:  12%|█▏        | 794/6470 [09:46<1:12:58,  1.30it/s]
4|795|Loss: 0.8224434852600098:  12%|█▏        | 794/6470 [09:46<1:12:58,  1.30it/s]
4|795|Loss: 0.8224434852600098:  12%|█▏        | 795/6470 [09:46<1:13:41,  1.28it/s]
4|796|Loss: 0.877180814743042:  12%|█▏        | 795/6470 [09:47<1:13:41,  1.28it/s] 
4|796|Loss: 0.877180814743042:  12%|█▏        | 796/6470 [09:47<1:15:35,  1.25it/s]
4|797|Loss: 0.8024370670318604:  12%|█▏        | 796/6470 [09:48<1:15:35,  1.25it/s]
4|797|Loss: 0.8024370670318604:  12%|█▏        | 797/6470 [09:48<1:12:58,  1.30it/s]
4|798|Loss: 0.9262685179710388:  12%|█▏        | 797/6470 [09:48<1:12:58,  1.30it/s]
4|798|Loss: 0.9262685179710388:  12%|█▏        | 798/6470 [09:49<1:13:02,  1.29it/s]
4|799|Loss: 0.790971577167511:  12%|█▏        | 798/6470 [09:49<1:13:02,  1.29it/s] 
4|799|Loss: 0.790971577167511:  12%|█▏        | 799/6470 [09:49<1:14:05,  1.28it/s]
4|800|Loss: 0.7476719617843628:  12%|█▏        | 799/6470 [09:50<1:14:05,  1.28it/s]
4|800|Loss: 0.7476719617843628:  12%|█▏        | 800/6470 [09:50<1:11:56,  1.31it/s]
4|801|Loss: 0.8802697062492371:  12%|█▏        | 800/6470 [09:51<1:11:56,  1.31it/s]
4|801|Loss: 0.8802697062492371:  12%|█▏        | 801/6470 [09:51<1:10:04,  1.35it/s]
4|802|Loss: 0.8295203447341919:  12%|█▏        | 801/6470 [09:51<1:10:04,  1.35it/s]
4|802|Loss: 0.8295203447341919:  12%|█▏        | 802/6470 [09:51<1:05:14,  1.45it/s]
4|803|Loss: 0.8032475113868713:  12%|█▏        | 802/6470 [09:52<1:05:14,  1.45it/s]
4|803|Loss: 0.8032475113868713:  12%|█▏        | 803/6470 [09:52<1:02:32,  1.51it/s]
4|804|Loss: 0.8165227770805359:  12%|█▏        | 803/6470 [09:52<1:02:32,  1.51it/s]
4|804|Loss: 0.8165227770805359:  12%|█▏        | 804/6470 [09:53<1:04:27,  1.47it/s]
4|805|Loss: 0.9093520641326904:  12%|█▏        | 804/6470 [09:53<1:04:27,  1.47it/s]
4|805|Loss: 0.9093520641326904:  12%|█▏        | 805/6470 [09:54<1:08:35,  1.38it/s]
4|806|Loss: 0.8213821649551392:  12%|█▏        | 805/6470 [09:54<1:08:35,  1.38it/s]
4|806|Loss: 0.8213821649551392:  12%|█▏        | 806/6470 [09:54<1:11:28,  1.32it/s]
4|807|Loss: 0.7929429411888123:  12%|█▏        | 806/6470 [09:55<1:11:28,  1.32it/s]
4|807|Loss: 0.7929429411888123:  12%|█▏        | 807/6470 [09:55<1:09:14,  1.36it/s]
4|808|Loss: 0.867382287979126:  12%|█▏        | 807/6470 [09:56<1:09:14,  1.36it/s] 
4|808|Loss: 0.867382287979126:  12%|█▏        | 808/6470 [09:56<1:08:54,  1.37it/s]
4|809|Loss: 0.746764063835144:  12%|█▏        | 808/6470 [09:56<1:08:54,  1.37it/s]
4|809|Loss: 0.746764063835144:  13%|█▎        | 809/6470 [09:56<1:05:21,  1.44it/s]
4|810|Loss: 0.8113009929656982:  13%|█▎        | 809/6470 [09:57<1:05:21,  1.44it/s]
4|810|Loss: 0.8113009929656982:  13%|█▎        | 810/6470 [09:57<1:11:56,  1.31it/s]
4|811|Loss: 0.7127644419670105:  13%|█▎        | 810/6470 [09:58<1:11:56,  1.31it/s]
4|811|Loss: 0.7127644419670105:  13%|█▎        | 811/6470 [09:58<1:16:08,  1.24it/s]
4|812|Loss: 0.8782476186752319:  13%|█▎        | 811/6470 [09:59<1:16:08,  1.24it/s]
4|812|Loss: 0.8782476186752319:  13%|█▎        | 812/6470 [09:59<1:16:45,  1.23it/s]
4|813|Loss: 0.7770428657531738:  13%|█▎        | 812/6470 [10:00<1:16:45,  1.23it/s]
4|813|Loss: 0.7770428657531738:  13%|█▎        | 813/6470 [10:00<1:14:23,  1.27it/s]
4|814|Loss: 0.7979109883308411:  13%|█▎        | 813/6470 [10:00<1:14:23,  1.27it/s]
4|814|Loss: 0.7979109883308411:  13%|█▎        | 814/6470 [10:01<1:13:08,  1.29it/s]
4|815|Loss: 0.9720483422279358:  13%|█▎        | 814/6470 [10:01<1:13:08,  1.29it/s]
4|815|Loss: 0.9720483422279358:  13%|█▎        | 815/6470 [10:01<1:10:26,  1.34it/s]
4|816|Loss: 0.6485617756843567:  13%|█▎        | 815/6470 [10:02<1:10:26,  1.34it/s]
4|816|Loss: 0.6485617756843567:  13%|█▎        | 816/6470 [10:02<1:16:54,  1.23it/s]
4|817|Loss: 0.7715104818344116:  13%|█▎        | 816/6470 [10:03<1:16:54,  1.23it/s]
4|817|Loss: 0.7715104818344116:  13%|█▎        | 817/6470 [10:03<1:15:10,  1.25it/s]
4|818|Loss: 0.792961597442627:  13%|█▎        | 817/6470 [10:03<1:15:10,  1.25it/s] 
4|818|Loss: 0.792961597442627:  13%|█▎        | 818/6470 [10:04<1:08:46,  1.37it/s]
4|819|Loss: 0.7981677651405334:  13%|█▎        | 818/6470 [10:04<1:08:46,  1.37it/s]
4|819|Loss: 0.7981677651405334:  13%|█▎        | 819/6470 [10:04<1:04:02,  1.47it/s]
4|820|Loss: 0.8779422640800476:  13%|█▎        | 819/6470 [10:05<1:04:02,  1.47it/s]
4|820|Loss: 0.8779422640800476:  13%|█▎        | 820/6470 [10:05<1:08:33,  1.37it/s]
4|821|Loss: 0.7049645185470581:  13%|█▎        | 820/6470 [10:05<1:08:33,  1.37it/s]
4|821|Loss: 0.7049645185470581:  13%|█▎        | 821/6470 [10:06<1:13:42,  1.28it/s]
4|822|Loss: 0.9518687129020691:  13%|█▎        | 821/6470 [10:06<1:13:42,  1.28it/s]
4|822|Loss: 0.9518687129020691:  13%|█▎        | 822/6470 [10:07<1:10:15,  1.34it/s]
4|823|Loss: 0.8267115354537964:  13%|█▎        | 822/6470 [10:07<1:10:15,  1.34it/s]
4|823|Loss: 0.8267115354537964:  13%|█▎        | 823/6470 [10:07<1:06:21,  1.42it/s]
4|824|Loss: 0.775377631187439:  13%|█▎        | 823/6470 [10:08<1:06:21,  1.42it/s] 
4|824|Loss: 0.775377631187439:  13%|█▎        | 824/6470 [10:08<1:03:59,  1.47it/s]
4|825|Loss: 0.5953782796859741:  13%|█▎        | 824/6470 [10:08<1:03:59,  1.47it/s]
4|825|Loss: 0.5953782796859741:  13%|█▎        | 825/6470 [10:08<1:03:12,  1.49it/s]
4|826|Loss: 0.728018045425415:  13%|█▎        | 825/6470 [10:09<1:03:12,  1.49it/s] 
4|826|Loss: 0.728018045425415:  13%|█▎        | 826/6470 [10:09<1:08:41,  1.37it/s]
4|827|Loss: 0.7637767791748047:  13%|█▎        | 826/6470 [10:10<1:08:41,  1.37it/s]
4|827|Loss: 0.7637767791748047:  13%|█▎        | 827/6470 [10:10<1:08:21,  1.38it/s]
4|828|Loss: 0.8564839959144592:  13%|█▎        | 827/6470 [10:10<1:08:21,  1.38it/s]
4|828|Loss: 0.8564839959144592:  13%|█▎        | 828/6470 [10:11<1:08:50,  1.37it/s]
4|829|Loss: 0.7819325923919678:  13%|█▎        | 828/6470 [10:11<1:08:50,  1.37it/s]
4|829|Loss: 0.7819325923919678:  13%|█▎        | 829/6470 [10:11<1:07:28,  1.39it/s]
4|830|Loss: 0.9017903804779053:  13%|█▎        | 829/6470 [10:12<1:07:28,  1.39it/s]
4|830|Loss: 0.9017903804779053:  13%|█▎        | 830/6470 [10:12<1:10:18,  1.34it/s]
4|831|Loss: 0.8794463872909546:  13%|█▎        | 830/6470 [10:13<1:10:18,  1.34it/s]
4|831|Loss: 0.8794463872909546:  13%|█▎        | 831/6470 [10:13<1:13:34,  1.28it/s]
4|832|Loss: 0.7082238793373108:  13%|█▎        | 831/6470 [10:14<1:13:34,  1.28it/s]
4|832|Loss: 0.7082238793373108:  13%|█▎        | 832/6470 [10:14<1:19:49,  1.18it/s]
4|833|Loss: 0.8377586007118225:  13%|█▎        | 832/6470 [10:15<1:19:49,  1.18it/s]
4|833|Loss: 0.8377586007118225:  13%|█▎        | 833/6470 [10:15<1:19:26,  1.18it/s]
4|834|Loss: 0.9639278054237366:  13%|█▎        | 833/6470 [10:15<1:19:26,  1.18it/s]
4|834|Loss: 0.9639278054237366:  13%|█▎        | 834/6470 [10:16<1:18:17,  1.20it/s]
4|835|Loss: 0.6781343817710876:  13%|█▎        | 834/6470 [10:16<1:18:17,  1.20it/s]
4|835|Loss: 0.6781343817710876:  13%|█▎        | 835/6470 [10:17<1:20:30,  1.17it/s]
4|836|Loss: 0.7229600548744202:  13%|█▎        | 835/6470 [10:17<1:20:30,  1.17it/s]
4|836|Loss: 0.7229600548744202:  13%|█▎        | 836/6470 [10:18<1:23:27,  1.13it/s]
4|837|Loss: 0.9021198153495789:  13%|█▎        | 836/6470 [10:18<1:23:27,  1.13it/s]
4|837|Loss: 0.9021198153495789:  13%|█▎        | 837/6470 [10:19<1:21:57,  1.15it/s]
4|838|Loss: 0.9245957732200623:  13%|█▎        | 837/6470 [10:19<1:21:57,  1.15it/s]
4|838|Loss: 0.9245957732200623:  13%|█▎        | 838/6470 [10:19<1:21:34,  1.15it/s]
4|839|Loss: 0.7123227715492249:  13%|█▎        | 838/6470 [10:20<1:21:34,  1.15it/s]
4|839|Loss: 0.7123227715492249:  13%|█▎        | 839/6470 [10:20<1:19:55,  1.17it/s]
4|840|Loss: 0.9835386872291565:  13%|█▎        | 839/6470 [10:21<1:19:55,  1.17it/s]
4|840|Loss: 0.9835386872291565:  13%|█▎        | 840/6470 [10:21<1:20:31,  1.17it/s]
4|841|Loss: 0.8117516040802002:  13%|█▎        | 840/6470 [10:22<1:20:31,  1.17it/s]
4|841|Loss: 0.8117516040802002:  13%|█▎        | 841/6470 [10:22<1:19:31,  1.18it/s]
4|842|Loss: 0.661485493183136:  13%|█▎        | 841/6470 [10:22<1:19:31,  1.18it/s] 
4|842|Loss: 0.661485493183136:  13%|█▎        | 842/6470 [10:23<1:17:41,  1.21it/s]
4|843|Loss: 0.6261244416236877:  13%|█▎        | 842/6470 [10:23<1:17:41,  1.21it/s]
4|843|Loss: 0.6261244416236877:  13%|█▎        | 843/6470 [10:23<1:09:20,  1.35it/s]
4|844|Loss: 0.6682590842247009:  13%|█▎        | 843/6470 [10:24<1:09:20,  1.35it/s]
4|844|Loss: 0.6682590842247009:  13%|█▎        | 844/6470 [10:24<1:11:28,  1.31it/s]
4|845|Loss: 0.9656345248222351:  13%|█▎        | 844/6470 [10:24<1:11:28,  1.31it/s]
4|845|Loss: 0.9656345248222351:  13%|█▎        | 845/6470 [10:25<1:08:16,  1.37it/s]
4|846|Loss: 0.7940083146095276:  13%|█▎        | 845/6470 [10:25<1:08:16,  1.37it/s]
4|846|Loss: 0.7940083146095276:  13%|█▎        | 846/6470 [10:25<1:06:46,  1.40it/s]
4|847|Loss: 0.9123256206512451:  13%|█▎        | 846/6470 [10:26<1:06:46,  1.40it/s]
4|847|Loss: 0.9123256206512451:  13%|█▎        | 847/6470 [10:26<1:05:51,  1.42it/s]
4|848|Loss: 0.771308183670044:  13%|█▎        | 847/6470 [10:26<1:05:51,  1.42it/s] 
4|848|Loss: 0.771308183670044:  13%|█▎        | 848/6470 [10:27<1:07:03,  1.40it/s]
4|849|Loss: 0.7869184017181396:  13%|█▎        | 848/6470 [10:27<1:07:03,  1.40it/s]
4|849|Loss: 0.7869184017181396:  13%|█▎        | 849/6470 [10:28<1:11:53,  1.30it/s]
4|850|Loss: 0.8721351027488708:  13%|█▎        | 849/6470 [10:28<1:11:53,  1.30it/s]
4|850|Loss: 0.8721351027488708:  13%|█▎        | 850/6470 [10:28<1:11:44,  1.31it/s]
4|851|Loss: 0.8923883438110352:  13%|█▎        | 850/6470 [10:29<1:11:44,  1.31it/s]
4|851|Loss: 0.8923883438110352:  13%|█▎        | 851/6470 [10:29<1:04:59,  1.44it/s]
4|852|Loss: 1.0355842113494873:  13%|█▎        | 851/6470 [10:29<1:04:59,  1.44it/s]
4|852|Loss: 1.0355842113494873:  13%|█▎        | 852/6470 [10:29<57:12,  1.64it/s]  
4|853|Loss: 0.7462512850761414:  13%|█▎        | 852/6470 [10:30<57:12,  1.64it/s]
4|853|Loss: 0.7462512850761414:  13%|█▎        | 853/6470 [10:30<1:00:10,  1.56it/s]
4|854|Loss: 0.81967693567276:  13%|█▎        | 853/6470 [10:30<1:00:10,  1.56it/s]  
4|854|Loss: 0.81967693567276:  13%|█▎        | 854/6470 [10:31<1:02:43,  1.49it/s]
4|855|Loss: 0.8658446073532104:  13%|█▎        | 854/6470 [10:31<1:02:43,  1.49it/s]
4|855|Loss: 0.8658446073532104:  13%|█▎        | 855/6470 [10:32<1:08:36,  1.36it/s]
4|856|Loss: 0.9519598484039307:  13%|█▎        | 855/6470 [10:32<1:08:36,  1.36it/s]
4|856|Loss: 0.9519598484039307:  13%|█▎        | 856/6470 [10:33<1:12:29,  1.29it/s]
4|857|Loss: 0.9759409427642822:  13%|█▎        | 856/6470 [10:33<1:12:29,  1.29it/s]
4|857|Loss: 0.9759409427642822:  13%|█▎        | 857/6470 [10:33<1:09:38,  1.34it/s]
4|858|Loss: 0.8609223365783691:  13%|█▎        | 857/6470 [10:34<1:09:38,  1.34it/s]
4|858|Loss: 0.8609223365783691:  13%|█▎        | 858/6470 [10:34<1:14:09,  1.26it/s]
4|859|Loss: 0.864994466304779:  13%|█▎        | 858/6470 [10:35<1:14:09,  1.26it/s] 
4|859|Loss: 0.864994466304779:  13%|█▎        | 859/6470 [10:35<1:20:05,  1.17it/s]
4|860|Loss: 0.9006288647651672:  13%|█▎        | 859/6470 [10:36<1:20:05,  1.17it/s]
4|860|Loss: 0.9006288647651672:  13%|█▎        | 860/6470 [10:36<1:16:34,  1.22it/s]
4|861|Loss: 0.7993617653846741:  13%|█▎        | 860/6470 [10:36<1:16:34,  1.22it/s]
4|861|Loss: 0.7993617653846741:  13%|█▎        | 861/6470 [10:37<1:18:10,  1.20it/s]
4|862|Loss: 0.7604817748069763:  13%|█▎        | 861/6470 [10:37<1:18:10,  1.20it/s]
4|862|Loss: 0.7604817748069763:  13%|█▎        | 862/6470 [10:38<1:16:35,  1.22it/s]
4|863|Loss: 0.8909853100776672:  13%|█▎        | 862/6470 [10:38<1:16:35,  1.22it/s]
4|863|Loss: 0.8909853100776672:  13%|█▎        | 863/6470 [10:38<1:09:16,  1.35it/s]
4|864|Loss: 0.9179238677024841:  13%|█▎        | 863/6470 [10:38<1:09:16,  1.35it/s]
4|864|Loss: 0.9179238677024841:  13%|█▎        | 864/6470 [10:39<1:08:52,  1.36it/s]
4|865|Loss: 0.7802256345748901:  13%|█▎        | 864/6470 [10:39<1:08:52,  1.36it/s]
4|865|Loss: 0.7802256345748901:  13%|█▎        | 865/6470 [10:39<1:03:20,  1.47it/s]
4|866|Loss: 0.8562290072441101:  13%|█▎        | 865/6470 [10:40<1:03:20,  1.47it/s]
4|866|Loss: 0.8562290072441101:  13%|█▎        | 866/6470 [10:40<1:03:28,  1.47it/s]
4|867|Loss: 0.7595273852348328:  13%|█▎        | 866/6470 [10:40<1:03:28,  1.47it/s]
4|867|Loss: 0.7595273852348328:  13%|█▎        | 867/6470 [10:41<1:07:15,  1.39it/s]
4|868|Loss: 0.8627128601074219:  13%|█▎        | 867/6470 [10:41<1:07:15,  1.39it/s]
4|868|Loss: 0.8627128601074219:  13%|█▎        | 868/6470 [10:42<1:08:48,  1.36it/s]
4|869|Loss: 0.8170015811920166:  13%|█▎        | 868/6470 [10:42<1:08:48,  1.36it/s]
4|869|Loss: 0.8170015811920166:  13%|█▎        | 869/6470 [10:42<1:07:58,  1.37it/s]
4|870|Loss: 0.8974818587303162:  13%|█▎        | 869/6470 [10:43<1:07:58,  1.37it/s]
4|870|Loss: 0.8974818587303162:  13%|█▎        | 870/6470 [10:43<1:07:01,  1.39it/s]
4|871|Loss: 0.7980165481567383:  13%|█▎        | 870/6470 [10:43<1:07:01,  1.39it/s]
4|871|Loss: 0.7980165481567383:  13%|█▎        | 871/6470 [10:44<1:11:17,  1.31it/s]
4|872|Loss: 0.8631923198699951:  13%|█▎        | 871/6470 [10:44<1:11:17,  1.31it/s]
4|872|Loss: 0.8631923198699951:  13%|█▎        | 872/6470 [10:45<1:08:05,  1.37it/s]
4|873|Loss: 0.8858770728111267:  13%|█▎        | 872/6470 [10:45<1:08:05,  1.37it/s]
4|873|Loss: 0.8858770728111267:  13%|█▎        | 873/6470 [10:45<1:02:18,  1.50it/s]
4|874|Loss: 0.8815532922744751:  13%|█▎        | 873/6470 [10:45<1:02:18,  1.50it/s]
4|874|Loss: 0.8815532922744751:  14%|█▎        | 874/6470 [10:46<1:03:21,  1.47it/s]
4|875|Loss: 0.7431610226631165:  14%|█▎        | 874/6470 [10:46<1:03:21,  1.47it/s]
4|875|Loss: 0.7431610226631165:  14%|█▎        | 875/6470 [10:47<1:05:14,  1.43it/s]
4|876|Loss: 0.8440995216369629:  14%|█▎        | 875/6470 [10:47<1:05:14,  1.43it/s]
4|876|Loss: 0.8440995216369629:  14%|█▎        | 876/6470 [10:47<1:04:39,  1.44it/s]
4|877|Loss: 0.9971557855606079:  14%|█▎        | 876/6470 [10:48<1:04:39,  1.44it/s]
4|877|Loss: 0.9971557855606079:  14%|█▎        | 877/6470 [10:48<1:04:24,  1.45it/s]
4|878|Loss: 0.9456613063812256:  14%|█▎        | 877/6470 [10:48<1:04:24,  1.45it/s]
4|878|Loss: 0.9456613063812256:  14%|█▎        | 878/6470 [10:49<1:07:12,  1.39it/s]
4|879|Loss: 0.9121481776237488:  14%|█▎        | 878/6470 [10:49<1:07:12,  1.39it/s]
4|879|Loss: 0.9121481776237488:  14%|█▎        | 879/6470 [10:49<1:08:33,  1.36it/s]
4|880|Loss: 0.6894911527633667:  14%|█▎        | 879/6470 [10:50<1:08:33,  1.36it/s]
4|880|Loss: 0.6894911527633667:  14%|█▎        | 880/6470 [10:50<1:05:21,  1.43it/s]
4|881|Loss: 1.0370744466781616:  14%|█▎        | 880/6470 [10:50<1:05:21,  1.43it/s]
4|881|Loss: 1.0370744466781616:  14%|█▎        | 881/6470 [10:51<1:09:14,  1.35it/s]
4|882|Loss: 0.7975921630859375:  14%|█▎        | 881/6470 [10:51<1:09:14,  1.35it/s]
4|882|Loss: 0.7975921630859375:  14%|█▎        | 882/6470 [10:52<1:10:20,  1.32it/s]
4|883|Loss: 0.6788688898086548:  14%|█▎        | 882/6470 [10:52<1:10:20,  1.32it/s]
4|883|Loss: 0.6788688898086548:  14%|█▎        | 883/6470 [10:53<1:17:25,  1.20it/s]
4|884|Loss: 0.7362907528877258:  14%|█▎        | 883/6470 [10:53<1:17:25,  1.20it/s]
4|884|Loss: 0.7362907528877258:  14%|█▎        | 884/6470 [10:53<1:10:54,  1.31it/s]
4|885|Loss: 0.8586698770523071:  14%|█▎        | 884/6470 [10:54<1:10:54,  1.31it/s]
4|885|Loss: 0.8586698770523071:  14%|█▎        | 885/6470 [10:54<1:13:22,  1.27it/s]
4|886|Loss: 0.7555059194564819:  14%|█▎        | 885/6470 [10:55<1:13:22,  1.27it/s]
4|886|Loss: 0.7555059194564819:  14%|█▎        | 886/6470 [10:55<1:14:22,  1.25it/s]
4|887|Loss: 0.7854548692703247:  14%|█▎        | 886/6470 [10:55<1:14:22,  1.25it/s]
4|887|Loss: 0.7854548692703247:  14%|█▎        | 887/6470 [10:56<1:16:41,  1.21it/s]
4|888|Loss: 0.7076136469841003:  14%|█▎        | 887/6470 [10:56<1:16:41,  1.21it/s]
4|888|Loss: 0.7076136469841003:  14%|█▎        | 888/6470 [10:57<1:16:45,  1.21it/s]
4|889|Loss: 0.745722234249115:  14%|█▎        | 888/6470 [10:57<1:16:45,  1.21it/s] 
4|889|Loss: 0.745722234249115:  14%|█▎        | 889/6470 [10:57<1:14:14,  1.25it/s]
4|890|Loss: 0.7622777819633484:  14%|█▎        | 889/6470 [10:58<1:14:14,  1.25it/s]
4|890|Loss: 0.7622777819633484:  14%|█▍        | 890/6470 [10:58<1:10:04,  1.33it/s]
4|891|Loss: 0.7849211096763611:  14%|█▍        | 890/6470 [10:58<1:10:04,  1.33it/s]
4|891|Loss: 0.7849211096763611:  14%|█▍        | 891/6470 [10:59<1:11:34,  1.30it/s]
4|892|Loss: 0.8027702569961548:  14%|█▍        | 891/6470 [10:59<1:11:34,  1.30it/s]
4|892|Loss: 0.8027702569961548:  14%|█▍        | 892/6470 [10:59<1:06:58,  1.39it/s]
4|893|Loss: 0.8045393824577332:  14%|█▍        | 892/6470 [11:00<1:06:58,  1.39it/s]
4|893|Loss: 0.8045393824577332:  14%|█▍        | 893/6470 [11:00<1:12:06,  1.29it/s]
4|894|Loss: 0.7611568570137024:  14%|█▍        | 893/6470 [11:01<1:12:06,  1.29it/s]
4|894|Loss: 0.7611568570137024:  14%|█▍        | 894/6470 [11:01<1:09:51,  1.33it/s]
4|895|Loss: 0.7470704913139343:  14%|█▍        | 894/6470 [11:01<1:09:51,  1.33it/s]
4|895|Loss: 0.7470704913139343:  14%|█▍        | 895/6470 [11:02<1:10:16,  1.32it/s]
4|896|Loss: 0.783597469329834:  14%|█▍        | 895/6470 [11:02<1:10:16,  1.32it/s] 
4|896|Loss: 0.783597469329834:  14%|█▍        | 896/6470 [11:02<1:06:42,  1.39it/s]
4|897|Loss: 0.9990230202674866:  14%|█▍        | 896/6470 [11:03<1:06:42,  1.39it/s]
4|897|Loss: 0.9990230202674866:  14%|█▍        | 897/6470 [11:03<1:05:34,  1.42it/s]
4|898|Loss: 0.8972842693328857:  14%|█▍        | 897/6470 [11:04<1:05:34,  1.42it/s]
4|898|Loss: 0.8972842693328857:  14%|█▍        | 898/6470 [11:04<1:10:59,  1.31it/s]
4|899|Loss: 0.9042283892631531:  14%|█▍        | 898/6470 [11:05<1:10:59,  1.31it/s]
4|899|Loss: 0.9042283892631531:  14%|█▍        | 899/6470 [11:05<1:15:10,  1.24it/s]
4|900|Loss: 0.9353869557380676:  14%|█▍        | 899/6470 [11:05<1:15:10,  1.24it/s]
4|900|Loss: 0.9353869557380676:  14%|█▍        | 900/6470 [11:06<1:08:09,  1.36it/s]
4|901|Loss: 0.7252366542816162:  14%|█▍        | 900/6470 [11:06<1:08:09,  1.36it/s]
4|901|Loss: 0.7252366542816162:  14%|█▍        | 901/6470 [11:06<1:07:35,  1.37it/s]
4|902|Loss: 0.8746719360351562:  14%|█▍        | 901/6470 [11:07<1:07:35,  1.37it/s]
4|902|Loss: 0.8746719360351562:  14%|█▍        | 902/6470 [11:07<1:07:07,  1.38it/s]
4|903|Loss: 0.8991291522979736:  14%|█▍        | 902/6470 [11:07<1:07:07,  1.38it/s]
4|903|Loss: 0.8991291522979736:  14%|█▍        | 903/6470 [11:08<1:09:13,  1.34it/s]
4|904|Loss: 0.9391062259674072:  14%|█▍        | 903/6470 [11:08<1:09:13,  1.34it/s]
4|904|Loss: 0.9391062259674072:  14%|█▍        | 904/6470 [11:08<1:08:32,  1.35it/s]
4|905|Loss: 0.8690761923789978:  14%|█▍        | 904/6470 [11:09<1:08:32,  1.35it/s]
4|905|Loss: 0.8690761923789978:  14%|█▍        | 905/6470 [11:09<1:11:58,  1.29it/s]
4|906|Loss: 0.5521435141563416:  14%|█▍        | 905/6470 [11:10<1:11:58,  1.29it/s]
4|906|Loss: 0.5521435141563416:  14%|█▍        | 906/6470 [11:11<1:30:42,  1.02it/s]
4|907|Loss: 0.784996747970581:  14%|█▍        | 906/6470 [11:11<1:30:42,  1.02it/s] 
4|907|Loss: 0.784996747970581:  14%|█▍        | 907/6470 [11:12<1:30:02,  1.03it/s]
4|908|Loss: 0.75095134973526:  14%|█▍        | 907/6470 [11:12<1:30:02,  1.03it/s] 
4|908|Loss: 0.75095134973526:  14%|█▍        | 908/6470 [11:13<1:25:55,  1.08it/s]
4|909|Loss: 0.8649941086769104:  14%|█▍        | 908/6470 [11:13<1:25:55,  1.08it/s]
4|909|Loss: 0.8649941086769104:  14%|█▍        | 909/6470 [11:13<1:20:33,  1.15it/s]
4|910|Loss: 0.7396011352539062:  14%|█▍        | 909/6470 [11:14<1:20:33,  1.15it/s]
4|910|Loss: 0.7396011352539062:  14%|█▍        | 910/6470 [11:14<1:20:11,  1.16it/s]
4|911|Loss: 0.8527868986129761:  14%|█▍        | 910/6470 [11:15<1:20:11,  1.16it/s]
4|911|Loss: 0.8527868986129761:  14%|█▍        | 911/6470 [11:15<1:15:40,  1.22it/s]
4|912|Loss: 0.7564906477928162:  14%|█▍        | 911/6470 [11:15<1:15:40,  1.22it/s]
4|912|Loss: 0.7564906477928162:  14%|█▍        | 912/6470 [11:16<1:13:29,  1.26it/s]
4|913|Loss: 0.8750666975975037:  14%|█▍        | 912/6470 [11:16<1:13:29,  1.26it/s]
4|913|Loss: 0.8750666975975037:  14%|█▍        | 913/6470 [11:16<1:13:05,  1.27it/s]
4|914|Loss: 0.7278225421905518:  14%|█▍        | 913/6470 [11:17<1:13:05,  1.27it/s]
4|914|Loss: 0.7278225421905518:  14%|█▍        | 914/6470 [11:17<1:11:36,  1.29it/s]
4|915|Loss: 0.8432438373565674:  14%|█▍        | 914/6470 [11:17<1:11:36,  1.29it/s]
4|915|Loss: 0.8432438373565674:  14%|█▍        | 915/6470 [11:18<1:07:59,  1.36it/s]
4|916|Loss: 0.7523171305656433:  14%|█▍        | 915/6470 [11:18<1:07:59,  1.36it/s]
4|916|Loss: 0.7523171305656433:  14%|█▍        | 916/6470 [11:19<1:08:41,  1.35it/s]
4|917|Loss: 0.7376247644424438:  14%|█▍        | 916/6470 [11:19<1:08:41,  1.35it/s]
4|917|Loss: 0.7376247644424438:  14%|█▍        | 917/6470 [11:19<1:05:28,  1.41it/s]
4|918|Loss: 0.8692553043365479:  14%|█▍        | 917/6470 [11:20<1:05:28,  1.41it/s]
4|918|Loss: 0.8692553043365479:  14%|█▍        | 918/6470 [11:20<1:05:42,  1.41it/s]
4|919|Loss: 0.9692937135696411:  14%|█▍        | 918/6470 [11:20<1:05:42,  1.41it/s]
4|919|Loss: 0.9692937135696411:  14%|█▍        | 919/6470 [11:21<1:10:40,  1.31it/s]
4|920|Loss: 0.738886833190918:  14%|█▍        | 919/6470 [11:21<1:10:40,  1.31it/s] 
4|920|Loss: 0.738886833190918:  14%|█▍        | 920/6470 [11:21<1:06:55,  1.38it/s]
4|921|Loss: 0.8366957902908325:  14%|█▍        | 920/6470 [11:22<1:06:55,  1.38it/s]
4|921|Loss: 0.8366957902908325:  14%|█▍        | 921/6470 [11:22<1:09:46,  1.33it/s]
4|922|Loss: 0.7882480621337891:  14%|█▍        | 921/6470 [11:23<1:09:46,  1.33it/s]
4|922|Loss: 0.7882480621337891:  14%|█▍        | 922/6470 [11:23<1:12:34,  1.27it/s]
4|923|Loss: 0.8387179374694824:  14%|█▍        | 922/6470 [11:24<1:12:34,  1.27it/s]
4|923|Loss: 0.8387179374694824:  14%|█▍        | 923/6470 [11:24<1:12:30,  1.28it/s]
4|924|Loss: 0.6937243342399597:  14%|█▍        | 923/6470 [11:24<1:12:30,  1.28it/s]
4|924|Loss: 0.6937243342399597:  14%|█▍        | 924/6470 [11:25<1:15:06,  1.23it/s]
4|925|Loss: 0.803171694278717:  14%|█▍        | 924/6470 [11:25<1:15:06,  1.23it/s] 
4|925|Loss: 0.803171694278717:  14%|█▍        | 925/6470 [11:25<1:12:30,  1.27it/s]
4|926|Loss: 0.8320766687393188:  14%|█▍        | 925/6470 [11:26<1:12:30,  1.27it/s]
4|926|Loss: 0.8320766687393188:  14%|█▍        | 926/6470 [11:26<1:11:48,  1.29it/s]
4|927|Loss: 0.6719513535499573:  14%|█▍        | 926/6470 [11:27<1:11:48,  1.29it/s]
4|927|Loss: 0.6719513535499573:  14%|█▍        | 927/6470 [11:27<1:11:59,  1.28it/s]
4|928|Loss: 0.5802533030509949:  14%|█▍        | 927/6470 [11:27<1:11:59,  1.28it/s]
4|928|Loss: 0.5802533030509949:  14%|█▍        | 928/6470 [11:28<1:15:30,  1.22it/s]
4|929|Loss: 0.6906569004058838:  14%|█▍        | 928/6470 [11:28<1:15:30,  1.22it/s]
4|929|Loss: 0.6906569004058838:  14%|█▍        | 929/6470 [11:29<1:12:57,  1.27it/s]
4|930|Loss: 0.8872374892234802:  14%|█▍        | 929/6470 [11:29<1:12:57,  1.27it/s]
4|930|Loss: 0.8872374892234802:  14%|█▍        | 930/6470 [11:29<1:09:29,  1.33it/s]
4|931|Loss: 0.9462597370147705:  14%|█▍        | 930/6470 [11:30<1:09:29,  1.33it/s]
4|931|Loss: 0.9462597370147705:  14%|█▍        | 931/6470 [11:30<1:09:05,  1.34it/s]
4|932|Loss: 0.9401436448097229:  14%|█▍        | 931/6470 [11:30<1:09:05,  1.34it/s]
4|932|Loss: 0.9401436448097229:  14%|█▍        | 932/6470 [11:31<1:10:20,  1.31it/s]
4|933|Loss: 0.8123957514762878:  14%|█▍        | 932/6470 [11:31<1:10:20,  1.31it/s]
4|933|Loss: 0.8123957514762878:  14%|█▍        | 933/6470 [11:32<1:11:27,  1.29it/s]
4|934|Loss: 0.9905700087547302:  14%|█▍        | 933/6470 [11:32<1:11:27,  1.29it/s]
4|934|Loss: 0.9905700087547302:  14%|█▍        | 934/6470 [11:32<1:11:04,  1.30it/s]
4|935|Loss: 0.8801954388618469:  14%|█▍        | 934/6470 [11:33<1:11:04,  1.30it/s]
4|935|Loss: 0.8801954388618469:  14%|█▍        | 935/6470 [11:33<1:10:32,  1.31it/s]
4|936|Loss: 0.8216437101364136:  14%|█▍        | 935/6470 [11:34<1:10:32,  1.31it/s]
4|936|Loss: 0.8216437101364136:  14%|█▍        | 936/6470 [11:34<1:08:27,  1.35it/s]
4|937|Loss: 0.8319584727287292:  14%|█▍        | 936/6470 [11:34<1:08:27,  1.35it/s]
4|937|Loss: 0.8319584727287292:  14%|█▍        | 937/6470 [11:35<1:07:04,  1.37it/s]
4|938|Loss: 0.8302540183067322:  14%|█▍        | 937/6470 [11:35<1:07:04,  1.37it/s]
4|938|Loss: 0.8302540183067322:  14%|█▍        | 938/6470 [11:35<1:08:13,  1.35it/s]
4|939|Loss: 0.8257436752319336:  14%|█▍        | 938/6470 [11:36<1:08:13,  1.35it/s]
4|939|Loss: 0.8257436752319336:  15%|█▍        | 939/6470 [11:36<1:12:01,  1.28it/s]
4|940|Loss: 0.9092278480529785:  15%|█▍        | 939/6470 [11:37<1:12:01,  1.28it/s]
4|940|Loss: 0.9092278480529785:  15%|█▍        | 940/6470 [11:37<1:15:57,  1.21it/s]
4|941|Loss: 0.9390132427215576:  15%|█▍        | 940/6470 [11:38<1:15:57,  1.21it/s]
4|941|Loss: 0.9390132427215576:  15%|█▍        | 941/6470 [11:38<1:12:25,  1.27it/s]
4|942|Loss: 0.7527705430984497:  15%|█▍        | 941/6470 [11:38<1:12:25,  1.27it/s]
4|942|Loss: 0.7527705430984497:  15%|█▍        | 942/6470 [11:39<1:10:35,  1.31it/s]
4|943|Loss: 0.8348471522331238:  15%|█▍        | 942/6470 [11:39<1:10:35,  1.31it/s]
4|943|Loss: 0.8348471522331238:  15%|█▍        | 943/6470 [11:39<1:09:01,  1.33it/s]
4|944|Loss: 0.7277026176452637:  15%|█▍        | 943/6470 [11:40<1:09:01,  1.33it/s]
4|944|Loss: 0.7277026176452637:  15%|█▍        | 944/6470 [11:40<1:11:25,  1.29it/s]
4|945|Loss: 0.9464455842971802:  15%|█▍        | 944/6470 [11:40<1:11:25,  1.29it/s]
4|945|Loss: 0.9464455842971802:  15%|█▍        | 945/6470 [11:41<1:07:06,  1.37it/s]
4|946|Loss: 0.73781818151474:  15%|█▍        | 945/6470 [11:41<1:07:06,  1.37it/s]  
4|946|Loss: 0.73781818151474:  15%|█▍        | 946/6470 [11:41<1:00:26,  1.52it/s]
4|947|Loss: 0.7533165812492371:  15%|█▍        | 946/6470 [11:41<1:00:26,  1.52it/s]
4|947|Loss: 0.7533165812492371:  15%|█▍        | 947/6470 [11:42<55:09,  1.67it/s]  
4|948|Loss: 0.8833178877830505:  15%|█▍        | 947/6470 [11:42<55:09,  1.67it/s]
4|948|Loss: 0.8833178877830505:  15%|█▍        | 948/6470 [11:42<49:55,  1.84it/s]
4|949|Loss: 0.7685377597808838:  15%|█▍        | 948/6470 [11:42<49:55,  1.84it/s]
4|949|Loss: 0.7685377597808838:  15%|█▍        | 949/6470 [11:43<49:42,  1.85it/s]
4|950|Loss: 0.7929676175117493:  15%|█▍        | 949/6470 [11:43<49:42,  1.85it/s]
4|950|Loss: 0.7929676175117493:  15%|█▍        | 950/6470 [11:43<50:54,  1.81it/s]
4|951|Loss: 0.8860626816749573:  15%|█▍        | 950/6470 [11:44<50:54,  1.81it/s]
4|951|Loss: 0.8860626816749573:  15%|█▍        | 951/6470 [11:44<55:45,  1.65it/s]
4|952|Loss: 0.7768915295600891:  15%|█▍        | 951/6470 [11:44<55:45,  1.65it/s]
4|952|Loss: 0.7768915295600891:  15%|█▍        | 952/6470 [11:45<1:03:36,  1.45it/s]
4|953|Loss: 0.8645035624504089:  15%|█▍        | 952/6470 [11:45<1:03:36,  1.45it/s]
4|953|Loss: 0.8645035624504089:  15%|█▍        | 953/6470 [11:46<1:05:18,  1.41it/s]
4|954|Loss: 0.7993756532669067:  15%|█▍        | 953/6470 [11:46<1:05:18,  1.41it/s]
4|954|Loss: 0.7993756532669067:  15%|█▍        | 954/6470 [11:46<1:06:46,  1.38it/s]
4|955|Loss: 0.869300901889801:  15%|█▍        | 954/6470 [11:47<1:06:46,  1.38it/s] 
4|955|Loss: 0.869300901889801:  15%|█▍        | 955/6470 [11:47<1:02:54,  1.46it/s]
4|956|Loss: 0.8197455406188965:  15%|█▍        | 955/6470 [11:47<1:02:54,  1.46it/s]
4|956|Loss: 0.8197455406188965:  15%|█▍        | 956/6470 [11:48<1:01:26,  1.50it/s]
4|957|Loss: 0.8094410300254822:  15%|█▍        | 956/6470 [11:48<1:01:26,  1.50it/s]
4|957|Loss: 0.8094410300254822:  15%|█▍        | 957/6470 [11:48<1:01:11,  1.50it/s]
4|958|Loss: 0.9358359575271606:  15%|█▍        | 957/6470 [11:49<1:01:11,  1.50it/s]
4|958|Loss: 0.9358359575271606:  15%|█▍        | 958/6470 [11:49<1:03:49,  1.44it/s]
4|959|Loss: 0.679550051689148:  15%|█▍        | 958/6470 [11:49<1:03:49,  1.44it/s] 
4|959|Loss: 0.679550051689148:  15%|█▍        | 959/6470 [11:50<1:03:32,  1.45it/s]
4|960|Loss: 0.7203666567802429:  15%|█▍        | 959/6470 [11:50<1:03:32,  1.45it/s]
4|960|Loss: 0.7203666567802429:  15%|█▍        | 960/6470 [11:50<1:05:20,  1.41it/s]
4|961|Loss: 1.0654692649841309:  15%|█▍        | 960/6470 [11:51<1:05:20,  1.41it/s]
4|961|Loss: 1.0654692649841309:  15%|█▍        | 961/6470 [11:51<1:06:27,  1.38it/s]
4|962|Loss: 0.7876749634742737:  15%|█▍        | 961/6470 [11:52<1:06:27,  1.38it/s]
4|962|Loss: 0.7876749634742737:  15%|█▍        | 962/6470 [11:52<1:08:02,  1.35it/s]
4|963|Loss: 0.7087562084197998:  15%|█▍        | 962/6470 [11:52<1:08:02,  1.35it/s]
4|963|Loss: 0.7087562084197998:  15%|█▍        | 963/6470 [11:53<1:05:35,  1.40it/s]
4|964|Loss: 0.7216621041297913:  15%|█▍        | 963/6470 [11:53<1:05:35,  1.40it/s]
4|964|Loss: 0.7216621041297913:  15%|█▍        | 964/6470 [11:53<1:06:18,  1.38it/s]
4|965|Loss: 0.7948905229568481:  15%|█▍        | 964/6470 [11:54<1:06:18,  1.38it/s]
4|965|Loss: 0.7948905229568481:  15%|█▍        | 965/6470 [11:54<1:01:07,  1.50it/s]
4|966|Loss: 0.8404701948165894:  15%|█▍        | 965/6470 [11:54<1:01:07,  1.50it/s]
4|966|Loss: 0.8404701948165894:  15%|█▍        | 966/6470 [11:55<1:02:28,  1.47it/s]
4|967|Loss: 0.8549808263778687:  15%|█▍        | 966/6470 [11:55<1:02:28,  1.47it/s]
4|967|Loss: 0.8549808263778687:  15%|█▍        | 967/6470 [11:56<1:10:19,  1.30it/s]
4|968|Loss: 0.7568413019180298:  15%|█▍        | 967/6470 [11:56<1:10:19,  1.30it/s]
4|968|Loss: 0.7568413019180298:  15%|█▍        | 968/6470 [11:56<1:11:48,  1.28it/s]
4|969|Loss: 0.9141197800636292:  15%|█▍        | 968/6470 [11:57<1:11:48,  1.28it/s]
4|969|Loss: 0.9141197800636292:  15%|█▍        | 969/6470 [11:57<1:11:22,  1.28it/s]
4|970|Loss: 0.8674160838127136:  15%|█▍        | 969/6470 [11:57<1:11:22,  1.28it/s]
4|970|Loss: 0.8674160838127136:  15%|█▍        | 970/6470 [11:58<1:05:53,  1.39it/s]
4|971|Loss: 0.8998396992683411:  15%|█▍        | 970/6470 [11:58<1:05:53,  1.39it/s]
4|971|Loss: 0.8998396992683411:  15%|█▌        | 971/6470 [11:58<1:06:14,  1.38it/s]
4|972|Loss: 0.6959550380706787:  15%|█▌        | 971/6470 [11:59<1:06:14,  1.38it/s]
4|972|Loss: 0.6959550380706787:  15%|█▌        | 972/6470 [11:59<1:11:24,  1.28it/s]
4|973|Loss: 0.857329785823822:  15%|█▌        | 972/6470 [12:00<1:11:24,  1.28it/s] 
4|973|Loss: 0.857329785823822:  15%|█▌        | 973/6470 [12:00<1:09:53,  1.31it/s]
4|974|Loss: 0.8375672698020935:  15%|█▌        | 973/6470 [12:00<1:09:53,  1.31it/s]
4|974|Loss: 0.8375672698020935:  15%|█▌        | 974/6470 [12:01<1:04:22,  1.42it/s]
4|975|Loss: 0.7498138546943665:  15%|█▌        | 974/6470 [12:01<1:04:22,  1.42it/s]
4|975|Loss: 0.7498138546943665:  15%|█▌        | 975/6470 [12:01<1:05:38,  1.40it/s]
4|976|Loss: 0.8341744542121887:  15%|█▌        | 975/6470 [12:02<1:05:38,  1.40it/s]
4|976|Loss: 0.8341744542121887:  15%|█▌        | 976/6470 [12:02<1:07:33,  1.36it/s]
4|977|Loss: 0.893774151802063:  15%|█▌        | 976/6470 [12:03<1:07:33,  1.36it/s] 
4|977|Loss: 0.893774151802063:  15%|█▌        | 977/6470 [12:03<1:05:48,  1.39it/s]
4|978|Loss: 1.0084645748138428:  15%|█▌        | 977/6470 [12:03<1:05:48,  1.39it/s]
4|978|Loss: 1.0084645748138428:  15%|█▌        | 978/6470 [12:04<1:06:47,  1.37it/s]
4|979|Loss: 0.7579783201217651:  15%|█▌        | 978/6470 [12:04<1:06:47,  1.37it/s]
4|979|Loss: 0.7579783201217651:  15%|█▌        | 979/6470 [12:04<1:07:30,  1.36it/s]
4|980|Loss: 0.8926861882209778:  15%|█▌        | 979/6470 [12:05<1:07:30,  1.36it/s]
4|980|Loss: 0.8926861882209778:  15%|█▌        | 980/6470 [12:05<1:08:55,  1.33it/s]
4|981|Loss: 0.8520171046257019:  15%|█▌        | 980/6470 [12:06<1:08:55,  1.33it/s]
4|981|Loss: 0.8520171046257019:  15%|█▌        | 981/6470 [12:06<1:07:58,  1.35it/s]
4|982|Loss: 0.831820011138916:  15%|█▌        | 981/6470 [12:06<1:07:58,  1.35it/s] 
4|982|Loss: 0.831820011138916:  15%|█▌        | 982/6470 [12:07<1:08:36,  1.33it/s]
4|983|Loss: 0.8369404673576355:  15%|█▌        | 982/6470 [12:07<1:08:36,  1.33it/s]
4|983|Loss: 0.8369404673576355:  15%|█▌        | 983/6470 [12:07<1:09:43,  1.31it/s]
4|984|Loss: 0.787009060382843:  15%|█▌        | 983/6470 [12:08<1:09:43,  1.31it/s] 
4|984|Loss: 0.787009060382843:  15%|█▌        | 984/6470 [12:08<1:10:43,  1.29it/s]
4|985|Loss: 0.7950665950775146:  15%|█▌        | 984/6470 [12:09<1:10:43,  1.29it/s]
4|985|Loss: 0.7950665950775146:  15%|█▌        | 985/6470 [12:09<1:07:27,  1.36it/s]
4|986|Loss: 0.8267322182655334:  15%|█▌        | 985/6470 [12:09<1:07:27,  1.36it/s]
4|986|Loss: 0.8267322182655334:  15%|█▌        | 986/6470 [12:10<1:05:53,  1.39it/s]
4|987|Loss: 0.7483841180801392:  15%|█▌        | 986/6470 [12:10<1:05:53,  1.39it/s]
4|987|Loss: 0.7483841180801392:  15%|█▌        | 987/6470 [12:10<1:06:26,  1.38it/s]
4|988|Loss: 0.6841374039649963:  15%|█▌        | 987/6470 [12:11<1:06:26,  1.38it/s]
4|988|Loss: 0.6841374039649963:  15%|█▌        | 988/6470 [12:11<1:06:06,  1.38it/s]
4|989|Loss: 0.686140775680542:  15%|█▌        | 988/6470 [12:11<1:06:06,  1.38it/s] 
4|989|Loss: 0.686140775680542:  15%|█▌        | 989/6470 [12:12<1:09:54,  1.31it/s]
4|990|Loss: 0.6935585737228394:  15%|█▌        | 989/6470 [12:12<1:09:54,  1.31it/s]
4|990|Loss: 0.6935585737228394:  15%|█▌        | 990/6470 [12:13<1:14:34,  1.22it/s]
4|991|Loss: 0.9259264469146729:  15%|█▌        | 990/6470 [12:13<1:14:34,  1.22it/s]
4|991|Loss: 0.9259264469146729:  15%|█▌        | 991/6470 [12:14<1:17:35,  1.18it/s]
4|992|Loss: 0.8603025674819946:  15%|█▌        | 991/6470 [12:14<1:17:35,  1.18it/s]
4|992|Loss: 0.8603025674819946:  15%|█▌        | 992/6470 [12:14<1:14:52,  1.22it/s]
4|993|Loss: 0.7762035131454468:  15%|█▌        | 992/6470 [12:15<1:14:52,  1.22it/s]
4|993|Loss: 0.7762035131454468:  15%|█▌        | 993/6470 [12:15<1:17:07,  1.18it/s]
4|994|Loss: 0.7497415542602539:  15%|█▌        | 993/6470 [12:16<1:17:07,  1.18it/s]
4|994|Loss: 0.7497415542602539:  15%|█▌        | 994/6470 [12:16<1:16:59,  1.19it/s]
4|995|Loss: 0.8432850241661072:  15%|█▌        | 994/6470 [12:17<1:16:59,  1.19it/s]
4|995|Loss: 0.8432850241661072:  15%|█▌        | 995/6470 [12:17<1:13:24,  1.24it/s]
4|996|Loss: 0.8130394816398621:  15%|█▌        | 995/6470 [12:17<1:13:24,  1.24it/s]
4|996|Loss: 0.8130394816398621:  15%|█▌        | 996/6470 [12:18<1:07:58,  1.34it/s]
4|997|Loss: 0.9311935901641846:  15%|█▌        | 996/6470 [12:18<1:07:58,  1.34it/s]
4|997|Loss: 0.9311935901641846:  15%|█▌        | 997/6470 [12:18<1:08:23,  1.33it/s]
4|998|Loss: 0.8296804428100586:  15%|█▌        | 997/6470 [12:19<1:08:23,  1.33it/s]
4|998|Loss: 0.8296804428100586:  15%|█▌        | 998/6470 [12:19<1:08:48,  1.33it/s]
4|999|Loss: 0.7770558595657349:  15%|█▌        | 998/6470 [12:19<1:08:48,  1.33it/s]
4|999|Loss: 0.7770558595657349:  15%|█▌        | 999/6470 [12:20<1:07:46,  1.35it/s]
4|1000|Loss: 0.8693813681602478:  15%|█▌        | 999/6470 [12:20<1:07:46,  1.35it/s]
4|1000|Loss: 0.8693813681602478:  15%|█▌        | 1000/6470 [12:20<1:03:42,  1.43it/s]
4|1001|Loss: 0.8420459628105164:  15%|█▌        | 1000/6470 [12:21<1:03:42,  1.43it/s]
4|1001|Loss: 0.8420459628105164:  15%|█▌        | 1001/6470 [12:21<1:05:23,  1.39it/s]
4|1002|Loss: 0.8323960900306702:  15%|█▌        | 1001/6470 [12:22<1:05:23,  1.39it/s]
4|1002|Loss: 0.8323960900306702:  15%|█▌        | 1002/6470 [12:22<1:07:41,  1.35it/s]
4|1003|Loss: 0.7914741039276123:  15%|█▌        | 1002/6470 [12:22<1:07:41,  1.35it/s]
4|1003|Loss: 0.7914741039276123:  16%|█▌        | 1003/6470 [12:22<1:02:14,  1.46it/s]
4|1004|Loss: 0.8259944319725037:  16%|█▌        | 1003/6470 [12:23<1:02:14,  1.46it/s]
4|1004|Loss: 0.8259944319725037:  16%|█▌        | 1004/6470 [12:23<1:03:11,  1.44it/s]
4|1005|Loss: 0.8089702725410461:  16%|█▌        | 1004/6470 [12:24<1:03:11,  1.44it/s]
4|1005|Loss: 0.8089702725410461:  16%|█▌        | 1005/6470 [12:24<1:07:29,  1.35it/s]
4|1006|Loss: 0.8707815408706665:  16%|█▌        | 1005/6470 [12:24<1:07:29,  1.35it/s]
4|1006|Loss: 0.8707815408706665:  16%|█▌        | 1006/6470 [12:25<1:07:32,  1.35it/s]
4|1007|Loss: 0.768463671207428:  16%|█▌        | 1006/6470 [12:25<1:07:32,  1.35it/s] 
4|1007|Loss: 0.768463671207428:  16%|█▌        | 1007/6470 [12:26<1:06:32,  1.37it/s]
4|1008|Loss: 0.9547364115715027:  16%|█▌        | 1007/6470 [12:26<1:06:32,  1.37it/s]
4|1008|Loss: 0.9547364115715027:  16%|█▌        | 1008/6470 [12:26<1:09:11,  1.32it/s]
4|1009|Loss: 0.8236809968948364:  16%|█▌        | 1008/6470 [12:27<1:09:11,  1.32it/s]
4|1009|Loss: 0.8236809968948364:  16%|█▌        | 1009/6470 [12:27<1:09:29,  1.31it/s]
4|1010|Loss: 0.9403937458992004:  16%|█▌        | 1009/6470 [12:27<1:09:29,  1.31it/s]
4|1010|Loss: 0.9403937458992004:  16%|█▌        | 1010/6470 [12:28<1:07:20,  1.35it/s]
4|1011|Loss: 0.7428346276283264:  16%|█▌        | 1010/6470 [12:28<1:07:20,  1.35it/s]
4|1011|Loss: 0.7428346276283264:  16%|█▌        | 1011/6470 [12:29<1:10:45,  1.29it/s]
4|1012|Loss: 0.8430352210998535:  16%|█▌        | 1011/6470 [12:29<1:10:45,  1.29it/s]
4|1012|Loss: 0.8430352210998535:  16%|█▌        | 1012/6470 [12:29<1:10:50,  1.28it/s]
4|1013|Loss: 0.852361261844635:  16%|█▌        | 1012/6470 [12:30<1:10:50,  1.28it/s] 
4|1013|Loss: 0.852361261844635:  16%|█▌        | 1013/6470 [12:30<1:07:21,  1.35it/s]
4|1014|Loss: 0.7282972931861877:  16%|█▌        | 1013/6470 [12:30<1:07:21,  1.35it/s]
4|1014|Loss: 0.7282972931861877:  16%|█▌        | 1014/6470 [12:31<1:06:57,  1.36it/s]
4|1015|Loss: 0.7939746379852295:  16%|█▌        | 1014/6470 [12:31<1:06:57,  1.36it/s]
4|1015|Loss: 0.7939746379852295:  16%|█▌        | 1015/6470 [12:32<1:07:57,  1.34it/s]
4|1016|Loss: 0.7624698281288147:  16%|█▌        | 1015/6470 [12:32<1:07:57,  1.34it/s]
4|1016|Loss: 0.7624698281288147:  16%|█▌        | 1016/6470 [12:32<1:04:24,  1.41it/s]
4|1017|Loss: 0.8670405745506287:  16%|█▌        | 1016/6470 [12:33<1:04:24,  1.41it/s]
4|1017|Loss: 0.8670405745506287:  16%|█▌        | 1017/6470 [12:33<1:08:19,  1.33it/s]
4|1018|Loss: 0.8512210845947266:  16%|█▌        | 1017/6470 [12:34<1:08:19,  1.33it/s]
4|1018|Loss: 0.8512210845947266:  16%|█▌        | 1018/6470 [12:34<1:10:33,  1.29it/s]
4|1019|Loss: 0.7578984498977661:  16%|█▌        | 1018/6470 [12:34<1:10:33,  1.29it/s]
4|1019|Loss: 0.7578984498977661:  16%|█▌        | 1019/6470 [12:35<1:06:24,  1.37it/s]
4|1020|Loss: 0.7784336805343628:  16%|█▌        | 1019/6470 [12:35<1:06:24,  1.37it/s]
4|1020|Loss: 0.7784336805343628:  16%|█▌        | 1020/6470 [12:35<1:04:14,  1.41it/s]
4|1021|Loss: 0.8603195548057556:  16%|█▌        | 1020/6470 [12:36<1:04:14,  1.41it/s]
4|1021|Loss: 0.8603195548057556:  16%|█▌        | 1021/6470 [12:36<1:07:53,  1.34it/s]
4|1022|Loss: 0.8251081705093384:  16%|█▌        | 1021/6470 [12:36<1:07:53,  1.34it/s]
4|1022|Loss: 0.8251081705093384:  16%|█▌        | 1022/6470 [12:37<1:11:50,  1.26it/s]
4|1023|Loss: 0.9739486575126648:  16%|█▌        | 1022/6470 [12:37<1:11:50,  1.26it/s]
4|1023|Loss: 0.9739486575126648:  16%|█▌        | 1023/6470 [12:38<1:08:50,  1.32it/s]
4|1024|Loss: 0.8687252998352051:  16%|█▌        | 1023/6470 [12:38<1:08:50,  1.32it/s]
4|1024|Loss: 0.8687252998352051:  16%|█▌        | 1024/6470 [12:38<1:10:23,  1.29it/s]
4|1025|Loss: 0.6407496929168701:  16%|█▌        | 1024/6470 [12:39<1:10:23,  1.29it/s]
4|1025|Loss: 0.6407496929168701:  16%|█▌        | 1025/6470 [12:39<1:11:33,  1.27it/s]
4|1026|Loss: 0.6846047043800354:  16%|█▌        | 1025/6470 [12:40<1:11:33,  1.27it/s]
4|1026|Loss: 0.6846047043800354:  16%|█▌        | 1026/6470 [12:40<1:06:51,  1.36it/s]
4|1027|Loss: 0.941990852355957:  16%|█▌        | 1026/6470 [12:40<1:06:51,  1.36it/s] 
4|1027|Loss: 0.941990852355957:  16%|█▌        | 1027/6470 [12:41<1:08:37,  1.32it/s]
4|1028|Loss: 0.7874335050582886:  16%|█▌        | 1027/6470 [12:41<1:08:37,  1.32it/s]
4|1028|Loss: 0.7874335050582886:  16%|█▌        | 1028/6470 [12:41<1:10:03,  1.29it/s]
4|1029|Loss: 0.8181101083755493:  16%|█▌        | 1028/6470 [12:42<1:10:03,  1.29it/s]
4|1029|Loss: 0.8181101083755493:  16%|█▌        | 1029/6470 [12:42<1:10:53,  1.28it/s]
4|1030|Loss: 0.7233690619468689:  16%|█▌        | 1029/6470 [12:43<1:10:53,  1.28it/s]
4|1030|Loss: 0.7233690619468689:  16%|█▌        | 1030/6470 [12:43<1:11:08,  1.27it/s]
4|1031|Loss: 0.8720899224281311:  16%|█▌        | 1030/6470 [12:43<1:11:08,  1.27it/s]
4|1031|Loss: 0.8720899224281311:  16%|█▌        | 1031/6470 [12:44<1:11:24,  1.27it/s]
4|1032|Loss: 0.8641990423202515:  16%|█▌        | 1031/6470 [12:44<1:11:24,  1.27it/s]
4|1032|Loss: 0.8641990423202515:  16%|█▌        | 1032/6470 [12:45<1:11:23,  1.27it/s]
4|1033|Loss: 0.8370313048362732:  16%|█▌        | 1032/6470 [12:45<1:11:23,  1.27it/s]
4|1033|Loss: 0.8370313048362732:  16%|█▌        | 1033/6470 [12:45<1:11:08,  1.27it/s]
4|1034|Loss: 0.8553738594055176:  16%|█▌        | 1033/6470 [12:46<1:11:08,  1.27it/s]
4|1034|Loss: 0.8553738594055176:  16%|█▌        | 1034/6470 [12:46<1:10:41,  1.28it/s]
4|1035|Loss: 0.7061870694160461:  16%|█▌        | 1034/6470 [12:47<1:10:41,  1.28it/s]
4|1035|Loss: 0.7061870694160461:  16%|█▌        | 1035/6470 [12:47<1:09:46,  1.30it/s]
4|1036|Loss: 0.829003095626831:  16%|█▌        | 1035/6470 [12:47<1:09:46,  1.30it/s] 
4|1036|Loss: 0.829003095626831:  16%|█▌        | 1036/6470 [12:48<1:10:01,  1.29it/s]
4|1037|Loss: 0.8777469992637634:  16%|█▌        | 1036/6470 [12:48<1:10:01,  1.29it/s]
4|1037|Loss: 0.8777469992637634:  16%|█▌        | 1037/6470 [12:49<1:10:52,  1.28it/s]
4|1038|Loss: 0.7316020131111145:  16%|█▌        | 1037/6470 [12:49<1:10:52,  1.28it/s]
4|1038|Loss: 0.7316020131111145:  16%|█▌        | 1038/6470 [12:49<1:15:54,  1.19it/s]
4|1039|Loss: 0.6405149698257446:  16%|█▌        | 1038/6470 [12:50<1:15:54,  1.19it/s]
4|1039|Loss: 0.6405149698257446:  16%|█▌        | 1039/6470 [12:50<1:13:02,  1.24it/s]
4|1040|Loss: 1.374046802520752:  16%|█▌        | 1039/6470 [12:51<1:13:02,  1.24it/s] 
4|1040|Loss: 1.374046802520752:  16%|█▌        | 1040/6470 [12:51<1:13:51,  1.23it/s]
4|1041|Loss: 0.8275312185287476:  16%|█▌        | 1040/6470 [12:51<1:13:51,  1.23it/s]
4|1041|Loss: 0.8275312185287476:  16%|█▌        | 1041/6470 [12:52<1:12:40,  1.25it/s]
4|1042|Loss: 0.6589670777320862:  16%|█▌        | 1041/6470 [12:52<1:12:40,  1.25it/s]
4|1042|Loss: 0.6589670777320862:  16%|█▌        | 1042/6470 [12:52<1:05:55,  1.37it/s]
4|1043|Loss: 0.8794853091239929:  16%|█▌        | 1042/6470 [12:53<1:05:55,  1.37it/s]
4|1043|Loss: 0.8794853091239929:  16%|█▌        | 1043/6470 [12:53<1:07:34,  1.34it/s]
4|1044|Loss: 0.5703882575035095:  16%|█▌        | 1043/6470 [12:54<1:07:34,  1.34it/s]
4|1044|Loss: 0.5703882575035095:  16%|█▌        | 1044/6470 [12:55<1:29:53,  1.01it/s]
4|1045|Loss: 0.7622454762458801:  16%|█▌        | 1044/6470 [12:55<1:29:53,  1.01it/s]
4|1045|Loss: 0.7622454762458801:  16%|█▌        | 1045/6470 [12:56<1:27:44,  1.03it/s]
4|1046|Loss: 0.7367291450500488:  16%|█▌        | 1045/6470 [12:56<1:27:44,  1.03it/s]
4|1046|Loss: 0.7367291450500488:  16%|█▌        | 1046/6470 [12:56<1:24:06,  1.07it/s]
4|1047|Loss: 0.7747135758399963:  16%|█▌        | 1046/6470 [12:57<1:24:06,  1.07it/s]
4|1047|Loss: 0.7747135758399963:  16%|█▌        | 1047/6470 [12:57<1:18:24,  1.15it/s]
4|1048|Loss: 0.7794704437255859:  16%|█▌        | 1047/6470 [12:58<1:18:24,  1.15it/s]
4|1048|Loss: 0.7794704437255859:  16%|█▌        | 1048/6470 [12:58<1:15:06,  1.20it/s]
4|1049|Loss: 0.7779130935668945:  16%|█▌        | 1048/6470 [12:58<1:15:06,  1.20it/s]
4|1049|Loss: 0.7779130935668945:  16%|█▌        | 1049/6470 [12:59<1:13:11,  1.23it/s]
4|1050|Loss: 0.7734444737434387:  16%|█▌        | 1049/6470 [12:59<1:13:11,  1.23it/s]
4|1050|Loss: 0.7734444737434387:  16%|█▌        | 1050/6470 [13:00<1:15:42,  1.19it/s]
4|1051|Loss: 0.690707266330719:  16%|█▌        | 1050/6470 [13:00<1:15:42,  1.19it/s] 
4|1051|Loss: 0.690707266330719:  16%|█▌        | 1051/6470 [13:00<1:06:32,  1.36it/s]
4|1052|Loss: 0.790343165397644:  16%|█▌        | 1051/6470 [13:00<1:06:32,  1.36it/s]
4|1052|Loss: 0.790343165397644:  16%|█▋        | 1052/6470 [13:01<58:53,  1.53it/s]  
4|1053|Loss: 0.8991800546646118:  16%|█▋        | 1052/6470 [13:01<58:53,  1.53it/s]
4|1053|Loss: 0.8991800546646118:  16%|█▋        | 1053/6470 [13:01<55:19,  1.63it/s]
4|1054|Loss: 0.8548673391342163:  16%|█▋        | 1053/6470 [13:02<55:19,  1.63it/s]
4|1054|Loss: 0.8548673391342163:  16%|█▋        | 1054/6470 [13:02<1:02:45,  1.44it/s]
4|1055|Loss: 0.7696655988693237:  16%|█▋        | 1054/6470 [13:02<1:02:45,  1.44it/s]
4|1055|Loss: 0.7696655988693237:  16%|█▋        | 1055/6470 [13:03<1:01:03,  1.48it/s]
4|1056|Loss: 0.9198808073997498:  16%|█▋        | 1055/6470 [13:03<1:01:03,  1.48it/s]
4|1056|Loss: 0.9198808073997498:  16%|█▋        | 1056/6470 [13:03<1:02:30,  1.44it/s]
4|1057|Loss: 0.8302757740020752:  16%|█▋        | 1056/6470 [13:04<1:02:30,  1.44it/s]
4|1057|Loss: 0.8302757740020752:  16%|█▋        | 1057/6470 [13:04<1:04:19,  1.40it/s]
4|1058|Loss: 0.7844552993774414:  16%|█▋        | 1057/6470 [13:04<1:04:19,  1.40it/s]
4|1058|Loss: 0.7844552993774414:  16%|█▋        | 1058/6470 [13:05<1:03:38,  1.42it/s]
4|1059|Loss: 0.7823039889335632:  16%|█▋        | 1058/6470 [13:05<1:03:38,  1.42it/s]
4|1059|Loss: 0.7823039889335632:  16%|█▋        | 1059/6470 [13:05<1:01:48,  1.46it/s]
4|1060|Loss: 0.8336520195007324:  16%|█▋        | 1059/6470 [13:06<1:01:48,  1.46it/s]
4|1060|Loss: 0.8336520195007324:  16%|█▋        | 1060/6470 [13:06<1:07:40,  1.33it/s]
4|1061|Loss: 0.9245319962501526:  16%|█▋        | 1060/6470 [13:07<1:07:40,  1.33it/s]
4|1061|Loss: 0.9245319962501526:  16%|█▋        | 1061/6470 [13:07<1:12:31,  1.24it/s]
4|1062|Loss: 0.7171484231948853:  16%|█▋        | 1061/6470 [13:08<1:12:31,  1.24it/s]
4|1062|Loss: 0.7171484231948853:  16%|█▋        | 1062/6470 [13:08<1:11:05,  1.27it/s]
4|1063|Loss: 0.8013713955879211:  16%|█▋        | 1062/6470 [13:08<1:11:05,  1.27it/s]
4|1063|Loss: 0.8013713955879211:  16%|█▋        | 1063/6470 [13:09<1:13:19,  1.23it/s]
4|1064|Loss: 0.780325174331665:  16%|█▋        | 1063/6470 [13:09<1:13:19,  1.23it/s] 
4|1064|Loss: 0.780325174331665:  16%|█▋        | 1064/6470 [13:09<1:07:19,  1.34it/s]
4|1065|Loss: 0.7634044289588928:  16%|█▋        | 1064/6470 [13:10<1:07:19,  1.34it/s]
4|1065|Loss: 0.7634044289588928:  16%|█▋        | 1065/6470 [13:10<1:04:28,  1.40it/s]
4|1066|Loss: 0.8635739088058472:  16%|█▋        | 1065/6470 [13:11<1:04:28,  1.40it/s]
4|1066|Loss: 0.8635739088058472:  16%|█▋        | 1066/6470 [13:11<1:06:22,  1.36it/s]
4|1067|Loss: 0.853157639503479:  16%|█▋        | 1066/6470 [13:11<1:06:22,  1.36it/s] 
4|1067|Loss: 0.853157639503479:  16%|█▋        | 1067/6470 [13:11<1:01:25,  1.47it/s]
4|1068|Loss: 0.7718223333358765:  16%|█▋        | 1067/6470 [13:12<1:01:25,  1.47it/s]
4|1068|Loss: 0.7718223333358765:  17%|█▋        | 1068/6470 [13:12<58:20,  1.54it/s]  
4|1069|Loss: 0.8737335801124573:  17%|█▋        | 1068/6470 [13:12<58:20,  1.54it/s]
4|1069|Loss: 0.8737335801124573:  17%|█▋        | 1069/6470 [13:13<1:00:11,  1.50it/s]
4|1070|Loss: 0.9531909823417664:  17%|█▋        | 1069/6470 [13:13<1:00:11,  1.50it/s]
4|1070|Loss: 0.9531909823417664:  17%|█▋        | 1070/6470 [13:13<59:01,  1.52it/s]  
4|1071|Loss: 0.7041446566581726:  17%|█▋        | 1070/6470 [13:14<59:01,  1.52it/s]
4|1071|Loss: 0.7041446566581726:  17%|█▋        | 1071/6470 [13:14<59:31,  1.51it/s]
4|1072|Loss: 0.6879333257675171:  17%|█▋        | 1071/6470 [13:14<59:31,  1.51it/s]
4|1072|Loss: 0.6879333257675171:  17%|█▋        | 1072/6470 [13:15<1:01:46,  1.46it/s]
4|1073|Loss: 0.8119572401046753:  17%|█▋        | 1072/6470 [13:15<1:01:46,  1.46it/s]
4|1073|Loss: 0.8119572401046753:  17%|█▋        | 1073/6470 [13:16<1:06:57,  1.34it/s]
4|1074|Loss: 0.7048894762992859:  17%|█▋        | 1073/6470 [13:16<1:06:57,  1.34it/s]
4|1074|Loss: 0.7048894762992859:  17%|█▋        | 1074/6470 [13:17<1:11:40,  1.25it/s]
4|1075|Loss: 0.7387048602104187:  17%|█▋        | 1074/6470 [13:17<1:11:40,  1.25it/s]
4|1075|Loss: 0.7387048602104187:  17%|█▋        | 1075/6470 [13:17<1:10:30,  1.28it/s]
4|1076|Loss: 0.8564977645874023:  17%|█▋        | 1075/6470 [13:18<1:10:30,  1.28it/s]
4|1076|Loss: 0.8564977645874023:  17%|█▋        | 1076/6470 [13:18<1:08:19,  1.32it/s]
4|1077|Loss: 0.8602068424224854:  17%|█▋        | 1076/6470 [13:18<1:08:19,  1.32it/s]
4|1077|Loss: 0.8602068424224854:  17%|█▋        | 1077/6470 [13:19<1:05:13,  1.38it/s]
4|1078|Loss: 0.8593742847442627:  17%|█▋        | 1077/6470 [13:19<1:05:13,  1.38it/s]
4|1078|Loss: 0.8593742847442627:  17%|█▋        | 1078/6470 [13:19<1:05:50,  1.36it/s]
4|1079|Loss: 0.7468849420547485:  17%|█▋        | 1078/6470 [13:20<1:05:50,  1.36it/s]
4|1079|Loss: 0.7468849420547485:  17%|█▋        | 1079/6470 [13:20<1:03:53,  1.41it/s]
4|1080|Loss: 0.7790297865867615:  17%|█▋        | 1079/6470 [13:20<1:03:53,  1.41it/s]
4|1080|Loss: 0.7790297865867615:  17%|█▋        | 1080/6470 [13:21<1:03:01,  1.43it/s]
4|1081|Loss: 0.6710362434387207:  17%|█▋        | 1080/6470 [13:21<1:03:01,  1.43it/s]
4|1081|Loss: 0.6710362434387207:  17%|█▋        | 1081/6470 [13:22<1:04:55,  1.38it/s]
4|1082|Loss: 0.8022007942199707:  17%|█▋        | 1081/6470 [13:22<1:04:55,  1.38it/s]
4|1082|Loss: 0.8022007942199707:  17%|█▋        | 1082/6470 [13:22<1:06:48,  1.34it/s]
4|1083|Loss: 0.9526239633560181:  17%|█▋        | 1082/6470 [13:23<1:06:48,  1.34it/s]
4|1083|Loss: 0.9526239633560181:  17%|█▋        | 1083/6470 [13:23<1:06:05,  1.36it/s]
4|1084|Loss: 0.9113252758979797:  17%|█▋        | 1083/6470 [13:24<1:06:05,  1.36it/s]
4|1084|Loss: 0.9113252758979797:  17%|█▋        | 1084/6470 [13:24<1:10:56,  1.27it/s]
4|1085|Loss: 0.8432502746582031:  17%|█▋        | 1084/6470 [13:24<1:10:56,  1.27it/s]
4|1085|Loss: 0.8432502746582031:  17%|█▋        | 1085/6470 [13:25<1:09:04,  1.30it/s]
4|1086|Loss: 0.873538076877594:  17%|█▋        | 1085/6470 [13:25<1:09:04,  1.30it/s] 
4|1086|Loss: 0.873538076877594:  17%|█▋        | 1086/6470 [13:25<1:08:00,  1.32it/s]
4|1087|Loss: 0.8140411972999573:  17%|█▋        | 1086/6470 [13:26<1:08:00,  1.32it/s]
4|1087|Loss: 0.8140411972999573:  17%|█▋        | 1087/6470 [13:26<1:11:45,  1.25it/s]
4|1088|Loss: 0.8789450526237488:  17%|█▋        | 1087/6470 [13:27<1:11:45,  1.25it/s]
4|1088|Loss: 0.8789450526237488:  17%|█▋        | 1088/6470 [13:27<1:12:39,  1.23it/s]
4|1089|Loss: 0.7952690720558167:  17%|█▋        | 1088/6470 [13:28<1:12:39,  1.23it/s]
4|1089|Loss: 0.7952690720558167:  17%|█▋        | 1089/6470 [13:28<1:13:44,  1.22it/s]
4|1090|Loss: 0.8354696035385132:  17%|█▋        | 1089/6470 [13:28<1:13:44,  1.22it/s]
4|1090|Loss: 0.8354696035385132:  17%|█▋        | 1090/6470 [13:29<1:07:25,  1.33it/s]
4|1091|Loss: 0.8786830902099609:  17%|█▋        | 1090/6470 [13:29<1:07:25,  1.33it/s]
4|1091|Loss: 0.8786830902099609:  17%|█▋        | 1091/6470 [13:29<1:09:51,  1.28it/s]
4|1092|Loss: 0.6954776048660278:  17%|█▋        | 1091/6470 [13:30<1:09:51,  1.28it/s]
4|1092|Loss: 0.6954776048660278:  17%|█▋        | 1092/6470 [13:30<1:10:16,  1.28it/s]
4|1093|Loss: 0.8397197723388672:  17%|█▋        | 1092/6470 [13:31<1:10:16,  1.28it/s]
4|1093|Loss: 0.8397197723388672:  17%|█▋        | 1093/6470 [13:31<1:08:35,  1.31it/s]
4|1094|Loss: 0.8660203814506531:  17%|█▋        | 1093/6470 [13:31<1:08:35,  1.31it/s]
4|1094|Loss: 0.8660203814506531:  17%|█▋        | 1094/6470 [13:32<1:11:53,  1.25it/s]
4|1095|Loss: 0.6230179667472839:  17%|█▋        | 1094/6470 [13:32<1:11:53,  1.25it/s]
4|1095|Loss: 0.6230179667472839:  17%|█▋        | 1095/6470 [13:33<1:15:33,  1.19it/s]
4|1096|Loss: 0.7855156660079956:  17%|█▋        | 1095/6470 [13:33<1:15:33,  1.19it/s]
4|1096|Loss: 0.7855156660079956:  17%|█▋        | 1096/6470 [13:34<1:15:43,  1.18it/s]
4|1097|Loss: 0.7596703767776489:  17%|█▋        | 1096/6470 [13:34<1:15:43,  1.18it/s]
4|1097|Loss: 0.7596703767776489:  17%|█▋        | 1097/6470 [13:34<1:14:37,  1.20it/s]
4|1098|Loss: 0.8241612911224365:  17%|█▋        | 1097/6470 [13:35<1:14:37,  1.20it/s]
4|1098|Loss: 0.8241612911224365:  17%|█▋        | 1098/6470 [13:35<1:15:43,  1.18it/s]
4|1099|Loss: 1.1028462648391724:  17%|█▋        | 1098/6470 [13:36<1:15:43,  1.18it/s]
4|1099|Loss: 1.1028462648391724:  17%|█▋        | 1099/6470 [13:36<1:16:51,  1.16it/s]
4|1100|Loss: 0.7837198972702026:  17%|█▋        | 1099/6470 [13:37<1:16:51,  1.16it/s]
4|1100|Loss: 0.7837198972702026:  17%|█▋        | 1100/6470 [13:37<1:09:29,  1.29it/s]
4|1101|Loss: 0.6915702819824219:  17%|█▋        | 1100/6470 [13:37<1:09:29,  1.29it/s]
4|1101|Loss: 0.6915702819824219:  17%|█▋        | 1101/6470 [13:37<1:04:59,  1.38it/s]
4|1102|Loss: 0.9174213409423828:  17%|█▋        | 1101/6470 [13:38<1:04:59,  1.38it/s]
4|1102|Loss: 0.9174213409423828:  17%|█▋        | 1102/6470 [13:38<59:13,  1.51it/s]  
4|1103|Loss: 0.9965564012527466:  17%|█▋        | 1102/6470 [13:38<59:13,  1.51it/s]
4|1103|Loss: 0.9965564012527466:  17%|█▋        | 1103/6470 [13:39<1:00:30,  1.48it/s]
4|1104|Loss: 0.8404546976089478:  17%|█▋        | 1103/6470 [13:39<1:00:30,  1.48it/s]
4|1104|Loss: 0.8404546976089478:  17%|█▋        | 1104/6470 [13:39<1:03:33,  1.41it/s]
4|1105|Loss: 0.9893719553947449:  17%|█▋        | 1104/6470 [13:40<1:03:33,  1.41it/s]
4|1105|Loss: 0.9893719553947449:  17%|█▋        | 1105/6470 [13:40<1:05:16,  1.37it/s]
4|1106|Loss: 0.8923233151435852:  17%|█▋        | 1105/6470 [13:41<1:05:16,  1.37it/s]
4|1106|Loss: 0.8923233151435852:  17%|█▋        | 1106/6470 [13:41<1:05:56,  1.36it/s]
4|1107|Loss: 0.7839664220809937:  17%|█▋        | 1106/6470 [13:41<1:05:56,  1.36it/s]
4|1107|Loss: 0.7839664220809937:  17%|█▋        | 1107/6470 [13:42<1:06:57,  1.33it/s]
4|1108|Loss: 0.7894318699836731:  17%|█▋        | 1107/6470 [13:42<1:06:57,  1.33it/s]
4|1108|Loss: 0.7894318699836731:  17%|█▋        | 1108/6470 [13:42<58:56,  1.52it/s]  
4|1109|Loss: 0.8320731520652771:  17%|█▋        | 1108/6470 [13:43<58:56,  1.52it/s]
4|1109|Loss: 0.8320731520652771:  17%|█▋        | 1109/6470 [13:43<59:49,  1.49it/s]
4|1110|Loss: 0.9859963655471802:  17%|█▋        | 1109/6470 [13:43<59:49,  1.49it/s]
4|1110|Loss: 0.9859963655471802:  17%|█▋        | 1110/6470 [13:44<1:06:18,  1.35it/s]
4|1111|Loss: 0.8084204792976379:  17%|█▋        | 1110/6470 [13:44<1:06:18,  1.35it/s]
4|1111|Loss: 0.8084204792976379:  17%|█▋        | 1111/6470 [13:45<1:06:13,  1.35it/s]
4|1112|Loss: 0.9668964147567749:  17%|█▋        | 1111/6470 [13:45<1:06:13,  1.35it/s]
4|1112|Loss: 0.9668964147567749:  17%|█▋        | 1112/6470 [13:45<1:07:13,  1.33it/s]
4|1113|Loss: 0.7940772771835327:  17%|█▋        | 1112/6470 [13:46<1:07:13,  1.33it/s]
4|1113|Loss: 0.7940772771835327:  17%|█▋        | 1113/6470 [13:46<1:07:35,  1.32it/s]
4|1114|Loss: 0.7990503907203674:  17%|█▋        | 1113/6470 [13:47<1:07:35,  1.32it/s]
4|1114|Loss: 0.7990503907203674:  17%|█▋        | 1114/6470 [13:47<1:08:32,  1.30it/s]
4|1115|Loss: 0.8532265424728394:  17%|█▋        | 1114/6470 [13:47<1:08:32,  1.30it/s]
4|1115|Loss: 0.8532265424728394:  17%|█▋        | 1115/6470 [13:48<1:12:45,  1.23it/s]
4|1116|Loss: 0.7313548922538757:  17%|█▋        | 1115/6470 [13:48<1:12:45,  1.23it/s]
4|1116|Loss: 0.7313548922538757:  17%|█▋        | 1116/6470 [13:49<1:11:28,  1.25it/s]
4|1117|Loss: 0.6762080788612366:  17%|█▋        | 1116/6470 [13:49<1:11:28,  1.25it/s]
4|1117|Loss: 0.6762080788612366:  17%|█▋        | 1117/6470 [13:49<1:12:43,  1.23it/s]
4|1118|Loss: 0.9060590267181396:  17%|█▋        | 1117/6470 [13:50<1:12:43,  1.23it/s]
4|1118|Loss: 0.9060590267181396:  17%|█▋        | 1118/6470 [13:50<1:09:01,  1.29it/s]
4|1119|Loss: 0.8621609807014465:  17%|█▋        | 1118/6470 [13:50<1:09:01,  1.29it/s]
4|1119|Loss: 0.8621609807014465:  17%|█▋        | 1119/6470 [13:51<1:07:31,  1.32it/s]
4|1120|Loss: 0.8654320240020752:  17%|█▋        | 1119/6470 [13:51<1:07:31,  1.32it/s]
4|1120|Loss: 0.8654320240020752:  17%|█▋        | 1120/6470 [13:52<1:07:10,  1.33it/s]
4|1121|Loss: 0.7575569748878479:  17%|█▋        | 1120/6470 [13:52<1:07:10,  1.33it/s]
4|1121|Loss: 0.7575569748878479:  17%|█▋        | 1121/6470 [13:52<1:05:35,  1.36it/s]
4|1122|Loss: 0.8102129697799683:  17%|█▋        | 1121/6470 [13:53<1:05:35,  1.36it/s]
4|1122|Loss: 0.8102129697799683:  17%|█▋        | 1122/6470 [13:53<1:05:42,  1.36it/s]
4|1123|Loss: 0.8804015517234802:  17%|█▋        | 1122/6470 [13:53<1:05:42,  1.36it/s]
4|1123|Loss: 0.8804015517234802:  17%|█▋        | 1123/6470 [13:54<1:06:13,  1.35it/s]
4|1124|Loss: 0.9061926007270813:  17%|█▋        | 1123/6470 [13:54<1:06:13,  1.35it/s]
4|1124|Loss: 0.9061926007270813:  17%|█▋        | 1124/6470 [13:55<1:08:02,  1.31it/s]
4|1125|Loss: 0.9287247657775879:  17%|█▋        | 1124/6470 [13:55<1:08:02,  1.31it/s]
4|1125|Loss: 0.9287247657775879:  17%|█▋        | 1125/6470 [13:55<1:12:50,  1.22it/s]
4|1126|Loss: 0.93141108751297:  17%|█▋        | 1125/6470 [13:56<1:12:50,  1.22it/s]  
4|1126|Loss: 0.93141108751297:  17%|█▋        | 1126/6470 [13:56<1:15:44,  1.18it/s]
4|1127|Loss: 0.7690038681030273:  17%|█▋        | 1126/6470 [13:57<1:15:44,  1.18it/s]
4|1127|Loss: 0.7690038681030273:  17%|█▋        | 1127/6470 [13:57<1:09:33,  1.28it/s]
4|1128|Loss: 0.9020141363143921:  17%|█▋        | 1127/6470 [13:57<1:09:33,  1.28it/s]
4|1128|Loss: 0.9020141363143921:  17%|█▋        | 1128/6470 [13:58<1:05:58,  1.35it/s]
4|1129|Loss: 0.8621684312820435:  17%|█▋        | 1128/6470 [13:58<1:05:58,  1.35it/s]
4|1129|Loss: 0.8621684312820435:  17%|█▋        | 1129/6470 [13:58<1:06:10,  1.35it/s]
4|1130|Loss: 0.7717357277870178:  17%|█▋        | 1129/6470 [13:59<1:06:10,  1.35it/s]
4|1130|Loss: 0.7717357277870178:  17%|█▋        | 1130/6470 [13:59<1:06:36,  1.34it/s]
4|1131|Loss: 0.8232749104499817:  17%|█▋        | 1130/6470 [14:00<1:06:36,  1.34it/s]
4|1131|Loss: 0.8232749104499817:  17%|█▋        | 1131/6470 [14:00<1:10:00,  1.27it/s]
4|1132|Loss: 0.9807244539260864:  17%|█▋        | 1131/6470 [14:01<1:10:00,  1.27it/s]
4|1132|Loss: 0.9807244539260864:  17%|█▋        | 1132/6470 [14:01<1:10:49,  1.26it/s]
4|1133|Loss: 0.9207720160484314:  17%|█▋        | 1132/6470 [14:01<1:10:49,  1.26it/s]
4|1133|Loss: 0.9207720160484314:  18%|█▊        | 1133/6470 [14:02<1:08:15,  1.30it/s]
4|1134|Loss: 0.724789559841156:  18%|█▊        | 1133/6470 [14:02<1:08:15,  1.30it/s] 
4|1134|Loss: 0.724789559841156:  18%|█▊        | 1134/6470 [14:02<1:10:34,  1.26it/s]
4|1135|Loss: 0.7124801874160767:  18%|█▊        | 1134/6470 [14:03<1:10:34,  1.26it/s]
4|1135|Loss: 0.7124801874160767:  18%|█▊        | 1135/6470 [14:03<1:08:15,  1.30it/s]
4|1136|Loss: 0.7738182544708252:  18%|█▊        | 1135/6470 [14:04<1:08:15,  1.30it/s]
4|1136|Loss: 0.7738182544708252:  18%|█▊        | 1136/6470 [14:04<1:09:53,  1.27it/s]
4|1137|Loss: 0.6748626232147217:  18%|█▊        | 1136/6470 [14:04<1:09:53,  1.27it/s]
4|1137|Loss: 0.6748626232147217:  18%|█▊        | 1137/6470 [14:05<1:11:55,  1.24it/s]
4|1138|Loss: 0.6278111934661865:  18%|█▊        | 1137/6470 [14:05<1:11:55,  1.24it/s]
4|1138|Loss: 0.6278111934661865:  18%|█▊        | 1138/6470 [14:06<1:12:10,  1.23it/s]
4|1139|Loss: 0.6966562867164612:  18%|█▊        | 1138/6470 [14:06<1:12:10,  1.23it/s]
4|1139|Loss: 0.6966562867164612:  18%|█▊        | 1139/6470 [14:06<1:11:00,  1.25it/s]
4|1140|Loss: 0.8159719109535217:  18%|█▊        | 1139/6470 [14:07<1:11:00,  1.25it/s]
4|1140|Loss: 0.8159719109535217:  18%|█▊        | 1140/6470 [14:07<1:13:52,  1.20it/s]
4|1141|Loss: 0.8218340873718262:  18%|█▊        | 1140/6470 [14:08<1:13:52,  1.20it/s]
4|1141|Loss: 0.8218340873718262:  18%|█▊        | 1141/6470 [14:08<1:08:46,  1.29it/s]
4|1142|Loss: 0.8270794749259949:  18%|█▊        | 1141/6470 [14:08<1:08:46,  1.29it/s]
4|1142|Loss: 0.8270794749259949:  18%|█▊        | 1142/6470 [14:09<1:07:18,  1.32it/s]
4|1143|Loss: 0.8338584303855896:  18%|█▊        | 1142/6470 [14:09<1:07:18,  1.32it/s]
4|1143|Loss: 0.8338584303855896:  18%|█▊        | 1143/6470 [14:10<1:09:59,  1.27it/s]
4|1144|Loss: 0.9159141778945923:  18%|█▊        | 1143/6470 [14:10<1:09:59,  1.27it/s]
4|1144|Loss: 0.9159141778945923:  18%|█▊        | 1144/6470 [14:10<1:12:59,  1.22it/s]
4|1145|Loss: 0.866112232208252:  18%|█▊        | 1144/6470 [14:11<1:12:59,  1.22it/s] 
4|1145|Loss: 0.866112232208252:  18%|█▊        | 1145/6470 [14:11<1:12:10,  1.23it/s]
4|1146|Loss: 0.7176641225814819:  18%|█▊        | 1145/6470 [14:12<1:12:10,  1.23it/s]
4|1146|Loss: 0.7176641225814819:  18%|█▊        | 1146/6470 [14:12<1:08:52,  1.29it/s]
4|1147|Loss: 0.8074071407318115:  18%|█▊        | 1146/6470 [14:12<1:08:52,  1.29it/s]
4|1147|Loss: 0.8074071407318115:  18%|█▊        | 1147/6470 [14:13<1:05:12,  1.36it/s]
4|1148|Loss: 0.8332763910293579:  18%|█▊        | 1147/6470 [14:13<1:05:12,  1.36it/s]
4|1148|Loss: 0.8332763910293579:  18%|█▊        | 1148/6470 [14:13<1:05:50,  1.35it/s]
4|1149|Loss: 0.7744536995887756:  18%|█▊        | 1148/6470 [14:14<1:05:50,  1.35it/s]
4|1149|Loss: 0.7744536995887756:  18%|█▊        | 1149/6470 [14:14<1:04:25,  1.38it/s]
4|1150|Loss: 0.7851229906082153:  18%|█▊        | 1149/6470 [14:14<1:04:25,  1.38it/s]
4|1150|Loss: 0.7851229906082153:  18%|█▊        | 1150/6470 [14:15<1:00:32,  1.46it/s]
4|1151|Loss: 1.0284110307693481:  18%|█▊        | 1150/6470 [14:15<1:00:32,  1.46it/s]
4|1151|Loss: 1.0284110307693481:  18%|█▊        | 1151/6470 [14:15<59:27,  1.49it/s]  
4|1152|Loss: 0.7660476565361023:  18%|█▊        | 1151/6470 [14:16<59:27,  1.49it/s]
4|1152|Loss: 0.7660476565361023:  18%|█▊        | 1152/6470 [14:16<59:53,  1.48it/s]
4|1153|Loss: 0.8488819599151611:  18%|█▊        | 1152/6470 [14:16<59:53,  1.48it/s]
4|1153|Loss: 0.8488819599151611:  18%|█▊        | 1153/6470 [14:17<57:30,  1.54it/s]
4|1154|Loss: 0.9482869505882263:  18%|█▊        | 1153/6470 [14:17<57:30,  1.54it/s]
4|1154|Loss: 0.9482869505882263:  18%|█▊        | 1154/6470 [14:17<1:00:10,  1.47it/s]
4|1155|Loss: 0.8573058843612671:  18%|█▊        | 1154/6470 [14:18<1:00:10,  1.47it/s]
4|1155|Loss: 0.8573058843612671:  18%|█▊        | 1155/6470 [14:18<1:05:14,  1.36it/s]
4|1156|Loss: 0.8480603098869324:  18%|█▊        | 1155/6470 [14:19<1:05:14,  1.36it/s]
4|1156|Loss: 0.8480603098869324:  18%|█▊        | 1156/6470 [14:19<1:06:22,  1.33it/s]
4|1157|Loss: 0.7815740704536438:  18%|█▊        | 1156/6470 [14:19<1:06:22,  1.33it/s]
4|1157|Loss: 0.7815740704536438:  18%|█▊        | 1157/6470 [14:20<1:05:20,  1.36it/s]
4|1158|Loss: 0.9161911606788635:  18%|█▊        | 1157/6470 [14:20<1:05:20,  1.36it/s]
4|1158|Loss: 0.9161911606788635:  18%|█▊        | 1158/6470 [14:20<1:03:46,  1.39it/s]
4|1159|Loss: 0.7629721760749817:  18%|█▊        | 1158/6470 [14:21<1:03:46,  1.39it/s]
4|1159|Loss: 0.7629721760749817:  18%|█▊        | 1159/6470 [14:21<1:10:36,  1.25it/s]
4|1160|Loss: 0.6833993196487427:  18%|█▊        | 1159/6470 [14:22<1:10:36,  1.25it/s]
4|1160|Loss: 0.6833993196487427:  18%|█▊        | 1160/6470 [14:22<1:05:19,  1.35it/s]
4|1161|Loss: 0.6864874362945557:  18%|█▊        | 1160/6470 [14:22<1:05:19,  1.35it/s]
4|1161|Loss: 0.6864874362945557:  18%|█▊        | 1161/6470 [14:22<1:01:04,  1.45it/s]
4|1162|Loss: 0.9774573445320129:  18%|█▊        | 1161/6470 [14:23<1:01:04,  1.45it/s]
4|1162|Loss: 0.9774573445320129:  18%|█▊        | 1162/6470 [14:23<1:00:47,  1.46it/s]
4|1163|Loss: 0.8525024056434631:  18%|█▊        | 1162/6470 [14:24<1:00:47,  1.46it/s]
4|1163|Loss: 0.8525024056434631:  18%|█▊        | 1163/6470 [14:24<1:03:02,  1.40it/s]
4|1164|Loss: 0.7763397693634033:  18%|█▊        | 1163/6470 [14:24<1:03:02,  1.40it/s]
4|1164|Loss: 0.7763397693634033:  18%|█▊        | 1164/6470 [14:25<1:05:13,  1.36it/s]
4|1165|Loss: 0.8827177882194519:  18%|█▊        | 1164/6470 [14:25<1:05:13,  1.36it/s]
4|1165|Loss: 0.8827177882194519:  18%|█▊        | 1165/6470 [14:25<1:06:17,  1.33it/s]
4|1166|Loss: 0.7630476951599121:  18%|█▊        | 1165/6470 [14:26<1:06:17,  1.33it/s]
4|1166|Loss: 0.7630476951599121:  18%|█▊        | 1166/6470 [14:26<1:10:41,  1.25it/s]
4|1167|Loss: 0.8623405694961548:  18%|█▊        | 1166/6470 [14:27<1:10:41,  1.25it/s]
4|1167|Loss: 0.8623405694961548:  18%|█▊        | 1167/6470 [14:27<1:08:45,  1.29it/s]
4|1168|Loss: 0.7886019945144653:  18%|█▊        | 1167/6470 [14:28<1:08:45,  1.29it/s]
4|1168|Loss: 0.7886019945144653:  18%|█▊        | 1168/6470 [14:28<1:11:00,  1.24it/s]
4|1169|Loss: 0.8159909248352051:  18%|█▊        | 1168/6470 [14:28<1:11:00,  1.24it/s]
4|1169|Loss: 0.8159909248352051:  18%|█▊        | 1169/6470 [14:29<1:10:15,  1.26it/s]
4|1170|Loss: 0.7928725481033325:  18%|█▊        | 1169/6470 [14:29<1:10:15,  1.26it/s]
4|1170|Loss: 0.7928725481033325:  18%|█▊        | 1170/6470 [14:30<1:10:15,  1.26it/s]
4|1171|Loss: 0.8538004755973816:  18%|█▊        | 1170/6470 [14:30<1:10:15,  1.26it/s]
4|1171|Loss: 0.8538004755973816:  18%|█▊        | 1171/6470 [14:30<1:05:45,  1.34it/s]
4|1172|Loss: 0.665126621723175:  18%|█▊        | 1171/6470 [14:31<1:05:45,  1.34it/s] 
4|1172|Loss: 0.665126621723175:  18%|█▊        | 1172/6470 [14:31<1:03:44,  1.39it/s]
4|1173|Loss: 0.8416582942008972:  18%|█▊        | 1172/6470 [14:31<1:03:44,  1.39it/s]
4|1173|Loss: 0.8416582942008972:  18%|█▊        | 1173/6470 [14:32<1:04:24,  1.37it/s]
4|1174|Loss: 0.8284921646118164:  18%|█▊        | 1173/6470 [14:32<1:04:24,  1.37it/s]
4|1174|Loss: 0.8284921646118164:  18%|█▊        | 1174/6470 [14:32<1:04:03,  1.38it/s]
4|1175|Loss: 0.904323160648346:  18%|█▊        | 1174/6470 [14:33<1:04:03,  1.38it/s] 
4|1175|Loss: 0.904323160648346:  18%|█▊        | 1175/6470 [14:33<1:05:38,  1.34it/s]
4|1176|Loss: 0.7959201335906982:  18%|█▊        | 1175/6470 [14:34<1:05:38,  1.34it/s]
4|1176|Loss: 0.7959201335906982:  18%|█▊        | 1176/6470 [14:34<1:03:25,  1.39it/s]
4|1177|Loss: 0.905471920967102:  18%|█▊        | 1176/6470 [14:34<1:03:25,  1.39it/s] 
4|1177|Loss: 0.905471920967102:  18%|█▊        | 1177/6470 [14:34<1:02:18,  1.42it/s]
4|1178|Loss: 0.8208345770835876:  18%|█▊        | 1177/6470 [14:35<1:02:18,  1.42it/s]
4|1178|Loss: 0.8208345770835876:  18%|█▊        | 1178/6470 [14:35<1:00:28,  1.46it/s]
4|1179|Loss: 0.9785987138748169:  18%|█▊        | 1178/6470 [14:35<1:00:28,  1.46it/s]
4|1179|Loss: 0.9785987138748169:  18%|█▊        | 1179/6470 [14:36<59:35,  1.48it/s]  
4|1180|Loss: 0.8394034504890442:  18%|█▊        | 1179/6470 [14:36<59:35,  1.48it/s]
4|1180|Loss: 0.8394034504890442:  18%|█▊        | 1180/6470 [14:37<1:05:19,  1.35it/s]
4|1181|Loss: 0.8786845803260803:  18%|█▊        | 1180/6470 [14:37<1:05:19,  1.35it/s]
4|1181|Loss: 0.8786845803260803:  18%|█▊        | 1181/6470 [14:38<1:09:13,  1.27it/s]
4|1182|Loss: 0.7518749237060547:  18%|█▊        | 1181/6470 [14:38<1:09:13,  1.27it/s]
4|1182|Loss: 0.7518749237060547:  18%|█▊        | 1182/6470 [14:38<1:07:44,  1.30it/s]
4|1183|Loss: 0.8155100345611572:  18%|█▊        | 1182/6470 [14:39<1:07:44,  1.30it/s]
4|1183|Loss: 0.8155100345611572:  18%|█▊        | 1183/6470 [14:39<1:08:09,  1.29it/s]
4|1184|Loss: 0.8614198565483093:  18%|█▊        | 1183/6470 [14:39<1:08:09,  1.29it/s]
4|1184|Loss: 0.8614198565483093:  18%|█▊        | 1184/6470 [14:40<1:06:40,  1.32it/s]
4|1185|Loss: 0.7515748143196106:  18%|█▊        | 1184/6470 [14:40<1:06:40,  1.32it/s]
4|1185|Loss: 0.7515748143196106:  18%|█▊        | 1185/6470 [14:41<1:07:12,  1.31it/s]
4|1186|Loss: 0.8448191285133362:  18%|█▊        | 1185/6470 [14:41<1:07:12,  1.31it/s]
4|1186|Loss: 0.8448191285133362:  18%|█▊        | 1186/6470 [14:41<1:11:33,  1.23it/s]
4|1187|Loss: 0.7886976599693298:  18%|█▊        | 1186/6470 [14:42<1:11:33,  1.23it/s]
4|1187|Loss: 0.7886976599693298:  18%|█▊        | 1187/6470 [14:42<1:12:07,  1.22it/s]
4|1188|Loss: 0.8414890766143799:  18%|█▊        | 1187/6470 [14:43<1:12:07,  1.22it/s]
4|1188|Loss: 0.8414890766143799:  18%|█▊        | 1188/6470 [14:43<1:10:42,  1.25it/s]
4|1189|Loss: 0.7757285833358765:  18%|█▊        | 1188/6470 [14:43<1:10:42,  1.25it/s]
4|1189|Loss: 0.7757285833358765:  18%|█▊        | 1189/6470 [14:44<1:07:21,  1.31it/s]
4|1190|Loss: 0.8694002032279968:  18%|█▊        | 1189/6470 [14:44<1:07:21,  1.31it/s]
4|1190|Loss: 0.8694002032279968:  18%|█▊        | 1190/6470 [14:44<1:06:30,  1.32it/s]
4|1191|Loss: 0.7839235663414001:  18%|█▊        | 1190/6470 [14:45<1:06:30,  1.32it/s]
4|1191|Loss: 0.7839235663414001:  18%|█▊        | 1191/6470 [14:45<1:05:00,  1.35it/s]
4|1192|Loss: 0.8617229461669922:  18%|█▊        | 1191/6470 [14:46<1:05:00,  1.35it/s]
4|1192|Loss: 0.8617229461669922:  18%|█▊        | 1192/6470 [14:46<59:07,  1.49it/s]  
4|1193|Loss: 0.7632171511650085:  18%|█▊        | 1192/6470 [14:46<59:07,  1.49it/s]
4|1193|Loss: 0.7632171511650085:  18%|█▊        | 1193/6470 [14:46<55:16,  1.59it/s]
4|1194|Loss: 0.7730128765106201:  18%|█▊        | 1193/6470 [14:46<55:16,  1.59it/s]
4|1194|Loss: 0.7730128765106201:  18%|█▊        | 1194/6470 [14:47<50:06,  1.76it/s]
4|1195|Loss: 0.8376869559288025:  18%|█▊        | 1194/6470 [14:47<50:06,  1.76it/s]
4|1195|Loss: 0.8376869559288025:  18%|█▊        | 1195/6470 [14:47<51:41,  1.70it/s]
4|1196|Loss: 0.7738021612167358:  18%|█▊        | 1195/6470 [14:48<51:41,  1.70it/s]
4|1196|Loss: 0.7738021612167358:  18%|█▊        | 1196/6470 [14:48<53:19,  1.65it/s]
4|1197|Loss: 0.8611698150634766:  18%|█▊        | 1196/6470 [14:48<53:19,  1.65it/s]
4|1197|Loss: 0.8611698150634766:  19%|█▊        | 1197/6470 [14:49<55:35,  1.58it/s]
4|1198|Loss: 0.8068439364433289:  19%|█▊        | 1197/6470 [14:49<55:35,  1.58it/s]
4|1198|Loss: 0.8068439364433289:  19%|█▊        | 1198/6470 [14:49<59:42,  1.47it/s]
4|1199|Loss: 0.8111881017684937:  19%|█▊        | 1198/6470 [14:50<59:42,  1.47it/s]
4|1199|Loss: 0.8111881017684937:  19%|█▊        | 1199/6470 [14:50<1:03:09,  1.39it/s]
4|1200|Loss: 0.8947739601135254:  19%|█▊        | 1199/6470 [14:51<1:03:09,  1.39it/s]
4|1200|Loss: 0.8947739601135254:  19%|█▊        | 1200/6470 [14:51<1:06:49,  1.31it/s]
4|1201|Loss: 0.8843374848365784:  19%|█▊        | 1200/6470 [14:52<1:06:49,  1.31it/s]
4|1201|Loss: 0.8843374848365784:  19%|█▊        | 1201/6470 [14:52<1:08:39,  1.28it/s]
4|1202|Loss: 0.6440024375915527:  19%|█▊        | 1201/6470 [14:52<1:08:39,  1.28it/s]
4|1202|Loss: 0.6440024375915527:  19%|█▊        | 1202/6470 [14:53<1:05:49,  1.33it/s]
4|1203|Loss: 0.7551230788230896:  19%|█▊        | 1202/6470 [14:53<1:05:49,  1.33it/s]
4|1203|Loss: 0.7551230788230896:  19%|█▊        | 1203/6470 [14:53<1:03:54,  1.37it/s]
4|1204|Loss: 0.7200168371200562:  19%|█▊        | 1203/6470 [14:54<1:03:54,  1.37it/s]
4|1204|Loss: 0.7200168371200562:  19%|█▊        | 1204/6470 [14:54<1:00:48,  1.44it/s]
4|1205|Loss: 0.8784139156341553:  19%|█▊        | 1204/6470 [14:54<1:00:48,  1.44it/s]
4|1205|Loss: 0.8784139156341553:  19%|█▊        | 1205/6470 [14:54<56:55,  1.54it/s]  
4|1206|Loss: 0.8484113812446594:  19%|█▊        | 1205/6470 [14:55<56:55,  1.54it/s]
4|1206|Loss: 0.8484113812446594:  19%|█▊        | 1206/6470 [14:55<58:33,  1.50it/s]
4|1207|Loss: 0.8882936835289001:  19%|█▊        | 1206/6470 [14:56<58:33,  1.50it/s]
4|1207|Loss: 0.8882936835289001:  19%|█▊        | 1207/6470 [14:56<1:01:32,  1.43it/s]
4|1208|Loss: 0.779495894908905:  19%|█▊        | 1207/6470 [14:56<1:01:32,  1.43it/s] 
4|1208|Loss: 0.779495894908905:  19%|█▊        | 1208/6470 [14:57<59:30,  1.47it/s]  
4|1209|Loss: 0.8755274415016174:  19%|█▊        | 1208/6470 [14:57<59:30,  1.47it/s]
4|1209|Loss: 0.8755274415016174:  19%|█▊        | 1209/6470 [14:57<59:20,  1.48it/s]
4|1210|Loss: 0.8037373423576355:  19%|█▊        | 1209/6470 [14:58<59:20,  1.48it/s]
4|1210|Loss: 0.8037373423576355:  19%|█▊        | 1210/6470 [14:58<59:18,  1.48it/s]
4|1211|Loss: 0.845386266708374:  19%|█▊        | 1210/6470 [14:58<59:18,  1.48it/s] 
4|1211|Loss: 0.845386266708374:  19%|█▊        | 1211/6470 [14:59<1:01:43,  1.42it/s]
4|1212|Loss: 0.6892693638801575:  19%|█▊        | 1211/6470 [14:59<1:01:43,  1.42it/s]
4|1212|Loss: 0.6892693638801575:  19%|█▊        | 1212/6470 [14:59<1:05:13,  1.34it/s]
4|1213|Loss: 0.8663884997367859:  19%|█▊        | 1212/6470 [15:00<1:05:13,  1.34it/s]
4|1213|Loss: 0.8663884997367859:  19%|█▊        | 1213/6470 [15:00<1:06:03,  1.33it/s]
4|1214|Loss: 0.832810640335083:  19%|█▊        | 1213/6470 [15:01<1:06:03,  1.33it/s] 
4|1214|Loss: 0.832810640335083:  19%|█▉        | 1214/6470 [15:01<1:08:14,  1.28it/s]
4|1215|Loss: 0.8615590333938599:  19%|█▉        | 1214/6470 [15:02<1:08:14,  1.28it/s]
4|1215|Loss: 0.8615590333938599:  19%|█▉        | 1215/6470 [15:02<1:11:26,  1.23it/s]
4|1216|Loss: 0.7798892855644226:  19%|█▉        | 1215/6470 [15:03<1:11:26,  1.23it/s]
4|1216|Loss: 0.7798892855644226:  19%|█▉        | 1216/6470 [15:03<1:16:08,  1.15it/s]
4|1217|Loss: 0.7359407544136047:  19%|█▉        | 1216/6470 [15:04<1:16:08,  1.15it/s]
4|1217|Loss: 0.7359407544136047:  19%|█▉        | 1217/6470 [15:04<1:16:05,  1.15it/s]
4|1218|Loss: 0.80219966173172:  19%|█▉        | 1217/6470 [15:04<1:16:05,  1.15it/s]  
4|1218|Loss: 0.80219966173172:  19%|█▉        | 1218/6470 [15:05<1:15:44,  1.16it/s]
4|1219|Loss: 0.8969675302505493:  19%|█▉        | 1218/6470 [15:05<1:15:44,  1.16it/s]
4|1219|Loss: 0.8969675302505493:  19%|█▉        | 1219/6470 [15:06<1:14:17,  1.18it/s]
4|1220|Loss: 0.7790826559066772:  19%|█▉        | 1219/6470 [15:06<1:14:17,  1.18it/s]
4|1220|Loss: 0.7790826559066772:  19%|█▉        | 1220/6470 [15:06<1:10:41,  1.24it/s]
4|1221|Loss: 0.8487012982368469:  19%|█▉        | 1220/6470 [15:07<1:10:41,  1.24it/s]
4|1221|Loss: 0.8487012982368469:  19%|█▉        | 1221/6470 [15:07<1:10:22,  1.24it/s]
4|1222|Loss: 0.9087368845939636:  19%|█▉        | 1221/6470 [15:07<1:10:22,  1.24it/s]
4|1222|Loss: 0.9087368845939636:  19%|█▉        | 1222/6470 [15:08<1:10:10,  1.25it/s]
4|1223|Loss: 0.8167430758476257:  19%|█▉        | 1222/6470 [15:08<1:10:10,  1.25it/s]
4|1223|Loss: 0.8167430758476257:  19%|█▉        | 1223/6470 [15:08<1:05:02,  1.34it/s]
4|1224|Loss: 0.8441992402076721:  19%|█▉        | 1223/6470 [15:09<1:05:02,  1.34it/s]
4|1224|Loss: 0.8441992402076721:  19%|█▉        | 1224/6470 [15:09<1:05:19,  1.34it/s]
4|1225|Loss: 0.7416380047798157:  19%|█▉        | 1224/6470 [15:10<1:05:19,  1.34it/s]
4|1225|Loss: 0.7416380047798157:  19%|█▉        | 1225/6470 [15:10<1:04:08,  1.36it/s]
4|1226|Loss: 0.9357408285140991:  19%|█▉        | 1225/6470 [15:10<1:04:08,  1.36it/s]
4|1226|Loss: 0.9357408285140991:  19%|█▉        | 1226/6470 [15:11<1:02:21,  1.40it/s]
4|1227|Loss: 1.0043503046035767:  19%|█▉        | 1226/6470 [15:11<1:02:21,  1.40it/s]
4|1227|Loss: 1.0043503046035767:  19%|█▉        | 1227/6470 [15:11<59:32,  1.47it/s]  
4|1228|Loss: 0.8308904767036438:  19%|█▉        | 1227/6470 [15:12<59:32,  1.47it/s]
4|1228|Loss: 0.8308904767036438:  19%|█▉        | 1228/6470 [15:12<58:54,  1.48it/s]
4|1229|Loss: 0.8343703150749207:  19%|█▉        | 1228/6470 [15:12<58:54,  1.48it/s]
4|1229|Loss: 0.8343703150749207:  19%|█▉        | 1229/6470 [15:13<1:00:57,  1.43it/s]
4|1230|Loss: 0.7866750359535217:  19%|█▉        | 1229/6470 [15:13<1:00:57,  1.43it/s]
4|1230|Loss: 0.7866750359535217:  19%|█▉        | 1230/6470 [15:13<1:02:14,  1.40it/s]
4|1231|Loss: 0.9129685163497925:  19%|█▉        | 1230/6470 [15:14<1:02:14,  1.40it/s]
4|1231|Loss: 0.9129685163497925:  19%|█▉        | 1231/6470 [15:14<59:25,  1.47it/s]  
4|1232|Loss: 0.8580056428909302:  19%|█▉        | 1231/6470 [15:14<59:25,  1.47it/s]
4|1232|Loss: 0.8580056428909302:  19%|█▉        | 1232/6470 [15:15<1:00:23,  1.45it/s]
4|1233|Loss: 0.9137120842933655:  19%|█▉        | 1232/6470 [15:15<1:00:23,  1.45it/s]
4|1233|Loss: 0.9137120842933655:  19%|█▉        | 1233/6470 [15:15<59:33,  1.47it/s]  
4|1234|Loss: 0.8681668043136597:  19%|█▉        | 1233/6470 [15:16<59:33,  1.47it/s]
4|1234|Loss: 0.8681668043136597:  19%|█▉        | 1234/6470 [15:16<1:03:34,  1.37it/s]
4|1235|Loss: 0.7470893859863281:  19%|█▉        | 1234/6470 [15:17<1:03:34,  1.37it/s]
4|1235|Loss: 0.7470893859863281:  19%|█▉        | 1235/6470 [15:17<1:04:57,  1.34it/s]
4|1236|Loss: 0.7988431453704834:  19%|█▉        | 1235/6470 [15:17<1:04:57,  1.34it/s]
4|1236|Loss: 0.7988431453704834:  19%|█▉        | 1236/6470 [15:18<1:05:23,  1.33it/s]
4|1237|Loss: 0.8194697499275208:  19%|█▉        | 1236/6470 [15:18<1:05:23,  1.33it/s]
4|1237|Loss: 0.8194697499275208:  19%|█▉        | 1237/6470 [15:18<1:03:44,  1.37it/s]
4|1238|Loss: 0.8443198800086975:  19%|█▉        | 1237/6470 [15:19<1:03:44,  1.37it/s]
4|1238|Loss: 0.8443198800086975:  19%|█▉        | 1238/6470 [15:19<1:04:58,  1.34it/s]
4|1239|Loss: 0.8807807564735413:  19%|█▉        | 1238/6470 [15:20<1:04:58,  1.34it/s]
4|1239|Loss: 0.8807807564735413:  19%|█▉        | 1239/6470 [15:20<1:06:36,  1.31it/s]
4|1240|Loss: 0.8460238575935364:  19%|█▉        | 1239/6470 [15:20<1:06:36,  1.31it/s]
4|1240|Loss: 0.8460238575935364:  19%|█▉        | 1240/6470 [15:21<1:10:06,  1.24it/s]
4|1241|Loss: 0.7057288885116577:  19%|█▉        | 1240/6470 [15:21<1:10:06,  1.24it/s]
4|1241|Loss: 0.7057288885116577:  19%|█▉        | 1241/6470 [15:22<1:13:57,  1.18it/s]
4|1242|Loss: 0.9497042298316956:  19%|█▉        | 1241/6470 [15:22<1:13:57,  1.18it/s]
4|1242|Loss: 0.9497042298316956:  19%|█▉        | 1242/6470 [15:23<1:11:14,  1.22it/s]
4|1243|Loss: 0.8155965209007263:  19%|█▉        | 1242/6470 [15:23<1:11:14,  1.22it/s]
4|1243|Loss: 0.8155965209007263:  19%|█▉        | 1243/6470 [15:23<1:08:13,  1.28it/s]
4|1244|Loss: 0.7530734539031982:  19%|█▉        | 1243/6470 [15:24<1:08:13,  1.28it/s]
4|1244|Loss: 0.7530734539031982:  19%|█▉        | 1244/6470 [15:24<1:08:17,  1.28it/s]
4|1245|Loss: 1.0048582553863525:  19%|█▉        | 1244/6470 [15:24<1:08:17,  1.28it/s]
4|1245|Loss: 1.0048582553863525:  19%|█▉        | 1245/6470 [15:25<1:06:41,  1.31it/s]
4|1246|Loss: 1.031406283378601:  19%|█▉        | 1245/6470 [15:25<1:06:41,  1.31it/s] 
4|1246|Loss: 1.031406283378601:  19%|█▉        | 1246/6470 [15:26<1:07:12,  1.30it/s]
4|1247|Loss: 0.9657678008079529:  19%|█▉        | 1246/6470 [15:26<1:07:12,  1.30it/s]
4|1247|Loss: 0.9657678008079529:  19%|█▉        | 1247/6470 [15:26<1:05:08,  1.34it/s]
4|1248|Loss: 0.7976193428039551:  19%|█▉        | 1247/6470 [15:27<1:05:08,  1.34it/s]
4|1248|Loss: 0.7976193428039551:  19%|█▉        | 1248/6470 [15:27<1:04:56,  1.34it/s]
4|1249|Loss: 0.8180097937583923:  19%|█▉        | 1248/6470 [15:27<1:04:56,  1.34it/s]
4|1249|Loss: 0.8180097937583923:  19%|█▉        | 1249/6470 [15:28<1:07:29,  1.29it/s]
4|1250|Loss: 0.9820300340652466:  19%|█▉        | 1249/6470 [15:28<1:07:29,  1.29it/s]
4|1250|Loss: 0.9820300340652466:  19%|█▉        | 1250/6470 [15:29<1:06:19,  1.31it/s]
4|1251|Loss: 0.814081072807312:  19%|█▉        | 1250/6470 [15:29<1:06:19,  1.31it/s] 
4|1251|Loss: 0.814081072807312:  19%|█▉        | 1251/6470 [15:29<1:01:13,  1.42it/s]
4|1252|Loss: 0.8467079997062683:  19%|█▉        | 1251/6470 [15:30<1:01:13,  1.42it/s]
4|1252|Loss: 0.8467079997062683:  19%|█▉        | 1252/6470 [15:30<1:01:32,  1.41it/s]
4|1253|Loss: 0.7777412533760071:  19%|█▉        | 1252/6470 [15:30<1:01:32,  1.41it/s]
4|1253|Loss: 0.7777412533760071:  19%|█▉        | 1253/6470 [15:31<1:05:43,  1.32it/s]
4|1254|Loss: 0.905362069606781:  19%|█▉        | 1253/6470 [15:31<1:05:43,  1.32it/s] 
4|1254|Loss: 0.905362069606781:  19%|█▉        | 1254/6470 [15:31<1:04:34,  1.35it/s]
4|1255|Loss: 0.7855010628700256:  19%|█▉        | 1254/6470 [15:32<1:04:34,  1.35it/s]
4|1255|Loss: 0.7855010628700256:  19%|█▉        | 1255/6470 [15:32<1:05:03,  1.34it/s]
4|1256|Loss: 0.7998024821281433:  19%|█▉        | 1255/6470 [15:33<1:05:03,  1.34it/s]
4|1256|Loss: 0.7998024821281433:  19%|█▉        | 1256/6470 [15:33<1:01:57,  1.40it/s]
4|1257|Loss: 0.8349491953849792:  19%|█▉        | 1256/6470 [15:33<1:01:57,  1.40it/s]
4|1257|Loss: 0.8349491953849792:  19%|█▉        | 1257/6470 [15:34<1:03:32,  1.37it/s]
4|1258|Loss: 0.8078402876853943:  19%|█▉        | 1257/6470 [15:34<1:03:32,  1.37it/s]
4|1258|Loss: 0.8078402876853943:  19%|█▉        | 1258/6470 [15:34<1:05:04,  1.33it/s]
4|1259|Loss: 0.8390306830406189:  19%|█▉        | 1258/6470 [15:35<1:05:04,  1.33it/s]
4|1259|Loss: 0.8390306830406189:  19%|█▉        | 1259/6470 [15:35<1:01:27,  1.41it/s]
4|1260|Loss: 0.8895736932754517:  19%|█▉        | 1259/6470 [15:35<1:01:27,  1.41it/s]
4|1260|Loss: 0.8895736932754517:  19%|█▉        | 1260/6470 [15:36<1:01:34,  1.41it/s]
4|1261|Loss: 0.7872686386108398:  19%|█▉        | 1260/6470 [15:36<1:01:34,  1.41it/s]
4|1261|Loss: 0.7872686386108398:  19%|█▉        | 1261/6470 [15:36<1:01:10,  1.42it/s]
4|1262|Loss: 0.8536379933357239:  19%|█▉        | 1261/6470 [15:37<1:01:10,  1.42it/s]
4|1262|Loss: 0.8536379933357239:  20%|█▉        | 1262/6470 [15:37<1:02:55,  1.38it/s]
4|1263|Loss: 0.7796528339385986:  20%|█▉        | 1262/6470 [15:38<1:02:55,  1.38it/s]
4|1263|Loss: 0.7796528339385986:  20%|█▉        | 1263/6470 [15:38<1:04:01,  1.36it/s]
4|1264|Loss: 0.7096273899078369:  20%|█▉        | 1263/6470 [15:38<1:04:01,  1.36it/s]
4|1264|Loss: 0.7096273899078369:  20%|█▉        | 1264/6470 [15:39<1:03:18,  1.37it/s]
4|1265|Loss: 0.7928411364555359:  20%|█▉        | 1264/6470 [15:39<1:03:18,  1.37it/s]
4|1265|Loss: 0.7928411364555359:  20%|█▉        | 1265/6470 [15:40<1:06:20,  1.31it/s]
4|1266|Loss: 0.7677082419395447:  20%|█▉        | 1265/6470 [15:40<1:06:20,  1.31it/s]
4|1266|Loss: 0.7677082419395447:  20%|█▉        | 1266/6470 [15:40<1:01:01,  1.42it/s]
4|1267|Loss: 0.9576069712638855:  20%|█▉        | 1266/6470 [15:40<1:01:01,  1.42it/s]
4|1267|Loss: 0.9576069712638855:  20%|█▉        | 1267/6470 [15:41<1:00:26,  1.43it/s]
4|1268|Loss: 0.8912214636802673:  20%|█▉        | 1267/6470 [15:41<1:00:26,  1.43it/s]
4|1268|Loss: 0.8912214636802673:  20%|█▉        | 1268/6470 [15:42<1:05:30,  1.32it/s]
4|1269|Loss: 0.8523988127708435:  20%|█▉        | 1268/6470 [15:42<1:05:30,  1.32it/s]
4|1269|Loss: 0.8523988127708435:  20%|█▉        | 1269/6470 [15:42<58:53,  1.47it/s]  
4|1270|Loss: 0.6320819854736328:  20%|█▉        | 1269/6470 [15:43<58:53,  1.47it/s]
4|1270|Loss: 0.6320819854736328:  20%|█▉        | 1270/6470 [15:43<58:47,  1.47it/s]
4|1271|Loss: 0.7745955586433411:  20%|█▉        | 1270/6470 [15:43<58:47,  1.47it/s]
4|1271|Loss: 0.7745955586433411:  20%|█▉        | 1271/6470 [15:44<1:04:55,  1.33it/s]
4|1272|Loss: 0.7970468401908875:  20%|█▉        | 1271/6470 [15:44<1:04:55,  1.33it/s]
4|1272|Loss: 0.7970468401908875:  20%|█▉        | 1272/6470 [15:44<1:04:51,  1.34it/s]
4|1273|Loss: 0.7957823276519775:  20%|█▉        | 1272/6470 [15:45<1:04:51,  1.34it/s]
4|1273|Loss: 0.7957823276519775:  20%|█▉        | 1273/6470 [15:45<1:05:21,  1.33it/s]
4|1274|Loss: 0.7871866822242737:  20%|█▉        | 1273/6470 [15:46<1:05:21,  1.33it/s]
4|1274|Loss: 0.7871866822242737:  20%|█▉        | 1274/6470 [15:46<1:09:33,  1.24it/s]
4|1275|Loss: 0.8414697051048279:  20%|█▉        | 1274/6470 [15:47<1:09:33,  1.24it/s]
4|1275|Loss: 0.8414697051048279:  20%|█▉        | 1275/6470 [15:47<1:06:02,  1.31it/s]
4|1276|Loss: 0.9091108441352844:  20%|█▉        | 1275/6470 [15:47<1:06:02,  1.31it/s]
4|1276|Loss: 0.9091108441352844:  20%|█▉        | 1276/6470 [15:48<1:03:17,  1.37it/s]
4|1277|Loss: 0.8255301713943481:  20%|█▉        | 1276/6470 [15:48<1:03:17,  1.37it/s]
4|1277|Loss: 0.8255301713943481:  20%|█▉        | 1277/6470 [15:48<1:03:38,  1.36it/s]
4|1278|Loss: 0.7164769172668457:  20%|█▉        | 1277/6470 [15:49<1:03:38,  1.36it/s]
4|1278|Loss: 0.7164769172668457:  20%|█▉        | 1278/6470 [15:49<56:16,  1.54it/s]  
4|1279|Loss: 0.8571999073028564:  20%|█▉        | 1278/6470 [15:49<56:16,  1.54it/s]
4|1279|Loss: 0.8571999073028564:  20%|█▉        | 1279/6470 [15:49<55:27,  1.56it/s]
4|1280|Loss: 0.958812415599823:  20%|█▉        | 1279/6470 [15:50<55:27,  1.56it/s] 
4|1280|Loss: 0.958812415599823:  20%|█▉        | 1280/6470 [15:50<56:33,  1.53it/s]
4|1281|Loss: 0.9134283065795898:  20%|█▉        | 1280/6470 [15:50<56:33,  1.53it/s]
4|1281|Loss: 0.9134283065795898:  20%|█▉        | 1281/6470 [15:51<54:14,  1.59it/s]
4|1282|Loss: 0.9045072793960571:  20%|█▉        | 1281/6470 [15:51<54:14,  1.59it/s]
4|1282|Loss: 0.9045072793960571:  20%|█▉        | 1282/6470 [15:51<56:17,  1.54it/s]
4|1283|Loss: 0.7271885275840759:  20%|█▉        | 1282/6470 [15:52<56:17,  1.54it/s]
4|1283|Loss: 0.7271885275840759:  20%|█▉        | 1283/6470 [15:52<57:00,  1.52it/s]
4|1284|Loss: 0.7862247228622437:  20%|█▉        | 1283/6470 [15:52<57:00,  1.52it/s]
4|1284|Loss: 0.7862247228622437:  20%|█▉        | 1284/6470 [15:53<54:26,  1.59it/s]
4|1285|Loss: 0.7355088591575623:  20%|█▉        | 1284/6470 [15:53<54:26,  1.59it/s]
4|1285|Loss: 0.7355088591575623:  20%|█▉        | 1285/6470 [15:53<50:54,  1.70it/s]
4|1286|Loss: 0.9740778803825378:  20%|█▉        | 1285/6470 [15:53<50:54,  1.70it/s]
4|1286|Loss: 0.9740778803825378:  20%|█▉        | 1286/6470 [15:54<54:52,  1.57it/s]
4|1287|Loss: 0.9589335918426514:  20%|█▉        | 1286/6470 [15:54<54:52,  1.57it/s]
4|1287|Loss: 0.9589335918426514:  20%|█▉        | 1287/6470 [15:55<1:00:46,  1.42it/s]
4|1288|Loss: 0.7855803966522217:  20%|█▉        | 1287/6470 [15:55<1:00:46,  1.42it/s]
4|1288|Loss: 0.7855803966522217:  20%|█▉        | 1288/6470 [15:55<1:01:45,  1.40it/s]
4|1289|Loss: 0.7738462090492249:  20%|█▉        | 1288/6470 [15:56<1:01:45,  1.40it/s]
4|1289|Loss: 0.7738462090492249:  20%|█▉        | 1289/6470 [15:56<1:04:58,  1.33it/s]
4|1290|Loss: 0.6325975656509399:  20%|█▉        | 1289/6470 [15:57<1:04:58,  1.33it/s]
4|1290|Loss: 0.6325975656509399:  20%|█▉        | 1290/6470 [15:57<1:02:40,  1.38it/s]
4|1291|Loss: 0.9777576923370361:  20%|█▉        | 1290/6470 [15:57<1:02:40,  1.38it/s]
4|1291|Loss: 0.9777576923370361:  20%|█▉        | 1291/6470 [15:57<56:41,  1.52it/s]  
4|1292|Loss: 0.9940213561058044:  20%|█▉        | 1291/6470 [15:58<56:41,  1.52it/s]
4|1292|Loss: 0.9940213561058044:  20%|█▉        | 1292/6470 [15:58<57:31,  1.50it/s]
4|1293|Loss: 0.7593404650688171:  20%|█▉        | 1292/6470 [15:59<57:31,  1.50it/s]
4|1293|Loss: 0.7593404650688171:  20%|█▉        | 1293/6470 [15:59<1:02:17,  1.39it/s]
4|1294|Loss: 0.878631591796875:  20%|█▉        | 1293/6470 [15:59<1:02:17,  1.39it/s] 
4|1294|Loss: 0.878631591796875:  20%|██        | 1294/6470 [16:00<1:04:50,  1.33it/s]
4|1295|Loss: 0.8959680199623108:  20%|██        | 1294/6470 [16:00<1:04:50,  1.33it/s]
4|1295|Loss: 0.8959680199623108:  20%|██        | 1295/6470 [16:00<1:03:31,  1.36it/s]
4|1296|Loss: 0.854403555393219:  20%|██        | 1295/6470 [16:01<1:03:31,  1.36it/s] 
4|1296|Loss: 0.854403555393219:  20%|██        | 1296/6470 [16:01<58:29,  1.47it/s]  
4|1297|Loss: 0.8424050807952881:  20%|██        | 1296/6470 [16:01<58:29,  1.47it/s]
4|1297|Loss: 0.8424050807952881:  20%|██        | 1297/6470 [16:02<59:28,  1.45it/s]
4|1298|Loss: 0.6709161996841431:  20%|██        | 1297/6470 [16:02<59:28,  1.45it/s]
4|1298|Loss: 0.6709161996841431:  20%|██        | 1298/6470 [16:02<1:02:09,  1.39it/s]
4|1299|Loss: 0.9006649255752563:  20%|██        | 1298/6470 [16:03<1:02:09,  1.39it/s]
4|1299|Loss: 0.9006649255752563:  20%|██        | 1299/6470 [16:03<1:04:13,  1.34it/s]
4|1300|Loss: 0.8466836214065552:  20%|██        | 1299/6470 [16:04<1:04:13,  1.34it/s]
4|1300|Loss: 0.8466836214065552:  20%|██        | 1300/6470 [16:04<1:03:02,  1.37it/s]
4|1301|Loss: 0.6583859920501709:  20%|██        | 1300/6470 [16:04<1:03:02,  1.37it/s]
4|1301|Loss: 0.6583859920501709:  20%|██        | 1301/6470 [16:04<55:18,  1.56it/s]  
4|1302|Loss: 0.6684784293174744:  20%|██        | 1301/6470 [16:05<55:18,  1.56it/s]
4|1302|Loss: 0.6684784293174744:  20%|██        | 1302/6470 [16:05<50:39,  1.70it/s]
4|1303|Loss: 0.5703203082084656:  20%|██        | 1302/6470 [16:05<50:39,  1.70it/s]
4|1303|Loss: 0.5703203082084656:  20%|██        | 1303/6470 [16:06<53:35,  1.61it/s]
4|1304|Loss: 0.85301274061203:  20%|██        | 1303/6470 [16:06<53:35,  1.61it/s]  
4|1304|Loss: 0.85301274061203:  20%|██        | 1304/6470 [16:06<57:17,  1.50it/s]
4|1305|Loss: 0.9767171144485474:  20%|██        | 1304/6470 [16:07<57:17,  1.50it/s]
4|1305|Loss: 0.9767171144485474:  20%|██        | 1305/6470 [16:07<1:01:47,  1.39it/s]
4|1306|Loss: 0.8144192099571228:  20%|██        | 1305/6470 [16:08<1:01:47,  1.39it/s]
4|1306|Loss: 0.8144192099571228:  20%|██        | 1306/6470 [16:08<1:07:30,  1.28it/s]
4|1307|Loss: 0.8989256024360657:  20%|██        | 1306/6470 [16:09<1:07:30,  1.28it/s]
4|1307|Loss: 0.8989256024360657:  20%|██        | 1307/6470 [16:09<1:08:28,  1.26it/s]
4|1308|Loss: 0.9140294194221497:  20%|██        | 1307/6470 [16:09<1:08:28,  1.26it/s]
4|1308|Loss: 0.9140294194221497:  20%|██        | 1308/6470 [16:10<1:11:22,  1.21it/s]
4|1309|Loss: 0.7471016645431519:  20%|██        | 1308/6470 [16:10<1:11:22,  1.21it/s]
4|1309|Loss: 0.7471016645431519:  20%|██        | 1309/6470 [16:11<1:11:39,  1.20it/s]
4|1310|Loss: 0.828397274017334:  20%|██        | 1309/6470 [16:11<1:11:39,  1.20it/s] 
4|1310|Loss: 0.828397274017334:  20%|██        | 1310/6470 [16:12<1:13:12,  1.17it/s]
4|1311|Loss: 0.8519372344017029:  20%|██        | 1310/6470 [16:12<1:13:12,  1.17it/s]
4|1311|Loss: 0.8519372344017029:  20%|██        | 1311/6470 [16:13<1:14:58,  1.15it/s]
4|1312|Loss: 0.8236116766929626:  20%|██        | 1311/6470 [16:13<1:14:58,  1.15it/s]
4|1312|Loss: 0.8236116766929626:  20%|██        | 1312/6470 [16:13<1:13:45,  1.17it/s]
4|1313|Loss: 0.7397411465644836:  20%|██        | 1312/6470 [16:14<1:13:45,  1.17it/s]
4|1313|Loss: 0.7397411465644836:  20%|██        | 1313/6470 [16:14<1:10:03,  1.23it/s]
4|1314|Loss: 0.9571182727813721:  20%|██        | 1313/6470 [16:14<1:10:03,  1.23it/s]
4|1314|Loss: 0.9571182727813721:  20%|██        | 1314/6470 [16:15<1:07:09,  1.28it/s]
4|1315|Loss: 0.8351476192474365:  20%|██        | 1314/6470 [16:15<1:07:09,  1.28it/s]
4|1315|Loss: 0.8351476192474365:  20%|██        | 1315/6470 [16:16<1:07:00,  1.28it/s]
4|1316|Loss: 0.8261898756027222:  20%|██        | 1315/6470 [16:16<1:07:00,  1.28it/s]
4|1316|Loss: 0.8261898756027222:  20%|██        | 1316/6470 [16:16<1:07:07,  1.28it/s]
4|1317|Loss: 0.7103065252304077:  20%|██        | 1316/6470 [16:17<1:07:07,  1.28it/s]
4|1317|Loss: 0.7103065252304077:  20%|██        | 1317/6470 [16:17<1:09:28,  1.24it/s]
4|1318|Loss: 0.9047812819480896:  20%|██        | 1317/6470 [16:18<1:09:28,  1.24it/s]
4|1318|Loss: 0.9047812819480896:  20%|██        | 1318/6470 [16:18<1:07:17,  1.28it/s]
4|1319|Loss: 0.8310370445251465:  20%|██        | 1318/6470 [16:18<1:07:17,  1.28it/s]
4|1319|Loss: 0.8310370445251465:  20%|██        | 1319/6470 [16:19<1:07:10,  1.28it/s]
4|1320|Loss: 0.7888336181640625:  20%|██        | 1319/6470 [16:19<1:07:10,  1.28it/s]
4|1320|Loss: 0.7888336181640625:  20%|██        | 1320/6470 [16:20<1:10:24,  1.22it/s]
4|1321|Loss: 0.733268678188324:  20%|██        | 1320/6470 [16:20<1:10:24,  1.22it/s] 
4|1321|Loss: 0.733268678188324:  20%|██        | 1321/6470 [16:20<1:05:26,  1.31it/s]
4|1322|Loss: 0.8638075590133667:  20%|██        | 1321/6470 [16:21<1:05:26,  1.31it/s]
4|1322|Loss: 0.8638075590133667:  20%|██        | 1322/6470 [16:21<1:01:59,  1.38it/s]
4|1323|Loss: 0.7807558178901672:  20%|██        | 1322/6470 [16:21<1:01:59,  1.38it/s]
4|1323|Loss: 0.7807558178901672:  20%|██        | 1323/6470 [16:22<1:02:33,  1.37it/s]
4|1324|Loss: 0.977439820766449:  20%|██        | 1323/6470 [16:22<1:02:33,  1.37it/s] 
4|1324|Loss: 0.977439820766449:  20%|██        | 1324/6470 [16:22<1:05:39,  1.31it/s]
4|1325|Loss: 0.8913553357124329:  20%|██        | 1324/6470 [16:23<1:05:39,  1.31it/s]
4|1325|Loss: 0.8913553357124329:  20%|██        | 1325/6470 [16:23<1:03:26,  1.35it/s]
4|1326|Loss: 0.7325114607810974:  20%|██        | 1325/6470 [16:24<1:03:26,  1.35it/s]
4|1326|Loss: 0.7325114607810974:  20%|██        | 1326/6470 [16:24<1:04:02,  1.34it/s]
4|1327|Loss: 0.7137928009033203:  20%|██        | 1326/6470 [16:24<1:04:02,  1.34it/s]
4|1327|Loss: 0.7137928009033203:  21%|██        | 1327/6470 [16:25<1:02:43,  1.37it/s]
4|1328|Loss: 0.7078324556350708:  21%|██        | 1327/6470 [16:25<1:02:43,  1.37it/s]
4|1328|Loss: 0.7078324556350708:  21%|██        | 1328/6470 [16:26<1:08:04,  1.26it/s]
4|1329|Loss: 0.9670385122299194:  21%|██        | 1328/6470 [16:26<1:08:04,  1.26it/s]
4|1329|Loss: 0.9670385122299194:  21%|██        | 1329/6470 [16:26<1:10:36,  1.21it/s]
4|1330|Loss: 0.8574774861335754:  21%|██        | 1329/6470 [16:27<1:10:36,  1.21it/s]
4|1330|Loss: 0.8574774861335754:  21%|██        | 1330/6470 [16:27<1:08:28,  1.25it/s]
4|1331|Loss: 0.6567527055740356:  21%|██        | 1330/6470 [16:28<1:08:28,  1.25it/s]
4|1331|Loss: 0.6567527055740356:  21%|██        | 1331/6470 [16:28<1:10:57,  1.21it/s]
4|1332|Loss: 0.8446195721626282:  21%|██        | 1331/6470 [16:29<1:10:57,  1.21it/s]
4|1332|Loss: 0.8446195721626282:  21%|██        | 1332/6470 [16:29<1:09:08,  1.24it/s]
4|1333|Loss: 0.9177056550979614:  21%|██        | 1332/6470 [16:29<1:09:08,  1.24it/s]
4|1333|Loss: 0.9177056550979614:  21%|██        | 1333/6470 [16:30<1:11:10,  1.20it/s]
4|1334|Loss: 0.7211582660675049:  21%|██        | 1333/6470 [16:30<1:11:10,  1.20it/s]
4|1334|Loss: 0.7211582660675049:  21%|██        | 1334/6470 [16:30<1:09:10,  1.24it/s]
4|1335|Loss: 0.8257545232772827:  21%|██        | 1334/6470 [16:31<1:09:10,  1.24it/s]
4|1335|Loss: 0.8257545232772827:  21%|██        | 1335/6470 [16:31<1:05:34,  1.31it/s]
4|1336|Loss: 0.8406002521514893:  21%|██        | 1335/6470 [16:32<1:05:34,  1.31it/s]
4|1336|Loss: 0.8406002521514893:  21%|██        | 1336/6470 [16:32<1:05:51,  1.30it/s]
4|1337|Loss: 0.7675965428352356:  21%|██        | 1336/6470 [16:32<1:05:51,  1.30it/s]
4|1337|Loss: 0.7675965428352356:  21%|██        | 1337/6470 [16:33<1:08:01,  1.26it/s]
4|1338|Loss: 0.8025742769241333:  21%|██        | 1337/6470 [16:33<1:08:01,  1.26it/s]
4|1338|Loss: 0.8025742769241333:  21%|██        | 1338/6470 [16:33<1:02:04,  1.38it/s]
4|1339|Loss: 0.8565097451210022:  21%|██        | 1338/6470 [16:34<1:02:04,  1.38it/s]
4|1339|Loss: 0.8565097451210022:  21%|██        | 1339/6470 [16:34<1:03:35,  1.34it/s]
4|1340|Loss: 1.0333523750305176:  21%|██        | 1339/6470 [16:35<1:03:35,  1.34it/s]
4|1340|Loss: 1.0333523750305176:  21%|██        | 1340/6470 [16:35<1:05:28,  1.31it/s]
4|1341|Loss: 0.8346982598304749:  21%|██        | 1340/6470 [16:35<1:05:28,  1.31it/s]
4|1341|Loss: 0.8346982598304749:  21%|██        | 1341/6470 [16:36<1:08:06,  1.26it/s]
4|1342|Loss: 0.8854570984840393:  21%|██        | 1341/6470 [16:36<1:08:06,  1.26it/s]
4|1342|Loss: 0.8854570984840393:  21%|██        | 1342/6470 [16:37<1:10:39,  1.21it/s]
4|1343|Loss: 0.7215994000434875:  21%|██        | 1342/6470 [16:37<1:10:39,  1.21it/s]
4|1343|Loss: 0.7215994000434875:  21%|██        | 1343/6470 [16:37<1:09:54,  1.22it/s]
4|1344|Loss: 0.8238639235496521:  21%|██        | 1343/6470 [16:38<1:09:54,  1.22it/s]
4|1344|Loss: 0.8238639235496521:  21%|██        | 1344/6470 [16:38<1:09:14,  1.23it/s]
4|1345|Loss: 0.8986921906471252:  21%|██        | 1344/6470 [16:39<1:09:14,  1.23it/s]
4|1345|Loss: 0.8986921906471252:  21%|██        | 1345/6470 [16:39<1:06:27,  1.29it/s]
4|1346|Loss: 0.8262414336204529:  21%|██        | 1345/6470 [16:39<1:06:27,  1.29it/s]
4|1346|Loss: 0.8262414336204529:  21%|██        | 1346/6470 [16:40<1:06:13,  1.29it/s]
4|1347|Loss: 0.9496098160743713:  21%|██        | 1346/6470 [16:40<1:06:13,  1.29it/s]
4|1347|Loss: 0.9496098160743713:  21%|██        | 1347/6470 [16:40<1:04:04,  1.33it/s]
4|1348|Loss: 0.7831484079360962:  21%|██        | 1347/6470 [16:41<1:04:04,  1.33it/s]
4|1348|Loss: 0.7831484079360962:  21%|██        | 1348/6470 [16:41<1:00:58,  1.40it/s]
4|1349|Loss: 0.9954366683959961:  21%|██        | 1348/6470 [16:41<1:00:58,  1.40it/s]
4|1349|Loss: 0.9954366683959961:  21%|██        | 1349/6470 [16:42<1:00:06,  1.42it/s]
4|1350|Loss: 0.9077973365783691:  21%|██        | 1349/6470 [16:42<1:00:06,  1.42it/s]
4|1350|Loss: 0.9077973365783691:  21%|██        | 1350/6470 [16:43<1:01:03,  1.40it/s]
4|1351|Loss: 0.8321816921234131:  21%|██        | 1350/6470 [16:43<1:01:03,  1.40it/s]
4|1351|Loss: 0.8321816921234131:  21%|██        | 1351/6470 [16:43<58:27,  1.46it/s]  
4|1352|Loss: 0.7034986615180969:  21%|██        | 1351/6470 [16:43<58:27,  1.46it/s]
4|1352|Loss: 0.7034986615180969:  21%|██        | 1352/6470 [16:44<56:34,  1.51it/s]
4|1353|Loss: 0.8648783564567566:  21%|██        | 1352/6470 [16:44<56:34,  1.51it/s]
4|1353|Loss: 0.8648783564567566:  21%|██        | 1353/6470 [16:44<58:26,  1.46it/s]
4|1354|Loss: 0.7745988368988037:  21%|██        | 1353/6470 [16:45<58:26,  1.46it/s]
4|1354|Loss: 0.7745988368988037:  21%|██        | 1354/6470 [16:45<56:22,  1.51it/s]
4|1355|Loss: 1.0203076601028442:  21%|██        | 1354/6470 [16:45<56:22,  1.51it/s]
4|1355|Loss: 1.0203076601028442:  21%|██        | 1355/6470 [16:46<59:46,  1.43it/s]
4|1356|Loss: 0.8000577092170715:  21%|██        | 1355/6470 [16:46<59:46,  1.43it/s]
4|1356|Loss: 0.8000577092170715:  21%|██        | 1356/6470 [16:47<1:02:34,  1.36it/s]
4|1357|Loss: 0.9369646906852722:  21%|██        | 1356/6470 [16:47<1:02:34,  1.36it/s]
4|1357|Loss: 0.9369646906852722:  21%|██        | 1357/6470 [16:48<1:05:40,  1.30it/s]
4|1358|Loss: 0.8459784984588623:  21%|██        | 1357/6470 [16:48<1:05:40,  1.30it/s]
4|1358|Loss: 0.8459784984588623:  21%|██        | 1358/6470 [16:48<1:01:48,  1.38it/s]
4|1359|Loss: 0.8397412896156311:  21%|██        | 1358/6470 [16:49<1:01:48,  1.38it/s]
4|1359|Loss: 0.8397412896156311:  21%|██        | 1359/6470 [16:49<1:02:33,  1.36it/s]
4|1360|Loss: 0.8907959461212158:  21%|██        | 1359/6470 [16:49<1:02:33,  1.36it/s]
4|1360|Loss: 0.8907959461212158:  21%|██        | 1360/6470 [16:50<1:04:21,  1.32it/s]
4|1361|Loss: 0.8162540197372437:  21%|██        | 1360/6470 [16:50<1:04:21,  1.32it/s]
4|1361|Loss: 0.8162540197372437:  21%|██        | 1361/6470 [16:51<1:05:38,  1.30it/s]
4|1362|Loss: 0.854564368724823:  21%|██        | 1361/6470 [16:51<1:05:38,  1.30it/s] 
4|1362|Loss: 0.854564368724823:  21%|██        | 1362/6470 [16:51<1:06:46,  1.28it/s]
4|1363|Loss: 0.6759475469589233:  21%|██        | 1362/6470 [16:52<1:06:46,  1.28it/s]
4|1363|Loss: 0.6759475469589233:  21%|██        | 1363/6470 [16:52<1:09:19,  1.23it/s]
4|1364|Loss: 0.7296050786972046:  21%|██        | 1363/6470 [16:53<1:09:19,  1.23it/s]
4|1364|Loss: 0.7296050786972046:  21%|██        | 1364/6470 [16:53<1:08:17,  1.25it/s]
4|1365|Loss: 0.8162671327590942:  21%|██        | 1364/6470 [16:53<1:08:17,  1.25it/s]
4|1365|Loss: 0.8162671327590942:  21%|██        | 1365/6470 [16:54<1:09:35,  1.22it/s]
4|1366|Loss: 0.844512403011322:  21%|██        | 1365/6470 [16:54<1:09:35,  1.22it/s] 
4|1366|Loss: 0.844512403011322:  21%|██        | 1366/6470 [16:55<1:07:26,  1.26it/s]
4|1367|Loss: 0.8191385269165039:  21%|██        | 1366/6470 [16:55<1:07:26,  1.26it/s]
4|1367|Loss: 0.8191385269165039:  21%|██        | 1367/6470 [16:55<1:06:59,  1.27it/s]
4|1368|Loss: 0.7712127566337585:  21%|██        | 1367/6470 [16:56<1:06:59,  1.27it/s]
4|1368|Loss: 0.7712127566337585:  21%|██        | 1368/6470 [16:56<1:11:13,  1.19it/s]
4|1369|Loss: 0.8240827322006226:  21%|██        | 1368/6470 [16:57<1:11:13,  1.19it/s]
4|1369|Loss: 0.8240827322006226:  21%|██        | 1369/6470 [16:57<1:06:07,  1.29it/s]
4|1370|Loss: 0.661794126033783:  21%|██        | 1369/6470 [16:57<1:06:07,  1.29it/s] 
4|1370|Loss: 0.661794126033783:  21%|██        | 1370/6470 [16:58<1:06:54,  1.27it/s]
4|1371|Loss: 0.6800804734230042:  21%|██        | 1370/6470 [16:58<1:06:54,  1.27it/s]
4|1371|Loss: 0.6800804734230042:  21%|██        | 1371/6470 [16:59<1:07:38,  1.26it/s]
4|1372|Loss: 0.789461076259613:  21%|██        | 1371/6470 [16:59<1:07:38,  1.26it/s] 
4|1372|Loss: 0.789461076259613:  21%|██        | 1372/6470 [16:59<1:02:42,  1.36it/s]
4|1373|Loss: 0.7667345404624939:  21%|██        | 1372/6470 [16:59<1:02:42,  1.36it/s]
4|1373|Loss: 0.7667345404624939:  21%|██        | 1373/6470 [17:00<54:15,  1.57it/s]  
4|1374|Loss: 0.8532776832580566:  21%|██        | 1373/6470 [17:00<54:15,  1.57it/s]
4|1374|Loss: 0.8532776832580566:  21%|██        | 1374/6470 [17:00<53:48,  1.58it/s]
4|1375|Loss: 0.8200355172157288:  21%|██        | 1374/6470 [17:01<53:48,  1.58it/s]
4|1375|Loss: 0.8200355172157288:  21%|██▏       | 1375/6470 [17:01<57:38,  1.47it/s]
4|1376|Loss: 0.7274832725524902:  21%|██▏       | 1375/6470 [17:01<57:38,  1.47it/s]
4|1376|Loss: 0.7274832725524902:  21%|██▏       | 1376/6470 [17:02<58:17,  1.46it/s]
4|1377|Loss: 0.7629091739654541:  21%|██▏       | 1376/6470 [17:02<58:17,  1.46it/s]
4|1377|Loss: 0.7629091739654541:  21%|██▏       | 1377/6470 [17:03<1:02:35,  1.36it/s]
4|1378|Loss: 0.9352480173110962:  21%|██▏       | 1377/6470 [17:03<1:02:35,  1.36it/s]
4|1378|Loss: 0.9352480173110962:  21%|██▏       | 1378/6470 [17:03<1:03:41,  1.33it/s]
4|1379|Loss: 0.8990603685379028:  21%|██▏       | 1378/6470 [17:04<1:03:41,  1.33it/s]
4|1379|Loss: 0.8990603685379028:  21%|██▏       | 1379/6470 [17:04<1:05:59,  1.29it/s]
4|1380|Loss: 0.8833738565444946:  21%|██▏       | 1379/6470 [17:05<1:05:59,  1.29it/s]
4|1380|Loss: 0.8833738565444946:  21%|██▏       | 1380/6470 [17:05<1:02:49,  1.35it/s]
4|1381|Loss: 0.6980940699577332:  21%|██▏       | 1380/6470 [17:05<1:02:49,  1.35it/s]
4|1381|Loss: 0.6980940699577332:  21%|██▏       | 1381/6470 [17:05<57:49,  1.47it/s]  
4|1382|Loss: 0.8296399116516113:  21%|██▏       | 1381/6470 [17:06<57:49,  1.47it/s]
4|1382|Loss: 0.8296399116516113:  21%|██▏       | 1382/6470 [17:06<59:45,  1.42it/s]
4|1383|Loss: 0.8356814980506897:  21%|██▏       | 1382/6470 [17:07<59:45,  1.42it/s]
4|1383|Loss: 0.8356814980506897:  21%|██▏       | 1383/6470 [17:07<58:13,  1.46it/s]
4|1384|Loss: 0.7865592837333679:  21%|██▏       | 1383/6470 [17:07<58:13,  1.46it/s]
4|1384|Loss: 0.7865592837333679:  21%|██▏       | 1384/6470 [17:08<1:06:01,  1.28it/s]
4|1385|Loss: 0.9262673258781433:  21%|██▏       | 1384/6470 [17:08<1:06:01,  1.28it/s]
4|1385|Loss: 0.9262673258781433:  21%|██▏       | 1385/6470 [17:09<1:05:38,  1.29it/s]
4|1386|Loss: 0.8797269463539124:  21%|██▏       | 1385/6470 [17:09<1:05:38,  1.29it/s]
4|1386|Loss: 0.8797269463539124:  21%|██▏       | 1386/6470 [17:09<1:08:31,  1.24it/s]
4|1387|Loss: 0.6604740023612976:  21%|██▏       | 1386/6470 [17:10<1:08:31,  1.24it/s]
4|1387|Loss: 0.6604740023612976:  21%|██▏       | 1387/6470 [17:10<1:11:41,  1.18it/s]
4|1388|Loss: 0.8617011904716492:  21%|██▏       | 1387/6470 [17:11<1:11:41,  1.18it/s]
4|1388|Loss: 0.8617011904716492:  21%|██▏       | 1388/6470 [17:11<1:10:15,  1.21it/s]
4|1389|Loss: 0.7361803650856018:  21%|██▏       | 1388/6470 [17:12<1:10:15,  1.21it/s]
4|1389|Loss: 0.7361803650856018:  21%|██▏       | 1389/6470 [17:12<1:10:37,  1.20it/s]
4|1390|Loss: 0.85166996717453:  21%|██▏       | 1389/6470 [17:12<1:10:37,  1.20it/s]  
4|1390|Loss: 0.85166996717453:  21%|██▏       | 1390/6470 [17:13<1:09:55,  1.21it/s]
4|1391|Loss: 0.6326907277107239:  21%|██▏       | 1390/6470 [17:13<1:09:55,  1.21it/s]
4|1391|Loss: 0.6326907277107239:  21%|██▏       | 1391/6470 [17:13<1:04:48,  1.31it/s]
4|1392|Loss: 0.842315137386322:  21%|██▏       | 1391/6470 [17:14<1:04:48,  1.31it/s] 
4|1392|Loss: 0.842315137386322:  22%|██▏       | 1392/6470 [17:14<1:03:40,  1.33it/s]
4|1393|Loss: 0.8131726980209351:  22%|██▏       | 1392/6470 [17:15<1:03:40,  1.33it/s]
4|1393|Loss: 0.8131726980209351:  22%|██▏       | 1393/6470 [17:15<1:04:27,  1.31it/s]
4|1394|Loss: 0.8091186285018921:  22%|██▏       | 1393/6470 [17:15<1:04:27,  1.31it/s]
4|1394|Loss: 0.8091186285018921:  22%|██▏       | 1394/6470 [17:16<1:01:58,  1.37it/s]
4|1395|Loss: 0.9617631435394287:  22%|██▏       | 1394/6470 [17:16<1:01:58,  1.37it/s]
4|1395|Loss: 0.9617631435394287:  22%|██▏       | 1395/6470 [17:16<1:02:36,  1.35it/s]
4|1396|Loss: 1.04702889919281:  22%|██▏       | 1395/6470 [17:17<1:02:36,  1.35it/s]  
4|1396|Loss: 1.04702889919281:  22%|██▏       | 1396/6470 [17:17<1:02:14,  1.36it/s]
4|1397|Loss: 0.8504824042320251:  22%|██▏       | 1396/6470 [17:17<1:02:14,  1.36it/s]
4|1397|Loss: 0.8504824042320251:  22%|██▏       | 1397/6470 [17:18<1:03:06,  1.34it/s]
4|1398|Loss: 0.8280578255653381:  22%|██▏       | 1397/6470 [17:18<1:03:06,  1.34it/s]
4|1398|Loss: 0.8280578255653381:  22%|██▏       | 1398/6470 [17:19<1:04:45,  1.31it/s]
4|1399|Loss: 0.8421422839164734:  22%|██▏       | 1398/6470 [17:19<1:04:45,  1.31it/s]
4|1399|Loss: 0.8421422839164734:  22%|██▏       | 1399/6470 [17:19<1:03:52,  1.32it/s]
4|1400|Loss: 0.9099001884460449:  22%|██▏       | 1399/6470 [17:20<1:03:52,  1.32it/s]
4|1400|Loss: 0.9099001884460449:  22%|██▏       | 1400/6470 [17:20<1:04:40,  1.31it/s]
4|1401|Loss: 0.8721693754196167:  22%|██▏       | 1400/6470 [17:21<1:04:40,  1.31it/s]
4|1401|Loss: 0.8721693754196167:  22%|██▏       | 1401/6470 [17:21<1:09:01,  1.22it/s]
4|1402|Loss: 0.7906495928764343:  22%|██▏       | 1401/6470 [17:22<1:09:01,  1.22it/s]
4|1402|Loss: 0.7906495928764343:  22%|██▏       | 1402/6470 [17:22<1:03:06,  1.34it/s]
4|1403|Loss: 1.0375980138778687:  22%|██▏       | 1402/6470 [17:22<1:03:06,  1.34it/s]
4|1403|Loss: 1.0375980138778687:  22%|██▏       | 1403/6470 [17:22<57:45,  1.46it/s]  
4|1404|Loss: 0.7620214819908142:  22%|██▏       | 1403/6470 [17:23<57:45,  1.46it/s]
4|1404|Loss: 0.7620214819908142:  22%|██▏       | 1404/6470 [17:23<58:38,  1.44it/s]
4|1405|Loss: 0.847200334072113:  22%|██▏       | 1404/6470 [17:23<58:38,  1.44it/s] 
4|1405|Loss: 0.847200334072113:  22%|██▏       | 1405/6470 [17:24<1:03:31,  1.33it/s]
4|1406|Loss: 0.8121577501296997:  22%|██▏       | 1405/6470 [17:24<1:03:31,  1.33it/s]
4|1406|Loss: 0.8121577501296997:  22%|██▏       | 1406/6470 [17:25<1:03:39,  1.33it/s]
4|1407|Loss: 0.8027742505073547:  22%|██▏       | 1406/6470 [17:25<1:03:39,  1.33it/s]
4|1407|Loss: 0.8027742505073547:  22%|██▏       | 1407/6470 [17:25<1:04:18,  1.31it/s]
4|1408|Loss: 0.8157339096069336:  22%|██▏       | 1407/6470 [17:26<1:04:18,  1.31it/s]
4|1408|Loss: 0.8157339096069336:  22%|██▏       | 1408/6470 [17:26<1:04:34,  1.31it/s]
4|1409|Loss: 0.8635572791099548:  22%|██▏       | 1408/6470 [17:27<1:04:34,  1.31it/s]
4|1409|Loss: 0.8635572791099548:  22%|██▏       | 1409/6470 [17:27<1:03:19,  1.33it/s]
4|1410|Loss: 0.8758322596549988:  22%|██▏       | 1409/6470 [17:27<1:03:19,  1.33it/s]
4|1410|Loss: 0.8758322596549988:  22%|██▏       | 1410/6470 [17:28<1:02:03,  1.36it/s]
4|1411|Loss: 0.9993685483932495:  22%|██▏       | 1410/6470 [17:28<1:02:03,  1.36it/s]
4|1411|Loss: 0.9993685483932495:  22%|██▏       | 1411/6470 [17:28<1:03:14,  1.33it/s]
4|1412|Loss: 0.8010733723640442:  22%|██▏       | 1411/6470 [17:29<1:03:14,  1.33it/s]
4|1412|Loss: 0.8010733723640442:  22%|██▏       | 1412/6470 [17:29<1:02:13,  1.35it/s]
4|1413|Loss: 0.741158127784729:  22%|██▏       | 1412/6470 [17:29<1:02:13,  1.35it/s] 
4|1413|Loss: 0.741158127784729:  22%|██▏       | 1413/6470 [17:30<1:03:28,  1.33it/s]
4|1414|Loss: 0.7662928104400635:  22%|██▏       | 1413/6470 [17:30<1:03:28,  1.33it/s]
4|1414|Loss: 0.7662928104400635:  22%|██▏       | 1414/6470 [17:31<1:01:12,  1.38it/s]
4|1415|Loss: 0.7364689111709595:  22%|██▏       | 1414/6470 [17:31<1:01:12,  1.38it/s]
4|1415|Loss: 0.7364689111709595:  22%|██▏       | 1415/6470 [17:31<59:59,  1.40it/s]  
4|1416|Loss: 0.8462948203086853:  22%|██▏       | 1415/6470 [17:32<59:59,  1.40it/s]
4|1416|Loss: 0.8462948203086853:  22%|██▏       | 1416/6470 [17:32<58:53,  1.43it/s]
4|1417|Loss: 0.8318487405776978:  22%|██▏       | 1416/6470 [17:32<58:53,  1.43it/s]
4|1417|Loss: 0.8318487405776978:  22%|██▏       | 1417/6470 [17:33<58:15,  1.45it/s]
4|1418|Loss: 0.9447497129440308:  22%|██▏       | 1417/6470 [17:33<58:15,  1.45it/s]
4|1418|Loss: 0.9447497129440308:  22%|██▏       | 1418/6470 [17:33<1:00:20,  1.40it/s]
4|1419|Loss: 0.8085336685180664:  22%|██▏       | 1418/6470 [17:34<1:00:20,  1.40it/s]
4|1419|Loss: 0.8085336685180664:  22%|██▏       | 1419/6470 [17:34<1:02:46,  1.34it/s]
4|1420|Loss: 0.6239413619041443:  22%|██▏       | 1419/6470 [17:35<1:02:46,  1.34it/s]
4|1420|Loss: 0.6239413619041443:  22%|██▏       | 1420/6470 [17:35<1:04:05,  1.31it/s]
4|1421|Loss: 0.8240483403205872:  22%|██▏       | 1420/6470 [17:35<1:04:05,  1.31it/s]
4|1421|Loss: 0.8240483403205872:  22%|██▏       | 1421/6470 [17:36<1:05:23,  1.29it/s]
4|1422|Loss: 0.8279584050178528:  22%|██▏       | 1421/6470 [17:36<1:05:23,  1.29it/s]
4|1422|Loss: 0.8279584050178528:  22%|██▏       | 1422/6470 [17:36<1:02:01,  1.36it/s]
4|1423|Loss: 0.8970350027084351:  22%|██▏       | 1422/6470 [17:37<1:02:01,  1.36it/s]
4|1423|Loss: 0.8970350027084351:  22%|██▏       | 1423/6470 [17:37<59:27,  1.41it/s]  
4|1424|Loss: 0.8070740103721619:  22%|██▏       | 1423/6470 [17:37<59:27,  1.41it/s]
4|1424|Loss: 0.8070740103721619:  22%|██▏       | 1424/6470 [17:38<58:31,  1.44it/s]
4|1425|Loss: 0.7006825804710388:  22%|██▏       | 1424/6470 [17:38<58:31,  1.44it/s]
4|1425|Loss: 0.7006825804710388:  22%|██▏       | 1425/6470 [17:38<54:31,  1.54it/s]
4|1426|Loss: 0.8197901844978333:  22%|██▏       | 1425/6470 [17:39<54:31,  1.54it/s]
4|1426|Loss: 0.8197901844978333:  22%|██▏       | 1426/6470 [17:39<51:57,  1.62it/s]
4|1427|Loss: 0.8738398551940918:  22%|██▏       | 1426/6470 [17:39<51:57,  1.62it/s]
4|1427|Loss: 0.8738398551940918:  22%|██▏       | 1427/6470 [17:40<57:34,  1.46it/s]
4|1428|Loss: 0.8640472292900085:  22%|██▏       | 1427/6470 [17:40<57:34,  1.46it/s]
4|1428|Loss: 0.8640472292900085:  22%|██▏       | 1428/6470 [17:41<1:03:07,  1.33it/s]
4|1429|Loss: 0.8034029603004456:  22%|██▏       | 1428/6470 [17:41<1:03:07,  1.33it/s]
4|1429|Loss: 0.8034029603004456:  22%|██▏       | 1429/6470 [17:41<1:01:54,  1.36it/s]
4|1430|Loss: 0.916537344455719:  22%|██▏       | 1429/6470 [17:42<1:01:54,  1.36it/s] 
4|1430|Loss: 0.916537344455719:  22%|██▏       | 1430/6470 [17:42<56:48,  1.48it/s]  
4|1431|Loss: 0.8318015336990356:  22%|██▏       | 1430/6470 [17:42<56:48,  1.48it/s]
4|1431|Loss: 0.8318015336990356:  22%|██▏       | 1431/6470 [17:42<55:08,  1.52it/s]
4|1432|Loss: 0.5297960638999939:  22%|██▏       | 1431/6470 [17:43<55:08,  1.52it/s]
4|1432|Loss: 0.5297960638999939:  22%|██▏       | 1432/6470 [17:43<52:51,  1.59it/s]
4|1433|Loss: 0.9317647218704224:  22%|██▏       | 1432/6470 [17:43<52:51,  1.59it/s]
4|1433|Loss: 0.9317647218704224:  22%|██▏       | 1433/6470 [17:44<57:03,  1.47it/s]
4|1434|Loss: 0.8363319635391235:  22%|██▏       | 1433/6470 [17:44<57:03,  1.47it/s]
4|1434|Loss: 0.8363319635391235:  22%|██▏       | 1434/6470 [17:45<1:02:40,  1.34it/s]
4|1435|Loss: 0.8732887506484985:  22%|██▏       | 1434/6470 [17:45<1:02:40,  1.34it/s]
4|1435|Loss: 0.8732887506484985:  22%|██▏       | 1435/6470 [17:45<1:02:12,  1.35it/s]
4|1436|Loss: 0.7011840343475342:  22%|██▏       | 1435/6470 [17:46<1:02:12,  1.35it/s]
4|1436|Loss: 0.7011840343475342:  22%|██▏       | 1436/6470 [17:46<59:38,  1.41it/s]  
4|1437|Loss: 0.8928691744804382:  22%|██▏       | 1436/6470 [17:46<59:38,  1.41it/s]
4|1437|Loss: 0.8928691744804382:  22%|██▏       | 1437/6470 [17:47<1:00:27,  1.39it/s]
4|1438|Loss: 0.6705120205879211:  22%|██▏       | 1437/6470 [17:47<1:00:27,  1.39it/s]
4|1438|Loss: 0.6705120205879211:  22%|██▏       | 1438/6470 [17:48<1:05:29,  1.28it/s]
4|1439|Loss: 0.8504469394683838:  22%|██▏       | 1438/6470 [17:48<1:05:29,  1.28it/s]
4|1439|Loss: 0.8504469394683838:  22%|██▏       | 1439/6470 [17:48<1:03:50,  1.31it/s]
4|1440|Loss: 0.7772132754325867:  22%|██▏       | 1439/6470 [17:49<1:03:50,  1.31it/s]
4|1440|Loss: 0.7772132754325867:  22%|██▏       | 1440/6470 [17:49<55:22,  1.51it/s]  
4|1441|Loss: 0.8278718590736389:  22%|██▏       | 1440/6470 [17:49<55:22,  1.51it/s]
4|1441|Loss: 0.8278718590736389:  22%|██▏       | 1441/6470 [17:50<58:34,  1.43it/s]
4|1442|Loss: 1.0057066679000854:  22%|██▏       | 1441/6470 [17:50<58:34,  1.43it/s]
4|1442|Loss: 1.0057066679000854:  22%|██▏       | 1442/6470 [17:50<58:36,  1.43it/s]
4|1443|Loss: 0.9669004082679749:  22%|██▏       | 1442/6470 [17:51<58:36,  1.43it/s]
4|1443|Loss: 0.9669004082679749:  22%|██▏       | 1443/6470 [17:51<1:00:40,  1.38it/s]
4|1444|Loss: 0.8039323091506958:  22%|██▏       | 1443/6470 [17:52<1:00:40,  1.38it/s]
4|1444|Loss: 0.8039323091506958:  22%|██▏       | 1444/6470 [17:52<1:02:32,  1.34it/s]
4|1445|Loss: 0.8781991600990295:  22%|██▏       | 1444/6470 [17:52<1:02:32,  1.34it/s]
4|1445|Loss: 0.8781991600990295:  22%|██▏       | 1445/6470 [17:53<1:08:01,  1.23it/s]
4|1446|Loss: 0.892690122127533:  22%|██▏       | 1445/6470 [17:53<1:08:01,  1.23it/s] 
4|1446|Loss: 0.892690122127533:  22%|██▏       | 1446/6470 [17:54<1:07:11,  1.25it/s]
4|1447|Loss: 0.7538881301879883:  22%|██▏       | 1446/6470 [17:54<1:07:11,  1.25it/s]
4|1447|Loss: 0.7538881301879883:  22%|██▏       | 1447/6470 [17:54<1:04:05,  1.31it/s]
4|1448|Loss: 0.7403733134269714:  22%|██▏       | 1447/6470 [17:55<1:04:05,  1.31it/s]
4|1448|Loss: 0.7403733134269714:  22%|██▏       | 1448/6470 [17:55<1:01:33,  1.36it/s]
4|1449|Loss: 0.8566568493843079:  22%|██▏       | 1448/6470 [17:55<1:01:33,  1.36it/s]
4|1449|Loss: 0.8566568493843079:  22%|██▏       | 1449/6470 [17:56<1:02:48,  1.33it/s]
4|1450|Loss: 0.6445786356925964:  22%|██▏       | 1449/6470 [17:56<1:02:48,  1.33it/s]
4|1450|Loss: 0.6445786356925964:  22%|██▏       | 1450/6470 [17:56<1:01:40,  1.36it/s]
4|1451|Loss: 0.805019736289978:  22%|██▏       | 1450/6470 [17:57<1:01:40,  1.36it/s] 
4|1451|Loss: 0.805019736289978:  22%|██▏       | 1451/6470 [17:57<1:02:24,  1.34it/s]
4|1452|Loss: 0.7768165469169617:  22%|██▏       | 1451/6470 [17:58<1:02:24,  1.34it/s]
4|1452|Loss: 0.7768165469169617:  22%|██▏       | 1452/6470 [17:58<1:00:01,  1.39it/s]
4|1453|Loss: 0.7586533427238464:  22%|██▏       | 1452/6470 [17:58<1:00:01,  1.39it/s]
4|1453|Loss: 0.7586533427238464:  22%|██▏       | 1453/6470 [17:59<59:03,  1.42it/s]  
4|1454|Loss: 0.8740220665931702:  22%|██▏       | 1453/6470 [17:59<59:03,  1.42it/s]
4|1454|Loss: 0.8740220665931702:  22%|██▏       | 1454/6470 [17:59<1:02:07,  1.35it/s]
4|1455|Loss: 0.7895129919052124:  22%|██▏       | 1454/6470 [18:00<1:02:07,  1.35it/s]
4|1455|Loss: 0.7895129919052124:  22%|██▏       | 1455/6470 [18:00<1:01:29,  1.36it/s]
4|1456|Loss: 0.9972501993179321:  22%|██▏       | 1455/6470 [18:01<1:01:29,  1.36it/s]
4|1456|Loss: 0.9972501993179321:  23%|██▎       | 1456/6470 [18:01<1:02:42,  1.33it/s]
4|1457|Loss: 0.7132071256637573:  23%|██▎       | 1456/6470 [18:01<1:02:42,  1.33it/s]
4|1457|Loss: 0.7132071256637573:  23%|██▎       | 1457/6470 [18:02<1:00:33,  1.38it/s]
4|1458|Loss: 0.8455787897109985:  23%|██▎       | 1457/6470 [18:02<1:00:33,  1.38it/s]
4|1458|Loss: 0.8455787897109985:  23%|██▎       | 1458/6470 [18:02<56:34,  1.48it/s]  
4|1459|Loss: 0.8946162462234497:  23%|██▎       | 1458/6470 [18:03<56:34,  1.48it/s]
4|1459|Loss: 0.8946162462234497:  23%|██▎       | 1459/6470 [18:03<57:45,  1.45it/s]
4|1460|Loss: 0.8638758659362793:  23%|██▎       | 1459/6470 [18:03<57:45,  1.45it/s]
4|1460|Loss: 0.8638758659362793:  23%|██▎       | 1460/6470 [18:04<58:22,  1.43it/s]
4|1461|Loss: 0.8674123287200928:  23%|██▎       | 1460/6470 [18:04<58:22,  1.43it/s]
4|1461|Loss: 0.8674123287200928:  23%|██▎       | 1461/6470 [18:04<58:57,  1.42it/s]
4|1462|Loss: 0.8012685179710388:  23%|██▎       | 1461/6470 [18:05<58:57,  1.42it/s]
4|1462|Loss: 0.8012685179710388:  23%|██▎       | 1462/6470 [18:05<59:58,  1.39it/s]
4|1463|Loss: 0.8090663552284241:  23%|██▎       | 1462/6470 [18:05<59:58,  1.39it/s]
4|1463|Loss: 0.8090663552284241:  23%|██▎       | 1463/6470 [18:06<56:42,  1.47it/s]
4|1464|Loss: 0.7546768188476562:  23%|██▎       | 1463/6470 [18:06<56:42,  1.47it/s]
4|1464|Loss: 0.7546768188476562:  23%|██▎       | 1464/6470 [18:06<52:35,  1.59it/s]
4|1465|Loss: 0.756533682346344:  23%|██▎       | 1464/6470 [18:06<52:35,  1.59it/s] 
4|1465|Loss: 0.756533682346344:  23%|██▎       | 1465/6470 [18:07<53:12,  1.57it/s]
4|1466|Loss: 0.8584883213043213:  23%|██▎       | 1465/6470 [18:07<53:12,  1.57it/s]
4|1466|Loss: 0.8584883213043213:  23%|██▎       | 1466/6470 [18:08<56:04,  1.49it/s]
4|1467|Loss: 0.8609293699264526:  23%|██▎       | 1466/6470 [18:08<56:04,  1.49it/s]
4|1467|Loss: 0.8609293699264526:  23%|██▎       | 1467/6470 [18:08<1:02:09,  1.34it/s]
4|1468|Loss: 0.7149357199668884:  23%|██▎       | 1467/6470 [18:09<1:02:09,  1.34it/s]
4|1468|Loss: 0.7149357199668884:  23%|██▎       | 1468/6470 [18:09<1:02:01,  1.34it/s]
4|1469|Loss: 0.7962511777877808:  23%|██▎       | 1468/6470 [18:10<1:02:01,  1.34it/s]
4|1469|Loss: 0.7962511777877808:  23%|██▎       | 1469/6470 [18:10<1:03:06,  1.32it/s]
4|1470|Loss: 0.8819721341133118:  23%|██▎       | 1469/6470 [18:10<1:03:06,  1.32it/s]
4|1470|Loss: 0.8819721341133118:  23%|██▎       | 1470/6470 [18:11<1:02:19,  1.34it/s]
4|1471|Loss: 0.8950507044792175:  23%|██▎       | 1470/6470 [18:11<1:02:19,  1.34it/s]
4|1471|Loss: 0.8950507044792175:  23%|██▎       | 1471/6470 [18:12<1:05:48,  1.27it/s]
4|1472|Loss: 0.8817692995071411:  23%|██▎       | 1471/6470 [18:12<1:05:48,  1.27it/s]
4|1472|Loss: 0.8817692995071411:  23%|██▎       | 1472/6470 [18:13<1:09:40,  1.20it/s]
4|1473|Loss: 0.8876848220825195:  23%|██▎       | 1472/6470 [18:13<1:09:40,  1.20it/s]
4|1473|Loss: 0.8876848220825195:  23%|██▎       | 1473/6470 [18:13<1:11:16,  1.17it/s]
4|1474|Loss: 0.8766805529594421:  23%|██▎       | 1473/6470 [18:14<1:11:16,  1.17it/s]
4|1474|Loss: 0.8766805529594421:  23%|██▎       | 1474/6470 [18:14<1:08:18,  1.22it/s]
4|1475|Loss: 1.0563526153564453:  23%|██▎       | 1474/6470 [18:15<1:08:18,  1.22it/s]
4|1475|Loss: 1.0563526153564453:  23%|██▎       | 1475/6470 [18:15<1:07:30,  1.23it/s]
4|1476|Loss: 0.958717942237854:  23%|██▎       | 1475/6470 [18:15<1:07:30,  1.23it/s] 
4|1476|Loss: 0.958717942237854:  23%|██▎       | 1476/6470 [18:16<1:05:11,  1.28it/s]
4|1477|Loss: 0.8161993622779846:  23%|██▎       | 1476/6470 [18:16<1:05:11,  1.28it/s]
4|1477|Loss: 0.8161993622779846:  23%|██▎       | 1477/6470 [18:16<58:39,  1.42it/s]  
4|1478|Loss: 0.8975861668586731:  23%|██▎       | 1477/6470 [18:17<58:39,  1.42it/s]
4|1478|Loss: 0.8975861668586731:  23%|██▎       | 1478/6470 [18:17<58:30,  1.42it/s]
4|1479|Loss: 0.8896711468696594:  23%|██▎       | 1478/6470 [18:17<58:30,  1.42it/s]
4|1479|Loss: 0.8896711468696594:  23%|██▎       | 1479/6470 [18:18<1:00:00,  1.39it/s]
4|1480|Loss: 0.8131563067436218:  23%|██▎       | 1479/6470 [18:18<1:00:00,  1.39it/s]
4|1480|Loss: 0.8131563067436218:  23%|██▎       | 1480/6470 [18:18<56:32,  1.47it/s]  
4|1481|Loss: 0.9422401785850525:  23%|██▎       | 1480/6470 [18:19<56:32,  1.47it/s]
4|1481|Loss: 0.9422401785850525:  23%|██▎       | 1481/6470 [18:19<57:05,  1.46it/s]
4|1482|Loss: 0.7882680892944336:  23%|██▎       | 1481/6470 [18:19<57:05,  1.46it/s]
4|1482|Loss: 0.7882680892944336:  23%|██▎       | 1482/6470 [18:20<53:16,  1.56it/s]
4|1483|Loss: 0.8134840726852417:  23%|██▎       | 1482/6470 [18:20<53:16,  1.56it/s]
4|1483|Loss: 0.8134840726852417:  23%|██▎       | 1483/6470 [18:20<54:21,  1.53it/s]
4|1484|Loss: 0.7215040326118469:  23%|██▎       | 1483/6470 [18:21<54:21,  1.53it/s]
4|1484|Loss: 0.7215040326118469:  23%|██▎       | 1484/6470 [18:21<57:36,  1.44it/s]
4|1485|Loss: 0.7357494235038757:  23%|██▎       | 1484/6470 [18:21<57:36,  1.44it/s]
4|1485|Loss: 0.7357494235038757:  23%|██▎       | 1485/6470 [18:22<57:37,  1.44it/s]
4|1486|Loss: 0.773647129535675:  23%|██▎       | 1485/6470 [18:22<57:37,  1.44it/s] 
4|1486|Loss: 0.773647129535675:  23%|██▎       | 1486/6470 [18:22<56:57,  1.46it/s]
4|1487|Loss: 0.9557655453681946:  23%|██▎       | 1486/6470 [18:23<56:57,  1.46it/s]
4|1487|Loss: 0.9557655453681946:  23%|██▎       | 1487/6470 [18:23<1:00:55,  1.36it/s]
4|1488|Loss: 0.7718651294708252:  23%|██▎       | 1487/6470 [18:24<1:00:55,  1.36it/s]
4|1488|Loss: 0.7718651294708252:  23%|██▎       | 1488/6470 [18:24<57:58,  1.43it/s]  
4|1489|Loss: 0.8574004173278809:  23%|██▎       | 1488/6470 [18:24<57:58,  1.43it/s]
4|1489|Loss: 0.8574004173278809:  23%|██▎       | 1489/6470 [18:25<1:00:43,  1.37it/s]
4|1490|Loss: 0.8829843401908875:  23%|██▎       | 1489/6470 [18:25<1:00:43,  1.37it/s]
4|1490|Loss: 0.8829843401908875:  23%|██▎       | 1490/6470 [18:25<1:00:13,  1.38it/s]
4|1491|Loss: 0.7949249148368835:  23%|██▎       | 1490/6470 [18:26<1:00:13,  1.38it/s]
4|1491|Loss: 0.7949249148368835:  23%|██▎       | 1491/6470 [18:26<1:00:35,  1.37it/s]
4|1492|Loss: 0.7395908236503601:  23%|██▎       | 1491/6470 [18:27<1:00:35,  1.37it/s]
4|1492|Loss: 0.7395908236503601:  23%|██▎       | 1492/6470 [18:27<1:05:18,  1.27it/s]
4|1493|Loss: 0.7276595830917358:  23%|██▎       | 1492/6470 [18:27<1:05:18,  1.27it/s]
4|1493|Loss: 0.7276595830917358:  23%|██▎       | 1493/6470 [18:28<1:05:33,  1.27it/s]
4|1494|Loss: 0.9196435809135437:  23%|██▎       | 1493/6470 [18:28<1:05:33,  1.27it/s]
4|1494|Loss: 0.9196435809135437:  23%|██▎       | 1494/6470 [18:29<1:06:49,  1.24it/s]
4|1495|Loss: 0.9265115261077881:  23%|██▎       | 1494/6470 [18:29<1:06:49,  1.24it/s]
4|1495|Loss: 0.9265115261077881:  23%|██▎       | 1495/6470 [18:29<1:06:51,  1.24it/s]
4|1496|Loss: 0.6676235795021057:  23%|██▎       | 1495/6470 [18:30<1:06:51,  1.24it/s]
4|1496|Loss: 0.6676235795021057:  23%|██▎       | 1496/6470 [18:31<1:33:09,  1.12s/it]
4|1497|Loss: 0.8769233822822571:  23%|██▎       | 1496/6470 [18:32<1:33:09,  1.12s/it]
4|1497|Loss: 0.8769233822822571:  23%|██▎       | 1497/6470 [18:32<1:26:18,  1.04s/it]
4|1498|Loss: 0.852565348148346:  23%|██▎       | 1497/6470 [18:33<1:26:18,  1.04s/it] 
4|1498|Loss: 0.852565348148346:  23%|██▎       | 1498/6470 [18:33<1:18:24,  1.06it/s]
4|1499|Loss: 0.8951641917228699:  23%|██▎       | 1498/6470 [18:33<1:18:24,  1.06it/s]
4|1499|Loss: 0.8951641917228699:  23%|██▎       | 1499/6470 [18:33<1:08:07,  1.22it/s]
4|1500|Loss: 0.8526086211204529:  23%|██▎       | 1499/6470 [18:34<1:08:07,  1.22it/s]
4|1500|Loss: 0.8526086211204529:  23%|██▎       | 1500/6470 [18:34<1:07:19,  1.23it/s]
4|1501|Loss: 0.983328640460968:  23%|██▎       | 1500/6470 [18:35<1:07:19,  1.23it/s] 
4|1501|Loss: 0.983328640460968:  23%|██▎       | 1501/6470 [18:35<1:03:55,  1.30it/s]
4|1502|Loss: 0.81294846534729:  23%|██▎       | 1501/6470 [18:35<1:03:55,  1.30it/s] 
4|1502|Loss: 0.81294846534729:  23%|██▎       | 1502/6470 [18:36<1:01:47,  1.34it/s]
4|1503|Loss: 0.8285088539123535:  23%|██▎       | 1502/6470 [18:36<1:01:47,  1.34it/s]
4|1503|Loss: 0.8285088539123535:  23%|██▎       | 1503/6470 [18:36<1:05:31,  1.26it/s]
4|1504|Loss: 0.9550809264183044:  23%|██▎       | 1503/6470 [18:37<1:05:31,  1.26it/s]
4|1504|Loss: 0.9550809264183044:  23%|██▎       | 1504/6470 [18:37<1:02:40,  1.32it/s]
4|1505|Loss: 0.5861697196960449:  23%|██▎       | 1504/6470 [18:37<1:02:40,  1.32it/s]
4|1505|Loss: 0.5861697196960449:  23%|██▎       | 1505/6470 [18:38<58:57,  1.40it/s]  
4|1506|Loss: 0.8401213884353638:  23%|██▎       | 1505/6470 [18:38<58:57,  1.40it/s]
4|1506|Loss: 0.8401213884353638:  23%|██▎       | 1506/6470 [18:38<59:36,  1.39it/s]
4|1507|Loss: 0.8482810854911804:  23%|██▎       | 1506/6470 [18:39<59:36,  1.39it/s]
4|1507|Loss: 0.8482810854911804:  23%|██▎       | 1507/6470 [18:39<58:44,  1.41it/s]
4|1508|Loss: 0.8690400719642639:  23%|██▎       | 1507/6470 [18:40<58:44,  1.41it/s]
4|1508|Loss: 0.8690400719642639:  23%|██▎       | 1508/6470 [18:40<57:31,  1.44it/s]
4|1509|Loss: 0.9496192336082458:  23%|██▎       | 1508/6470 [18:40<57:31,  1.44it/s]
4|1509|Loss: 0.9496192336082458:  23%|██▎       | 1509/6470 [18:41<1:00:33,  1.37it/s]
4|1510|Loss: 0.9544191360473633:  23%|██▎       | 1509/6470 [18:41<1:00:33,  1.37it/s]
4|1510|Loss: 0.9544191360473633:  23%|██▎       | 1510/6470 [18:41<1:02:29,  1.32it/s]
4|1511|Loss: 0.7701153755187988:  23%|██▎       | 1510/6470 [18:42<1:02:29,  1.32it/s]
4|1511|Loss: 0.7701153755187988:  23%|██▎       | 1511/6470 [18:42<1:01:37,  1.34it/s]
4|1512|Loss: 0.8998638987541199:  23%|██▎       | 1511/6470 [18:43<1:01:37,  1.34it/s]
4|1512|Loss: 0.8998638987541199:  23%|██▎       | 1512/6470 [18:43<58:17,  1.42it/s]  
4|1513|Loss: 0.9258418679237366:  23%|██▎       | 1512/6470 [18:43<58:17,  1.42it/s]
4|1513|Loss: 0.9258418679237366:  23%|██▎       | 1513/6470 [18:44<1:02:10,  1.33it/s]
4|1514|Loss: 0.7504119277000427:  23%|██▎       | 1513/6470 [18:44<1:02:10,  1.33it/s]
4|1514|Loss: 0.7504119277000427:  23%|██▎       | 1514/6470 [18:44<1:03:20,  1.30it/s]
4|1515|Loss: 0.843921422958374:  23%|██▎       | 1514/6470 [18:45<1:03:20,  1.30it/s] 
4|1515|Loss: 0.843921422958374:  23%|██▎       | 1515/6470 [18:45<1:06:53,  1.23it/s]
4|1516|Loss: 0.9430661201477051:  23%|██▎       | 1515/6470 [18:46<1:06:53,  1.23it/s]
4|1516|Loss: 0.9430661201477051:  23%|██▎       | 1516/6470 [18:46<1:04:54,  1.27it/s]
4|1517|Loss: 0.8827274441719055:  23%|██▎       | 1516/6470 [18:46<1:04:54,  1.27it/s]
4|1517|Loss: 0.8827274441719055:  23%|██▎       | 1517/6470 [18:47<1:02:33,  1.32it/s]
4|1518|Loss: 0.7076531052589417:  23%|██▎       | 1517/6470 [18:47<1:02:33,  1.32it/s]
4|1518|Loss: 0.7076531052589417:  23%|██▎       | 1518/6470 [18:48<1:05:39,  1.26it/s]
4|1519|Loss: 0.8256353139877319:  23%|██▎       | 1518/6470 [18:48<1:05:39,  1.26it/s]
4|1519|Loss: 0.8256353139877319:  23%|██▎       | 1519/6470 [18:48<1:03:53,  1.29it/s]
4|1520|Loss: 0.8787373304367065:  23%|██▎       | 1519/6470 [18:49<1:03:53,  1.29it/s]
4|1520|Loss: 0.8787373304367065:  23%|██▎       | 1520/6470 [18:49<1:06:46,  1.24it/s]
4|1521|Loss: 0.8733163475990295:  23%|██▎       | 1520/6470 [18:50<1:06:46,  1.24it/s]
4|1521|Loss: 0.8733163475990295:  24%|██▎       | 1521/6470 [18:50<1:07:19,  1.23it/s]
4|1522|Loss: 0.7701850533485413:  24%|██▎       | 1521/6470 [18:50<1:07:19,  1.23it/s]
4|1522|Loss: 0.7701850533485413:  24%|██▎       | 1522/6470 [18:51<1:03:31,  1.30it/s]
4|1523|Loss: 0.8581629395484924:  24%|██▎       | 1522/6470 [18:51<1:03:31,  1.30it/s]
4|1523|Loss: 0.8581629395484924:  24%|██▎       | 1523/6470 [18:52<1:02:33,  1.32it/s]
4|1524|Loss: 0.848851203918457:  24%|██▎       | 1523/6470 [18:52<1:02:33,  1.32it/s] 
4|1524|Loss: 0.848851203918457:  24%|██▎       | 1524/6470 [18:52<1:03:23,  1.30it/s]
4|1525|Loss: 0.7472813725471497:  24%|██▎       | 1524/6470 [18:53<1:03:23,  1.30it/s]
4|1525|Loss: 0.7472813725471497:  24%|██▎       | 1525/6470 [18:53<1:05:28,  1.26it/s]
4|1526|Loss: 0.9169584512710571:  24%|██▎       | 1525/6470 [18:54<1:05:28,  1.26it/s]
4|1526|Loss: 0.9169584512710571:  24%|██▎       | 1526/6470 [18:54<1:03:50,  1.29it/s]
4|1527|Loss: 0.8674688935279846:  24%|██▎       | 1526/6470 [18:54<1:03:50,  1.29it/s]
4|1527|Loss: 0.8674688935279846:  24%|██▎       | 1527/6470 [18:55<1:01:11,  1.35it/s]
4|1528|Loss: 0.824423611164093:  24%|██▎       | 1527/6470 [18:55<1:01:11,  1.35it/s] 
4|1528|Loss: 0.824423611164093:  24%|██▎       | 1528/6470 [18:55<58:07,  1.42it/s]  
4|1529|Loss: 0.8799722194671631:  24%|██▎       | 1528/6470 [18:56<58:07,  1.42it/s]
4|1529|Loss: 0.8799722194671631:  24%|██▎       | 1529/6470 [18:56<59:38,  1.38it/s]
4|1530|Loss: 0.5562072992324829:  24%|██▎       | 1529/6470 [18:56<59:38,  1.38it/s]
4|1530|Loss: 0.5562072992324829:  24%|██▎       | 1530/6470 [18:57<1:04:42,  1.27it/s]
4|1531|Loss: 0.6118602156639099:  24%|██▎       | 1530/6470 [18:57<1:04:42,  1.27it/s]
4|1531|Loss: 0.6118602156639099:  24%|██▎       | 1531/6470 [18:58<1:03:29,  1.30it/s]
4|1532|Loss: 0.8324236273765564:  24%|██▎       | 1531/6470 [18:58<1:03:29,  1.30it/s]
4|1532|Loss: 0.8324236273765564:  24%|██▎       | 1532/6470 [18:58<1:05:54,  1.25it/s]
4|1533|Loss: 0.9501847624778748:  24%|██▎       | 1532/6470 [18:59<1:05:54,  1.25it/s]
4|1533|Loss: 0.9501847624778748:  24%|██▎       | 1533/6470 [18:59<1:09:08,  1.19it/s]
4|1534|Loss: 0.9350531697273254:  24%|██▎       | 1533/6470 [19:00<1:09:08,  1.19it/s]
4|1534|Loss: 0.9350531697273254:  24%|██▎       | 1534/6470 [19:00<1:10:09,  1.17it/s]
4|1535|Loss: 0.7661478519439697:  24%|██▎       | 1534/6470 [19:01<1:10:09,  1.17it/s]
4|1535|Loss: 0.7661478519439697:  24%|██▎       | 1535/6470 [19:01<1:01:00,  1.35it/s]
4|1536|Loss: 0.8547035455703735:  24%|██▎       | 1535/6470 [19:01<1:01:00,  1.35it/s]
4|1536|Loss: 0.8547035455703735:  24%|██▎       | 1536/6470 [19:01<54:35,  1.51it/s]  
4|1537|Loss: 0.834739625453949:  24%|██▎       | 1536/6470 [19:02<54:35,  1.51it/s] 
4|1537|Loss: 0.834739625453949:  24%|██▍       | 1537/6470 [19:02<54:10,  1.52it/s]
4|1538|Loss: 0.7432547211647034:  24%|██▍       | 1537/6470 [19:02<54:10,  1.52it/s]
4|1538|Loss: 0.7432547211647034:  24%|██▍       | 1538/6470 [19:03<55:19,  1.49it/s]
4|1539|Loss: 0.6577603816986084:  24%|██▍       | 1538/6470 [19:03<55:19,  1.49it/s]
4|1539|Loss: 0.6577603816986084:  24%|██▍       | 1539/6470 [19:04<1:03:06,  1.30it/s]
4|1540|Loss: 0.8952391147613525:  24%|██▍       | 1539/6470 [19:04<1:03:06,  1.30it/s]
4|1540|Loss: 0.8952391147613525:  24%|██▍       | 1540/6470 [19:04<1:02:28,  1.32it/s]
4|1541|Loss: 0.6945169568061829:  24%|██▍       | 1540/6470 [19:05<1:02:28,  1.32it/s]
4|1541|Loss: 0.6945169568061829:  24%|██▍       | 1541/6470 [19:05<1:02:53,  1.31it/s]
4|1542|Loss: 0.7564511895179749:  24%|██▍       | 1541/6470 [19:05<1:02:53,  1.31it/s]
4|1542|Loss: 0.7564511895179749:  24%|██▍       | 1542/6470 [19:06<58:39,  1.40it/s]  
4|1543|Loss: 0.739456832408905:  24%|██▍       | 1542/6470 [19:06<58:39,  1.40it/s] 
4|1543|Loss: 0.739456832408905:  24%|██▍       | 1543/6470 [19:07<1:02:29,  1.31it/s]
4|1544|Loss: 0.7407404184341431:  24%|██▍       | 1543/6470 [19:07<1:02:29,  1.31it/s]
4|1544|Loss: 0.7407404184341431:  24%|██▍       | 1544/6470 [19:07<1:01:37,  1.33it/s]
4|1545|Loss: 0.9116736054420471:  24%|██▍       | 1544/6470 [19:08<1:01:37,  1.33it/s]
4|1545|Loss: 0.9116736054420471:  24%|██▍       | 1545/6470 [19:08<1:01:30,  1.33it/s]
4|1546|Loss: 0.7080993056297302:  24%|██▍       | 1545/6470 [19:08<1:01:30,  1.33it/s]
4|1546|Loss: 0.7080993056297302:  24%|██▍       | 1546/6470 [19:09<1:00:26,  1.36it/s]
4|1547|Loss: 0.7237390279769897:  24%|██▍       | 1546/6470 [19:09<1:00:26,  1.36it/s]
4|1547|Loss: 0.7237390279769897:  24%|██▍       | 1547/6470 [19:10<1:01:48,  1.33it/s]
4|1548|Loss: 0.843951404094696:  24%|██▍       | 1547/6470 [19:10<1:01:48,  1.33it/s] 
4|1548|Loss: 0.843951404094696:  24%|██▍       | 1548/6470 [19:10<1:00:36,  1.35it/s]
4|1549|Loss: 0.7822685837745667:  24%|██▍       | 1548/6470 [19:11<1:00:36,  1.35it/s]
4|1549|Loss: 0.7822685837745667:  24%|██▍       | 1549/6470 [19:11<55:50,  1.47it/s]  
4|1550|Loss: 0.8603363633155823:  24%|██▍       | 1549/6470 [19:11<55:50,  1.47it/s]
4|1550|Loss: 0.8603363633155823:  24%|██▍       | 1550/6470 [19:12<57:06,  1.44it/s]
4|1551|Loss: 0.7737563848495483:  24%|██▍       | 1550/6470 [19:12<57:06,  1.44it/s]
4|1551|Loss: 0.7737563848495483:  24%|██▍       | 1551/6470 [19:13<1:03:48,  1.28it/s]
4|1552|Loss: 0.9465213418006897:  24%|██▍       | 1551/6470 [19:13<1:03:48,  1.28it/s]
4|1552|Loss: 0.9465213418006897:  24%|██▍       | 1552/6470 [19:13<59:59,  1.37it/s]  
4|1553|Loss: 0.8203846216201782:  24%|██▍       | 1552/6470 [19:13<59:59,  1.37it/s]
4|1553|Loss: 0.8203846216201782:  24%|██▍       | 1553/6470 [19:14<58:12,  1.41it/s]
4|1554|Loss: 0.8033318519592285:  24%|██▍       | 1553/6470 [19:14<58:12,  1.41it/s]
4|1554|Loss: 0.8033318519592285:  24%|██▍       | 1554/6470 [19:14<54:15,  1.51it/s]
4|1555|Loss: 0.8314281105995178:  24%|██▍       | 1554/6470 [19:15<54:15,  1.51it/s]
4|1555|Loss: 0.8314281105995178:  24%|██▍       | 1555/6470 [19:15<52:34,  1.56it/s]
4|1556|Loss: 0.8209104537963867:  24%|██▍       | 1555/6470 [19:15<52:34,  1.56it/s]
4|1556|Loss: 0.8209104537963867:  24%|██▍       | 1556/6470 [19:16<51:21,  1.59it/s]
4|1557|Loss: 0.8196926116943359:  24%|██▍       | 1556/6470 [19:16<51:21,  1.59it/s]
4|1557|Loss: 0.8196926116943359:  24%|██▍       | 1557/6470 [19:16<54:37,  1.50it/s]
4|1558|Loss: 0.8155307769775391:  24%|██▍       | 1557/6470 [19:17<54:37,  1.50it/s]
4|1558|Loss: 0.8155307769775391:  24%|██▍       | 1558/6470 [19:17<59:00,  1.39it/s]
4|1559|Loss: 0.8083020448684692:  24%|██▍       | 1558/6470 [19:18<59:00,  1.39it/s]
4|1559|Loss: 0.8083020448684692:  24%|██▍       | 1559/6470 [19:18<1:01:28,  1.33it/s]
4|1560|Loss: 0.7706409096717834:  24%|██▍       | 1559/6470 [19:18<1:01:28,  1.33it/s]
4|1560|Loss: 0.7706409096717834:  24%|██▍       | 1560/6470 [19:19<1:06:40,  1.23it/s]
4|1561|Loss: 0.8466009497642517:  24%|██▍       | 1560/6470 [19:19<1:06:40,  1.23it/s]
4|1561|Loss: 0.8466009497642517:  24%|██▍       | 1561/6470 [19:20<1:08:38,  1.19it/s]
4|1562|Loss: 0.8019170761108398:  24%|██▍       | 1561/6470 [19:20<1:08:38,  1.19it/s]
4|1562|Loss: 0.8019170761108398:  24%|██▍       | 1562/6470 [19:21<1:07:54,  1.20it/s]
4|1563|Loss: 0.8179092407226562:  24%|██▍       | 1562/6470 [19:21<1:07:54,  1.20it/s]
4|1563|Loss: 0.8179092407226562:  24%|██▍       | 1563/6470 [19:22<1:11:19,  1.15it/s]
4|1564|Loss: 0.7662711143493652:  24%|██▍       | 1563/6470 [19:22<1:11:19,  1.15it/s]
4|1564|Loss: 0.7662711143493652:  24%|██▍       | 1564/6470 [19:22<1:12:13,  1.13it/s]
4|1565|Loss: 0.8324472308158875:  24%|██▍       | 1564/6470 [19:23<1:12:13,  1.13it/s]
4|1565|Loss: 0.8324472308158875:  24%|██▍       | 1565/6470 [19:23<1:08:37,  1.19it/s]
4|1566|Loss: 0.6853874921798706:  24%|██▍       | 1565/6470 [19:24<1:08:37,  1.19it/s]
4|1566|Loss: 0.6853874921798706:  24%|██▍       | 1566/6470 [19:24<1:09:38,  1.17it/s]
4|1567|Loss: 0.8242993354797363:  24%|██▍       | 1566/6470 [19:25<1:09:38,  1.17it/s]
4|1567|Loss: 0.8242993354797363:  24%|██▍       | 1567/6470 [19:25<1:09:10,  1.18it/s]
4|1568|Loss: 0.7579809427261353:  24%|██▍       | 1567/6470 [19:25<1:09:10,  1.18it/s]
4|1568|Loss: 0.7579809427261353:  24%|██▍       | 1568/6470 [19:26<1:07:11,  1.22it/s]
4|1569|Loss: 0.6902647018432617:  24%|██▍       | 1568/6470 [19:26<1:07:11,  1.22it/s]
4|1569|Loss: 0.6902647018432617:  24%|██▍       | 1569/6470 [19:26<1:05:39,  1.24it/s]
4|1570|Loss: 0.8392812013626099:  24%|██▍       | 1569/6470 [19:27<1:05:39,  1.24it/s]
4|1570|Loss: 0.8392812013626099:  24%|██▍       | 1570/6470 [19:27<1:02:05,  1.32it/s]
4|1571|Loss: 0.9293627738952637:  24%|██▍       | 1570/6470 [19:27<1:02:05,  1.32it/s]
4|1571|Loss: 0.9293627738952637:  24%|██▍       | 1571/6470 [19:28<57:51,  1.41it/s]  
4|1572|Loss: 0.7124785780906677:  24%|██▍       | 1571/6470 [19:28<57:51,  1.41it/s]
4|1572|Loss: 0.7124785780906677:  24%|██▍       | 1572/6470 [19:28<57:02,  1.43it/s]
4|1573|Loss: 0.9895244240760803:  24%|██▍       | 1572/6470 [19:29<57:02,  1.43it/s]
4|1573|Loss: 0.9895244240760803:  24%|██▍       | 1573/6470 [19:29<54:30,  1.50it/s]
4|1574|Loss: 0.818638801574707:  24%|██▍       | 1573/6470 [19:29<54:30,  1.50it/s] 
4|1574|Loss: 0.818638801574707:  24%|██▍       | 1574/6470 [19:30<54:41,  1.49it/s]
4|1575|Loss: 0.8119557499885559:  24%|██▍       | 1574/6470 [19:30<54:41,  1.49it/s]
4|1575|Loss: 0.8119557499885559:  24%|██▍       | 1575/6470 [19:30<55:37,  1.47it/s]
4|1576|Loss: 0.8643958568572998:  24%|██▍       | 1575/6470 [19:31<55:37,  1.47it/s]
4|1576|Loss: 0.8643958568572998:  24%|██▍       | 1576/6470 [19:31<57:22,  1.42it/s]
4|1577|Loss: 0.7967098355293274:  24%|██▍       | 1576/6470 [19:32<57:22,  1.42it/s]
4|1577|Loss: 0.7967098355293274:  24%|██▍       | 1577/6470 [19:32<58:19,  1.40it/s]
4|1578|Loss: 0.8791390061378479:  24%|██▍       | 1577/6470 [19:32<58:19,  1.40it/s]
4|1578|Loss: 0.8791390061378479:  24%|██▍       | 1578/6470 [19:33<57:25,  1.42it/s]
4|1579|Loss: 0.7817196846008301:  24%|██▍       | 1578/6470 [19:33<57:25,  1.42it/s]
4|1579|Loss: 0.7817196846008301:  24%|██▍       | 1579/6470 [19:33<50:53,  1.60it/s]
4|1580|Loss: 0.991826057434082:  24%|██▍       | 1579/6470 [19:33<50:53,  1.60it/s] 
4|1580|Loss: 0.991826057434082:  24%|██▍       | 1580/6470 [19:34<55:20,  1.47it/s]
4|1581|Loss: 0.8593391180038452:  24%|██▍       | 1580/6470 [19:34<55:20,  1.47it/s]
4|1581|Loss: 0.8593391180038452:  24%|██▍       | 1581/6470 [19:34<54:09,  1.50it/s]
4|1582|Loss: 0.8096416592597961:  24%|██▍       | 1581/6470 [19:35<54:09,  1.50it/s]
4|1582|Loss: 0.8096416592597961:  24%|██▍       | 1582/6470 [19:35<56:20,  1.45it/s]
4|1583|Loss: 0.813816487789154:  24%|██▍       | 1582/6470 [19:36<56:20,  1.45it/s] 
4|1583|Loss: 0.813816487789154:  24%|██▍       | 1583/6470 [19:36<55:48,  1.46it/s]
4|1584|Loss: 0.7605027556419373:  24%|██▍       | 1583/6470 [19:36<55:48,  1.46it/s]
4|1584|Loss: 0.7605027556419373:  24%|██▍       | 1584/6470 [19:36<53:26,  1.52it/s]
4|1585|Loss: 0.8708431124687195:  24%|██▍       | 1584/6470 [19:37<53:26,  1.52it/s]
4|1585|Loss: 0.8708431124687195:  24%|██▍       | 1585/6470 [19:37<57:03,  1.43it/s]
4|1586|Loss: 0.9262117147445679:  24%|██▍       | 1585/6470 [19:38<57:03,  1.43it/s]
4|1586|Loss: 0.9262117147445679:  25%|██▍       | 1586/6470 [19:38<1:00:06,  1.35it/s]
4|1587|Loss: 0.8820820450782776:  25%|██▍       | 1586/6470 [19:39<1:00:06,  1.35it/s]
4|1587|Loss: 0.8820820450782776:  25%|██▍       | 1587/6470 [19:39<1:00:45,  1.34it/s]
4|1588|Loss: 0.9664595127105713:  25%|██▍       | 1587/6470 [19:39<1:00:45,  1.34it/s]
4|1588|Loss: 0.9664595127105713:  25%|██▍       | 1588/6470 [19:40<1:00:44,  1.34it/s]
4|1589|Loss: 0.6359821557998657:  25%|██▍       | 1588/6470 [19:40<1:00:44,  1.34it/s]
4|1589|Loss: 0.6359821557998657:  25%|██▍       | 1589/6470 [19:40<59:49,  1.36it/s]  
4|1590|Loss: 0.5878118276596069:  25%|██▍       | 1589/6470 [19:41<59:49,  1.36it/s]
4|1590|Loss: 0.5878118276596069:  25%|██▍       | 1590/6470 [19:41<1:00:01,  1.36it/s]
4|1591|Loss: 0.814436137676239:  25%|██▍       | 1590/6470 [19:41<1:00:01,  1.36it/s] 
4|1591|Loss: 0.814436137676239:  25%|██▍       | 1591/6470 [19:42<1:00:27,  1.34it/s]
4|1592|Loss: 0.696445107460022:  25%|██▍       | 1591/6470 [19:42<1:00:27,  1.34it/s]
4|1592|Loss: 0.696445107460022:  25%|██▍       | 1592/6470 [19:43<1:00:37,  1.34it/s]
4|1593|Loss: 0.9170188307762146:  25%|██▍       | 1592/6470 [19:43<1:00:37,  1.34it/s]
4|1593|Loss: 0.9170188307762146:  25%|██▍       | 1593/6470 [19:43<57:32,  1.41it/s]  
4|1594|Loss: 1.004106044769287:  25%|██▍       | 1593/6470 [19:44<57:32,  1.41it/s] 
4|1594|Loss: 1.004106044769287:  25%|██▍       | 1594/6470 [19:44<58:41,  1.38it/s]
4|1595|Loss: 0.708480715751648:  25%|██▍       | 1594/6470 [19:44<58:41,  1.38it/s]
4|1595|Loss: 0.708480715751648:  25%|██▍       | 1595/6470 [19:45<1:05:31,  1.24it/s]
4|1596|Loss: 0.9182177186012268:  25%|██▍       | 1595/6470 [19:45<1:05:31,  1.24it/s]
4|1596|Loss: 0.9182177186012268:  25%|██▍       | 1596/6470 [19:46<1:06:28,  1.22it/s]
4|1597|Loss: 0.8008718490600586:  25%|██▍       | 1596/6470 [19:46<1:06:28,  1.22it/s]
4|1597|Loss: 0.8008718490600586:  25%|██▍       | 1597/6470 [19:47<1:05:46,  1.23it/s]
4|1598|Loss: 0.7780573964118958:  25%|██▍       | 1597/6470 [19:47<1:05:46,  1.23it/s]
4|1598|Loss: 0.7780573964118958:  25%|██▍       | 1598/6470 [19:47<1:03:23,  1.28it/s]
4|1599|Loss: 0.7639585733413696:  25%|██▍       | 1598/6470 [19:48<1:03:23,  1.28it/s]
4|1599|Loss: 0.7639585733413696:  25%|██▍       | 1599/6470 [19:48<1:00:40,  1.34it/s]
4|1600|Loss: 0.8265788555145264:  25%|██▍       | 1599/6470 [19:48<1:00:40,  1.34it/s]
4|1600|Loss: 0.8265788555145264:  25%|██▍       | 1600/6470 [19:49<1:00:56,  1.33it/s]
4|1601|Loss: 0.909420907497406:  25%|██▍       | 1600/6470 [19:49<1:00:56,  1.33it/s] 
4|1601|Loss: 0.909420907497406:  25%|██▍       | 1601/6470 [19:49<1:01:00,  1.33it/s]
4|1602|Loss: 0.8260364532470703:  25%|██▍       | 1601/6470 [19:50<1:01:00,  1.33it/s]
4|1602|Loss: 0.8260364532470703:  25%|██▍       | 1602/6470 [19:50<58:35,  1.38it/s]  
4|1603|Loss: 0.7875979542732239:  25%|██▍       | 1602/6470 [19:51<58:35,  1.38it/s]
4|1603|Loss: 0.7875979542732239:  25%|██▍       | 1603/6470 [19:51<1:01:18,  1.32it/s]
4|1604|Loss: 0.8823065757751465:  25%|██▍       | 1603/6470 [19:51<1:01:18,  1.32it/s]
4|1604|Loss: 0.8823065757751465:  25%|██▍       | 1604/6470 [19:52<1:01:44,  1.31it/s]
4|1605|Loss: 0.9827647805213928:  25%|██▍       | 1604/6470 [19:52<1:01:44,  1.31it/s]
4|1605|Loss: 0.9827647805213928:  25%|██▍       | 1605/6470 [19:52<59:54,  1.35it/s]  
4|1606|Loss: 0.8305255174636841:  25%|██▍       | 1605/6470 [19:53<59:54,  1.35it/s]
4|1606|Loss: 0.8305255174636841:  25%|██▍       | 1606/6470 [19:53<1:02:51,  1.29it/s]
4|1607|Loss: 0.8321353197097778:  25%|██▍       | 1606/6470 [19:54<1:02:51,  1.29it/s]
4|1607|Loss: 0.8321353197097778:  25%|██▍       | 1607/6470 [19:54<1:02:51,  1.29it/s]
4|1608|Loss: 0.9114704132080078:  25%|██▍       | 1607/6470 [19:54<1:02:51,  1.29it/s]
4|1608|Loss: 0.9114704132080078:  25%|██▍       | 1608/6470 [19:55<1:04:13,  1.26it/s]
4|1609|Loss: 0.913245677947998:  25%|██▍       | 1608/6470 [19:55<1:04:13,  1.26it/s] 
4|1609|Loss: 0.913245677947998:  25%|██▍       | 1609/6470 [19:56<1:07:53,  1.19it/s]
4|1610|Loss: 0.8963416814804077:  25%|██▍       | 1609/6470 [19:56<1:07:53,  1.19it/s]
4|1610|Loss: 0.8963416814804077:  25%|██▍       | 1610/6470 [19:57<1:07:12,  1.21it/s]
4|1611|Loss: 0.7900172472000122:  25%|██▍       | 1610/6470 [19:57<1:07:12,  1.21it/s]
4|1611|Loss: 0.7900172472000122:  25%|██▍       | 1611/6470 [19:57<1:07:35,  1.20it/s]
4|1612|Loss: 0.6996545195579529:  25%|██▍       | 1611/6470 [19:58<1:07:35,  1.20it/s]
4|1612|Loss: 0.6996545195579529:  25%|██▍       | 1612/6470 [19:58<1:02:54,  1.29it/s]
4|1613|Loss: 0.8908411264419556:  25%|██▍       | 1612/6470 [19:58<1:02:54,  1.29it/s]
4|1613|Loss: 0.8908411264419556:  25%|██▍       | 1613/6470 [19:59<1:00:22,  1.34it/s]
4|1614|Loss: 0.8289174437522888:  25%|██▍       | 1613/6470 [19:59<1:00:22,  1.34it/s]
4|1614|Loss: 0.8289174437522888:  25%|██▍       | 1614/6470 [20:00<1:04:22,  1.26it/s]
4|1615|Loss: 0.7956563830375671:  25%|██▍       | 1614/6470 [20:00<1:04:22,  1.26it/s]
4|1615|Loss: 0.7956563830375671:  25%|██▍       | 1615/6470 [20:00<1:00:27,  1.34it/s]
4|1616|Loss: 0.717291533946991:  25%|██▍       | 1615/6470 [20:01<1:00:27,  1.34it/s] 
4|1616|Loss: 0.717291533946991:  25%|██▍       | 1616/6470 [20:01<1:00:09,  1.34it/s]
4|1617|Loss: 0.910338819026947:  25%|██▍       | 1616/6470 [20:01<1:00:09,  1.34it/s]
4|1617|Loss: 0.910338819026947:  25%|██▍       | 1617/6470 [20:02<1:00:27,  1.34it/s]
4|1618|Loss: 0.8378255367279053:  25%|██▍       | 1617/6470 [20:02<1:00:27,  1.34it/s]
4|1618|Loss: 0.8378255367279053:  25%|██▌       | 1618/6470 [20:03<1:01:31,  1.31it/s]
4|1619|Loss: 0.7317569255828857:  25%|██▌       | 1618/6470 [20:03<1:01:31,  1.31it/s]
4|1619|Loss: 0.7317569255828857:  25%|██▌       | 1619/6470 [20:03<59:11,  1.37it/s]  
4|1620|Loss: 0.7454720735549927:  25%|██▌       | 1619/6470 [20:04<59:11,  1.37it/s]
4|1620|Loss: 0.7454720735549927:  25%|██▌       | 1620/6470 [20:04<59:55,  1.35it/s]
4|1621|Loss: 0.6591222882270813:  25%|██▌       | 1620/6470 [20:04<59:55,  1.35it/s]
4|1621|Loss: 0.6591222882270813:  25%|██▌       | 1621/6470 [20:05<1:00:40,  1.33it/s]
4|1622|Loss: 0.8565660715103149:  25%|██▌       | 1621/6470 [20:05<1:00:40,  1.33it/s]
4|1622|Loss: 0.8565660715103149:  25%|██▌       | 1622/6470 [20:06<59:35,  1.36it/s]  
4|1623|Loss: 0.7737783789634705:  25%|██▌       | 1622/6470 [20:06<59:35,  1.36it/s]
4|1623|Loss: 0.7737783789634705:  25%|██▌       | 1623/6470 [20:06<1:03:21,  1.28it/s]
4|1624|Loss: 0.8409426212310791:  25%|██▌       | 1623/6470 [20:07<1:03:21,  1.28it/s]
4|1624|Loss: 0.8409426212310791:  25%|██▌       | 1624/6470 [20:07<1:06:27,  1.22it/s]
4|1625|Loss: 0.8556215763092041:  25%|██▌       | 1624/6470 [20:08<1:06:27,  1.22it/s]
4|1625|Loss: 0.8556215763092041:  25%|██▌       | 1625/6470 [20:08<1:03:50,  1.26it/s]
4|1626|Loss: 0.9346433281898499:  25%|██▌       | 1625/6470 [20:08<1:03:50,  1.26it/s]
4|1626|Loss: 0.9346433281898499:  25%|██▌       | 1626/6470 [20:09<1:05:43,  1.23it/s]
4|1627|Loss: 0.8403955698013306:  25%|██▌       | 1626/6470 [20:09<1:05:43,  1.23it/s]
4|1627|Loss: 0.8403955698013306:  25%|██▌       | 1627/6470 [20:10<1:02:55,  1.28it/s]
4|1628|Loss: 0.8389020562171936:  25%|██▌       | 1627/6470 [20:10<1:02:55,  1.28it/s]
4|1628|Loss: 0.8389020562171936:  25%|██▌       | 1628/6470 [20:10<59:44,  1.35it/s]  
4|1629|Loss: 0.7911883592605591:  25%|██▌       | 1628/6470 [20:11<59:44,  1.35it/s]
4|1629|Loss: 0.7911883592605591:  25%|██▌       | 1629/6470 [20:11<1:00:30,  1.33it/s]
4|1630|Loss: 0.7460105419158936:  25%|██▌       | 1629/6470 [20:11<1:00:30,  1.33it/s]
4|1630|Loss: 0.7460105419158936:  25%|██▌       | 1630/6470 [20:12<1:01:28,  1.31it/s]
4|1631|Loss: 0.6187615990638733:  25%|██▌       | 1630/6470 [20:12<1:01:28,  1.31it/s]
4|1631|Loss: 0.6187615990638733:  25%|██▌       | 1631/6470 [20:13<1:00:13,  1.34it/s]
4|1632|Loss: 0.8823735117912292:  25%|██▌       | 1631/6470 [20:13<1:00:13,  1.34it/s]
4|1632|Loss: 0.8823735117912292:  25%|██▌       | 1632/6470 [20:13<59:59,  1.34it/s]  
4|1633|Loss: 1.0487935543060303:  25%|██▌       | 1632/6470 [20:14<59:59,  1.34it/s]
4|1633|Loss: 1.0487935543060303:  25%|██▌       | 1633/6470 [20:14<56:33,  1.43it/s]
4|1634|Loss: 0.9386714696884155:  25%|██▌       | 1633/6470 [20:14<56:33,  1.43it/s]
4|1634|Loss: 0.9386714696884155:  25%|██▌       | 1634/6470 [20:14<54:26,  1.48it/s]
4|1635|Loss: 0.949948787689209:  25%|██▌       | 1634/6470 [20:15<54:26,  1.48it/s] 
4|1635|Loss: 0.949948787689209:  25%|██▌       | 1635/6470 [20:15<58:24,  1.38it/s]
4|1636|Loss: 0.8734380602836609:  25%|██▌       | 1635/6470 [20:16<58:24,  1.38it/s]
4|1636|Loss: 0.8734380602836609:  25%|██▌       | 1636/6470 [20:16<59:48,  1.35it/s]
4|1637|Loss: 0.7848193049430847:  25%|██▌       | 1636/6470 [20:17<59:48,  1.35it/s]
4|1637|Loss: 0.7848193049430847:  25%|██▌       | 1637/6470 [20:17<1:01:25,  1.31it/s]
4|1638|Loss: 0.8520113229751587:  25%|██▌       | 1637/6470 [20:17<1:01:25,  1.31it/s]
4|1638|Loss: 0.8520113229751587:  25%|██▌       | 1638/6470 [20:18<1:04:30,  1.25it/s]
4|1639|Loss: 0.8784466981887817:  25%|██▌       | 1638/6470 [20:18<1:04:30,  1.25it/s]
4|1639|Loss: 0.8784466981887817:  25%|██▌       | 1639/6470 [20:19<1:10:10,  1.15it/s]
4|1640|Loss: 0.6960512399673462:  25%|██▌       | 1639/6470 [20:19<1:10:10,  1.15it/s]
4|1640|Loss: 0.6960512399673462:  25%|██▌       | 1640/6470 [20:20<1:05:02,  1.24it/s]
4|1641|Loss: 0.8186333179473877:  25%|██▌       | 1640/6470 [20:20<1:05:02,  1.24it/s]
4|1641|Loss: 0.8186333179473877:  25%|██▌       | 1641/6470 [20:20<1:03:53,  1.26it/s]
4|1642|Loss: 1.0572247505187988:  25%|██▌       | 1641/6470 [20:21<1:03:53,  1.26it/s]
4|1642|Loss: 1.0572247505187988:  25%|██▌       | 1642/6470 [20:21<1:06:02,  1.22it/s]
4|1643|Loss: 0.769740641117096:  25%|██▌       | 1642/6470 [20:22<1:06:02,  1.22it/s] 
4|1643|Loss: 0.769740641117096:  25%|██▌       | 1643/6470 [20:22<59:55,  1.34it/s]  
4|1644|Loss: 0.8470367193222046:  25%|██▌       | 1643/6470 [20:22<59:55,  1.34it/s]
4|1644|Loss: 0.8470367193222046:  25%|██▌       | 1644/6470 [20:22<59:21,  1.36it/s]
4|1645|Loss: 0.8912187814712524:  25%|██▌       | 1644/6470 [20:23<59:21,  1.36it/s]
4|1645|Loss: 0.8912187814712524:  25%|██▌       | 1645/6470 [20:23<1:01:08,  1.32it/s]
4|1646|Loss: 0.7134471535682678:  25%|██▌       | 1645/6470 [20:24<1:01:08,  1.32it/s]
4|1646|Loss: 0.7134471535682678:  25%|██▌       | 1646/6470 [20:24<57:24,  1.40it/s]  
4|1647|Loss: 0.7993385791778564:  25%|██▌       | 1646/6470 [20:24<57:24,  1.40it/s]
4|1647|Loss: 0.7993385791778564:  25%|██▌       | 1647/6470 [20:24<53:35,  1.50it/s]
4|1648|Loss: 0.8718730807304382:  25%|██▌       | 1647/6470 [20:25<53:35,  1.50it/s]
4|1648|Loss: 0.8718730807304382:  25%|██▌       | 1648/6470 [20:25<52:57,  1.52it/s]
4|1649|Loss: 0.7994375228881836:  25%|██▌       | 1648/6470 [20:25<52:57,  1.52it/s]
4|1649|Loss: 0.7994375228881836:  25%|██▌       | 1649/6470 [20:26<53:51,  1.49it/s]
4|1650|Loss: 0.6854060888290405:  25%|██▌       | 1649/6470 [20:26<53:51,  1.49it/s]
4|1650|Loss: 0.6854060888290405:  26%|██▌       | 1650/6470 [20:26<53:51,  1.49it/s]
4|1651|Loss: 0.8926379680633545:  26%|██▌       | 1650/6470 [20:27<53:51,  1.49it/s]
4|1651|Loss: 0.8926379680633545:  26%|██▌       | 1651/6470 [20:27<52:47,  1.52it/s]
4|1652|Loss: 0.7179545164108276:  26%|██▌       | 1651/6470 [20:27<52:47,  1.52it/s]
4|1652|Loss: 0.7179545164108276:  26%|██▌       | 1652/6470 [20:28<53:30,  1.50it/s]
4|1653|Loss: 0.8347862362861633:  26%|██▌       | 1652/6470 [20:28<53:30,  1.50it/s]
4|1653|Loss: 0.8347862362861633:  26%|██▌       | 1653/6470 [20:29<58:19,  1.38it/s]
4|1654|Loss: 0.8120583295822144:  26%|██▌       | 1653/6470 [20:29<58:19,  1.38it/s]
4|1654|Loss: 0.8120583295822144:  26%|██▌       | 1654/6470 [20:29<1:01:55,  1.30it/s]
4|1655|Loss: 0.8294523358345032:  26%|██▌       | 1654/6470 [20:30<1:01:55,  1.30it/s]
4|1655|Loss: 0.8294523358345032:  26%|██▌       | 1655/6470 [20:30<1:00:27,  1.33it/s]
4|1656|Loss: 0.6633445620536804:  26%|██▌       | 1655/6470 [20:31<1:00:27,  1.33it/s]
4|1656|Loss: 0.6633445620536804:  26%|██▌       | 1656/6470 [20:31<1:03:33,  1.26it/s]
4|1657|Loss: 0.8688987493515015:  26%|██▌       | 1656/6470 [20:31<1:03:33,  1.26it/s]
4|1657|Loss: 0.8688987493515015:  26%|██▌       | 1657/6470 [20:32<58:18,  1.38it/s]  
4|1658|Loss: 0.80014568567276:  26%|██▌       | 1657/6470 [20:32<58:18,  1.38it/s]  
4|1658|Loss: 0.80014568567276:  26%|██▌       | 1658/6470 [20:32<55:57,  1.43it/s]
4|1659|Loss: 0.8698044419288635:  26%|██▌       | 1658/6470 [20:33<55:57,  1.43it/s]
4|1659|Loss: 0.8698044419288635:  26%|██▌       | 1659/6470 [20:33<53:24,  1.50it/s]
4|1660|Loss: 0.7956491708755493:  26%|██▌       | 1659/6470 [20:33<53:24,  1.50it/s]
4|1660|Loss: 0.7956491708755493:  26%|██▌       | 1660/6470 [20:34<52:49,  1.52it/s]
4|1661|Loss: 0.9524756073951721:  26%|██▌       | 1660/6470 [20:34<52:49,  1.52it/s]
4|1661|Loss: 0.9524756073951721:  26%|██▌       | 1661/6470 [20:34<49:28,  1.62it/s]
4|1662|Loss: 0.8552903532981873:  26%|██▌       | 1661/6470 [20:34<49:28,  1.62it/s]
4|1662|Loss: 0.8552903532981873:  26%|██▌       | 1662/6470 [20:35<52:02,  1.54it/s]
4|1663|Loss: 0.9247159361839294:  26%|██▌       | 1662/6470 [20:35<52:02,  1.54it/s]
4|1663|Loss: 0.9247159361839294:  26%|██▌       | 1663/6470 [20:35<50:54,  1.57it/s]
4|1664|Loss: 0.6873880624771118:  26%|██▌       | 1663/6470 [20:36<50:54,  1.57it/s]
4|1664|Loss: 0.6873880624771118:  26%|██▌       | 1664/6470 [20:36<54:57,  1.46it/s]
4|1665|Loss: 0.6619834303855896:  26%|██▌       | 1664/6470 [20:37<54:57,  1.46it/s]
4|1665|Loss: 0.6619834303855896:  26%|██▌       | 1665/6470 [20:37<1:00:15,  1.33it/s]
4|1666|Loss: 0.8073267340660095:  26%|██▌       | 1665/6470 [20:38<1:00:15,  1.33it/s]
4|1666|Loss: 0.8073267340660095:  26%|██▌       | 1666/6470 [20:38<1:01:16,  1.31it/s]
4|1667|Loss: 0.8370413184165955:  26%|██▌       | 1666/6470 [20:38<1:01:16,  1.31it/s]
4|1667|Loss: 0.8370413184165955:  26%|██▌       | 1667/6470 [20:39<1:07:36,  1.18it/s]
4|1668|Loss: 0.9318261742591858:  26%|██▌       | 1667/6470 [20:39<1:07:36,  1.18it/s]
4|1668|Loss: 0.9318261742591858:  26%|██▌       | 1668/6470 [20:40<1:10:24,  1.14it/s]
4|1669|Loss: 0.8112756609916687:  26%|██▌       | 1668/6470 [20:40<1:10:24,  1.14it/s]
4|1669|Loss: 0.8112756609916687:  26%|██▌       | 1669/6470 [20:41<1:14:13,  1.08it/s]
4|1670|Loss: 0.9738894104957581:  26%|██▌       | 1669/6470 [20:41<1:14:13,  1.08it/s]
4|1670|Loss: 0.9738894104957581:  26%|██▌       | 1670/6470 [20:42<1:07:12,  1.19it/s]
4|1671|Loss: 0.8798415064811707:  26%|██▌       | 1670/6470 [20:42<1:07:12,  1.19it/s]
4|1671|Loss: 0.8798415064811707:  26%|██▌       | 1671/6470 [20:42<1:07:07,  1.19it/s]
4|1672|Loss: 0.7810587882995605:  26%|██▌       | 1671/6470 [20:43<1:07:07,  1.19it/s]
4|1672|Loss: 0.7810587882995605:  26%|██▌       | 1672/6470 [20:43<1:04:20,  1.24it/s]
4|1673|Loss: 0.8087567090988159:  26%|██▌       | 1672/6470 [20:44<1:04:20,  1.24it/s]
4|1673|Loss: 0.8087567090988159:  26%|██▌       | 1673/6470 [20:44<1:03:35,  1.26it/s]
4|1674|Loss: 0.7478570342063904:  26%|██▌       | 1673/6470 [20:44<1:03:35,  1.26it/s]
4|1674|Loss: 0.7478570342063904:  26%|██▌       | 1674/6470 [20:45<1:01:01,  1.31it/s]
4|1675|Loss: 0.9029887914657593:  26%|██▌       | 1674/6470 [20:45<1:01:01,  1.31it/s]
4|1675|Loss: 0.9029887914657593:  26%|██▌       | 1675/6470 [20:45<59:36,  1.34it/s]  
4|1676|Loss: 0.9207977056503296:  26%|██▌       | 1675/6470 [20:46<59:36,  1.34it/s]
4|1676|Loss: 0.9207977056503296:  26%|██▌       | 1676/6470 [20:46<1:02:07,  1.29it/s]
4|1677|Loss: 0.7475018501281738:  26%|██▌       | 1676/6470 [20:47<1:02:07,  1.29it/s]
4|1677|Loss: 0.7475018501281738:  26%|██▌       | 1677/6470 [20:47<1:02:01,  1.29it/s]
4|1678|Loss: 0.8507909774780273:  26%|██▌       | 1677/6470 [20:47<1:02:01,  1.29it/s]
4|1678|Loss: 0.8507909774780273:  26%|██▌       | 1678/6470 [20:48<1:00:16,  1.32it/s]
4|1679|Loss: 0.7408227324485779:  26%|██▌       | 1678/6470 [20:48<1:00:16,  1.32it/s]
4|1679|Loss: 0.7408227324485779:  26%|██▌       | 1679/6470 [20:48<57:14,  1.40it/s]  
4|1680|Loss: 0.8439866304397583:  26%|██▌       | 1679/6470 [20:49<57:14,  1.40it/s]
4|1680|Loss: 0.8439866304397583:  26%|██▌       | 1680/6470 [20:49<1:00:53,  1.31it/s]
4|1681|Loss: 0.7474639415740967:  26%|██▌       | 1680/6470 [20:50<1:00:53,  1.31it/s]
4|1681|Loss: 0.7474639415740967:  26%|██▌       | 1681/6470 [20:50<1:00:55,  1.31it/s]
4|1682|Loss: 0.8010830283164978:  26%|██▌       | 1681/6470 [20:50<1:00:55,  1.31it/s]
4|1682|Loss: 0.8010830283164978:  26%|██▌       | 1682/6470 [20:51<1:01:07,  1.31it/s]
4|1683|Loss: 0.8101274371147156:  26%|██▌       | 1682/6470 [20:51<1:01:07,  1.31it/s]
4|1683|Loss: 0.8101274371147156:  26%|██▌       | 1683/6470 [20:51<1:01:21,  1.30it/s]
4|1684|Loss: 0.7609941363334656:  26%|██▌       | 1683/6470 [20:52<1:01:21,  1.30it/s]
4|1684|Loss: 0.7609941363334656:  26%|██▌       | 1684/6470 [20:52<1:01:08,  1.30it/s]
4|1685|Loss: 0.7698671817779541:  26%|██▌       | 1684/6470 [20:53<1:01:08,  1.30it/s]
4|1685|Loss: 0.7698671817779541:  26%|██▌       | 1685/6470 [20:53<59:38,  1.34it/s]  
4|1686|Loss: 0.7795512080192566:  26%|██▌       | 1685/6470 [20:53<59:38,  1.34it/s]
4|1686|Loss: 0.7795512080192566:  26%|██▌       | 1686/6470 [20:54<1:00:17,  1.32it/s]
4|1687|Loss: 0.7205751538276672:  26%|██▌       | 1686/6470 [20:54<1:00:17,  1.32it/s]
4|1687|Loss: 0.7205751538276672:  26%|██▌       | 1687/6470 [20:55<1:03:02,  1.26it/s]
4|1688|Loss: 0.7822573184967041:  26%|██▌       | 1687/6470 [20:55<1:03:02,  1.26it/s]
4|1688|Loss: 0.7822573184967041:  26%|██▌       | 1688/6470 [20:55<1:00:25,  1.32it/s]
4|1689|Loss: 0.869040310382843:  26%|██▌       | 1688/6470 [20:56<1:00:25,  1.32it/s] 
4|1689|Loss: 0.869040310382843:  26%|██▌       | 1689/6470 [20:56<1:02:13,  1.28it/s]
4|1690|Loss: 0.8317465782165527:  26%|██▌       | 1689/6470 [20:56<1:02:13,  1.28it/s]
4|1690|Loss: 0.8317465782165527:  26%|██▌       | 1690/6470 [20:57<1:01:53,  1.29it/s]
4|1691|Loss: 0.8123891949653625:  26%|██▌       | 1690/6470 [20:57<1:01:53,  1.29it/s]
4|1691|Loss: 0.8123891949653625:  26%|██▌       | 1691/6470 [20:58<1:00:16,  1.32it/s]
4|1692|Loss: 0.644195020198822:  26%|██▌       | 1691/6470 [20:58<1:00:16,  1.32it/s] 
4|1692|Loss: 0.644195020198822:  26%|██▌       | 1692/6470 [20:58<59:54,  1.33it/s]  
4|1693|Loss: 0.8216854333877563:  26%|██▌       | 1692/6470 [20:59<59:54,  1.33it/s]
4|1693|Loss: 0.8216854333877563:  26%|██▌       | 1693/6470 [20:59<59:40,  1.33it/s]
4|1694|Loss: 0.7961689233779907:  26%|██▌       | 1693/6470 [20:59<59:40,  1.33it/s]
4|1694|Loss: 0.7961689233779907:  26%|██▌       | 1694/6470 [21:00<57:12,  1.39it/s]
4|1695|Loss: 0.8078585863113403:  26%|██▌       | 1694/6470 [21:00<57:12,  1.39it/s]
4|1695|Loss: 0.8078585863113403:  26%|██▌       | 1695/6470 [21:00<56:08,  1.42it/s]
4|1696|Loss: 0.7713518738746643:  26%|██▌       | 1695/6470 [21:01<56:08,  1.42it/s]
4|1696|Loss: 0.7713518738746643:  26%|██▌       | 1696/6470 [21:01<57:34,  1.38it/s]
4|1697|Loss: 1.0032342672348022:  26%|██▌       | 1696/6470 [21:01<57:34,  1.38it/s]
4|1697|Loss: 1.0032342672348022:  26%|██▌       | 1697/6470 [21:02<54:22,  1.46it/s]
4|1698|Loss: 0.9051164984703064:  26%|██▌       | 1697/6470 [21:02<54:22,  1.46it/s]
4|1698|Loss: 0.9051164984703064:  26%|██▌       | 1698/6470 [21:02<54:45,  1.45it/s]
4|1699|Loss: 0.7130301594734192:  26%|██▌       | 1698/6470 [21:03<54:45,  1.45it/s]
4|1699|Loss: 0.7130301594734192:  26%|██▋       | 1699/6470 [21:03<55:08,  1.44it/s]
4|1700|Loss: 0.9623769521713257:  26%|██▋       | 1699/6470 [21:04<55:08,  1.44it/s]
4|1700|Loss: 0.9623769521713257:  26%|██▋       | 1700/6470 [21:04<59:55,  1.33it/s]
4|1701|Loss: 0.811759352684021:  26%|██▋       | 1700/6470 [21:05<59:55,  1.33it/s] 
4|1701|Loss: 0.811759352684021:  26%|██▋       | 1701/6470 [21:05<1:04:49,  1.23it/s]
4|1702|Loss: 0.8181709051132202:  26%|██▋       | 1701/6470 [21:05<1:04:49,  1.23it/s]
4|1702|Loss: 0.8181709051132202:  26%|██▋       | 1702/6470 [21:06<1:05:37,  1.21it/s]
4|1703|Loss: 0.7660722732543945:  26%|██▋       | 1702/6470 [21:06<1:05:37,  1.21it/s]
4|1703|Loss: 0.7660722732543945:  26%|██▋       | 1703/6470 [21:07<1:04:10,  1.24it/s]
4|1704|Loss: 0.9773126244544983:  26%|██▋       | 1703/6470 [21:07<1:04:10,  1.24it/s]
4|1704|Loss: 0.9773126244544983:  26%|██▋       | 1704/6470 [21:07<1:03:34,  1.25it/s]
4|1705|Loss: 0.8167772889137268:  26%|██▋       | 1704/6470 [21:08<1:03:34,  1.25it/s]
4|1705|Loss: 0.8167772889137268:  26%|██▋       | 1705/6470 [21:08<1:04:50,  1.22it/s]
4|1706|Loss: 0.9190932512283325:  26%|██▋       | 1705/6470 [21:09<1:04:50,  1.22it/s]
4|1706|Loss: 0.9190932512283325:  26%|██▋       | 1706/6470 [21:09<1:05:32,  1.21it/s]
4|1707|Loss: 0.9128628373146057:  26%|██▋       | 1706/6470 [21:10<1:05:32,  1.21it/s]
4|1707|Loss: 0.9128628373146057:  26%|██▋       | 1707/6470 [21:10<1:07:12,  1.18it/s]
4|1708|Loss: 0.7560524344444275:  26%|██▋       | 1707/6470 [21:10<1:07:12,  1.18it/s]
4|1708|Loss: 0.7560524344444275:  26%|██▋       | 1708/6470 [21:11<1:06:43,  1.19it/s]
4|1709|Loss: 0.8577532768249512:  26%|██▋       | 1708/6470 [21:11<1:06:43,  1.19it/s]
4|1709|Loss: 0.8577532768249512:  26%|██▋       | 1709/6470 [21:12<1:05:40,  1.21it/s]
4|1710|Loss: 0.7792964577674866:  26%|██▋       | 1709/6470 [21:12<1:05:40,  1.21it/s]
4|1710|Loss: 0.7792964577674866:  26%|██▋       | 1710/6470 [21:12<59:51,  1.33it/s]  
4|1711|Loss: 0.8578594326972961:  26%|██▋       | 1710/6470 [21:12<59:51,  1.33it/s]
4|1711|Loss: 0.8578594326972961:  26%|██▋       | 1711/6470 [21:13<57:31,  1.38it/s]
4|1712|Loss: 0.7292693257331848:  26%|██▋       | 1711/6470 [21:13<57:31,  1.38it/s]
4|1712|Loss: 0.7292693257331848:  26%|██▋       | 1712/6470 [21:14<58:12,  1.36it/s]
4|1713|Loss: 0.9012797474861145:  26%|██▋       | 1712/6470 [21:14<58:12,  1.36it/s]
4|1713|Loss: 0.9012797474861145:  26%|██▋       | 1713/6470 [21:14<56:54,  1.39it/s]
4|1714|Loss: 0.7304233908653259:  26%|██▋       | 1713/6470 [21:15<56:54,  1.39it/s]
4|1714|Loss: 0.7304233908653259:  26%|██▋       | 1714/6470 [21:15<54:54,  1.44it/s]
4|1715|Loss: 0.6337776780128479:  26%|██▋       | 1714/6470 [21:15<54:54,  1.44it/s]
4|1715|Loss: 0.6337776780128479:  27%|██▋       | 1715/6470 [21:15<52:25,  1.51it/s]
4|1716|Loss: 0.846488893032074:  27%|██▋       | 1715/6470 [21:16<52:25,  1.51it/s] 
4|1716|Loss: 0.846488893032074:  27%|██▋       | 1716/6470 [21:16<52:13,  1.52it/s]
4|1717|Loss: 0.8546700477600098:  27%|██▋       | 1716/6470 [21:17<52:13,  1.52it/s]
4|1717|Loss: 0.8546700477600098:  27%|██▋       | 1717/6470 [21:17<53:24,  1.48it/s]
4|1718|Loss: 0.8368247747421265:  27%|██▋       | 1717/6470 [21:17<53:24,  1.48it/s]
4|1718|Loss: 0.8368247747421265:  27%|██▋       | 1718/6470 [21:18<58:49,  1.35it/s]
4|1719|Loss: 0.8046863079071045:  27%|██▋       | 1718/6470 [21:18<58:49,  1.35it/s]
4|1719|Loss: 0.8046863079071045:  27%|██▋       | 1719/6470 [21:18<57:46,  1.37it/s]
4|1720|Loss: 0.9525986909866333:  27%|██▋       | 1719/6470 [21:19<57:46,  1.37it/s]
4|1720|Loss: 0.9525986909866333:  27%|██▋       | 1720/6470 [21:19<58:40,  1.35it/s]
4|1721|Loss: 0.802962064743042:  27%|██▋       | 1720/6470 [21:20<58:40,  1.35it/s] 
4|1721|Loss: 0.802962064743042:  27%|██▋       | 1721/6470 [21:20<1:00:37,  1.31it/s]
4|1722|Loss: 0.7423186302185059:  27%|██▋       | 1721/6470 [21:20<1:00:37,  1.31it/s]
4|1722|Loss: 0.7423186302185059:  27%|██▋       | 1722/6470 [21:21<59:40,  1.33it/s]  
4|1723|Loss: 0.885343611240387:  27%|██▋       | 1722/6470 [21:21<59:40,  1.33it/s] 
4|1723|Loss: 0.885343611240387:  27%|██▋       | 1723/6470 [21:21<57:40,  1.37it/s]
4|1724|Loss: 0.8622082471847534:  27%|██▋       | 1723/6470 [21:22<57:40,  1.37it/s]
4|1724|Loss: 0.8622082471847534:  27%|██▋       | 1724/6470 [21:22<1:02:10,  1.27it/s]
4|1725|Loss: 0.8400046825408936:  27%|██▋       | 1724/6470 [21:23<1:02:10,  1.27it/s]
4|1725|Loss: 0.8400046825408936:  27%|██▋       | 1725/6470 [21:23<1:02:51,  1.26it/s]
4|1726|Loss: 0.7511017918586731:  27%|██▋       | 1725/6470 [21:24<1:02:51,  1.26it/s]
4|1726|Loss: 0.7511017918586731:  27%|██▋       | 1726/6470 [21:24<1:00:12,  1.31it/s]
4|1727|Loss: 0.9976650476455688:  27%|██▋       | 1726/6470 [21:24<1:00:12,  1.31it/s]
4|1727|Loss: 0.9976650476455688:  27%|██▋       | 1727/6470 [21:24<57:16,  1.38it/s]  
4|1728|Loss: 0.8251561522483826:  27%|██▋       | 1727/6470 [21:25<57:16,  1.38it/s]
4|1728|Loss: 0.8251561522483826:  27%|██▋       | 1728/6470 [21:25<1:00:15,  1.31it/s]
4|1729|Loss: 0.9454470872879028:  27%|██▋       | 1728/6470 [21:26<1:00:15,  1.31it/s]
4|1729|Loss: 0.9454470872879028:  27%|██▋       | 1729/6470 [21:26<1:01:38,  1.28it/s]
4|1730|Loss: 0.8315046429634094:  27%|██▋       | 1729/6470 [21:27<1:01:38,  1.28it/s]
4|1730|Loss: 0.8315046429634094:  27%|██▋       | 1730/6470 [21:27<1:02:26,  1.27it/s]
4|1731|Loss: 0.8108832836151123:  27%|██▋       | 1730/6470 [21:27<1:02:26,  1.27it/s]
4|1731|Loss: 0.8108832836151123:  27%|██▋       | 1731/6470 [21:28<1:04:13,  1.23it/s]
4|1732|Loss: 0.7359916567802429:  27%|██▋       | 1731/6470 [21:28<1:04:13,  1.23it/s]
4|1732|Loss: 0.7359916567802429:  27%|██▋       | 1732/6470 [21:29<1:04:18,  1.23it/s]
4|1733|Loss: 0.734923779964447:  27%|██▋       | 1732/6470 [21:29<1:04:18,  1.23it/s] 
4|1733|Loss: 0.734923779964447:  27%|██▋       | 1733/6470 [21:30<1:07:19,  1.17it/s]
4|1734|Loss: 0.8791877031326294:  27%|██▋       | 1733/6470 [21:30<1:07:19,  1.17it/s]
4|1734|Loss: 0.8791877031326294:  27%|██▋       | 1734/6470 [21:30<1:05:00,  1.21it/s]
4|1735|Loss: 0.862404465675354:  27%|██▋       | 1734/6470 [21:31<1:05:00,  1.21it/s] 
4|1735|Loss: 0.862404465675354:  27%|██▋       | 1735/6470 [21:31<1:06:25,  1.19it/s]
4|1736|Loss: 0.7907761931419373:  27%|██▋       | 1735/6470 [21:32<1:06:25,  1.19it/s]
4|1736|Loss: 0.7907761931419373:  27%|██▋       | 1736/6470 [21:32<1:03:50,  1.24it/s]
4|1737|Loss: 0.9256742000579834:  27%|██▋       | 1736/6470 [21:32<1:03:50,  1.24it/s]
4|1737|Loss: 0.9256742000579834:  27%|██▋       | 1737/6470 [21:33<1:02:56,  1.25it/s]
4|1738|Loss: 0.8630731105804443:  27%|██▋       | 1737/6470 [21:33<1:02:56,  1.25it/s]
4|1738|Loss: 0.8630731105804443:  27%|██▋       | 1738/6470 [21:33<1:00:15,  1.31it/s]
4|1739|Loss: 0.8123031854629517:  27%|██▋       | 1738/6470 [21:34<1:00:15,  1.31it/s]
4|1739|Loss: 0.8123031854629517:  27%|██▋       | 1739/6470 [21:34<57:53,  1.36it/s]  
4|1740|Loss: 0.8714960217475891:  27%|██▋       | 1739/6470 [21:34<57:53,  1.36it/s]
4|1740|Loss: 0.8714960217475891:  27%|██▋       | 1740/6470 [21:35<59:18,  1.33it/s]
4|1741|Loss: 0.8213720917701721:  27%|██▋       | 1740/6470 [21:35<59:18,  1.33it/s]
4|1741|Loss: 0.8213720917701721:  27%|██▋       | 1741/6470 [21:36<1:03:10,  1.25it/s]
4|1742|Loss: 0.8477656245231628:  27%|██▋       | 1741/6470 [21:36<1:03:10,  1.25it/s]
4|1742|Loss: 0.8477656245231628:  27%|██▋       | 1742/6470 [21:37<1:01:31,  1.28it/s]
4|1743|Loss: 0.8828286528587341:  27%|██▋       | 1742/6470 [21:37<1:01:31,  1.28it/s]
4|1743|Loss: 0.8828286528587341:  27%|██▋       | 1743/6470 [21:37<1:01:34,  1.28it/s]
4|1744|Loss: 0.8035128712654114:  27%|██▋       | 1743/6470 [21:38<1:01:34,  1.28it/s]
4|1744|Loss: 0.8035128712654114:  27%|██▋       | 1744/6470 [21:38<59:55,  1.31it/s]  
4|1745|Loss: 0.9030771851539612:  27%|██▋       | 1744/6470 [21:38<59:55,  1.31it/s]
4|1745|Loss: 0.9030771851539612:  27%|██▋       | 1745/6470 [21:39<58:19,  1.35it/s]
4|1746|Loss: 0.9022371768951416:  27%|██▋       | 1745/6470 [21:39<58:19,  1.35it/s]
4|1746|Loss: 0.9022371768951416:  27%|██▋       | 1746/6470 [21:39<59:09,  1.33it/s]
4|1747|Loss: 0.767269492149353:  27%|██▋       | 1746/6470 [21:40<59:09,  1.33it/s] 
4|1747|Loss: 0.767269492149353:  27%|██▋       | 1747/6470 [21:40<59:54,  1.31it/s]
4|1748|Loss: 0.982542097568512:  27%|██▋       | 1747/6470 [21:41<59:54,  1.31it/s]
4|1748|Loss: 0.982542097568512:  27%|██▋       | 1748/6470 [21:41<1:00:47,  1.29it/s]
4|1749|Loss: 0.7289184331893921:  27%|██▋       | 1748/6470 [21:41<1:00:47,  1.29it/s]
4|1749|Loss: 0.7289184331893921:  27%|██▋       | 1749/6470 [21:42<1:00:35,  1.30it/s]
4|1750|Loss: 0.8775060176849365:  27%|██▋       | 1749/6470 [21:42<1:00:35,  1.30it/s]
4|1750|Loss: 0.8775060176849365:  27%|██▋       | 1750/6470 [21:43<59:57,  1.31it/s]  
4|1751|Loss: 0.9816742539405823:  27%|██▋       | 1750/6470 [21:43<59:57,  1.31it/s]
4|1751|Loss: 0.9816742539405823:  27%|██▋       | 1751/6470 [21:43<57:07,  1.38it/s]
4|1752|Loss: 0.854947030544281:  27%|██▋       | 1751/6470 [21:44<57:07,  1.38it/s] 
4|1752|Loss: 0.854947030544281:  27%|██▋       | 1752/6470 [21:44<56:55,  1.38it/s]
4|1753|Loss: 0.8425132036209106:  27%|██▋       | 1752/6470 [21:44<56:55,  1.38it/s]
4|1753|Loss: 0.8425132036209106:  27%|██▋       | 1753/6470 [21:45<55:48,  1.41it/s]
4|1754|Loss: 0.7479966878890991:  27%|██▋       | 1753/6470 [21:45<55:48,  1.41it/s]
4|1754|Loss: 0.7479966878890991:  27%|██▋       | 1754/6470 [21:45<54:43,  1.44it/s]
4|1755|Loss: 0.8998536467552185:  27%|██▋       | 1754/6470 [21:46<54:43,  1.44it/s]
4|1755|Loss: 0.8998536467552185:  27%|██▋       | 1755/6470 [21:46<54:04,  1.45it/s]
4|1756|Loss: 1.0273041725158691:  27%|██▋       | 1755/6470 [21:46<54:04,  1.45it/s]
4|1756|Loss: 1.0273041725158691:  27%|██▋       | 1756/6470 [21:47<54:02,  1.45it/s]
4|1757|Loss: 0.8078989386558533:  27%|██▋       | 1756/6470 [21:47<54:02,  1.45it/s]
4|1757|Loss: 0.8078989386558533:  27%|██▋       | 1757/6470 [21:47<54:21,  1.44it/s]
4|1758|Loss: 0.7466166615486145:  27%|██▋       | 1757/6470 [21:48<54:21,  1.44it/s]
4|1758|Loss: 0.7466166615486145:  27%|██▋       | 1758/6470 [21:48<53:29,  1.47it/s]
4|1759|Loss: 0.8229579329490662:  27%|██▋       | 1758/6470 [21:48<53:29,  1.47it/s]
4|1759|Loss: 0.8229579329490662:  27%|██▋       | 1759/6470 [21:49<55:25,  1.42it/s]
4|1760|Loss: 0.8264991641044617:  27%|██▋       | 1759/6470 [21:49<55:25,  1.42it/s]
4|1760|Loss: 0.8264991641044617:  27%|██▋       | 1760/6470 [21:50<58:23,  1.34it/s]
4|1761|Loss: 0.8276206254959106:  27%|██▋       | 1760/6470 [21:50<58:23,  1.34it/s]
4|1761|Loss: 0.8276206254959106:  27%|██▋       | 1761/6470 [21:50<55:10,  1.42it/s]
4|1762|Loss: 0.7407200336456299:  27%|██▋       | 1761/6470 [21:51<55:10,  1.42it/s]
4|1762|Loss: 0.7407200336456299:  27%|██▋       | 1762/6470 [21:51<59:02,  1.33it/s]
4|1763|Loss: 0.951905369758606:  27%|██▋       | 1762/6470 [21:52<59:02,  1.33it/s] 
4|1763|Loss: 0.951905369758606:  27%|██▋       | 1763/6470 [21:52<1:00:14,  1.30it/s]
4|1764|Loss: 0.8565377593040466:  27%|██▋       | 1763/6470 [21:52<1:00:14,  1.30it/s]
4|1764|Loss: 0.8565377593040466:  27%|██▋       | 1764/6470 [21:52<56:18,  1.39it/s]  
4|1765|Loss: 0.8126722574234009:  27%|██▋       | 1764/6470 [21:53<56:18,  1.39it/s]
4|1765|Loss: 0.8126722574234009:  27%|██▋       | 1765/6470 [21:53<58:00,  1.35it/s]
4|1766|Loss: 0.8275787234306335:  27%|██▋       | 1765/6470 [21:54<58:00,  1.35it/s]
4|1766|Loss: 0.8275787234306335:  27%|██▋       | 1766/6470 [21:54<59:04,  1.33it/s]
4|1767|Loss: 0.8490824103355408:  27%|██▋       | 1766/6470 [21:54<59:04,  1.33it/s]
4|1767|Loss: 0.8490824103355408:  27%|██▋       | 1767/6470 [21:55<59:19,  1.32it/s]
4|1768|Loss: 0.8664774894714355:  27%|██▋       | 1767/6470 [21:55<59:19,  1.32it/s]
4|1768|Loss: 0.8664774894714355:  27%|██▋       | 1768/6470 [21:56<1:00:12,  1.30it/s]
4|1769|Loss: 0.8167500495910645:  27%|██▋       | 1768/6470 [21:56<1:00:12,  1.30it/s]
4|1769|Loss: 0.8167500495910645:  27%|██▋       | 1769/6470 [21:56<1:00:43,  1.29it/s]
4|1770|Loss: 0.8058118224143982:  27%|██▋       | 1769/6470 [21:57<1:00:43,  1.29it/s]
4|1770|Loss: 0.8058118224143982:  27%|██▋       | 1770/6470 [21:57<1:00:20,  1.30it/s]
4|1771|Loss: 0.8835838437080383:  27%|██▋       | 1770/6470 [21:58<1:00:20,  1.30it/s]
4|1771|Loss: 0.8835838437080383:  27%|██▋       | 1771/6470 [21:58<1:02:25,  1.25it/s]
4|1772|Loss: 0.9027954936027527:  27%|██▋       | 1771/6470 [21:58<1:02:25,  1.25it/s]
4|1772|Loss: 0.9027954936027527:  27%|██▋       | 1772/6470 [21:59<1:00:14,  1.30it/s]
4|1773|Loss: 0.9126024842262268:  27%|██▋       | 1772/6470 [21:59<1:00:14,  1.30it/s]
4|1773|Loss: 0.9126024842262268:  27%|██▋       | 1773/6470 [21:59<1:00:33,  1.29it/s]
4|1774|Loss: 0.8993557095527649:  27%|██▋       | 1773/6470 [22:00<1:00:33,  1.29it/s]
4|1774|Loss: 0.8993557095527649:  27%|██▋       | 1774/6470 [22:00<1:01:35,  1.27it/s]
4|1775|Loss: 0.8665287494659424:  27%|██▋       | 1774/6470 [22:01<1:01:35,  1.27it/s]
4|1775|Loss: 0.8665287494659424:  27%|██▋       | 1775/6470 [22:01<1:02:12,  1.26it/s]
4|1776|Loss: 0.7730793356895447:  27%|██▋       | 1775/6470 [22:02<1:02:12,  1.26it/s]
4|1776|Loss: 0.7730793356895447:  27%|██▋       | 1776/6470 [22:02<59:55,  1.31it/s]  
4|1777|Loss: 0.8038002848625183:  27%|██▋       | 1776/6470 [22:02<59:55,  1.31it/s]
4|1777|Loss: 0.8038002848625183:  27%|██▋       | 1777/6470 [22:03<1:02:48,  1.25it/s]
4|1778|Loss: 0.8163368105888367:  27%|██▋       | 1777/6470 [22:03<1:02:48,  1.25it/s]
4|1778|Loss: 0.8163368105888367:  27%|██▋       | 1778/6470 [22:04<1:03:30,  1.23it/s]
4|1779|Loss: 0.7426874041557312:  27%|██▋       | 1778/6470 [22:04<1:03:30,  1.23it/s]
4|1779|Loss: 0.7426874041557312:  27%|██▋       | 1779/6470 [22:04<57:16,  1.37it/s]  
4|1780|Loss: 0.9793094992637634:  27%|██▋       | 1779/6470 [22:04<57:16,  1.37it/s]
4|1780|Loss: 0.9793094992637634:  28%|██▊       | 1780/6470 [22:05<54:37,  1.43it/s]
4|1781|Loss: 0.9823104739189148:  28%|██▊       | 1780/6470 [22:05<54:37,  1.43it/s]
4|1781|Loss: 0.9823104739189148:  28%|██▊       | 1781/6470 [22:05<56:23,  1.39it/s]
4|1782|Loss: 0.897168755531311:  28%|██▊       | 1781/6470 [22:06<56:23,  1.39it/s] 
4|1782|Loss: 0.897168755531311:  28%|██▊       | 1782/6470 [22:06<55:49,  1.40it/s]
4|1783|Loss: 0.760424792766571:  28%|██▊       | 1782/6470 [22:07<55:49,  1.40it/s]
4|1783|Loss: 0.760424792766571:  28%|██▊       | 1783/6470 [22:07<55:19,  1.41it/s]
4|1784|Loss: 0.8862729668617249:  28%|██▊       | 1783/6470 [22:07<55:19,  1.41it/s]
4|1784|Loss: 0.8862729668617249:  28%|██▊       | 1784/6470 [22:08<58:08,  1.34it/s]
4|1785|Loss: 0.834470808506012:  28%|██▊       | 1784/6470 [22:08<58:08,  1.34it/s] 
4|1785|Loss: 0.834470808506012:  28%|██▊       | 1785/6470 [22:08<59:01,  1.32it/s]
4|1786|Loss: 0.6565034985542297:  28%|██▊       | 1785/6470 [22:09<59:01,  1.32it/s]
4|1786|Loss: 0.6565034985542297:  28%|██▊       | 1786/6470 [22:09<56:18,  1.39it/s]
4|1787|Loss: 0.9337120056152344:  28%|██▊       | 1786/6470 [22:10<56:18,  1.39it/s]
4|1787|Loss: 0.9337120056152344:  28%|██▊       | 1787/6470 [22:10<59:21,  1.31it/s]
4|1788|Loss: 0.7856873273849487:  28%|██▊       | 1787/6470 [22:10<59:21,  1.31it/s]
4|1788|Loss: 0.7856873273849487:  28%|██▊       | 1788/6470 [22:11<1:03:02,  1.24it/s]
4|1789|Loss: 0.8687255382537842:  28%|██▊       | 1788/6470 [22:11<1:03:02,  1.24it/s]
4|1789|Loss: 0.8687255382537842:  28%|██▊       | 1789/6470 [22:12<1:03:25,  1.23it/s]
4|1790|Loss: 0.7874463796615601:  28%|██▊       | 1789/6470 [22:12<1:03:25,  1.23it/s]
4|1790|Loss: 0.7874463796615601:  28%|██▊       | 1790/6470 [22:13<1:04:24,  1.21it/s]
4|1791|Loss: 0.8015871047973633:  28%|██▊       | 1790/6470 [22:13<1:04:24,  1.21it/s]
4|1791|Loss: 0.8015871047973633:  28%|██▊       | 1791/6470 [22:13<59:43,  1.31it/s]  
4|1792|Loss: 0.882673442363739:  28%|██▊       | 1791/6470 [22:14<59:43,  1.31it/s] 
4|1792|Loss: 0.882673442363739:  28%|██▊       | 1792/6470 [22:14<56:46,  1.37it/s]
4|1793|Loss: 0.8637519478797913:  28%|██▊       | 1792/6470 [22:14<56:46,  1.37it/s]
4|1793|Loss: 0.8637519478797913:  28%|██▊       | 1793/6470 [22:15<57:26,  1.36it/s]
4|1794|Loss: 0.8007586002349854:  28%|██▊       | 1793/6470 [22:15<57:26,  1.36it/s]
4|1794|Loss: 0.8007586002349854:  28%|██▊       | 1794/6470 [22:15<54:00,  1.44it/s]
4|1795|Loss: 0.8657217025756836:  28%|██▊       | 1794/6470 [22:16<54:00,  1.44it/s]
4|1795|Loss: 0.8657217025756836:  28%|██▊       | 1795/6470 [22:16<50:39,  1.54it/s]
4|1796|Loss: 0.9122822284698486:  28%|██▊       | 1795/6470 [22:16<50:39,  1.54it/s]
4|1796|Loss: 0.9122822284698486:  28%|██▊       | 1796/6470 [22:16<52:08,  1.49it/s]
4|1797|Loss: 0.9195377230644226:  28%|██▊       | 1796/6470 [22:17<52:08,  1.49it/s]
4|1797|Loss: 0.9195377230644226:  28%|██▊       | 1797/6470 [22:17<54:08,  1.44it/s]
4|1798|Loss: 0.7725081443786621:  28%|██▊       | 1797/6470 [22:18<54:08,  1.44it/s]
4|1798|Loss: 0.7725081443786621:  28%|██▊       | 1798/6470 [22:18<50:25,  1.54it/s]
4|1799|Loss: 0.8965702056884766:  28%|██▊       | 1798/6470 [22:18<50:25,  1.54it/s]
4|1799|Loss: 0.8965702056884766:  28%|██▊       | 1799/6470 [22:18<52:07,  1.49it/s]
4|1800|Loss: 0.7198194265365601:  28%|██▊       | 1799/6470 [22:19<52:07,  1.49it/s]
4|1800|Loss: 0.7198194265365601:  28%|██▊       | 1800/6470 [22:19<53:32,  1.45it/s]
4|1801|Loss: 0.9215729832649231:  28%|██▊       | 1800/6470 [22:20<53:32,  1.45it/s]
4|1801|Loss: 0.9215729832649231:  28%|██▊       | 1801/6470 [22:20<56:01,  1.39it/s]
4|1802|Loss: 1.0087453126907349:  28%|██▊       | 1801/6470 [22:20<56:01,  1.39it/s]
4|1802|Loss: 1.0087453126907349:  28%|██▊       | 1802/6470 [22:21<56:07,  1.39it/s]
4|1803|Loss: 0.7937755584716797:  28%|██▊       | 1802/6470 [22:21<56:07,  1.39it/s]
4|1803|Loss: 0.7937755584716797:  28%|██▊       | 1803/6470 [22:22<59:39,  1.30it/s]
4|1804|Loss: 0.8399311304092407:  28%|██▊       | 1803/6470 [22:22<59:39,  1.30it/s]
4|1804|Loss: 0.8399311304092407:  28%|██▊       | 1804/6470 [22:22<55:06,  1.41it/s]
4|1805|Loss: 0.7299519181251526:  28%|██▊       | 1804/6470 [22:22<55:06,  1.41it/s]
4|1805|Loss: 0.7299519181251526:  28%|██▊       | 1805/6470 [22:23<49:43,  1.56it/s]
4|1806|Loss: 0.956695556640625:  28%|██▊       | 1805/6470 [22:23<49:43,  1.56it/s] 
4|1806|Loss: 0.956695556640625:  28%|██▊       | 1806/6470 [22:23<52:02,  1.49it/s]
4|1807|Loss: 0.7346449494361877:  28%|██▊       | 1806/6470 [22:24<52:02,  1.49it/s]
4|1807|Loss: 0.7346449494361877:  28%|██▊       | 1807/6470 [22:24<54:48,  1.42it/s]
4|1808|Loss: 0.7981534600257874:  28%|██▊       | 1807/6470 [22:25<54:48,  1.42it/s]
4|1808|Loss: 0.7981534600257874:  28%|██▊       | 1808/6470 [22:25<56:02,  1.39it/s]
4|1809|Loss: 0.8473464846611023:  28%|██▊       | 1808/6470 [22:25<56:02,  1.39it/s]
4|1809|Loss: 0.8473464846611023:  28%|██▊       | 1809/6470 [22:26<55:44,  1.39it/s]
4|1810|Loss: 0.8362085819244385:  28%|██▊       | 1809/6470 [22:26<55:44,  1.39it/s]
4|1810|Loss: 0.8362085819244385:  28%|██▊       | 1810/6470 [22:26<54:42,  1.42it/s]
4|1811|Loss: 0.9411799311637878:  28%|██▊       | 1810/6470 [22:27<54:42,  1.42it/s]
4|1811|Loss: 0.9411799311637878:  28%|██▊       | 1811/6470 [22:27<55:31,  1.40it/s]
4|1812|Loss: 0.75466388463974:  28%|██▊       | 1811/6470 [22:28<55:31,  1.40it/s]  
4|1812|Loss: 0.75466388463974:  28%|██▊       | 1812/6470 [22:28<58:21,  1.33it/s]
4|1813|Loss: 0.9496538639068604:  28%|██▊       | 1812/6470 [22:28<58:21,  1.33it/s]
4|1813|Loss: 0.9496538639068604:  28%|██▊       | 1813/6470 [22:29<57:19,  1.35it/s]
4|1814|Loss: 0.6773547530174255:  28%|██▊       | 1813/6470 [22:29<57:19,  1.35it/s]
4|1814|Loss: 0.6773547530174255:  28%|██▊       | 1814/6470 [22:30<1:02:36,  1.24it/s]
4|1815|Loss: 0.8097416162490845:  28%|██▊       | 1814/6470 [22:30<1:02:36,  1.24it/s]
4|1815|Loss: 0.8097416162490845:  28%|██▊       | 1815/6470 [22:30<1:01:17,  1.27it/s]
4|1816|Loss: 0.8983699679374695:  28%|██▊       | 1815/6470 [22:31<1:01:17,  1.27it/s]
4|1816|Loss: 0.8983699679374695:  28%|██▊       | 1816/6470 [22:31<58:25,  1.33it/s]  
4|1817|Loss: 0.7505391836166382:  28%|██▊       | 1816/6470 [22:31<58:25,  1.33it/s]
4|1817|Loss: 0.7505391836166382:  28%|██▊       | 1817/6470 [22:32<1:00:56,  1.27it/s]
4|1818|Loss: 1.031437635421753:  28%|██▊       | 1817/6470 [22:32<1:00:56,  1.27it/s] 
4|1818|Loss: 1.031437635421753:  28%|██▊       | 1818/6470 [22:33<1:03:46,  1.22it/s]
4|1819|Loss: 0.713217556476593:  28%|██▊       | 1818/6470 [22:33<1:03:46,  1.22it/s]
4|1819|Loss: 0.713217556476593:  28%|██▊       | 1819/6470 [22:34<1:03:30,  1.22it/s]
4|1820|Loss: 0.7491217255592346:  28%|██▊       | 1819/6470 [22:34<1:03:30,  1.22it/s]
4|1820|Loss: 0.7491217255592346:  28%|██▊       | 1820/6470 [22:34<59:36,  1.30it/s]  
4|1821|Loss: 0.8923768997192383:  28%|██▊       | 1820/6470 [22:35<59:36,  1.30it/s]
4|1821|Loss: 0.8923768997192383:  28%|██▊       | 1821/6470 [22:35<56:50,  1.36it/s]
4|1822|Loss: 0.7683119177818298:  28%|██▊       | 1821/6470 [22:35<56:50,  1.36it/s]
4|1822|Loss: 0.7683119177818298:  28%|██▊       | 1822/6470 [22:36<57:04,  1.36it/s]
4|1823|Loss: 0.7999147176742554:  28%|██▊       | 1822/6470 [22:36<57:04,  1.36it/s]
4|1823|Loss: 0.7999147176742554:  28%|██▊       | 1823/6470 [22:36<57:26,  1.35it/s]
4|1824|Loss: 0.7994516491889954:  28%|██▊       | 1823/6470 [22:37<57:26,  1.35it/s]
4|1824|Loss: 0.7994516491889954:  28%|██▊       | 1824/6470 [22:37<58:23,  1.33it/s]
4|1825|Loss: 1.0059177875518799:  28%|██▊       | 1824/6470 [22:38<58:23,  1.33it/s]
4|1825|Loss: 1.0059177875518799:  28%|██▊       | 1825/6470 [22:38<58:22,  1.33it/s]
4|1826|Loss: 0.7898257374763489:  28%|██▊       | 1825/6470 [22:38<58:22,  1.33it/s]
4|1826|Loss: 0.7898257374763489:  28%|██▊       | 1826/6470 [22:39<58:34,  1.32it/s]
4|1827|Loss: 0.8168433308601379:  28%|██▊       | 1826/6470 [22:39<58:34,  1.32it/s]
4|1827|Loss: 0.8168433308601379:  28%|██▊       | 1827/6470 [22:39<57:21,  1.35it/s]
4|1828|Loss: 0.8430385589599609:  28%|██▊       | 1827/6470 [22:40<57:21,  1.35it/s]
4|1828|Loss: 0.8430385589599609:  28%|██▊       | 1828/6470 [22:40<54:20,  1.42it/s]
4|1829|Loss: 0.8854670524597168:  28%|██▊       | 1828/6470 [22:40<54:20,  1.42it/s]
4|1829|Loss: 0.8854670524597168:  28%|██▊       | 1829/6470 [22:41<54:55,  1.41it/s]
4|1830|Loss: 0.9697872996330261:  28%|██▊       | 1829/6470 [22:41<54:55,  1.41it/s]
4|1830|Loss: 0.9697872996330261:  28%|██▊       | 1830/6470 [22:41<56:11,  1.38it/s]
4|1831|Loss: 0.9701718091964722:  28%|██▊       | 1830/6470 [22:42<56:11,  1.38it/s]
4|1831|Loss: 0.9701718091964722:  28%|██▊       | 1831/6470 [22:42<57:14,  1.35it/s]
4|1832|Loss: 0.9375645518302917:  28%|██▊       | 1831/6470 [22:43<57:14,  1.35it/s]
4|1832|Loss: 0.9375645518302917:  28%|██▊       | 1832/6470 [22:43<57:38,  1.34it/s]
4|1833|Loss: 0.6621602773666382:  28%|██▊       | 1832/6470 [22:43<57:38,  1.34it/s]
4|1833|Loss: 0.6621602773666382:  28%|██▊       | 1833/6470 [22:44<1:00:03,  1.29it/s]
4|1834|Loss: 0.7933923006057739:  28%|██▊       | 1833/6470 [22:44<1:00:03,  1.29it/s]
4|1834|Loss: 0.7933923006057739:  28%|██▊       | 1834/6470 [22:45<57:25,  1.35it/s]  
4|1835|Loss: 0.9152134656906128:  28%|██▊       | 1834/6470 [22:45<57:25,  1.35it/s]
4|1835|Loss: 0.9152134656906128:  28%|██▊       | 1835/6470 [22:45<57:47,  1.34it/s]
4|1836|Loss: 0.7250607013702393:  28%|██▊       | 1835/6470 [22:46<57:47,  1.34it/s]
4|1836|Loss: 0.7250607013702393:  28%|██▊       | 1836/6470 [22:46<1:00:05,  1.29it/s]
4|1837|Loss: 0.9425899386405945:  28%|██▊       | 1836/6470 [22:47<1:00:05,  1.29it/s]
4|1837|Loss: 0.9425899386405945:  28%|██▊       | 1837/6470 [22:47<57:47,  1.34it/s]  
4|1838|Loss: 0.7710936665534973:  28%|██▊       | 1837/6470 [22:47<57:47,  1.34it/s]
4|1838|Loss: 0.7710936665534973:  28%|██▊       | 1838/6470 [22:48<57:23,  1.35it/s]
4|1839|Loss: 0.667313277721405:  28%|██▊       | 1838/6470 [22:48<57:23,  1.35it/s] 
4|1839|Loss: 0.667313277721405:  28%|██▊       | 1839/6470 [22:48<56:42,  1.36it/s]
4|1840|Loss: 0.7934530377388:  28%|██▊       | 1839/6470 [22:49<56:42,  1.36it/s]  
4|1840|Loss: 0.7934530377388:  28%|██▊       | 1840/6470 [22:49<55:37,  1.39it/s]
4|1841|Loss: 0.7492510080337524:  28%|██▊       | 1840/6470 [22:49<55:37,  1.39it/s]
4|1841|Loss: 0.7492510080337524:  28%|██▊       | 1841/6470 [22:50<1:01:17,  1.26it/s]
4|1842|Loss: 0.8803473711013794:  28%|██▊       | 1841/6470 [22:50<1:01:17,  1.26it/s]
4|1842|Loss: 0.8803473711013794:  28%|██▊       | 1842/6470 [22:51<1:02:41,  1.23it/s]
4|1843|Loss: 0.960848867893219:  28%|██▊       | 1842/6470 [22:51<1:02:41,  1.23it/s] 
4|1843|Loss: 0.960848867893219:  28%|██▊       | 1843/6470 [22:51<1:00:06,  1.28it/s]
4|1844|Loss: 0.844988226890564:  28%|██▊       | 1843/6470 [22:52<1:00:06,  1.28it/s]
4|1844|Loss: 0.844988226890564:  29%|██▊       | 1844/6470 [22:52<59:00,  1.31it/s]  
4|1845|Loss: 0.7537608742713928:  29%|██▊       | 1844/6470 [22:53<59:00,  1.31it/s]
4|1845|Loss: 0.7537608742713928:  29%|██▊       | 1845/6470 [22:53<55:19,  1.39it/s]
4|1846|Loss: 0.9345216751098633:  29%|██▊       | 1845/6470 [22:53<55:19,  1.39it/s]
4|1846|Loss: 0.9345216751098633:  29%|██▊       | 1846/6470 [22:54<58:11,  1.32it/s]
4|1847|Loss: 0.885629415512085:  29%|██▊       | 1846/6470 [22:54<58:11,  1.32it/s] 
4|1847|Loss: 0.885629415512085:  29%|██▊       | 1847/6470 [22:54<59:44,  1.29it/s]
4|1848|Loss: 0.7949715852737427:  29%|██▊       | 1847/6470 [22:55<59:44,  1.29it/s]
4|1848|Loss: 0.7949715852737427:  29%|██▊       | 1848/6470 [22:55<57:43,  1.33it/s]
4|1849|Loss: 0.7931889295578003:  29%|██▊       | 1848/6470 [22:56<57:43,  1.33it/s]
4|1849|Loss: 0.7931889295578003:  29%|██▊       | 1849/6470 [22:56<57:17,  1.34it/s]
4|1850|Loss: 0.8001697063446045:  29%|██▊       | 1849/6470 [22:56<57:17,  1.34it/s]
4|1850|Loss: 0.8001697063446045:  29%|██▊       | 1850/6470 [22:57<55:04,  1.40it/s]
4|1851|Loss: 0.9804060459136963:  29%|██▊       | 1850/6470 [22:57<55:04,  1.40it/s]
4|1851|Loss: 0.9804060459136963:  29%|██▊       | 1851/6470 [22:57<56:00,  1.37it/s]
4|1852|Loss: 0.6467853784561157:  29%|██▊       | 1851/6470 [22:58<56:00,  1.37it/s]
4|1852|Loss: 0.6467853784561157:  29%|██▊       | 1852/6470 [22:58<57:21,  1.34it/s]
4|1853|Loss: 0.6926017999649048:  29%|██▊       | 1852/6470 [22:58<57:21,  1.34it/s]
4|1853|Loss: 0.6926017999649048:  29%|██▊       | 1853/6470 [22:59<56:35,  1.36it/s]
4|1854|Loss: 0.7962740063667297:  29%|██▊       | 1853/6470 [22:59<56:35,  1.36it/s]
4|1854|Loss: 0.7962740063667297:  29%|██▊       | 1854/6470 [22:59<53:17,  1.44it/s]
4|1855|Loss: 0.7819033861160278:  29%|██▊       | 1854/6470 [23:00<53:17,  1.44it/s]
4|1855|Loss: 0.7819033861160278:  29%|██▊       | 1855/6470 [23:00<49:29,  1.55it/s]
4|1856|Loss: 0.8445579409599304:  29%|██▊       | 1855/6470 [23:00<49:29,  1.55it/s]
4|1856|Loss: 0.8445579409599304:  29%|██▊       | 1856/6470 [23:01<49:02,  1.57it/s]
4|1857|Loss: 0.8631697297096252:  29%|██▊       | 1856/6470 [23:01<49:02,  1.57it/s]
4|1857|Loss: 0.8631697297096252:  29%|██▊       | 1857/6470 [23:01<51:25,  1.50it/s]
4|1858|Loss: 0.8135127425193787:  29%|██▊       | 1857/6470 [23:02<51:25,  1.50it/s]
4|1858|Loss: 0.8135127425193787:  29%|██▊       | 1858/6470 [23:02<53:20,  1.44it/s]
4|1859|Loss: 0.7126370668411255:  29%|██▊       | 1858/6470 [23:02<53:20,  1.44it/s]
4|1859|Loss: 0.7126370668411255:  29%|██▊       | 1859/6470 [23:03<54:53,  1.40it/s]
4|1860|Loss: 0.88924241065979:  29%|██▊       | 1859/6470 [23:03<54:53,  1.40it/s]  
4|1860|Loss: 0.88924241065979:  29%|██▊       | 1860/6470 [23:04<55:41,  1.38it/s]
4|1861|Loss: 0.7017927169799805:  29%|██▊       | 1860/6470 [23:04<55:41,  1.38it/s]
4|1861|Loss: 0.7017927169799805:  29%|██▉       | 1861/6470 [23:04<57:06,  1.35it/s]
4|1862|Loss: 0.7673969864845276:  29%|██▉       | 1861/6470 [23:05<57:06,  1.35it/s]
4|1862|Loss: 0.7673969864845276:  29%|██▉       | 1862/6470 [23:05<56:26,  1.36it/s]
4|1863|Loss: 0.8007610440254211:  29%|██▉       | 1862/6470 [23:06<56:26,  1.36it/s]
4|1863|Loss: 0.8007610440254211:  29%|██▉       | 1863/6470 [23:06<59:39,  1.29it/s]
4|1864|Loss: 0.7441806793212891:  29%|██▉       | 1863/6470 [23:06<59:39,  1.29it/s]
4|1864|Loss: 0.7441806793212891:  29%|██▉       | 1864/6470 [23:07<59:02,  1.30it/s]
4|1865|Loss: 0.8375948071479797:  29%|██▉       | 1864/6470 [23:07<59:02,  1.30it/s]
4|1865|Loss: 0.8375948071479797:  29%|██▉       | 1865/6470 [23:07<58:38,  1.31it/s]
4|1866|Loss: 0.755462646484375:  29%|██▉       | 1865/6470 [23:08<58:38,  1.31it/s] 
4|1866|Loss: 0.755462646484375:  29%|██▉       | 1866/6470 [23:08<1:04:39,  1.19it/s]
4|1867|Loss: 0.7561100125312805:  29%|██▉       | 1866/6470 [23:09<1:04:39,  1.19it/s]
4|1867|Loss: 0.7561100125312805:  29%|██▉       | 1867/6470 [23:09<1:03:06,  1.22it/s]
4|1868|Loss: 0.7966798543930054:  29%|██▉       | 1867/6470 [23:10<1:03:06,  1.22it/s]
4|1868|Loss: 0.7966798543930054:  29%|██▉       | 1868/6470 [23:10<1:02:16,  1.23it/s]
4|1869|Loss: 0.6911371350288391:  29%|██▉       | 1868/6470 [23:10<1:02:16,  1.23it/s]
4|1869|Loss: 0.6911371350288391:  29%|██▉       | 1869/6470 [23:11<57:10,  1.34it/s]  
4|1870|Loss: 0.800862729549408:  29%|██▉       | 1869/6470 [23:11<57:10,  1.34it/s] 
4|1870|Loss: 0.800862729549408:  29%|██▉       | 1870/6470 [23:11<57:28,  1.33it/s]
4|1871|Loss: 0.7680225372314453:  29%|██▉       | 1870/6470 [23:12<57:28,  1.33it/s]
4|1871|Loss: 0.7680225372314453:  29%|██▉       | 1871/6470 [23:12<54:52,  1.40it/s]
4|1872|Loss: 0.8638864159584045:  29%|██▉       | 1871/6470 [23:12<54:52,  1.40it/s]
4|1872|Loss: 0.8638864159584045:  29%|██▉       | 1872/6470 [23:13<55:49,  1.37it/s]
4|1873|Loss: 0.7839495539665222:  29%|██▉       | 1872/6470 [23:13<55:49,  1.37it/s]
4|1873|Loss: 0.7839495539665222:  29%|██▉       | 1873/6470 [23:14<57:06,  1.34it/s]
4|1874|Loss: 0.7393567562103271:  29%|██▉       | 1873/6470 [23:14<57:06,  1.34it/s]
4|1874|Loss: 0.7393567562103271:  29%|██▉       | 1874/6470 [23:14<1:00:26,  1.27it/s]
4|1875|Loss: 0.6724263429641724:  29%|██▉       | 1874/6470 [23:15<1:00:26,  1.27it/s]
4|1875|Loss: 0.6724263429641724:  29%|██▉       | 1875/6470 [23:15<57:45,  1.33it/s]  
4|1876|Loss: 0.7268975973129272:  29%|██▉       | 1875/6470 [23:16<57:45,  1.33it/s]
4|1876|Loss: 0.7268975973129272:  29%|██▉       | 1876/6470 [23:16<57:28,  1.33it/s]
4|1877|Loss: 0.8790029287338257:  29%|██▉       | 1876/6470 [23:16<57:28,  1.33it/s]
4|1877|Loss: 0.8790029287338257:  29%|██▉       | 1877/6470 [23:17<58:04,  1.32it/s]
4|1878|Loss: 0.8436223864555359:  29%|██▉       | 1877/6470 [23:17<58:04,  1.32it/s]
4|1878|Loss: 0.8436223864555359:  29%|██▉       | 1878/6470 [23:17<58:44,  1.30it/s]
4|1879|Loss: 0.7796676158905029:  29%|██▉       | 1878/6470 [23:18<58:44,  1.30it/s]
4|1879|Loss: 0.7796676158905029:  29%|██▉       | 1879/6470 [23:18<59:06,  1.29it/s]
4|1880|Loss: 0.8185944557189941:  29%|██▉       | 1879/6470 [23:19<59:06,  1.29it/s]
4|1880|Loss: 0.8185944557189941:  29%|██▉       | 1880/6470 [23:19<1:01:27,  1.24it/s]
4|1881|Loss: 0.7892117500305176:  29%|██▉       | 1880/6470 [23:19<1:01:27,  1.24it/s]
4|1881|Loss: 0.7892117500305176:  29%|██▉       | 1881/6470 [23:20<57:56,  1.32it/s]  
4|1882|Loss: 0.8186734914779663:  29%|██▉       | 1881/6470 [23:20<57:56,  1.32it/s]
4|1882|Loss: 0.8186734914779663:  29%|██▉       | 1882/6470 [23:20<53:32,  1.43it/s]
4|1883|Loss: 0.9551906585693359:  29%|██▉       | 1882/6470 [23:21<53:32,  1.43it/s]
4|1883|Loss: 0.9551906585693359:  29%|██▉       | 1883/6470 [23:21<53:57,  1.42it/s]
4|1884|Loss: 0.8048966526985168:  29%|██▉       | 1883/6470 [23:21<53:57,  1.42it/s]
4|1884|Loss: 0.8048966526985168:  29%|██▉       | 1884/6470 [23:22<55:27,  1.38it/s]
4|1885|Loss: 0.6891707181930542:  29%|██▉       | 1884/6470 [23:22<55:27,  1.38it/s]
4|1885|Loss: 0.6891707181930542:  29%|██▉       | 1885/6470 [23:23<56:51,  1.34it/s]
4|1886|Loss: 0.6264324188232422:  29%|██▉       | 1885/6470 [23:23<56:51,  1.34it/s]
4|1886|Loss: 0.6264324188232422:  29%|██▉       | 1886/6470 [23:23<54:33,  1.40it/s]
4|1887|Loss: 0.8873085975646973:  29%|██▉       | 1886/6470 [23:24<54:33,  1.40it/s]
4|1887|Loss: 0.8873085975646973:  29%|██▉       | 1887/6470 [23:24<56:57,  1.34it/s]
4|1888|Loss: 0.8938327431678772:  29%|██▉       | 1887/6470 [23:24<56:57,  1.34it/s]
4|1888|Loss: 0.8938327431678772:  29%|██▉       | 1888/6470 [23:25<56:21,  1.36it/s]
4|1889|Loss: 0.9428045749664307:  29%|██▉       | 1888/6470 [23:25<56:21,  1.36it/s]
4|1889|Loss: 0.9428045749664307:  29%|██▉       | 1889/6470 [23:26<57:22,  1.33it/s]
4|1890|Loss: 0.7685770988464355:  29%|██▉       | 1889/6470 [23:26<57:22,  1.33it/s]
4|1890|Loss: 0.7685770988464355:  29%|██▉       | 1890/6470 [23:26<58:14,  1.31it/s]
4|1891|Loss: 0.7337886691093445:  29%|██▉       | 1890/6470 [23:27<58:14,  1.31it/s]
4|1891|Loss: 0.7337886691093445:  29%|██▉       | 1891/6470 [23:27<56:38,  1.35it/s]
4|1892|Loss: 0.8866985440254211:  29%|██▉       | 1891/6470 [23:27<56:38,  1.35it/s]
4|1892|Loss: 0.8866985440254211:  29%|██▉       | 1892/6470 [23:28<57:02,  1.34it/s]
4|1893|Loss: 0.8772034645080566:  29%|██▉       | 1892/6470 [23:28<57:02,  1.34it/s]
4|1893|Loss: 0.8772034645080566:  29%|██▉       | 1893/6470 [23:28<50:46,  1.50it/s]
4|1894|Loss: 0.7748043537139893:  29%|██▉       | 1893/6470 [23:29<50:46,  1.50it/s]
4|1894|Loss: 0.7748043537139893:  29%|██▉       | 1894/6470 [23:29<52:11,  1.46it/s]
4|1895|Loss: 0.8305273652076721:  29%|██▉       | 1894/6470 [23:29<52:11,  1.46it/s]
4|1895|Loss: 0.8305273652076721:  29%|██▉       | 1895/6470 [23:30<56:54,  1.34it/s]
4|1896|Loss: 0.8766133785247803:  29%|██▉       | 1895/6470 [23:30<56:54,  1.34it/s]
4|1896|Loss: 0.8766133785247803:  29%|██▉       | 1896/6470 [23:31<58:17,  1.31it/s]
4|1897|Loss: 0.8405258059501648:  29%|██▉       | 1896/6470 [23:31<58:17,  1.31it/s]
4|1897|Loss: 0.8405258059501648:  29%|██▉       | 1897/6470 [23:31<58:41,  1.30it/s]
4|1898|Loss: 1.1645939350128174:  29%|██▉       | 1897/6470 [23:32<58:41,  1.30it/s]
4|1898|Loss: 1.1645939350128174:  29%|██▉       | 1898/6470 [23:32<56:01,  1.36it/s]
4|1899|Loss: 0.8433823585510254:  29%|██▉       | 1898/6470 [23:32<56:01,  1.36it/s]
4|1899|Loss: 0.8433823585510254:  29%|██▉       | 1899/6470 [23:33<52:14,  1.46it/s]
4|1900|Loss: 0.773382306098938:  29%|██▉       | 1899/6470 [23:33<52:14,  1.46it/s] 
4|1900|Loss: 0.773382306098938:  29%|██▉       | 1900/6470 [23:33<50:14,  1.52it/s]
4|1901|Loss: 0.7117952704429626:  29%|██▉       | 1900/6470 [23:34<50:14,  1.52it/s]
4|1901|Loss: 0.7117952704429626:  29%|██▉       | 1901/6470 [23:34<54:31,  1.40it/s]
4|1902|Loss: 0.867965579032898:  29%|██▉       | 1901/6470 [23:35<54:31,  1.40it/s] 
4|1902|Loss: 0.867965579032898:  29%|██▉       | 1902/6470 [23:35<55:00,  1.38it/s]
4|1903|Loss: 0.7876233458518982:  29%|██▉       | 1902/6470 [23:35<55:00,  1.38it/s]
4|1903|Loss: 0.7876233458518982:  29%|██▉       | 1903/6470 [23:36<57:32,  1.32it/s]
4|1904|Loss: 0.9676950573921204:  29%|██▉       | 1903/6470 [23:36<57:32,  1.32it/s]
4|1904|Loss: 0.9676950573921204:  29%|██▉       | 1904/6470 [23:36<56:39,  1.34it/s]
4|1905|Loss: 0.843024492263794:  29%|██▉       | 1904/6470 [23:37<56:39,  1.34it/s] 
4|1905|Loss: 0.843024492263794:  29%|██▉       | 1905/6470 [23:37<59:47,  1.27it/s]
4|1906|Loss: 0.7492303252220154:  29%|██▉       | 1905/6470 [23:38<59:47,  1.27it/s]
4|1906|Loss: 0.7492303252220154:  29%|██▉       | 1906/6470 [23:38<59:02,  1.29it/s]
4|1907|Loss: 0.7570546865463257:  29%|██▉       | 1906/6470 [23:39<59:02,  1.29it/s]
4|1907|Loss: 0.7570546865463257:  29%|██▉       | 1907/6470 [23:39<1:00:27,  1.26it/s]
4|1908|Loss: 0.7376127243041992:  29%|██▉       | 1907/6470 [23:39<1:00:27,  1.26it/s]
4|1908|Loss: 0.7376127243041992:  29%|██▉       | 1908/6470 [23:40<1:00:12,  1.26it/s]
4|1909|Loss: 0.8850108981132507:  29%|██▉       | 1908/6470 [23:40<1:00:12,  1.26it/s]
4|1909|Loss: 0.8850108981132507:  30%|██▉       | 1909/6470 [23:41<1:02:35,  1.21it/s]
4|1910|Loss: 1.044002652168274:  30%|██▉       | 1909/6470 [23:41<1:02:35,  1.21it/s] 
4|1910|Loss: 1.044002652168274:  30%|██▉       | 1910/6470 [23:41<1:02:39,  1.21it/s]
4|1911|Loss: 0.8312298059463501:  30%|██▉       | 1910/6470 [23:42<1:02:39,  1.21it/s]
4|1911|Loss: 0.8312298059463501:  30%|██▉       | 1911/6470 [23:42<1:03:07,  1.20it/s]
4|1912|Loss: 0.9623613953590393:  30%|██▉       | 1911/6470 [23:43<1:03:07,  1.20it/s]
4|1912|Loss: 0.9623613953590393:  30%|██▉       | 1912/6470 [23:43<1:04:37,  1.18it/s]
4|1913|Loss: 0.7800191640853882:  30%|██▉       | 1912/6470 [23:44<1:04:37,  1.18it/s]
4|1913|Loss: 0.7800191640853882:  30%|██▉       | 1913/6470 [23:44<59:29,  1.28it/s]  
4|1914|Loss: 0.811485767364502:  30%|██▉       | 1913/6470 [23:44<59:29,  1.28it/s] 
4|1914|Loss: 0.811485767364502:  30%|██▉       | 1914/6470 [23:44<55:24,  1.37it/s]
4|1915|Loss: 0.865536093711853:  30%|██▉       | 1914/6470 [23:45<55:24,  1.37it/s]
4|1915|Loss: 0.865536093711853:  30%|██▉       | 1915/6470 [23:45<51:00,  1.49it/s]
4|1916|Loss: 0.6092542409896851:  30%|██▉       | 1915/6470 [23:45<51:00,  1.49it/s]
4|1916|Loss: 0.6092542409896851:  30%|██▉       | 1916/6470 [23:46<49:07,  1.55it/s]
4|1917|Loss: 0.7783057689666748:  30%|██▉       | 1916/6470 [23:46<49:07,  1.55it/s]
4|1917|Loss: 0.7783057689666748:  30%|██▉       | 1917/6470 [23:46<48:18,  1.57it/s]
4|1918|Loss: 0.7466763257980347:  30%|██▉       | 1917/6470 [23:47<48:18,  1.57it/s]
4|1918|Loss: 0.7466763257980347:  30%|██▉       | 1918/6470 [23:47<53:43,  1.41it/s]
4|1919|Loss: 0.9397940039634705:  30%|██▉       | 1918/6470 [23:47<53:43,  1.41it/s]
4|1919|Loss: 0.9397940039634705:  30%|██▉       | 1919/6470 [23:48<52:16,  1.45it/s]
4|1920|Loss: 0.8873058557510376:  30%|██▉       | 1919/6470 [23:48<52:16,  1.45it/s]
4|1920|Loss: 0.8873058557510376:  30%|██▉       | 1920/6470 [23:48<51:03,  1.49it/s]
4|1921|Loss: 0.8480095863342285:  30%|██▉       | 1920/6470 [23:49<51:03,  1.49it/s]
4|1921|Loss: 0.8480095863342285:  30%|██▉       | 1921/6470 [23:49<53:34,  1.42it/s]
4|1922|Loss: 0.8835448026657104:  30%|██▉       | 1921/6470 [23:49<53:34,  1.42it/s]
4|1922|Loss: 0.8835448026657104:  30%|██▉       | 1922/6470 [23:50<53:33,  1.42it/s]
4|1923|Loss: 0.9107846617698669:  30%|██▉       | 1922/6470 [23:50<53:33,  1.42it/s]
4|1923|Loss: 0.9107846617698669:  30%|██▉       | 1923/6470 [23:51<54:21,  1.39it/s]
4|1924|Loss: 0.9368690252304077:  30%|██▉       | 1923/6470 [23:51<54:21,  1.39it/s]
4|1924|Loss: 0.9368690252304077:  30%|██▉       | 1924/6470 [23:51<54:14,  1.40it/s]
4|1925|Loss: 0.8593318462371826:  30%|██▉       | 1924/6470 [23:52<54:14,  1.40it/s]
4|1925|Loss: 0.8593318462371826:  30%|██▉       | 1925/6470 [23:52<49:53,  1.52it/s]
4|1926|Loss: 0.9385225772857666:  30%|██▉       | 1925/6470 [23:52<49:53,  1.52it/s]
4|1926|Loss: 0.9385225772857666:  30%|██▉       | 1926/6470 [23:52<49:04,  1.54it/s]
4|1927|Loss: 0.8271985650062561:  30%|██▉       | 1926/6470 [23:53<49:04,  1.54it/s]
4|1927|Loss: 0.8271985650062561:  30%|██▉       | 1927/6470 [23:53<50:16,  1.51it/s]
4|1928|Loss: 1.01598060131073:  30%|██▉       | 1927/6470 [23:54<50:16,  1.51it/s]  
4|1928|Loss: 1.01598060131073:  30%|██▉       | 1928/6470 [23:54<55:00,  1.38it/s]
4|1929|Loss: 0.7448065876960754:  30%|██▉       | 1928/6470 [23:54<55:00,  1.38it/s]
4|1929|Loss: 0.7448065876960754:  30%|██▉       | 1929/6470 [23:55<54:42,  1.38it/s]
4|1930|Loss: 0.9340773224830627:  30%|██▉       | 1929/6470 [23:55<54:42,  1.38it/s]
4|1930|Loss: 0.9340773224830627:  30%|██▉       | 1930/6470 [23:56<59:35,  1.27it/s]
4|1931|Loss: 0.8853336572647095:  30%|██▉       | 1930/6470 [23:56<59:35,  1.27it/s]
4|1931|Loss: 0.8853336572647095:  30%|██▉       | 1931/6470 [23:56<56:09,  1.35it/s]
4|1932|Loss: 0.9158622026443481:  30%|██▉       | 1931/6470 [23:57<56:09,  1.35it/s]
4|1932|Loss: 0.9158622026443481:  30%|██▉       | 1932/6470 [23:57<56:22,  1.34it/s]
4|1933|Loss: 0.8843656182289124:  30%|██▉       | 1932/6470 [23:57<56:22,  1.34it/s]
4|1933|Loss: 0.8843656182289124:  30%|██▉       | 1933/6470 [23:58<57:28,  1.32it/s]
4|1934|Loss: 0.9191659092903137:  30%|██▉       | 1933/6470 [23:58<57:28,  1.32it/s]
4|1934|Loss: 0.9191659092903137:  30%|██▉       | 1934/6470 [23:58<55:24,  1.36it/s]
4|1935|Loss: 0.8205408453941345:  30%|██▉       | 1934/6470 [23:59<55:24,  1.36it/s]
4|1935|Loss: 0.8205408453941345:  30%|██▉       | 1935/6470 [23:59<55:45,  1.36it/s]
4|1936|Loss: 0.7736664414405823:  30%|██▉       | 1935/6470 [24:00<55:45,  1.36it/s]
4|1936|Loss: 0.7736664414405823:  30%|██▉       | 1936/6470 [24:00<53:18,  1.42it/s]
4|1937|Loss: 0.7915468811988831:  30%|██▉       | 1936/6470 [24:00<53:18,  1.42it/s]
4|1937|Loss: 0.7915468811988831:  30%|██▉       | 1937/6470 [24:01<53:37,  1.41it/s]
4|1938|Loss: 0.8277097940444946:  30%|██▉       | 1937/6470 [24:01<53:37,  1.41it/s]
4|1938|Loss: 0.8277097940444946:  30%|██▉       | 1938/6470 [24:01<52:21,  1.44it/s]
4|1939|Loss: 0.9687834978103638:  30%|██▉       | 1938/6470 [24:02<52:21,  1.44it/s]
4|1939|Loss: 0.9687834978103638:  30%|██▉       | 1939/6470 [24:02<55:05,  1.37it/s]
4|1940|Loss: 0.6784160137176514:  30%|██▉       | 1939/6470 [24:03<55:05,  1.37it/s]
4|1940|Loss: 0.6784160137176514:  30%|██▉       | 1940/6470 [24:03<1:00:27,  1.25it/s]
4|1941|Loss: 0.7640228867530823:  30%|██▉       | 1940/6470 [24:03<1:00:27,  1.25it/s]
4|1941|Loss: 0.7640228867530823:  30%|███       | 1941/6470 [24:04<59:44,  1.26it/s]  
4|1942|Loss: 0.9523054957389832:  30%|███       | 1941/6470 [24:04<59:44,  1.26it/s]
4|1942|Loss: 0.9523054957389832:  30%|███       | 1942/6470 [24:04<56:49,  1.33it/s]
4|1943|Loss: 0.8724360466003418:  30%|███       | 1942/6470 [24:05<56:49,  1.33it/s]
4|1943|Loss: 0.8724360466003418:  30%|███       | 1943/6470 [24:05<59:33,  1.27it/s]
4|1944|Loss: 0.5895971655845642:  30%|███       | 1943/6470 [24:06<59:33,  1.27it/s]
4|1944|Loss: 0.5895971655845642:  30%|███       | 1944/6470 [24:06<1:03:36,  1.19it/s]
4|1945|Loss: 0.9252298474311829:  30%|███       | 1944/6470 [24:07<1:03:36,  1.19it/s]
4|1945|Loss: 0.9252298474311829:  30%|███       | 1945/6470 [24:07<1:01:56,  1.22it/s]
4|1946|Loss: 1.0064092874526978:  30%|███       | 1945/6470 [24:07<1:01:56,  1.22it/s]
4|1946|Loss: 1.0064092874526978:  30%|███       | 1946/6470 [24:08<59:13,  1.27it/s]  
4|1947|Loss: 0.7349417805671692:  30%|███       | 1946/6470 [24:08<59:13,  1.27it/s]
4|1947|Loss: 0.7349417805671692:  30%|███       | 1947/6470 [24:09<59:16,  1.27it/s]
4|1948|Loss: 0.7630777955055237:  30%|███       | 1947/6470 [24:09<59:16,  1.27it/s]
4|1948|Loss: 0.7630777955055237:  30%|███       | 1948/6470 [24:09<58:59,  1.28it/s]
4|1949|Loss: 0.8237141370773315:  30%|███       | 1948/6470 [24:10<58:59,  1.28it/s]
4|1949|Loss: 0.8237141370773315:  30%|███       | 1949/6470 [24:10<58:47,  1.28it/s]
4|1950|Loss: 0.849601149559021:  30%|███       | 1949/6470 [24:10<58:47,  1.28it/s] 
4|1950|Loss: 0.849601149559021:  30%|███       | 1950/6470 [24:11<57:11,  1.32it/s]
4|1951|Loss: 0.87232905626297:  30%|███       | 1950/6470 [24:11<57:11,  1.32it/s] 
4|1951|Loss: 0.87232905626297:  30%|███       | 1951/6470 [24:11<55:48,  1.35it/s]
4|1952|Loss: 0.8135247230529785:  30%|███       | 1951/6470 [24:12<55:48,  1.35it/s]
4|1952|Loss: 0.8135247230529785:  30%|███       | 1952/6470 [24:12<56:56,  1.32it/s]
4|1953|Loss: 0.923675000667572:  30%|███       | 1952/6470 [24:13<56:56,  1.32it/s] 
4|1953|Loss: 0.923675000667572:  30%|███       | 1953/6470 [24:13<1:00:18,  1.25it/s]
4|1954|Loss: 0.7669087052345276:  30%|███       | 1953/6470 [24:14<1:00:18,  1.25it/s]
4|1954|Loss: 0.7669087052345276:  30%|███       | 1954/6470 [24:14<1:01:00,  1.23it/s]
4|1955|Loss: 0.905433714389801:  30%|███       | 1954/6470 [24:14<1:01:00,  1.23it/s] 
4|1955|Loss: 0.905433714389801:  30%|███       | 1955/6470 [24:15<1:01:40,  1.22it/s]
4|1956|Loss: 0.7733878493309021:  30%|███       | 1955/6470 [24:15<1:01:40,  1.22it/s]
4|1956|Loss: 0.7733878493309021:  30%|███       | 1956/6470 [24:16<59:38,  1.26it/s]  
4|1957|Loss: 0.8100019693374634:  30%|███       | 1956/6470 [24:16<59:38,  1.26it/s]
4|1957|Loss: 0.8100019693374634:  30%|███       | 1957/6470 [24:16<58:42,  1.28it/s]
4|1958|Loss: 0.7586679458618164:  30%|███       | 1957/6470 [24:17<58:42,  1.28it/s]
4|1958|Loss: 0.7586679458618164:  30%|███       | 1958/6470 [24:17<1:00:25,  1.24it/s]
4|1959|Loss: 0.9647476077079773:  30%|███       | 1958/6470 [24:18<1:00:25,  1.24it/s]
4|1959|Loss: 0.9647476077079773:  30%|███       | 1959/6470 [24:18<59:45,  1.26it/s]  
4|1960|Loss: 1.0065793991088867:  30%|███       | 1959/6470 [24:18<59:45,  1.26it/s]
4|1960|Loss: 1.0065793991088867:  30%|███       | 1960/6470 [24:19<59:27,  1.26it/s]
4|1961|Loss: 0.802658200263977:  30%|███       | 1960/6470 [24:19<59:27,  1.26it/s] 
4|1961|Loss: 0.802658200263977:  30%|███       | 1961/6470 [24:19<57:21,  1.31it/s]
4|1962|Loss: 0.9677535891532898:  30%|███       | 1961/6470 [24:20<57:21,  1.31it/s]
4|1962|Loss: 0.9677535891532898:  30%|███       | 1962/6470 [24:20<56:29,  1.33it/s]
4|1963|Loss: 0.9091021418571472:  30%|███       | 1962/6470 [24:21<56:29,  1.33it/s]
4|1963|Loss: 0.9091021418571472:  30%|███       | 1963/6470 [24:21<56:26,  1.33it/s]
4|1964|Loss: 0.8191503882408142:  30%|███       | 1963/6470 [24:21<56:26,  1.33it/s]
4|1964|Loss: 0.8191503882408142:  30%|███       | 1964/6470 [24:22<59:38,  1.26it/s]
4|1965|Loss: 0.839728832244873:  30%|███       | 1964/6470 [24:22<59:38,  1.26it/s] 
4|1965|Loss: 0.839728832244873:  30%|███       | 1965/6470 [24:23<1:03:27,  1.18it/s]
4|1966|Loss: 0.8488351106643677:  30%|███       | 1965/6470 [24:23<1:03:27,  1.18it/s]
4|1966|Loss: 0.8488351106643677:  30%|███       | 1966/6470 [24:23<59:46,  1.26it/s]  
4|1967|Loss: 0.7853499054908752:  30%|███       | 1966/6470 [24:24<59:46,  1.26it/s]
4|1967|Loss: 0.7853499054908752:  30%|███       | 1967/6470 [24:24<56:16,  1.33it/s]
4|1968|Loss: 0.8195019960403442:  30%|███       | 1967/6470 [24:24<56:16,  1.33it/s]
4|1968|Loss: 0.8195019960403442:  30%|███       | 1968/6470 [24:25<54:27,  1.38it/s]
4|1969|Loss: 0.7262147665023804:  30%|███       | 1968/6470 [24:25<54:27,  1.38it/s]
4|1969|Loss: 0.7262147665023804:  30%|███       | 1969/6470 [24:25<49:30,  1.52it/s]
4|1970|Loss: 0.7994058132171631:  30%|███       | 1969/6470 [24:26<49:30,  1.52it/s]
4|1970|Loss: 0.7994058132171631:  30%|███       | 1970/6470 [24:26<50:08,  1.50it/s]
4|1971|Loss: 0.8002476692199707:  30%|███       | 1970/6470 [24:26<50:08,  1.50it/s]
4|1971|Loss: 0.8002476692199707:  30%|███       | 1971/6470 [24:27<51:14,  1.46it/s]
4|1972|Loss: 0.928558349609375:  30%|███       | 1971/6470 [24:27<51:14,  1.46it/s] 
4|1972|Loss: 0.928558349609375:  30%|███       | 1972/6470 [24:27<51:08,  1.47it/s]
4|1973|Loss: 0.823633074760437:  30%|███       | 1972/6470 [24:28<51:08,  1.47it/s]
4|1973|Loss: 0.823633074760437:  30%|███       | 1973/6470 [24:28<53:26,  1.40it/s]
4|1974|Loss: 0.9924452304840088:  30%|███       | 1973/6470 [24:29<53:26,  1.40it/s]
4|1974|Loss: 0.9924452304840088:  31%|███       | 1974/6470 [24:29<56:30,  1.33it/s]
4|1975|Loss: 0.9405999779701233:  31%|███       | 1974/6470 [24:29<56:30,  1.33it/s]
4|1975|Loss: 0.9405999779701233:  31%|███       | 1975/6470 [24:30<56:11,  1.33it/s]
4|1976|Loss: 0.8228442072868347:  31%|███       | 1975/6470 [24:30<56:11,  1.33it/s]
4|1976|Loss: 0.8228442072868347:  31%|███       | 1976/6470 [24:30<53:43,  1.39it/s]
4|1977|Loss: 0.852844774723053:  31%|███       | 1976/6470 [24:31<53:43,  1.39it/s] 
4|1977|Loss: 0.852844774723053:  31%|███       | 1977/6470 [24:31<52:20,  1.43it/s]
4|1978|Loss: 0.7897956967353821:  31%|███       | 1977/6470 [24:32<52:20,  1.43it/s]
4|1978|Loss: 0.7897956967353821:  31%|███       | 1978/6470 [24:32<56:20,  1.33it/s]
4|1979|Loss: 0.8773922920227051:  31%|███       | 1978/6470 [24:32<56:20,  1.33it/s]
4|1979|Loss: 0.8773922920227051:  31%|███       | 1979/6470 [24:33<55:45,  1.34it/s]
4|1980|Loss: 0.8819288015365601:  31%|███       | 1979/6470 [24:33<55:45,  1.34it/s]
4|1980|Loss: 0.8819288015365601:  31%|███       | 1980/6470 [24:33<56:34,  1.32it/s]
4|1981|Loss: 0.8138763904571533:  31%|███       | 1980/6470 [24:34<56:34,  1.32it/s]
4|1981|Loss: 0.8138763904571533:  31%|███       | 1981/6470 [24:34<52:44,  1.42it/s]
4|1982|Loss: 0.8118832111358643:  31%|███       | 1981/6470 [24:34<52:44,  1.42it/s]
4|1982|Loss: 0.8118832111358643:  31%|███       | 1982/6470 [24:35<53:04,  1.41it/s]
4|1983|Loss: 0.6641116738319397:  31%|███       | 1982/6470 [24:35<53:04,  1.41it/s]
4|1983|Loss: 0.6641116738319397:  31%|███       | 1983/6470 [24:36<56:50,  1.32it/s]
4|1984|Loss: 0.7180290222167969:  31%|███       | 1983/6470 [24:36<56:50,  1.32it/s]
4|1984|Loss: 0.7180290222167969:  31%|███       | 1984/6470 [24:36<55:52,  1.34it/s]
4|1985|Loss: 0.7583748698234558:  31%|███       | 1984/6470 [24:37<55:52,  1.34it/s]
4|1985|Loss: 0.7583748698234558:  31%|███       | 1985/6470 [24:37<53:26,  1.40it/s]
4|1986|Loss: 0.7908832430839539:  31%|███       | 1985/6470 [24:37<53:26,  1.40it/s]
4|1986|Loss: 0.7908832430839539:  31%|███       | 1986/6470 [24:38<53:05,  1.41it/s]
4|1987|Loss: 0.8755322694778442:  31%|███       | 1986/6470 [24:38<53:05,  1.41it/s]
4|1987|Loss: 0.8755322694778442:  31%|███       | 1987/6470 [24:39<57:08,  1.31it/s]
4|1988|Loss: 0.8585378527641296:  31%|███       | 1987/6470 [24:39<57:08,  1.31it/s]
4|1988|Loss: 0.8585378527641296:  31%|███       | 1988/6470 [24:39<54:32,  1.37it/s]
4|1989|Loss: 0.8853811025619507:  31%|███       | 1988/6470 [24:40<54:32,  1.37it/s]
4|1989|Loss: 0.8853811025619507:  31%|███       | 1989/6470 [24:40<57:49,  1.29it/s]
4|1990|Loss: 0.83219975233078:  31%|███       | 1989/6470 [24:41<57:49,  1.29it/s]  
4|1990|Loss: 0.83219975233078:  31%|███       | 1990/6470 [24:41<57:42,  1.29it/s]
4|1991|Loss: 0.7864075303077698:  31%|███       | 1990/6470 [24:41<57:42,  1.29it/s]
4|1991|Loss: 0.7864075303077698:  31%|███       | 1991/6470 [24:42<56:26,  1.32it/s]
4|1992|Loss: 0.8795180916786194:  31%|███       | 1991/6470 [24:42<56:26,  1.32it/s]
4|1992|Loss: 0.8795180916786194:  31%|███       | 1992/6470 [24:42<56:09,  1.33it/s]
4|1993|Loss: 0.9232504367828369:  31%|███       | 1992/6470 [24:43<56:09,  1.33it/s]
4|1993|Loss: 0.9232504367828369:  31%|███       | 1993/6470 [24:43<52:17,  1.43it/s]
4|1994|Loss: 0.6960883736610413:  31%|███       | 1993/6470 [24:43<52:17,  1.43it/s]
4|1994|Loss: 0.6960883736610413:  31%|███       | 1994/6470 [24:44<55:55,  1.33it/s]
4|1995|Loss: 0.6283543705940247:  31%|███       | 1994/6470 [24:44<55:55,  1.33it/s]
4|1995|Loss: 0.6283543705940247:  31%|███       | 1995/6470 [24:45<56:55,  1.31it/s]
4|1996|Loss: 0.9027230739593506:  31%|███       | 1995/6470 [24:45<56:55,  1.31it/s]
4|1996|Loss: 0.9027230739593506:  31%|███       | 1996/6470 [24:45<55:31,  1.34it/s]
4|1997|Loss: 0.8660346269607544:  31%|███       | 1996/6470 [24:46<55:31,  1.34it/s]
4|1997|Loss: 0.8660346269607544:  31%|███       | 1997/6470 [24:46<54:33,  1.37it/s]
4|1998|Loss: 0.7181031107902527:  31%|███       | 1997/6470 [24:46<54:33,  1.37it/s]
4|1998|Loss: 0.7181031107902527:  31%|███       | 1998/6470 [24:47<54:51,  1.36it/s]
4|1999|Loss: 0.7961620092391968:  31%|███       | 1998/6470 [24:47<54:51,  1.36it/s]
4|1999|Loss: 0.7961620092391968:  31%|███       | 1999/6470 [24:47<53:39,  1.39it/s]
4|2000|Loss: 0.8491620421409607:  31%|███       | 1999/6470 [24:48<53:39,  1.39it/s]
4|2000|Loss: 0.8491620421409607:  31%|███       | 2000/6470 [24:48<58:23,  1.28it/s]
4|2001|Loss: 0.8787484169006348:  31%|███       | 2000/6470 [24:49<58:23,  1.28it/s]
4|2001|Loss: 0.8787484169006348:  31%|███       | 2001/6470 [24:49<59:42,  1.25it/s]
4|2002|Loss: 0.789515495300293:  31%|███       | 2001/6470 [24:50<59:42,  1.25it/s] 
4|2002|Loss: 0.789515495300293:  31%|███       | 2002/6470 [24:50<59:57,  1.24it/s]
4|2003|Loss: 0.9468375444412231:  31%|███       | 2002/6470 [24:50<59:57,  1.24it/s]
4|2003|Loss: 0.9468375444412231:  31%|███       | 2003/6470 [24:51<55:28,  1.34it/s]
4|2004|Loss: 0.7301271557807922:  31%|███       | 2003/6470 [24:51<55:28,  1.34it/s]
4|2004|Loss: 0.7301271557807922:  31%|███       | 2004/6470 [24:52<1:00:00,  1.24it/s]
4|2005|Loss: 0.8662443161010742:  31%|███       | 2004/6470 [24:52<1:00:00,  1.24it/s]
4|2005|Loss: 0.8662443161010742:  31%|███       | 2005/6470 [24:52<58:49,  1.27it/s]  
4|2006|Loss: 0.7454823851585388:  31%|███       | 2005/6470 [24:53<58:49,  1.27it/s]
4|2006|Loss: 0.7454823851585388:  31%|███       | 2006/6470 [24:53<1:00:25,  1.23it/s]
4|2007|Loss: 0.7462120056152344:  31%|███       | 2006/6470 [24:54<1:00:25,  1.23it/s]
4|2007|Loss: 0.7462120056152344:  31%|███       | 2007/6470 [24:54<57:50,  1.29it/s]  
4|2008|Loss: 0.5184406638145447:  31%|███       | 2007/6470 [24:54<57:50,  1.29it/s]
4|2008|Loss: 0.5184406638145447:  31%|███       | 2008/6470 [24:55<1:03:42,  1.17it/s]
4|2009|Loss: 0.7054738998413086:  31%|███       | 2008/6470 [24:55<1:03:42,  1.17it/s]
4|2009|Loss: 0.7054738998413086:  31%|███       | 2009/6470 [24:56<1:05:11,  1.14it/s]
4|2010|Loss: 0.9318962693214417:  31%|███       | 2009/6470 [24:56<1:05:11,  1.14it/s]
4|2010|Loss: 0.9318962693214417:  31%|███       | 2010/6470 [24:56<59:02,  1.26it/s]  
4|2011|Loss: 0.9742519855499268:  31%|███       | 2010/6470 [24:57<59:02,  1.26it/s]
4|2011|Loss: 0.9742519855499268:  31%|███       | 2011/6470 [24:57<59:21,  1.25it/s]
4|2012|Loss: 0.8145232796669006:  31%|███       | 2011/6470 [24:58<59:21,  1.25it/s]
4|2012|Loss: 0.8145232796669006:  31%|███       | 2012/6470 [24:58<55:53,  1.33it/s]
4|2013|Loss: 1.0366296768188477:  31%|███       | 2012/6470 [24:58<55:53,  1.33it/s]
4|2013|Loss: 1.0366296768188477:  31%|███       | 2013/6470 [24:59<53:18,  1.39it/s]
4|2014|Loss: 0.8115473389625549:  31%|███       | 2013/6470 [24:59<53:18,  1.39it/s]
4|2014|Loss: 0.8115473389625549:  31%|███       | 2014/6470 [24:59<51:30,  1.44it/s]
4|2015|Loss: 0.9429175853729248:  31%|███       | 2014/6470 [24:59<51:30,  1.44it/s]
4|2015|Loss: 0.9429175853729248:  31%|███       | 2015/6470 [25:00<48:18,  1.54it/s]
4|2016|Loss: 0.923092246055603:  31%|███       | 2015/6470 [25:00<48:18,  1.54it/s] 
4|2016|Loss: 0.923092246055603:  31%|███       | 2016/6470 [25:01<52:49,  1.41it/s]
4|2017|Loss: 0.9974385499954224:  31%|███       | 2016/6470 [25:01<52:49,  1.41it/s]
4|2017|Loss: 0.9974385499954224:  31%|███       | 2017/6470 [25:01<57:16,  1.30it/s]
4|2018|Loss: 0.8232897520065308:  31%|███       | 2017/6470 [25:02<57:16,  1.30it/s]
4|2018|Loss: 0.8232897520065308:  31%|███       | 2018/6470 [25:02<1:02:38,  1.18it/s]
4|2019|Loss: 0.9468725323677063:  31%|███       | 2018/6470 [25:03<1:02:38,  1.18it/s]
4|2019|Loss: 0.9468725323677063:  31%|███       | 2019/6470 [25:03<59:09,  1.25it/s]  
4|2020|Loss: 0.783587634563446:  31%|███       | 2019/6470 [25:04<59:09,  1.25it/s] 
4|2020|Loss: 0.783587634563446:  31%|███       | 2020/6470 [25:04<59:20,  1.25it/s]
4|2021|Loss: 0.7062343955039978:  31%|███       | 2020/6470 [25:04<59:20,  1.25it/s]
4|2021|Loss: 0.7062343955039978:  31%|███       | 2021/6470 [25:05<58:44,  1.26it/s]
4|2022|Loss: 0.7279635667800903:  31%|███       | 2021/6470 [25:05<58:44,  1.26it/s]
4|2022|Loss: 0.7279635667800903:  31%|███▏      | 2022/6470 [25:06<58:52,  1.26it/s]
4|2023|Loss: 0.8733462691307068:  31%|███▏      | 2022/6470 [25:06<58:52,  1.26it/s]
4|2023|Loss: 0.8733462691307068:  31%|███▏      | 2023/6470 [25:06<54:03,  1.37it/s]
4|2024|Loss: 0.891818106174469:  31%|███▏      | 2023/6470 [25:06<54:03,  1.37it/s] 
4|2024|Loss: 0.891818106174469:  31%|███▏      | 2024/6470 [25:07<53:20,  1.39it/s]
4|2025|Loss: 1.023083209991455:  31%|███▏      | 2024/6470 [25:07<53:20,  1.39it/s]
4|2025|Loss: 1.023083209991455:  31%|███▏      | 2025/6470 [25:07<51:46,  1.43it/s]
4|2026|Loss: 0.8525316119194031:  31%|███▏      | 2025/6470 [25:08<51:46,  1.43it/s]
4|2026|Loss: 0.8525316119194031:  31%|███▏      | 2026/6470 [25:08<52:53,  1.40it/s]
4|2027|Loss: 0.771841824054718:  31%|███▏      | 2026/6470 [25:09<52:53,  1.40it/s] 
4|2027|Loss: 0.771841824054718:  31%|███▏      | 2027/6470 [25:09<53:46,  1.38it/s]
4|2028|Loss: 0.8135948181152344:  31%|███▏      | 2027/6470 [25:09<53:46,  1.38it/s]
4|2028|Loss: 0.8135948181152344:  31%|███▏      | 2028/6470 [25:10<52:56,  1.40it/s]
4|2029|Loss: 0.8565847873687744:  31%|███▏      | 2028/6470 [25:10<52:56,  1.40it/s]
4|2029|Loss: 0.8565847873687744:  31%|███▏      | 2029/6470 [25:11<56:40,  1.31it/s]
4|2030|Loss: 0.8680029511451721:  31%|███▏      | 2029/6470 [25:11<56:40,  1.31it/s]
4|2030|Loss: 0.8680029511451721:  31%|███▏      | 2030/6470 [25:11<57:18,  1.29it/s]
4|2031|Loss: 0.7457388043403625:  31%|███▏      | 2030/6470 [25:12<57:18,  1.29it/s]
4|2031|Loss: 0.7457388043403625:  31%|███▏      | 2031/6470 [25:12<56:56,  1.30it/s]
4|2032|Loss: 0.8044440150260925:  31%|███▏      | 2031/6470 [25:12<56:56,  1.30it/s]
4|2032|Loss: 0.8044440150260925:  31%|███▏      | 2032/6470 [25:13<54:11,  1.36it/s]
4|2033|Loss: 0.7739140391349792:  31%|███▏      | 2032/6470 [25:13<54:11,  1.36it/s]
4|2033|Loss: 0.7739140391349792:  31%|███▏      | 2033/6470 [25:13<50:30,  1.46it/s]
4|2034|Loss: 0.889553964138031:  31%|███▏      | 2033/6470 [25:14<50:30,  1.46it/s] 
4|2034|Loss: 0.889553964138031:  31%|███▏      | 2034/6470 [25:14<52:07,  1.42it/s]
4|2035|Loss: 0.8411655426025391:  31%|███▏      | 2034/6470 [25:14<52:07,  1.42it/s]
4|2035|Loss: 0.8411655426025391:  31%|███▏      | 2035/6470 [25:15<51:20,  1.44it/s]
4|2036|Loss: 0.7858854532241821:  31%|███▏      | 2035/6470 [25:15<51:20,  1.44it/s]
4|2036|Loss: 0.7858854532241821:  31%|███▏      | 2036/6470 [25:16<54:15,  1.36it/s]
4|2037|Loss: 1.059838891029358:  31%|███▏      | 2036/6470 [25:16<54:15,  1.36it/s] 
4|2037|Loss: 1.059838891029358:  31%|███▏      | 2037/6470 [25:16<56:23,  1.31it/s]
4|2038|Loss: 0.8998652696609497:  31%|███▏      | 2037/6470 [25:17<56:23,  1.31it/s]
4|2038|Loss: 0.8998652696609497:  31%|███▏      | 2038/6470 [25:17<50:25,  1.46it/s]
4|2039|Loss: 0.9549004435539246:  31%|███▏      | 2038/6470 [25:17<50:25,  1.46it/s]
4|2039|Loss: 0.9549004435539246:  32%|███▏      | 2039/6470 [25:17<48:06,  1.54it/s]
4|2040|Loss: 0.7828238010406494:  32%|███▏      | 2039/6470 [25:18<48:06,  1.54it/s]
4|2040|Loss: 0.7828238010406494:  32%|███▏      | 2040/6470 [25:18<53:37,  1.38it/s]
4|2041|Loss: 0.8897847533226013:  32%|███▏      | 2040/6470 [25:19<53:37,  1.38it/s]
4|2041|Loss: 0.8897847533226013:  32%|███▏      | 2041/6470 [25:19<56:11,  1.31it/s]
4|2042|Loss: 0.8262911438941956:  32%|███▏      | 2041/6470 [25:20<56:11,  1.31it/s]
4|2042|Loss: 0.8262911438941956:  32%|███▏      | 2042/6470 [25:20<1:00:43,  1.22it/s]
4|2043|Loss: 0.7908398509025574:  32%|███▏      | 2042/6470 [25:21<1:00:43,  1.22it/s]
4|2043|Loss: 0.7908398509025574:  32%|███▏      | 2043/6470 [25:21<59:31,  1.24it/s]  
4|2044|Loss: 0.8636482357978821:  32%|███▏      | 2043/6470 [25:21<59:31,  1.24it/s]
4|2044|Loss: 0.8636482357978821:  32%|███▏      | 2044/6470 [25:22<58:26,  1.26it/s]
4|2045|Loss: 0.7927283048629761:  32%|███▏      | 2044/6470 [25:22<58:26,  1.26it/s]
4|2045|Loss: 0.7927283048629761:  32%|███▏      | 2045/6470 [25:23<1:01:26,  1.20it/s]
4|2046|Loss: 0.8123471736907959:  32%|███▏      | 2045/6470 [25:23<1:01:26,  1.20it/s]
4|2046|Loss: 0.8123471736907959:  32%|███▏      | 2046/6470 [25:23<55:56,  1.32it/s]  
4|2047|Loss: 0.8364515900611877:  32%|███▏      | 2046/6470 [25:24<55:56,  1.32it/s]
4|2047|Loss: 0.8364515900611877:  32%|███▏      | 2047/6470 [25:24<53:23,  1.38it/s]
4|2048|Loss: 0.8715517520904541:  32%|███▏      | 2047/6470 [25:24<53:23,  1.38it/s]
4|2048|Loss: 0.8715517520904541:  32%|███▏      | 2048/6470 [25:25<52:28,  1.40it/s]
4|2049|Loss: 0.7898104190826416:  32%|███▏      | 2048/6470 [25:25<52:28,  1.40it/s]
4|2049|Loss: 0.7898104190826416:  32%|███▏      | 2049/6470 [25:25<52:50,  1.39it/s]
4|2050|Loss: 0.9001818895339966:  32%|███▏      | 2049/6470 [25:26<52:50,  1.39it/s]
4|2050|Loss: 0.9001818895339966:  32%|███▏      | 2050/6470 [25:26<54:07,  1.36it/s]
4|2051|Loss: 0.7876603603363037:  32%|███▏      | 2050/6470 [25:26<54:07,  1.36it/s]
4|2051|Loss: 0.7876603603363037:  32%|███▏      | 2051/6470 [25:27<52:23,  1.41it/s]
4|2052|Loss: 0.9927131533622742:  32%|███▏      | 2051/6470 [25:27<52:23,  1.41it/s]
4|2052|Loss: 0.9927131533622742:  32%|███▏      | 2052/6470 [25:28<54:39,  1.35it/s]
4|2053|Loss: 0.7359604239463806:  32%|███▏      | 2052/6470 [25:28<54:39,  1.35it/s]
4|2053|Loss: 0.7359604239463806:  32%|███▏      | 2053/6470 [25:28<51:35,  1.43it/s]
4|2054|Loss: 0.7634163498878479:  32%|███▏      | 2053/6470 [25:29<51:35,  1.43it/s]
4|2054|Loss: 0.7634163498878479:  32%|███▏      | 2054/6470 [25:29<52:40,  1.40it/s]
4|2055|Loss: 0.6740177273750305:  32%|███▏      | 2054/6470 [25:29<52:40,  1.40it/s]
4|2055|Loss: 0.6740177273750305:  32%|███▏      | 2055/6470 [25:29<47:16,  1.56it/s]
4|2056|Loss: 0.8127111792564392:  32%|███▏      | 2055/6470 [25:30<47:16,  1.56it/s]
4|2056|Loss: 0.8127111792564392:  32%|███▏      | 2056/6470 [25:30<48:59,  1.50it/s]
4|2057|Loss: 0.7607905268669128:  32%|███▏      | 2056/6470 [25:31<48:59,  1.50it/s]
4|2057|Loss: 0.7607905268669128:  32%|███▏      | 2057/6470 [25:31<51:31,  1.43it/s]
4|2058|Loss: 0.8592513799667358:  32%|███▏      | 2057/6470 [25:31<51:31,  1.43it/s]
4|2058|Loss: 0.8592513799667358:  32%|███▏      | 2058/6470 [25:32<51:37,  1.42it/s]
4|2059|Loss: 0.7585176825523376:  32%|███▏      | 2058/6470 [25:32<51:37,  1.42it/s]
4|2059|Loss: 0.7585176825523376:  32%|███▏      | 2059/6470 [25:32<53:27,  1.38it/s]
4|2060|Loss: 0.8861185312271118:  32%|███▏      | 2059/6470 [25:33<53:27,  1.38it/s]
4|2060|Loss: 0.8861185312271118:  32%|███▏      | 2060/6470 [25:33<54:32,  1.35it/s]
4|2061|Loss: 0.6997451186180115:  32%|███▏      | 2060/6470 [25:34<54:32,  1.35it/s]
4|2061|Loss: 0.6997451186180115:  32%|███▏      | 2061/6470 [25:34<53:38,  1.37it/s]
4|2062|Loss: 0.7912450432777405:  32%|███▏      | 2061/6470 [25:34<53:38,  1.37it/s]
4|2062|Loss: 0.7912450432777405:  32%|███▏      | 2062/6470 [25:34<52:12,  1.41it/s]
4|2063|Loss: 0.7074536085128784:  32%|███▏      | 2062/6470 [25:35<52:12,  1.41it/s]
4|2063|Loss: 0.7074536085128784:  32%|███▏      | 2063/6470 [25:35<51:30,  1.43it/s]
4|2064|Loss: 0.7596228122711182:  32%|███▏      | 2063/6470 [25:36<51:30,  1.43it/s]
4|2064|Loss: 0.7596228122711182:  32%|███▏      | 2064/6470 [25:36<55:22,  1.33it/s]
4|2065|Loss: 0.8841187953948975:  32%|███▏      | 2064/6470 [25:36<55:22,  1.33it/s]
4|2065|Loss: 0.8841187953948975:  32%|███▏      | 2065/6470 [25:37<54:46,  1.34it/s]
4|2066|Loss: 0.7821117639541626:  32%|███▏      | 2065/6470 [25:37<54:46,  1.34it/s]
4|2066|Loss: 0.7821117639541626:  32%|███▏      | 2066/6470 [25:38<55:24,  1.32it/s]
4|2067|Loss: 0.6907771825790405:  32%|███▏      | 2066/6470 [25:38<55:24,  1.32it/s]
4|2067|Loss: 0.6907771825790405:  32%|███▏      | 2067/6470 [25:38<56:18,  1.30it/s]
4|2068|Loss: 0.85007244348526:  32%|███▏      | 2067/6470 [25:39<56:18,  1.30it/s]  
4|2068|Loss: 0.85007244348526:  32%|███▏      | 2068/6470 [25:39<56:03,  1.31it/s]
4|2069|Loss: 0.8809168338775635:  32%|███▏      | 2068/6470 [25:39<56:03,  1.31it/s]
4|2069|Loss: 0.8809168338775635:  32%|███▏      | 2069/6470 [25:40<53:29,  1.37it/s]
4|2070|Loss: 0.9377211332321167:  32%|███▏      | 2069/6470 [25:40<53:29,  1.37it/s]
4|2070|Loss: 0.9377211332321167:  32%|███▏      | 2070/6470 [25:40<53:16,  1.38it/s]
4|2071|Loss: 0.7655046582221985:  32%|███▏      | 2070/6470 [25:41<53:16,  1.38it/s]
4|2071|Loss: 0.7655046582221985:  32%|███▏      | 2071/6470 [25:41<57:11,  1.28it/s]
4|2072|Loss: 0.8917945623397827:  32%|███▏      | 2071/6470 [25:42<57:11,  1.28it/s]
4|2072|Loss: 0.8917945623397827:  32%|███▏      | 2072/6470 [25:42<57:38,  1.27it/s]
4|2073|Loss: 0.884896993637085:  32%|███▏      | 2072/6470 [25:43<57:38,  1.27it/s] 
4|2073|Loss: 0.884896993637085:  32%|███▏      | 2073/6470 [25:43<58:51,  1.25it/s]
4|2074|Loss: 0.9363471269607544:  32%|███▏      | 2073/6470 [25:43<58:51,  1.25it/s]
4|2074|Loss: 0.9363471269607544:  32%|███▏      | 2074/6470 [25:44<58:17,  1.26it/s]
4|2075|Loss: 0.7758197784423828:  32%|███▏      | 2074/6470 [25:44<58:17,  1.26it/s]
4|2075|Loss: 0.7758197784423828:  32%|███▏      | 2075/6470 [25:45<59:08,  1.24it/s]
4|2076|Loss: 0.8108106255531311:  32%|███▏      | 2075/6470 [25:45<59:08,  1.24it/s]
4|2076|Loss: 0.8108106255531311:  32%|███▏      | 2076/6470 [25:45<56:57,  1.29it/s]
4|2077|Loss: 0.9486438632011414:  32%|███▏      | 2076/6470 [25:46<56:57,  1.29it/s]
4|2077|Loss: 0.9486438632011414:  32%|███▏      | 2077/6470 [25:46<57:02,  1.28it/s]
4|2078|Loss: 0.7444533109664917:  32%|███▏      | 2077/6470 [25:47<57:02,  1.28it/s]
4|2078|Loss: 0.7444533109664917:  32%|███▏      | 2078/6470 [25:47<55:02,  1.33it/s]
4|2079|Loss: 0.8730195164680481:  32%|███▏      | 2078/6470 [25:47<55:02,  1.33it/s]
4|2079|Loss: 0.8730195164680481:  32%|███▏      | 2079/6470 [25:47<53:05,  1.38it/s]
4|2080|Loss: 0.7626507878303528:  32%|███▏      | 2079/6470 [25:48<53:05,  1.38it/s]
4|2080|Loss: 0.7626507878303528:  32%|███▏      | 2080/6470 [25:48<49:28,  1.48it/s]
4|2081|Loss: 0.8861615061759949:  32%|███▏      | 2080/6470 [25:48<49:28,  1.48it/s]
4|2081|Loss: 0.8861615061759949:  32%|███▏      | 2081/6470 [25:49<47:45,  1.53it/s]
4|2082|Loss: 0.813595712184906:  32%|███▏      | 2081/6470 [25:49<47:45,  1.53it/s] 
4|2082|Loss: 0.813595712184906:  32%|███▏      | 2082/6470 [25:49<47:58,  1.52it/s]
4|2083|Loss: 0.8833441734313965:  32%|███▏      | 2082/6470 [25:50<47:58,  1.52it/s]
4|2083|Loss: 0.8833441734313965:  32%|███▏      | 2083/6470 [25:50<48:48,  1.50it/s]
4|2084|Loss: 0.8889913558959961:  32%|███▏      | 2083/6470 [25:50<48:48,  1.50it/s]
4|2084|Loss: 0.8889913558959961:  32%|███▏      | 2084/6470 [25:51<49:36,  1.47it/s]
4|2085|Loss: 0.8190602660179138:  32%|███▏      | 2084/6470 [25:51<49:36,  1.47it/s]
4|2085|Loss: 0.8190602660179138:  32%|███▏      | 2085/6470 [25:51<51:07,  1.43it/s]
4|2086|Loss: 0.7519723773002625:  32%|███▏      | 2085/6470 [25:52<51:07,  1.43it/s]
4|2086|Loss: 0.7519723773002625:  32%|███▏      | 2086/6470 [25:52<54:07,  1.35it/s]
4|2087|Loss: 0.7691066861152649:  32%|███▏      | 2086/6470 [25:53<54:07,  1.35it/s]
4|2087|Loss: 0.7691066861152649:  32%|███▏      | 2087/6470 [25:53<51:23,  1.42it/s]
4|2088|Loss: 0.6886430978775024:  32%|███▏      | 2087/6470 [25:53<51:23,  1.42it/s]
4|2088|Loss: 0.6886430978775024:  32%|███▏      | 2088/6470 [25:54<49:45,  1.47it/s]
4|2089|Loss: 0.8132844567298889:  32%|███▏      | 2088/6470 [25:54<49:45,  1.47it/s]
4|2089|Loss: 0.8132844567298889:  32%|███▏      | 2089/6470 [25:54<51:58,  1.40it/s]
4|2090|Loss: 0.7769446969032288:  32%|███▏      | 2089/6470 [25:55<51:58,  1.40it/s]
4|2090|Loss: 0.7769446969032288:  32%|███▏      | 2090/6470 [25:55<51:41,  1.41it/s]
4|2091|Loss: 0.7943077087402344:  32%|███▏      | 2090/6470 [25:55<51:41,  1.41it/s]
4|2091|Loss: 0.7943077087402344:  32%|███▏      | 2091/6470 [25:56<54:14,  1.35it/s]
4|2092|Loss: 0.8956174254417419:  32%|███▏      | 2091/6470 [25:56<54:14,  1.35it/s]
4|2092|Loss: 0.8956174254417419:  32%|███▏      | 2092/6470 [25:57<54:13,  1.35it/s]
4|2093|Loss: 0.7549564838409424:  32%|███▏      | 2092/6470 [25:57<54:13,  1.35it/s]
4|2093|Loss: 0.7549564838409424:  32%|███▏      | 2093/6470 [25:57<53:00,  1.38it/s]
4|2094|Loss: 0.7632976770401001:  32%|███▏      | 2093/6470 [25:58<53:00,  1.38it/s]
4|2094|Loss: 0.7632976770401001:  32%|███▏      | 2094/6470 [25:58<53:55,  1.35it/s]
4|2095|Loss: 0.8444712162017822:  32%|███▏      | 2094/6470 [25:59<53:55,  1.35it/s]
4|2095|Loss: 0.8444712162017822:  32%|███▏      | 2095/6470 [25:59<59:45,  1.22it/s]
4|2096|Loss: 0.6710735559463501:  32%|███▏      | 2095/6470 [26:00<59:45,  1.22it/s]
4|2096|Loss: 0.6710735559463501:  32%|███▏      | 2096/6470 [26:00<59:42,  1.22it/s]
4|2097|Loss: 0.8388920426368713:  32%|███▏      | 2096/6470 [26:00<59:42,  1.22it/s]
4|2097|Loss: 0.8388920426368713:  32%|███▏      | 2097/6470 [26:01<58:17,  1.25it/s]
4|2098|Loss: 0.8044750094413757:  32%|███▏      | 2097/6470 [26:01<58:17,  1.25it/s]
4|2098|Loss: 0.8044750094413757:  32%|███▏      | 2098/6470 [26:01<55:32,  1.31it/s]
4|2099|Loss: 0.8055964112281799:  32%|███▏      | 2098/6470 [26:02<55:32,  1.31it/s]
4|2099|Loss: 0.8055964112281799:  32%|███▏      | 2099/6470 [26:02<58:03,  1.25it/s]
4|2100|Loss: 0.7177398204803467:  32%|███▏      | 2099/6470 [26:03<58:03,  1.25it/s]
4|2100|Loss: 0.7177398204803467:  32%|███▏      | 2100/6470 [26:03<1:00:52,  1.20it/s]
4|2101|Loss: 0.73419189453125:  32%|███▏      | 2100/6470 [26:04<1:00:52,  1.20it/s]  
4|2101|Loss: 0.73419189453125:  32%|███▏      | 2101/6470 [26:04<1:00:32,  1.20it/s]
4|2102|Loss: 0.80060213804245:  32%|███▏      | 2101/6470 [26:04<1:00:32,  1.20it/s]
4|2102|Loss: 0.80060213804245:  32%|███▏      | 2102/6470 [26:05<58:46,  1.24it/s]  
4|2103|Loss: 0.7744419574737549:  32%|███▏      | 2102/6470 [26:05<58:46,  1.24it/s]
4|2103|Loss: 0.7744419574737549:  33%|███▎      | 2103/6470 [26:05<55:52,  1.30it/s]
4|2104|Loss: 0.7453941106796265:  33%|███▎      | 2103/6470 [26:06<55:52,  1.30it/s]
4|2104|Loss: 0.7453941106796265:  33%|███▎      | 2104/6470 [26:06<51:59,  1.40it/s]
4|2105|Loss: 0.8552995324134827:  33%|███▎      | 2104/6470 [26:06<51:59,  1.40it/s]
4|2105|Loss: 0.8552995324134827:  33%|███▎      | 2105/6470 [26:07<54:37,  1.33it/s]
4|2106|Loss: 0.822472870349884:  33%|███▎      | 2105/6470 [26:07<54:37,  1.33it/s] 
4|2106|Loss: 0.822472870349884:  33%|███▎      | 2106/6470 [26:07<53:48,  1.35it/s]
4|2107|Loss: 0.8986155390739441:  33%|███▎      | 2106/6470 [26:08<53:48,  1.35it/s]
4|2107|Loss: 0.8986155390739441:  33%|███▎      | 2107/6470 [26:08<48:54,  1.49it/s]
4|2108|Loss: 0.8917621970176697:  33%|███▎      | 2107/6470 [26:08<48:54,  1.49it/s]
4|2108|Loss: 0.8917621970176697:  33%|███▎      | 2108/6470 [26:09<51:22,  1.41it/s]
4|2109|Loss: 0.8059878349304199:  33%|███▎      | 2108/6470 [26:09<51:22,  1.41it/s]
4|2109|Loss: 0.8059878349304199:  33%|███▎      | 2109/6470 [26:09<51:10,  1.42it/s]
4|2110|Loss: 0.8529397249221802:  33%|███▎      | 2109/6470 [26:10<51:10,  1.42it/s]
4|2110|Loss: 0.8529397249221802:  33%|███▎      | 2110/6470 [26:10<46:59,  1.55it/s]
4|2111|Loss: 0.825225293636322:  33%|███▎      | 2110/6470 [26:10<46:59,  1.55it/s] 
4|2111|Loss: 0.825225293636322:  33%|███▎      | 2111/6470 [26:11<48:43,  1.49it/s]
4|2112|Loss: 0.7537832856178284:  33%|███▎      | 2111/6470 [26:11<48:43,  1.49it/s]
4|2112|Loss: 0.7537832856178284:  33%|███▎      | 2112/6470 [26:11<46:33,  1.56it/s]
4|2113|Loss: 0.6745402812957764:  33%|███▎      | 2112/6470 [26:12<46:33,  1.56it/s]
4|2113|Loss: 0.6745402812957764:  33%|███▎      | 2113/6470 [26:12<47:41,  1.52it/s]
4|2114|Loss: 0.947745144367218:  33%|███▎      | 2113/6470 [26:12<47:41,  1.52it/s] 
4|2114|Loss: 0.947745144367218:  33%|███▎      | 2114/6470 [26:13<47:38,  1.52it/s]
4|2115|Loss: 0.6357797384262085:  33%|███▎      | 2114/6470 [26:13<47:38,  1.52it/s]
4|2115|Loss: 0.6357797384262085:  33%|███▎      | 2115/6470 [26:13<41:47,  1.74it/s]
4|2116|Loss: 0.8069300055503845:  33%|███▎      | 2115/6470 [26:13<41:47,  1.74it/s]
4|2116|Loss: 0.8069300055503845:  33%|███▎      | 2116/6470 [26:14<42:23,  1.71it/s]
4|2117|Loss: 0.7626478672027588:  33%|███▎      | 2116/6470 [26:14<42:23,  1.71it/s]
4|2117|Loss: 0.7626478672027588:  33%|███▎      | 2117/6470 [26:14<46:06,  1.57it/s]
4|2118|Loss: 0.9337345361709595:  33%|███▎      | 2117/6470 [26:15<46:06,  1.57it/s]
4|2118|Loss: 0.9337345361709595:  33%|███▎      | 2118/6470 [26:15<45:02,  1.61it/s]
4|2119|Loss: 0.7034934759140015:  33%|███▎      | 2118/6470 [26:15<45:02,  1.61it/s]
4|2119|Loss: 0.7034934759140015:  33%|███▎      | 2119/6470 [26:16<47:56,  1.51it/s]
4|2120|Loss: 0.9387049674987793:  33%|███▎      | 2119/6470 [26:16<47:56,  1.51it/s]
4|2120|Loss: 0.9387049674987793:  33%|███▎      | 2120/6470 [26:17<53:07,  1.36it/s]
4|2121|Loss: 0.9774654507637024:  33%|███▎      | 2120/6470 [26:17<53:07,  1.36it/s]
4|2121|Loss: 0.9774654507637024:  33%|███▎      | 2121/6470 [26:18<56:56,  1.27it/s]
4|2122|Loss: 0.8067300319671631:  33%|███▎      | 2121/6470 [26:18<56:56,  1.27it/s]
4|2122|Loss: 0.8067300319671631:  33%|███▎      | 2122/6470 [26:18<52:12,  1.39it/s]
4|2123|Loss: 0.953886091709137:  33%|███▎      | 2122/6470 [26:18<52:12,  1.39it/s] 
4|2123|Loss: 0.953886091709137:  33%|███▎      | 2123/6470 [26:19<48:35,  1.49it/s]
4|2124|Loss: 0.6932313442230225:  33%|███▎      | 2123/6470 [26:19<48:35,  1.49it/s]
4|2124|Loss: 0.6932313442230225:  33%|███▎      | 2124/6470 [26:19<42:38,  1.70it/s]
4|2125|Loss: 0.8327957987785339:  33%|███▎      | 2124/6470 [26:19<42:38,  1.70it/s]
4|2125|Loss: 0.8327957987785339:  33%|███▎      | 2125/6470 [26:20<43:35,  1.66it/s]
4|2126|Loss: 0.797852098941803:  33%|███▎      | 2125/6470 [26:20<43:35,  1.66it/s] 
4|2126|Loss: 0.797852098941803:  33%|███▎      | 2126/6470 [26:20<43:58,  1.65it/s]
4|2127|Loss: 0.8952783346176147:  33%|███▎      | 2126/6470 [26:21<43:58,  1.65it/s]
4|2127|Loss: 0.8952783346176147:  33%|███▎      | 2127/6470 [26:21<46:19,  1.56it/s]
4|2128|Loss: 0.8649299144744873:  33%|███▎      | 2127/6470 [26:21<46:19,  1.56it/s]
4|2128|Loss: 0.8649299144744873:  33%|███▎      | 2128/6470 [26:22<44:26,  1.63it/s]
4|2129|Loss: 0.9191033840179443:  33%|███▎      | 2128/6470 [26:22<44:26,  1.63it/s]
4|2129|Loss: 0.9191033840179443:  33%|███▎      | 2129/6470 [26:22<46:11,  1.57it/s]
4|2130|Loss: 0.6636278033256531:  33%|███▎      | 2129/6470 [26:23<46:11,  1.57it/s]
4|2130|Loss: 0.6636278033256531:  33%|███▎      | 2130/6470 [26:23<49:10,  1.47it/s]
4|2131|Loss: 0.8865037560462952:  33%|███▎      | 2130/6470 [26:23<49:10,  1.47it/s]
4|2131|Loss: 0.8865037560462952:  33%|███▎      | 2131/6470 [26:24<49:04,  1.47it/s]
4|2132|Loss: 0.7608456611633301:  33%|███▎      | 2131/6470 [26:24<49:04,  1.47it/s]
4|2132|Loss: 0.7608456611633301:  33%|███▎      | 2132/6470 [26:24<45:16,  1.60it/s]
4|2133|Loss: 1.0320649147033691:  33%|███▎      | 2132/6470 [26:25<45:16,  1.60it/s]
4|2133|Loss: 1.0320649147033691:  33%|███▎      | 2133/6470 [26:25<47:51,  1.51it/s]
4|2134|Loss: 0.7515804171562195:  33%|███▎      | 2133/6470 [26:25<47:51,  1.51it/s]
4|2134|Loss: 0.7515804171562195:  33%|███▎      | 2134/6470 [26:26<47:54,  1.51it/s]
4|2135|Loss: 0.8661611080169678:  33%|███▎      | 2134/6470 [26:26<47:54,  1.51it/s]
4|2135|Loss: 0.8661611080169678:  33%|███▎      | 2135/6470 [26:26<45:24,  1.59it/s]
4|2136|Loss: 0.6911318898200989:  33%|███▎      | 2135/6470 [26:27<45:24,  1.59it/s]
4|2136|Loss: 0.6911318898200989:  33%|███▎      | 2136/6470 [26:27<51:29,  1.40it/s]
4|2137|Loss: 0.8906702995300293:  33%|███▎      | 2136/6470 [26:27<51:29,  1.40it/s]
4|2137|Loss: 0.8906702995300293:  33%|███▎      | 2137/6470 [26:28<47:34,  1.52it/s]
4|2138|Loss: 0.8730273246765137:  33%|███▎      | 2137/6470 [26:28<47:34,  1.52it/s]
4|2138|Loss: 0.8730273246765137:  33%|███▎      | 2138/6470 [26:28<41:45,  1.73it/s]
4|2139|Loss: 0.8308306336402893:  33%|███▎      | 2138/6470 [26:28<41:45,  1.73it/s]
4|2139|Loss: 0.8308306336402893:  33%|███▎      | 2139/6470 [26:29<45:56,  1.57it/s]
4|2140|Loss: 0.9116330742835999:  33%|███▎      | 2139/6470 [26:29<45:56,  1.57it/s]
4|2140|Loss: 0.9116330742835999:  33%|███▎      | 2140/6470 [26:30<51:41,  1.40it/s]
4|2141|Loss: 0.8626223206520081:  33%|███▎      | 2140/6470 [26:30<51:41,  1.40it/s]
4|2141|Loss: 0.8626223206520081:  33%|███▎      | 2141/6470 [26:31<53:48,  1.34it/s]
4|2142|Loss: 0.8860257863998413:  33%|███▎      | 2141/6470 [26:31<53:48,  1.34it/s]
4|2142|Loss: 0.8860257863998413:  33%|███▎      | 2142/6470 [26:31<54:08,  1.33it/s]
4|2143|Loss: 0.8212732672691345:  33%|███▎      | 2142/6470 [26:32<54:08,  1.33it/s]
4|2143|Loss: 0.8212732672691345:  33%|███▎      | 2143/6470 [26:32<54:23,  1.33it/s]
4|2144|Loss: 0.7613203525543213:  33%|███▎      | 2143/6470 [26:32<54:23,  1.33it/s]
4|2144|Loss: 0.7613203525543213:  33%|███▎      | 2144/6470 [26:33<52:42,  1.37it/s]
4|2145|Loss: 0.8401252627372742:  33%|███▎      | 2144/6470 [26:33<52:42,  1.37it/s]
4|2145|Loss: 0.8401252627372742:  33%|███▎      | 2145/6470 [26:34<57:00,  1.26it/s]
4|2146|Loss: 0.8942760229110718:  33%|███▎      | 2145/6470 [26:34<57:00,  1.26it/s]
4|2146|Loss: 0.8942760229110718:  33%|███▎      | 2146/6470 [26:34<58:29,  1.23it/s]
4|2147|Loss: 0.736538827419281:  33%|███▎      | 2146/6470 [26:35<58:29,  1.23it/s] 
4|2147|Loss: 0.736538827419281:  33%|███▎      | 2147/6470 [26:35<56:42,  1.27it/s]
4|2148|Loss: 0.9344651103019714:  33%|███▎      | 2147/6470 [26:36<56:42,  1.27it/s]
4|2148|Loss: 0.9344651103019714:  33%|███▎      | 2148/6470 [26:36<54:17,  1.33it/s]
4|2149|Loss: 0.7447493672370911:  33%|███▎      | 2148/6470 [26:36<54:17,  1.33it/s]
4|2149|Loss: 0.7447493672370911:  33%|███▎      | 2149/6470 [26:37<52:57,  1.36it/s]
4|2150|Loss: 0.753165602684021:  33%|███▎      | 2149/6470 [26:37<52:57,  1.36it/s] 
4|2150|Loss: 0.753165602684021:  33%|███▎      | 2150/6470 [26:37<52:00,  1.38it/s]
4|2151|Loss: 0.7336019277572632:  33%|███▎      | 2150/6470 [26:38<52:00,  1.38it/s]
4|2151|Loss: 0.7336019277572632:  33%|███▎      | 2151/6470 [26:38<48:34,  1.48it/s]
4|2152|Loss: 0.6756236553192139:  33%|███▎      | 2151/6470 [26:38<48:34,  1.48it/s]
4|2152|Loss: 0.6756236553192139:  33%|███▎      | 2152/6470 [26:39<1:04:15,  1.12it/s]
4|2153|Loss: 0.8104891777038574:  33%|███▎      | 2152/6470 [26:40<1:04:15,  1.12it/s]
4|2153|Loss: 0.8104891777038574:  33%|███▎      | 2153/6470 [26:40<1:06:05,  1.09it/s]
4|2154|Loss: 0.9276654720306396:  33%|███▎      | 2153/6470 [26:41<1:06:05,  1.09it/s]
4|2154|Loss: 0.9276654720306396:  33%|███▎      | 2154/6470 [26:41<1:05:21,  1.10it/s]
4|2155|Loss: 0.7987594604492188:  33%|███▎      | 2154/6470 [26:42<1:05:21,  1.10it/s]
4|2155|Loss: 0.7987594604492188:  33%|███▎      | 2155/6470 [26:42<1:01:32,  1.17it/s]
4|2156|Loss: 0.892301619052887:  33%|███▎      | 2155/6470 [26:42<1:01:32,  1.17it/s] 
4|2156|Loss: 0.892301619052887:  33%|███▎      | 2156/6470 [26:42<55:25,  1.30it/s]  
4|2157|Loss: 0.8443776965141296:  33%|███▎      | 2156/6470 [26:43<55:25,  1.30it/s]
4|2157|Loss: 0.8443776965141296:  33%|███▎      | 2157/6470 [26:43<54:15,  1.32it/s]
4|2158|Loss: 0.7914339303970337:  33%|███▎      | 2157/6470 [26:44<54:15,  1.32it/s]
4|2158|Loss: 0.7914339303970337:  33%|███▎      | 2158/6470 [26:44<51:34,  1.39it/s]
4|2159|Loss: 0.9437639117240906:  33%|███▎      | 2158/6470 [26:44<51:34,  1.39it/s]
4|2159|Loss: 0.9437639117240906:  33%|███▎      | 2159/6470 [26:45<52:10,  1.38it/s]
4|2160|Loss: 0.8070550560951233:  33%|███▎      | 2159/6470 [26:45<52:10,  1.38it/s]
4|2160|Loss: 0.8070550560951233:  33%|███▎      | 2160/6470 [26:45<54:57,  1.31it/s]
4|2161|Loss: 0.8682554364204407:  33%|███▎      | 2160/6470 [26:46<54:57,  1.31it/s]
4|2161|Loss: 0.8682554364204407:  33%|███▎      | 2161/6470 [26:46<55:24,  1.30it/s]
4|2162|Loss: 0.8426726460456848:  33%|███▎      | 2161/6470 [26:47<55:24,  1.30it/s]
4|2162|Loss: 0.8426726460456848:  33%|███▎      | 2162/6470 [26:47<56:01,  1.28it/s]
4|2163|Loss: 0.7737650871276855:  33%|███▎      | 2162/6470 [26:47<56:01,  1.28it/s]
4|2163|Loss: 0.7737650871276855:  33%|███▎      | 2163/6470 [26:48<1:00:07,  1.19it/s]
4|2164|Loss: 0.8237563371658325:  33%|███▎      | 2163/6470 [26:48<1:00:07,  1.19it/s]
4|2164|Loss: 0.8237563371658325:  33%|███▎      | 2164/6470 [26:49<1:02:46,  1.14it/s]
4|2165|Loss: 0.819035530090332:  33%|███▎      | 2164/6470 [26:49<1:02:46,  1.14it/s] 
4|2165|Loss: 0.819035530090332:  33%|███▎      | 2165/6470 [26:50<58:36,  1.22it/s]  
4|2166|Loss: 0.7861112952232361:  33%|███▎      | 2165/6470 [26:50<58:36,  1.22it/s]
4|2166|Loss: 0.7861112952232361:  33%|███▎      | 2166/6470 [26:50<54:48,  1.31it/s]
4|2167|Loss: 0.7277815937995911:  33%|███▎      | 2166/6470 [26:51<54:48,  1.31it/s]
4|2167|Loss: 0.7277815937995911:  33%|███▎      | 2167/6470 [26:51<52:51,  1.36it/s]
4|2168|Loss: 0.9092592597007751:  33%|███▎      | 2167/6470 [26:51<52:51,  1.36it/s]
4|2168|Loss: 0.9092592597007751:  34%|███▎      | 2168/6470 [26:52<53:56,  1.33it/s]
4|2169|Loss: 0.8210142850875854:  34%|███▎      | 2168/6470 [26:52<53:56,  1.33it/s]
4|2169|Loss: 0.8210142850875854:  34%|███▎      | 2169/6470 [26:52<50:55,  1.41it/s]
4|2170|Loss: 0.6834468841552734:  34%|███▎      | 2169/6470 [26:53<50:55,  1.41it/s]
4|2170|Loss: 0.6834468841552734:  34%|███▎      | 2170/6470 [26:53<49:56,  1.43it/s]
4|2171|Loss: 1.0557914972305298:  34%|███▎      | 2170/6470 [26:53<49:56,  1.43it/s]
4|2171|Loss: 1.0557914972305298:  34%|███▎      | 2171/6470 [26:54<51:56,  1.38it/s]
4|2172|Loss: 0.7579192519187927:  34%|███▎      | 2171/6470 [26:54<51:56,  1.38it/s]
4|2172|Loss: 0.7579192519187927:  34%|███▎      | 2172/6470 [26:54<48:40,  1.47it/s]
4|2173|Loss: 1.0211915969848633:  34%|███▎      | 2172/6470 [26:55<48:40,  1.47it/s]
4|2173|Loss: 1.0211915969848633:  34%|███▎      | 2173/6470 [26:55<48:11,  1.49it/s]
4|2174|Loss: 0.8319110870361328:  34%|███▎      | 2173/6470 [26:55<48:11,  1.49it/s]
4|2174|Loss: 0.8319110870361328:  34%|███▎      | 2174/6470 [26:56<50:23,  1.42it/s]
4|2175|Loss: 0.8010624647140503:  34%|███▎      | 2174/6470 [26:56<50:23,  1.42it/s]
4|2175|Loss: 0.8010624647140503:  34%|███▎      | 2175/6470 [26:57<55:10,  1.30it/s]
4|2176|Loss: 0.8927357196807861:  34%|███▎      | 2175/6470 [26:57<55:10,  1.30it/s]
4|2176|Loss: 0.8927357196807861:  34%|███▎      | 2176/6470 [26:57<55:42,  1.28it/s]
4|2177|Loss: 1.001692533493042:  34%|███▎      | 2176/6470 [26:58<55:42,  1.28it/s] 
4|2177|Loss: 1.001692533493042:  34%|███▎      | 2177/6470 [26:58<55:16,  1.29it/s]
4|2178|Loss: 0.8011178374290466:  34%|███▎      | 2177/6470 [26:59<55:16,  1.29it/s]
4|2178|Loss: 0.8011178374290466:  34%|███▎      | 2178/6470 [26:59<55:15,  1.29it/s]
4|2179|Loss: 0.903998076915741:  34%|███▎      | 2178/6470 [27:00<55:15,  1.29it/s] 
4|2179|Loss: 0.903998076915741:  34%|███▎      | 2179/6470 [27:00<59:54,  1.19it/s]
4|2180|Loss: 0.9267953038215637:  34%|███▎      | 2179/6470 [27:00<59:54,  1.19it/s]
4|2180|Loss: 0.9267953038215637:  34%|███▎      | 2180/6470 [27:01<59:59,  1.19it/s]
4|2181|Loss: 0.8040094971656799:  34%|███▎      | 2180/6470 [27:01<59:59,  1.19it/s]
4|2181|Loss: 0.8040094971656799:  34%|███▎      | 2181/6470 [27:02<58:37,  1.22it/s]
4|2182|Loss: 0.8336507678031921:  34%|███▎      | 2181/6470 [27:02<58:37,  1.22it/s]
4|2182|Loss: 0.8336507678031921:  34%|███▎      | 2182/6470 [27:02<57:45,  1.24it/s]
4|2183|Loss: 0.8260272741317749:  34%|███▎      | 2182/6470 [27:03<57:45,  1.24it/s]
4|2183|Loss: 0.8260272741317749:  34%|███▎      | 2183/6470 [27:03<53:54,  1.33it/s]
4|2184|Loss: 0.8270469903945923:  34%|███▎      | 2183/6470 [27:03<53:54,  1.33it/s]
4|2184|Loss: 0.8270469903945923:  34%|███▍      | 2184/6470 [27:04<54:01,  1.32it/s]
4|2185|Loss: 0.7482239007949829:  34%|███▍      | 2184/6470 [27:04<54:01,  1.32it/s]
4|2185|Loss: 0.7482239007949829:  34%|███▍      | 2185/6470 [27:05<55:03,  1.30it/s]
4|2186|Loss: 0.7061065435409546:  34%|███▍      | 2185/6470 [27:05<55:03,  1.30it/s]
4|2186|Loss: 0.7061065435409546:  34%|███▍      | 2186/6470 [27:05<56:41,  1.26it/s]
4|2187|Loss: 0.7819998264312744:  34%|███▍      | 2186/6470 [27:06<56:41,  1.26it/s]
4|2187|Loss: 0.7819998264312744:  34%|███▍      | 2187/6470 [27:06<57:03,  1.25it/s]
4|2188|Loss: 0.801241934299469:  34%|███▍      | 2187/6470 [27:07<57:03,  1.25it/s] 
4|2188|Loss: 0.801241934299469:  34%|███▍      | 2188/6470 [27:07<56:51,  1.26it/s]
4|2189|Loss: 0.84955233335495:  34%|███▍      | 2188/6470 [27:07<56:51,  1.26it/s] 
4|2189|Loss: 0.84955233335495:  34%|███▍      | 2189/6470 [27:08<52:24,  1.36it/s]
4|2190|Loss: 0.8790267705917358:  34%|███▍      | 2189/6470 [27:08<52:24,  1.36it/s]
4|2190|Loss: 0.8790267705917358:  34%|███▍      | 2190/6470 [27:08<49:23,  1.44it/s]
4|2191|Loss: 0.6669913530349731:  34%|███▍      | 2190/6470 [27:09<49:23,  1.44it/s]
4|2191|Loss: 0.6669913530349731:  34%|███▍      | 2191/6470 [27:09<48:25,  1.47it/s]
4|2192|Loss: 0.46488887071609497:  34%|███▍      | 2191/6470 [27:09<48:25,  1.47it/s]
4|2192|Loss: 0.46488887071609497:  34%|███▍      | 2192/6470 [27:10<48:26,  1.47it/s]
4|2193|Loss: 0.7434597611427307:  34%|███▍      | 2192/6470 [27:10<48:26,  1.47it/s] 
4|2193|Loss: 0.7434597611427307:  34%|███▍      | 2193/6470 [27:10<50:32,  1.41it/s]
4|2194|Loss: 0.8284476399421692:  34%|███▍      | 2193/6470 [27:11<50:32,  1.41it/s]
4|2194|Loss: 0.8284476399421692:  34%|███▍      | 2194/6470 [27:11<50:21,  1.41it/s]
4|2195|Loss: 0.8155426383018494:  34%|███▍      | 2194/6470 [27:11<50:21,  1.41it/s]
4|2195|Loss: 0.8155426383018494:  34%|███▍      | 2195/6470 [27:12<50:20,  1.42it/s]
4|2196|Loss: 0.8385390043258667:  34%|███▍      | 2195/6470 [27:12<50:20,  1.42it/s]
4|2196|Loss: 0.8385390043258667:  34%|███▍      | 2196/6470 [27:13<53:55,  1.32it/s]
4|2197|Loss: 0.9059338569641113:  34%|███▍      | 2196/6470 [27:13<53:55,  1.32it/s]
4|2197|Loss: 0.9059338569641113:  34%|███▍      | 2197/6470 [27:13<52:38,  1.35it/s]
4|2198|Loss: 0.8885554671287537:  34%|███▍      | 2197/6470 [27:14<52:38,  1.35it/s]
4|2198|Loss: 0.8885554671287537:  34%|███▍      | 2198/6470 [27:14<52:17,  1.36it/s]
4|2199|Loss: 0.739363968372345:  34%|███▍      | 2198/6470 [27:14<52:17,  1.36it/s] 
4|2199|Loss: 0.739363968372345:  34%|███▍      | 2199/6470 [27:15<51:32,  1.38it/s]
4|2200|Loss: 0.9573283791542053:  34%|███▍      | 2199/6470 [27:15<51:32,  1.38it/s]
4|2200|Loss: 0.9573283791542053:  34%|███▍      | 2200/6470 [27:15<51:58,  1.37it/s]
4|2201|Loss: 0.884453535079956:  34%|███▍      | 2200/6470 [27:16<51:58,  1.37it/s] 
4|2201|Loss: 0.884453535079956:  34%|███▍      | 2201/6470 [27:16<50:33,  1.41it/s]
4|2202|Loss: 0.8100755214691162:  34%|███▍      | 2201/6470 [27:17<50:33,  1.41it/s]
4|2202|Loss: 0.8100755214691162:  34%|███▍      | 2202/6470 [27:17<51:50,  1.37it/s]
4|2203|Loss: 0.6327760219573975:  34%|███▍      | 2202/6470 [27:17<51:50,  1.37it/s]
4|2203|Loss: 0.6327760219573975:  34%|███▍      | 2203/6470 [27:18<51:18,  1.39it/s]
4|2204|Loss: 0.8106600046157837:  34%|███▍      | 2203/6470 [27:18<51:18,  1.39it/s]
4|2204|Loss: 0.8106600046157837:  34%|███▍      | 2204/6470 [27:18<52:37,  1.35it/s]
4|2205|Loss: 0.8327954411506653:  34%|███▍      | 2204/6470 [27:19<52:37,  1.35it/s]
4|2205|Loss: 0.8327954411506653:  34%|███▍      | 2205/6470 [27:19<51:51,  1.37it/s]
4|2206|Loss: 0.8571006655693054:  34%|███▍      | 2205/6470 [27:20<51:51,  1.37it/s]
4|2206|Loss: 0.8571006655693054:  34%|███▍      | 2206/6470 [27:20<51:47,  1.37it/s]
4|2207|Loss: 0.8669488430023193:  34%|███▍      | 2206/6470 [27:20<51:47,  1.37it/s]
4|2207|Loss: 0.8669488430023193:  34%|███▍      | 2207/6470 [27:21<54:01,  1.31it/s]
4|2208|Loss: 0.6619881391525269:  34%|███▍      | 2207/6470 [27:21<54:01,  1.31it/s]
4|2208|Loss: 0.6619881391525269:  34%|███▍      | 2208/6470 [27:21<54:28,  1.30it/s]
4|2209|Loss: 0.8737049698829651:  34%|███▍      | 2208/6470 [27:22<54:28,  1.30it/s]
4|2209|Loss: 0.8737049698829651:  34%|███▍      | 2209/6470 [27:22<51:30,  1.38it/s]
4|2210|Loss: 0.8275249004364014:  34%|███▍      | 2209/6470 [27:23<51:30,  1.38it/s]
4|2210|Loss: 0.8275249004364014:  34%|███▍      | 2210/6470 [27:23<54:29,  1.30it/s]
4|2211|Loss: 1.041508436203003:  34%|███▍      | 2210/6470 [27:23<54:29,  1.30it/s] 
4|2211|Loss: 1.041508436203003:  34%|███▍      | 2211/6470 [27:24<55:46,  1.27it/s]
4|2212|Loss: 0.7610536813735962:  34%|███▍      | 2211/6470 [27:24<55:46,  1.27it/s]
4|2212|Loss: 0.7610536813735962:  34%|███▍      | 2212/6470 [27:24<50:37,  1.40it/s]
4|2213|Loss: 0.7706703543663025:  34%|███▍      | 2212/6470 [27:25<50:37,  1.40it/s]
4|2213|Loss: 0.7706703543663025:  34%|███▍      | 2213/6470 [27:25<48:02,  1.48it/s]
4|2214|Loss: 0.8078384399414062:  34%|███▍      | 2213/6470 [27:25<48:02,  1.48it/s]
4|2214|Loss: 0.8078384399414062:  34%|███▍      | 2214/6470 [27:26<47:34,  1.49it/s]
4|2215|Loss: 0.8092681765556335:  34%|███▍      | 2214/6470 [27:26<47:34,  1.49it/s]
4|2215|Loss: 0.8092681765556335:  34%|███▍      | 2215/6470 [27:26<44:45,  1.58it/s]
4|2216|Loss: 0.7801311612129211:  34%|███▍      | 2215/6470 [27:26<44:45,  1.58it/s]
4|2216|Loss: 0.7801311612129211:  34%|███▍      | 2216/6470 [27:27<43:50,  1.62it/s]
4|2217|Loss: 0.9612446427345276:  34%|███▍      | 2216/6470 [27:27<43:50,  1.62it/s]
4|2217|Loss: 0.9612446427345276:  34%|███▍      | 2217/6470 [27:28<48:26,  1.46it/s]
4|2218|Loss: 0.7876235246658325:  34%|███▍      | 2217/6470 [27:28<48:26,  1.46it/s]
4|2218|Loss: 0.7876235246658325:  34%|███▍      | 2218/6470 [27:28<50:44,  1.40it/s]
4|2219|Loss: 0.8048943281173706:  34%|███▍      | 2218/6470 [27:29<50:44,  1.40it/s]
4|2219|Loss: 0.8048943281173706:  34%|███▍      | 2219/6470 [27:29<54:24,  1.30it/s]
4|2220|Loss: 0.8110970258712769:  34%|███▍      | 2219/6470 [27:30<54:24,  1.30it/s]
4|2220|Loss: 0.8110970258712769:  34%|███▍      | 2220/6470 [27:30<57:58,  1.22it/s]
4|2221|Loss: 0.8594101667404175:  34%|███▍      | 2220/6470 [27:31<57:58,  1.22it/s]
4|2221|Loss: 0.8594101667404175:  34%|███▍      | 2221/6470 [27:31<58:06,  1.22it/s]
4|2222|Loss: 0.8430331945419312:  34%|███▍      | 2221/6470 [27:31<58:06,  1.22it/s]
4|2222|Loss: 0.8430331945419312:  34%|███▍      | 2222/6470 [27:32<55:50,  1.27it/s]
4|2223|Loss: 0.9435222744941711:  34%|███▍      | 2222/6470 [27:32<55:50,  1.27it/s]
4|2223|Loss: 0.9435222744941711:  34%|███▍      | 2223/6470 [27:33<57:47,  1.22it/s]
4|2224|Loss: 0.9108710885047913:  34%|███▍      | 2223/6470 [27:33<57:47,  1.22it/s]
4|2224|Loss: 0.9108710885047913:  34%|███▍      | 2224/6470 [27:33<59:15,  1.19it/s]
4|2225|Loss: 0.7251272201538086:  34%|███▍      | 2224/6470 [27:34<59:15,  1.19it/s]
4|2225|Loss: 0.7251272201538086:  34%|███▍      | 2225/6470 [27:34<59:00,  1.20it/s]
4|2226|Loss: 0.806465744972229:  34%|███▍      | 2225/6470 [27:35<59:00,  1.20it/s] 
4|2226|Loss: 0.806465744972229:  34%|███▍      | 2226/6470 [27:35<59:31,  1.19it/s]
4|2227|Loss: 0.7944728136062622:  34%|███▍      | 2226/6470 [27:36<59:31,  1.19it/s]
4|2227|Loss: 0.7944728136062622:  34%|███▍      | 2227/6470 [27:36<1:01:32,  1.15it/s]
4|2228|Loss: 0.6523963212966919:  34%|███▍      | 2227/6470 [27:37<1:01:32,  1.15it/s]
4|2228|Loss: 0.6523963212966919:  34%|███▍      | 2228/6470 [27:37<1:00:12,  1.17it/s]
4|2229|Loss: 0.8781340718269348:  34%|███▍      | 2228/6470 [27:37<1:00:12,  1.17it/s]
4|2229|Loss: 0.8781340718269348:  34%|███▍      | 2229/6470 [27:38<58:57,  1.20it/s]  
4|2230|Loss: 0.8419268727302551:  34%|███▍      | 2229/6470 [27:38<58:57,  1.20it/s]
4|2230|Loss: 0.8419268727302551:  34%|███▍      | 2230/6470 [27:38<58:30,  1.21it/s]
4|2231|Loss: 0.7290470004081726:  34%|███▍      | 2230/6470 [27:39<58:30,  1.21it/s]
4|2231|Loss: 0.7290470004081726:  34%|███▍      | 2231/6470 [27:39<1:00:07,  1.18it/s]
4|2232|Loss: 0.8337278962135315:  34%|███▍      | 2231/6470 [27:40<1:00:07,  1.18it/s]
4|2232|Loss: 0.8337278962135315:  34%|███▍      | 2232/6470 [27:40<1:00:42,  1.16it/s]
4|2233|Loss: 0.8570763468742371:  34%|███▍      | 2232/6470 [27:41<1:00:42,  1.16it/s]
4|2233|Loss: 0.8570763468742371:  35%|███▍      | 2233/6470 [27:41<1:02:26,  1.13it/s]
4|2234|Loss: 0.8027149438858032:  35%|███▍      | 2233/6470 [27:42<1:02:26,  1.13it/s]
4|2234|Loss: 0.8027149438858032:  35%|███▍      | 2234/6470 [27:42<1:04:02,  1.10it/s]
4|2235|Loss: 0.837723433971405:  35%|███▍      | 2234/6470 [27:43<1:04:02,  1.10it/s] 
4|2235|Loss: 0.837723433971405:  35%|███▍      | 2235/6470 [27:43<1:00:41,  1.16it/s]
4|2236|Loss: 0.7617639899253845:  35%|███▍      | 2235/6470 [27:43<1:00:41,  1.16it/s]
4|2236|Loss: 0.7617639899253845:  35%|███▍      | 2236/6470 [27:44<57:12,  1.23it/s]  
4|2237|Loss: 0.7349727749824524:  35%|███▍      | 2236/6470 [27:44<57:12,  1.23it/s]
4|2237|Loss: 0.7349727749824524:  35%|███▍      | 2237/6470 [27:44<56:13,  1.25it/s]
4|2238|Loss: 0.7215489149093628:  35%|███▍      | 2237/6470 [27:45<56:13,  1.25it/s]
4|2238|Loss: 0.7215489149093628:  35%|███▍      | 2238/6470 [27:45<56:36,  1.25it/s]
4|2239|Loss: 0.8896818161010742:  35%|███▍      | 2238/6470 [27:46<56:36,  1.25it/s]
4|2239|Loss: 0.8896818161010742:  35%|███▍      | 2239/6470 [27:46<55:07,  1.28it/s]
4|2240|Loss: 0.7444139719009399:  35%|███▍      | 2239/6470 [27:46<55:07,  1.28it/s]
4|2240|Loss: 0.7444139719009399:  35%|███▍      | 2240/6470 [27:47<57:18,  1.23it/s]
4|2241|Loss: 0.733762800693512:  35%|███▍      | 2240/6470 [27:47<57:18,  1.23it/s] 
4|2241|Loss: 0.733762800693512:  35%|███▍      | 2241/6470 [27:47<53:15,  1.32it/s]
4|2242|Loss: 0.9861096143722534:  35%|███▍      | 2241/6470 [27:48<53:15,  1.32it/s]
4|2242|Loss: 0.9861096143722534:  35%|███▍      | 2242/6470 [27:48<51:15,  1.37it/s]
4|2243|Loss: 0.7127448320388794:  35%|███▍      | 2242/6470 [27:48<51:15,  1.37it/s]
4|2243|Loss: 0.7127448320388794:  35%|███▍      | 2243/6470 [27:49<49:24,  1.43it/s]
4|2244|Loss: 0.7701334953308105:  35%|███▍      | 2243/6470 [27:49<49:24,  1.43it/s]
4|2244|Loss: 0.7701334953308105:  35%|███▍      | 2244/6470 [27:49<49:51,  1.41it/s]
4|2245|Loss: 1.0623884201049805:  35%|███▍      | 2244/6470 [27:50<49:51,  1.41it/s]
4|2245|Loss: 1.0623884201049805:  35%|███▍      | 2245/6470 [27:50<51:41,  1.36it/s]
4|2246|Loss: 0.7424551248550415:  35%|███▍      | 2245/6470 [27:51<51:41,  1.36it/s]
4|2246|Loss: 0.7424551248550415:  35%|███▍      | 2246/6470 [27:51<51:40,  1.36it/s]
4|2247|Loss: 0.8191139698028564:  35%|███▍      | 2246/6470 [27:51<51:40,  1.36it/s]
4|2247|Loss: 0.8191139698028564:  35%|███▍      | 2247/6470 [27:52<51:06,  1.38it/s]
4|2248|Loss: 0.8276681900024414:  35%|███▍      | 2247/6470 [27:52<51:06,  1.38it/s]
4|2248|Loss: 0.8276681900024414:  35%|███▍      | 2248/6470 [27:52<50:10,  1.40it/s]
4|2249|Loss: 0.782789945602417:  35%|███▍      | 2248/6470 [27:53<50:10,  1.40it/s] 
4|2249|Loss: 0.782789945602417:  35%|███▍      | 2249/6470 [27:53<49:55,  1.41it/s]
4|2250|Loss: 0.5697925090789795:  35%|███▍      | 2249/6470 [27:54<49:55,  1.41it/s]
4|2250|Loss: 0.5697925090789795:  35%|███▍      | 2250/6470 [27:54<51:30,  1.37it/s]
4|2251|Loss: 0.8241732120513916:  35%|███▍      | 2250/6470 [27:54<51:30,  1.37it/s]
4|2251|Loss: 0.8241732120513916:  35%|███▍      | 2251/6470 [27:55<55:14,  1.27it/s]
4|2252|Loss: 0.7580110430717468:  35%|███▍      | 2251/6470 [27:55<55:14,  1.27it/s]
4|2252|Loss: 0.7580110430717468:  35%|███▍      | 2252/6470 [27:56<55:40,  1.26it/s]
4|2253|Loss: 0.9479915499687195:  35%|███▍      | 2252/6470 [27:56<55:40,  1.26it/s]
4|2253|Loss: 0.9479915499687195:  35%|███▍      | 2253/6470 [27:56<55:36,  1.26it/s]
4|2254|Loss: 1.0759440660476685:  35%|███▍      | 2253/6470 [27:57<55:36,  1.26it/s]
4|2254|Loss: 1.0759440660476685:  35%|███▍      | 2254/6470 [27:57<59:14,  1.19it/s]
4|2255|Loss: 0.6665465235710144:  35%|███▍      | 2254/6470 [27:58<59:14,  1.19it/s]
4|2255|Loss: 0.6665465235710144:  35%|███▍      | 2255/6470 [27:58<59:15,  1.19it/s]
4|2256|Loss: 0.9092649221420288:  35%|███▍      | 2255/6470 [27:59<59:15,  1.19it/s]
4|2256|Loss: 0.9092649221420288:  35%|███▍      | 2256/6470 [27:59<56:38,  1.24it/s]
4|2257|Loss: 0.7692093849182129:  35%|███▍      | 2256/6470 [27:59<56:38,  1.24it/s]
4|2257|Loss: 0.7692093849182129:  35%|███▍      | 2257/6470 [28:00<53:38,  1.31it/s]
4|2258|Loss: 0.7738734483718872:  35%|███▍      | 2257/6470 [28:00<53:38,  1.31it/s]
4|2258|Loss: 0.7738734483718872:  35%|███▍      | 2258/6470 [28:00<53:08,  1.32it/s]
4|2259|Loss: 0.8361595273017883:  35%|███▍      | 2258/6470 [28:01<53:08,  1.32it/s]
4|2259|Loss: 0.8361595273017883:  35%|███▍      | 2259/6470 [28:01<55:57,  1.25it/s]
4|2260|Loss: 0.8423491716384888:  35%|███▍      | 2259/6470 [28:02<55:57,  1.25it/s]
4|2260|Loss: 0.8423491716384888:  35%|███▍      | 2260/6470 [28:02<58:02,  1.21it/s]
4|2261|Loss: 0.7465595006942749:  35%|███▍      | 2260/6470 [28:03<58:02,  1.21it/s]
4|2261|Loss: 0.7465595006942749:  35%|███▍      | 2261/6470 [28:03<54:12,  1.29it/s]
4|2262|Loss: 0.7910459637641907:  35%|███▍      | 2261/6470 [28:03<54:12,  1.29it/s]
4|2262|Loss: 0.7910459637641907:  35%|███▍      | 2262/6470 [28:03<51:01,  1.37it/s]
4|2263|Loss: 1.058876633644104:  35%|███▍      | 2262/6470 [28:04<51:01,  1.37it/s] 
4|2263|Loss: 1.058876633644104:  35%|███▍      | 2263/6470 [28:04<47:30,  1.48it/s]
4|2264|Loss: 0.8909559845924377:  35%|███▍      | 2263/6470 [28:04<47:30,  1.48it/s]
4|2264|Loss: 0.8909559845924377:  35%|███▍      | 2264/6470 [28:04<44:47,  1.57it/s]
4|2265|Loss: 0.8787403702735901:  35%|███▍      | 2264/6470 [28:05<44:47,  1.57it/s]
4|2265|Loss: 0.8787403702735901:  35%|███▌      | 2265/6470 [28:05<49:07,  1.43it/s]
4|2266|Loss: 0.8869566917419434:  35%|███▌      | 2265/6470 [28:06<49:07,  1.43it/s]
4|2266|Loss: 0.8869566917419434:  35%|███▌      | 2266/6470 [28:06<48:14,  1.45it/s]
4|2267|Loss: 0.8764927387237549:  35%|███▌      | 2266/6470 [28:06<48:14,  1.45it/s]
4|2267|Loss: 0.8764927387237549:  35%|███▌      | 2267/6470 [28:07<51:47,  1.35it/s]
4|2268|Loss: 0.7605935335159302:  35%|███▌      | 2267/6470 [28:07<51:47,  1.35it/s]
4|2268|Loss: 0.7605935335159302:  35%|███▌      | 2268/6470 [28:08<51:19,  1.36it/s]
4|2269|Loss: 0.8135721683502197:  35%|███▌      | 2268/6470 [28:08<51:19,  1.36it/s]
4|2269|Loss: 0.8135721683502197:  35%|███▌      | 2269/6470 [28:08<50:24,  1.39it/s]
4|2270|Loss: 0.7765066027641296:  35%|███▌      | 2269/6470 [28:09<50:24,  1.39it/s]
4|2270|Loss: 0.7765066027641296:  35%|███▌      | 2270/6470 [28:09<48:34,  1.44it/s]
4|2271|Loss: 0.8163692951202393:  35%|███▌      | 2270/6470 [28:09<48:34,  1.44it/s]
4|2271|Loss: 0.8163692951202393:  35%|███▌      | 2271/6470 [28:10<51:42,  1.35it/s]
4|2272|Loss: 0.8668392300605774:  35%|███▌      | 2271/6470 [28:10<51:42,  1.35it/s]
4|2272|Loss: 0.8668392300605774:  35%|███▌      | 2272/6470 [28:10<49:45,  1.41it/s]
4|2273|Loss: 0.8315339684486389:  35%|███▌      | 2272/6470 [28:11<49:45,  1.41it/s]
4|2273|Loss: 0.8315339684486389:  35%|███▌      | 2273/6470 [28:11<51:55,  1.35it/s]
4|2274|Loss: 0.72963947057724:  35%|███▌      | 2273/6470 [28:12<51:55,  1.35it/s]  
4|2274|Loss: 0.72963947057724:  35%|███▌      | 2274/6470 [28:12<52:58,  1.32it/s]
4|2275|Loss: 0.8700150847434998:  35%|███▌      | 2274/6470 [28:12<52:58,  1.32it/s]
4|2275|Loss: 0.8700150847434998:  35%|███▌      | 2275/6470 [28:13<53:56,  1.30it/s]
4|2276|Loss: 0.6286802887916565:  35%|███▌      | 2275/6470 [28:13<53:56,  1.30it/s]
4|2276|Loss: 0.6286802887916565:  35%|███▌      | 2276/6470 [28:13<46:31,  1.50it/s]
4|2277|Loss: 0.8818191289901733:  35%|███▌      | 2276/6470 [28:14<46:31,  1.50it/s]
4|2277|Loss: 0.8818191289901733:  35%|███▌      | 2277/6470 [28:14<45:21,  1.54it/s]
4|2278|Loss: 0.8398091793060303:  35%|███▌      | 2277/6470 [28:14<45:21,  1.54it/s]
4|2278|Loss: 0.8398091793060303:  35%|███▌      | 2278/6470 [28:15<51:36,  1.35it/s]
4|2279|Loss: 0.7528622150421143:  35%|███▌      | 2278/6470 [28:15<51:36,  1.35it/s]
4|2279|Loss: 0.7528622150421143:  35%|███▌      | 2279/6470 [28:16<52:20,  1.33it/s]
4|2280|Loss: 0.8722243905067444:  35%|███▌      | 2279/6470 [28:16<52:20,  1.33it/s]
4|2280|Loss: 0.8722243905067444:  35%|███▌      | 2280/6470 [28:16<52:55,  1.32it/s]
4|2281|Loss: 0.9015217423439026:  35%|███▌      | 2280/6470 [28:17<52:55,  1.32it/s]
4|2281|Loss: 0.9015217423439026:  35%|███▌      | 2281/6470 [28:17<52:07,  1.34it/s]
4|2282|Loss: 0.8221998810768127:  35%|███▌      | 2281/6470 [28:17<52:07,  1.34it/s]
4|2282|Loss: 0.8221998810768127:  35%|███▌      | 2282/6470 [28:18<52:29,  1.33it/s]
4|2283|Loss: 0.7959752678871155:  35%|███▌      | 2282/6470 [28:18<52:29,  1.33it/s]
4|2283|Loss: 0.7959752678871155:  35%|███▌      | 2283/6470 [28:19<54:46,  1.27it/s]
4|2284|Loss: 0.8394510746002197:  35%|███▌      | 2283/6470 [28:19<54:46,  1.27it/s]
4|2284|Loss: 0.8394510746002197:  35%|███▌      | 2284/6470 [28:19<53:09,  1.31it/s]
4|2285|Loss: 0.9627828001976013:  35%|███▌      | 2284/6470 [28:20<53:09,  1.31it/s]
4|2285|Loss: 0.9627828001976013:  35%|███▌      | 2285/6470 [28:20<56:15,  1.24it/s]
4|2286|Loss: 0.9276025295257568:  35%|███▌      | 2285/6470 [28:21<56:15,  1.24it/s]
4|2286|Loss: 0.9276025295257568:  35%|███▌      | 2286/6470 [28:21<54:50,  1.27it/s]
4|2287|Loss: 0.6823663711547852:  35%|███▌      | 2286/6470 [28:21<54:50,  1.27it/s]
4|2287|Loss: 0.6823663711547852:  35%|███▌      | 2287/6470 [28:22<54:26,  1.28it/s]
4|2288|Loss: 0.6946588158607483:  35%|███▌      | 2287/6470 [28:22<54:26,  1.28it/s]
4|2288|Loss: 0.6946588158607483:  35%|███▌      | 2288/6470 [28:23<52:59,  1.32it/s]
4|2289|Loss: 0.7139668464660645:  35%|███▌      | 2288/6470 [28:23<52:59,  1.32it/s]
4|2289|Loss: 0.7139668464660645:  35%|███▌      | 2289/6470 [28:23<54:27,  1.28it/s]
4|2290|Loss: 0.8200843930244446:  35%|███▌      | 2289/6470 [28:24<54:27,  1.28it/s]
4|2290|Loss: 0.8200843930244446:  35%|███▌      | 2290/6470 [28:24<54:56,  1.27it/s]
4|2291|Loss: 0.8166114687919617:  35%|███▌      | 2290/6470 [28:25<54:56,  1.27it/s]
4|2291|Loss: 0.8166114687919617:  35%|███▌      | 2291/6470 [28:25<57:15,  1.22it/s]
4|2292|Loss: 0.6717320084571838:  35%|███▌      | 2291/6470 [28:26<57:15,  1.22it/s]
4|2292|Loss: 0.6717320084571838:  35%|███▌      | 2292/6470 [28:26<58:22,  1.19it/s]
4|2293|Loss: 0.8933113217353821:  35%|███▌      | 2292/6470 [28:26<58:22,  1.19it/s]
4|2293|Loss: 0.8933113217353821:  35%|███▌      | 2293/6470 [28:27<57:04,  1.22it/s]
4|2294|Loss: 0.8795949816703796:  35%|███▌      | 2293/6470 [28:27<57:04,  1.22it/s]
4|2294|Loss: 0.8795949816703796:  35%|███▌      | 2294/6470 [28:27<53:20,  1.30it/s]
4|2295|Loss: 0.9590516090393066:  35%|███▌      | 2294/6470 [28:28<53:20,  1.30it/s]
4|2295|Loss: 0.9590516090393066:  35%|███▌      | 2295/6470 [28:28<50:58,  1.36it/s]
4|2296|Loss: 0.892179548740387:  35%|███▌      | 2295/6470 [28:28<50:58,  1.36it/s] 
4|2296|Loss: 0.892179548740387:  35%|███▌      | 2296/6470 [28:29<50:23,  1.38it/s]
4|2297|Loss: 0.7296447157859802:  35%|███▌      | 2296/6470 [28:29<50:23,  1.38it/s]
4|2297|Loss: 0.7296447157859802:  36%|███▌      | 2297/6470 [28:29<48:04,  1.45it/s]
4|2298|Loss: 0.767318606376648:  36%|███▌      | 2297/6470 [28:30<48:04,  1.45it/s] 
4|2298|Loss: 0.767318606376648:  36%|███▌      | 2298/6470 [28:30<51:23,  1.35it/s]
4|2299|Loss: 0.9596685171127319:  36%|███▌      | 2298/6470 [28:31<51:23,  1.35it/s]
4|2299|Loss: 0.9596685171127319:  36%|███▌      | 2299/6470 [28:31<50:24,  1.38it/s]
4|2300|Loss: 0.8263359069824219:  36%|███▌      | 2299/6470 [28:31<50:24,  1.38it/s]
4|2300|Loss: 0.8263359069824219:  36%|███▌      | 2300/6470 [28:31<48:13,  1.44it/s]
4|2301|Loss: 0.7565180063247681:  36%|███▌      | 2300/6470 [28:32<48:13,  1.44it/s]
4|2301|Loss: 0.7565180063247681:  36%|███▌      | 2301/6470 [28:32<47:24,  1.47it/s]
4|2302|Loss: 0.7131391763687134:  36%|███▌      | 2301/6470 [28:33<47:24,  1.47it/s]
4|2302|Loss: 0.7131391763687134:  36%|███▌      | 2302/6470 [28:33<51:12,  1.36it/s]
4|2303|Loss: 0.7107193470001221:  36%|███▌      | 2302/6470 [28:33<51:12,  1.36it/s]
4|2303|Loss: 0.7107193470001221:  36%|███▌      | 2303/6470 [28:34<49:10,  1.41it/s]
4|2304|Loss: 1.0832592248916626:  36%|███▌      | 2303/6470 [28:34<49:10,  1.41it/s]
4|2304|Loss: 1.0832592248916626:  36%|███▌      | 2304/6470 [28:34<49:58,  1.39it/s]
4|2305|Loss: 0.8060811758041382:  36%|███▌      | 2304/6470 [28:35<49:58,  1.39it/s]
4|2305|Loss: 0.8060811758041382:  36%|███▌      | 2305/6470 [28:35<49:20,  1.41it/s]
4|2306|Loss: 0.9003667235374451:  36%|███▌      | 2305/6470 [28:35<49:20,  1.41it/s]
4|2306|Loss: 0.9003667235374451:  36%|███▌      | 2306/6470 [28:36<46:57,  1.48it/s]
4|2307|Loss: 0.7922872304916382:  36%|███▌      | 2306/6470 [28:36<46:57,  1.48it/s]
4|2307|Loss: 0.7922872304916382:  36%|███▌      | 2307/6470 [28:36<48:29,  1.43it/s]
4|2308|Loss: 0.787170946598053:  36%|███▌      | 2307/6470 [28:37<48:29,  1.43it/s] 
4|2308|Loss: 0.787170946598053:  36%|███▌      | 2308/6470 [28:37<48:57,  1.42it/s]
4|2309|Loss: 0.8690983653068542:  36%|███▌      | 2308/6470 [28:37<48:57,  1.42it/s]
4|2309|Loss: 0.8690983653068542:  36%|███▌      | 2309/6470 [28:38<45:17,  1.53it/s]
4|2310|Loss: 0.7183592915534973:  36%|███▌      | 2309/6470 [28:38<45:17,  1.53it/s]
4|2310|Loss: 0.7183592915534973:  36%|███▌      | 2310/6470 [28:38<45:16,  1.53it/s]
4|2311|Loss: 0.8126228451728821:  36%|███▌      | 2310/6470 [28:39<45:16,  1.53it/s]
4|2311|Loss: 0.8126228451728821:  36%|███▌      | 2311/6470 [28:39<48:08,  1.44it/s]
4|2312|Loss: 0.9063172340393066:  36%|███▌      | 2311/6470 [28:40<48:08,  1.44it/s]
4|2312|Loss: 0.9063172340393066:  36%|███▌      | 2312/6470 [28:40<52:59,  1.31it/s]
4|2313|Loss: 0.7994285225868225:  36%|███▌      | 2312/6470 [28:40<52:59,  1.31it/s]
4|2313|Loss: 0.7994285225868225:  36%|███▌      | 2313/6470 [28:41<51:55,  1.33it/s]
4|2314|Loss: 1.1319772005081177:  36%|███▌      | 2313/6470 [28:41<51:55,  1.33it/s]
4|2314|Loss: 1.1319772005081177:  36%|███▌      | 2314/6470 [28:42<56:32,  1.23it/s]
4|2315|Loss: 0.8973070979118347:  36%|███▌      | 2314/6470 [28:42<56:32,  1.23it/s]
4|2315|Loss: 0.8973070979118347:  36%|███▌      | 2315/6470 [28:42<55:19,  1.25it/s]
4|2316|Loss: 0.7160055041313171:  36%|███▌      | 2315/6470 [28:43<55:19,  1.25it/s]
4|2316|Loss: 0.7160055041313171:  36%|███▌      | 2316/6470 [28:43<54:35,  1.27it/s]
4|2317|Loss: 0.7728947997093201:  36%|███▌      | 2316/6470 [28:44<54:35,  1.27it/s]
4|2317|Loss: 0.7728947997093201:  36%|███▌      | 2317/6470 [28:44<52:59,  1.31it/s]
4|2318|Loss: 0.9131820797920227:  36%|███▌      | 2317/6470 [28:44<52:59,  1.31it/s]
4|2318|Loss: 0.9131820797920227:  36%|███▌      | 2318/6470 [28:45<55:17,  1.25it/s]
4|2319|Loss: 0.7887709736824036:  36%|███▌      | 2318/6470 [28:45<55:17,  1.25it/s]
4|2319|Loss: 0.7887709736824036:  36%|███▌      | 2319/6470 [28:46<54:03,  1.28it/s]
4|2320|Loss: 0.7267680168151855:  36%|███▌      | 2319/6470 [28:46<54:03,  1.28it/s]
4|2320|Loss: 0.7267680168151855:  36%|███▌      | 2320/6470 [28:47<57:27,  1.20it/s]
4|2321|Loss: 0.8248859643936157:  36%|███▌      | 2320/6470 [28:47<57:27,  1.20it/s]
4|2321|Loss: 0.8248859643936157:  36%|███▌      | 2321/6470 [28:47<57:32,  1.20it/s]
4|2322|Loss: 0.8403608798980713:  36%|███▌      | 2321/6470 [28:48<57:32,  1.20it/s]
4|2322|Loss: 0.8403608798980713:  36%|███▌      | 2322/6470 [28:48<58:20,  1.18it/s]
4|2323|Loss: 0.8903989195823669:  36%|███▌      | 2322/6470 [28:49<58:20,  1.18it/s]
4|2323|Loss: 0.8903989195823669:  36%|███▌      | 2323/6470 [28:49<56:55,  1.21it/s]
4|2324|Loss: 0.8368671536445618:  36%|███▌      | 2323/6470 [28:49<56:55,  1.21it/s]
4|2324|Loss: 0.8368671536445618:  36%|███▌      | 2324/6470 [28:50<53:54,  1.28it/s]
4|2325|Loss: 0.8411574363708496:  36%|███▌      | 2324/6470 [28:50<53:54,  1.28it/s]
4|2325|Loss: 0.8411574363708496:  36%|███▌      | 2325/6470 [28:50<53:27,  1.29it/s]
4|2326|Loss: 0.8521262407302856:  36%|███▌      | 2325/6470 [28:51<53:27,  1.29it/s]
4|2326|Loss: 0.8521262407302856:  36%|███▌      | 2326/6470 [28:51<53:48,  1.28it/s]
4|2327|Loss: 0.9227970242500305:  36%|███▌      | 2326/6470 [28:52<53:48,  1.28it/s]
4|2327|Loss: 0.9227970242500305:  36%|███▌      | 2327/6470 [28:52<52:16,  1.32it/s]
4|2328|Loss: 0.8099324703216553:  36%|███▌      | 2327/6470 [28:52<52:16,  1.32it/s]
4|2328|Loss: 0.8099324703216553:  36%|███▌      | 2328/6470 [28:53<54:03,  1.28it/s]
4|2329|Loss: 0.7914699912071228:  36%|███▌      | 2328/6470 [28:53<54:03,  1.28it/s]
4|2329|Loss: 0.7914699912071228:  36%|███▌      | 2329/6470 [28:53<50:18,  1.37it/s]
4|2330|Loss: 0.9647573232650757:  36%|███▌      | 2329/6470 [28:54<50:18,  1.37it/s]
4|2330|Loss: 0.9647573232650757:  36%|███▌      | 2330/6470 [28:54<50:24,  1.37it/s]
4|2331|Loss: 0.9160641431808472:  36%|███▌      | 2330/6470 [28:55<50:24,  1.37it/s]
4|2331|Loss: 0.9160641431808472:  36%|███▌      | 2331/6470 [28:55<51:59,  1.33it/s]
4|2332|Loss: 0.7857625484466553:  36%|███▌      | 2331/6470 [28:55<51:59,  1.33it/s]
4|2332|Loss: 0.7857625484466553:  36%|███▌      | 2332/6470 [28:55<47:54,  1.44it/s]
4|2333|Loss: 0.8162223100662231:  36%|███▌      | 2332/6470 [28:56<47:54,  1.44it/s]
4|2333|Loss: 0.8162223100662231:  36%|███▌      | 2333/6470 [28:56<46:19,  1.49it/s]
4|2334|Loss: 0.8314565420150757:  36%|███▌      | 2333/6470 [28:57<46:19,  1.49it/s]
4|2334|Loss: 0.8314565420150757:  36%|███▌      | 2334/6470 [28:57<48:14,  1.43it/s]
4|2335|Loss: 0.8767952919006348:  36%|███▌      | 2334/6470 [28:57<48:14,  1.43it/s]
4|2335|Loss: 0.8767952919006348:  36%|███▌      | 2335/6470 [28:58<53:06,  1.30it/s]
4|2336|Loss: 0.9353910684585571:  36%|███▌      | 2335/6470 [28:58<53:06,  1.30it/s]
4|2336|Loss: 0.9353910684585571:  36%|███▌      | 2336/6470 [28:59<51:55,  1.33it/s]
4|2337|Loss: 0.8363714218139648:  36%|███▌      | 2336/6470 [28:59<51:55,  1.33it/s]
4|2337|Loss: 0.8363714218139648:  36%|███▌      | 2337/6470 [28:59<48:35,  1.42it/s]
4|2338|Loss: 0.6832756996154785:  36%|███▌      | 2337/6470 [29:00<48:35,  1.42it/s]
4|2338|Loss: 0.6832756996154785:  36%|███▌      | 2338/6470 [29:00<48:59,  1.41it/s]
4|2339|Loss: 0.9841480851173401:  36%|███▌      | 2338/6470 [29:00<48:59,  1.41it/s]
4|2339|Loss: 0.9841480851173401:  36%|███▌      | 2339/6470 [29:00<46:21,  1.49it/s]
4|2340|Loss: 0.8837928175926208:  36%|███▌      | 2339/6470 [29:01<46:21,  1.49it/s]
4|2340|Loss: 0.8837928175926208:  36%|███▌      | 2340/6470 [29:01<47:19,  1.45it/s]
4|2341|Loss: 0.9491953253746033:  36%|███▌      | 2340/6470 [29:02<47:19,  1.45it/s]
4|2341|Loss: 0.9491953253746033:  36%|███▌      | 2341/6470 [29:02<52:06,  1.32it/s]
4|2342|Loss: 0.8708500266075134:  36%|███▌      | 2341/6470 [29:03<52:06,  1.32it/s]
4|2342|Loss: 0.8708500266075134:  36%|███▌      | 2342/6470 [29:03<55:01,  1.25it/s]
4|2343|Loss: 0.8986631631851196:  36%|███▌      | 2342/6470 [29:03<55:01,  1.25it/s]
4|2343|Loss: 0.8986631631851196:  36%|███▌      | 2343/6470 [29:04<51:41,  1.33it/s]
4|2344|Loss: 0.6970536708831787:  36%|███▌      | 2343/6470 [29:04<51:41,  1.33it/s]
4|2344|Loss: 0.6970536708831787:  36%|███▌      | 2344/6470 [29:04<47:18,  1.45it/s]
4|2345|Loss: 0.9947136640548706:  36%|███▌      | 2344/6470 [29:05<47:18,  1.45it/s]
4|2345|Loss: 0.9947136640548706:  36%|███▌      | 2345/6470 [29:05<48:40,  1.41it/s]
4|2346|Loss: 0.7792462706565857:  36%|███▌      | 2345/6470 [29:05<48:40,  1.41it/s]
4|2346|Loss: 0.7792462706565857:  36%|███▋      | 2346/6470 [29:06<50:04,  1.37it/s]
4|2347|Loss: 0.8201978802680969:  36%|███▋      | 2346/6470 [29:06<50:04,  1.37it/s]
4|2347|Loss: 0.8201978802680969:  36%|███▋      | 2347/6470 [29:06<51:19,  1.34it/s]
4|2348|Loss: 0.7957783937454224:  36%|███▋      | 2347/6470 [29:07<51:19,  1.34it/s]
4|2348|Loss: 0.7957783937454224:  36%|███▋      | 2348/6470 [29:07<51:45,  1.33it/s]
4|2349|Loss: 0.8169505596160889:  36%|███▋      | 2348/6470 [29:08<51:45,  1.33it/s]
4|2349|Loss: 0.8169505596160889:  36%|███▋      | 2349/6470 [29:08<51:37,  1.33it/s]
4|2350|Loss: 0.8343721032142639:  36%|███▋      | 2349/6470 [29:08<51:37,  1.33it/s]
4|2350|Loss: 0.8343721032142639:  36%|███▋      | 2350/6470 [29:09<55:10,  1.24it/s]
4|2351|Loss: 0.8700106739997864:  36%|███▋      | 2350/6470 [29:09<55:10,  1.24it/s]
4|2351|Loss: 0.8700106739997864:  36%|███▋      | 2351/6470 [29:10<54:41,  1.26it/s]
4|2352|Loss: 0.8087663650512695:  36%|███▋      | 2351/6470 [29:10<54:41,  1.26it/s]
4|2352|Loss: 0.8087663650512695:  36%|███▋      | 2352/6470 [29:11<56:41,  1.21it/s]
4|2353|Loss: 0.8088158965110779:  36%|███▋      | 2352/6470 [29:11<56:41,  1.21it/s]
4|2353|Loss: 0.8088158965110779:  36%|███▋      | 2353/6470 [29:11<55:03,  1.25it/s]
4|2354|Loss: 0.9495006203651428:  36%|███▋      | 2353/6470 [29:12<55:03,  1.25it/s]
4|2354|Loss: 0.9495006203651428:  36%|███▋      | 2354/6470 [29:12<54:37,  1.26it/s]
4|2355|Loss: 0.8760944604873657:  36%|███▋      | 2354/6470 [29:12<54:37,  1.26it/s]
4|2355|Loss: 0.8760944604873657:  36%|███▋      | 2355/6470 [29:13<50:38,  1.35it/s]
4|2356|Loss: 0.7417097091674805:  36%|███▋      | 2355/6470 [29:13<50:38,  1.35it/s]
4|2356|Loss: 0.7417097091674805:  36%|███▋      | 2356/6470 [29:13<48:26,  1.42it/s]
4|2357|Loss: 0.8121450543403625:  36%|███▋      | 2356/6470 [29:14<48:26,  1.42it/s]
4|2357|Loss: 0.8121450543403625:  36%|███▋      | 2357/6470 [29:14<49:46,  1.38it/s]
4|2358|Loss: 0.7638955116271973:  36%|███▋      | 2357/6470 [29:15<49:46,  1.38it/s]
4|2358|Loss: 0.7638955116271973:  36%|███▋      | 2358/6470 [29:15<48:10,  1.42it/s]
4|2359|Loss: 0.7021446228027344:  36%|███▋      | 2358/6470 [29:15<48:10,  1.42it/s]
4|2359|Loss: 0.7021446228027344:  36%|███▋      | 2359/6470 [29:15<43:57,  1.56it/s]
4|2360|Loss: 0.9050000905990601:  36%|███▋      | 2359/6470 [29:16<43:57,  1.56it/s]
4|2360|Loss: 0.9050000905990601:  36%|███▋      | 2360/6470 [29:16<45:09,  1.52it/s]
4|2361|Loss: 0.8367252349853516:  36%|███▋      | 2360/6470 [29:16<45:09,  1.52it/s]
4|2361|Loss: 0.8367252349853516:  36%|███▋      | 2361/6470 [29:17<49:21,  1.39it/s]
4|2362|Loss: 0.7744084596633911:  36%|███▋      | 2361/6470 [29:17<49:21,  1.39it/s]
4|2362|Loss: 0.7744084596633911:  37%|███▋      | 2362/6470 [29:18<49:47,  1.37it/s]
4|2363|Loss: 0.8568662405014038:  37%|███▋      | 2362/6470 [29:18<49:47,  1.37it/s]
4|2363|Loss: 0.8568662405014038:  37%|███▋      | 2363/6470 [29:19<54:17,  1.26it/s]
4|2364|Loss: 0.8776952028274536:  37%|███▋      | 2363/6470 [29:19<54:17,  1.26it/s]
4|2364|Loss: 0.8776952028274536:  37%|███▋      | 2364/6470 [29:19<55:40,  1.23it/s]
4|2365|Loss: 0.7209097743034363:  37%|███▋      | 2364/6470 [29:20<55:40,  1.23it/s]
4|2365|Loss: 0.7209097743034363:  37%|███▋      | 2365/6470 [29:20<52:13,  1.31it/s]
4|2366|Loss: 0.7693739533424377:  37%|███▋      | 2365/6470 [29:20<52:13,  1.31it/s]
4|2366|Loss: 0.7693739533424377:  37%|███▋      | 2366/6470 [29:21<49:39,  1.38it/s]
4|2367|Loss: 0.8535158038139343:  37%|███▋      | 2366/6470 [29:21<49:39,  1.38it/s]
4|2367|Loss: 0.8535158038139343:  37%|███▋      | 2367/6470 [29:21<45:19,  1.51it/s]
4|2368|Loss: 0.9956063628196716:  37%|███▋      | 2367/6470 [29:22<45:19,  1.51it/s]
4|2368|Loss: 0.9956063628196716:  37%|███▋      | 2368/6470 [29:22<49:09,  1.39it/s]
4|2369|Loss: 0.8018727898597717:  37%|███▋      | 2368/6470 [29:22<49:09,  1.39it/s]
4|2369|Loss: 0.8018727898597717:  37%|███▋      | 2369/6470 [29:23<50:01,  1.37it/s]
4|2370|Loss: 0.8586558699607849:  37%|███▋      | 2369/6470 [29:23<50:01,  1.37it/s]
4|2370|Loss: 0.8586558699607849:  37%|███▋      | 2370/6470 [29:23<48:03,  1.42it/s]
4|2371|Loss: 0.8270222544670105:  37%|███▋      | 2370/6470 [29:24<48:03,  1.42it/s]
4|2371|Loss: 0.8270222544670105:  37%|███▋      | 2371/6470 [29:24<48:55,  1.40it/s]
4|2372|Loss: 0.8450174927711487:  37%|███▋      | 2371/6470 [29:25<48:55,  1.40it/s]
4|2372|Loss: 0.8450174927711487:  37%|███▋      | 2372/6470 [29:25<49:50,  1.37it/s]
4|2373|Loss: 0.6686655282974243:  37%|███▋      | 2372/6470 [29:25<49:50,  1.37it/s]
4|2373|Loss: 0.6686655282974243:  37%|███▋      | 2373/6470 [29:26<49:31,  1.38it/s]
4|2374|Loss: 0.8108521103858948:  37%|███▋      | 2373/6470 [29:26<49:31,  1.38it/s]
4|2374|Loss: 0.8108521103858948:  37%|███▋      | 2374/6470 [29:26<50:34,  1.35it/s]
4|2375|Loss: 0.6522688269615173:  37%|███▋      | 2374/6470 [29:27<50:34,  1.35it/s]
4|2375|Loss: 0.6522688269615173:  37%|███▋      | 2375/6470 [29:27<51:27,  1.33it/s]
4|2376|Loss: 0.8543127179145813:  37%|███▋      | 2375/6470 [29:28<51:27,  1.33it/s]
4|2376|Loss: 0.8543127179145813:  37%|███▋      | 2376/6470 [29:28<52:13,  1.31it/s]
4|2377|Loss: 0.8320803046226501:  37%|███▋      | 2376/6470 [29:28<52:13,  1.31it/s]
4|2377|Loss: 0.8320803046226501:  37%|███▋      | 2377/6470 [29:29<49:25,  1.38it/s]
4|2378|Loss: 0.8037774562835693:  37%|███▋      | 2377/6470 [29:29<49:25,  1.38it/s]
4|2378|Loss: 0.8037774562835693:  37%|███▋      | 2378/6470 [29:29<48:15,  1.41it/s]
4|2379|Loss: 0.7116822600364685:  37%|███▋      | 2378/6470 [29:30<48:15,  1.41it/s]
4|2379|Loss: 0.7116822600364685:  37%|███▋      | 2379/6470 [29:30<54:09,  1.26it/s]
4|2380|Loss: 0.7247324585914612:  37%|███▋      | 2379/6470 [29:31<54:09,  1.26it/s]
4|2380|Loss: 0.7247324585914612:  37%|███▋      | 2380/6470 [29:31<52:16,  1.30it/s]
4|2381|Loss: 0.923970639705658:  37%|███▋      | 2380/6470 [29:31<52:16,  1.30it/s] 
4|2381|Loss: 0.923970639705658:  37%|███▋      | 2381/6470 [29:32<47:41,  1.43it/s]
4|2382|Loss: 0.7632931470870972:  37%|███▋      | 2381/6470 [29:32<47:41,  1.43it/s]
4|2382|Loss: 0.7632931470870972:  37%|███▋      | 2382/6470 [29:32<50:09,  1.36it/s]
4|2383|Loss: 0.8893030881881714:  37%|███▋      | 2382/6470 [29:33<50:09,  1.36it/s]
4|2383|Loss: 0.8893030881881714:  37%|███▋      | 2383/6470 [29:33<51:43,  1.32it/s]
4|2384|Loss: 0.8258703351020813:  37%|███▋      | 2383/6470 [29:34<51:43,  1.32it/s]
4|2384|Loss: 0.8258703351020813:  37%|███▋      | 2384/6470 [29:34<49:25,  1.38it/s]
4|2385|Loss: 0.7670599222183228:  37%|███▋      | 2384/6470 [29:34<49:25,  1.38it/s]
4|2385|Loss: 0.7670599222183228:  37%|███▋      | 2385/6470 [29:34<47:40,  1.43it/s]
4|2386|Loss: 0.8194769024848938:  37%|███▋      | 2385/6470 [29:35<47:40,  1.43it/s]
4|2386|Loss: 0.8194769024848938:  37%|███▋      | 2386/6470 [29:35<44:57,  1.51it/s]
4|2387|Loss: 0.8568202257156372:  37%|███▋      | 2386/6470 [29:35<44:57,  1.51it/s]
4|2387|Loss: 0.8568202257156372:  37%|███▋      | 2387/6470 [29:36<41:31,  1.64it/s]
4|2388|Loss: 0.8466989398002625:  37%|███▋      | 2387/6470 [29:36<41:31,  1.64it/s]
4|2388|Loss: 0.8466989398002625:  37%|███▋      | 2388/6470 [29:36<43:49,  1.55it/s]
4|2389|Loss: 0.7523013353347778:  37%|███▋      | 2388/6470 [29:37<43:49,  1.55it/s]
4|2389|Loss: 0.7523013353347778:  37%|███▋      | 2389/6470 [29:37<45:35,  1.49it/s]
4|2390|Loss: 0.8712176084518433:  37%|███▋      | 2389/6470 [29:37<45:35,  1.49it/s]
4|2390|Loss: 0.8712176084518433:  37%|███▋      | 2390/6470 [29:38<49:37,  1.37it/s]
4|2391|Loss: 0.6867832541465759:  37%|███▋      | 2390/6470 [29:38<49:37,  1.37it/s]
4|2391|Loss: 0.6867832541465759:  37%|███▋      | 2391/6470 [29:39<50:41,  1.34it/s]
4|2392|Loss: 0.8320180773735046:  37%|███▋      | 2391/6470 [29:39<50:41,  1.34it/s]
4|2392|Loss: 0.8320180773735046:  37%|███▋      | 2392/6470 [29:39<51:23,  1.32it/s]
4|2393|Loss: 0.7690340280532837:  37%|███▋      | 2392/6470 [29:40<51:23,  1.32it/s]
4|2393|Loss: 0.7690340280532837:  37%|███▋      | 2393/6470 [29:40<50:37,  1.34it/s]
4|2394|Loss: 0.6775826215744019:  37%|███▋      | 2393/6470 [29:41<50:37,  1.34it/s]
4|2394|Loss: 0.6775826215744019:  37%|███▋      | 2394/6470 [29:41<49:12,  1.38it/s]
4|2395|Loss: 0.7679378390312195:  37%|███▋      | 2394/6470 [29:41<49:12,  1.38it/s]
4|2395|Loss: 0.7679378390312195:  37%|███▋      | 2395/6470 [29:42<49:56,  1.36it/s]
4|2396|Loss: 0.8886780142784119:  37%|███▋      | 2395/6470 [29:42<49:56,  1.36it/s]
4|2396|Loss: 0.8886780142784119:  37%|███▋      | 2396/6470 [29:42<48:20,  1.40it/s]
4|2397|Loss: 0.9625375270843506:  37%|███▋      | 2396/6470 [29:43<48:20,  1.40it/s]
4|2397|Loss: 0.9625375270843506:  37%|███▋      | 2397/6470 [29:43<46:16,  1.47it/s]
4|2398|Loss: 0.7005969882011414:  37%|███▋      | 2397/6470 [29:43<46:16,  1.47it/s]
4|2398|Loss: 0.7005969882011414:  37%|███▋      | 2398/6470 [29:44<46:20,  1.46it/s]
4|2399|Loss: 0.7151402831077576:  37%|███▋      | 2398/6470 [29:44<46:20,  1.46it/s]
4|2399|Loss: 0.7151402831077576:  37%|███▋      | 2399/6470 [29:44<49:15,  1.38it/s]
4|2400|Loss: 0.7768864631652832:  37%|███▋      | 2399/6470 [29:45<49:15,  1.38it/s]
4|2400|Loss: 0.7768864631652832:  37%|███▋      | 2400/6470 [29:45<52:24,  1.29it/s]
4|2401|Loss: 0.9533717036247253:  37%|███▋      | 2400/6470 [29:46<52:24,  1.29it/s]
4|2401|Loss: 0.9533717036247253:  37%|███▋      | 2401/6470 [29:46<55:23,  1.22it/s]
4|2402|Loss: 0.8300288915634155:  37%|███▋      | 2401/6470 [29:47<55:23,  1.22it/s]
4|2402|Loss: 0.8300288915634155:  37%|███▋      | 2402/6470 [29:47<55:31,  1.22it/s]
4|2403|Loss: 0.8166216611862183:  37%|███▋      | 2402/6470 [29:47<55:31,  1.22it/s]
4|2403|Loss: 0.8166216611862183:  37%|███▋      | 2403/6470 [29:48<53:00,  1.28it/s]
4|2404|Loss: 0.8618463277816772:  37%|███▋      | 2403/6470 [29:48<53:00,  1.28it/s]
4|2404|Loss: 0.8618463277816772:  37%|███▋      | 2404/6470 [29:49<55:42,  1.22it/s]
4|2405|Loss: 0.9382061958312988:  37%|███▋      | 2404/6470 [29:49<55:42,  1.22it/s]
4|2405|Loss: 0.9382061958312988:  37%|███▋      | 2405/6470 [29:49<55:46,  1.21it/s]
4|2406|Loss: 0.8353605270385742:  37%|███▋      | 2405/6470 [29:50<55:46,  1.21it/s]
4|2406|Loss: 0.8353605270385742:  37%|███▋      | 2406/6470 [29:50<53:39,  1.26it/s]
4|2407|Loss: 1.0010569095611572:  37%|███▋      | 2406/6470 [29:51<53:39,  1.26it/s]
4|2407|Loss: 1.0010569095611572:  37%|███▋      | 2407/6470 [29:51<52:34,  1.29it/s]
4|2408|Loss: 0.881296694278717:  37%|███▋      | 2407/6470 [29:51<52:34,  1.29it/s] 
4|2408|Loss: 0.881296694278717:  37%|███▋      | 2408/6470 [29:52<52:37,  1.29it/s]
4|2409|Loss: 0.8201126456260681:  37%|███▋      | 2408/6470 [29:52<52:37,  1.29it/s]
4|2409|Loss: 0.8201126456260681:  37%|███▋      | 2409/6470 [29:52<51:01,  1.33it/s]
4|2410|Loss: 0.786210298538208:  37%|███▋      | 2409/6470 [29:53<51:01,  1.33it/s] 
4|2410|Loss: 0.786210298538208:  37%|███▋      | 2410/6470 [29:53<53:26,  1.27it/s]
4|2411|Loss: 0.7617685198783875:  37%|███▋      | 2410/6470 [29:54<53:26,  1.27it/s]
4|2411|Loss: 0.7617685198783875:  37%|███▋      | 2411/6470 [29:54<54:20,  1.25it/s]
4|2412|Loss: 0.8107337951660156:  37%|███▋      | 2411/6470 [29:55<54:20,  1.25it/s]
4|2412|Loss: 0.8107337951660156:  37%|███▋      | 2412/6470 [29:55<55:46,  1.21it/s]
4|2413|Loss: 0.941398024559021:  37%|███▋      | 2412/6470 [29:55<55:46,  1.21it/s] 
4|2413|Loss: 0.941398024559021:  37%|███▋      | 2413/6470 [29:56<57:42,  1.17it/s]
4|2414|Loss: 0.7795815467834473:  37%|███▋      | 2413/6470 [29:56<57:42,  1.17it/s]
4|2414|Loss: 0.7795815467834473:  37%|███▋      | 2414/6470 [29:56<52:58,  1.28it/s]
4|2415|Loss: 0.6673039793968201:  37%|███▋      | 2414/6470 [29:57<52:58,  1.28it/s]
4|2415|Loss: 0.6673039793968201:  37%|███▋      | 2415/6470 [29:57<51:46,  1.31it/s]
4|2416|Loss: 0.8344190120697021:  37%|███▋      | 2415/6470 [29:58<51:46,  1.31it/s]
4|2416|Loss: 0.8344190120697021:  37%|███▋      | 2416/6470 [29:58<54:11,  1.25it/s]
4|2417|Loss: 0.8230112791061401:  37%|███▋      | 2416/6470 [29:59<54:11,  1.25it/s]
4|2417|Loss: 0.8230112791061401:  37%|███▋      | 2417/6470 [29:59<54:30,  1.24it/s]
4|2418|Loss: 0.7603574991226196:  37%|███▋      | 2417/6470 [29:59<54:30,  1.24it/s]
4|2418|Loss: 0.7603574991226196:  37%|███▋      | 2418/6470 [30:00<54:15,  1.24it/s]
4|2419|Loss: 0.9002615213394165:  37%|███▋      | 2418/6470 [30:00<54:15,  1.24it/s]
4|2419|Loss: 0.9002615213394165:  37%|███▋      | 2419/6470 [30:01<56:45,  1.19it/s]
4|2420|Loss: 0.7768113017082214:  37%|███▋      | 2419/6470 [30:01<56:45,  1.19it/s]
4|2420|Loss: 0.7768113017082214:  37%|███▋      | 2420/6470 [30:02<58:55,  1.15it/s]
4|2421|Loss: 0.7517104744911194:  37%|███▋      | 2420/6470 [30:02<58:55,  1.15it/s]
4|2421|Loss: 0.7517104744911194:  37%|███▋      | 2421/6470 [30:02<54:45,  1.23it/s]
4|2422|Loss: 0.6820953488349915:  37%|███▋      | 2421/6470 [30:03<54:45,  1.23it/s]
4|2422|Loss: 0.6820953488349915:  37%|███▋      | 2422/6470 [30:03<56:46,  1.19it/s]
4|2423|Loss: 0.7742345929145813:  37%|███▋      | 2422/6470 [30:04<56:46,  1.19it/s]
4|2423|Loss: 0.7742345929145813:  37%|███▋      | 2423/6470 [30:04<53:25,  1.26it/s]
4|2424|Loss: 0.8336265683174133:  37%|███▋      | 2423/6470 [30:04<53:25,  1.26it/s]
4|2424|Loss: 0.8336265683174133:  37%|███▋      | 2424/6470 [30:05<52:43,  1.28it/s]
4|2425|Loss: 0.8823626041412354:  37%|███▋      | 2424/6470 [30:05<52:43,  1.28it/s]
4|2425|Loss: 0.8823626041412354:  37%|███▋      | 2425/6470 [30:05<54:06,  1.25it/s]
4|2426|Loss: 0.9637824296951294:  37%|███▋      | 2425/6470 [30:06<54:06,  1.25it/s]
4|2426|Loss: 0.9637824296951294:  37%|███▋      | 2426/6470 [30:06<55:58,  1.20it/s]
4|2427|Loss: 0.804996132850647:  37%|███▋      | 2426/6470 [30:07<55:58,  1.20it/s] 
4|2427|Loss: 0.804996132850647:  38%|███▊      | 2427/6470 [30:07<53:36,  1.26it/s]
4|2428|Loss: 0.8435605764389038:  38%|███▊      | 2427/6470 [30:07<53:36,  1.26it/s]
4|2428|Loss: 0.8435605764389038:  38%|███▊      | 2428/6470 [30:08<54:17,  1.24it/s]
4|2429|Loss: 0.6540727615356445:  38%|███▊      | 2428/6470 [30:08<54:17,  1.24it/s]
4|2429|Loss: 0.6540727615356445:  38%|███▊      | 2429/6470 [30:09<52:40,  1.28it/s]
4|2430|Loss: 0.891948401927948:  38%|███▊      | 2429/6470 [30:09<52:40,  1.28it/s] 
4|2430|Loss: 0.891948401927948:  38%|███▊      | 2430/6470 [30:09<52:16,  1.29it/s]
4|2431|Loss: 0.7473522424697876:  38%|███▊      | 2430/6470 [30:10<52:16,  1.29it/s]
4|2431|Loss: 0.7473522424697876:  38%|███▊      | 2431/6470 [30:10<49:50,  1.35it/s]
4|2432|Loss: 0.8589249849319458:  38%|███▊      | 2431/6470 [30:10<49:50,  1.35it/s]
4|2432|Loss: 0.8589249849319458:  38%|███▊      | 2432/6470 [30:11<51:50,  1.30it/s]
4|2433|Loss: 0.9419950246810913:  38%|███▊      | 2432/6470 [30:11<51:50,  1.30it/s]
4|2433|Loss: 0.9419950246810913:  38%|███▊      | 2433/6470 [30:12<54:02,  1.24it/s]
4|2434|Loss: 0.8230487108230591:  38%|███▊      | 2433/6470 [30:12<54:02,  1.24it/s]
4|2434|Loss: 0.8230487108230591:  38%|███▊      | 2434/6470 [30:13<54:16,  1.24it/s]
4|2435|Loss: 0.8919262290000916:  38%|███▊      | 2434/6470 [30:13<54:16,  1.24it/s]
4|2435|Loss: 0.8919262290000916:  38%|███▊      | 2435/6470 [30:13<53:47,  1.25it/s]
4|2436|Loss: 0.8288058638572693:  38%|███▊      | 2435/6470 [30:14<53:47,  1.25it/s]
4|2436|Loss: 0.8288058638572693:  38%|███▊      | 2436/6470 [30:14<54:52,  1.23it/s]
4|2437|Loss: 0.8363256454467773:  38%|███▊      | 2436/6470 [30:15<54:52,  1.23it/s]
4|2437|Loss: 0.8363256454467773:  38%|███▊      | 2437/6470 [30:15<51:32,  1.30it/s]
4|2438|Loss: 1.046870231628418:  38%|███▊      | 2437/6470 [30:15<51:32,  1.30it/s] 
4|2438|Loss: 1.046870231628418:  38%|███▊      | 2438/6470 [30:16<51:22,  1.31it/s]
4|2439|Loss: 0.7622548341751099:  38%|███▊      | 2438/6470 [30:16<51:22,  1.31it/s]
4|2439|Loss: 0.7622548341751099:  38%|███▊      | 2439/6470 [30:16<52:12,  1.29it/s]
4|2440|Loss: 0.9375810027122498:  38%|███▊      | 2439/6470 [30:17<52:12,  1.29it/s]
4|2440|Loss: 0.9375810027122498:  38%|███▊      | 2440/6470 [30:17<53:21,  1.26it/s]
4|2441|Loss: 0.8915072083473206:  38%|███▊      | 2440/6470 [30:18<53:21,  1.26it/s]
4|2441|Loss: 0.8915072083473206:  38%|███▊      | 2441/6470 [30:18<50:32,  1.33it/s]
4|2442|Loss: 0.7688480615615845:  38%|███▊      | 2441/6470 [30:18<50:32,  1.33it/s]
4|2442|Loss: 0.7688480615615845:  38%|███▊      | 2442/6470 [30:19<50:06,  1.34it/s]
4|2443|Loss: 0.9263543486595154:  38%|███▊      | 2442/6470 [30:19<50:06,  1.34it/s]
4|2443|Loss: 0.9263543486595154:  38%|███▊      | 2443/6470 [30:19<48:55,  1.37it/s]
4|2444|Loss: 0.713293731212616:  38%|███▊      | 2443/6470 [30:20<48:55,  1.37it/s] 
4|2444|Loss: 0.713293731212616:  38%|███▊      | 2444/6470 [30:20<45:57,  1.46it/s]
4|2445|Loss: 0.6881797313690186:  38%|███▊      | 2444/6470 [30:20<45:57,  1.46it/s]
4|2445|Loss: 0.6881797313690186:  38%|███▊      | 2445/6470 [30:21<45:44,  1.47it/s]
4|2446|Loss: 0.9397340416908264:  38%|███▊      | 2445/6470 [30:21<45:44,  1.47it/s]
4|2446|Loss: 0.9397340416908264:  38%|███▊      | 2446/6470 [30:21<47:58,  1.40it/s]
4|2447|Loss: 0.9844011068344116:  38%|███▊      | 2446/6470 [30:22<47:58,  1.40it/s]
4|2447|Loss: 0.9844011068344116:  38%|███▊      | 2447/6470 [30:22<49:03,  1.37it/s]
4|2448|Loss: 0.8605919480323792:  38%|███▊      | 2447/6470 [30:23<49:03,  1.37it/s]
4|2448|Loss: 0.8605919480323792:  38%|███▊      | 2448/6470 [30:23<49:28,  1.35it/s]
4|2449|Loss: 0.8095476627349854:  38%|███▊      | 2448/6470 [30:23<49:28,  1.35it/s]
4|2449|Loss: 0.8095476627349854:  38%|███▊      | 2449/6470 [30:24<49:43,  1.35it/s]
4|2450|Loss: 0.843993604183197:  38%|███▊      | 2449/6470 [30:24<49:43,  1.35it/s] 
4|2450|Loss: 0.843993604183197:  38%|███▊      | 2450/6470 [30:24<49:19,  1.36it/s]
4|2451|Loss: 0.8196472525596619:  38%|███▊      | 2450/6470 [30:25<49:19,  1.36it/s]
4|2451|Loss: 0.8196472525596619:  38%|███▊      | 2451/6470 [30:25<52:24,  1.28it/s]
4|2452|Loss: 0.8410623669624329:  38%|███▊      | 2451/6470 [30:26<52:24,  1.28it/s]
4|2452|Loss: 0.8410623669624329:  38%|███▊      | 2452/6470 [30:26<51:57,  1.29it/s]
4|2453|Loss: 0.9295052886009216:  38%|███▊      | 2452/6470 [30:26<51:57,  1.29it/s]
4|2453|Loss: 0.9295052886009216:  38%|███▊      | 2453/6470 [30:27<49:09,  1.36it/s]
4|2454|Loss: 0.9174935221672058:  38%|███▊      | 2453/6470 [30:27<49:09,  1.36it/s]
4|2454|Loss: 0.9174935221672058:  38%|███▊      | 2454/6470 [30:27<46:25,  1.44it/s]
4|2455|Loss: 0.8721523880958557:  38%|███▊      | 2454/6470 [30:28<46:25,  1.44it/s]
4|2455|Loss: 0.8721523880958557:  38%|███▊      | 2455/6470 [30:28<45:20,  1.48it/s]
4|2456|Loss: 0.8170492649078369:  38%|███▊      | 2455/6470 [30:28<45:20,  1.48it/s]
4|2456|Loss: 0.8170492649078369:  38%|███▊      | 2456/6470 [30:29<45:03,  1.48it/s]
4|2457|Loss: 0.7986108660697937:  38%|███▊      | 2456/6470 [30:29<45:03,  1.48it/s]
4|2457|Loss: 0.7986108660697937:  38%|███▊      | 2457/6470 [30:29<49:07,  1.36it/s]
4|2458|Loss: 0.9267072677612305:  38%|███▊      | 2457/6470 [30:30<49:07,  1.36it/s]
4|2458|Loss: 0.9267072677612305:  38%|███▊      | 2458/6470 [30:30<52:04,  1.28it/s]
4|2459|Loss: 0.7560868859291077:  38%|███▊      | 2458/6470 [30:31<52:04,  1.28it/s]
4|2459|Loss: 0.7560868859291077:  38%|███▊      | 2459/6470 [30:31<51:58,  1.29it/s]
4|2460|Loss: 0.8213651776313782:  38%|███▊      | 2459/6470 [30:32<51:58,  1.29it/s]
4|2460|Loss: 0.8213651776313782:  38%|███▊      | 2460/6470 [30:32<54:31,  1.23it/s]
4|2461|Loss: 0.8927240371704102:  38%|███▊      | 2460/6470 [30:33<54:31,  1.23it/s]
4|2461|Loss: 0.8927240371704102:  38%|███▊      | 2461/6470 [30:33<56:30,  1.18it/s]
4|2462|Loss: 0.8046762347221375:  38%|███▊      | 2461/6470 [30:33<56:30,  1.18it/s]
4|2462|Loss: 0.8046762347221375:  38%|███▊      | 2462/6470 [30:34<52:06,  1.28it/s]
4|2463|Loss: 0.8577272295951843:  38%|███▊      | 2462/6470 [30:34<52:06,  1.28it/s]
4|2463|Loss: 0.8577272295951843:  38%|███▊      | 2463/6470 [30:34<48:06,  1.39it/s]
4|2464|Loss: 0.6776793599128723:  38%|███▊      | 2463/6470 [30:35<48:06,  1.39it/s]
4|2464|Loss: 0.6776793599128723:  38%|███▊      | 2464/6470 [30:35<55:27,  1.20it/s]
4|2465|Loss: 0.7873239517211914:  38%|███▊      | 2464/6470 [30:36<55:27,  1.20it/s]
4|2465|Loss: 0.7873239517211914:  38%|███▊      | 2465/6470 [30:36<56:29,  1.18it/s]
4|2466|Loss: 0.8564512729644775:  38%|███▊      | 2465/6470 [30:37<56:29,  1.18it/s]
4|2466|Loss: 0.8564512729644775:  38%|███▊      | 2466/6470 [30:37<55:48,  1.20it/s]
4|2467|Loss: 0.8997244238853455:  38%|███▊      | 2466/6470 [30:37<55:48,  1.20it/s]
4|2467|Loss: 0.8997244238853455:  38%|███▊      | 2467/6470 [30:37<50:26,  1.32it/s]
4|2468|Loss: 0.8390451073646545:  38%|███▊      | 2467/6470 [30:38<50:26,  1.32it/s]
4|2468|Loss: 0.8390451073646545:  38%|███▊      | 2468/6470 [30:38<47:58,  1.39it/s]
4|2469|Loss: 0.9814547300338745:  38%|███▊      | 2468/6470 [30:39<47:58,  1.39it/s]
4|2469|Loss: 0.9814547300338745:  38%|███▊      | 2469/6470 [30:39<50:11,  1.33it/s]
4|2470|Loss: 0.9315887689590454:  38%|███▊      | 2469/6470 [30:39<50:11,  1.33it/s]
4|2470|Loss: 0.9315887689590454:  38%|███▊      | 2470/6470 [30:40<48:56,  1.36it/s]
4|2471|Loss: 0.8829666972160339:  38%|███▊      | 2470/6470 [30:40<48:56,  1.36it/s]
4|2471|Loss: 0.8829666972160339:  38%|███▊      | 2471/6470 [30:40<49:25,  1.35it/s]
4|2472|Loss: 0.8351225852966309:  38%|███▊      | 2471/6470 [30:41<49:25,  1.35it/s]
4|2472|Loss: 0.8351225852966309:  38%|███▊      | 2472/6470 [30:41<50:45,  1.31it/s]
4|2473|Loss: 0.8300182819366455:  38%|███▊      | 2472/6470 [30:42<50:45,  1.31it/s]
4|2473|Loss: 0.8300182819366455:  38%|███▊      | 2473/6470 [30:42<52:49,  1.26it/s]
4|2474|Loss: 0.8833581209182739:  38%|███▊      | 2473/6470 [30:43<52:49,  1.26it/s]
4|2474|Loss: 0.8833581209182739:  38%|███▊      | 2474/6470 [30:43<54:53,  1.21it/s]
4|2475|Loss: 0.8084028363227844:  38%|███▊      | 2474/6470 [30:43<54:53,  1.21it/s]
4|2475|Loss: 0.8084028363227844:  38%|███▊      | 2475/6470 [30:44<55:35,  1.20it/s]
4|2476|Loss: 0.9750300645828247:  38%|███▊      | 2475/6470 [30:44<55:35,  1.20it/s]
4|2476|Loss: 0.9750300645828247:  38%|███▊      | 2476/6470 [30:45<53:43,  1.24it/s]
4|2477|Loss: 0.9123917818069458:  38%|███▊      | 2476/6470 [30:45<53:43,  1.24it/s]
4|2477|Loss: 0.9123917818069458:  38%|███▊      | 2477/6470 [30:45<48:10,  1.38it/s]
4|2478|Loss: 0.9083046317100525:  38%|███▊      | 2477/6470 [30:46<48:10,  1.38it/s]
4|2478|Loss: 0.9083046317100525:  38%|███▊      | 2478/6470 [30:46<50:12,  1.33it/s]
4|2479|Loss: 0.8559502363204956:  38%|███▊      | 2478/6470 [30:46<50:12,  1.33it/s]
4|2479|Loss: 0.8559502363204956:  38%|███▊      | 2479/6470 [30:47<54:51,  1.21it/s]
4|2480|Loss: 0.8323483467102051:  38%|███▊      | 2479/6470 [30:47<54:51,  1.21it/s]
4|2480|Loss: 0.8323483467102051:  38%|███▊      | 2480/6470 [30:48<50:19,  1.32it/s]
4|2481|Loss: 0.756483256816864:  38%|███▊      | 2480/6470 [30:48<50:19,  1.32it/s] 
4|2481|Loss: 0.756483256816864:  38%|███▊      | 2481/6470 [30:48<51:11,  1.30it/s]
4|2482|Loss: 0.7531753182411194:  38%|███▊      | 2481/6470 [30:49<51:11,  1.30it/s]
4|2482|Loss: 0.7531753182411194:  38%|███▊      | 2482/6470 [30:49<52:20,  1.27it/s]
4|2483|Loss: 0.8357492685317993:  38%|███▊      | 2482/6470 [30:50<52:20,  1.27it/s]
4|2483|Loss: 0.8357492685317993:  38%|███▊      | 2483/6470 [30:50<50:29,  1.32it/s]
4|2484|Loss: 0.7698259949684143:  38%|███▊      | 2483/6470 [30:50<50:29,  1.32it/s]
4|2484|Loss: 0.7698259949684143:  38%|███▊      | 2484/6470 [30:51<50:47,  1.31it/s]
4|2485|Loss: 1.0547798871994019:  38%|███▊      | 2484/6470 [30:51<50:47,  1.31it/s]
4|2485|Loss: 1.0547798871994019:  38%|███▊      | 2485/6470 [30:51<50:41,  1.31it/s]
4|2486|Loss: 0.6923924088478088:  38%|███▊      | 2485/6470 [30:52<50:41,  1.31it/s]
4|2486|Loss: 0.6923924088478088:  38%|███▊      | 2486/6470 [30:52<50:29,  1.31it/s]
4|2487|Loss: 0.7354021668434143:  38%|███▊      | 2486/6470 [30:53<50:29,  1.31it/s]
4|2487|Loss: 0.7354021668434143:  38%|███▊      | 2487/6470 [30:53<48:07,  1.38it/s]
4|2488|Loss: 0.7537446618080139:  38%|███▊      | 2487/6470 [30:53<48:07,  1.38it/s]
4|2488|Loss: 0.7537446618080139:  38%|███▊      | 2488/6470 [30:53<44:37,  1.49it/s]
4|2489|Loss: 0.8207107186317444:  38%|███▊      | 2488/6470 [30:54<44:37,  1.49it/s]
4|2489|Loss: 0.8207107186317444:  38%|███▊      | 2489/6470 [30:54<43:57,  1.51it/s]
4|2490|Loss: 0.8035341501235962:  38%|███▊      | 2489/6470 [30:54<43:57,  1.51it/s]
4|2490|Loss: 0.8035341501235962:  38%|███▊      | 2490/6470 [30:55<42:31,  1.56it/s]
4|2491|Loss: 0.9050580859184265:  38%|███▊      | 2490/6470 [30:55<42:31,  1.56it/s]
4|2491|Loss: 0.9050580859184265:  39%|███▊      | 2491/6470 [30:55<38:32,  1.72it/s]
4|2492|Loss: 0.7378320693969727:  39%|███▊      | 2491/6470 [30:55<38:32,  1.72it/s]
4|2492|Loss: 0.7378320693969727:  39%|███▊      | 2492/6470 [30:56<37:25,  1.77it/s]
4|2493|Loss: 0.6565386652946472:  39%|███▊      | 2492/6470 [30:56<37:25,  1.77it/s]
4|2493|Loss: 0.6565386652946472:  39%|███▊      | 2493/6470 [30:56<40:43,  1.63it/s]
4|2494|Loss: 0.8136835694313049:  39%|███▊      | 2493/6470 [30:57<40:43,  1.63it/s]
4|2494|Loss: 0.8136835694313049:  39%|███▊      | 2494/6470 [30:57<42:12,  1.57it/s]
4|2495|Loss: 0.8105717897415161:  39%|███▊      | 2494/6470 [30:57<42:12,  1.57it/s]
4|2495|Loss: 0.8105717897415161:  39%|███▊      | 2495/6470 [30:58<44:58,  1.47it/s]
4|2496|Loss: 0.9374728798866272:  39%|███▊      | 2495/6470 [30:58<44:58,  1.47it/s]
4|2496|Loss: 0.9374728798866272:  39%|███▊      | 2496/6470 [30:58<44:00,  1.51it/s]
4|2497|Loss: 0.7738091349601746:  39%|███▊      | 2496/6470 [30:59<44:00,  1.51it/s]
4|2497|Loss: 0.7738091349601746:  39%|███▊      | 2497/6470 [30:59<47:51,  1.38it/s]
4|2498|Loss: 0.7922207117080688:  39%|███▊      | 2497/6470 [31:00<47:51,  1.38it/s]
4|2498|Loss: 0.7922207117080688:  39%|███▊      | 2498/6470 [31:00<47:08,  1.40it/s]
4|2499|Loss: 0.9079247713088989:  39%|███▊      | 2498/6470 [31:00<47:08,  1.40it/s]
4|2499|Loss: 0.9079247713088989:  39%|███▊      | 2499/6470 [31:01<49:09,  1.35it/s]
4|2500|Loss: 0.752850353717804:  39%|███▊      | 2499/6470 [31:01<49:09,  1.35it/s] 
4|2500|Loss: 0.752850353717804:  39%|███▊      | 2500/6470 [31:02<50:15,  1.32it/s]
4|2501|Loss: 0.777603268623352:  39%|███▊      | 2500/6470 [31:02<50:15,  1.32it/s]
4|2501|Loss: 0.777603268623352:  39%|███▊      | 2501/6470 [31:02<51:45,  1.28it/s]
4|2502|Loss: 0.8186136484146118:  39%|███▊      | 2501/6470 [31:03<51:45,  1.28it/s]
4|2502|Loss: 0.8186136484146118:  39%|███▊      | 2502/6470 [31:03<50:14,  1.32it/s]
4|2503|Loss: 0.8132683634757996:  39%|███▊      | 2502/6470 [31:03<50:14,  1.32it/s]
4|2503|Loss: 0.8132683634757996:  39%|███▊      | 2503/6470 [31:04<52:37,  1.26it/s]
4|2504|Loss: 0.6998813152313232:  39%|███▊      | 2503/6470 [31:04<52:37,  1.26it/s]
4|2504|Loss: 0.6998813152313232:  39%|███▊      | 2504/6470 [31:04<47:12,  1.40it/s]
4|2505|Loss: 0.7526431083679199:  39%|███▊      | 2504/6470 [31:05<47:12,  1.40it/s]
4|2505|Loss: 0.7526431083679199:  39%|███▊      | 2505/6470 [31:05<44:30,  1.48it/s]
4|2506|Loss: 0.7600330710411072:  39%|███▊      | 2505/6470 [31:05<44:30,  1.48it/s]
4|2506|Loss: 0.7600330710411072:  39%|███▊      | 2506/6470 [31:06<44:38,  1.48it/s]
4|2507|Loss: 0.7982640862464905:  39%|███▊      | 2506/6470 [31:06<44:38,  1.48it/s]
4|2507|Loss: 0.7982640862464905:  39%|███▊      | 2507/6470 [31:06<42:34,  1.55it/s]
4|2508|Loss: 0.6744095683097839:  39%|███▊      | 2507/6470 [31:07<42:34,  1.55it/s]
4|2508|Loss: 0.6744095683097839:  39%|███▉      | 2508/6470 [31:07<42:21,  1.56it/s]
4|2509|Loss: 0.7269864082336426:  39%|███▉      | 2508/6470 [31:07<42:21,  1.56it/s]
4|2509|Loss: 0.7269864082336426:  39%|███▉      | 2509/6470 [31:08<44:46,  1.47it/s]
4|2510|Loss: 0.8196067214012146:  39%|███▉      | 2509/6470 [31:08<44:46,  1.47it/s]
4|2510|Loss: 0.8196067214012146:  39%|███▉      | 2510/6470 [31:09<49:24,  1.34it/s]
4|2511|Loss: 0.8854947090148926:  39%|███▉      | 2510/6470 [31:09<49:24,  1.34it/s]
4|2511|Loss: 0.8854947090148926:  39%|███▉      | 2511/6470 [31:09<47:10,  1.40it/s]
4|2512|Loss: 0.7027276754379272:  39%|███▉      | 2511/6470 [31:10<47:10,  1.40it/s]
4|2512|Loss: 0.7027276754379272:  39%|███▉      | 2512/6470 [31:10<47:57,  1.38it/s]
4|2513|Loss: 0.824605405330658:  39%|███▉      | 2512/6470 [31:10<47:57,  1.38it/s] 
4|2513|Loss: 0.824605405330658:  39%|███▉      | 2513/6470 [31:11<49:09,  1.34it/s]
4|2514|Loss: 0.7474828958511353:  39%|███▉      | 2513/6470 [31:11<49:09,  1.34it/s]
4|2514|Loss: 0.7474828958511353:  39%|███▉      | 2514/6470 [31:11<48:18,  1.36it/s]
4|2515|Loss: 0.9442025423049927:  39%|███▉      | 2514/6470 [31:12<48:18,  1.36it/s]
4|2515|Loss: 0.9442025423049927:  39%|███▉      | 2515/6470 [31:12<46:30,  1.42it/s]
4|2516|Loss: 0.7495946288108826:  39%|███▉      | 2515/6470 [31:13<46:30,  1.42it/s]
4|2516|Loss: 0.7495946288108826:  39%|███▉      | 2516/6470 [31:13<50:16,  1.31it/s]
4|2517|Loss: 0.7441997528076172:  39%|███▉      | 2516/6470 [31:14<50:16,  1.31it/s]
4|2517|Loss: 0.7441997528076172:  39%|███▉      | 2517/6470 [31:14<52:05,  1.26it/s]
4|2518|Loss: 0.862993597984314:  39%|███▉      | 2517/6470 [31:14<52:05,  1.26it/s] 
4|2518|Loss: 0.862993597984314:  39%|███▉      | 2518/6470 [31:14<48:24,  1.36it/s]
4|2519|Loss: 0.5956994295120239:  39%|███▉      | 2518/6470 [31:15<48:24,  1.36it/s]
4|2519|Loss: 0.5956994295120239:  39%|███▉      | 2519/6470 [31:15<48:39,  1.35it/s]
4|2520|Loss: 0.8480901718139648:  39%|███▉      | 2519/6470 [31:16<48:39,  1.35it/s]
4|2520|Loss: 0.8480901718139648:  39%|███▉      | 2520/6470 [31:16<50:41,  1.30it/s]
4|2521|Loss: 0.7586483359336853:  39%|███▉      | 2520/6470 [31:16<50:41,  1.30it/s]
4|2521|Loss: 0.7586483359336853:  39%|███▉      | 2521/6470 [31:17<49:39,  1.33it/s]
4|2522|Loss: 0.743217945098877:  39%|███▉      | 2521/6470 [31:17<49:39,  1.33it/s] 
4|2522|Loss: 0.743217945098877:  39%|███▉      | 2522/6470 [31:18<52:20,  1.26it/s]
4|2523|Loss: 0.8876203298568726:  39%|███▉      | 2522/6470 [31:18<52:20,  1.26it/s]
4|2523|Loss: 0.8876203298568726:  39%|███▉      | 2523/6470 [31:19<55:02,  1.20it/s]
4|2524|Loss: 0.7221093773841858:  39%|███▉      | 2523/6470 [31:19<55:02,  1.20it/s]
4|2524|Loss: 0.7221093773841858:  39%|███▉      | 2524/6470 [31:19<48:06,  1.37it/s]
4|2525|Loss: 0.7200868725776672:  39%|███▉      | 2524/6470 [31:20<48:06,  1.37it/s]
4|2525|Loss: 0.7200868725776672:  39%|███▉      | 2525/6470 [31:20<51:13,  1.28it/s]
4|2526|Loss: 0.8360804915428162:  39%|███▉      | 2525/6470 [31:20<51:13,  1.28it/s]
4|2526|Loss: 0.8360804915428162:  39%|███▉      | 2526/6470 [31:21<49:44,  1.32it/s]
4|2527|Loss: 0.7326361536979675:  39%|███▉      | 2526/6470 [31:21<49:44,  1.32it/s]
4|2527|Loss: 0.7326361536979675:  39%|███▉      | 2527/6470 [31:21<47:52,  1.37it/s]
4|2528|Loss: 0.8414164185523987:  39%|███▉      | 2527/6470 [31:22<47:52,  1.37it/s]
4|2528|Loss: 0.8414164185523987:  39%|███▉      | 2528/6470 [31:22<46:40,  1.41it/s]
4|2529|Loss: 0.8557156920433044:  39%|███▉      | 2528/6470 [31:22<46:40,  1.41it/s]
4|2529|Loss: 0.8557156920433044:  39%|███▉      | 2529/6470 [31:23<48:52,  1.34it/s]
4|2530|Loss: 0.9977198839187622:  39%|███▉      | 2529/6470 [31:23<48:52,  1.34it/s]
4|2530|Loss: 0.9977198839187622:  39%|███▉      | 2530/6470 [31:24<48:34,  1.35it/s]
4|2531|Loss: 0.84991854429245:  39%|███▉      | 2530/6470 [31:24<48:34,  1.35it/s]  
4|2531|Loss: 0.84991854429245:  39%|███▉      | 2531/6470 [31:24<49:07,  1.34it/s]
4|2532|Loss: 0.7246612906455994:  39%|███▉      | 2531/6470 [31:25<49:07,  1.34it/s]
4|2532|Loss: 0.7246612906455994:  39%|███▉      | 2532/6470 [31:25<48:05,  1.36it/s]
4|2533|Loss: 0.8638198971748352:  39%|███▉      | 2532/6470 [31:25<48:05,  1.36it/s]
4|2533|Loss: 0.8638198971748352:  39%|███▉      | 2533/6470 [31:26<51:26,  1.28it/s]
4|2534|Loss: 0.6739019155502319:  39%|███▉      | 2533/6470 [31:26<51:26,  1.28it/s]
4|2534|Loss: 0.6739019155502319:  39%|███▉      | 2534/6470 [31:27<50:19,  1.30it/s]
4|2535|Loss: 0.8712647557258606:  39%|███▉      | 2534/6470 [31:27<50:19,  1.30it/s]
4|2535|Loss: 0.8712647557258606:  39%|███▉      | 2535/6470 [31:28<51:41,  1.27it/s]
4|2536|Loss: 0.7362114787101746:  39%|███▉      | 2535/6470 [31:28<51:41,  1.27it/s]
4|2536|Loss: 0.7362114787101746:  39%|███▉      | 2536/6470 [31:28<51:33,  1.27it/s]
4|2537|Loss: 0.8524137139320374:  39%|███▉      | 2536/6470 [31:29<51:33,  1.27it/s]
4|2537|Loss: 0.8524137139320374:  39%|███▉      | 2537/6470 [31:29<48:44,  1.34it/s]
4|2538|Loss: 0.9005897045135498:  39%|███▉      | 2537/6470 [31:29<48:44,  1.34it/s]
4|2538|Loss: 0.9005897045135498:  39%|███▉      | 2538/6470 [31:30<47:34,  1.38it/s]
4|2539|Loss: 0.8626947999000549:  39%|███▉      | 2538/6470 [31:30<47:34,  1.38it/s]
4|2539|Loss: 0.8626947999000549:  39%|███▉      | 2539/6470 [31:30<48:02,  1.36it/s]
4|2540|Loss: 0.6312300562858582:  39%|███▉      | 2539/6470 [31:31<48:02,  1.36it/s]
4|2540|Loss: 0.6312300562858582:  39%|███▉      | 2540/6470 [31:31<54:31,  1.20it/s]
4|2541|Loss: 0.8908570408821106:  39%|███▉      | 2540/6470 [31:32<54:31,  1.20it/s]
4|2541|Loss: 0.8908570408821106:  39%|███▉      | 2541/6470 [31:32<53:47,  1.22it/s]
4|2542|Loss: 0.7256405353546143:  39%|███▉      | 2541/6470 [31:33<53:47,  1.22it/s]
4|2542|Loss: 0.7256405353546143:  39%|███▉      | 2542/6470 [31:33<54:15,  1.21it/s]
4|2543|Loss: 0.7949625849723816:  39%|███▉      | 2542/6470 [31:33<54:15,  1.21it/s]
4|2543|Loss: 0.7949625849723816:  39%|███▉      | 2543/6470 [31:34<53:08,  1.23it/s]
4|2544|Loss: 0.8514997363090515:  39%|███▉      | 2543/6470 [31:34<53:08,  1.23it/s]
4|2544|Loss: 0.8514997363090515:  39%|███▉      | 2544/6470 [31:34<49:57,  1.31it/s]
4|2545|Loss: 0.8222445845603943:  39%|███▉      | 2544/6470 [31:35<49:57,  1.31it/s]
4|2545|Loss: 0.8222445845603943:  39%|███▉      | 2545/6470 [31:35<52:32,  1.24it/s]
4|2546|Loss: 0.9151063561439514:  39%|███▉      | 2545/6470 [31:36<52:32,  1.24it/s]
4|2546|Loss: 0.9151063561439514:  39%|███▉      | 2546/6470 [31:36<55:03,  1.19it/s]
4|2547|Loss: 0.7827670574188232:  39%|███▉      | 2546/6470 [31:37<55:03,  1.19it/s]
4|2547|Loss: 0.7827670574188232:  39%|███▉      | 2547/6470 [31:37<52:51,  1.24it/s]
4|2548|Loss: 0.8203566670417786:  39%|███▉      | 2547/6470 [31:37<52:51,  1.24it/s]
4|2548|Loss: 0.8203566670417786:  39%|███▉      | 2548/6470 [31:38<51:26,  1.27it/s]
4|2549|Loss: 0.9932191371917725:  39%|███▉      | 2548/6470 [31:38<51:26,  1.27it/s]
4|2549|Loss: 0.9932191371917725:  39%|███▉      | 2549/6470 [31:39<51:10,  1.28it/s]
4|2550|Loss: 0.7006568908691406:  39%|███▉      | 2549/6470 [31:39<51:10,  1.28it/s]
4|2550|Loss: 0.7006568908691406:  39%|███▉      | 2550/6470 [31:39<51:30,  1.27it/s]
4|2551|Loss: 0.8820849061012268:  39%|███▉      | 2550/6470 [31:40<51:30,  1.27it/s]
4|2551|Loss: 0.8820849061012268:  39%|███▉      | 2551/6470 [31:40<51:32,  1.27it/s]
4|2552|Loss: 0.8163671493530273:  39%|███▉      | 2551/6470 [31:41<51:32,  1.27it/s]
4|2552|Loss: 0.8163671493530273:  39%|███▉      | 2552/6470 [31:41<49:15,  1.33it/s]
4|2553|Loss: 0.7523868680000305:  39%|███▉      | 2552/6470 [31:41<49:15,  1.33it/s]
4|2553|Loss: 0.7523868680000305:  39%|███▉      | 2553/6470 [31:42<48:57,  1.33it/s]
4|2554|Loss: 0.8136857748031616:  39%|███▉      | 2553/6470 [31:42<48:57,  1.33it/s]
4|2554|Loss: 0.8136857748031616:  39%|███▉      | 2554/6470 [31:42<46:02,  1.42it/s]
4|2555|Loss: 0.8776251673698425:  39%|███▉      | 2554/6470 [31:43<46:02,  1.42it/s]
4|2555|Loss: 0.8776251673698425:  39%|███▉      | 2555/6470 [31:43<48:01,  1.36it/s]
4|2556|Loss: 0.7863628268241882:  39%|███▉      | 2555/6470 [31:43<48:01,  1.36it/s]
4|2556|Loss: 0.7863628268241882:  40%|███▉      | 2556/6470 [31:44<51:47,  1.26it/s]
4|2557|Loss: 0.742088794708252:  40%|███▉      | 2556/6470 [31:44<51:47,  1.26it/s] 
4|2557|Loss: 0.742088794708252:  40%|███▉      | 2557/6470 [31:45<54:00,  1.21it/s]
4|2558|Loss: 0.7852988839149475:  40%|███▉      | 2557/6470 [31:45<54:00,  1.21it/s]
4|2558|Loss: 0.7852988839149475:  40%|███▉      | 2558/6470 [31:45<50:35,  1.29it/s]
4|2559|Loss: 0.8975014090538025:  40%|███▉      | 2558/6470 [31:46<50:35,  1.29it/s]
4|2559|Loss: 0.8975014090538025:  40%|███▉      | 2559/6470 [31:46<48:22,  1.35it/s]
4|2560|Loss: 0.8334222435951233:  40%|███▉      | 2559/6470 [31:47<48:22,  1.35it/s]
4|2560|Loss: 0.8334222435951233:  40%|███▉      | 2560/6470 [31:47<47:17,  1.38it/s]
4|2561|Loss: 0.8160392045974731:  40%|███▉      | 2560/6470 [31:47<47:17,  1.38it/s]
4|2561|Loss: 0.8160392045974731:  40%|███▉      | 2561/6470 [31:48<49:15,  1.32it/s]
4|2562|Loss: 0.875458300113678:  40%|███▉      | 2561/6470 [31:48<49:15,  1.32it/s] 
4|2562|Loss: 0.875458300113678:  40%|███▉      | 2562/6470 [31:49<52:07,  1.25it/s]
4|2563|Loss: 0.9144154787063599:  40%|███▉      | 2562/6470 [31:49<52:07,  1.25it/s]
4|2563|Loss: 0.9144154787063599:  40%|███▉      | 2563/6470 [31:49<49:02,  1.33it/s]
4|2564|Loss: 0.8043147921562195:  40%|███▉      | 2563/6470 [31:50<49:02,  1.33it/s]
4|2564|Loss: 0.8043147921562195:  40%|███▉      | 2564/6470 [31:50<49:02,  1.33it/s]
4|2565|Loss: 0.8122122883796692:  40%|███▉      | 2564/6470 [31:50<49:02,  1.33it/s]
4|2565|Loss: 0.8122122883796692:  40%|███▉      | 2565/6470 [31:51<51:03,  1.27it/s]
4|2566|Loss: 0.6827914118766785:  40%|███▉      | 2565/6470 [31:51<51:03,  1.27it/s]
4|2566|Loss: 0.6827914118766785:  40%|███▉      | 2566/6470 [31:51<45:02,  1.44it/s]
4|2567|Loss: 0.9043145775794983:  40%|███▉      | 2566/6470 [31:52<45:02,  1.44it/s]
4|2567|Loss: 0.9043145775794983:  40%|███▉      | 2567/6470 [31:52<45:22,  1.43it/s]
4|2568|Loss: 0.8703908920288086:  40%|███▉      | 2567/6470 [31:52<45:22,  1.43it/s]
4|2568|Loss: 0.8703908920288086:  40%|███▉      | 2568/6470 [31:53<49:02,  1.33it/s]
4|2569|Loss: 0.7094289660453796:  40%|███▉      | 2568/6470 [31:53<49:02,  1.33it/s]
4|2569|Loss: 0.7094289660453796:  40%|███▉      | 2569/6470 [31:54<53:21,  1.22it/s]
4|2570|Loss: 0.957428514957428:  40%|███▉      | 2569/6470 [31:54<53:21,  1.22it/s] 
4|2570|Loss: 0.957428514957428:  40%|███▉      | 2570/6470 [31:55<55:28,  1.17it/s]
4|2571|Loss: 0.8944857120513916:  40%|███▉      | 2570/6470 [31:55<55:28,  1.17it/s]
4|2571|Loss: 0.8944857120513916:  40%|███▉      | 2571/6470 [31:56<53:14,  1.22it/s]
4|2572|Loss: 0.7256959080696106:  40%|███▉      | 2571/6470 [31:56<53:14,  1.22it/s]
4|2572|Loss: 0.7256959080696106:  40%|███▉      | 2572/6470 [31:56<48:13,  1.35it/s]
4|2573|Loss: 0.7767626643180847:  40%|███▉      | 2572/6470 [31:56<48:13,  1.35it/s]
4|2573|Loss: 0.7767626643180847:  40%|███▉      | 2573/6470 [31:57<45:20,  1.43it/s]
4|2574|Loss: 0.6931795477867126:  40%|███▉      | 2573/6470 [31:57<45:20,  1.43it/s]
4|2574|Loss: 0.6931795477867126:  40%|███▉      | 2574/6470 [31:57<41:36,  1.56it/s]
4|2575|Loss: 0.8402474522590637:  40%|███▉      | 2574/6470 [31:58<41:36,  1.56it/s]
4|2575|Loss: 0.8402474522590637:  40%|███▉      | 2575/6470 [31:58<43:06,  1.51it/s]
4|2576|Loss: 0.935724675655365:  40%|███▉      | 2575/6470 [31:58<43:06,  1.51it/s] 
4|2576|Loss: 0.935724675655365:  40%|███▉      | 2576/6470 [31:59<45:38,  1.42it/s]
4|2577|Loss: 0.9456064105033875:  40%|███▉      | 2576/6470 [31:59<45:38,  1.42it/s]
4|2577|Loss: 0.9456064105033875:  40%|███▉      | 2577/6470 [31:59<44:37,  1.45it/s]
4|2578|Loss: 0.8771663308143616:  40%|███▉      | 2577/6470 [32:00<44:37,  1.45it/s]
4|2578|Loss: 0.8771663308143616:  40%|███▉      | 2578/6470 [32:00<44:19,  1.46it/s]
4|2579|Loss: 0.8305701017379761:  40%|███▉      | 2578/6470 [32:00<44:19,  1.46it/s]
4|2579|Loss: 0.8305701017379761:  40%|███▉      | 2579/6470 [32:01<48:05,  1.35it/s]
4|2580|Loss: 0.8813244700431824:  40%|███▉      | 2579/6470 [32:01<48:05,  1.35it/s]
4|2580|Loss: 0.8813244700431824:  40%|███▉      | 2580/6470 [32:01<44:59,  1.44it/s]
4|2581|Loss: 0.7947913408279419:  40%|███▉      | 2580/6470 [32:02<44:59,  1.44it/s]
4|2581|Loss: 0.7947913408279419:  40%|███▉      | 2581/6470 [32:02<43:49,  1.48it/s]
4|2582|Loss: 0.9011470079421997:  40%|███▉      | 2581/6470 [32:03<43:49,  1.48it/s]
4|2582|Loss: 0.9011470079421997:  40%|███▉      | 2582/6470 [32:03<47:56,  1.35it/s]
4|2583|Loss: 0.8666675686836243:  40%|███▉      | 2582/6470 [32:04<47:56,  1.35it/s]
4|2583|Loss: 0.8666675686836243:  40%|███▉      | 2583/6470 [32:04<51:49,  1.25it/s]
4|2584|Loss: 0.900445818901062:  40%|███▉      | 2583/6470 [32:04<51:49,  1.25it/s] 
4|2584|Loss: 0.900445818901062:  40%|███▉      | 2584/6470 [32:05<52:34,  1.23it/s]
4|2585|Loss: 0.9471758604049683:  40%|███▉      | 2584/6470 [32:05<52:34,  1.23it/s]
4|2585|Loss: 0.9471758604049683:  40%|███▉      | 2585/6470 [32:06<52:30,  1.23it/s]
4|2586|Loss: 0.8532141447067261:  40%|███▉      | 2585/6470 [32:06<52:30,  1.23it/s]
4|2586|Loss: 0.8532141447067261:  40%|███▉      | 2586/6470 [32:06<50:23,  1.28it/s]
4|2587|Loss: 0.8221306800842285:  40%|███▉      | 2586/6470 [32:07<50:23,  1.28it/s]
4|2587|Loss: 0.8221306800842285:  40%|███▉      | 2587/6470 [32:07<48:35,  1.33it/s]
4|2588|Loss: 0.7811102271080017:  40%|███▉      | 2587/6470 [32:07<48:35,  1.33it/s]
4|2588|Loss: 0.7811102271080017:  40%|████      | 2588/6470 [32:08<46:12,  1.40it/s]
4|2589|Loss: 0.8634262084960938:  40%|████      | 2588/6470 [32:08<46:12,  1.40it/s]
4|2589|Loss: 0.8634262084960938:  40%|████      | 2589/6470 [32:08<40:50,  1.58it/s]
4|2590|Loss: 0.8639461994171143:  40%|████      | 2589/6470 [32:08<40:50,  1.58it/s]
4|2590|Loss: 0.8639461994171143:  40%|████      | 2590/6470 [32:09<42:34,  1.52it/s]
4|2591|Loss: 0.9872321486473083:  40%|████      | 2590/6470 [32:09<42:34,  1.52it/s]
4|2591|Loss: 0.9872321486473083:  40%|████      | 2591/6470 [32:10<45:15,  1.43it/s]
4|2592|Loss: 0.876424252986908:  40%|████      | 2591/6470 [32:10<45:15,  1.43it/s] 
4|2592|Loss: 0.876424252986908:  40%|████      | 2592/6470 [32:10<49:02,  1.32it/s]
4|2593|Loss: 0.7234876155853271:  40%|████      | 2592/6470 [32:11<49:02,  1.32it/s]
4|2593|Loss: 0.7234876155853271:  40%|████      | 2593/6470 [32:11<50:12,  1.29it/s]
4|2594|Loss: 0.8266159892082214:  40%|████      | 2593/6470 [32:12<50:12,  1.29it/s]
4|2594|Loss: 0.8266159892082214:  40%|████      | 2594/6470 [32:12<47:43,  1.35it/s]
4|2595|Loss: 0.7694756388664246:  40%|████      | 2594/6470 [32:12<47:43,  1.35it/s]
4|2595|Loss: 0.7694756388664246:  40%|████      | 2595/6470 [32:13<45:10,  1.43it/s]
4|2596|Loss: 0.8097375631332397:  40%|████      | 2595/6470 [32:13<45:10,  1.43it/s]
4|2596|Loss: 0.8097375631332397:  40%|████      | 2596/6470 [32:13<43:34,  1.48it/s]
4|2597|Loss: 0.8047747611999512:  40%|████      | 2596/6470 [32:14<43:34,  1.48it/s]
4|2597|Loss: 0.8047747611999512:  40%|████      | 2597/6470 [32:14<45:14,  1.43it/s]
4|2598|Loss: 0.9071511626243591:  40%|████      | 2597/6470 [32:14<45:14,  1.43it/s]
4|2598|Loss: 0.9071511626243591:  40%|████      | 2598/6470 [32:15<48:06,  1.34it/s]
4|2599|Loss: 0.8617207407951355:  40%|████      | 2598/6470 [32:15<48:06,  1.34it/s]
4|2599|Loss: 0.8617207407951355:  40%|████      | 2599/6470 [32:16<51:59,  1.24it/s]
4|2600|Loss: 0.7042218446731567:  40%|████      | 2599/6470 [32:16<51:59,  1.24it/s]
4|2600|Loss: 0.7042218446731567:  40%|████      | 2600/6470 [32:17<53:51,  1.20it/s]
4|2601|Loss: 0.8056291341781616:  40%|████      | 2600/6470 [32:17<53:51,  1.20it/s]
4|2601|Loss: 0.8056291341781616:  40%|████      | 2601/6470 [32:17<53:28,  1.21it/s]
4|2602|Loss: 0.8636050820350647:  40%|████      | 2601/6470 [32:18<53:28,  1.21it/s]
4|2602|Loss: 0.8636050820350647:  40%|████      | 2602/6470 [32:18<51:17,  1.26it/s]
4|2603|Loss: 0.948198676109314:  40%|████      | 2602/6470 [32:19<51:17,  1.26it/s] 
4|2603|Loss: 0.948198676109314:  40%|████      | 2603/6470 [32:19<49:02,  1.31it/s]
4|2604|Loss: 0.8648128509521484:  40%|████      | 2603/6470 [32:19<49:02,  1.31it/s]
4|2604|Loss: 0.8648128509521484:  40%|████      | 2604/6470 [32:20<50:02,  1.29it/s]
4|2605|Loss: 0.9417228102684021:  40%|████      | 2604/6470 [32:20<50:02,  1.29it/s]
4|2605|Loss: 0.9417228102684021:  40%|████      | 2605/6470 [32:20<48:42,  1.32it/s]
4|2606|Loss: 0.7193130850791931:  40%|████      | 2605/6470 [32:21<48:42,  1.32it/s]
4|2606|Loss: 0.7193130850791931:  40%|████      | 2606/6470 [32:21<46:13,  1.39it/s]
4|2607|Loss: 0.7652928829193115:  40%|████      | 2606/6470 [32:21<46:13,  1.39it/s]
4|2607|Loss: 0.7652928829193115:  40%|████      | 2607/6470 [32:22<47:08,  1.37it/s]
4|2608|Loss: 1.0793850421905518:  40%|████      | 2607/6470 [32:22<47:08,  1.37it/s]
4|2608|Loss: 1.0793850421905518:  40%|████      | 2608/6470 [32:22<45:09,  1.43it/s]
4|2609|Loss: 0.7531368136405945:  40%|████      | 2608/6470 [32:23<45:09,  1.43it/s]
4|2609|Loss: 0.7531368136405945:  40%|████      | 2609/6470 [32:23<43:58,  1.46it/s]
4|2610|Loss: 0.7605393528938293:  40%|████      | 2609/6470 [32:23<43:58,  1.46it/s]
4|2610|Loss: 0.7605393528938293:  40%|████      | 2610/6470 [32:24<45:17,  1.42it/s]
4|2611|Loss: 0.8231877684593201:  40%|████      | 2610/6470 [32:24<45:17,  1.42it/s]
4|2611|Loss: 0.8231877684593201:  40%|████      | 2611/6470 [32:24<43:27,  1.48it/s]
4|2612|Loss: 0.857991099357605:  40%|████      | 2611/6470 [32:25<43:27,  1.48it/s] 
4|2612|Loss: 0.857991099357605:  40%|████      | 2612/6470 [32:25<44:39,  1.44it/s]
4|2613|Loss: 0.885818600654602:  40%|████      | 2612/6470 [32:26<44:39,  1.44it/s]
4|2613|Loss: 0.885818600654602:  40%|████      | 2613/6470 [32:26<45:33,  1.41it/s]
4|2614|Loss: 0.7067122459411621:  40%|████      | 2613/6470 [32:26<45:33,  1.41it/s]
4|2614|Loss: 0.7067122459411621:  40%|████      | 2614/6470 [32:27<48:30,  1.32it/s]
4|2615|Loss: 0.8898617029190063:  40%|████      | 2614/6470 [32:27<48:30,  1.32it/s]
4|2615|Loss: 0.8898617029190063:  40%|████      | 2615/6470 [32:27<47:35,  1.35it/s]
4|2616|Loss: 0.9034838676452637:  40%|████      | 2615/6470 [32:28<47:35,  1.35it/s]
4|2616|Loss: 0.9034838676452637:  40%|████      | 2616/6470 [32:28<51:30,  1.25it/s]
4|2617|Loss: 0.8849964141845703:  40%|████      | 2616/6470 [32:29<51:30,  1.25it/s]
4|2617|Loss: 0.8849964141845703:  40%|████      | 2617/6470 [32:29<54:47,  1.17it/s]
4|2618|Loss: 0.7722479104995728:  40%|████      | 2617/6470 [32:30<54:47,  1.17it/s]
4|2618|Loss: 0.7722479104995728:  40%|████      | 2618/6470 [32:30<52:36,  1.22it/s]
4|2619|Loss: 0.9443938732147217:  40%|████      | 2618/6470 [32:30<52:36,  1.22it/s]
4|2619|Loss: 0.9443938732147217:  40%|████      | 2619/6470 [32:31<50:11,  1.28it/s]
4|2620|Loss: 0.939449667930603:  40%|████      | 2619/6470 [32:31<50:11,  1.28it/s] 
4|2620|Loss: 0.939449667930603:  40%|████      | 2620/6470 [32:32<51:40,  1.24it/s]
4|2621|Loss: 0.783163845539093:  40%|████      | 2620/6470 [32:32<51:40,  1.24it/s]
4|2621|Loss: 0.783163845539093:  41%|████      | 2621/6470 [32:32<48:41,  1.32it/s]
4|2622|Loss: 0.760250449180603:  41%|████      | 2621/6470 [32:33<48:41,  1.32it/s]
4|2622|Loss: 0.760250449180603:  41%|████      | 2622/6470 [32:33<47:57,  1.34it/s]
4|2623|Loss: 0.8166214823722839:  41%|████      | 2622/6470 [32:33<47:57,  1.34it/s]
4|2623|Loss: 0.8166214823722839:  41%|████      | 2623/6470 [32:34<48:35,  1.32it/s]
4|2624|Loss: 0.7930346727371216:  41%|████      | 2623/6470 [32:34<48:35,  1.32it/s]
4|2624|Loss: 0.7930346727371216:  41%|████      | 2624/6470 [32:34<44:26,  1.44it/s]
4|2625|Loss: 0.6471037268638611:  41%|████      | 2624/6470 [32:35<44:26,  1.44it/s]
4|2625|Loss: 0.6471037268638611:  41%|████      | 2625/6470 [32:35<42:09,  1.52it/s]
4|2626|Loss: 0.9206341505050659:  41%|████      | 2625/6470 [32:35<42:09,  1.52it/s]
4|2626|Loss: 0.9206341505050659:  41%|████      | 2626/6470 [32:35<40:03,  1.60it/s]
4|2627|Loss: 0.8400079011917114:  41%|████      | 2626/6470 [32:36<40:03,  1.60it/s]
4|2627|Loss: 0.8400079011917114:  41%|████      | 2627/6470 [32:36<42:00,  1.52it/s]
4|2628|Loss: 0.8771514892578125:  41%|████      | 2627/6470 [32:37<42:00,  1.52it/s]
4|2628|Loss: 0.8771514892578125:  41%|████      | 2628/6470 [32:37<44:48,  1.43it/s]
4|2629|Loss: 0.8894198536872864:  41%|████      | 2628/6470 [32:37<44:48,  1.43it/s]
4|2629|Loss: 0.8894198536872864:  41%|████      | 2629/6470 [32:38<44:43,  1.43it/s]
4|2630|Loss: 1.054399847984314:  41%|████      | 2629/6470 [32:38<44:43,  1.43it/s] 
4|2630|Loss: 1.054399847984314:  41%|████      | 2630/6470 [32:39<47:01,  1.36it/s]
4|2631|Loss: 0.8544666767120361:  41%|████      | 2630/6470 [32:39<47:01,  1.36it/s]
4|2631|Loss: 0.8544666767120361:  41%|████      | 2631/6470 [32:39<45:14,  1.41it/s]
4|2632|Loss: 0.7871882915496826:  41%|████      | 2631/6470 [32:40<45:14,  1.41it/s]
4|2632|Loss: 0.7871882915496826:  41%|████      | 2632/6470 [32:40<45:54,  1.39it/s]
4|2633|Loss: 0.8954652547836304:  41%|████      | 2632/6470 [32:40<45:54,  1.39it/s]
4|2633|Loss: 0.8954652547836304:  41%|████      | 2633/6470 [32:41<45:45,  1.40it/s]
4|2634|Loss: 0.8102497458457947:  41%|████      | 2633/6470 [32:41<45:45,  1.40it/s]
4|2634|Loss: 0.8102497458457947:  41%|████      | 2634/6470 [32:41<44:30,  1.44it/s]
4|2635|Loss: 0.8447103500366211:  41%|████      | 2634/6470 [32:42<44:30,  1.44it/s]
4|2635|Loss: 0.8447103500366211:  41%|████      | 2635/6470 [32:42<46:38,  1.37it/s]
4|2636|Loss: 0.9236735701560974:  41%|████      | 2635/6470 [32:42<46:38,  1.37it/s]
4|2636|Loss: 0.9236735701560974:  41%|████      | 2636/6470 [32:43<46:14,  1.38it/s]
4|2637|Loss: 0.728133499622345:  41%|████      | 2636/6470 [32:43<46:14,  1.38it/s] 
4|2637|Loss: 0.728133499622345:  41%|████      | 2637/6470 [32:44<47:20,  1.35it/s]
4|2638|Loss: 0.746854305267334:  41%|████      | 2637/6470 [32:44<47:20,  1.35it/s]
4|2638|Loss: 0.746854305267334:  41%|████      | 2638/6470 [32:44<43:57,  1.45it/s]
4|2639|Loss: 0.8150307536125183:  41%|████      | 2638/6470 [32:44<43:57,  1.45it/s]
4|2639|Loss: 0.8150307536125183:  41%|████      | 2639/6470 [32:45<43:12,  1.48it/s]
4|2640|Loss: 0.772402286529541:  41%|████      | 2639/6470 [32:45<43:12,  1.48it/s] 
4|2640|Loss: 0.772402286529541:  41%|████      | 2640/6470 [32:46<47:44,  1.34it/s]
4|2641|Loss: 0.7904266119003296:  41%|████      | 2640/6470 [32:46<47:44,  1.34it/s]
4|2641|Loss: 0.7904266119003296:  41%|████      | 2641/6470 [32:46<48:56,  1.30it/s]
4|2642|Loss: 0.858155369758606:  41%|████      | 2641/6470 [32:47<48:56,  1.30it/s] 
4|2642|Loss: 0.858155369758606:  41%|████      | 2642/6470 [32:47<50:14,  1.27it/s]
4|2643|Loss: 0.8644008636474609:  41%|████      | 2642/6470 [32:48<50:14,  1.27it/s]
4|2643|Loss: 0.8644008636474609:  41%|████      | 2643/6470 [32:48<52:20,  1.22it/s]
4|2644|Loss: 0.739952802658081:  41%|████      | 2643/6470 [32:49<52:20,  1.22it/s] 
4|2644|Loss: 0.739952802658081:  41%|████      | 2644/6470 [32:49<50:09,  1.27it/s]
4|2645|Loss: 0.8404684662818909:  41%|████      | 2644/6470 [32:49<50:09,  1.27it/s]
4|2645|Loss: 0.8404684662818909:  41%|████      | 2645/6470 [32:49<44:58,  1.42it/s]
4|2646|Loss: 0.6780513525009155:  41%|████      | 2645/6470 [32:50<44:58,  1.42it/s]
4|2646|Loss: 0.6780513525009155:  41%|████      | 2646/6470 [32:50<47:47,  1.33it/s]
4|2647|Loss: 0.8785955905914307:  41%|████      | 2646/6470 [32:51<47:47,  1.33it/s]
4|2647|Loss: 0.8785955905914307:  41%|████      | 2647/6470 [32:51<47:20,  1.35it/s]
4|2648|Loss: 0.8437676429748535:  41%|████      | 2647/6470 [32:51<47:20,  1.35it/s]
4|2648|Loss: 0.8437676429748535:  41%|████      | 2648/6470 [32:52<43:52,  1.45it/s]
4|2649|Loss: 0.7886615991592407:  41%|████      | 2648/6470 [32:52<43:52,  1.45it/s]
4|2649|Loss: 0.7886615991592407:  41%|████      | 2649/6470 [32:52<44:15,  1.44it/s]
4|2650|Loss: 0.819595456123352:  41%|████      | 2649/6470 [32:53<44:15,  1.44it/s] 
4|2650|Loss: 0.819595456123352:  41%|████      | 2650/6470 [32:53<45:58,  1.38it/s]
4|2651|Loss: 0.7739056944847107:  41%|████      | 2650/6470 [32:53<45:58,  1.38it/s]
4|2651|Loss: 0.7739056944847107:  41%|████      | 2651/6470 [32:54<43:56,  1.45it/s]
4|2652|Loss: 0.9054567217826843:  41%|████      | 2651/6470 [32:54<43:56,  1.45it/s]
4|2652|Loss: 0.9054567217826843:  41%|████      | 2652/6470 [32:54<38:36,  1.65it/s]
4|2653|Loss: 0.8226377367973328:  41%|████      | 2652/6470 [32:54<38:36,  1.65it/s]
4|2653|Loss: 0.8226377367973328:  41%|████      | 2653/6470 [32:55<38:40,  1.64it/s]
4|2654|Loss: 0.7304880023002625:  41%|████      | 2653/6470 [32:55<38:40,  1.64it/s]
4|2654|Loss: 0.7304880023002625:  41%|████      | 2654/6470 [32:55<39:50,  1.60it/s]
4|2655|Loss: 0.5840744972229004:  41%|████      | 2654/6470 [32:56<39:50,  1.60it/s]
4|2655|Loss: 0.5840744972229004:  41%|████      | 2655/6470 [32:56<45:49,  1.39it/s]
4|2656|Loss: 0.8445786237716675:  41%|████      | 2655/6470 [32:57<45:49,  1.39it/s]
4|2656|Loss: 0.8445786237716675:  41%|████      | 2656/6470 [32:57<48:17,  1.32it/s]
4|2657|Loss: 0.8902220726013184:  41%|████      | 2656/6470 [32:58<48:17,  1.32it/s]
4|2657|Loss: 0.8902220726013184:  41%|████      | 2657/6470 [32:58<46:42,  1.36it/s]
4|2658|Loss: 0.8549185991287231:  41%|████      | 2657/6470 [32:58<46:42,  1.36it/s]
4|2658|Loss: 0.8549185991287231:  41%|████      | 2658/6470 [32:58<43:46,  1.45it/s]
4|2659|Loss: 1.0811245441436768:  41%|████      | 2658/6470 [32:59<43:46,  1.45it/s]
4|2659|Loss: 1.0811245441436768:  41%|████      | 2659/6470 [32:59<43:31,  1.46it/s]
4|2660|Loss: 0.7130205035209656:  41%|████      | 2659/6470 [33:00<43:31,  1.46it/s]
4|2660|Loss: 0.7130205035209656:  41%|████      | 2660/6470 [33:00<44:55,  1.41it/s]
4|2661|Loss: 0.6768850088119507:  41%|████      | 2660/6470 [33:00<44:55,  1.41it/s]
4|2661|Loss: 0.6768850088119507:  41%|████      | 2661/6470 [33:01<44:28,  1.43it/s]
4|2662|Loss: 0.8883328437805176:  41%|████      | 2661/6470 [33:01<44:28,  1.43it/s]
4|2662|Loss: 0.8883328437805176:  41%|████      | 2662/6470 [33:01<46:07,  1.38it/s]
4|2663|Loss: 0.8797128200531006:  41%|████      | 2662/6470 [33:02<46:07,  1.38it/s]
4|2663|Loss: 0.8797128200531006:  41%|████      | 2663/6470 [33:02<48:35,  1.31it/s]
4|2664|Loss: 0.6844900846481323:  41%|████      | 2663/6470 [33:03<48:35,  1.31it/s]
4|2664|Loss: 0.6844900846481323:  41%|████      | 2664/6470 [33:03<51:29,  1.23it/s]
4|2665|Loss: 0.8095640540122986:  41%|████      | 2664/6470 [33:04<51:29,  1.23it/s]
4|2665|Loss: 0.8095640540122986:  41%|████      | 2665/6470 [33:04<51:53,  1.22it/s]
4|2666|Loss: 0.8387815356254578:  41%|████      | 2665/6470 [33:04<51:53,  1.22it/s]
4|2666|Loss: 0.8387815356254578:  41%|████      | 2666/6470 [33:05<51:39,  1.23it/s]
4|2667|Loss: 0.7602846026420593:  41%|████      | 2666/6470 [33:05<51:39,  1.23it/s]
4|2667|Loss: 0.7602846026420593:  41%|████      | 2667/6470 [33:06<51:28,  1.23it/s]
4|2668|Loss: 0.8001100420951843:  41%|████      | 2667/6470 [33:06<51:28,  1.23it/s]
4|2668|Loss: 0.8001100420951843:  41%|████      | 2668/6470 [33:06<52:42,  1.20it/s]
4|2669|Loss: 0.8329395055770874:  41%|████      | 2668/6470 [33:07<52:42,  1.20it/s]
4|2669|Loss: 0.8329395055770874:  41%|████▏     | 2669/6470 [33:07<50:59,  1.24it/s]
4|2670|Loss: 0.8647141456604004:  41%|████▏     | 2669/6470 [33:08<50:59,  1.24it/s]
4|2670|Loss: 0.8647141456604004:  41%|████▏     | 2670/6470 [33:08<49:11,  1.29it/s]
4|2671|Loss: 0.8588658571243286:  41%|████▏     | 2670/6470 [33:08<49:11,  1.29it/s]
4|2671|Loss: 0.8588658571243286:  41%|████▏     | 2671/6470 [33:08<45:18,  1.40it/s]
4|2672|Loss: 0.8718485236167908:  41%|████▏     | 2671/6470 [33:09<45:18,  1.40it/s]
4|2672|Loss: 0.8718485236167908:  41%|████▏     | 2672/6470 [33:09<47:36,  1.33it/s]
4|2673|Loss: 0.9776809215545654:  41%|████▏     | 2672/6470 [33:10<47:36,  1.33it/s]
4|2673|Loss: 0.9776809215545654:  41%|████▏     | 2673/6470 [33:10<47:19,  1.34it/s]
4|2674|Loss: 0.8004482388496399:  41%|████▏     | 2673/6470 [33:10<47:19,  1.34it/s]
4|2674|Loss: 0.8004482388496399:  41%|████▏     | 2674/6470 [33:11<47:43,  1.33it/s]
4|2675|Loss: 0.8366995453834534:  41%|████▏     | 2674/6470 [33:11<47:43,  1.33it/s]
4|2675|Loss: 0.8366995453834534:  41%|████▏     | 2675/6470 [33:12<47:01,  1.35it/s]
4|2676|Loss: 0.7525758743286133:  41%|████▏     | 2675/6470 [33:12<47:01,  1.35it/s]
4|2676|Loss: 0.7525758743286133:  41%|████▏     | 2676/6470 [33:12<44:30,  1.42it/s]
4|2677|Loss: 0.8136458992958069:  41%|████▏     | 2676/6470 [33:13<44:30,  1.42it/s]
4|2677|Loss: 0.8136458992958069:  41%|████▏     | 2677/6470 [33:13<46:17,  1.37it/s]
4|2678|Loss: 0.7679800987243652:  41%|████▏     | 2677/6470 [33:13<46:17,  1.37it/s]
4|2678|Loss: 0.7679800987243652:  41%|████▏     | 2678/6470 [33:14<47:48,  1.32it/s]
4|2679|Loss: 0.8389807343482971:  41%|████▏     | 2678/6470 [33:14<47:48,  1.32it/s]
4|2679|Loss: 0.8389807343482971:  41%|████▏     | 2679/6470 [33:15<48:59,  1.29it/s]
4|2680|Loss: 0.9757292866706848:  41%|████▏     | 2679/6470 [33:15<48:59,  1.29it/s]
4|2680|Loss: 0.9757292866706848:  41%|████▏     | 2680/6470 [33:15<48:56,  1.29it/s]
4|2681|Loss: 0.7838860750198364:  41%|████▏     | 2680/6470 [33:16<48:56,  1.29it/s]
4|2681|Loss: 0.7838860750198364:  41%|████▏     | 2681/6470 [33:16<49:31,  1.27it/s]
4|2682|Loss: 0.8417304158210754:  41%|████▏     | 2681/6470 [33:17<49:31,  1.27it/s]
4|2682|Loss: 0.8417304158210754:  41%|████▏     | 2682/6470 [33:17<47:22,  1.33it/s]
4|2683|Loss: 0.7014586329460144:  41%|████▏     | 2682/6470 [33:17<47:22,  1.33it/s]
4|2683|Loss: 0.7014586329460144:  41%|████▏     | 2683/6470 [33:18<47:30,  1.33it/s]
4|2684|Loss: 0.9355469942092896:  41%|████▏     | 2683/6470 [33:18<47:30,  1.33it/s]
4|2684|Loss: 0.9355469942092896:  41%|████▏     | 2684/6470 [33:18<44:48,  1.41it/s]
4|2685|Loss: 0.7347274422645569:  41%|████▏     | 2684/6470 [33:19<44:48,  1.41it/s]
4|2685|Loss: 0.7347274422645569:  41%|████▏     | 2685/6470 [33:19<45:51,  1.38it/s]
4|2686|Loss: 0.8865830898284912:  41%|████▏     | 2685/6470 [33:19<45:51,  1.38it/s]
4|2686|Loss: 0.8865830898284912:  42%|████▏     | 2686/6470 [33:20<49:07,  1.28it/s]
4|2687|Loss: 0.8018752336502075:  42%|████▏     | 2686/6470 [33:20<49:07,  1.28it/s]
4|2687|Loss: 0.8018752336502075:  42%|████▏     | 2687/6470 [33:21<51:42,  1.22it/s]
4|2688|Loss: 0.7969303131103516:  42%|████▏     | 2687/6470 [33:21<51:42,  1.22it/s]
4|2688|Loss: 0.7969303131103516:  42%|████▏     | 2688/6470 [33:22<52:28,  1.20it/s]
4|2689|Loss: 0.7521083354949951:  42%|████▏     | 2688/6470 [33:22<52:28,  1.20it/s]
4|2689|Loss: 0.7521083354949951:  42%|████▏     | 2689/6470 [33:22<49:24,  1.28it/s]
4|2690|Loss: 0.7683063745498657:  42%|████▏     | 2689/6470 [33:23<49:24,  1.28it/s]
4|2690|Loss: 0.7683063745498657:  42%|████▏     | 2690/6470 [33:23<43:00,  1.47it/s]
4|2691|Loss: 0.9696788787841797:  42%|████▏     | 2690/6470 [33:23<43:00,  1.47it/s]
4|2691|Loss: 0.9696788787841797:  42%|████▏     | 2691/6470 [33:23<41:48,  1.51it/s]
4|2692|Loss: 0.8423773050308228:  42%|████▏     | 2691/6470 [33:24<41:48,  1.51it/s]
4|2692|Loss: 0.8423773050308228:  42%|████▏     | 2692/6470 [33:24<43:40,  1.44it/s]
4|2693|Loss: 0.6844859719276428:  42%|████▏     | 2692/6470 [33:25<43:40,  1.44it/s]
4|2693|Loss: 0.6844859719276428:  42%|████▏     | 2693/6470 [33:25<42:01,  1.50it/s]
4|2694|Loss: 0.803547203540802:  42%|████▏     | 2693/6470 [33:25<42:01,  1.50it/s] 
4|2694|Loss: 0.803547203540802:  42%|████▏     | 2694/6470 [33:25<41:26,  1.52it/s]
4|2695|Loss: 0.9608846306800842:  42%|████▏     | 2694/6470 [33:26<41:26,  1.52it/s]
4|2695|Loss: 0.9608846306800842:  42%|████▏     | 2695/6470 [33:26<43:16,  1.45it/s]
4|2696|Loss: 0.8906967639923096:  42%|████▏     | 2695/6470 [33:27<43:16,  1.45it/s]
4|2696|Loss: 0.8906967639923096:  42%|████▏     | 2696/6470 [33:27<46:20,  1.36it/s]
4|2697|Loss: 0.9018710851669312:  42%|████▏     | 2696/6470 [33:27<46:20,  1.36it/s]
4|2697|Loss: 0.9018710851669312:  42%|████▏     | 2697/6470 [33:28<49:24,  1.27it/s]
4|2698|Loss: 0.8527793288230896:  42%|████▏     | 2697/6470 [33:28<49:24,  1.27it/s]
4|2698|Loss: 0.8527793288230896:  42%|████▏     | 2698/6470 [33:29<51:08,  1.23it/s]
4|2699|Loss: 0.6792477369308472:  42%|████▏     | 2698/6470 [33:29<51:08,  1.23it/s]
4|2699|Loss: 0.6792477369308472:  42%|████▏     | 2699/6470 [33:30<51:05,  1.23it/s]
4|2700|Loss: 0.6835488677024841:  42%|████▏     | 2699/6470 [33:30<51:05,  1.23it/s]
4|2700|Loss: 0.6835488677024841:  42%|████▏     | 2700/6470 [33:30<50:50,  1.24it/s]
4|2701|Loss: 0.8072136044502258:  42%|████▏     | 2700/6470 [33:31<50:50,  1.24it/s]
4|2701|Loss: 0.8072136044502258:  42%|████▏     | 2701/6470 [33:31<46:42,  1.34it/s]
4|2702|Loss: 0.7246329188346863:  42%|████▏     | 2701/6470 [33:31<46:42,  1.34it/s]
4|2702|Loss: 0.7246329188346863:  42%|████▏     | 2702/6470 [33:32<46:45,  1.34it/s]
4|2703|Loss: 0.7820931673049927:  42%|████▏     | 2702/6470 [33:32<46:45,  1.34it/s]
4|2703|Loss: 0.7820931673049927:  42%|████▏     | 2703/6470 [33:33<47:52,  1.31it/s]
4|2704|Loss: 0.9201949238777161:  42%|████▏     | 2703/6470 [33:33<47:52,  1.31it/s]
4|2704|Loss: 0.9201949238777161:  42%|████▏     | 2704/6470 [33:33<46:35,  1.35it/s]
4|2705|Loss: 0.6744552850723267:  42%|████▏     | 2704/6470 [33:34<46:35,  1.35it/s]
4|2705|Loss: 0.6744552850723267:  42%|████▏     | 2705/6470 [33:34<49:11,  1.28it/s]
4|2706|Loss: 0.8185498714447021:  42%|████▏     | 2705/6470 [33:35<49:11,  1.28it/s]
4|2706|Loss: 0.8185498714447021:  42%|████▏     | 2706/6470 [33:35<46:20,  1.35it/s]
4|2707|Loss: 0.8383516669273376:  42%|████▏     | 2706/6470 [33:35<46:20,  1.35it/s]
4|2707|Loss: 0.8383516669273376:  42%|████▏     | 2707/6470 [33:35<42:55,  1.46it/s]
4|2708|Loss: 0.8419131636619568:  42%|████▏     | 2707/6470 [33:36<42:55,  1.46it/s]
4|2708|Loss: 0.8419131636619568:  42%|████▏     | 2708/6470 [33:36<45:13,  1.39it/s]
4|2709|Loss: 0.8475755453109741:  42%|████▏     | 2708/6470 [33:37<45:13,  1.39it/s]
4|2709|Loss: 0.8475755453109741:  42%|████▏     | 2709/6470 [33:37<48:38,  1.29it/s]
4|2710|Loss: 0.7406995296478271:  42%|████▏     | 2709/6470 [33:38<48:38,  1.29it/s]
4|2710|Loss: 0.7406995296478271:  42%|████▏     | 2710/6470 [33:38<50:34,  1.24it/s]
4|2711|Loss: 0.794127345085144:  42%|████▏     | 2710/6470 [33:38<50:34,  1.24it/s] 
4|2711|Loss: 0.794127345085144:  42%|████▏     | 2711/6470 [33:39<52:23,  1.20it/s]
4|2712|Loss: 0.8585222959518433:  42%|████▏     | 2711/6470 [33:39<52:23,  1.20it/s]
4|2712|Loss: 0.8585222959518433:  42%|████▏     | 2712/6470 [33:40<53:17,  1.18it/s]
4|2713|Loss: 0.8030498027801514:  42%|████▏     | 2712/6470 [33:40<53:17,  1.18it/s]
4|2713|Loss: 0.8030498027801514:  42%|████▏     | 2713/6470 [33:40<49:59,  1.25it/s]
4|2714|Loss: 0.8666301369667053:  42%|████▏     | 2713/6470 [33:41<49:59,  1.25it/s]
4|2714|Loss: 0.8666301369667053:  42%|████▏     | 2714/6470 [33:41<48:03,  1.30it/s]
4|2715|Loss: 0.7954885959625244:  42%|████▏     | 2714/6470 [33:41<48:03,  1.30it/s]
4|2715|Loss: 0.7954885959625244:  42%|████▏     | 2715/6470 [33:42<46:54,  1.33it/s]
4|2716|Loss: 0.8370429873466492:  42%|████▏     | 2715/6470 [33:42<46:54,  1.33it/s]
4|2716|Loss: 0.8370429873466492:  42%|████▏     | 2716/6470 [33:43<47:48,  1.31it/s]
4|2717|Loss: 0.7648382782936096:  42%|████▏     | 2716/6470 [33:43<47:48,  1.31it/s]
4|2717|Loss: 0.7648382782936096:  42%|████▏     | 2717/6470 [33:44<54:01,  1.16it/s]
4|2718|Loss: 0.9553834795951843:  42%|████▏     | 2717/6470 [33:44<54:01,  1.16it/s]
4|2718|Loss: 0.9553834795951843:  42%|████▏     | 2718/6470 [33:44<51:41,  1.21it/s]
4|2719|Loss: 0.8090784549713135:  42%|████▏     | 2718/6470 [33:45<51:41,  1.21it/s]
4|2719|Loss: 0.8090784549713135:  42%|████▏     | 2719/6470 [33:45<48:47,  1.28it/s]
4|2720|Loss: 0.6815879344940186:  42%|████▏     | 2719/6470 [33:45<48:47,  1.28it/s]
4|2720|Loss: 0.6815879344940186:  42%|████▏     | 2720/6470 [33:46<47:14,  1.32it/s]
4|2721|Loss: 0.8515914678573608:  42%|████▏     | 2720/6470 [33:46<47:14,  1.32it/s]
4|2721|Loss: 0.8515914678573608:  42%|████▏     | 2721/6470 [33:47<47:15,  1.32it/s]
4|2722|Loss: 0.8412298560142517:  42%|████▏     | 2721/6470 [33:47<47:15,  1.32it/s]
4|2722|Loss: 0.8412298560142517:  42%|████▏     | 2722/6470 [33:47<46:28,  1.34it/s]
4|2723|Loss: 0.7542777061462402:  42%|████▏     | 2722/6470 [33:48<46:28,  1.34it/s]
4|2723|Loss: 0.7542777061462402:  42%|████▏     | 2723/6470 [33:48<45:37,  1.37it/s]
4|2724|Loss: 0.7823567986488342:  42%|████▏     | 2723/6470 [33:48<45:37,  1.37it/s]
4|2724|Loss: 0.7823567986488342:  42%|████▏     | 2724/6470 [33:49<46:03,  1.36it/s]
4|2725|Loss: 0.794299304485321:  42%|████▏     | 2724/6470 [33:49<46:03,  1.36it/s] 
4|2725|Loss: 0.794299304485321:  42%|████▏     | 2725/6470 [33:49<45:33,  1.37it/s]
4|2726|Loss: 1.0737162828445435:  42%|████▏     | 2725/6470 [33:50<45:33,  1.37it/s]
4|2726|Loss: 1.0737162828445435:  42%|████▏     | 2726/6470 [33:50<48:26,  1.29it/s]
4|2727|Loss: 0.7193604111671448:  42%|████▏     | 2726/6470 [33:51<48:26,  1.29it/s]
4|2727|Loss: 0.7193604111671448:  42%|████▏     | 2727/6470 [33:51<47:35,  1.31it/s]
4|2728|Loss: 1.0660001039505005:  42%|████▏     | 2727/6470 [33:51<47:35,  1.31it/s]
4|2728|Loss: 1.0660001039505005:  42%|████▏     | 2728/6470 [33:52<47:15,  1.32it/s]
4|2729|Loss: 0.6781780123710632:  42%|████▏     | 2728/6470 [33:52<47:15,  1.32it/s]
4|2729|Loss: 0.6781780123710632:  42%|████▏     | 2729/6470 [33:52<43:00,  1.45it/s]
4|2730|Loss: 0.8508949279785156:  42%|████▏     | 2729/6470 [33:53<43:00,  1.45it/s]
4|2730|Loss: 0.8508949279785156:  42%|████▏     | 2730/6470 [33:53<42:53,  1.45it/s]
4|2731|Loss: 0.6475576162338257:  42%|████▏     | 2730/6470 [33:53<42:53,  1.45it/s]
4|2731|Loss: 0.6475576162338257:  42%|████▏     | 2731/6470 [33:54<41:47,  1.49it/s]
4|2732|Loss: 0.793208658695221:  42%|████▏     | 2731/6470 [33:54<41:47,  1.49it/s] 
4|2732|Loss: 0.793208658695221:  42%|████▏     | 2732/6470 [33:54<38:29,  1.62it/s]
4|2733|Loss: 0.9183521270751953:  42%|████▏     | 2732/6470 [33:54<38:29,  1.62it/s]
4|2733|Loss: 0.9183521270751953:  42%|████▏     | 2733/6470 [33:55<40:39,  1.53it/s]
4|2734|Loss: 0.7144490480422974:  42%|████▏     | 2733/6470 [33:55<40:39,  1.53it/s]
4|2734|Loss: 0.7144490480422974:  42%|████▏     | 2734/6470 [33:56<44:10,  1.41it/s]
4|2735|Loss: 0.9638200402259827:  42%|████▏     | 2734/6470 [33:56<44:10,  1.41it/s]
4|2735|Loss: 0.9638200402259827:  42%|████▏     | 2735/6470 [33:57<46:14,  1.35it/s]
4|2736|Loss: 1.0622625350952148:  42%|████▏     | 2735/6470 [33:57<46:14,  1.35it/s]
4|2736|Loss: 1.0622625350952148:  42%|████▏     | 2736/6470 [33:57<43:32,  1.43it/s]
4|2737|Loss: 0.8088569641113281:  42%|████▏     | 2736/6470 [33:57<43:32,  1.43it/s]
4|2737|Loss: 0.8088569641113281:  42%|████▏     | 2737/6470 [33:58<42:51,  1.45it/s]
4|2738|Loss: 0.7414950728416443:  42%|████▏     | 2737/6470 [33:58<42:51,  1.45it/s]
4|2738|Loss: 0.7414950728416443:  42%|████▏     | 2738/6470 [33:59<46:23,  1.34it/s]
4|2739|Loss: 0.9374028444290161:  42%|████▏     | 2738/6470 [33:59<46:23,  1.34it/s]
4|2739|Loss: 0.9374028444290161:  42%|████▏     | 2739/6470 [33:59<46:09,  1.35it/s]
4|2740|Loss: 0.8517879247665405:  42%|████▏     | 2739/6470 [34:00<46:09,  1.35it/s]
4|2740|Loss: 0.8517879247665405:  42%|████▏     | 2740/6470 [34:00<46:51,  1.33it/s]
4|2741|Loss: 0.8822281360626221:  42%|████▏     | 2740/6470 [34:01<46:51,  1.33it/s]
4|2741|Loss: 0.8822281360626221:  42%|████▏     | 2741/6470 [34:01<47:54,  1.30it/s]
4|2742|Loss: 0.913982093334198:  42%|████▏     | 2741/6470 [34:01<47:54,  1.30it/s] 
4|2742|Loss: 0.913982093334198:  42%|████▏     | 2742/6470 [34:02<47:09,  1.32it/s]
4|2743|Loss: 0.665292501449585:  42%|████▏     | 2742/6470 [34:02<47:09,  1.32it/s]
4|2743|Loss: 0.665292501449585:  42%|████▏     | 2743/6470 [34:03<50:29,  1.23it/s]
4|2744|Loss: 0.8039444088935852:  42%|████▏     | 2743/6470 [34:03<50:29,  1.23it/s]
4|2744|Loss: 0.8039444088935852:  42%|████▏     | 2744/6470 [34:04<53:16,  1.17it/s]
4|2745|Loss: 0.8066844940185547:  42%|████▏     | 2744/6470 [34:04<53:16,  1.17it/s]
4|2745|Loss: 0.8066844940185547:  42%|████▏     | 2745/6470 [34:04<52:47,  1.18it/s]
4|2746|Loss: 0.9897096753120422:  42%|████▏     | 2745/6470 [34:05<52:47,  1.18it/s]
4|2746|Loss: 0.9897096753120422:  42%|████▏     | 2746/6470 [34:05<51:47,  1.20it/s]
4|2747|Loss: 0.8200591206550598:  42%|████▏     | 2746/6470 [34:06<51:47,  1.20it/s]
4|2747|Loss: 0.8200591206550598:  42%|████▏     | 2747/6470 [34:06<47:20,  1.31it/s]
4|2748|Loss: 0.8425292372703552:  42%|████▏     | 2747/6470 [34:06<47:20,  1.31it/s]
4|2748|Loss: 0.8425292372703552:  42%|████▏     | 2748/6470 [34:07<47:51,  1.30it/s]
4|2749|Loss: 0.8590203523635864:  42%|████▏     | 2748/6470 [34:07<47:51,  1.30it/s]
4|2749|Loss: 0.8590203523635864:  42%|████▏     | 2749/6470 [34:07<48:47,  1.27it/s]
4|2750|Loss: 0.8286933302879333:  42%|████▏     | 2749/6470 [34:08<48:47,  1.27it/s]
4|2750|Loss: 0.8286933302879333:  43%|████▎     | 2750/6470 [34:08<47:14,  1.31it/s]
4|2751|Loss: 0.7909665703773499:  43%|████▎     | 2750/6470 [34:09<47:14,  1.31it/s]
4|2751|Loss: 0.7909665703773499:  43%|████▎     | 2751/6470 [34:09<51:12,  1.21it/s]
4|2752|Loss: 0.9359262585639954:  43%|████▎     | 2751/6470 [34:10<51:12,  1.21it/s]
4|2752|Loss: 0.9359262585639954:  43%|████▎     | 2752/6470 [34:10<51:42,  1.20it/s]
4|2753|Loss: 0.909325897693634:  43%|████▎     | 2752/6470 [34:10<51:42,  1.20it/s] 
4|2753|Loss: 0.909325897693634:  43%|████▎     | 2753/6470 [34:11<52:52,  1.17it/s]
4|2754|Loss: 0.7623975276947021:  43%|████▎     | 2753/6470 [34:11<52:52,  1.17it/s]
4|2754|Loss: 0.7623975276947021:  43%|████▎     | 2754/6470 [34:12<50:41,  1.22it/s]
4|2755|Loss: 0.9481661915779114:  43%|████▎     | 2754/6470 [34:12<50:41,  1.22it/s]
4|2755|Loss: 0.9481661915779114:  43%|████▎     | 2755/6470 [34:12<48:35,  1.27it/s]
4|2756|Loss: 0.9925084114074707:  43%|████▎     | 2755/6470 [34:13<48:35,  1.27it/s]
4|2756|Loss: 0.9925084114074707:  43%|████▎     | 2756/6470 [34:13<45:54,  1.35it/s]
4|2757|Loss: 0.9440883994102478:  43%|████▎     | 2756/6470 [34:13<45:54,  1.35it/s]
4|2757|Loss: 0.9440883994102478:  43%|████▎     | 2757/6470 [34:14<45:59,  1.35it/s]
4|2758|Loss: 0.9436575174331665:  43%|████▎     | 2757/6470 [34:14<45:59,  1.35it/s]
4|2758|Loss: 0.9436575174331665:  43%|████▎     | 2758/6470 [34:15<48:48,  1.27it/s]
4|2759|Loss: 0.9285135865211487:  43%|████▎     | 2758/6470 [34:15<48:48,  1.27it/s]
4|2759|Loss: 0.9285135865211487:  43%|████▎     | 2759/6470 [34:15<47:40,  1.30it/s]
4|2760|Loss: 0.9164815545082092:  43%|████▎     | 2759/6470 [34:16<47:40,  1.30it/s]
4|2760|Loss: 0.9164815545082092:  43%|████▎     | 2760/6470 [34:16<50:26,  1.23it/s]
4|2761|Loss: 0.8792234063148499:  43%|████▎     | 2760/6470 [34:17<50:26,  1.23it/s]
4|2761|Loss: 0.8792234063148499:  43%|████▎     | 2761/6470 [34:17<47:41,  1.30it/s]
4|2762|Loss: 0.7732754945755005:  43%|████▎     | 2761/6470 [34:17<47:41,  1.30it/s]
4|2762|Loss: 0.7732754945755005:  43%|████▎     | 2762/6470 [34:18<47:34,  1.30it/s]
4|2763|Loss: 0.8708017468452454:  43%|████▎     | 2762/6470 [34:18<47:34,  1.30it/s]
4|2763|Loss: 0.8708017468452454:  43%|████▎     | 2763/6470 [34:18<46:19,  1.33it/s]
4|2764|Loss: 0.8405143022537231:  43%|████▎     | 2763/6470 [34:19<46:19,  1.33it/s]
4|2764|Loss: 0.8405143022537231:  43%|████▎     | 2764/6470 [34:19<46:53,  1.32it/s]
4|2765|Loss: 0.93076491355896:  43%|████▎     | 2764/6470 [34:20<46:53,  1.32it/s]  
4|2765|Loss: 0.93076491355896:  43%|████▎     | 2765/6470 [34:20<49:49,  1.24it/s]
4|2766|Loss: 0.8466949462890625:  43%|████▎     | 2765/6470 [34:21<49:49,  1.24it/s]
4|2766|Loss: 0.8466949462890625:  43%|████▎     | 2766/6470 [34:21<48:19,  1.28it/s]
4|2767|Loss: 0.816853940486908:  43%|████▎     | 2766/6470 [34:21<48:19,  1.28it/s] 
4|2767|Loss: 0.816853940486908:  43%|████▎     | 2767/6470 [34:21<44:37,  1.38it/s]
4|2768|Loss: 0.8361439108848572:  43%|████▎     | 2767/6470 [34:22<44:37,  1.38it/s]
4|2768|Loss: 0.8361439108848572:  43%|████▎     | 2768/6470 [34:22<46:10,  1.34it/s]
4|2769|Loss: 0.8348132967948914:  43%|████▎     | 2768/6470 [34:23<46:10,  1.34it/s]
4|2769|Loss: 0.8348132967948914:  43%|████▎     | 2769/6470 [34:23<47:28,  1.30it/s]
4|2770|Loss: 0.9115298986434937:  43%|████▎     | 2769/6470 [34:23<47:28,  1.30it/s]
4|2770|Loss: 0.9115298986434937:  43%|████▎     | 2770/6470 [34:24<44:26,  1.39it/s]
4|2771|Loss: 0.7693688869476318:  43%|████▎     | 2770/6470 [34:24<44:26,  1.39it/s]
4|2771|Loss: 0.7693688869476318:  43%|████▎     | 2771/6470 [34:24<42:50,  1.44it/s]
4|2772|Loss: 0.8652997612953186:  43%|████▎     | 2771/6470 [34:25<42:50,  1.44it/s]
4|2772|Loss: 0.8652997612953186:  43%|████▎     | 2772/6470 [34:25<39:35,  1.56it/s]
4|2773|Loss: 0.7897849082946777:  43%|████▎     | 2772/6470 [34:25<39:35,  1.56it/s]
4|2773|Loss: 0.7897849082946777:  43%|████▎     | 2773/6470 [34:26<43:41,  1.41it/s]
4|2774|Loss: 1.0919471979141235:  43%|████▎     | 2773/6470 [34:26<43:41,  1.41it/s]
4|2774|Loss: 1.0919471979141235:  43%|████▎     | 2774/6470 [34:26<44:25,  1.39it/s]
4|2775|Loss: 0.6899818778038025:  43%|████▎     | 2774/6470 [34:27<44:25,  1.39it/s]
4|2775|Loss: 0.6899818778038025:  43%|████▎     | 2775/6470 [34:27<43:48,  1.41it/s]
4|2776|Loss: 0.7324063181877136:  43%|████▎     | 2775/6470 [34:27<43:48,  1.41it/s]
4|2776|Loss: 0.7324063181877136:  43%|████▎     | 2776/6470 [34:28<42:40,  1.44it/s]
4|2777|Loss: 0.972379744052887:  43%|████▎     | 2776/6470 [34:28<42:40,  1.44it/s] 
4|2777|Loss: 0.972379744052887:  43%|████▎     | 2777/6470 [34:28<42:17,  1.46it/s]
4|2778|Loss: 0.7741307020187378:  43%|████▎     | 2777/6470 [34:29<42:17,  1.46it/s]
4|2778|Loss: 0.7741307020187378:  43%|████▎     | 2778/6470 [34:29<42:28,  1.45it/s]
4|2779|Loss: 0.814732015132904:  43%|████▎     | 2778/6470 [34:30<42:28,  1.45it/s] 
4|2779|Loss: 0.814732015132904:  43%|████▎     | 2779/6470 [34:30<43:29,  1.41it/s]
4|2780|Loss: 0.7154233455657959:  43%|████▎     | 2779/6470 [34:30<43:29,  1.41it/s]
4|2780|Loss: 0.7154233455657959:  43%|████▎     | 2780/6470 [34:30<42:03,  1.46it/s]
4|2781|Loss: 0.9697644114494324:  43%|████▎     | 2780/6470 [34:31<42:03,  1.46it/s]
4|2781|Loss: 0.9697644114494324:  43%|████▎     | 2781/6470 [34:31<40:07,  1.53it/s]
4|2782|Loss: 0.8749731779098511:  43%|████▎     | 2781/6470 [34:31<40:07,  1.53it/s]
4|2782|Loss: 0.8749731779098511:  43%|████▎     | 2782/6470 [34:32<41:24,  1.48it/s]
4|2783|Loss: 0.8613808751106262:  43%|████▎     | 2782/6470 [34:32<41:24,  1.48it/s]
4|2783|Loss: 0.8613808751106262:  43%|████▎     | 2783/6470 [34:33<45:01,  1.36it/s]
4|2784|Loss: 0.8429118394851685:  43%|████▎     | 2783/6470 [34:33<45:01,  1.36it/s]
4|2784|Loss: 0.8429118394851685:  43%|████▎     | 2784/6470 [34:34<47:22,  1.30it/s]
4|2785|Loss: 0.9365001320838928:  43%|████▎     | 2784/6470 [34:34<47:22,  1.30it/s]
4|2785|Loss: 0.9365001320838928:  43%|████▎     | 2785/6470 [34:34<47:33,  1.29it/s]
4|2786|Loss: 0.7164850234985352:  43%|████▎     | 2785/6470 [34:35<47:33,  1.29it/s]
4|2786|Loss: 0.7164850234985352:  43%|████▎     | 2786/6470 [34:35<48:54,  1.26it/s]
4|2787|Loss: 0.6929576992988586:  43%|████▎     | 2786/6470 [34:36<48:54,  1.26it/s]
4|2787|Loss: 0.6929576992988586:  43%|████▎     | 2787/6470 [34:36<47:48,  1.28it/s]
4|2788|Loss: 0.821394145488739:  43%|████▎     | 2787/6470 [34:36<47:48,  1.28it/s] 
4|2788|Loss: 0.821394145488739:  43%|████▎     | 2788/6470 [34:37<47:10,  1.30it/s]
4|2789|Loss: 0.7867769598960876:  43%|████▎     | 2788/6470 [34:37<47:10,  1.30it/s]
4|2789|Loss: 0.7867769598960876:  43%|████▎     | 2789/6470 [34:37<45:34,  1.35it/s]
4|2790|Loss: 0.8755922913551331:  43%|████▎     | 2789/6470 [34:38<45:34,  1.35it/s]
4|2790|Loss: 0.8755922913551331:  43%|████▎     | 2790/6470 [34:38<43:38,  1.41it/s]
4|2791|Loss: 0.7070736289024353:  43%|████▎     | 2790/6470 [34:38<43:38,  1.41it/s]
4|2791|Loss: 0.7070736289024353:  43%|████▎     | 2791/6470 [34:39<42:31,  1.44it/s]
4|2792|Loss: 0.7968135476112366:  43%|████▎     | 2791/6470 [34:39<42:31,  1.44it/s]
4|2792|Loss: 0.7968135476112366:  43%|████▎     | 2792/6470 [34:39<43:49,  1.40it/s]
4|2793|Loss: 0.9487319588661194:  43%|████▎     | 2792/6470 [34:40<43:49,  1.40it/s]
4|2793|Loss: 0.9487319588661194:  43%|████▎     | 2793/6470 [34:40<46:52,  1.31it/s]
4|2794|Loss: 0.863038182258606:  43%|████▎     | 2793/6470 [34:41<46:52,  1.31it/s] 
4|2794|Loss: 0.863038182258606:  43%|████▎     | 2794/6470 [34:41<47:20,  1.29it/s]
4|2795|Loss: 0.8541251420974731:  43%|████▎     | 2794/6470 [34:41<47:20,  1.29it/s]
4|2795|Loss: 0.8541251420974731:  43%|████▎     | 2795/6470 [34:42<47:09,  1.30it/s]
4|2796|Loss: 0.7078651785850525:  43%|████▎     | 2795/6470 [34:42<47:09,  1.30it/s]
4|2796|Loss: 0.7078651785850525:  43%|████▎     | 2796/6470 [34:42<43:39,  1.40it/s]
4|2797|Loss: 0.845450222492218:  43%|████▎     | 2796/6470 [34:43<43:39,  1.40it/s] 
4|2797|Loss: 0.845450222492218:  43%|████▎     | 2797/6470 [34:43<44:25,  1.38it/s]
4|2798|Loss: 0.7755870223045349:  43%|████▎     | 2797/6470 [34:44<44:25,  1.38it/s]
4|2798|Loss: 0.7755870223045349:  43%|████▎     | 2798/6470 [34:44<45:48,  1.34it/s]
4|2799|Loss: 0.8507340550422668:  43%|████▎     | 2798/6470 [34:44<45:48,  1.34it/s]
4|2799|Loss: 0.8507340550422668:  43%|████▎     | 2799/6470 [34:45<47:43,  1.28it/s]
4|2800|Loss: 0.9804426431655884:  43%|████▎     | 2799/6470 [34:45<47:43,  1.28it/s]
4|2800|Loss: 0.9804426431655884:  43%|████▎     | 2800/6470 [34:46<49:11,  1.24it/s]
4|2801|Loss: 0.8458245992660522:  43%|████▎     | 2800/6470 [34:46<49:11,  1.24it/s]
4|2801|Loss: 0.8458245992660522:  43%|████▎     | 2801/6470 [34:46<45:59,  1.33it/s]
4|2802|Loss: 0.859356701374054:  43%|████▎     | 2801/6470 [34:47<45:59,  1.33it/s] 
4|2802|Loss: 0.859356701374054:  43%|████▎     | 2802/6470 [34:47<43:58,  1.39it/s]
4|2803|Loss: 0.8305717706680298:  43%|████▎     | 2802/6470 [34:47<43:58,  1.39it/s]
4|2803|Loss: 0.8305717706680298:  43%|████▎     | 2803/6470 [34:47<40:15,  1.52it/s]
4|2804|Loss: 0.8284960985183716:  43%|████▎     | 2803/6470 [34:48<40:15,  1.52it/s]
4|2804|Loss: 0.8284960985183716:  43%|████▎     | 2804/6470 [34:48<38:29,  1.59it/s]
4|2805|Loss: 0.8922119736671448:  43%|████▎     | 2804/6470 [34:48<38:29,  1.59it/s]
4|2805|Loss: 0.8922119736671448:  43%|████▎     | 2805/6470 [34:49<38:55,  1.57it/s]
4|2806|Loss: 0.8859903216362:  43%|████▎     | 2805/6470 [34:49<38:55,  1.57it/s]   
4|2806|Loss: 0.8859903216362:  43%|████▎     | 2806/6470 [34:49<41:16,  1.48it/s]
4|2807|Loss: 0.5621800422668457:  43%|████▎     | 2806/6470 [34:50<41:16,  1.48it/s]
4|2807|Loss: 0.5621800422668457:  43%|████▎     | 2807/6470 [34:50<47:02,  1.30it/s]
4|2808|Loss: 0.8419321179389954:  43%|████▎     | 2807/6470 [34:51<47:02,  1.30it/s]
4|2808|Loss: 0.8419321179389954:  43%|████▎     | 2808/6470 [34:51<46:31,  1.31it/s]
4|2809|Loss: 0.7753220796585083:  43%|████▎     | 2808/6470 [34:52<46:31,  1.31it/s]
4|2809|Loss: 0.7753220796585083:  43%|████▎     | 2809/6470 [34:52<42:55,  1.42it/s]
4|2810|Loss: 0.7552493214607239:  43%|████▎     | 2809/6470 [34:52<42:55,  1.42it/s]
4|2810|Loss: 0.7552493214607239:  43%|████▎     | 2810/6470 [34:52<44:02,  1.39it/s]
4|2811|Loss: 0.6936249732971191:  43%|████▎     | 2810/6470 [34:53<44:02,  1.39it/s]
4|2811|Loss: 0.6936249732971191:  43%|████▎     | 2811/6470 [34:53<45:28,  1.34it/s]
4|2812|Loss: 0.9410867094993591:  43%|████▎     | 2811/6470 [34:54<45:28,  1.34it/s]
4|2812|Loss: 0.9410867094993591:  43%|████▎     | 2812/6470 [34:54<46:43,  1.30it/s]
4|2813|Loss: 0.7041919827461243:  43%|████▎     | 2812/6470 [34:55<46:43,  1.30it/s]
4|2813|Loss: 0.7041919827461243:  43%|████▎     | 2813/6470 [34:55<49:17,  1.24it/s]
4|2814|Loss: 0.7391405701637268:  43%|████▎     | 2813/6470 [34:55<49:17,  1.24it/s]
4|2814|Loss: 0.7391405701637268:  43%|████▎     | 2814/6470 [34:56<43:55,  1.39it/s]
4|2815|Loss: 0.8648681640625:  43%|████▎     | 2814/6470 [34:56<43:55,  1.39it/s]   
4|2815|Loss: 0.8648681640625:  44%|████▎     | 2815/6470 [34:56<43:24,  1.40it/s]
4|2816|Loss: 0.9249404072761536:  44%|████▎     | 2815/6470 [34:57<43:24,  1.40it/s]
4|2816|Loss: 0.9249404072761536:  44%|████▎     | 2816/6470 [34:57<42:54,  1.42it/s]
4|2817|Loss: 0.7787672877311707:  44%|████▎     | 2816/6470 [34:57<42:54,  1.42it/s]
4|2817|Loss: 0.7787672877311707:  44%|████▎     | 2817/6470 [34:57<40:08,  1.52it/s]
4|2818|Loss: 0.8045161962509155:  44%|████▎     | 2817/6470 [34:58<40:08,  1.52it/s]
4|2818|Loss: 0.8045161962509155:  44%|████▎     | 2818/6470 [34:58<41:14,  1.48it/s]
4|2819|Loss: 0.708369255065918:  44%|████▎     | 2818/6470 [34:59<41:14,  1.48it/s] 
4|2819|Loss: 0.708369255065918:  44%|████▎     | 2819/6470 [34:59<43:14,  1.41it/s]
4|2820|Loss: 0.8033577799797058:  44%|████▎     | 2819/6470 [34:59<43:14,  1.41it/s]
4|2820|Loss: 0.8033577799797058:  44%|████▎     | 2820/6470 [35:00<41:42,  1.46it/s]
4|2821|Loss: 0.7681169509887695:  44%|████▎     | 2820/6470 [35:00<41:42,  1.46it/s]
4|2821|Loss: 0.7681169509887695:  44%|████▎     | 2821/6470 [35:00<42:15,  1.44it/s]
4|2822|Loss: 0.8061754703521729:  44%|████▎     | 2821/6470 [35:01<42:15,  1.44it/s]
4|2822|Loss: 0.8061754703521729:  44%|████▎     | 2822/6470 [35:01<44:06,  1.38it/s]
4|2823|Loss: 0.7881460785865784:  44%|████▎     | 2822/6470 [35:01<44:06,  1.38it/s]
4|2823|Loss: 0.7881460785865784:  44%|████▎     | 2823/6470 [35:02<40:59,  1.48it/s]
4|2824|Loss: 0.7456461191177368:  44%|████▎     | 2823/6470 [35:02<40:59,  1.48it/s]
4|2824|Loss: 0.7456461191177368:  44%|████▎     | 2824/6470 [35:02<41:53,  1.45it/s]
4|2825|Loss: 0.8494754433631897:  44%|████▎     | 2824/6470 [35:03<41:53,  1.45it/s]
4|2825|Loss: 0.8494754433631897:  44%|████▎     | 2825/6470 [35:03<41:59,  1.45it/s]
4|2826|Loss: 0.7362316846847534:  44%|████▎     | 2825/6470 [35:03<41:59,  1.45it/s]
4|2826|Loss: 0.7362316846847534:  44%|████▎     | 2826/6470 [35:04<40:21,  1.50it/s]
4|2827|Loss: 0.8387520909309387:  44%|████▎     | 2826/6470 [35:04<40:21,  1.50it/s]
4|2827|Loss: 0.8387520909309387:  44%|████▎     | 2827/6470 [35:04<42:20,  1.43it/s]
4|2828|Loss: 0.842704176902771:  44%|████▎     | 2827/6470 [35:05<42:20,  1.43it/s] 
4|2828|Loss: 0.842704176902771:  44%|████▎     | 2828/6470 [35:05<44:24,  1.37it/s]
4|2829|Loss: 0.876983106136322:  44%|████▎     | 2828/6470 [35:06<44:24,  1.37it/s]
4|2829|Loss: 0.876983106136322:  44%|████▎     | 2829/6470 [35:06<43:25,  1.40it/s]
4|2830|Loss: 0.7940518856048584:  44%|████▎     | 2829/6470 [35:06<43:25,  1.40it/s]
4|2830|Loss: 0.7940518856048584:  44%|████▎     | 2830/6470 [35:07<46:00,  1.32it/s]
4|2831|Loss: 0.8728880882263184:  44%|████▎     | 2830/6470 [35:07<46:00,  1.32it/s]
4|2831|Loss: 0.8728880882263184:  44%|████▍     | 2831/6470 [35:08<45:15,  1.34it/s]
4|2832|Loss: 0.7250617742538452:  44%|████▍     | 2831/6470 [35:08<45:15,  1.34it/s]
4|2832|Loss: 0.7250617742538452:  44%|████▍     | 2832/6470 [35:08<48:04,  1.26it/s]
4|2833|Loss: 0.7730897665023804:  44%|████▍     | 2832/6470 [35:09<48:04,  1.26it/s]
4|2833|Loss: 0.7730897665023804:  44%|████▍     | 2833/6470 [35:09<45:44,  1.33it/s]
4|2834|Loss: 0.913762092590332:  44%|████▍     | 2833/6470 [35:09<45:44,  1.33it/s] 
4|2834|Loss: 0.913762092590332:  44%|████▍     | 2834/6470 [35:10<45:57,  1.32it/s]
4|2835|Loss: 0.7983895540237427:  44%|████▍     | 2834/6470 [35:10<45:57,  1.32it/s]
4|2835|Loss: 0.7983895540237427:  44%|████▍     | 2835/6470 [35:11<48:05,  1.26it/s]
4|2836|Loss: 0.796425461769104:  44%|████▍     | 2835/6470 [35:11<48:05,  1.26it/s] 
4|2836|Loss: 0.796425461769104:  44%|████▍     | 2836/6470 [35:12<48:32,  1.25it/s]
4|2837|Loss: 0.781592607498169:  44%|████▍     | 2836/6470 [35:12<48:32,  1.25it/s]
4|2837|Loss: 0.781592607498169:  44%|████▍     | 2837/6470 [35:13<51:39,  1.17it/s]
4|2838|Loss: 0.8499327301979065:  44%|████▍     | 2837/6470 [35:13<51:39,  1.17it/s]
4|2838|Loss: 0.8499327301979065:  44%|████▍     | 2838/6470 [35:13<50:03,  1.21it/s]
4|2839|Loss: 0.7926557660102844:  44%|████▍     | 2838/6470 [35:14<50:03,  1.21it/s]
4|2839|Loss: 0.7926557660102844:  44%|████▍     | 2839/6470 [35:14<46:50,  1.29it/s]
4|2840|Loss: 0.8096361756324768:  44%|████▍     | 2839/6470 [35:14<46:50,  1.29it/s]
4|2840|Loss: 0.8096361756324768:  44%|████▍     | 2840/6470 [35:15<45:01,  1.34it/s]
4|2841|Loss: 0.8741258382797241:  44%|████▍     | 2840/6470 [35:15<45:01,  1.34it/s]
4|2841|Loss: 0.8741258382797241:  44%|████▍     | 2841/6470 [35:15<43:15,  1.40it/s]
4|2842|Loss: 0.9604905843734741:  44%|████▍     | 2841/6470 [35:16<43:15,  1.40it/s]
4|2842|Loss: 0.9604905843734741:  44%|████▍     | 2842/6470 [35:16<46:05,  1.31it/s]
4|2843|Loss: 0.832065999507904:  44%|████▍     | 2842/6470 [35:17<46:05,  1.31it/s] 
4|2843|Loss: 0.832065999507904:  44%|████▍     | 2843/6470 [35:17<47:09,  1.28it/s]
4|2844|Loss: 0.7334181070327759:  44%|████▍     | 2843/6470 [35:17<47:09,  1.28it/s]
4|2844|Loss: 0.7334181070327759:  44%|████▍     | 2844/6470 [35:18<48:15,  1.25it/s]
4|2845|Loss: 0.7533316016197205:  44%|████▍     | 2844/6470 [35:18<48:15,  1.25it/s]
4|2845|Loss: 0.7533316016197205:  44%|████▍     | 2845/6470 [35:19<47:17,  1.28it/s]
4|2846|Loss: 0.7904527187347412:  44%|████▍     | 2845/6470 [35:19<47:17,  1.28it/s]
4|2846|Loss: 0.7904527187347412:  44%|████▍     | 2846/6470 [35:19<45:39,  1.32it/s]
4|2847|Loss: 0.7995487451553345:  44%|████▍     | 2846/6470 [35:20<45:39,  1.32it/s]
4|2847|Loss: 0.7995487451553345:  44%|████▍     | 2847/6470 [35:20<47:04,  1.28it/s]
4|2848|Loss: 0.9086943864822388:  44%|████▍     | 2847/6470 [35:21<47:04,  1.28it/s]
4|2848|Loss: 0.9086943864822388:  44%|████▍     | 2848/6470 [35:21<48:33,  1.24it/s]
4|2849|Loss: 0.8537577986717224:  44%|████▍     | 2848/6470 [35:21<48:33,  1.24it/s]
4|2849|Loss: 0.8537577986717224:  44%|████▍     | 2849/6470 [35:22<51:31,  1.17it/s]
4|2850|Loss: 0.6348326206207275:  44%|████▍     | 2849/6470 [35:22<51:31,  1.17it/s]
4|2850|Loss: 0.6348326206207275:  44%|████▍     | 2850/6470 [35:23<48:32,  1.24it/s]
4|2851|Loss: 0.7195534110069275:  44%|████▍     | 2850/6470 [35:23<48:32,  1.24it/s]
4|2851|Loss: 0.7195534110069275:  44%|████▍     | 2851/6470 [35:23<47:11,  1.28it/s]
4|2852|Loss: 0.9181563854217529:  44%|████▍     | 2851/6470 [35:24<47:11,  1.28it/s]
4|2852|Loss: 0.9181563854217529:  44%|████▍     | 2852/6470 [35:24<47:23,  1.27it/s]
4|2853|Loss: 0.8933258056640625:  44%|████▍     | 2852/6470 [35:25<47:23,  1.27it/s]
4|2853|Loss: 0.8933258056640625:  44%|████▍     | 2853/6470 [35:25<47:42,  1.26it/s]
4|2854|Loss: 0.9636869430541992:  44%|████▍     | 2853/6470 [35:25<47:42,  1.26it/s]
4|2854|Loss: 0.9636869430541992:  44%|████▍     | 2854/6470 [35:26<47:35,  1.27it/s]
4|2855|Loss: 0.7514989972114563:  44%|████▍     | 2854/6470 [35:26<47:35,  1.27it/s]
4|2855|Loss: 0.7514989972114563:  44%|████▍     | 2855/6470 [35:26<46:09,  1.31it/s]
4|2856|Loss: 0.7729990482330322:  44%|████▍     | 2855/6470 [35:27<46:09,  1.31it/s]
4|2856|Loss: 0.7729990482330322:  44%|████▍     | 2856/6470 [35:27<46:10,  1.30it/s]
4|2857|Loss: 0.9055305123329163:  44%|████▍     | 2856/6470 [35:28<46:10,  1.30it/s]
4|2857|Loss: 0.9055305123329163:  44%|████▍     | 2857/6470 [35:28<46:09,  1.30it/s]
4|2858|Loss: 0.7811504006385803:  44%|████▍     | 2857/6470 [35:28<46:09,  1.30it/s]
4|2858|Loss: 0.7811504006385803:  44%|████▍     | 2858/6470 [35:29<46:21,  1.30it/s]
4|2859|Loss: 0.8912989497184753:  44%|████▍     | 2858/6470 [35:29<46:21,  1.30it/s]
4|2859|Loss: 0.8912989497184753:  44%|████▍     | 2859/6470 [35:30<48:39,  1.24it/s]
4|2860|Loss: 0.7360996007919312:  44%|████▍     | 2859/6470 [35:30<48:39,  1.24it/s]
4|2860|Loss: 0.7360996007919312:  44%|████▍     | 2860/6470 [35:30<47:05,  1.28it/s]
4|2861|Loss: 0.8181595206260681:  44%|████▍     | 2860/6470 [35:31<47:05,  1.28it/s]
4|2861|Loss: 0.8181595206260681:  44%|████▍     | 2861/6470 [35:31<48:45,  1.23it/s]
4|2862|Loss: 0.8355125784873962:  44%|████▍     | 2861/6470 [35:32<48:45,  1.23it/s]
4|2862|Loss: 0.8355125784873962:  44%|████▍     | 2862/6470 [35:32<48:34,  1.24it/s]
4|2863|Loss: 0.8411098718643188:  44%|████▍     | 2862/6470 [35:32<48:34,  1.24it/s]
4|2863|Loss: 0.8411098718643188:  44%|████▍     | 2863/6470 [35:33<45:38,  1.32it/s]
4|2864|Loss: 0.8531818389892578:  44%|████▍     | 2863/6470 [35:33<45:38,  1.32it/s]
4|2864|Loss: 0.8531818389892578:  44%|████▍     | 2864/6470 [35:33<43:41,  1.38it/s]
4|2865|Loss: 0.8640114665031433:  44%|████▍     | 2864/6470 [35:34<43:41,  1.38it/s]
4|2865|Loss: 0.8640114665031433:  44%|████▍     | 2865/6470 [35:34<43:03,  1.40it/s]
4|2866|Loss: 0.689893364906311:  44%|████▍     | 2865/6470 [35:34<43:03,  1.40it/s] 
4|2866|Loss: 0.689893364906311:  44%|████▍     | 2866/6470 [35:35<46:02,  1.30it/s]
4|2867|Loss: 0.7477482557296753:  44%|████▍     | 2866/6470 [35:35<46:02,  1.30it/s]
4|2867|Loss: 0.7477482557296753:  44%|████▍     | 2867/6470 [35:36<45:17,  1.33it/s]
4|2868|Loss: 0.9617844820022583:  44%|████▍     | 2867/6470 [35:36<45:17,  1.33it/s]
4|2868|Loss: 0.9617844820022583:  44%|████▍     | 2868/6470 [35:36<47:04,  1.28it/s]
4|2869|Loss: 0.7742206454277039:  44%|████▍     | 2868/6470 [35:37<47:04,  1.28it/s]
4|2869|Loss: 0.7742206454277039:  44%|████▍     | 2869/6470 [35:37<46:52,  1.28it/s]
4|2870|Loss: 0.9035700559616089:  44%|████▍     | 2869/6470 [35:38<46:52,  1.28it/s]
4|2870|Loss: 0.9035700559616089:  44%|████▍     | 2870/6470 [35:38<46:47,  1.28it/s]
4|2871|Loss: 0.7222694754600525:  44%|████▍     | 2870/6470 [35:39<46:47,  1.28it/s]
4|2871|Loss: 0.7222694754600525:  44%|████▍     | 2871/6470 [35:39<48:36,  1.23it/s]
4|2872|Loss: 0.8837766647338867:  44%|████▍     | 2871/6470 [35:39<48:36,  1.23it/s]
4|2872|Loss: 0.8837766647338867:  44%|████▍     | 2872/6470 [35:40<49:41,  1.21it/s]
4|2873|Loss: 0.9339435696601868:  44%|████▍     | 2872/6470 [35:40<49:41,  1.21it/s]
4|2873|Loss: 0.9339435696601868:  44%|████▍     | 2873/6470 [35:41<48:09,  1.24it/s]
4|2874|Loss: 0.8648245334625244:  44%|████▍     | 2873/6470 [35:41<48:09,  1.24it/s]
4|2874|Loss: 0.8648245334625244:  44%|████▍     | 2874/6470 [35:41<46:04,  1.30it/s]
4|2875|Loss: 0.7496154308319092:  44%|████▍     | 2874/6470 [35:42<46:04,  1.30it/s]
4|2875|Loss: 0.7496154308319092:  44%|████▍     | 2875/6470 [35:42<44:39,  1.34it/s]
4|2876|Loss: 0.603939414024353:  44%|████▍     | 2875/6470 [35:42<44:39,  1.34it/s] 
4|2876|Loss: 0.603939414024353:  44%|████▍     | 2876/6470 [35:43<45:08,  1.33it/s]
4|2877|Loss: 0.8566286563873291:  44%|████▍     | 2876/6470 [35:43<45:08,  1.33it/s]
4|2877|Loss: 0.8566286563873291:  44%|████▍     | 2877/6470 [35:43<44:08,  1.36it/s]
4|2878|Loss: 0.8974202275276184:  44%|████▍     | 2877/6470 [35:44<44:08,  1.36it/s]
4|2878|Loss: 0.8974202275276184:  44%|████▍     | 2878/6470 [35:44<42:32,  1.41it/s]
4|2879|Loss: 0.7128276824951172:  44%|████▍     | 2878/6470 [35:44<42:32,  1.41it/s]
4|2879|Loss: 0.7128276824951172:  44%|████▍     | 2879/6470 [35:44<37:24,  1.60it/s]
4|2880|Loss: 0.8369202613830566:  44%|████▍     | 2879/6470 [35:45<37:24,  1.60it/s]
4|2880|Loss: 0.8369202613830566:  45%|████▍     | 2880/6470 [35:45<37:29,  1.60it/s]
4|2881|Loss: 0.7929573059082031:  45%|████▍     | 2880/6470 [35:45<37:29,  1.60it/s]
4|2881|Loss: 0.7929573059082031:  45%|████▍     | 2881/6470 [35:46<37:53,  1.58it/s]
4|2882|Loss: 0.7866140604019165:  45%|████▍     | 2881/6470 [35:46<37:53,  1.58it/s]
4|2882|Loss: 0.7866140604019165:  45%|████▍     | 2882/6470 [35:46<37:26,  1.60it/s]
4|2883|Loss: 0.7213919758796692:  45%|████▍     | 2882/6470 [35:47<37:26,  1.60it/s]
4|2883|Loss: 0.7213919758796692:  45%|████▍     | 2883/6470 [35:47<37:50,  1.58it/s]
4|2884|Loss: 0.7239941954612732:  45%|████▍     | 2883/6470 [35:47<37:50,  1.58it/s]
4|2884|Loss: 0.7239941954612732:  45%|████▍     | 2884/6470 [35:48<38:22,  1.56it/s]
4|2885|Loss: 0.7181366682052612:  45%|████▍     | 2884/6470 [35:48<38:22,  1.56it/s]
4|2885|Loss: 0.7181366682052612:  45%|████▍     | 2885/6470 [35:48<41:31,  1.44it/s]
4|2886|Loss: 0.869488537311554:  45%|████▍     | 2885/6470 [35:49<41:31,  1.44it/s] 
4|2886|Loss: 0.869488537311554:  45%|████▍     | 2886/6470 [35:49<41:46,  1.43it/s]
4|2887|Loss: 0.9715788960456848:  45%|████▍     | 2886/6470 [35:50<41:46,  1.43it/s]
4|2887|Loss: 0.9715788960456848:  45%|████▍     | 2887/6470 [35:50<40:27,  1.48it/s]
4|2888|Loss: 0.833402156829834:  45%|████▍     | 2887/6470 [35:50<40:27,  1.48it/s] 
4|2888|Loss: 0.833402156829834:  45%|████▍     | 2888/6470 [35:51<43:43,  1.37it/s]
4|2889|Loss: 0.9154816269874573:  45%|████▍     | 2888/6470 [35:51<43:43,  1.37it/s]
4|2889|Loss: 0.9154816269874573:  45%|████▍     | 2889/6470 [35:51<44:05,  1.35it/s]
4|2890|Loss: 0.8295792937278748:  45%|████▍     | 2889/6470 [35:52<44:05,  1.35it/s]
4|2890|Loss: 0.8295792937278748:  45%|████▍     | 2890/6470 [35:52<43:12,  1.38it/s]
4|2891|Loss: 0.755260169506073:  45%|████▍     | 2890/6470 [35:53<43:12,  1.38it/s] 
4|2891|Loss: 0.755260169506073:  45%|████▍     | 2891/6470 [35:53<42:08,  1.42it/s]
4|2892|Loss: 0.8544180989265442:  45%|████▍     | 2891/6470 [35:53<42:08,  1.42it/s]
4|2892|Loss: 0.8544180989265442:  45%|████▍     | 2892/6470 [35:53<40:16,  1.48it/s]
4|2893|Loss: 0.7945126891136169:  45%|████▍     | 2892/6470 [35:54<40:16,  1.48it/s]
4|2893|Loss: 0.7945126891136169:  45%|████▍     | 2893/6470 [35:54<42:42,  1.40it/s]
4|2894|Loss: 0.770499587059021:  45%|████▍     | 2893/6470 [35:55<42:42,  1.40it/s] 
4|2894|Loss: 0.770499587059021:  45%|████▍     | 2894/6470 [35:55<43:06,  1.38it/s]
4|2895|Loss: 0.7380971908569336:  45%|████▍     | 2894/6470 [35:55<43:06,  1.38it/s]
4|2895|Loss: 0.7380971908569336:  45%|████▍     | 2895/6470 [35:56<45:18,  1.32it/s]
4|2896|Loss: 0.7705066204071045:  45%|████▍     | 2895/6470 [35:56<45:18,  1.32it/s]
4|2896|Loss: 0.7705066204071045:  45%|████▍     | 2896/6470 [35:57<47:29,  1.25it/s]
4|2897|Loss: 0.8585205674171448:  45%|████▍     | 2896/6470 [35:57<47:29,  1.25it/s]
4|2897|Loss: 0.8585205674171448:  45%|████▍     | 2897/6470 [35:57<45:32,  1.31it/s]
4|2898|Loss: 0.7717611193656921:  45%|████▍     | 2897/6470 [35:58<45:32,  1.31it/s]
4|2898|Loss: 0.7717611193656921:  45%|████▍     | 2898/6470 [35:58<42:23,  1.40it/s]
4|2899|Loss: 0.9193779230117798:  45%|████▍     | 2898/6470 [35:58<42:23,  1.40it/s]
4|2899|Loss: 0.9193779230117798:  45%|████▍     | 2899/6470 [35:59<43:22,  1.37it/s]
4|2900|Loss: 0.788676381111145:  45%|████▍     | 2899/6470 [35:59<43:22,  1.37it/s] 
4|2900|Loss: 0.788676381111145:  45%|████▍     | 2900/6470 [35:59<38:51,  1.53it/s]
4|2901|Loss: 0.7603756189346313:  45%|████▍     | 2900/6470 [36:00<38:51,  1.53it/s]
4|2901|Loss: 0.7603756189346313:  45%|████▍     | 2901/6470 [36:00<42:05,  1.41it/s]
4|2902|Loss: 0.8167058825492859:  45%|████▍     | 2901/6470 [36:00<42:05,  1.41it/s]
4|2902|Loss: 0.8167058825492859:  45%|████▍     | 2902/6470 [36:01<41:34,  1.43it/s]
4|2903|Loss: 0.9303399324417114:  45%|████▍     | 2902/6470 [36:01<41:34,  1.43it/s]
4|2903|Loss: 0.9303399324417114:  45%|████▍     | 2903/6470 [36:01<42:33,  1.40it/s]
4|2904|Loss: 0.7407613396644592:  45%|████▍     | 2903/6470 [36:02<42:33,  1.40it/s]
4|2904|Loss: 0.7407613396644592:  45%|████▍     | 2904/6470 [36:02<46:20,  1.28it/s]
4|2905|Loss: 0.7923627495765686:  45%|████▍     | 2904/6470 [36:03<46:20,  1.28it/s]
4|2905|Loss: 0.7923627495765686:  45%|████▍     | 2905/6470 [36:03<48:51,  1.22it/s]
4|2906|Loss: 0.7750311493873596:  45%|████▍     | 2905/6470 [36:04<48:51,  1.22it/s]
4|2906|Loss: 0.7750311493873596:  45%|████▍     | 2906/6470 [36:04<47:04,  1.26it/s]
4|2907|Loss: 0.9070676565170288:  45%|████▍     | 2906/6470 [36:04<47:04,  1.26it/s]
4|2907|Loss: 0.9070676565170288:  45%|████▍     | 2907/6470 [36:05<46:39,  1.27it/s]
4|2908|Loss: 0.8060460090637207:  45%|████▍     | 2907/6470 [36:05<46:39,  1.27it/s]
4|2908|Loss: 0.8060460090637207:  45%|████▍     | 2908/6470 [36:06<47:48,  1.24it/s]
4|2909|Loss: 0.9309715032577515:  45%|████▍     | 2908/6470 [36:06<47:48,  1.24it/s]
4|2909|Loss: 0.9309715032577515:  45%|████▍     | 2909/6470 [36:06<48:05,  1.23it/s]
4|2910|Loss: 0.78396075963974:  45%|████▍     | 2909/6470 [36:07<48:05,  1.23it/s]  
4|2910|Loss: 0.78396075963974:  45%|████▍     | 2910/6470 [36:07<45:15,  1.31it/s]
4|2911|Loss: 0.861010730266571:  45%|████▍     | 2910/6470 [36:08<45:15,  1.31it/s]
4|2911|Loss: 0.861010730266571:  45%|████▍     | 2911/6470 [36:08<44:33,  1.33it/s]
4|2912|Loss: 0.8148558735847473:  45%|████▍     | 2911/6470 [36:08<44:33,  1.33it/s]
4|2912|Loss: 0.8148558735847473:  45%|████▌     | 2912/6470 [36:09<46:57,  1.26it/s]
4|2913|Loss: 0.8755513429641724:  45%|████▌     | 2912/6470 [36:09<46:57,  1.26it/s]
4|2913|Loss: 0.8755513429641724:  45%|████▌     | 2913/6470 [36:10<46:30,  1.27it/s]
4|2914|Loss: 0.6735799908638:  45%|████▌     | 2913/6470 [36:10<46:30,  1.27it/s]   
4|2914|Loss: 0.6735799908638:  45%|████▌     | 2914/6470 [36:10<44:46,  1.32it/s]
4|2915|Loss: 0.8908933401107788:  45%|████▌     | 2914/6470 [36:11<44:46,  1.32it/s]
4|2915|Loss: 0.8908933401107788:  45%|████▌     | 2915/6470 [36:11<44:56,  1.32it/s]
4|2916|Loss: 0.805753231048584:  45%|████▌     | 2915/6470 [36:11<44:56,  1.32it/s] 
4|2916|Loss: 0.805753231048584:  45%|████▌     | 2916/6470 [36:12<46:37,  1.27it/s]
4|2917|Loss: 0.7554118037223816:  45%|████▌     | 2916/6470 [36:12<46:37,  1.27it/s]
4|2917|Loss: 0.7554118037223816:  45%|████▌     | 2917/6470 [36:13<48:28,  1.22it/s]
4|2918|Loss: 0.9260739684104919:  45%|████▌     | 2917/6470 [36:13<48:28,  1.22it/s]
4|2918|Loss: 0.9260739684104919:  45%|████▌     | 2918/6470 [36:14<49:35,  1.19it/s]
4|2919|Loss: 0.8676064610481262:  45%|████▌     | 2918/6470 [36:14<49:35,  1.19it/s]
4|2919|Loss: 0.8676064610481262:  45%|████▌     | 2919/6470 [36:14<49:12,  1.20it/s]
4|2920|Loss: 0.801708996295929:  45%|████▌     | 2919/6470 [36:15<49:12,  1.20it/s] 
4|2920|Loss: 0.801708996295929:  45%|████▌     | 2920/6470 [36:15<50:25,  1.17it/s]
4|2921|Loss: 0.8161838054656982:  45%|████▌     | 2920/6470 [36:16<50:25,  1.17it/s]
4|2921|Loss: 0.8161838054656982:  45%|████▌     | 2921/6470 [36:16<50:51,  1.16it/s]
4|2922|Loss: 0.8420681953430176:  45%|████▌     | 2921/6470 [36:17<50:51,  1.16it/s]
4|2922|Loss: 0.8420681953430176:  45%|████▌     | 2922/6470 [36:17<50:06,  1.18it/s]
4|2923|Loss: 0.7123846411705017:  45%|████▌     | 2922/6470 [36:17<50:06,  1.18it/s]
4|2923|Loss: 0.7123846411705017:  45%|████▌     | 2923/6470 [36:18<46:15,  1.28it/s]
4|2924|Loss: 0.8079045414924622:  45%|████▌     | 2923/6470 [36:18<46:15,  1.28it/s]
4|2924|Loss: 0.8079045414924622:  45%|████▌     | 2924/6470 [36:18<43:47,  1.35it/s]
4|2925|Loss: 0.8367817401885986:  45%|████▌     | 2924/6470 [36:19<43:47,  1.35it/s]
4|2925|Loss: 0.8367817401885986:  45%|████▌     | 2925/6470 [36:19<46:22,  1.27it/s]
4|2926|Loss: 0.8339799642562866:  45%|████▌     | 2925/6470 [36:20<46:22,  1.27it/s]
4|2926|Loss: 0.8339799642562866:  45%|████▌     | 2926/6470 [36:20<47:06,  1.25it/s]
4|2927|Loss: 0.7510016560554504:  45%|████▌     | 2926/6470 [36:20<47:06,  1.25it/s]
4|2927|Loss: 0.7510016560554504:  45%|████▌     | 2927/6470 [36:21<47:52,  1.23it/s]
4|2928|Loss: 0.8577556014060974:  45%|████▌     | 2927/6470 [36:21<47:52,  1.23it/s]
4|2928|Loss: 0.8577556014060974:  45%|████▌     | 2928/6470 [36:22<48:05,  1.23it/s]
4|2929|Loss: 0.8396115899085999:  45%|████▌     | 2928/6470 [36:22<48:05,  1.23it/s]
4|2929|Loss: 0.8396115899085999:  45%|████▌     | 2929/6470 [36:22<46:36,  1.27it/s]
4|2930|Loss: 0.6642401218414307:  45%|████▌     | 2929/6470 [36:23<46:36,  1.27it/s]
4|2930|Loss: 0.6642401218414307:  45%|████▌     | 2930/6470 [36:23<49:22,  1.19it/s]
4|2931|Loss: 0.8251063227653503:  45%|████▌     | 2930/6470 [36:24<49:22,  1.19it/s]
4|2931|Loss: 0.8251063227653503:  45%|████▌     | 2931/6470 [36:24<51:53,  1.14it/s]
4|2932|Loss: 0.884040117263794:  45%|████▌     | 2931/6470 [36:25<51:53,  1.14it/s] 
4|2932|Loss: 0.884040117263794:  45%|████▌     | 2932/6470 [36:25<50:46,  1.16it/s]
4|2933|Loss: 0.8426361083984375:  45%|████▌     | 2932/6470 [36:26<50:46,  1.16it/s]
4|2933|Loss: 0.8426361083984375:  45%|████▌     | 2933/6470 [36:26<48:59,  1.20it/s]
4|2934|Loss: 0.8096306920051575:  45%|████▌     | 2933/6470 [36:26<48:59,  1.20it/s]
4|2934|Loss: 0.8096306920051575:  45%|████▌     | 2934/6470 [36:27<46:57,  1.26it/s]
4|2935|Loss: 0.8021751642227173:  45%|████▌     | 2934/6470 [36:27<46:57,  1.26it/s]
4|2935|Loss: 0.8021751642227173:  45%|████▌     | 2935/6470 [36:27<42:09,  1.40it/s]
4|2936|Loss: 0.8619757294654846:  45%|████▌     | 2935/6470 [36:28<42:09,  1.40it/s]
4|2936|Loss: 0.8619757294654846:  45%|████▌     | 2936/6470 [36:28<42:14,  1.39it/s]
4|2937|Loss: 0.8272691369056702:  45%|████▌     | 2936/6470 [36:28<42:14,  1.39it/s]
4|2937|Loss: 0.8272691369056702:  45%|████▌     | 2937/6470 [36:29<43:42,  1.35it/s]
4|2938|Loss: 0.7022985816001892:  45%|████▌     | 2937/6470 [36:29<43:42,  1.35it/s]
4|2938|Loss: 0.7022985816001892:  45%|████▌     | 2938/6470 [36:29<42:44,  1.38it/s]
4|2939|Loss: 0.7502655386924744:  45%|████▌     | 2938/6470 [36:30<42:44,  1.38it/s]
4|2939|Loss: 0.7502655386924744:  45%|████▌     | 2939/6470 [36:30<42:54,  1.37it/s]
4|2940|Loss: 0.9250187873840332:  45%|████▌     | 2939/6470 [36:31<42:54,  1.37it/s]
4|2940|Loss: 0.9250187873840332:  45%|████▌     | 2940/6470 [36:31<44:56,  1.31it/s]
4|2941|Loss: 0.8246738314628601:  45%|████▌     | 2940/6470 [36:31<44:56,  1.31it/s]
4|2941|Loss: 0.8246738314628601:  45%|████▌     | 2941/6470 [36:32<44:05,  1.33it/s]
4|2942|Loss: 0.6428831219673157:  45%|████▌     | 2941/6470 [36:32<44:05,  1.33it/s]
4|2942|Loss: 0.6428831219673157:  45%|████▌     | 2942/6470 [36:32<41:56,  1.40it/s]
4|2943|Loss: 0.8437696695327759:  45%|████▌     | 2942/6470 [36:33<41:56,  1.40it/s]
4|2943|Loss: 0.8437696695327759:  45%|████▌     | 2943/6470 [36:33<42:06,  1.40it/s]
4|2944|Loss: 0.8927451372146606:  45%|████▌     | 2943/6470 [36:33<42:06,  1.40it/s]
4|2944|Loss: 0.8927451372146606:  46%|████▌     | 2944/6470 [36:34<42:49,  1.37it/s]
4|2945|Loss: 0.8041138648986816:  46%|████▌     | 2944/6470 [36:34<42:49,  1.37it/s]
4|2945|Loss: 0.8041138648986816:  46%|████▌     | 2945/6470 [36:34<42:32,  1.38it/s]
4|2946|Loss: 0.8749329447746277:  46%|████▌     | 2945/6470 [36:35<42:32,  1.38it/s]
4|2946|Loss: 0.8749329447746277:  46%|████▌     | 2946/6470 [36:35<43:49,  1.34it/s]
4|2947|Loss: 1.0400053262710571:  46%|████▌     | 2946/6470 [36:36<43:49,  1.34it/s]
4|2947|Loss: 1.0400053262710571:  46%|████▌     | 2947/6470 [36:36<45:56,  1.28it/s]
4|2948|Loss: 1.0967646837234497:  46%|████▌     | 2947/6470 [36:37<45:56,  1.28it/s]
4|2948|Loss: 1.0967646837234497:  46%|████▌     | 2948/6470 [36:37<46:17,  1.27it/s]
4|2949|Loss: 0.9732558131217957:  46%|████▌     | 2948/6470 [36:37<46:17,  1.27it/s]
4|2949|Loss: 0.9732558131217957:  46%|████▌     | 2949/6470 [36:38<44:35,  1.32it/s]
4|2950|Loss: 0.7546617984771729:  46%|████▌     | 2949/6470 [36:38<44:35,  1.32it/s]
4|2950|Loss: 0.7546617984771729:  46%|████▌     | 2950/6470 [36:38<41:07,  1.43it/s]
4|2951|Loss: 0.8359339833259583:  46%|████▌     | 2950/6470 [36:39<41:07,  1.43it/s]
4|2951|Loss: 0.8359339833259583:  46%|████▌     | 2951/6470 [36:39<39:59,  1.47it/s]
4|2952|Loss: 0.7928621172904968:  46%|████▌     | 2951/6470 [36:39<39:59,  1.47it/s]
4|2952|Loss: 0.7928621172904968:  46%|████▌     | 2952/6470 [36:40<39:52,  1.47it/s]
4|2953|Loss: 0.7982972860336304:  46%|████▌     | 2952/6470 [36:40<39:52,  1.47it/s]
4|2953|Loss: 0.7982972860336304:  46%|████▌     | 2953/6470 [36:40<35:51,  1.63it/s]
4|2954|Loss: 0.9171172976493835:  46%|████▌     | 2953/6470 [36:40<35:51,  1.63it/s]
4|2954|Loss: 0.9171172976493835:  46%|████▌     | 2954/6470 [36:41<35:22,  1.66it/s]
4|2955|Loss: 0.8199097514152527:  46%|████▌     | 2954/6470 [36:41<35:22,  1.66it/s]
4|2955|Loss: 0.8199097514152527:  46%|████▌     | 2955/6470 [36:41<35:31,  1.65it/s]
4|2956|Loss: 0.7437796592712402:  46%|████▌     | 2955/6470 [36:42<35:31,  1.65it/s]
4|2956|Loss: 0.7437796592712402:  46%|████▌     | 2956/6470 [36:42<38:34,  1.52it/s]
4|2957|Loss: 0.8018940091133118:  46%|████▌     | 2956/6470 [36:42<38:34,  1.52it/s]
4|2957|Loss: 0.8018940091133118:  46%|████▌     | 2957/6470 [36:43<42:44,  1.37it/s]
4|2958|Loss: 0.8512210249900818:  46%|████▌     | 2957/6470 [36:43<42:44,  1.37it/s]
4|2958|Loss: 0.8512210249900818:  46%|████▌     | 2958/6470 [36:44<46:25,  1.26it/s]
4|2959|Loss: 0.8208275437355042:  46%|████▌     | 2958/6470 [36:44<46:25,  1.26it/s]
4|2959|Loss: 0.8208275437355042:  46%|████▌     | 2959/6470 [36:45<47:22,  1.24it/s]
4|2960|Loss: 0.9309270977973938:  46%|████▌     | 2959/6470 [36:45<47:22,  1.24it/s]
4|2960|Loss: 0.9309270977973938:  46%|████▌     | 2960/6470 [36:45<46:42,  1.25it/s]
4|2961|Loss: 0.9396733045578003:  46%|████▌     | 2960/6470 [36:46<46:42,  1.25it/s]
4|2961|Loss: 0.9396733045578003:  46%|████▌     | 2961/6470 [36:46<42:44,  1.37it/s]
4|2962|Loss: 0.9636439681053162:  46%|████▌     | 2961/6470 [36:46<42:44,  1.37it/s]
4|2962|Loss: 0.9636439681053162:  46%|████▌     | 2962/6470 [36:47<42:25,  1.38it/s]
4|2963|Loss: 0.8312050700187683:  46%|████▌     | 2962/6470 [36:47<42:25,  1.38it/s]
4|2963|Loss: 0.8312050700187683:  46%|████▌     | 2963/6470 [36:48<44:45,  1.31it/s]
4|2964|Loss: 0.879051685333252:  46%|████▌     | 2963/6470 [36:48<44:45,  1.31it/s] 
4|2964|Loss: 0.879051685333252:  46%|████▌     | 2964/6470 [36:48<47:00,  1.24it/s]
4|2965|Loss: 0.8932474851608276:  46%|████▌     | 2964/6470 [36:49<47:00,  1.24it/s]
4|2965|Loss: 0.8932474851608276:  46%|████▌     | 2965/6470 [36:49<47:05,  1.24it/s]
4|2966|Loss: 0.8311605453491211:  46%|████▌     | 2965/6470 [36:50<47:05,  1.24it/s]
4|2966|Loss: 0.8311605453491211:  46%|████▌     | 2966/6470 [36:50<47:00,  1.24it/s]
4|2967|Loss: 0.7318195104598999:  46%|████▌     | 2966/6470 [36:51<47:00,  1.24it/s]
4|2967|Loss: 0.7318195104598999:  46%|████▌     | 2967/6470 [36:51<48:14,  1.21it/s]
4|2968|Loss: 0.7477838397026062:  46%|████▌     | 2967/6470 [36:51<48:14,  1.21it/s]
4|2968|Loss: 0.7477838397026062:  46%|████▌     | 2968/6470 [36:52<50:21,  1.16it/s]
4|2969|Loss: 0.8649635910987854:  46%|████▌     | 2968/6470 [36:52<50:21,  1.16it/s]
4|2969|Loss: 0.8649635910987854:  46%|████▌     | 2969/6470 [36:53<50:11,  1.16it/s]
4|2970|Loss: 0.7622089982032776:  46%|████▌     | 2969/6470 [36:53<50:11,  1.16it/s]
4|2970|Loss: 0.7622089982032776:  46%|████▌     | 2970/6470 [36:54<48:49,  1.19it/s]
4|2971|Loss: 0.7696132659912109:  46%|████▌     | 2970/6470 [36:54<48:49,  1.19it/s]
4|2971|Loss: 0.7696132659912109:  46%|████▌     | 2971/6470 [36:54<45:30,  1.28it/s]
4|2972|Loss: 0.8248559832572937:  46%|████▌     | 2971/6470 [36:55<45:30,  1.28it/s]
4|2972|Loss: 0.8248559832572937:  46%|████▌     | 2972/6470 [36:55<46:32,  1.25it/s]
4|2973|Loss: 0.7829344272613525:  46%|████▌     | 2972/6470 [36:55<46:32,  1.25it/s]
4|2973|Loss: 0.7829344272613525:  46%|████▌     | 2973/6470 [36:56<46:40,  1.25it/s]
4|2974|Loss: 0.663936197757721:  46%|████▌     | 2973/6470 [36:56<46:40,  1.25it/s] 
4|2974|Loss: 0.663936197757721:  46%|████▌     | 2974/6470 [36:57<50:06,  1.16it/s]
4|2975|Loss: 0.8846868276596069:  46%|████▌     | 2974/6470 [36:57<50:06,  1.16it/s]
4|2975|Loss: 0.8846868276596069:  46%|████▌     | 2975/6470 [36:58<50:08,  1.16it/s]
4|2976|Loss: 0.8590793013572693:  46%|████▌     | 2975/6470 [36:58<50:08,  1.16it/s]
4|2976|Loss: 0.8590793013572693:  46%|████▌     | 2976/6470 [36:58<47:35,  1.22it/s]
4|2977|Loss: 0.8421075344085693:  46%|████▌     | 2976/6470 [36:59<47:35,  1.22it/s]
4|2977|Loss: 0.8421075344085693:  46%|████▌     | 2977/6470 [36:59<46:19,  1.26it/s]
4|2978|Loss: 0.776922345161438:  46%|████▌     | 2977/6470 [37:00<46:19,  1.26it/s] 
4|2978|Loss: 0.776922345161438:  46%|████▌     | 2978/6470 [37:00<44:37,  1.30it/s]
4|2979|Loss: 0.8489578366279602:  46%|████▌     | 2978/6470 [37:00<44:37,  1.30it/s]
4|2979|Loss: 0.8489578366279602:  46%|████▌     | 2979/6470 [37:01<44:57,  1.29it/s]
4|2980|Loss: 0.8064841628074646:  46%|████▌     | 2979/6470 [37:01<44:57,  1.29it/s]
4|2980|Loss: 0.8064841628074646:  46%|████▌     | 2980/6470 [37:01<43:58,  1.32it/s]
4|2981|Loss: 0.815989077091217:  46%|████▌     | 2980/6470 [37:02<43:58,  1.32it/s] 
4|2981|Loss: 0.815989077091217:  46%|████▌     | 2981/6470 [37:02<43:47,  1.33it/s]
4|2982|Loss: 0.8211648464202881:  46%|████▌     | 2981/6470 [37:03<43:47,  1.33it/s]
4|2982|Loss: 0.8211648464202881:  46%|████▌     | 2982/6470 [37:03<43:52,  1.32it/s]
4|2983|Loss: 0.8256873488426208:  46%|████▌     | 2982/6470 [37:03<43:52,  1.32it/s]
4|2983|Loss: 0.8256873488426208:  46%|████▌     | 2983/6470 [37:03<42:17,  1.37it/s]
4|2984|Loss: 0.927087128162384:  46%|████▌     | 2983/6470 [37:04<42:17,  1.37it/s] 
4|2984|Loss: 0.927087128162384:  46%|████▌     | 2984/6470 [37:04<43:10,  1.35it/s]
4|2985|Loss: 0.728580117225647:  46%|████▌     | 2984/6470 [37:05<43:10,  1.35it/s]
4|2985|Loss: 0.728580117225647:  46%|████▌     | 2985/6470 [37:05<44:30,  1.30it/s]
4|2986|Loss: 0.9302948713302612:  46%|████▌     | 2985/6470 [37:06<44:30,  1.30it/s]
4|2986|Loss: 0.9302948713302612:  46%|████▌     | 2986/6470 [37:06<47:01,  1.23it/s]
4|2987|Loss: 0.7631440758705139:  46%|████▌     | 2986/6470 [37:07<47:01,  1.23it/s]
4|2987|Loss: 0.7631440758705139:  46%|████▌     | 2987/6470 [37:07<49:39,  1.17it/s]
4|2988|Loss: 0.6417965888977051:  46%|████▌     | 2987/6470 [37:08<49:39,  1.17it/s]
4|2988|Loss: 0.6417965888977051:  46%|████▌     | 2988/6470 [37:08<53:16,  1.09it/s]
4|2989|Loss: 0.8119456768035889:  46%|████▌     | 2988/6470 [37:09<53:16,  1.09it/s]
4|2989|Loss: 0.8119456768035889:  46%|████▌     | 2989/6470 [37:09<50:51,  1.14it/s]
4|2990|Loss: 0.8013116121292114:  46%|████▌     | 2989/6470 [37:09<50:51,  1.14it/s]
4|2990|Loss: 0.8013116121292114:  46%|████▌     | 2990/6470 [37:10<48:43,  1.19it/s]
4|2991|Loss: 0.896291971206665:  46%|████▌     | 2990/6470 [37:10<48:43,  1.19it/s] 
4|2991|Loss: 0.896291971206665:  46%|████▌     | 2991/6470 [37:10<46:20,  1.25it/s]
4|2992|Loss: 0.6940135359764099:  46%|████▌     | 2991/6470 [37:11<46:20,  1.25it/s]
4|2992|Loss: 0.6940135359764099:  46%|████▌     | 2992/6470 [37:11<44:14,  1.31it/s]
4|2993|Loss: 0.8037525415420532:  46%|████▌     | 2992/6470 [37:11<44:14,  1.31it/s]
4|2993|Loss: 0.8037525415420532:  46%|████▋     | 2993/6470 [37:12<45:28,  1.27it/s]
4|2994|Loss: 0.7089758515357971:  46%|████▋     | 2993/6470 [37:12<45:28,  1.27it/s]
4|2994|Loss: 0.7089758515357971:  46%|████▋     | 2994/6470 [37:13<45:58,  1.26it/s]
4|2995|Loss: 0.9825396537780762:  46%|████▋     | 2994/6470 [37:13<45:58,  1.26it/s]
4|2995|Loss: 0.9825396537780762:  46%|████▋     | 2995/6470 [37:13<44:53,  1.29it/s]
4|2996|Loss: 0.7548432350158691:  46%|████▋     | 2995/6470 [37:14<44:53,  1.29it/s]
4|2996|Loss: 0.7548432350158691:  46%|████▋     | 2996/6470 [37:14<44:57,  1.29it/s]
4|2997|Loss: 0.7687451243400574:  46%|████▋     | 2996/6470 [37:15<44:57,  1.29it/s]
4|2997|Loss: 0.7687451243400574:  46%|████▋     | 2997/6470 [37:15<42:48,  1.35it/s]
4|2998|Loss: 0.7458398938179016:  46%|████▋     | 2997/6470 [37:15<42:48,  1.35it/s]
4|2998|Loss: 0.7458398938179016:  46%|████▋     | 2998/6470 [37:15<41:14,  1.40it/s]
4|2999|Loss: 0.6684101223945618:  46%|████▋     | 2998/6470 [37:16<41:14,  1.40it/s]
4|2999|Loss: 0.6684101223945618:  46%|████▋     | 2999/6470 [37:16<38:49,  1.49it/s]
4|3000|Loss: 0.8432095646858215:  46%|████▋     | 2999/6470 [37:16<38:49,  1.49it/s]
4|3000|Loss: 0.8432095646858215:  46%|████▋     | 3000/6470 [37:17<37:35,  1.54it/s]
4|3001|Loss: 0.8865488767623901:  46%|████▋     | 3000/6470 [37:17<37:35,  1.54it/s]
4|3001|Loss: 0.8865488767623901:  46%|████▋     | 3001/6470 [37:17<36:15,  1.59it/s]
4|3002|Loss: 0.8445498943328857:  46%|████▋     | 3001/6470 [37:18<36:15,  1.59it/s]
4|3002|Loss: 0.8445498943328857:  46%|████▋     | 3002/6470 [37:18<37:46,  1.53it/s]
4|3003|Loss: 0.846489667892456:  46%|████▋     | 3002/6470 [37:18<37:46,  1.53it/s] 
4|3003|Loss: 0.846489667892456:  46%|████▋     | 3003/6470 [37:19<42:17,  1.37it/s]
4|3004|Loss: 0.9348365068435669:  46%|████▋     | 3003/6470 [37:19<42:17,  1.37it/s]
4|3004|Loss: 0.9348365068435669:  46%|████▋     | 3004/6470 [37:20<42:29,  1.36it/s]
4|3005|Loss: 0.7533111572265625:  46%|████▋     | 3004/6470 [37:20<42:29,  1.36it/s]
4|3005|Loss: 0.7533111572265625:  46%|████▋     | 3005/6470 [37:20<42:43,  1.35it/s]
4|3006|Loss: 0.7721236944198608:  46%|████▋     | 3005/6470 [37:21<42:43,  1.35it/s]
4|3006|Loss: 0.7721236944198608:  46%|████▋     | 3006/6470 [37:21<43:29,  1.33it/s]
4|3007|Loss: 0.8024240732192993:  46%|████▋     | 3006/6470 [37:21<43:29,  1.33it/s]
4|3007|Loss: 0.8024240732192993:  46%|████▋     | 3007/6470 [37:22<39:58,  1.44it/s]
4|3008|Loss: 0.9327068328857422:  46%|████▋     | 3007/6470 [37:22<39:58,  1.44it/s]
4|3008|Loss: 0.9327068328857422:  46%|████▋     | 3008/6470 [37:22<40:23,  1.43it/s]
4|3009|Loss: 0.7359787821769714:  46%|████▋     | 3008/6470 [37:23<40:23,  1.43it/s]
4|3009|Loss: 0.7359787821769714:  47%|████▋     | 3009/6470 [37:23<39:45,  1.45it/s]
4|3010|Loss: 0.7181810736656189:  47%|████▋     | 3009/6470 [37:23<39:45,  1.45it/s]
4|3010|Loss: 0.7181810736656189:  47%|████▋     | 3010/6470 [37:24<36:37,  1.57it/s]
4|3011|Loss: 0.7064319849014282:  47%|████▋     | 3010/6470 [37:24<36:37,  1.57it/s]
4|3011|Loss: 0.7064319849014282:  47%|████▋     | 3011/6470 [37:24<37:38,  1.53it/s]
4|3012|Loss: 0.7233595848083496:  47%|████▋     | 3011/6470 [37:25<37:38,  1.53it/s]
4|3012|Loss: 0.7233595848083496:  47%|████▋     | 3012/6470 [37:25<38:28,  1.50it/s]
4|3013|Loss: 0.8146321177482605:  47%|████▋     | 3012/6470 [37:25<38:28,  1.50it/s]
4|3013|Loss: 0.8146321177482605:  47%|████▋     | 3013/6470 [37:26<40:18,  1.43it/s]
4|3014|Loss: 0.8252258896827698:  47%|████▋     | 3013/6470 [37:26<40:18,  1.43it/s]
4|3014|Loss: 0.8252258896827698:  47%|████▋     | 3014/6470 [37:26<41:32,  1.39it/s]
4|3015|Loss: 0.8989232778549194:  47%|████▋     | 3014/6470 [37:27<41:32,  1.39it/s]
4|3015|Loss: 0.8989232778549194:  47%|████▋     | 3015/6470 [37:27<36:56,  1.56it/s]
4|3016|Loss: 0.9371543526649475:  47%|████▋     | 3015/6470 [37:27<36:56,  1.56it/s]
4|3016|Loss: 0.9371543526649475:  47%|████▋     | 3016/6470 [37:27<35:02,  1.64it/s]
4|3017|Loss: 0.9110721945762634:  47%|████▋     | 3016/6470 [37:28<35:02,  1.64it/s]
4|3017|Loss: 0.9110721945762634:  47%|████▋     | 3017/6470 [37:28<33:57,  1.69it/s]
4|3018|Loss: 0.8004748821258545:  47%|████▋     | 3017/6470 [37:28<33:57,  1.69it/s]
4|3018|Loss: 0.8004748821258545:  47%|████▋     | 3018/6470 [37:29<34:01,  1.69it/s]
4|3019|Loss: 0.8785986304283142:  47%|████▋     | 3018/6470 [37:29<34:01,  1.69it/s]
4|3019|Loss: 0.8785986304283142:  47%|████▋     | 3019/6470 [37:29<37:02,  1.55it/s]
4|3020|Loss: 0.8510487079620361:  47%|████▋     | 3019/6470 [37:30<37:02,  1.55it/s]
4|3020|Loss: 0.8510487079620361:  47%|████▋     | 3020/6470 [37:30<36:55,  1.56it/s]
4|3021|Loss: 0.7075170874595642:  47%|████▋     | 3020/6470 [37:30<36:55,  1.56it/s]
4|3021|Loss: 0.7075170874595642:  47%|████▋     | 3021/6470 [37:31<38:53,  1.48it/s]
4|3022|Loss: 0.6625862121582031:  47%|████▋     | 3021/6470 [37:31<38:53,  1.48it/s]
4|3022|Loss: 0.6625862121582031:  47%|████▋     | 3022/6470 [37:32<40:50,  1.41it/s]
4|3023|Loss: 1.0944373607635498:  47%|████▋     | 3022/6470 [37:32<40:50,  1.41it/s]
4|3023|Loss: 1.0944373607635498:  47%|████▋     | 3023/6470 [37:32<40:59,  1.40it/s]
4|3024|Loss: 0.6988651156425476:  47%|████▋     | 3023/6470 [37:33<40:59,  1.40it/s]
4|3024|Loss: 0.6988651156425476:  47%|████▋     | 3024/6470 [37:33<40:34,  1.42it/s]
4|3025|Loss: 0.7282577157020569:  47%|████▋     | 3024/6470 [37:33<40:34,  1.42it/s]
4|3025|Loss: 0.7282577157020569:  47%|████▋     | 3025/6470 [37:33<37:26,  1.53it/s]
4|3026|Loss: 1.04373037815094:  47%|████▋     | 3025/6470 [37:34<37:26,  1.53it/s]  
4|3026|Loss: 1.04373037815094:  47%|████▋     | 3026/6470 [37:34<41:32,  1.38it/s]
4|3027|Loss: 0.6307310461997986:  47%|████▋     | 3026/6470 [37:35<41:32,  1.38it/s]
4|3027|Loss: 0.6307310461997986:  47%|████▋     | 3027/6470 [37:35<40:38,  1.41it/s]
4|3028|Loss: 0.7369455695152283:  47%|████▋     | 3027/6470 [37:35<40:38,  1.41it/s]
4|3028|Loss: 0.7369455695152283:  47%|████▋     | 3028/6470 [37:36<42:26,  1.35it/s]
4|3029|Loss: 0.8122168183326721:  47%|████▋     | 3028/6470 [37:36<42:26,  1.35it/s]
4|3029|Loss: 0.8122168183326721:  47%|████▋     | 3029/6470 [37:37<45:07,  1.27it/s]
4|3030|Loss: 0.7545482516288757:  47%|████▋     | 3029/6470 [37:37<45:07,  1.27it/s]
4|3030|Loss: 0.7545482516288757:  47%|████▋     | 3030/6470 [37:37<43:17,  1.32it/s]
4|3031|Loss: 0.8467237949371338:  47%|████▋     | 3030/6470 [37:38<43:17,  1.32it/s]
4|3031|Loss: 0.8467237949371338:  47%|████▋     | 3031/6470 [37:38<45:20,  1.26it/s]
4|3032|Loss: 0.9051183462142944:  47%|████▋     | 3031/6470 [37:39<45:20,  1.26it/s]
4|3032|Loss: 0.9051183462142944:  47%|████▋     | 3032/6470 [37:39<45:55,  1.25it/s]
4|3033|Loss: 0.8772674798965454:  47%|████▋     | 3032/6470 [37:40<45:55,  1.25it/s]
4|3033|Loss: 0.8772674798965454:  47%|████▋     | 3033/6470 [37:40<47:23,  1.21it/s]
4|3034|Loss: 0.8082059621810913:  47%|████▋     | 3033/6470 [37:40<47:23,  1.21it/s]
4|3034|Loss: 0.8082059621810913:  47%|████▋     | 3034/6470 [37:41<44:34,  1.28it/s]
4|3035|Loss: 0.7260330319404602:  47%|████▋     | 3034/6470 [37:41<44:34,  1.28it/s]
4|3035|Loss: 0.7260330319404602:  47%|████▋     | 3035/6470 [37:41<43:33,  1.31it/s]
4|3036|Loss: 0.8723663687705994:  47%|████▋     | 3035/6470 [37:42<43:33,  1.31it/s]
4|3036|Loss: 0.8723663687705994:  47%|████▋     | 3036/6470 [37:42<40:49,  1.40it/s]
4|3037|Loss: 0.8243576884269714:  47%|████▋     | 3036/6470 [37:42<40:49,  1.40it/s]
4|3037|Loss: 0.8243576884269714:  47%|████▋     | 3037/6470 [37:43<41:22,  1.38it/s]
4|3038|Loss: 0.8393977284431458:  47%|████▋     | 3037/6470 [37:43<41:22,  1.38it/s]
4|3038|Loss: 0.8393977284431458:  47%|████▋     | 3038/6470 [37:43<40:57,  1.40it/s]
4|3039|Loss: 0.7808014154434204:  47%|████▋     | 3038/6470 [37:44<40:57,  1.40it/s]
4|3039|Loss: 0.7808014154434204:  47%|████▋     | 3039/6470 [37:44<42:09,  1.36it/s]
4|3040|Loss: 0.7581902742385864:  47%|████▋     | 3039/6470 [37:45<42:09,  1.36it/s]
4|3040|Loss: 0.7581902742385864:  47%|████▋     | 3040/6470 [37:45<42:34,  1.34it/s]
4|3041|Loss: 0.9201527237892151:  47%|████▋     | 3040/6470 [37:45<42:34,  1.34it/s]
4|3041|Loss: 0.9201527237892151:  47%|████▋     | 3041/6470 [37:46<43:16,  1.32it/s]
4|3042|Loss: 0.6973669528961182:  47%|████▋     | 3041/6470 [37:46<43:16,  1.32it/s]
4|3042|Loss: 0.6973669528961182:  47%|████▋     | 3042/6470 [37:46<38:17,  1.49it/s]
4|3043|Loss: 0.926749587059021:  47%|████▋     | 3042/6470 [37:47<38:17,  1.49it/s] 
4|3043|Loss: 0.926749587059021:  47%|████▋     | 3043/6470 [37:47<37:25,  1.53it/s]
4|3044|Loss: 0.8057200908660889:  47%|████▋     | 3043/6470 [37:47<37:25,  1.53it/s]
4|3044|Loss: 0.8057200908660889:  47%|████▋     | 3044/6470 [37:48<37:07,  1.54it/s]
4|3045|Loss: 0.8026477694511414:  47%|████▋     | 3044/6470 [37:48<37:07,  1.54it/s]
4|3045|Loss: 0.8026477694511414:  47%|████▋     | 3045/6470 [37:48<38:52,  1.47it/s]
4|3046|Loss: 0.7317373752593994:  47%|████▋     | 3045/6470 [37:49<38:52,  1.47it/s]
4|3046|Loss: 0.7317373752593994:  47%|████▋     | 3046/6470 [37:49<40:37,  1.40it/s]
4|3047|Loss: 0.9024986624717712:  47%|████▋     | 3046/6470 [37:49<40:37,  1.40it/s]
4|3047|Loss: 0.9024986624717712:  47%|████▋     | 3047/6470 [37:50<40:09,  1.42it/s]
4|3048|Loss: 0.7864229679107666:  47%|████▋     | 3047/6470 [37:50<40:09,  1.42it/s]
4|3048|Loss: 0.7864229679107666:  47%|████▋     | 3048/6470 [37:50<38:43,  1.47it/s]
4|3049|Loss: 0.732460081577301:  47%|████▋     | 3048/6470 [37:51<38:43,  1.47it/s] 
4|3049|Loss: 0.732460081577301:  47%|████▋     | 3049/6470 [37:51<40:55,  1.39it/s]
4|3050|Loss: 0.8347632884979248:  47%|████▋     | 3049/6470 [37:52<40:55,  1.39it/s]
4|3050|Loss: 0.8347632884979248:  47%|████▋     | 3050/6470 [37:52<42:31,  1.34it/s]
4|3051|Loss: 0.7265677452087402:  47%|████▋     | 3050/6470 [37:52<42:31,  1.34it/s]
4|3051|Loss: 0.7265677452087402:  47%|████▋     | 3051/6470 [37:53<42:21,  1.35it/s]
4|3052|Loss: 0.7070634365081787:  47%|████▋     | 3051/6470 [37:53<42:21,  1.35it/s]
4|3052|Loss: 0.7070634365081787:  47%|████▋     | 3052/6470 [37:53<36:54,  1.54it/s]
4|3053|Loss: 0.8166489005088806:  47%|████▋     | 3052/6470 [37:53<36:54,  1.54it/s]
4|3053|Loss: 0.8166489005088806:  47%|████▋     | 3053/6470 [37:54<35:52,  1.59it/s]
4|3054|Loss: 0.8326999545097351:  47%|████▋     | 3053/6470 [37:54<35:52,  1.59it/s]
4|3054|Loss: 0.8326999545097351:  47%|████▋     | 3054/6470 [37:55<39:53,  1.43it/s]
4|3055|Loss: 0.8223328590393066:  47%|████▋     | 3054/6470 [37:55<39:53,  1.43it/s]
4|3055|Loss: 0.8223328590393066:  47%|████▋     | 3055/6470 [37:55<38:40,  1.47it/s]
4|3056|Loss: 0.8446487188339233:  47%|████▋     | 3055/6470 [37:56<38:40,  1.47it/s]
4|3056|Loss: 0.8446487188339233:  47%|████▋     | 3056/6470 [37:56<35:57,  1.58it/s]
4|3057|Loss: 0.9933039546012878:  47%|████▋     | 3056/6470 [37:56<35:57,  1.58it/s]
4|3057|Loss: 0.9933039546012878:  47%|████▋     | 3057/6470 [37:56<37:03,  1.54it/s]
4|3058|Loss: 0.7958163619041443:  47%|████▋     | 3057/6470 [37:57<37:03,  1.54it/s]
4|3058|Loss: 0.7958163619041443:  47%|████▋     | 3058/6470 [37:57<40:07,  1.42it/s]
4|3059|Loss: 1.0026919841766357:  47%|████▋     | 3058/6470 [37:58<40:07,  1.42it/s]
4|3059|Loss: 1.0026919841766357:  47%|████▋     | 3059/6470 [37:58<40:24,  1.41it/s]
4|3060|Loss: 0.7073928713798523:  47%|████▋     | 3059/6470 [37:58<40:24,  1.41it/s]
4|3060|Loss: 0.7073928713798523:  47%|████▋     | 3060/6470 [37:59<39:24,  1.44it/s]
4|3061|Loss: 0.8165919184684753:  47%|████▋     | 3060/6470 [37:59<39:24,  1.44it/s]
4|3061|Loss: 0.8165919184684753:  47%|████▋     | 3061/6470 [37:59<38:43,  1.47it/s]
4|3062|Loss: 0.8109993934631348:  47%|████▋     | 3061/6470 [38:00<38:43,  1.47it/s]
4|3062|Loss: 0.8109993934631348:  47%|████▋     | 3062/6470 [38:00<37:48,  1.50it/s]
4|3063|Loss: 0.8677054643630981:  47%|████▋     | 3062/6470 [38:00<37:48,  1.50it/s]
4|3063|Loss: 0.8677054643630981:  47%|████▋     | 3063/6470 [38:01<41:10,  1.38it/s]
4|3064|Loss: 0.8570853471755981:  47%|████▋     | 3063/6470 [38:01<41:10,  1.38it/s]
4|3064|Loss: 0.8570853471755981:  47%|████▋     | 3064/6470 [38:02<44:26,  1.28it/s]
4|3065|Loss: 0.8908810615539551:  47%|████▋     | 3064/6470 [38:02<44:26,  1.28it/s]
4|3065|Loss: 0.8908810615539551:  47%|████▋     | 3065/6470 [38:02<41:39,  1.36it/s]
4|3066|Loss: 0.838407039642334:  47%|████▋     | 3065/6470 [38:03<41:39,  1.36it/s] 
4|3066|Loss: 0.838407039642334:  47%|████▋     | 3066/6470 [38:03<41:43,  1.36it/s]
4|3067|Loss: 0.7944332957267761:  47%|████▋     | 3066/6470 [38:03<41:43,  1.36it/s]
4|3067|Loss: 0.7944332957267761:  47%|████▋     | 3067/6470 [38:04<37:30,  1.51it/s]
4|3068|Loss: 0.7532986998558044:  47%|████▋     | 3067/6470 [38:04<37:30,  1.51it/s]
4|3068|Loss: 0.7532986998558044:  47%|████▋     | 3068/6470 [38:04<36:58,  1.53it/s]
4|3069|Loss: 0.8218178749084473:  47%|████▋     | 3068/6470 [38:05<36:58,  1.53it/s]
4|3069|Loss: 0.8218178749084473:  47%|████▋     | 3069/6470 [38:05<38:24,  1.48it/s]
4|3070|Loss: 0.8885159492492676:  47%|████▋     | 3069/6470 [38:05<38:24,  1.48it/s]
4|3070|Loss: 0.8885159492492676:  47%|████▋     | 3070/6470 [38:05<36:15,  1.56it/s]
4|3071|Loss: 0.9102148413658142:  47%|████▋     | 3070/6470 [38:06<36:15,  1.56it/s]
4|3071|Loss: 0.9102148413658142:  47%|████▋     | 3071/6470 [38:06<36:33,  1.55it/s]
4|3072|Loss: 0.8210762143135071:  47%|████▋     | 3071/6470 [38:07<36:33,  1.55it/s]
4|3072|Loss: 0.8210762143135071:  47%|████▋     | 3072/6470 [38:07<36:41,  1.54it/s]
4|3073|Loss: 0.7401929497718811:  47%|████▋     | 3072/6470 [38:07<36:41,  1.54it/s]
4|3073|Loss: 0.7401929497718811:  47%|████▋     | 3073/6470 [38:08<39:24,  1.44it/s]
4|3074|Loss: 0.7841460108757019:  47%|████▋     | 3073/6470 [38:08<39:24,  1.44it/s]
4|3074|Loss: 0.7841460108757019:  48%|████▊     | 3074/6470 [38:08<39:45,  1.42it/s]
4|3075|Loss: 0.8768562078475952:  48%|████▊     | 3074/6470 [38:09<39:45,  1.42it/s]
4|3075|Loss: 0.8768562078475952:  48%|████▊     | 3075/6470 [38:09<39:47,  1.42it/s]
4|3076|Loss: 0.8523976802825928:  48%|████▊     | 3075/6470 [38:09<39:47,  1.42it/s]
4|3076|Loss: 0.8523976802825928:  48%|████▊     | 3076/6470 [38:10<40:38,  1.39it/s]
4|3077|Loss: 0.7995539307594299:  48%|████▊     | 3076/6470 [38:10<40:38,  1.39it/s]
4|3077|Loss: 0.7995539307594299:  48%|████▊     | 3077/6470 [38:10<35:23,  1.60it/s]
4|3078|Loss: 0.9432051777839661:  48%|████▊     | 3077/6470 [38:11<35:23,  1.60it/s]
4|3078|Loss: 0.9432051777839661:  48%|████▊     | 3078/6470 [38:11<37:09,  1.52it/s]
4|3079|Loss: 0.8212354779243469:  48%|████▊     | 3078/6470 [38:11<37:09,  1.52it/s]
4|3079|Loss: 0.8212354779243469:  48%|████▊     | 3079/6470 [38:12<40:04,  1.41it/s]
4|3080|Loss: 0.9183156490325928:  48%|████▊     | 3079/6470 [38:12<40:04,  1.41it/s]
4|3080|Loss: 0.9183156490325928:  48%|████▊     | 3080/6470 [38:13<43:39,  1.29it/s]
4|3081|Loss: 0.8099742531776428:  48%|████▊     | 3080/6470 [38:13<43:39,  1.29it/s]
4|3081|Loss: 0.8099742531776428:  48%|████▊     | 3081/6470 [38:13<40:22,  1.40it/s]
4|3082|Loss: 0.7206655740737915:  48%|████▊     | 3081/6470 [38:14<40:22,  1.40it/s]
4|3082|Loss: 0.7206655740737915:  48%|████▊     | 3082/6470 [38:14<40:41,  1.39it/s]
4|3083|Loss: 0.8645305037498474:  48%|████▊     | 3082/6470 [38:14<40:41,  1.39it/s]
4|3083|Loss: 0.8645305037498474:  48%|████▊     | 3083/6470 [38:15<40:51,  1.38it/s]
4|3084|Loss: 0.8824486136436462:  48%|████▊     | 3083/6470 [38:15<40:51,  1.38it/s]
4|3084|Loss: 0.8824486136436462:  48%|████▊     | 3084/6470 [38:15<39:55,  1.41it/s]
4|3085|Loss: 0.8519957661628723:  48%|████▊     | 3084/6470 [38:16<39:55,  1.41it/s]
4|3085|Loss: 0.8519957661628723:  48%|████▊     | 3085/6470 [38:16<43:01,  1.31it/s]
4|3086|Loss: 0.9370125532150269:  48%|████▊     | 3085/6470 [38:17<43:01,  1.31it/s]
4|3086|Loss: 0.9370125532150269:  48%|████▊     | 3086/6470 [38:17<42:46,  1.32it/s]
4|3087|Loss: 0.7780715823173523:  48%|████▊     | 3086/6470 [38:17<42:46,  1.32it/s]
4|3087|Loss: 0.7780715823173523:  48%|████▊     | 3087/6470 [38:18<41:41,  1.35it/s]
4|3088|Loss: 0.860231876373291:  48%|████▊     | 3087/6470 [38:18<41:41,  1.35it/s] 
4|3088|Loss: 0.860231876373291:  48%|████▊     | 3088/6470 [38:19<42:35,  1.32it/s]
4|3089|Loss: 0.893086850643158:  48%|████▊     | 3088/6470 [38:19<42:35,  1.32it/s]
4|3089|Loss: 0.893086850643158:  48%|████▊     | 3089/6470 [38:19<43:39,  1.29it/s]
4|3090|Loss: 0.9097044467926025:  48%|████▊     | 3089/6470 [38:20<43:39,  1.29it/s]
4|3090|Loss: 0.9097044467926025:  48%|████▊     | 3090/6470 [38:20<44:08,  1.28it/s]
4|3091|Loss: 0.7502738237380981:  48%|████▊     | 3090/6470 [38:21<44:08,  1.28it/s]
4|3091|Loss: 0.7502738237380981:  48%|████▊     | 3091/6470 [38:21<44:36,  1.26it/s]
4|3092|Loss: 0.7359820604324341:  48%|████▊     | 3091/6470 [38:21<44:36,  1.26it/s]
4|3092|Loss: 0.7359820604324341:  48%|████▊     | 3092/6470 [38:22<44:10,  1.27it/s]
4|3093|Loss: 0.8927397131919861:  48%|████▊     | 3092/6470 [38:22<44:10,  1.27it/s]
4|3093|Loss: 0.8927397131919861:  48%|████▊     | 3093/6470 [38:22<43:42,  1.29it/s]
4|3094|Loss: 0.9123629331588745:  48%|████▊     | 3093/6470 [38:23<43:42,  1.29it/s]
4|3094|Loss: 0.9123629331588745:  48%|████▊     | 3094/6470 [38:23<41:45,  1.35it/s]
4|3095|Loss: 0.8992792963981628:  48%|████▊     | 3094/6470 [38:24<41:45,  1.35it/s]
4|3095|Loss: 0.8992792963981628:  48%|████▊     | 3095/6470 [38:24<44:05,  1.28it/s]
4|3096|Loss: 0.6508603692054749:  48%|████▊     | 3095/6470 [38:24<44:05,  1.28it/s]
4|3096|Loss: 0.6508603692054749:  48%|████▊     | 3096/6470 [38:25<41:45,  1.35it/s]
4|3097|Loss: 0.8760615587234497:  48%|████▊     | 3096/6470 [38:25<41:45,  1.35it/s]
4|3097|Loss: 0.8760615587234497:  48%|████▊     | 3097/6470 [38:25<38:38,  1.45it/s]
4|3098|Loss: 0.796010434627533:  48%|████▊     | 3097/6470 [38:26<38:38,  1.45it/s] 
4|3098|Loss: 0.796010434627533:  48%|████▊     | 3098/6470 [38:26<37:51,  1.48it/s]
4|3099|Loss: 0.8164262771606445:  48%|████▊     | 3098/6470 [38:26<37:51,  1.48it/s]
4|3099|Loss: 0.8164262771606445:  48%|████▊     | 3099/6470 [38:27<39:26,  1.42it/s]
4|3100|Loss: 0.831942081451416:  48%|████▊     | 3099/6470 [38:27<39:26,  1.42it/s] 
4|3100|Loss: 0.831942081451416:  48%|████▊     | 3100/6470 [38:27<39:49,  1.41it/s]
4|3101|Loss: 0.7639443874359131:  48%|████▊     | 3100/6470 [38:28<39:49,  1.41it/s]
4|3101|Loss: 0.7639443874359131:  48%|████▊     | 3101/6470 [38:28<41:09,  1.36it/s]
4|3102|Loss: 0.8190954923629761:  48%|████▊     | 3101/6470 [38:29<41:09,  1.36it/s]
4|3102|Loss: 0.8190954923629761:  48%|████▊     | 3102/6470 [38:29<43:51,  1.28it/s]
4|3103|Loss: 0.6972643733024597:  48%|████▊     | 3102/6470 [38:29<43:51,  1.28it/s]
4|3103|Loss: 0.6972643733024597:  48%|████▊     | 3103/6470 [38:30<40:12,  1.40it/s]
4|3104|Loss: 0.9726200103759766:  48%|████▊     | 3103/6470 [38:30<40:12,  1.40it/s]
4|3104|Loss: 0.9726200103759766:  48%|████▊     | 3104/6470 [38:30<41:54,  1.34it/s]
4|3105|Loss: 0.8873673677444458:  48%|████▊     | 3104/6470 [38:31<41:54,  1.34it/s]
4|3105|Loss: 0.8873673677444458:  48%|████▊     | 3105/6470 [38:31<42:25,  1.32it/s]
4|3106|Loss: 0.784872829914093:  48%|████▊     | 3105/6470 [38:32<42:25,  1.32it/s] 
4|3106|Loss: 0.784872829914093:  48%|████▊     | 3106/6470 [38:32<41:13,  1.36it/s]
4|3107|Loss: 0.8598688840866089:  48%|████▊     | 3106/6470 [38:32<41:13,  1.36it/s]
4|3107|Loss: 0.8598688840866089:  48%|████▊     | 3107/6470 [38:33<39:44,  1.41it/s]
4|3108|Loss: 0.816703736782074:  48%|████▊     | 3107/6470 [38:33<39:44,  1.41it/s] 
4|3108|Loss: 0.816703736782074:  48%|████▊     | 3108/6470 [38:33<40:29,  1.38it/s]
4|3109|Loss: 0.7877517938613892:  48%|████▊     | 3108/6470 [38:34<40:29,  1.38it/s]
4|3109|Loss: 0.7877517938613892:  48%|████▊     | 3109/6470 [38:34<39:56,  1.40it/s]
4|3110|Loss: 0.936525821685791:  48%|████▊     | 3109/6470 [38:34<39:56,  1.40it/s] 
4|3110|Loss: 0.936525821685791:  48%|████▊     | 3110/6470 [38:35<40:14,  1.39it/s]
4|3111|Loss: 0.8711588978767395:  48%|████▊     | 3110/6470 [38:35<40:14,  1.39it/s]
4|3111|Loss: 0.8711588978767395:  48%|████▊     | 3111/6470 [38:35<37:28,  1.49it/s]
4|3112|Loss: 0.8780704140663147:  48%|████▊     | 3111/6470 [38:36<37:28,  1.49it/s]
4|3112|Loss: 0.8780704140663147:  48%|████▊     | 3112/6470 [38:36<36:28,  1.53it/s]
4|3113|Loss: 0.754207193851471:  48%|████▊     | 3112/6470 [38:36<36:28,  1.53it/s] 
4|3113|Loss: 0.754207193851471:  48%|████▊     | 3113/6470 [38:37<37:53,  1.48it/s]
4|3114|Loss: 0.7474137544631958:  48%|████▊     | 3113/6470 [38:37<37:53,  1.48it/s]
4|3114|Loss: 0.7474137544631958:  48%|████▊     | 3114/6470 [38:37<36:55,  1.51it/s]
4|3115|Loss: 0.7790315747261047:  48%|████▊     | 3114/6470 [38:38<36:55,  1.51it/s]
4|3115|Loss: 0.7790315747261047:  48%|████▊     | 3115/6470 [38:38<34:59,  1.60it/s]
4|3116|Loss: 0.8555548787117004:  48%|████▊     | 3115/6470 [38:38<34:59,  1.60it/s]
4|3116|Loss: 0.8555548787117004:  48%|████▊     | 3116/6470 [38:39<37:00,  1.51it/s]
4|3117|Loss: 0.9004371166229248:  48%|████▊     | 3116/6470 [38:39<37:00,  1.51it/s]
4|3117|Loss: 0.9004371166229248:  48%|████▊     | 3117/6470 [38:39<41:10,  1.36it/s]
4|3118|Loss: 0.8778249621391296:  48%|████▊     | 3117/6470 [38:40<41:10,  1.36it/s]
4|3118|Loss: 0.8778249621391296:  48%|████▊     | 3118/6470 [38:40<42:57,  1.30it/s]
4|3119|Loss: 0.6886575222015381:  48%|████▊     | 3118/6470 [38:41<42:57,  1.30it/s]
4|3119|Loss: 0.6886575222015381:  48%|████▊     | 3119/6470 [38:41<43:41,  1.28it/s]
4|3120|Loss: 0.9664125442504883:  48%|████▊     | 3119/6470 [38:42<43:41,  1.28it/s]
4|3120|Loss: 0.9664125442504883:  48%|████▊     | 3120/6470 [38:42<42:41,  1.31it/s]
4|3121|Loss: 0.7241983413696289:  48%|████▊     | 3120/6470 [38:42<42:41,  1.31it/s]
4|3121|Loss: 0.7241983413696289:  48%|████▊     | 3121/6470 [38:43<44:53,  1.24it/s]
4|3122|Loss: 0.9077198505401611:  48%|████▊     | 3121/6470 [38:43<44:53,  1.24it/s]
4|3122|Loss: 0.9077198505401611:  48%|████▊     | 3122/6470 [38:43<42:36,  1.31it/s]
4|3123|Loss: 0.6851974129676819:  48%|████▊     | 3122/6470 [38:44<42:36,  1.31it/s]
4|3123|Loss: 0.6851974129676819:  48%|████▊     | 3123/6470 [38:44<40:01,  1.39it/s]
4|3124|Loss: 0.9050506353378296:  48%|████▊     | 3123/6470 [38:44<40:01,  1.39it/s]
4|3124|Loss: 0.9050506353378296:  48%|████▊     | 3124/6470 [38:45<42:24,  1.32it/s]
4|3125|Loss: 0.943760097026825:  48%|████▊     | 3124/6470 [38:45<42:24,  1.32it/s] 
4|3125|Loss: 0.943760097026825:  48%|████▊     | 3125/6470 [38:45<38:02,  1.47it/s]
4|3126|Loss: 0.9365827441215515:  48%|████▊     | 3125/6470 [38:46<38:02,  1.47it/s]
4|3126|Loss: 0.9365827441215515:  48%|████▊     | 3126/6470 [38:46<41:20,  1.35it/s]
4|3127|Loss: 0.7178665995597839:  48%|████▊     | 3126/6470 [38:47<41:20,  1.35it/s]
4|3127|Loss: 0.7178665995597839:  48%|████▊     | 3127/6470 [38:47<40:40,  1.37it/s]
4|3128|Loss: 0.9771003723144531:  48%|████▊     | 3127/6470 [38:47<40:40,  1.37it/s]
4|3128|Loss: 0.9771003723144531:  48%|████▊     | 3128/6470 [38:48<41:10,  1.35it/s]
4|3129|Loss: 0.8458777070045471:  48%|████▊     | 3128/6470 [38:48<41:10,  1.35it/s]
4|3129|Loss: 0.8458777070045471:  48%|████▊     | 3129/6470 [38:49<42:28,  1.31it/s]
4|3130|Loss: 0.8951753973960876:  48%|████▊     | 3129/6470 [38:49<42:28,  1.31it/s]
4|3130|Loss: 0.8951753973960876:  48%|████▊     | 3130/6470 [38:49<43:19,  1.29it/s]
4|3131|Loss: 0.747677206993103:  48%|████▊     | 3130/6470 [38:50<43:19,  1.29it/s] 
4|3131|Loss: 0.747677206993103:  48%|████▊     | 3131/6470 [38:50<43:05,  1.29it/s]
4|3132|Loss: 0.897205650806427:  48%|████▊     | 3131/6470 [38:50<43:05,  1.29it/s]
4|3132|Loss: 0.897205650806427:  48%|████▊     | 3132/6470 [38:51<38:56,  1.43it/s]
4|3133|Loss: 0.7118703722953796:  48%|████▊     | 3132/6470 [38:51<38:56,  1.43it/s]
4|3133|Loss: 0.7118703722953796:  48%|████▊     | 3133/6470 [38:51<41:29,  1.34it/s]
4|3134|Loss: 0.9248408079147339:  48%|████▊     | 3133/6470 [38:52<41:29,  1.34it/s]
4|3134|Loss: 0.9248408079147339:  48%|████▊     | 3134/6470 [38:52<44:21,  1.25it/s]
4|3135|Loss: 0.7942559123039246:  48%|████▊     | 3134/6470 [38:53<44:21,  1.25it/s]
4|3135|Loss: 0.7942559123039246:  48%|████▊     | 3135/6470 [38:53<42:23,  1.31it/s]
4|3136|Loss: 0.8621853590011597:  48%|████▊     | 3135/6470 [38:53<42:23,  1.31it/s]
4|3136|Loss: 0.8621853590011597:  48%|████▊     | 3136/6470 [38:54<42:24,  1.31it/s]
4|3137|Loss: 0.7184581160545349:  48%|████▊     | 3136/6470 [38:54<42:24,  1.31it/s]
4|3137|Loss: 0.7184581160545349:  48%|████▊     | 3137/6470 [38:54<39:51,  1.39it/s]
4|3138|Loss: 0.7431229948997498:  48%|████▊     | 3137/6470 [38:55<39:51,  1.39it/s]
4|3138|Loss: 0.7431229948997498:  49%|████▊     | 3138/6470 [38:55<38:22,  1.45it/s]
4|3139|Loss: 0.9689367413520813:  49%|████▊     | 3138/6470 [38:56<38:22,  1.45it/s]
4|3139|Loss: 0.9689367413520813:  49%|████▊     | 3139/6470 [38:56<41:41,  1.33it/s]
4|3140|Loss: 0.8647716641426086:  49%|████▊     | 3139/6470 [38:56<41:41,  1.33it/s]
4|3140|Loss: 0.8647716641426086:  49%|████▊     | 3140/6470 [38:57<43:12,  1.28it/s]
4|3141|Loss: 0.7874706387519836:  49%|████▊     | 3140/6470 [38:57<43:12,  1.28it/s]
4|3141|Loss: 0.7874706387519836:  49%|████▊     | 3141/6470 [38:58<45:42,  1.21it/s]
4|3142|Loss: 0.8358125686645508:  49%|████▊     | 3141/6470 [38:58<45:42,  1.21it/s]
4|3142|Loss: 0.8358125686645508:  49%|████▊     | 3142/6470 [38:59<49:27,  1.12it/s]
4|3143|Loss: 0.666445255279541:  49%|████▊     | 3142/6470 [38:59<49:27,  1.12it/s] 
4|3143|Loss: 0.666445255279541:  49%|████▊     | 3143/6470 [39:00<46:54,  1.18it/s]
4|3144|Loss: 0.7321047186851501:  49%|████▊     | 3143/6470 [39:00<46:54,  1.18it/s]
4|3144|Loss: 0.7321047186851501:  49%|████▊     | 3144/6470 [39:00<44:21,  1.25it/s]
4|3145|Loss: 0.7292829155921936:  49%|████▊     | 3144/6470 [39:01<44:21,  1.25it/s]
4|3145|Loss: 0.7292829155921936:  49%|████▊     | 3145/6470 [39:01<42:23,  1.31it/s]
4|3146|Loss: 0.8313056230545044:  49%|████▊     | 3145/6470 [39:01<42:23,  1.31it/s]
4|3146|Loss: 0.8313056230545044:  49%|████▊     | 3146/6470 [39:02<41:18,  1.34it/s]
4|3147|Loss: 0.8256532549858093:  49%|████▊     | 3146/6470 [39:02<41:18,  1.34it/s]
4|3147|Loss: 0.8256532549858093:  49%|████▊     | 3147/6470 [39:02<43:35,  1.27it/s]
4|3148|Loss: 0.7222918272018433:  49%|████▊     | 3147/6470 [39:03<43:35,  1.27it/s]
4|3148|Loss: 0.7222918272018433:  49%|████▊     | 3148/6470 [39:03<41:21,  1.34it/s]
4|3149|Loss: 0.8607268929481506:  49%|████▊     | 3148/6470 [39:04<41:21,  1.34it/s]
4|3149|Loss: 0.8607268929481506:  49%|████▊     | 3149/6470 [39:04<40:57,  1.35it/s]
4|3150|Loss: 0.7882922291755676:  49%|████▊     | 3149/6470 [39:04<40:57,  1.35it/s]
4|3150|Loss: 0.7882922291755676:  49%|████▊     | 3150/6470 [39:05<39:37,  1.40it/s]
4|3151|Loss: 0.8337056040763855:  49%|████▊     | 3150/6470 [39:05<39:37,  1.40it/s]
4|3151|Loss: 0.8337056040763855:  49%|████▊     | 3151/6470 [39:05<41:33,  1.33it/s]
4|3152|Loss: 0.8471019268035889:  49%|████▊     | 3151/6470 [39:06<41:33,  1.33it/s]
4|3152|Loss: 0.8471019268035889:  49%|████▊     | 3152/6470 [39:06<39:06,  1.41it/s]
4|3153|Loss: 0.8804700374603271:  49%|████▊     | 3152/6470 [39:06<39:06,  1.41it/s]
4|3153|Loss: 0.8804700374603271:  49%|████▊     | 3153/6470 [39:07<39:55,  1.38it/s]
4|3154|Loss: 0.7712244391441345:  49%|████▊     | 3153/6470 [39:07<39:55,  1.38it/s]
4|3154|Loss: 0.7712244391441345:  49%|████▊     | 3154/6470 [39:07<39:39,  1.39it/s]
4|3155|Loss: 0.8052846789360046:  49%|████▊     | 3154/6470 [39:08<39:39,  1.39it/s]
4|3155|Loss: 0.8052846789360046:  49%|████▉     | 3155/6470 [39:08<41:37,  1.33it/s]
4|3156|Loss: 0.9468854069709778:  49%|████▉     | 3155/6470 [39:09<41:37,  1.33it/s]
4|3156|Loss: 0.9468854069709778:  49%|████▉     | 3156/6470 [39:09<43:29,  1.27it/s]
4|3157|Loss: 0.8008507490158081:  49%|████▉     | 3156/6470 [39:10<43:29,  1.27it/s]
4|3157|Loss: 0.8008507490158081:  49%|████▉     | 3157/6470 [39:10<42:13,  1.31it/s]
4|3158|Loss: 0.8306863903999329:  49%|████▉     | 3157/6470 [39:10<42:13,  1.31it/s]
4|3158|Loss: 0.8306863903999329:  49%|████▉     | 3158/6470 [39:11<40:57,  1.35it/s]
4|3159|Loss: 0.675607442855835:  49%|████▉     | 3158/6470 [39:11<40:57,  1.35it/s] 
4|3159|Loss: 0.675607442855835:  49%|████▉     | 3159/6470 [39:11<35:29,  1.55it/s]
4|3160|Loss: 0.7711327075958252:  49%|████▉     | 3159/6470 [39:11<35:29,  1.55it/s]
4|3160|Loss: 0.7711327075958252:  49%|████▉     | 3160/6470 [39:12<35:39,  1.55it/s]
4|3161|Loss: 0.7402481436729431:  49%|████▉     | 3160/6470 [39:12<35:39,  1.55it/s]
4|3161|Loss: 0.7402481436729431:  49%|████▉     | 3161/6470 [39:12<38:01,  1.45it/s]
4|3162|Loss: 0.6116653084754944:  49%|████▉     | 3161/6470 [39:13<38:01,  1.45it/s]
4|3162|Loss: 0.6116653084754944:  49%|████▉     | 3162/6470 [39:13<43:24,  1.27it/s]
4|3163|Loss: 0.7445148229598999:  49%|████▉     | 3162/6470 [39:14<43:24,  1.27it/s]
4|3163|Loss: 0.7445148229598999:  49%|████▉     | 3163/6470 [39:14<44:40,  1.23it/s]
4|3164|Loss: 0.8176466822624207:  49%|████▉     | 3163/6470 [39:15<44:40,  1.23it/s]
4|3164|Loss: 0.8176466822624207:  49%|████▉     | 3164/6470 [39:15<43:03,  1.28it/s]
4|3165|Loss: 0.7843199372291565:  49%|████▉     | 3164/6470 [39:15<43:03,  1.28it/s]
4|3165|Loss: 0.7843199372291565:  49%|████▉     | 3165/6470 [39:16<41:22,  1.33it/s]
4|3166|Loss: 0.8803490996360779:  49%|████▉     | 3165/6470 [39:16<41:22,  1.33it/s]
4|3166|Loss: 0.8803490996360779:  49%|████▉     | 3166/6470 [39:16<41:57,  1.31it/s]
4|3167|Loss: 0.8568823933601379:  49%|████▉     | 3166/6470 [39:17<41:57,  1.31it/s]
4|3167|Loss: 0.8568823933601379:  49%|████▉     | 3167/6470 [39:17<41:14,  1.33it/s]
4|3168|Loss: 0.6624377965927124:  49%|████▉     | 3167/6470 [39:18<41:14,  1.33it/s]
4|3168|Loss: 0.6624377965927124:  49%|████▉     | 3168/6470 [39:18<38:12,  1.44it/s]
4|3169|Loss: 0.784694492816925:  49%|████▉     | 3168/6470 [39:18<38:12,  1.44it/s] 
4|3169|Loss: 0.784694492816925:  49%|████▉     | 3169/6470 [39:18<36:43,  1.50it/s]
4|3170|Loss: 0.8769009113311768:  49%|████▉     | 3169/6470 [39:19<36:43,  1.50it/s]
4|3170|Loss: 0.8769009113311768:  49%|████▉     | 3170/6470 [39:19<36:40,  1.50it/s]
4|3171|Loss: 0.8435786962509155:  49%|████▉     | 3170/6470 [39:19<36:40,  1.50it/s]
4|3171|Loss: 0.8435786962509155:  49%|████▉     | 3171/6470 [39:20<38:35,  1.42it/s]
4|3172|Loss: 0.9196377396583557:  49%|████▉     | 3171/6470 [39:20<38:35,  1.42it/s]
4|3172|Loss: 0.9196377396583557:  49%|████▉     | 3172/6470 [39:21<41:52,  1.31it/s]
4|3173|Loss: 0.8880317211151123:  49%|████▉     | 3172/6470 [39:21<41:52,  1.31it/s]
4|3173|Loss: 0.8880317211151123:  49%|████▉     | 3173/6470 [39:22<42:40,  1.29it/s]
4|3174|Loss: 0.8128491640090942:  49%|████▉     | 3173/6470 [39:22<42:40,  1.29it/s]
4|3174|Loss: 0.8128491640090942:  49%|████▉     | 3174/6470 [39:22<43:54,  1.25it/s]
4|3175|Loss: 0.8474805355072021:  49%|████▉     | 3174/6470 [39:23<43:54,  1.25it/s]
4|3175|Loss: 0.8474805355072021:  49%|████▉     | 3175/6470 [39:23<45:42,  1.20it/s]
4|3176|Loss: 0.8833900690078735:  49%|████▉     | 3175/6470 [39:24<45:42,  1.20it/s]
4|3176|Loss: 0.8833900690078735:  49%|████▉     | 3176/6470 [39:24<45:18,  1.21it/s]
4|3177|Loss: 0.7662325501441956:  49%|████▉     | 3176/6470 [39:24<45:18,  1.21it/s]
4|3177|Loss: 0.7662325501441956:  49%|████▉     | 3177/6470 [39:25<43:00,  1.28it/s]
4|3178|Loss: 0.7954133152961731:  49%|████▉     | 3177/6470 [39:25<43:00,  1.28it/s]
4|3178|Loss: 0.7954133152961731:  49%|████▉     | 3178/6470 [39:26<43:46,  1.25it/s]
4|3179|Loss: 0.845893383026123:  49%|████▉     | 3178/6470 [39:26<43:46,  1.25it/s] 
4|3179|Loss: 0.845893383026123:  49%|████▉     | 3179/6470 [39:26<45:30,  1.21it/s]
4|3180|Loss: 0.85565584897995:  49%|████▉     | 3179/6470 [39:27<45:30,  1.21it/s] 
4|3180|Loss: 0.85565584897995:  49%|████▉     | 3180/6470 [39:27<44:00,  1.25it/s]
4|3181|Loss: 0.6472722291946411:  49%|████▉     | 3180/6470 [39:28<44:00,  1.25it/s]
4|3181|Loss: 0.6472722291946411:  49%|████▉     | 3181/6470 [39:28<45:11,  1.21it/s]
4|3182|Loss: 0.8567503690719604:  49%|████▉     | 3181/6470 [39:29<45:11,  1.21it/s]
4|3182|Loss: 0.8567503690719604:  49%|████▉     | 3182/6470 [39:29<45:58,  1.19it/s]
4|3183|Loss: 0.8880468606948853:  49%|████▉     | 3182/6470 [39:29<45:58,  1.19it/s]
4|3183|Loss: 0.8880468606948853:  49%|████▉     | 3183/6470 [39:30<43:53,  1.25it/s]
4|3184|Loss: 0.7931177616119385:  49%|████▉     | 3183/6470 [39:30<43:53,  1.25it/s]
4|3184|Loss: 0.7931177616119385:  49%|████▉     | 3184/6470 [39:31<46:58,  1.17it/s]
4|3185|Loss: 0.9187384247779846:  49%|████▉     | 3184/6470 [39:31<46:58,  1.17it/s]
4|3185|Loss: 0.9187384247779846:  49%|████▉     | 3185/6470 [39:31<44:48,  1.22it/s]
4|3186|Loss: 0.8849648833274841:  49%|████▉     | 3185/6470 [39:32<44:48,  1.22it/s]
4|3186|Loss: 0.8849648833274841:  49%|████▉     | 3186/6470 [39:32<45:26,  1.20it/s]
4|3187|Loss: 0.756079375743866:  49%|████▉     | 3186/6470 [39:33<45:26,  1.20it/s] 
4|3187|Loss: 0.756079375743866:  49%|████▉     | 3187/6470 [39:33<42:08,  1.30it/s]
4|3188|Loss: 0.6914311647415161:  49%|████▉     | 3187/6470 [39:33<42:08,  1.30it/s]
4|3188|Loss: 0.6914311647415161:  49%|████▉     | 3188/6470 [39:34<41:36,  1.31it/s]
4|3189|Loss: 0.8944915533065796:  49%|████▉     | 3188/6470 [39:34<41:36,  1.31it/s]
4|3189|Loss: 0.8944915533065796:  49%|████▉     | 3189/6470 [39:35<43:28,  1.26it/s]
4|3190|Loss: 0.7979593276977539:  49%|████▉     | 3189/6470 [39:35<43:28,  1.26it/s]
4|3190|Loss: 0.7979593276977539:  49%|████▉     | 3190/6470 [39:35<43:12,  1.26it/s]
4|3191|Loss: 0.6735543012619019:  49%|████▉     | 3190/6470 [39:36<43:12,  1.26it/s]
4|3191|Loss: 0.6735543012619019:  49%|████▉     | 3191/6470 [39:36<42:56,  1.27it/s]
4|3192|Loss: 0.8463695049285889:  49%|████▉     | 3191/6470 [39:36<42:56,  1.27it/s]
4|3192|Loss: 0.8463695049285889:  49%|████▉     | 3192/6470 [39:37<42:40,  1.28it/s]
4|3193|Loss: 0.8923929333686829:  49%|████▉     | 3192/6470 [39:37<42:40,  1.28it/s]
4|3193|Loss: 0.8923929333686829:  49%|████▉     | 3193/6470 [39:38<44:15,  1.23it/s]
4|3194|Loss: 0.843174159526825:  49%|████▉     | 3193/6470 [39:38<44:15,  1.23it/s] 
4|3194|Loss: 0.843174159526825:  49%|████▉     | 3194/6470 [39:39<45:49,  1.19it/s]
4|3195|Loss: 0.7553281188011169:  49%|████▉     | 3194/6470 [39:39<45:49,  1.19it/s]
4|3195|Loss: 0.7553281188011169:  49%|████▉     | 3195/6470 [39:39<44:22,  1.23it/s]
4|3196|Loss: 0.8390216827392578:  49%|████▉     | 3195/6470 [39:40<44:22,  1.23it/s]
4|3196|Loss: 0.8390216827392578:  49%|████▉     | 3196/6470 [39:40<40:21,  1.35it/s]
4|3197|Loss: 0.882040798664093:  49%|████▉     | 3196/6470 [39:40<40:21,  1.35it/s] 
4|3197|Loss: 0.882040798664093:  49%|████▉     | 3197/6470 [39:41<42:28,  1.28it/s]
4|3198|Loss: 0.8952052593231201:  49%|████▉     | 3197/6470 [39:41<42:28,  1.28it/s]
4|3198|Loss: 0.8952052593231201:  49%|████▉     | 3198/6470 [39:42<44:51,  1.22it/s]
4|3199|Loss: 0.6676508188247681:  49%|████▉     | 3198/6470 [39:42<44:51,  1.22it/s]
4|3199|Loss: 0.6676508188247681:  49%|████▉     | 3199/6470 [39:43<47:27,  1.15it/s]
4|3200|Loss: 0.8615162968635559:  49%|████▉     | 3199/6470 [39:43<47:27,  1.15it/s]
4|3200|Loss: 0.8615162968635559:  49%|████▉     | 3200/6470 [39:43<45:40,  1.19it/s]
4|3201|Loss: 0.8746494650840759:  49%|████▉     | 3200/6470 [39:44<45:40,  1.19it/s]
4|3201|Loss: 0.8746494650840759:  49%|████▉     | 3201/6470 [39:44<45:36,  1.19it/s]
4|3202|Loss: 0.7579627633094788:  49%|████▉     | 3201/6470 [39:45<45:36,  1.19it/s]
4|3202|Loss: 0.7579627633094788:  49%|████▉     | 3202/6470 [39:45<43:22,  1.26it/s]
4|3203|Loss: 0.7954134345054626:  49%|████▉     | 3202/6470 [39:45<43:22,  1.26it/s]
4|3203|Loss: 0.7954134345054626:  50%|████▉     | 3203/6470 [39:46<41:05,  1.33it/s]
4|3204|Loss: 0.8618353605270386:  50%|████▉     | 3203/6470 [39:46<41:05,  1.33it/s]
4|3204|Loss: 0.8618353605270386:  50%|████▉     | 3204/6470 [39:47<43:43,  1.25it/s]
4|3205|Loss: 0.8946722149848938:  50%|████▉     | 3204/6470 [39:47<43:43,  1.25it/s]
4|3205|Loss: 0.8946722149848938:  50%|████▉     | 3205/6470 [39:47<44:14,  1.23it/s]
4|3206|Loss: 0.8666535019874573:  50%|████▉     | 3205/6470 [39:48<44:14,  1.23it/s]
4|3206|Loss: 0.8666535019874573:  50%|████▉     | 3206/6470 [39:48<43:59,  1.24it/s]
4|3207|Loss: 1.0362125635147095:  50%|████▉     | 3206/6470 [39:49<43:59,  1.24it/s]
4|3207|Loss: 1.0362125635147095:  50%|████▉     | 3207/6470 [39:49<45:30,  1.20it/s]
4|3208|Loss: 0.7829562425613403:  50%|████▉     | 3207/6470 [39:50<45:30,  1.20it/s]
4|3208|Loss: 0.7829562425613403:  50%|████▉     | 3208/6470 [39:50<42:58,  1.27it/s]
4|3209|Loss: 0.8304684162139893:  50%|████▉     | 3208/6470 [39:50<42:58,  1.27it/s]
4|3209|Loss: 0.8304684162139893:  50%|████▉     | 3209/6470 [39:51<42:24,  1.28it/s]
4|3210|Loss: 0.7868226170539856:  50%|████▉     | 3209/6470 [39:51<42:24,  1.28it/s]
4|3210|Loss: 0.7868226170539856:  50%|████▉     | 3210/6470 [39:52<45:22,  1.20it/s]
4|3211|Loss: 0.8434018492698669:  50%|████▉     | 3210/6470 [39:52<45:22,  1.20it/s]
4|3211|Loss: 0.8434018492698669:  50%|████▉     | 3211/6470 [39:52<45:20,  1.20it/s]
4|3212|Loss: 0.8577150702476501:  50%|████▉     | 3211/6470 [39:53<45:20,  1.20it/s]
4|3212|Loss: 0.8577150702476501:  50%|████▉     | 3212/6470 [39:53<43:29,  1.25it/s]
4|3213|Loss: 0.7738864421844482:  50%|████▉     | 3212/6470 [39:53<43:29,  1.25it/s]
4|3213|Loss: 0.7738864421844482:  50%|████▉     | 3213/6470 [39:54<41:56,  1.29it/s]
4|3214|Loss: 0.7281693816184998:  50%|████▉     | 3213/6470 [39:54<41:56,  1.29it/s]
4|3214|Loss: 0.7281693816184998:  50%|████▉     | 3214/6470 [39:54<38:43,  1.40it/s]
4|3215|Loss: 0.7065607309341431:  50%|████▉     | 3214/6470 [39:55<38:43,  1.40it/s]
4|3215|Loss: 0.7065607309341431:  50%|████▉     | 3215/6470 [39:55<38:01,  1.43it/s]
4|3216|Loss: 0.9046570062637329:  50%|████▉     | 3215/6470 [39:56<38:01,  1.43it/s]
4|3216|Loss: 0.9046570062637329:  50%|████▉     | 3216/6470 [39:56<41:49,  1.30it/s]
4|3217|Loss: 0.6552952527999878:  50%|████▉     | 3216/6470 [39:56<41:49,  1.30it/s]
4|3217|Loss: 0.6552952527999878:  50%|████▉     | 3217/6470 [39:57<39:37,  1.37it/s]
4|3218|Loss: 0.8240653276443481:  50%|████▉     | 3217/6470 [39:57<39:37,  1.37it/s]
4|3218|Loss: 0.8240653276443481:  50%|████▉     | 3218/6470 [39:57<41:19,  1.31it/s]
4|3219|Loss: 0.8254079222679138:  50%|████▉     | 3218/6470 [39:58<41:19,  1.31it/s]
4|3219|Loss: 0.8254079222679138:  50%|████▉     | 3219/6470 [39:58<41:33,  1.30it/s]
4|3220|Loss: 0.9173709154129028:  50%|████▉     | 3219/6470 [39:59<41:33,  1.30it/s]
4|3220|Loss: 0.9173709154129028:  50%|████▉     | 3220/6470 [39:59<42:00,  1.29it/s]
4|3221|Loss: 0.8744500875473022:  50%|████▉     | 3220/6470 [39:59<42:00,  1.29it/s]
4|3221|Loss: 0.8744500875473022:  50%|████▉     | 3221/6470 [40:00<41:01,  1.32it/s]
4|3222|Loss: 0.7968320846557617:  50%|████▉     | 3221/6470 [40:00<41:01,  1.32it/s]
4|3222|Loss: 0.7968320846557617:  50%|████▉     | 3222/6470 [40:01<41:15,  1.31it/s]
4|3223|Loss: 0.7124857306480408:  50%|████▉     | 3222/6470 [40:01<41:15,  1.31it/s]
4|3223|Loss: 0.7124857306480408:  50%|████▉     | 3223/6470 [40:01<36:29,  1.48it/s]
4|3224|Loss: 0.946109414100647:  50%|████▉     | 3223/6470 [40:01<36:29,  1.48it/s] 
4|3224|Loss: 0.946109414100647:  50%|████▉     | 3224/6470 [40:02<36:51,  1.47it/s]
4|3225|Loss: 1.0837515592575073:  50%|████▉     | 3224/6470 [40:02<36:51,  1.47it/s]
4|3225|Loss: 1.0837515592575073:  50%|████▉     | 3225/6470 [40:03<40:28,  1.34it/s]
4|3226|Loss: 0.7454371452331543:  50%|████▉     | 3225/6470 [40:03<40:28,  1.34it/s]
4|3226|Loss: 0.7454371452331543:  50%|████▉     | 3226/6470 [40:03<41:12,  1.31it/s]
4|3227|Loss: 0.884649395942688:  50%|████▉     | 3226/6470 [40:04<41:12,  1.31it/s] 
4|3227|Loss: 0.884649395942688:  50%|████▉     | 3227/6470 [40:04<41:41,  1.30it/s]
4|3228|Loss: 0.737372100353241:  50%|████▉     | 3227/6470 [40:05<41:41,  1.30it/s]
4|3228|Loss: 0.737372100353241:  50%|████▉     | 3228/6470 [40:05<41:36,  1.30it/s]
4|3229|Loss: 0.9870430827140808:  50%|████▉     | 3228/6470 [40:05<41:36,  1.30it/s]
4|3229|Loss: 0.9870430827140808:  50%|████▉     | 3229/6470 [40:06<41:57,  1.29it/s]
4|3230|Loss: 0.8109663724899292:  50%|████▉     | 3229/6470 [40:06<41:57,  1.29it/s]
4|3230|Loss: 0.8109663724899292:  50%|████▉     | 3230/6470 [40:06<40:09,  1.34it/s]
4|3231|Loss: 0.9586156010627747:  50%|████▉     | 3230/6470 [40:07<40:09,  1.34it/s]
4|3231|Loss: 0.9586156010627747:  50%|████▉     | 3231/6470 [40:07<37:04,  1.46it/s]
4|3232|Loss: 0.8581384420394897:  50%|████▉     | 3231/6470 [40:07<37:04,  1.46it/s]
4|3232|Loss: 0.8581384420394897:  50%|████▉     | 3232/6470 [40:08<39:28,  1.37it/s]
4|3233|Loss: 0.8457409739494324:  50%|████▉     | 3232/6470 [40:08<39:28,  1.37it/s]
4|3233|Loss: 0.8457409739494324:  50%|████▉     | 3233/6470 [40:09<40:15,  1.34it/s]
4|3234|Loss: 0.8735829591751099:  50%|████▉     | 3233/6470 [40:09<40:15,  1.34it/s]
4|3234|Loss: 0.8735829591751099:  50%|████▉     | 3234/6470 [40:09<39:40,  1.36it/s]
4|3235|Loss: 0.9288548231124878:  50%|████▉     | 3234/6470 [40:10<39:40,  1.36it/s]
4|3235|Loss: 0.9288548231124878:  50%|█████     | 3235/6470 [40:10<40:16,  1.34it/s]
4|3236|Loss: 0.8109672665596008:  50%|█████     | 3235/6470 [40:10<40:16,  1.34it/s]
4|3236|Loss: 0.8109672665596008:  50%|█████     | 3236/6470 [40:11<40:57,  1.32it/s]
4|3237|Loss: 0.7613577246665955:  50%|█████     | 3236/6470 [40:11<40:57,  1.32it/s]
4|3237|Loss: 0.7613577246665955:  50%|█████     | 3237/6470 [40:12<41:04,  1.31it/s]
4|3238|Loss: 0.8409156799316406:  50%|█████     | 3237/6470 [40:12<41:04,  1.31it/s]
4|3238|Loss: 0.8409156799316406:  50%|█████     | 3238/6470 [40:12<42:18,  1.27it/s]
4|3239|Loss: 0.6286566257476807:  50%|█████     | 3238/6470 [40:13<42:18,  1.27it/s]
4|3239|Loss: 0.6286566257476807:  50%|█████     | 3239/6470 [40:13<46:18,  1.16it/s]
4|3240|Loss: 0.5569819211959839:  50%|█████     | 3239/6470 [40:14<46:18,  1.16it/s]
4|3240|Loss: 0.5569819211959839:  50%|█████     | 3240/6470 [40:14<46:29,  1.16it/s]
4|3241|Loss: 0.854308545589447:  50%|█████     | 3240/6470 [40:15<46:29,  1.16it/s] 
4|3241|Loss: 0.854308545589447:  50%|█████     | 3241/6470 [40:15<45:46,  1.18it/s]
4|3242|Loss: 0.6971675157546997:  50%|█████     | 3241/6470 [40:16<45:46,  1.18it/s]
4|3242|Loss: 0.6971675157546997:  50%|█████     | 3242/6470 [40:16<44:56,  1.20it/s]
4|3243|Loss: 0.8417400121688843:  50%|█████     | 3242/6470 [40:16<44:56,  1.20it/s]
4|3243|Loss: 0.8417400121688843:  50%|█████     | 3243/6470 [40:17<43:03,  1.25it/s]
4|3244|Loss: 0.8584304451942444:  50%|█████     | 3243/6470 [40:17<43:03,  1.25it/s]
4|3244|Loss: 0.8584304451942444:  50%|█████     | 3244/6470 [40:17<42:18,  1.27it/s]
4|3245|Loss: 0.7972713708877563:  50%|█████     | 3244/6470 [40:18<42:18,  1.27it/s]
4|3245|Loss: 0.7972713708877563:  50%|█████     | 3245/6470 [40:18<36:56,  1.45it/s]
4|3246|Loss: 0.8100211024284363:  50%|█████     | 3245/6470 [40:18<36:56,  1.45it/s]
4|3246|Loss: 0.8100211024284363:  50%|█████     | 3246/6470 [40:19<36:08,  1.49it/s]
4|3247|Loss: 0.8173936605453491:  50%|█████     | 3246/6470 [40:19<36:08,  1.49it/s]
4|3247|Loss: 0.8173936605453491:  50%|█████     | 3247/6470 [40:19<36:42,  1.46it/s]
4|3248|Loss: 0.8306129574775696:  50%|█████     | 3247/6470 [40:20<36:42,  1.46it/s]
4|3248|Loss: 0.8306129574775696:  50%|█████     | 3248/6470 [40:20<37:04,  1.45it/s]
4|3249|Loss: 1.0380702018737793:  50%|█████     | 3248/6470 [40:20<37:04,  1.45it/s]
4|3249|Loss: 1.0380702018737793:  50%|█████     | 3249/6470 [40:21<39:25,  1.36it/s]
4|3250|Loss: 0.7777626514434814:  50%|█████     | 3249/6470 [40:21<39:25,  1.36it/s]
4|3250|Loss: 0.7777626514434814:  50%|█████     | 3250/6470 [40:22<40:43,  1.32it/s]
4|3251|Loss: 0.7761080861091614:  50%|█████     | 3250/6470 [40:22<40:43,  1.32it/s]
4|3251|Loss: 0.7761080861091614:  50%|█████     | 3251/6470 [40:22<42:26,  1.26it/s]
4|3252|Loss: 0.8949984908103943:  50%|█████     | 3251/6470 [40:23<42:26,  1.26it/s]
4|3252|Loss: 0.8949984908103943:  50%|█████     | 3252/6470 [40:23<42:44,  1.25it/s]
4|3253|Loss: 0.8160423636436462:  50%|█████     | 3252/6470 [40:24<42:44,  1.25it/s]
4|3253|Loss: 0.8160423636436462:  50%|█████     | 3253/6470 [40:24<42:16,  1.27it/s]
4|3254|Loss: 0.8489552140235901:  50%|█████     | 3253/6470 [40:25<42:16,  1.27it/s]
4|3254|Loss: 0.8489552140235901:  50%|█████     | 3254/6470 [40:25<44:10,  1.21it/s]
4|3255|Loss: 0.9172426462173462:  50%|█████     | 3254/6470 [40:25<44:10,  1.21it/s]
4|3255|Loss: 0.9172426462173462:  50%|█████     | 3255/6470 [40:26<43:46,  1.22it/s]
4|3256|Loss: 0.8530573844909668:  50%|█████     | 3255/6470 [40:26<43:46,  1.22it/s]
4|3256|Loss: 0.8530573844909668:  50%|█████     | 3256/6470 [40:27<42:56,  1.25it/s]
4|3257|Loss: 0.6138693690299988:  50%|█████     | 3256/6470 [40:27<42:56,  1.25it/s]
4|3257|Loss: 0.6138693690299988:  50%|█████     | 3257/6470 [40:27<45:22,  1.18it/s]
4|3258|Loss: 0.8039469122886658:  50%|█████     | 3257/6470 [40:28<45:22,  1.18it/s]
4|3258|Loss: 0.8039469122886658:  50%|█████     | 3258/6470 [40:28<44:10,  1.21it/s]
4|3259|Loss: 0.8091673851013184:  50%|█████     | 3258/6470 [40:29<44:10,  1.21it/s]
4|3259|Loss: 0.8091673851013184:  50%|█████     | 3259/6470 [40:29<43:01,  1.24it/s]
4|3260|Loss: 0.8397189378738403:  50%|█████     | 3259/6470 [40:29<43:01,  1.24it/s]
4|3260|Loss: 0.8397189378738403:  50%|█████     | 3260/6470 [40:30<39:35,  1.35it/s]
4|3261|Loss: 0.7885605692863464:  50%|█████     | 3260/6470 [40:30<39:35,  1.35it/s]
4|3261|Loss: 0.7885605692863464:  50%|█████     | 3261/6470 [40:30<36:57,  1.45it/s]
4|3262|Loss: 0.868485689163208:  50%|█████     | 3261/6470 [40:31<36:57,  1.45it/s] 
4|3262|Loss: 0.868485689163208:  50%|█████     | 3262/6470 [40:31<36:34,  1.46it/s]
4|3263|Loss: 0.8135731220245361:  50%|█████     | 3262/6470 [40:31<36:34,  1.46it/s]
4|3263|Loss: 0.8135731220245361:  50%|█████     | 3263/6470 [40:32<37:43,  1.42it/s]
4|3264|Loss: 0.8077149987220764:  50%|█████     | 3263/6470 [40:32<37:43,  1.42it/s]
4|3264|Loss: 0.8077149987220764:  50%|█████     | 3264/6470 [40:32<38:58,  1.37it/s]
4|3265|Loss: 0.8125438094139099:  50%|█████     | 3264/6470 [40:33<38:58,  1.37it/s]
4|3265|Loss: 0.8125438094139099:  50%|█████     | 3265/6470 [40:33<40:50,  1.31it/s]
4|3266|Loss: 0.8501870632171631:  50%|█████     | 3265/6470 [40:34<40:50,  1.31it/s]
4|3266|Loss: 0.8501870632171631:  50%|█████     | 3266/6470 [40:34<41:04,  1.30it/s]
4|3267|Loss: 0.8809873461723328:  50%|█████     | 3266/6470 [40:34<41:04,  1.30it/s]
4|3267|Loss: 0.8809873461723328:  50%|█████     | 3267/6470 [40:35<42:06,  1.27it/s]
4|3268|Loss: 0.7801896333694458:  50%|█████     | 3267/6470 [40:35<42:06,  1.27it/s]
4|3268|Loss: 0.7801896333694458:  51%|█████     | 3268/6470 [40:35<39:43,  1.34it/s]
4|3269|Loss: 0.8253291249275208:  51%|█████     | 3268/6470 [40:36<39:43,  1.34it/s]
4|3269|Loss: 0.8253291249275208:  51%|█████     | 3269/6470 [40:36<39:15,  1.36it/s]
4|3270|Loss: 0.8163527250289917:  51%|█████     | 3269/6470 [40:37<39:15,  1.36it/s]
4|3270|Loss: 0.8163527250289917:  51%|█████     | 3270/6470 [40:37<37:55,  1.41it/s]
4|3271|Loss: 0.8049175143241882:  51%|█████     | 3270/6470 [40:37<37:55,  1.41it/s]
4|3271|Loss: 0.8049175143241882:  51%|█████     | 3271/6470 [40:38<38:20,  1.39it/s]
4|3272|Loss: 0.7888426780700684:  51%|█████     | 3271/6470 [40:38<38:20,  1.39it/s]
4|3272|Loss: 0.7888426780700684:  51%|█████     | 3272/6470 [40:38<39:19,  1.36it/s]
4|3273|Loss: 0.8409355878829956:  51%|█████     | 3272/6470 [40:39<39:19,  1.36it/s]
4|3273|Loss: 0.8409355878829956:  51%|█████     | 3273/6470 [40:39<39:04,  1.36it/s]
4|3274|Loss: 0.9617341160774231:  51%|█████     | 3273/6470 [40:40<39:04,  1.36it/s]
4|3274|Loss: 0.9617341160774231:  51%|█████     | 3274/6470 [40:40<39:11,  1.36it/s]
4|3275|Loss: 0.8508656620979309:  51%|█████     | 3274/6470 [40:40<39:11,  1.36it/s]
4|3275|Loss: 0.8508656620979309:  51%|█████     | 3275/6470 [40:41<41:20,  1.29it/s]
4|3276|Loss: 0.8696551322937012:  51%|█████     | 3275/6470 [40:41<41:20,  1.29it/s]
4|3276|Loss: 0.8696551322937012:  51%|█████     | 3276/6470 [40:42<41:50,  1.27it/s]
4|3277|Loss: 0.8029205203056335:  51%|█████     | 3276/6470 [40:42<41:50,  1.27it/s]
4|3277|Loss: 0.8029205203056335:  51%|█████     | 3277/6470 [40:42<43:43,  1.22it/s]
4|3278|Loss: 0.7503288388252258:  51%|█████     | 3277/6470 [40:43<43:43,  1.22it/s]
4|3278|Loss: 0.7503288388252258:  51%|█████     | 3278/6470 [40:43<42:34,  1.25it/s]
4|3279|Loss: 0.8122884035110474:  51%|█████     | 3278/6470 [40:44<42:34,  1.25it/s]
4|3279|Loss: 0.8122884035110474:  51%|█████     | 3279/6470 [40:44<40:32,  1.31it/s]
4|3280|Loss: 0.8118143081665039:  51%|█████     | 3279/6470 [40:44<40:32,  1.31it/s]
4|3280|Loss: 0.8118143081665039:  51%|█████     | 3280/6470 [40:45<38:53,  1.37it/s]
4|3281|Loss: 0.7469901442527771:  51%|█████     | 3280/6470 [40:45<38:53,  1.37it/s]
4|3281|Loss: 0.7469901442527771:  51%|█████     | 3281/6470 [40:45<38:14,  1.39it/s]
4|3282|Loss: 0.8371869325637817:  51%|█████     | 3281/6470 [40:46<38:14,  1.39it/s]
4|3282|Loss: 0.8371869325637817:  51%|█████     | 3282/6470 [40:46<40:04,  1.33it/s]
4|3283|Loss: 0.9014908075332642:  51%|█████     | 3282/6470 [40:46<40:04,  1.33it/s]
4|3283|Loss: 0.9014908075332642:  51%|█████     | 3283/6470 [40:47<39:00,  1.36it/s]
4|3284|Loss: 0.8349798321723938:  51%|█████     | 3283/6470 [40:47<39:00,  1.36it/s]
4|3284|Loss: 0.8349798321723938:  51%|█████     | 3284/6470 [40:47<37:42,  1.41it/s]
4|3285|Loss: 0.6826892495155334:  51%|█████     | 3284/6470 [40:48<37:42,  1.41it/s]
4|3285|Loss: 0.6826892495155334:  51%|█████     | 3285/6470 [40:48<40:13,  1.32it/s]
4|3286|Loss: 0.7592428922653198:  51%|█████     | 3285/6470 [40:49<40:13,  1.32it/s]
4|3286|Loss: 0.7592428922653198:  51%|█████     | 3286/6470 [40:49<38:00,  1.40it/s]
4|3287|Loss: 0.6372936367988586:  51%|█████     | 3286/6470 [40:49<38:00,  1.40it/s]
4|3287|Loss: 0.6372936367988586:  51%|█████     | 3287/6470 [40:50<37:54,  1.40it/s]
4|3288|Loss: 0.6749626994132996:  51%|█████     | 3287/6470 [40:50<37:54,  1.40it/s]
4|3288|Loss: 0.6749626994132996:  51%|█████     | 3288/6470 [40:50<37:23,  1.42it/s]
4|3289|Loss: 0.6910447478294373:  51%|█████     | 3288/6470 [40:51<37:23,  1.42it/s]
4|3289|Loss: 0.6910447478294373:  51%|█████     | 3289/6470 [40:51<36:41,  1.45it/s]
4|3290|Loss: 0.9533280730247498:  51%|█████     | 3289/6470 [40:51<36:41,  1.45it/s]
4|3290|Loss: 0.9533280730247498:  51%|█████     | 3290/6470 [40:52<37:46,  1.40it/s]
4|3291|Loss: 0.8950618505477905:  51%|█████     | 3290/6470 [40:52<37:46,  1.40it/s]
4|3291|Loss: 0.8950618505477905:  51%|█████     | 3291/6470 [40:52<35:53,  1.48it/s]
4|3292|Loss: 0.9051346182823181:  51%|█████     | 3291/6470 [40:53<35:53,  1.48it/s]
4|3292|Loss: 0.9051346182823181:  51%|█████     | 3292/6470 [40:53<39:27,  1.34it/s]
4|3293|Loss: 0.8235246539115906:  51%|█████     | 3292/6470 [40:54<39:27,  1.34it/s]
4|3293|Loss: 0.8235246539115906:  51%|█████     | 3293/6470 [40:54<42:47,  1.24it/s]
4|3294|Loss: 0.7421454787254333:  51%|█████     | 3293/6470 [40:55<42:47,  1.24it/s]
4|3294|Loss: 0.7421454787254333:  51%|█████     | 3294/6470 [40:55<41:36,  1.27it/s]
4|3295|Loss: 0.7206256985664368:  51%|█████     | 3294/6470 [40:55<41:36,  1.27it/s]
4|3295|Loss: 0.7206256985664368:  51%|█████     | 3295/6470 [40:55<36:19,  1.46it/s]
4|3296|Loss: 0.833663284778595:  51%|█████     | 3295/6470 [40:56<36:19,  1.46it/s] 
4|3296|Loss: 0.833663284778595:  51%|█████     | 3296/6470 [40:56<39:41,  1.33it/s]
4|3297|Loss: 0.785265326499939:  51%|█████     | 3296/6470 [40:57<39:41,  1.33it/s]
4|3297|Loss: 0.785265326499939:  51%|█████     | 3297/6470 [40:57<41:07,  1.29it/s]
4|3298|Loss: 0.6550049781799316:  51%|█████     | 3297/6470 [40:58<41:07,  1.29it/s]
4|3298|Loss: 0.6550049781799316:  51%|█████     | 3298/6470 [40:58<43:10,  1.22it/s]
4|3299|Loss: 0.9064932465553284:  51%|█████     | 3298/6470 [40:58<43:10,  1.22it/s]
4|3299|Loss: 0.9064932465553284:  51%|█████     | 3299/6470 [40:59<44:47,  1.18it/s]
4|3300|Loss: 0.8221122026443481:  51%|█████     | 3299/6470 [40:59<44:47,  1.18it/s]
4|3300|Loss: 0.8221122026443481:  51%|█████     | 3300/6470 [41:00<42:07,  1.25it/s]
4|3301|Loss: 0.8123329281806946:  51%|█████     | 3300/6470 [41:00<42:07,  1.25it/s]
4|3301|Loss: 0.8123329281806946:  51%|█████     | 3301/6470 [41:00<41:31,  1.27it/s]
4|3302|Loss: 0.9426867365837097:  51%|█████     | 3301/6470 [41:01<41:31,  1.27it/s]
4|3302|Loss: 0.9426867365837097:  51%|█████     | 3302/6470 [41:01<40:31,  1.30it/s]
4|3303|Loss: 0.7650545835494995:  51%|█████     | 3302/6470 [41:01<40:31,  1.30it/s]
4|3303|Loss: 0.7650545835494995:  51%|█████     | 3303/6470 [41:02<40:45,  1.30it/s]
4|3304|Loss: 0.7289113402366638:  51%|█████     | 3303/6470 [41:02<40:45,  1.30it/s]
4|3304|Loss: 0.7289113402366638:  51%|█████     | 3304/6470 [41:03<40:43,  1.30it/s]
4|3305|Loss: 0.9551491141319275:  51%|█████     | 3304/6470 [41:03<40:43,  1.30it/s]
4|3305|Loss: 0.9551491141319275:  51%|█████     | 3305/6470 [41:03<37:01,  1.42it/s]
4|3306|Loss: 0.7636744976043701:  51%|█████     | 3305/6470 [41:04<37:01,  1.42it/s]
4|3306|Loss: 0.7636744976043701:  51%|█████     | 3306/6470 [41:04<37:27,  1.41it/s]
4|3307|Loss: 0.7749438881874084:  51%|█████     | 3306/6470 [41:04<37:27,  1.41it/s]
4|3307|Loss: 0.7749438881874084:  51%|█████     | 3307/6470 [41:05<38:47,  1.36it/s]
4|3308|Loss: 0.8956665396690369:  51%|█████     | 3307/6470 [41:05<38:47,  1.36it/s]
4|3308|Loss: 0.8956665396690369:  51%|█████     | 3308/6470 [41:05<39:18,  1.34it/s]
4|3309|Loss: 0.8983729481697083:  51%|█████     | 3308/6470 [41:06<39:18,  1.34it/s]
4|3309|Loss: 0.8983729481697083:  51%|█████     | 3309/6470 [41:06<40:49,  1.29it/s]
4|3310|Loss: 0.8640739321708679:  51%|█████     | 3309/6470 [41:07<40:49,  1.29it/s]
4|3310|Loss: 0.8640739321708679:  51%|█████     | 3310/6470 [41:07<41:32,  1.27it/s]
4|3311|Loss: 0.9323095083236694:  51%|█████     | 3310/6470 [41:08<41:32,  1.27it/s]
4|3311|Loss: 0.9323095083236694:  51%|█████     | 3311/6470 [41:08<40:27,  1.30it/s]
4|3312|Loss: 0.6359903812408447:  51%|█████     | 3311/6470 [41:08<40:27,  1.30it/s]
4|3312|Loss: 0.6359903812408447:  51%|█████     | 3312/6470 [41:08<37:17,  1.41it/s]
4|3313|Loss: 0.7491816878318787:  51%|█████     | 3312/6470 [41:09<37:17,  1.41it/s]
4|3313|Loss: 0.7491816878318787:  51%|█████     | 3313/6470 [41:09<36:25,  1.44it/s]
4|3314|Loss: 0.9587273001670837:  51%|█████     | 3313/6470 [41:09<36:25,  1.44it/s]
4|3314|Loss: 0.9587273001670837:  51%|█████     | 3314/6470 [41:10<37:08,  1.42it/s]
4|3315|Loss: 0.852729320526123:  51%|█████     | 3314/6470 [41:10<37:08,  1.42it/s] 
4|3315|Loss: 0.852729320526123:  51%|█████     | 3315/6470 [41:11<40:11,  1.31it/s]
4|3316|Loss: 0.8456062078475952:  51%|█████     | 3315/6470 [41:11<40:11,  1.31it/s]
4|3316|Loss: 0.8456062078475952:  51%|█████▏    | 3316/6470 [41:11<39:12,  1.34it/s]
4|3317|Loss: 0.8803107738494873:  51%|█████▏    | 3316/6470 [41:12<39:12,  1.34it/s]
4|3317|Loss: 0.8803107738494873:  51%|█████▏    | 3317/6470 [41:12<39:21,  1.34it/s]
4|3318|Loss: 0.8540902137756348:  51%|█████▏    | 3317/6470 [41:13<39:21,  1.34it/s]
4|3318|Loss: 0.8540902137756348:  51%|█████▏    | 3318/6470 [41:13<40:16,  1.30it/s]
4|3319|Loss: 0.7792322039604187:  51%|█████▏    | 3318/6470 [41:13<40:16,  1.30it/s]
4|3319|Loss: 0.7792322039604187:  51%|█████▏    | 3319/6470 [41:14<41:45,  1.26it/s]
4|3320|Loss: 0.7643697261810303:  51%|█████▏    | 3319/6470 [41:14<41:45,  1.26it/s]
4|3320|Loss: 0.7643697261810303:  51%|█████▏    | 3320/6470 [41:14<38:40,  1.36it/s]
4|3321|Loss: 0.8309500217437744:  51%|█████▏    | 3320/6470 [41:15<38:40,  1.36it/s]
4|3321|Loss: 0.8309500217437744:  51%|█████▏    | 3321/6470 [41:15<38:52,  1.35it/s]
4|3322|Loss: 0.7684810161590576:  51%|█████▏    | 3321/6470 [41:16<38:52,  1.35it/s]
4|3322|Loss: 0.7684810161590576:  51%|█████▏    | 3322/6470 [41:16<39:21,  1.33it/s]
4|3323|Loss: 0.7468475699424744:  51%|█████▏    | 3322/6470 [41:16<39:21,  1.33it/s]
4|3323|Loss: 0.7468475699424744:  51%|█████▏    | 3323/6470 [41:17<38:23,  1.37it/s]
4|3324|Loss: 0.6525688171386719:  51%|█████▏    | 3323/6470 [41:17<38:23,  1.37it/s]
4|3324|Loss: 0.6525688171386719:  51%|█████▏    | 3324/6470 [41:17<35:43,  1.47it/s]
4|3325|Loss: 0.8279569149017334:  51%|█████▏    | 3324/6470 [41:18<35:43,  1.47it/s]
4|3325|Loss: 0.8279569149017334:  51%|█████▏    | 3325/6470 [41:18<35:06,  1.49it/s]
4|3326|Loss: 0.753745973110199:  51%|█████▏    | 3325/6470 [41:18<35:06,  1.49it/s] 
4|3326|Loss: 0.753745973110199:  51%|█████▏    | 3326/6470 [41:19<37:35,  1.39it/s]
4|3327|Loss: 0.7732579112052917:  51%|█████▏    | 3326/6470 [41:19<37:35,  1.39it/s]
4|3327|Loss: 0.7732579112052917:  51%|█████▏    | 3327/6470 [41:20<39:48,  1.32it/s]
4|3328|Loss: 0.7562766671180725:  51%|█████▏    | 3327/6470 [41:20<39:48,  1.32it/s]
4|3328|Loss: 0.7562766671180725:  51%|█████▏    | 3328/6470 [41:20<42:38,  1.23it/s]
4|3329|Loss: 0.8815791010856628:  51%|█████▏    | 3328/6470 [41:21<42:38,  1.23it/s]
4|3329|Loss: 0.8815791010856628:  51%|█████▏    | 3329/6470 [41:21<39:24,  1.33it/s]
4|3330|Loss: 0.730056643486023:  51%|█████▏    | 3329/6470 [41:21<39:24,  1.33it/s] 
4|3330|Loss: 0.730056643486023:  51%|█████▏    | 3330/6470 [41:22<40:34,  1.29it/s]
4|3331|Loss: 0.8930615782737732:  51%|█████▏    | 3330/6470 [41:22<40:34,  1.29it/s]
4|3331|Loss: 0.8930615782737732:  51%|█████▏    | 3331/6470 [41:23<40:43,  1.28it/s]
4|3332|Loss: 0.8354455828666687:  51%|█████▏    | 3331/6470 [41:23<40:43,  1.28it/s]
4|3332|Loss: 0.8354455828666687:  51%|█████▏    | 3332/6470 [41:23<41:03,  1.27it/s]
4|3333|Loss: 0.8853644728660583:  51%|█████▏    | 3332/6470 [41:24<41:03,  1.27it/s]
4|3333|Loss: 0.8853644728660583:  52%|█████▏    | 3333/6470 [41:24<41:18,  1.27it/s]
4|3334|Loss: 0.7505329251289368:  52%|█████▏    | 3333/6470 [41:25<41:18,  1.27it/s]
4|3334|Loss: 0.7505329251289368:  52%|█████▏    | 3334/6470 [41:25<40:43,  1.28it/s]
4|3335|Loss: 0.7277048230171204:  52%|█████▏    | 3334/6470 [41:26<40:43,  1.28it/s]
4|3335|Loss: 0.7277048230171204:  52%|█████▏    | 3335/6470 [41:26<42:24,  1.23it/s]
4|3336|Loss: 0.8233067989349365:  52%|█████▏    | 3335/6470 [41:26<42:24,  1.23it/s]
4|3336|Loss: 0.8233067989349365:  52%|█████▏    | 3336/6470 [41:27<42:38,  1.23it/s]
4|3337|Loss: 0.8778821229934692:  52%|█████▏    | 3336/6470 [41:27<42:38,  1.23it/s]
4|3337|Loss: 0.8778821229934692:  52%|█████▏    | 3337/6470 [41:28<41:54,  1.25it/s]
4|3338|Loss: 0.7724689245223999:  52%|█████▏    | 3337/6470 [41:28<41:54,  1.25it/s]
4|3338|Loss: 0.7724689245223999:  52%|█████▏    | 3338/6470 [41:28<42:49,  1.22it/s]
4|3339|Loss: 0.8090581297874451:  52%|█████▏    | 3338/6470 [41:29<42:49,  1.22it/s]
4|3339|Loss: 0.8090581297874451:  52%|█████▏    | 3339/6470 [41:29<42:00,  1.24it/s]
4|3340|Loss: 0.8001118898391724:  52%|█████▏    | 3339/6470 [41:30<42:00,  1.24it/s]
4|3340|Loss: 0.8001118898391724:  52%|█████▏    | 3340/6470 [41:30<41:10,  1.27it/s]
4|3341|Loss: 0.8114455938339233:  52%|█████▏    | 3340/6470 [41:30<41:10,  1.27it/s]
4|3341|Loss: 0.8114455938339233:  52%|█████▏    | 3341/6470 [41:31<41:08,  1.27it/s]
4|3342|Loss: 0.7041366696357727:  52%|█████▏    | 3341/6470 [41:31<41:08,  1.27it/s]
4|3342|Loss: 0.7041366696357727:  52%|█████▏    | 3342/6470 [41:32<43:31,  1.20it/s]
4|3343|Loss: 0.9526440501213074:  52%|█████▏    | 3342/6470 [41:32<43:31,  1.20it/s]
4|3343|Loss: 0.9526440501213074:  52%|█████▏    | 3343/6470 [41:32<40:25,  1.29it/s]
4|3344|Loss: 0.9251824021339417:  52%|█████▏    | 3343/6470 [41:33<40:25,  1.29it/s]
4|3344|Loss: 0.9251824021339417:  52%|█████▏    | 3344/6470 [41:33<41:19,  1.26it/s]
4|3345|Loss: 0.8440364003181458:  52%|█████▏    | 3344/6470 [41:34<41:19,  1.26it/s]
4|3345|Loss: 0.8440364003181458:  52%|█████▏    | 3345/6470 [41:34<41:47,  1.25it/s]
4|3346|Loss: 0.8286576867103577:  52%|█████▏    | 3345/6470 [41:34<41:47,  1.25it/s]
4|3346|Loss: 0.8286576867103577:  52%|█████▏    | 3346/6470 [41:35<40:12,  1.30it/s]
4|3347|Loss: 0.8662599921226501:  52%|█████▏    | 3346/6470 [41:35<40:12,  1.30it/s]
4|3347|Loss: 0.8662599921226501:  52%|█████▏    | 3347/6470 [41:35<39:30,  1.32it/s]
4|3348|Loss: 0.7517071962356567:  52%|█████▏    | 3347/6470 [41:36<39:30,  1.32it/s]
4|3348|Loss: 0.7517071962356567:  52%|█████▏    | 3348/6470 [41:36<37:55,  1.37it/s]
4|3349|Loss: 0.7625854015350342:  52%|█████▏    | 3348/6470 [41:36<37:55,  1.37it/s]
4|3349|Loss: 0.7625854015350342:  52%|█████▏    | 3349/6470 [41:37<37:54,  1.37it/s]
4|3350|Loss: 0.7766035199165344:  52%|█████▏    | 3349/6470 [41:37<37:54,  1.37it/s]
4|3350|Loss: 0.7766035199165344:  52%|█████▏    | 3350/6470 [41:38<38:17,  1.36it/s]
4|3351|Loss: 0.6936293244361877:  52%|█████▏    | 3350/6470 [41:38<38:17,  1.36it/s]
4|3351|Loss: 0.6936293244361877:  52%|█████▏    | 3351/6470 [41:38<35:32,  1.46it/s]
4|3352|Loss: 0.7819410562515259:  52%|█████▏    | 3351/6470 [41:38<35:32,  1.46it/s]
4|3352|Loss: 0.7819410562515259:  52%|█████▏    | 3352/6470 [41:39<34:54,  1.49it/s]
4|3353|Loss: 0.7560133337974548:  52%|█████▏    | 3352/6470 [41:39<34:54,  1.49it/s]
4|3353|Loss: 0.7560133337974548:  52%|█████▏    | 3353/6470 [41:39<31:40,  1.64it/s]
4|3354|Loss: 0.8983263969421387:  52%|█████▏    | 3353/6470 [41:40<31:40,  1.64it/s]
4|3354|Loss: 0.8983263969421387:  52%|█████▏    | 3354/6470 [41:40<33:30,  1.55it/s]
4|3355|Loss: 0.8486814498901367:  52%|█████▏    | 3354/6470 [41:40<33:30,  1.55it/s]
4|3355|Loss: 0.8486814498901367:  52%|█████▏    | 3355/6470 [41:41<38:01,  1.37it/s]
4|3356|Loss: 0.8393373489379883:  52%|█████▏    | 3355/6470 [41:41<38:01,  1.37it/s]
4|3356|Loss: 0.8393373489379883:  52%|█████▏    | 3356/6470 [41:42<38:30,  1.35it/s]
4|3357|Loss: 0.8686328530311584:  52%|█████▏    | 3356/6470 [41:42<38:30,  1.35it/s]
4|3357|Loss: 0.8686328530311584:  52%|█████▏    | 3357/6470 [41:42<38:41,  1.34it/s]
4|3358|Loss: 0.823742151260376:  52%|█████▏    | 3357/6470 [41:43<38:41,  1.34it/s] 
4|3358|Loss: 0.823742151260376:  52%|█████▏    | 3358/6470 [41:43<38:11,  1.36it/s]
4|3359|Loss: 0.855223536491394:  52%|█████▏    | 3358/6470 [41:43<38:11,  1.36it/s]
4|3359|Loss: 0.855223536491394:  52%|█████▏    | 3359/6470 [41:44<35:38,  1.45it/s]
4|3360|Loss: 0.7232655882835388:  52%|█████▏    | 3359/6470 [41:44<35:38,  1.45it/s]
4|3360|Loss: 0.7232655882835388:  52%|█████▏    | 3360/6470 [41:45<38:27,  1.35it/s]
4|3361|Loss: 1.0007517337799072:  52%|█████▏    | 3360/6470 [41:45<38:27,  1.35it/s]
4|3361|Loss: 1.0007517337799072:  52%|█████▏    | 3361/6470 [41:46<42:26,  1.22it/s]
4|3362|Loss: 0.7582244277000427:  52%|█████▏    | 3361/6470 [41:46<42:26,  1.22it/s]
4|3362|Loss: 0.7582244277000427:  52%|█████▏    | 3362/6470 [41:46<41:27,  1.25it/s]
4|3363|Loss: 0.7995207905769348:  52%|█████▏    | 3362/6470 [41:47<41:27,  1.25it/s]
4|3363|Loss: 0.7995207905769348:  52%|█████▏    | 3363/6470 [41:47<41:06,  1.26it/s]
4|3364|Loss: 0.9184430241584778:  52%|█████▏    | 3363/6470 [41:47<41:06,  1.26it/s]
4|3364|Loss: 0.9184430241584778:  52%|█████▏    | 3364/6470 [41:48<40:50,  1.27it/s]
4|3365|Loss: 0.8405724763870239:  52%|█████▏    | 3364/6470 [41:48<40:50,  1.27it/s]
4|3365|Loss: 0.8405724763870239:  52%|█████▏    | 3365/6470 [41:49<42:24,  1.22it/s]
4|3366|Loss: 0.9717814922332764:  52%|█████▏    | 3365/6470 [41:49<42:24,  1.22it/s]
4|3366|Loss: 0.9717814922332764:  52%|█████▏    | 3366/6470 [41:49<41:14,  1.25it/s]
4|3367|Loss: 0.7677833437919617:  52%|█████▏    | 3366/6470 [41:50<41:14,  1.25it/s]
4|3367|Loss: 0.7677833437919617:  52%|█████▏    | 3367/6470 [41:50<40:29,  1.28it/s]
4|3368|Loss: 0.7617691159248352:  52%|█████▏    | 3367/6470 [41:51<40:29,  1.28it/s]
4|3368|Loss: 0.7617691159248352:  52%|█████▏    | 3368/6470 [41:51<38:39,  1.34it/s]
4|3369|Loss: 0.8390536904335022:  52%|█████▏    | 3368/6470 [41:51<38:39,  1.34it/s]
4|3369|Loss: 0.8390536904335022:  52%|█████▏    | 3369/6470 [41:52<37:09,  1.39it/s]
4|3370|Loss: 0.8396891951560974:  52%|█████▏    | 3369/6470 [41:52<37:09,  1.39it/s]
4|3370|Loss: 0.8396891951560974:  52%|█████▏    | 3370/6470 [41:52<34:25,  1.50it/s]
4|3371|Loss: 0.8895297050476074:  52%|█████▏    | 3370/6470 [41:52<34:25,  1.50it/s]
4|3371|Loss: 0.8895297050476074:  52%|█████▏    | 3371/6470 [41:53<35:19,  1.46it/s]
4|3372|Loss: 0.8652833104133606:  52%|█████▏    | 3371/6470 [41:53<35:19,  1.46it/s]
4|3372|Loss: 0.8652833104133606:  52%|█████▏    | 3372/6470 [41:54<38:55,  1.33it/s]
4|3373|Loss: 0.8426909446716309:  52%|█████▏    | 3372/6470 [41:54<38:55,  1.33it/s]
4|3373|Loss: 0.8426909446716309:  52%|█████▏    | 3373/6470 [41:54<38:43,  1.33it/s]
4|3374|Loss: 0.7175228595733643:  52%|█████▏    | 3373/6470 [41:55<38:43,  1.33it/s]
4|3374|Loss: 0.7175228595733643:  52%|█████▏    | 3374/6470 [41:55<36:43,  1.41it/s]
4|3375|Loss: 0.7537019848823547:  52%|█████▏    | 3374/6470 [41:55<36:43,  1.41it/s]
4|3375|Loss: 0.7537019848823547:  52%|█████▏    | 3375/6470 [41:56<32:20,  1.59it/s]
4|3376|Loss: 0.790779709815979:  52%|█████▏    | 3375/6470 [41:56<32:20,  1.59it/s] 
4|3376|Loss: 0.790779709815979:  52%|█████▏    | 3376/6470 [41:56<28:31,  1.81it/s]
4|3377|Loss: 0.8579244017601013:  52%|█████▏    | 3376/6470 [41:56<28:31,  1.81it/s]
4|3377|Loss: 0.8579244017601013:  52%|█████▏    | 3377/6470 [41:57<32:36,  1.58it/s]
4|3378|Loss: 0.737250030040741:  52%|█████▏    | 3377/6470 [41:57<32:36,  1.58it/s] 
4|3378|Loss: 0.737250030040741:  52%|█████▏    | 3378/6470 [41:58<35:38,  1.45it/s]
4|3379|Loss: 0.8230806589126587:  52%|█████▏    | 3378/6470 [41:58<35:38,  1.45it/s]
4|3379|Loss: 0.8230806589126587:  52%|█████▏    | 3379/6470 [41:58<37:21,  1.38it/s]
4|3380|Loss: 0.7768240571022034:  52%|█████▏    | 3379/6470 [41:59<37:21,  1.38it/s]
4|3380|Loss: 0.7768240571022034:  52%|█████▏    | 3380/6470 [41:59<41:01,  1.26it/s]
4|3381|Loss: 0.8373361825942993:  52%|█████▏    | 3380/6470 [42:00<41:01,  1.26it/s]
4|3381|Loss: 0.8373361825942993:  52%|█████▏    | 3381/6470 [42:00<42:34,  1.21it/s]
4|3382|Loss: 0.836355447769165:  52%|█████▏    | 3381/6470 [42:01<42:34,  1.21it/s] 
4|3382|Loss: 0.836355447769165:  52%|█████▏    | 3382/6470 [42:01<39:55,  1.29it/s]
4|3383|Loss: 0.7259470820426941:  52%|█████▏    | 3382/6470 [42:01<39:55,  1.29it/s]
4|3383|Loss: 0.7259470820426941:  52%|█████▏    | 3383/6470 [42:01<36:43,  1.40it/s]
4|3384|Loss: 0.7650455236434937:  52%|█████▏    | 3383/6470 [42:02<36:43,  1.40it/s]
4|3384|Loss: 0.7650455236434937:  52%|█████▏    | 3384/6470 [42:02<33:18,  1.54it/s]
4|3385|Loss: 0.9291366338729858:  52%|█████▏    | 3384/6470 [42:02<33:18,  1.54it/s]
4|3385|Loss: 0.9291366338729858:  52%|█████▏    | 3385/6470 [42:03<33:14,  1.55it/s]
4|3386|Loss: 0.770214319229126:  52%|█████▏    | 3385/6470 [42:03<33:14,  1.55it/s] 
4|3386|Loss: 0.770214319229126:  52%|█████▏    | 3386/6470 [42:03<33:16,  1.54it/s]
4|3387|Loss: 0.8608588576316833:  52%|█████▏    | 3386/6470 [42:04<33:16,  1.54it/s]
4|3387|Loss: 0.8608588576316833:  52%|█████▏    | 3387/6470 [42:04<33:31,  1.53it/s]
4|3388|Loss: 1.02424955368042:  52%|█████▏    | 3387/6470 [42:04<33:31,  1.53it/s]  
4|3388|Loss: 1.02424955368042:  52%|█████▏    | 3388/6470 [42:05<35:36,  1.44it/s]
4|3389|Loss: 0.8154096603393555:  52%|█████▏    | 3388/6470 [42:05<35:36,  1.44it/s]
4|3389|Loss: 0.8154096603393555:  52%|█████▏    | 3389/6470 [42:05<36:03,  1.42it/s]
4|3390|Loss: 0.8310431241989136:  52%|█████▏    | 3389/6470 [42:06<36:03,  1.42it/s]
4|3390|Loss: 0.8310431241989136:  52%|█████▏    | 3390/6470 [42:06<35:35,  1.44it/s]
4|3391|Loss: 0.8516402244567871:  52%|█████▏    | 3390/6470 [42:06<35:35,  1.44it/s]
4|3391|Loss: 0.8516402244567871:  52%|█████▏    | 3391/6470 [42:07<33:54,  1.51it/s]
4|3392|Loss: 0.8830411434173584:  52%|█████▏    | 3391/6470 [42:07<33:54,  1.51it/s]
4|3392|Loss: 0.8830411434173584:  52%|█████▏    | 3392/6470 [42:07<34:06,  1.50it/s]
4|3393|Loss: 0.7620053887367249:  52%|█████▏    | 3392/6470 [42:08<34:06,  1.50it/s]
4|3393|Loss: 0.7620053887367249:  52%|█████▏    | 3393/6470 [42:08<36:01,  1.42it/s]
4|3394|Loss: 0.7878955602645874:  52%|█████▏    | 3393/6470 [42:09<36:01,  1.42it/s]
4|3394|Loss: 0.7878955602645874:  52%|█████▏    | 3394/6470 [42:09<37:46,  1.36it/s]
4|3395|Loss: 0.7509148716926575:  52%|█████▏    | 3394/6470 [42:09<37:46,  1.36it/s]
4|3395|Loss: 0.7509148716926575:  52%|█████▏    | 3395/6470 [42:10<39:00,  1.31it/s]
4|3396|Loss: 0.8103011250495911:  52%|█████▏    | 3395/6470 [42:10<39:00,  1.31it/s]
4|3396|Loss: 0.8103011250495911:  52%|█████▏    | 3396/6470 [42:11<42:11,  1.21it/s]
4|3397|Loss: 0.6767023205757141:  52%|█████▏    | 3396/6470 [42:11<42:11,  1.21it/s]
4|3397|Loss: 0.6767023205757141:  53%|█████▎    | 3397/6470 [42:11<38:38,  1.33it/s]
4|3398|Loss: 0.7792235612869263:  53%|█████▎    | 3397/6470 [42:12<38:38,  1.33it/s]
4|3398|Loss: 0.7792235612869263:  53%|█████▎    | 3398/6470 [42:12<37:34,  1.36it/s]
4|3399|Loss: 0.7398368120193481:  53%|█████▎    | 3398/6470 [42:12<37:34,  1.36it/s]
4|3399|Loss: 0.7398368120193481:  53%|█████▎    | 3399/6470 [42:13<39:12,  1.31it/s]
4|3400|Loss: 0.8943412899971008:  53%|█████▎    | 3399/6470 [42:13<39:12,  1.31it/s]
4|3400|Loss: 0.8943412899971008:  53%|█████▎    | 3400/6470 [42:14<38:32,  1.33it/s]
4|3401|Loss: 0.8091281652450562:  53%|█████▎    | 3400/6470 [42:14<38:32,  1.33it/s]
4|3401|Loss: 0.8091281652450562:  53%|█████▎    | 3401/6470 [42:14<38:24,  1.33it/s]
4|3402|Loss: 0.6878812909126282:  53%|█████▎    | 3401/6470 [42:15<38:24,  1.33it/s]
4|3402|Loss: 0.6878812909126282:  53%|█████▎    | 3402/6470 [42:15<41:01,  1.25it/s]
4|3403|Loss: 0.8683595061302185:  53%|█████▎    | 3402/6470 [42:16<41:01,  1.25it/s]
4|3403|Loss: 0.8683595061302185:  53%|█████▎    | 3403/6470 [42:16<40:14,  1.27it/s]
4|3404|Loss: 0.8102218508720398:  53%|█████▎    | 3403/6470 [42:16<40:14,  1.27it/s]
4|3404|Loss: 0.8102218508720398:  53%|█████▎    | 3404/6470 [42:17<41:59,  1.22it/s]
4|3405|Loss: 0.8536432385444641:  53%|█████▎    | 3404/6470 [42:17<41:59,  1.22it/s]
4|3405|Loss: 0.8536432385444641:  53%|█████▎    | 3405/6470 [42:18<39:39,  1.29it/s]
4|3406|Loss: 0.9644433259963989:  53%|█████▎    | 3405/6470 [42:18<39:39,  1.29it/s]
4|3406|Loss: 0.9644433259963989:  53%|█████▎    | 3406/6470 [42:18<37:54,  1.35it/s]
4|3407|Loss: 0.7718086838722229:  53%|█████▎    | 3406/6470 [42:19<37:54,  1.35it/s]
4|3407|Loss: 0.7718086838722229:  53%|█████▎    | 3407/6470 [42:19<37:59,  1.34it/s]
4|3408|Loss: 0.7298219203948975:  53%|█████▎    | 3407/6470 [42:19<37:59,  1.34it/s]
4|3408|Loss: 0.7298219203948975:  53%|█████▎    | 3408/6470 [42:20<36:47,  1.39it/s]
4|3409|Loss: 0.8429104089736938:  53%|█████▎    | 3408/6470 [42:20<36:47,  1.39it/s]
4|3409|Loss: 0.8429104089736938:  53%|█████▎    | 3409/6470 [42:20<34:12,  1.49it/s]
4|3410|Loss: 0.9379739165306091:  53%|█████▎    | 3409/6470 [42:21<34:12,  1.49it/s]
4|3410|Loss: 0.9379739165306091:  53%|█████▎    | 3410/6470 [42:21<33:47,  1.51it/s]
4|3411|Loss: 0.7559140920639038:  53%|█████▎    | 3410/6470 [42:21<33:47,  1.51it/s]
4|3411|Loss: 0.7559140920639038:  53%|█████▎    | 3411/6470 [42:22<37:09,  1.37it/s]
4|3412|Loss: 0.9259077310562134:  53%|█████▎    | 3411/6470 [42:22<37:09,  1.37it/s]
4|3412|Loss: 0.9259077310562134:  53%|█████▎    | 3412/6470 [42:22<33:41,  1.51it/s]
4|3413|Loss: 0.83062744140625:  53%|█████▎    | 3412/6470 [42:23<33:41,  1.51it/s]  
4|3413|Loss: 0.83062744140625:  53%|█████▎    | 3413/6470 [42:23<31:05,  1.64it/s]
4|3414|Loss: 0.7645508646965027:  53%|█████▎    | 3413/6470 [42:23<31:05,  1.64it/s]
4|3414|Loss: 0.7645508646965027:  53%|█████▎    | 3414/6470 [42:24<34:24,  1.48it/s]
4|3415|Loss: 0.734554648399353:  53%|█████▎    | 3414/6470 [42:24<34:24,  1.48it/s] 
4|3415|Loss: 0.734554648399353:  53%|█████▎    | 3415/6470 [42:24<35:21,  1.44it/s]
4|3416|Loss: 0.9057414531707764:  53%|█████▎    | 3415/6470 [42:25<35:21,  1.44it/s]
4|3416|Loss: 0.9057414531707764:  53%|█████▎    | 3416/6470 [42:25<36:35,  1.39it/s]
4|3417|Loss: 0.8888720273971558:  53%|█████▎    | 3416/6470 [42:25<36:35,  1.39it/s]
4|3417|Loss: 0.8888720273971558:  53%|█████▎    | 3417/6470 [42:26<37:52,  1.34it/s]
4|3418|Loss: 0.7578927874565125:  53%|█████▎    | 3417/6470 [42:26<37:52,  1.34it/s]
4|3418|Loss: 0.7578927874565125:  53%|█████▎    | 3418/6470 [42:27<40:29,  1.26it/s]
4|3419|Loss: 0.8685542941093445:  53%|█████▎    | 3418/6470 [42:27<40:29,  1.26it/s]
4|3419|Loss: 0.8685542941093445:  53%|█████▎    | 3419/6470 [42:28<41:04,  1.24it/s]
4|3420|Loss: 0.8995104432106018:  53%|█████▎    | 3419/6470 [42:28<41:04,  1.24it/s]
4|3420|Loss: 0.8995104432106018:  53%|█████▎    | 3420/6470 [42:28<37:08,  1.37it/s]
4|3421|Loss: 0.752170741558075:  53%|█████▎    | 3420/6470 [42:28<37:08,  1.37it/s] 
4|3421|Loss: 0.752170741558075:  53%|█████▎    | 3421/6470 [42:29<35:05,  1.45it/s]
4|3422|Loss: 0.7463611364364624:  53%|█████▎    | 3421/6470 [42:29<35:05,  1.45it/s]
4|3422|Loss: 0.7463611364364624:  53%|█████▎    | 3422/6470 [42:29<34:06,  1.49it/s]
4|3423|Loss: 0.7174434065818787:  53%|█████▎    | 3422/6470 [42:30<34:06,  1.49it/s]
4|3423|Loss: 0.7174434065818787:  53%|█████▎    | 3423/6470 [42:30<34:25,  1.48it/s]
4|3424|Loss: 0.7531614303588867:  53%|█████▎    | 3423/6470 [42:30<34:25,  1.48it/s]
4|3424|Loss: 0.7531614303588867:  53%|█████▎    | 3424/6470 [42:31<34:49,  1.46it/s]
4|3425|Loss: 0.5999622941017151:  53%|█████▎    | 3424/6470 [42:31<34:49,  1.46it/s]
4|3425|Loss: 0.5999622941017151:  53%|█████▎    | 3425/6470 [42:32<37:49,  1.34it/s]
4|3426|Loss: 0.7013598084449768:  53%|█████▎    | 3425/6470 [42:32<37:49,  1.34it/s]
4|3426|Loss: 0.7013598084449768:  53%|█████▎    | 3426/6470 [42:33<40:24,  1.26it/s]
4|3427|Loss: 0.9074162244796753:  53%|█████▎    | 3426/6470 [42:33<40:24,  1.26it/s]
4|3427|Loss: 0.9074162244796753:  53%|█████▎    | 3427/6470 [42:33<38:06,  1.33it/s]
4|3428|Loss: 0.8792481422424316:  53%|█████▎    | 3427/6470 [42:34<38:06,  1.33it/s]
4|3428|Loss: 0.8792481422424316:  53%|█████▎    | 3428/6470 [42:34<37:41,  1.35it/s]
4|3429|Loss: 0.7074483633041382:  53%|█████▎    | 3428/6470 [42:34<37:41,  1.35it/s]
4|3429|Loss: 0.7074483633041382:  53%|█████▎    | 3429/6470 [42:35<38:13,  1.33it/s]
4|3430|Loss: 0.834176778793335:  53%|█████▎    | 3429/6470 [42:35<38:13,  1.33it/s] 
4|3430|Loss: 0.834176778793335:  53%|█████▎    | 3430/6470 [42:36<38:49,  1.30it/s]
4|3431|Loss: 0.7835224270820618:  53%|█████▎    | 3430/6470 [42:36<38:49,  1.30it/s]
4|3431|Loss: 0.7835224270820618:  53%|█████▎    | 3431/6470 [42:36<40:14,  1.26it/s]
4|3432|Loss: 0.8048684597015381:  53%|█████▎    | 3431/6470 [42:37<40:14,  1.26it/s]
4|3432|Loss: 0.8048684597015381:  53%|█████▎    | 3432/6470 [42:37<38:25,  1.32it/s]
4|3433|Loss: 0.8116129040718079:  53%|█████▎    | 3432/6470 [42:37<38:25,  1.32it/s]
4|3433|Loss: 0.8116129040718079:  53%|█████▎    | 3433/6470 [42:38<39:09,  1.29it/s]
4|3434|Loss: 0.688515841960907:  53%|█████▎    | 3433/6470 [42:38<39:09,  1.29it/s] 
4|3434|Loss: 0.688515841960907:  53%|█████▎    | 3434/6470 [42:39<40:28,  1.25it/s]
4|3435|Loss: 0.8055569529533386:  53%|█████▎    | 3434/6470 [42:39<40:28,  1.25it/s]
4|3435|Loss: 0.8055569529533386:  53%|█████▎    | 3435/6470 [42:40<40:33,  1.25it/s]
4|3436|Loss: 0.6957275867462158:  53%|█████▎    | 3435/6470 [42:40<40:33,  1.25it/s]
4|3436|Loss: 0.6957275867462158:  53%|█████▎    | 3436/6470 [42:40<40:17,  1.25it/s]
4|3437|Loss: 0.6966093182563782:  53%|█████▎    | 3436/6470 [42:41<40:17,  1.25it/s]
4|3437|Loss: 0.6966093182563782:  53%|█████▎    | 3437/6470 [42:41<40:04,  1.26it/s]
4|3438|Loss: 0.8552128672599792:  53%|█████▎    | 3437/6470 [42:41<40:04,  1.26it/s]
4|3438|Loss: 0.8552128672599792:  53%|█████▎    | 3438/6470 [42:42<37:01,  1.36it/s]
4|3439|Loss: 0.7376763820648193:  53%|█████▎    | 3438/6470 [42:42<37:01,  1.36it/s]
4|3439|Loss: 0.7376763820648193:  53%|█████▎    | 3439/6470 [42:42<32:28,  1.56it/s]
4|3440|Loss: 0.8160023093223572:  53%|█████▎    | 3439/6470 [42:43<32:28,  1.56it/s]
4|3440|Loss: 0.8160023093223572:  53%|█████▎    | 3440/6470 [42:43<33:44,  1.50it/s]
4|3441|Loss: 0.8311301469802856:  53%|█████▎    | 3440/6470 [42:43<33:44,  1.50it/s]
4|3441|Loss: 0.8311301469802856:  53%|█████▎    | 3441/6470 [42:44<35:56,  1.40it/s]
4|3442|Loss: 0.9477881193161011:  53%|█████▎    | 3441/6470 [42:44<35:56,  1.40it/s]
4|3442|Loss: 0.9477881193161011:  53%|█████▎    | 3442/6470 [42:44<36:51,  1.37it/s]
4|3443|Loss: 0.8417266011238098:  53%|█████▎    | 3442/6470 [42:45<36:51,  1.37it/s]
4|3443|Loss: 0.8417266011238098:  53%|█████▎    | 3443/6470 [42:45<37:07,  1.36it/s]
4|3444|Loss: 0.8001118302345276:  53%|█████▎    | 3443/6470 [42:46<37:07,  1.36it/s]
4|3444|Loss: 0.8001118302345276:  53%|█████▎    | 3444/6470 [42:46<38:05,  1.32it/s]
4|3445|Loss: 1.0011024475097656:  53%|█████▎    | 3444/6470 [42:46<38:05,  1.32it/s]
4|3445|Loss: 1.0011024475097656:  53%|█████▎    | 3445/6470 [42:47<37:49,  1.33it/s]
4|3446|Loss: 0.8926774263381958:  53%|█████▎    | 3445/6470 [42:47<37:49,  1.33it/s]
4|3446|Loss: 0.8926774263381958:  53%|█████▎    | 3446/6470 [42:47<36:44,  1.37it/s]
4|3447|Loss: 0.9545901417732239:  53%|█████▎    | 3446/6470 [42:48<36:44,  1.37it/s]
4|3447|Loss: 0.9545901417732239:  53%|█████▎    | 3447/6470 [42:48<35:43,  1.41it/s]
4|3448|Loss: 0.7884563207626343:  53%|█████▎    | 3447/6470 [42:49<35:43,  1.41it/s]
4|3448|Loss: 0.7884563207626343:  53%|█████▎    | 3448/6470 [42:49<38:38,  1.30it/s]
4|3449|Loss: 0.8418014645576477:  53%|█████▎    | 3448/6470 [42:49<38:38,  1.30it/s]
4|3449|Loss: 0.8418014645576477:  53%|█████▎    | 3449/6470 [42:50<37:17,  1.35it/s]
4|3450|Loss: 0.8848291635513306:  53%|█████▎    | 3449/6470 [42:50<37:17,  1.35it/s]
4|3450|Loss: 0.8848291635513306:  53%|█████▎    | 3450/6470 [42:50<37:24,  1.35it/s]
4|3451|Loss: 0.7376195788383484:  53%|█████▎    | 3450/6470 [42:51<37:24,  1.35it/s]
4|3451|Loss: 0.7376195788383484:  53%|█████▎    | 3451/6470 [42:51<36:04,  1.39it/s]
4|3452|Loss: 0.9189997315406799:  53%|█████▎    | 3451/6470 [42:51<36:04,  1.39it/s]
4|3452|Loss: 0.9189997315406799:  53%|█████▎    | 3452/6470 [42:52<33:15,  1.51it/s]
4|3453|Loss: 0.816964328289032:  53%|█████▎    | 3452/6470 [42:52<33:15,  1.51it/s] 
4|3453|Loss: 0.816964328289032:  53%|█████▎    | 3453/6470 [42:52<33:50,  1.49it/s]
4|3454|Loss: 0.8989797234535217:  53%|█████▎    | 3453/6470 [42:53<33:50,  1.49it/s]
4|3454|Loss: 0.8989797234535217:  53%|█████▎    | 3454/6470 [42:53<33:38,  1.49it/s]
4|3455|Loss: 0.8428103923797607:  53%|█████▎    | 3454/6470 [42:53<33:38,  1.49it/s]
4|3455|Loss: 0.8428103923797607:  53%|█████▎    | 3455/6470 [42:54<32:49,  1.53it/s]
4|3456|Loss: 0.8653401136398315:  53%|█████▎    | 3455/6470 [42:54<32:49,  1.53it/s]
4|3456|Loss: 0.8653401136398315:  53%|█████▎    | 3456/6470 [42:54<33:22,  1.51it/s]
4|3457|Loss: 0.8271580934524536:  53%|█████▎    | 3456/6470 [42:55<33:22,  1.51it/s]
4|3457|Loss: 0.8271580934524536:  53%|█████▎    | 3457/6470 [42:55<33:31,  1.50it/s]
4|3458|Loss: 1.0242395401000977:  53%|█████▎    | 3457/6470 [42:55<33:31,  1.50it/s]
4|3458|Loss: 1.0242395401000977:  53%|█████▎    | 3458/6470 [42:56<34:54,  1.44it/s]
4|3459|Loss: 0.9729607701301575:  53%|█████▎    | 3458/6470 [42:56<34:54,  1.44it/s]
4|3459|Loss: 0.9729607701301575:  53%|█████▎    | 3459/6470 [42:56<34:17,  1.46it/s]
4|3460|Loss: 0.7232381701469421:  53%|█████▎    | 3459/6470 [42:57<34:17,  1.46it/s]
4|3460|Loss: 0.7232381701469421:  53%|█████▎    | 3460/6470 [42:57<33:19,  1.51it/s]
4|3461|Loss: 0.8424381017684937:  53%|█████▎    | 3460/6470 [42:57<33:19,  1.51it/s]
4|3461|Loss: 0.8424381017684937:  53%|█████▎    | 3461/6470 [42:58<34:31,  1.45it/s]
4|3462|Loss: 0.8377171158790588:  53%|█████▎    | 3461/6470 [42:58<34:31,  1.45it/s]
4|3462|Loss: 0.8377171158790588:  54%|█████▎    | 3462/6470 [42:58<34:53,  1.44it/s]
4|3463|Loss: 0.808985710144043:  54%|█████▎    | 3462/6470 [42:59<34:53,  1.44it/s] 
4|3463|Loss: 0.808985710144043:  54%|█████▎    | 3463/6470 [42:59<36:01,  1.39it/s]
4|3464|Loss: 0.6688594818115234:  54%|█████▎    | 3463/6470 [43:00<36:01,  1.39it/s]
4|3464|Loss: 0.6688594818115234:  54%|█████▎    | 3464/6470 [43:00<35:39,  1.40it/s]
4|3465|Loss: 0.8361346125602722:  54%|█████▎    | 3464/6470 [43:00<35:39,  1.40it/s]
4|3465|Loss: 0.8361346125602722:  54%|█████▎    | 3465/6470 [43:00<33:27,  1.50it/s]
4|3466|Loss: 0.7578637003898621:  54%|█████▎    | 3465/6470 [43:01<33:27,  1.50it/s]
4|3466|Loss: 0.7578637003898621:  54%|█████▎    | 3466/6470 [43:01<33:25,  1.50it/s]
4|3467|Loss: 0.8548870086669922:  54%|█████▎    | 3466/6470 [43:02<33:25,  1.50it/s]
4|3467|Loss: 0.8548870086669922:  54%|█████▎    | 3467/6470 [43:02<33:18,  1.50it/s]
4|3468|Loss: 0.6801679134368896:  54%|█████▎    | 3467/6470 [43:02<33:18,  1.50it/s]
4|3468|Loss: 0.6801679134368896:  54%|█████▎    | 3468/6470 [43:03<34:42,  1.44it/s]
4|3469|Loss: 0.7780816555023193:  54%|█████▎    | 3468/6470 [43:03<34:42,  1.44it/s]
4|3469|Loss: 0.7780816555023193:  54%|█████▎    | 3469/6470 [43:03<36:11,  1.38it/s]
4|3470|Loss: 0.9074513912200928:  54%|█████▎    | 3469/6470 [43:04<36:11,  1.38it/s]
4|3470|Loss: 0.9074513912200928:  54%|█████▎    | 3470/6470 [43:04<37:27,  1.33it/s]
4|3471|Loss: 0.7111818790435791:  54%|█████▎    | 3470/6470 [43:05<37:27,  1.33it/s]
4|3471|Loss: 0.7111818790435791:  54%|█████▎    | 3471/6470 [43:05<38:19,  1.30it/s]
4|3472|Loss: 0.7251687049865723:  54%|█████▎    | 3471/6470 [43:05<38:19,  1.30it/s]
4|3472|Loss: 0.7251687049865723:  54%|█████▎    | 3472/6470 [43:06<39:00,  1.28it/s]
4|3473|Loss: 1.012480616569519:  54%|█████▎    | 3472/6470 [43:06<39:00,  1.28it/s] 
4|3473|Loss: 1.012480616569519:  54%|█████▎    | 3473/6470 [43:07<39:17,  1.27it/s]
4|3474|Loss: 0.6976465582847595:  54%|█████▎    | 3473/6470 [43:07<39:17,  1.27it/s]
4|3474|Loss: 0.6976465582847595:  54%|█████▎    | 3474/6470 [43:07<37:52,  1.32it/s]
4|3475|Loss: 0.785589337348938:  54%|█████▎    | 3474/6470 [43:08<37:52,  1.32it/s] 
4|3475|Loss: 0.785589337348938:  54%|█████▎    | 3475/6470 [43:08<38:00,  1.31it/s]
4|3476|Loss: 0.975444495677948:  54%|█████▎    | 3475/6470 [43:08<38:00,  1.31it/s]
4|3476|Loss: 0.975444495677948:  54%|█████▎    | 3476/6470 [43:09<36:57,  1.35it/s]
4|3477|Loss: 0.8381333351135254:  54%|█████▎    | 3476/6470 [43:09<36:57,  1.35it/s]
4|3477|Loss: 0.8381333351135254:  54%|█████▎    | 3477/6470 [43:09<36:11,  1.38it/s]
4|3478|Loss: 0.7445635795593262:  54%|█████▎    | 3477/6470 [43:10<36:11,  1.38it/s]
4|3478|Loss: 0.7445635795593262:  54%|█████▍    | 3478/6470 [43:10<35:19,  1.41it/s]
4|3479|Loss: 0.7900067567825317:  54%|█████▍    | 3478/6470 [43:10<35:19,  1.41it/s]
4|3479|Loss: 0.7900067567825317:  54%|█████▍    | 3479/6470 [43:11<35:44,  1.40it/s]
4|3480|Loss: 0.8595154285430908:  54%|█████▍    | 3479/6470 [43:11<35:44,  1.40it/s]
4|3480|Loss: 0.8595154285430908:  54%|█████▍    | 3480/6470 [43:12<36:18,  1.37it/s]
4|3481|Loss: 0.8477351665496826:  54%|█████▍    | 3480/6470 [43:12<36:18,  1.37it/s]
4|3481|Loss: 0.8477351665496826:  54%|█████▍    | 3481/6470 [43:12<38:38,  1.29it/s]
4|3482|Loss: 0.8620650768280029:  54%|█████▍    | 3481/6470 [43:13<38:38,  1.29it/s]
4|3482|Loss: 0.8620650768280029:  54%|█████▍    | 3482/6470 [43:13<38:54,  1.28it/s]
4|3483|Loss: 0.7099494338035583:  54%|█████▍    | 3482/6470 [43:14<38:54,  1.28it/s]
4|3483|Loss: 0.7099494338035583:  54%|█████▍    | 3483/6470 [43:14<38:40,  1.29it/s]
4|3484|Loss: 0.9256194829940796:  54%|█████▍    | 3483/6470 [43:14<38:40,  1.29it/s]
4|3484|Loss: 0.9256194829940796:  54%|█████▍    | 3484/6470 [43:15<37:33,  1.32it/s]
4|3485|Loss: 0.8203582763671875:  54%|█████▍    | 3484/6470 [43:15<37:33,  1.32it/s]
4|3485|Loss: 0.8203582763671875:  54%|█████▍    | 3485/6470 [43:16<38:42,  1.29it/s]
4|3486|Loss: 0.8330930471420288:  54%|█████▍    | 3485/6470 [43:16<38:42,  1.29it/s]
4|3486|Loss: 0.8330930471420288:  54%|█████▍    | 3486/6470 [43:16<37:44,  1.32it/s]
4|3487|Loss: 0.592178463935852:  54%|█████▍    | 3486/6470 [43:17<37:44,  1.32it/s] 
4|3487|Loss: 0.592178463935852:  54%|█████▍    | 3487/6470 [43:17<35:22,  1.41it/s]
4|3488|Loss: 0.766411304473877:  54%|█████▍    | 3487/6470 [43:17<35:22,  1.41it/s]
4|3488|Loss: 0.766411304473877:  54%|█████▍    | 3488/6470 [43:17<32:35,  1.53it/s]
4|3489|Loss: 0.78276127576828:  54%|█████▍    | 3488/6470 [43:18<32:35,  1.53it/s] 
4|3489|Loss: 0.78276127576828:  54%|█████▍    | 3489/6470 [43:18<32:38,  1.52it/s]
4|3490|Loss: 0.8546851277351379:  54%|█████▍    | 3489/6470 [43:18<32:38,  1.52it/s]
4|3490|Loss: 0.8546851277351379:  54%|█████▍    | 3490/6470 [43:19<33:20,  1.49it/s]
4|3491|Loss: 0.6934459209442139:  54%|█████▍    | 3490/6470 [43:19<33:20,  1.49it/s]
4|3491|Loss: 0.6934459209442139:  54%|█████▍    | 3491/6470 [43:19<31:26,  1.58it/s]
4|3492|Loss: 0.8714075088500977:  54%|█████▍    | 3491/6470 [43:20<31:26,  1.58it/s]
4|3492|Loss: 0.8714075088500977:  54%|█████▍    | 3492/6470 [43:20<32:18,  1.54it/s]
4|3493|Loss: 0.7556195855140686:  54%|█████▍    | 3492/6470 [43:20<32:18,  1.54it/s]
4|3493|Loss: 0.7556195855140686:  54%|█████▍    | 3493/6470 [43:21<32:50,  1.51it/s]
4|3494|Loss: 0.7662428021430969:  54%|█████▍    | 3493/6470 [43:21<32:50,  1.51it/s]
4|3494|Loss: 0.7662428021430969:  54%|█████▍    | 3494/6470 [43:21<31:52,  1.56it/s]
4|3495|Loss: 0.7598478198051453:  54%|█████▍    | 3494/6470 [43:22<31:52,  1.56it/s]
4|3495|Loss: 0.7598478198051453:  54%|█████▍    | 3495/6470 [43:22<32:21,  1.53it/s]
4|3496|Loss: 0.8798479437828064:  54%|█████▍    | 3495/6470 [43:22<32:21,  1.53it/s]
4|3496|Loss: 0.8798479437828064:  54%|█████▍    | 3496/6470 [43:23<32:25,  1.53it/s]
4|3497|Loss: 0.6521293520927429:  54%|█████▍    | 3496/6470 [43:23<32:25,  1.53it/s]
4|3497|Loss: 0.6521293520927429:  54%|█████▍    | 3497/6470 [43:23<33:45,  1.47it/s]
4|3498|Loss: 0.7426450848579407:  54%|█████▍    | 3497/6470 [43:24<33:45,  1.47it/s]
4|3498|Loss: 0.7426450848579407:  54%|█████▍    | 3498/6470 [43:24<36:55,  1.34it/s]
4|3499|Loss: 0.9996614456176758:  54%|█████▍    | 3498/6470 [43:25<36:55,  1.34it/s]
4|3499|Loss: 0.9996614456176758:  54%|█████▍    | 3499/6470 [43:25<36:33,  1.35it/s]
4|3500|Loss: 0.8878433108329773:  54%|█████▍    | 3499/6470 [43:25<36:33,  1.35it/s]
4|3500|Loss: 0.8878433108329773:  54%|█████▍    | 3500/6470 [43:26<37:04,  1.33it/s]
4|3501|Loss: 0.7451034188270569:  54%|█████▍    | 3500/6470 [43:26<37:04,  1.33it/s]
4|3501|Loss: 0.7451034188270569:  54%|█████▍    | 3501/6470 [43:26<35:11,  1.41it/s]
4|3502|Loss: 0.7746375203132629:  54%|█████▍    | 3501/6470 [43:27<35:11,  1.41it/s]
4|3502|Loss: 0.7746375203132629:  54%|█████▍    | 3502/6470 [43:27<34:09,  1.45it/s]
4|3503|Loss: 0.7042778730392456:  54%|█████▍    | 3502/6470 [43:27<34:09,  1.45it/s]
4|3503|Loss: 0.7042778730392456:  54%|█████▍    | 3503/6470 [43:28<33:24,  1.48it/s]
4|3504|Loss: 0.7735493779182434:  54%|█████▍    | 3503/6470 [43:28<33:24,  1.48it/s]
4|3504|Loss: 0.7735493779182434:  54%|█████▍    | 3504/6470 [43:29<36:13,  1.36it/s]
4|3505|Loss: 0.7287415862083435:  54%|█████▍    | 3504/6470 [43:29<36:13,  1.36it/s]
4|3505|Loss: 0.7287415862083435:  54%|█████▍    | 3505/6470 [43:29<38:53,  1.27it/s]
4|3506|Loss: 0.9386833906173706:  54%|█████▍    | 3505/6470 [43:30<38:53,  1.27it/s]
4|3506|Loss: 0.9386833906173706:  54%|█████▍    | 3506/6470 [43:30<40:56,  1.21it/s]
4|3507|Loss: 0.734786331653595:  54%|█████▍    | 3506/6470 [43:31<40:56,  1.21it/s] 
4|3507|Loss: 0.734786331653595:  54%|█████▍    | 3507/6470 [43:31<40:26,  1.22it/s]
4|3508|Loss: 0.8152141571044922:  54%|█████▍    | 3507/6470 [43:32<40:26,  1.22it/s]
4|3508|Loss: 0.8152141571044922:  54%|█████▍    | 3508/6470 [43:32<39:29,  1.25it/s]
4|3509|Loss: 0.8253459334373474:  54%|█████▍    | 3508/6470 [43:32<39:29,  1.25it/s]
4|3509|Loss: 0.8253459334373474:  54%|█████▍    | 3509/6470 [43:33<38:16,  1.29it/s]
4|3510|Loss: 0.73056560754776:  54%|█████▍    | 3509/6470 [43:33<38:16,  1.29it/s]  
4|3510|Loss: 0.73056560754776:  54%|█████▍    | 3510/6470 [43:33<36:10,  1.36it/s]
4|3511|Loss: 0.8512681126594543:  54%|█████▍    | 3510/6470 [43:34<36:10,  1.36it/s]
4|3511|Loss: 0.8512681126594543:  54%|█████▍    | 3511/6470 [43:34<34:55,  1.41it/s]
4|3512|Loss: 0.961277186870575:  54%|█████▍    | 3511/6470 [43:34<34:55,  1.41it/s] 
4|3512|Loss: 0.961277186870575:  54%|█████▍    | 3512/6470 [43:35<35:23,  1.39it/s]
4|3513|Loss: 0.7885122299194336:  54%|█████▍    | 3512/6470 [43:35<35:23,  1.39it/s]
4|3513|Loss: 0.7885122299194336:  54%|█████▍    | 3513/6470 [43:35<35:55,  1.37it/s]
4|3514|Loss: 0.8682165145874023:  54%|█████▍    | 3513/6470 [43:36<35:55,  1.37it/s]
4|3514|Loss: 0.8682165145874023:  54%|█████▍    | 3514/6470 [43:36<35:28,  1.39it/s]
4|3515|Loss: 0.9315046072006226:  54%|█████▍    | 3514/6470 [43:36<35:28,  1.39it/s]
4|3515|Loss: 0.9315046072006226:  54%|█████▍    | 3515/6470 [43:37<33:11,  1.48it/s]
4|3516|Loss: 0.8637990951538086:  54%|█████▍    | 3515/6470 [43:37<33:11,  1.48it/s]
4|3516|Loss: 0.8637990951538086:  54%|█████▍    | 3516/6470 [43:37<31:30,  1.56it/s]
4|3517|Loss: 0.6960822939872742:  54%|█████▍    | 3516/6470 [43:38<31:30,  1.56it/s]
4|3517|Loss: 0.6960822939872742:  54%|█████▍    | 3517/6470 [43:38<34:42,  1.42it/s]
4|3518|Loss: 0.9000400304794312:  54%|█████▍    | 3517/6470 [43:39<34:42,  1.42it/s]
4|3518|Loss: 0.9000400304794312:  54%|█████▍    | 3518/6470 [43:39<36:26,  1.35it/s]
4|3519|Loss: 0.7485566735267639:  54%|█████▍    | 3518/6470 [43:39<36:26,  1.35it/s]
4|3519|Loss: 0.7485566735267639:  54%|█████▍    | 3519/6470 [43:40<38:26,  1.28it/s]
4|3520|Loss: 0.899276614189148:  54%|█████▍    | 3519/6470 [43:40<38:26,  1.28it/s] 
4|3520|Loss: 0.899276614189148:  54%|█████▍    | 3520/6470 [43:40<35:57,  1.37it/s]
4|3521|Loss: 0.7183340787887573:  54%|█████▍    | 3520/6470 [43:41<35:57,  1.37it/s]
4|3521|Loss: 0.7183340787887573:  54%|█████▍    | 3521/6470 [43:41<34:05,  1.44it/s]
4|3522|Loss: 0.7041187882423401:  54%|█████▍    | 3521/6470 [43:41<34:05,  1.44it/s]
4|3522|Loss: 0.7041187882423401:  54%|█████▍    | 3522/6470 [43:42<32:08,  1.53it/s]
4|3523|Loss: 0.797124981880188:  54%|█████▍    | 3522/6470 [43:42<32:08,  1.53it/s] 
4|3523|Loss: 0.797124981880188:  54%|█████▍    | 3523/6470 [43:42<32:18,  1.52it/s]
4|3524|Loss: 0.8248416185379028:  54%|█████▍    | 3523/6470 [43:43<32:18,  1.52it/s]
4|3524|Loss: 0.8248416185379028:  54%|█████▍    | 3524/6470 [43:43<34:31,  1.42it/s]
4|3525|Loss: 0.8719522356987:  54%|█████▍    | 3524/6470 [43:43<34:31,  1.42it/s]   
4|3525|Loss: 0.8719522356987:  54%|█████▍    | 3525/6470 [43:44<34:31,  1.42it/s]
4|3526|Loss: 0.7886689901351929:  54%|█████▍    | 3525/6470 [43:44<34:31,  1.42it/s]
4|3526|Loss: 0.7886689901351929:  54%|█████▍    | 3526/6470 [43:44<33:30,  1.46it/s]
4|3527|Loss: 0.7951411008834839:  54%|█████▍    | 3526/6470 [43:45<33:30,  1.46it/s]
4|3527|Loss: 0.7951411008834839:  55%|█████▍    | 3527/6470 [43:45<34:09,  1.44it/s]
4|3528|Loss: 0.9253420829772949:  55%|█████▍    | 3527/6470 [43:46<34:09,  1.44it/s]
4|3528|Loss: 0.9253420829772949:  55%|█████▍    | 3528/6470 [43:46<36:49,  1.33it/s]
4|3529|Loss: 0.7016746401786804:  55%|█████▍    | 3528/6470 [43:46<36:49,  1.33it/s]
4|3529|Loss: 0.7016746401786804:  55%|█████▍    | 3529/6470 [43:47<36:42,  1.34it/s]
4|3530|Loss: 0.9269037842750549:  55%|█████▍    | 3529/6470 [43:47<36:42,  1.34it/s]
4|3530|Loss: 0.9269037842750549:  55%|█████▍    | 3530/6470 [43:47<35:05,  1.40it/s]
4|3531|Loss: 0.6920256614685059:  55%|█████▍    | 3530/6470 [43:48<35:05,  1.40it/s]
4|3531|Loss: 0.6920256614685059:  55%|█████▍    | 3531/6470 [43:48<35:50,  1.37it/s]
4|3532|Loss: 0.7206078171730042:  55%|█████▍    | 3531/6470 [43:49<35:50,  1.37it/s]
4|3532|Loss: 0.7206078171730042:  55%|█████▍    | 3532/6470 [43:49<36:59,  1.32it/s]
4|3533|Loss: 0.8876847624778748:  55%|█████▍    | 3532/6470 [43:49<36:59,  1.32it/s]
4|3533|Loss: 0.8876847624778748:  55%|█████▍    | 3533/6470 [43:50<37:28,  1.31it/s]
4|3534|Loss: 0.8735692501068115:  55%|█████▍    | 3533/6470 [43:50<37:28,  1.31it/s]
4|3534|Loss: 0.8735692501068115:  55%|█████▍    | 3534/6470 [43:50<36:21,  1.35it/s]
4|3535|Loss: 0.807170569896698:  55%|█████▍    | 3534/6470 [43:51<36:21,  1.35it/s] 
4|3535|Loss: 0.807170569896698:  55%|█████▍    | 3535/6470 [43:51<36:58,  1.32it/s]
4|3536|Loss: 0.7766774892807007:  55%|█████▍    | 3535/6470 [43:52<36:58,  1.32it/s]
4|3536|Loss: 0.7766774892807007:  55%|█████▍    | 3536/6470 [43:52<37:32,  1.30it/s]
4|3537|Loss: 0.7309522032737732:  55%|█████▍    | 3536/6470 [43:52<37:32,  1.30it/s]
4|3537|Loss: 0.7309522032737732:  55%|█████▍    | 3537/6470 [43:53<35:50,  1.36it/s]
4|3538|Loss: 0.7941573858261108:  55%|█████▍    | 3537/6470 [43:53<35:50,  1.36it/s]
4|3538|Loss: 0.7941573858261108:  55%|█████▍    | 3538/6470 [43:54<37:24,  1.31it/s]
4|3539|Loss: 0.7160357236862183:  55%|█████▍    | 3538/6470 [43:54<37:24,  1.31it/s]
4|3539|Loss: 0.7160357236862183:  55%|█████▍    | 3539/6470 [43:54<33:14,  1.47it/s]
4|3540|Loss: 0.6162717938423157:  55%|█████▍    | 3539/6470 [43:54<33:14,  1.47it/s]
4|3540|Loss: 0.6162717938423157:  55%|█████▍    | 3540/6470 [43:55<35:00,  1.39it/s]
4|3541|Loss: 0.7070905566215515:  55%|█████▍    | 3540/6470 [43:55<35:00,  1.39it/s]
4|3541|Loss: 0.7070905566215515:  55%|█████▍    | 3541/6470 [43:56<36:15,  1.35it/s]
4|3542|Loss: 1.033753514289856:  55%|█████▍    | 3541/6470 [43:56<36:15,  1.35it/s] 
4|3542|Loss: 1.033753514289856:  55%|█████▍    | 3542/6470 [43:56<35:49,  1.36it/s]
4|3543|Loss: 0.8022315502166748:  55%|█████▍    | 3542/6470 [43:57<35:49,  1.36it/s]
4|3543|Loss: 0.8022315502166748:  55%|█████▍    | 3543/6470 [43:57<37:36,  1.30it/s]
4|3544|Loss: 0.9002453088760376:  55%|█████▍    | 3543/6470 [43:58<37:36,  1.30it/s]
4|3544|Loss: 0.9002453088760376:  55%|█████▍    | 3544/6470 [43:58<36:44,  1.33it/s]
4|3545|Loss: 0.6565040946006775:  55%|█████▍    | 3544/6470 [43:58<36:44,  1.33it/s]
4|3545|Loss: 0.6565040946006775:  55%|█████▍    | 3545/6470 [43:59<35:55,  1.36it/s]
4|3546|Loss: 0.7262365818023682:  55%|█████▍    | 3545/6470 [43:59<35:55,  1.36it/s]
4|3546|Loss: 0.7262365818023682:  55%|█████▍    | 3546/6470 [43:59<36:02,  1.35it/s]
4|3547|Loss: 0.933185338973999:  55%|█████▍    | 3546/6470 [44:00<36:02,  1.35it/s] 
4|3547|Loss: 0.933185338973999:  55%|█████▍    | 3547/6470 [44:00<36:25,  1.34it/s]
4|3548|Loss: 0.8676807880401611:  55%|█████▍    | 3547/6470 [44:01<36:25,  1.34it/s]
4|3548|Loss: 0.8676807880401611:  55%|█████▍    | 3548/6470 [44:01<37:49,  1.29it/s]
4|3549|Loss: 0.81035977602005:  55%|█████▍    | 3548/6470 [44:01<37:49,  1.29it/s]  
4|3549|Loss: 0.81035977602005:  55%|█████▍    | 3549/6470 [44:02<34:34,  1.41it/s]
4|3550|Loss: 0.8957394361495972:  55%|█████▍    | 3549/6470 [44:02<34:34,  1.41it/s]
4|3550|Loss: 0.8957394361495972:  55%|█████▍    | 3550/6470 [44:02<33:12,  1.47it/s]
4|3551|Loss: 0.9704233407974243:  55%|█████▍    | 3550/6470 [44:03<33:12,  1.47it/s]
4|3551|Loss: 0.9704233407974243:  55%|█████▍    | 3551/6470 [44:03<34:47,  1.40it/s]
4|3552|Loss: 0.8161945343017578:  55%|█████▍    | 3551/6470 [44:03<34:47,  1.40it/s]
4|3552|Loss: 0.8161945343017578:  55%|█████▍    | 3552/6470 [44:04<36:10,  1.34it/s]
4|3553|Loss: 0.805711030960083:  55%|█████▍    | 3552/6470 [44:04<36:10,  1.34it/s] 
4|3553|Loss: 0.805711030960083:  55%|█████▍    | 3553/6470 [44:05<37:42,  1.29it/s]
4|3554|Loss: 0.830802321434021:  55%|█████▍    | 3553/6470 [44:05<37:42,  1.29it/s]
4|3554|Loss: 0.830802321434021:  55%|█████▍    | 3554/6470 [44:05<37:39,  1.29it/s]
4|3555|Loss: 0.8910005688667297:  55%|█████▍    | 3554/6470 [44:06<37:39,  1.29it/s]
4|3555|Loss: 0.8910005688667297:  55%|█████▍    | 3555/6470 [44:06<37:42,  1.29it/s]
4|3556|Loss: 0.7783899903297424:  55%|█████▍    | 3555/6470 [44:07<37:42,  1.29it/s]
4|3556|Loss: 0.7783899903297424:  55%|█████▍    | 3556/6470 [44:07<39:07,  1.24it/s]
4|3557|Loss: 0.928106427192688:  55%|█████▍    | 3556/6470 [44:07<39:07,  1.24it/s] 
4|3557|Loss: 0.928106427192688:  55%|█████▍    | 3557/6470 [44:08<38:08,  1.27it/s]
4|3558|Loss: 0.7796393036842346:  55%|█████▍    | 3557/6470 [44:08<38:08,  1.27it/s]
4|3558|Loss: 0.7796393036842346:  55%|█████▍    | 3558/6470 [44:08<36:10,  1.34it/s]
4|3559|Loss: 0.9131190776824951:  55%|█████▍    | 3558/6470 [44:09<36:10,  1.34it/s]
4|3559|Loss: 0.9131190776824951:  55%|█████▌    | 3559/6470 [44:09<34:12,  1.42it/s]
4|3560|Loss: 0.8091118335723877:  55%|█████▌    | 3559/6470 [44:09<34:12,  1.42it/s]
4|3560|Loss: 0.8091118335723877:  55%|█████▌    | 3560/6470 [44:10<35:10,  1.38it/s]
4|3561|Loss: 0.797967791557312:  55%|█████▌    | 3560/6470 [44:10<35:10,  1.38it/s] 
4|3561|Loss: 0.797967791557312:  55%|█████▌    | 3561/6470 [44:10<34:13,  1.42it/s]
4|3562|Loss: 0.875962495803833:  55%|█████▌    | 3561/6470 [44:11<34:13,  1.42it/s]
4|3562|Loss: 0.875962495803833:  55%|█████▌    | 3562/6470 [44:11<33:22,  1.45it/s]
4|3563|Loss: 0.8791365027427673:  55%|█████▌    | 3562/6470 [44:12<33:22,  1.45it/s]
4|3563|Loss: 0.8791365027427673:  55%|█████▌    | 3563/6470 [44:12<35:35,  1.36it/s]
4|3564|Loss: 0.8382870554924011:  55%|█████▌    | 3563/6470 [44:12<35:35,  1.36it/s]
4|3564|Loss: 0.8382870554924011:  55%|█████▌    | 3564/6470 [44:13<36:14,  1.34it/s]
4|3565|Loss: 0.895947277545929:  55%|█████▌    | 3564/6470 [44:13<36:14,  1.34it/s] 
4|3565|Loss: 0.895947277545929:  55%|█████▌    | 3565/6470 [44:13<35:44,  1.35it/s]
4|3566|Loss: 0.801773726940155:  55%|█████▌    | 3565/6470 [44:14<35:44,  1.35it/s]
4|3566|Loss: 0.801773726940155:  55%|█████▌    | 3566/6470 [44:14<31:26,  1.54it/s]
4|3567|Loss: 0.790346622467041:  55%|█████▌    | 3566/6470 [44:14<31:26,  1.54it/s]
4|3567|Loss: 0.790346622467041:  55%|█████▌    | 3567/6470 [44:14<30:50,  1.57it/s]
4|3568|Loss: 0.8395817279815674:  55%|█████▌    | 3567/6470 [44:15<30:50,  1.57it/s]
4|3568|Loss: 0.8395817279815674:  55%|█████▌    | 3568/6470 [44:15<29:24,  1.65it/s]
4|3569|Loss: 0.8509359955787659:  55%|█████▌    | 3568/6470 [44:15<29:24,  1.65it/s]
4|3569|Loss: 0.8509359955787659:  55%|█████▌    | 3569/6470 [44:16<32:27,  1.49it/s]
4|3570|Loss: 1.0351557731628418:  55%|█████▌    | 3569/6470 [44:16<32:27,  1.49it/s]
4|3570|Loss: 1.0351557731628418:  55%|█████▌    | 3570/6470 [44:17<32:59,  1.46it/s]
4|3571|Loss: 0.7478011250495911:  55%|█████▌    | 3570/6470 [44:17<32:59,  1.46it/s]
4|3571|Loss: 0.7478011250495911:  55%|█████▌    | 3571/6470 [44:17<36:33,  1.32it/s]
4|3572|Loss: 0.8510814309120178:  55%|█████▌    | 3571/6470 [44:18<36:33,  1.32it/s]
4|3572|Loss: 0.8510814309120178:  55%|█████▌    | 3572/6470 [44:18<36:34,  1.32it/s]
4|3573|Loss: 0.9473503828048706:  55%|█████▌    | 3572/6470 [44:19<36:34,  1.32it/s]
4|3573|Loss: 0.9473503828048706:  55%|█████▌    | 3573/6470 [44:19<35:52,  1.35it/s]
4|3574|Loss: 0.8473882079124451:  55%|█████▌    | 3573/6470 [44:19<35:52,  1.35it/s]
4|3574|Loss: 0.8473882079124451:  55%|█████▌    | 3574/6470 [44:20<35:21,  1.37it/s]
4|3575|Loss: 0.5813382267951965:  55%|█████▌    | 3574/6470 [44:20<35:21,  1.37it/s]
4|3575|Loss: 0.5813382267951965:  55%|█████▌    | 3575/6470 [44:20<35:27,  1.36it/s]
4|3576|Loss: 0.876980185508728:  55%|█████▌    | 3575/6470 [44:21<35:27,  1.36it/s] 
4|3576|Loss: 0.876980185508728:  55%|█████▌    | 3576/6470 [44:21<34:42,  1.39it/s]
4|3577|Loss: 0.8748232126235962:  55%|█████▌    | 3576/6470 [44:21<34:42,  1.39it/s]
4|3577|Loss: 0.8748232126235962:  55%|█████▌    | 3577/6470 [44:22<32:41,  1.47it/s]
4|3578|Loss: 0.8396714329719543:  55%|█████▌    | 3577/6470 [44:22<32:41,  1.47it/s]
4|3578|Loss: 0.8396714329719543:  55%|█████▌    | 3578/6470 [44:22<33:36,  1.43it/s]
4|3579|Loss: 0.8383564949035645:  55%|█████▌    | 3578/6470 [44:23<33:36,  1.43it/s]
4|3579|Loss: 0.8383564949035645:  55%|█████▌    | 3579/6470 [44:23<36:27,  1.32it/s]
4|3580|Loss: 0.8907909989356995:  55%|█████▌    | 3579/6470 [44:24<36:27,  1.32it/s]
4|3580|Loss: 0.8907909989356995:  55%|█████▌    | 3580/6470 [44:24<37:33,  1.28it/s]
4|3581|Loss: 0.7591726779937744:  55%|█████▌    | 3580/6470 [44:25<37:33,  1.28it/s]
4|3581|Loss: 0.7591726779937744:  55%|█████▌    | 3581/6470 [44:25<38:47,  1.24it/s]
4|3582|Loss: 0.7432985901832581:  55%|█████▌    | 3581/6470 [44:25<38:47,  1.24it/s]
4|3582|Loss: 0.7432985901832581:  55%|█████▌    | 3582/6470 [44:26<38:40,  1.24it/s]
4|3583|Loss: 0.82098788022995:  55%|█████▌    | 3582/6470 [44:26<38:40,  1.24it/s]  
4|3583|Loss: 0.82098788022995:  55%|█████▌    | 3583/6470 [44:27<38:25,  1.25it/s]
4|3584|Loss: 0.823479950428009:  55%|█████▌    | 3583/6470 [44:27<38:25,  1.25it/s]
4|3584|Loss: 0.823479950428009:  55%|█████▌    | 3584/6470 [44:27<37:07,  1.30it/s]
4|3585|Loss: 0.9692570567131042:  55%|█████▌    | 3584/6470 [44:28<37:07,  1.30it/s]
4|3585|Loss: 0.9692570567131042:  55%|█████▌    | 3585/6470 [44:28<36:03,  1.33it/s]
4|3586|Loss: 0.7085671424865723:  55%|█████▌    | 3585/6470 [44:28<36:03,  1.33it/s]
4|3586|Loss: 0.7085671424865723:  55%|█████▌    | 3586/6470 [44:29<34:33,  1.39it/s]
4|3587|Loss: 0.8024019598960876:  55%|█████▌    | 3586/6470 [44:29<34:33,  1.39it/s]
4|3587|Loss: 0.8024019598960876:  55%|█████▌    | 3587/6470 [44:29<35:59,  1.34it/s]
4|3588|Loss: 0.7493823170661926:  55%|█████▌    | 3587/6470 [44:30<35:59,  1.34it/s]
4|3588|Loss: 0.7493823170661926:  55%|█████▌    | 3588/6470 [44:30<34:55,  1.38it/s]
4|3589|Loss: 0.7665094137191772:  55%|█████▌    | 3588/6470 [44:31<34:55,  1.38it/s]
4|3589|Loss: 0.7665094137191772:  55%|█████▌    | 3589/6470 [44:31<35:09,  1.37it/s]
4|3590|Loss: 0.9958321452140808:  55%|█████▌    | 3589/6470 [44:31<35:09,  1.37it/s]
4|3590|Loss: 0.9958321452140808:  55%|█████▌    | 3590/6470 [44:32<37:08,  1.29it/s]
4|3591|Loss: 0.8601551055908203:  55%|█████▌    | 3590/6470 [44:32<37:08,  1.29it/s]
4|3591|Loss: 0.8601551055908203:  56%|█████▌    | 3591/6470 [44:33<37:09,  1.29it/s]
4|3592|Loss: 0.7698338031768799:  56%|█████▌    | 3591/6470 [44:33<37:09,  1.29it/s]
4|3592|Loss: 0.7698338031768799:  56%|█████▌    | 3592/6470 [44:33<38:15,  1.25it/s]
4|3593|Loss: 0.7835767865180969:  56%|█████▌    | 3592/6470 [44:34<38:15,  1.25it/s]
4|3593|Loss: 0.7835767865180969:  56%|█████▌    | 3593/6470 [44:34<37:13,  1.29it/s]
4|3594|Loss: 0.8400527238845825:  56%|█████▌    | 3593/6470 [44:34<37:13,  1.29it/s]
4|3594|Loss: 0.8400527238845825:  56%|█████▌    | 3594/6470 [44:35<36:44,  1.30it/s]
4|3595|Loss: 0.735113263130188:  56%|█████▌    | 3594/6470 [44:35<36:44,  1.30it/s] 
4|3595|Loss: 0.735113263130188:  56%|█████▌    | 3595/6470 [44:36<36:49,  1.30it/s]
4|3596|Loss: 0.8316267728805542:  56%|█████▌    | 3595/6470 [44:36<36:49,  1.30it/s]
4|3596|Loss: 0.8316267728805542:  56%|█████▌    | 3596/6470 [44:37<38:47,  1.24it/s]
4|3597|Loss: 0.8477799892425537:  56%|█████▌    | 3596/6470 [44:37<38:47,  1.24it/s]
4|3597|Loss: 0.8477799892425537:  56%|█████▌    | 3597/6470 [44:37<36:16,  1.32it/s]
4|3598|Loss: 0.835438072681427:  56%|█████▌    | 3597/6470 [44:37<36:16,  1.32it/s] 
4|3598|Loss: 0.835438072681427:  56%|█████▌    | 3598/6470 [44:38<34:00,  1.41it/s]
4|3599|Loss: 0.8261454701423645:  56%|█████▌    | 3598/6470 [44:38<34:00,  1.41it/s]
4|3599|Loss: 0.8261454701423645:  56%|█████▌    | 3599/6470 [44:38<31:04,  1.54it/s]
4|3600|Loss: 0.887688398361206:  56%|█████▌    | 3599/6470 [44:39<31:04,  1.54it/s] 
4|3600|Loss: 0.887688398361206:  56%|█████▌    | 3600/6470 [44:39<32:50,  1.46it/s]
4|3601|Loss: 0.715141773223877:  56%|█████▌    | 3600/6470 [44:39<32:50,  1.46it/s]
4|3601|Loss: 0.715141773223877:  56%|█████▌    | 3601/6470 [44:40<34:28,  1.39it/s]
4|3602|Loss: 0.8662573099136353:  56%|█████▌    | 3601/6470 [44:40<34:28,  1.39it/s]
4|3602|Loss: 0.8662573099136353:  56%|█████▌    | 3602/6470 [44:41<35:25,  1.35it/s]
4|3603|Loss: 0.78896164894104:  56%|█████▌    | 3602/6470 [44:41<35:25,  1.35it/s]  
4|3603|Loss: 0.78896164894104:  56%|█████▌    | 3603/6470 [44:41<36:00,  1.33it/s]
4|3604|Loss: 0.7655639052391052:  56%|█████▌    | 3603/6470 [44:42<36:00,  1.33it/s]
4|3604|Loss: 0.7655639052391052:  56%|█████▌    | 3604/6470 [44:42<38:57,  1.23it/s]
4|3605|Loss: 0.7544569373130798:  56%|█████▌    | 3604/6470 [44:43<38:57,  1.23it/s]
4|3605|Loss: 0.7544569373130798:  56%|█████▌    | 3605/6470 [44:43<41:23,  1.15it/s]
4|3606|Loss: 0.8449112772941589:  56%|█████▌    | 3605/6470 [44:44<41:23,  1.15it/s]
4|3606|Loss: 0.8449112772941589:  56%|█████▌    | 3606/6470 [44:44<41:11,  1.16it/s]
4|3607|Loss: 0.8975139856338501:  56%|█████▌    | 3606/6470 [44:45<41:11,  1.16it/s]
4|3607|Loss: 0.8975139856338501:  56%|█████▌    | 3607/6470 [44:45<37:30,  1.27it/s]
4|3608|Loss: 0.8818790912628174:  56%|█████▌    | 3607/6470 [44:45<37:30,  1.27it/s]
4|3608|Loss: 0.8818790912628174:  56%|█████▌    | 3608/6470 [44:46<36:59,  1.29it/s]
4|3609|Loss: 0.8827066421508789:  56%|█████▌    | 3608/6470 [44:46<36:59,  1.29it/s]
4|3609|Loss: 0.8827066421508789:  56%|█████▌    | 3609/6470 [44:46<34:14,  1.39it/s]
4|3610|Loss: 0.8766258358955383:  56%|█████▌    | 3609/6470 [44:47<34:14,  1.39it/s]
4|3610|Loss: 0.8766258358955383:  56%|█████▌    | 3610/6470 [44:47<35:09,  1.36it/s]
4|3611|Loss: 0.8238121867179871:  56%|█████▌    | 3610/6470 [44:47<35:09,  1.36it/s]
4|3611|Loss: 0.8238121867179871:  56%|█████▌    | 3611/6470 [44:48<37:50,  1.26it/s]
4|3612|Loss: 0.8296750783920288:  56%|█████▌    | 3611/6470 [44:48<37:50,  1.26it/s]
4|3612|Loss: 0.8296750783920288:  56%|█████▌    | 3612/6470 [44:49<36:57,  1.29it/s]
4|3613|Loss: 0.7535191178321838:  56%|█████▌    | 3612/6470 [44:49<36:57,  1.29it/s]
4|3613|Loss: 0.7535191178321838:  56%|█████▌    | 3613/6470 [44:49<38:34,  1.23it/s]
4|3614|Loss: 0.826280415058136:  56%|█████▌    | 3613/6470 [44:50<38:34,  1.23it/s] 
4|3614|Loss: 0.826280415058136:  56%|█████▌    | 3614/6470 [44:50<37:42,  1.26it/s]
4|3615|Loss: 0.8692267537117004:  56%|█████▌    | 3614/6470 [44:51<37:42,  1.26it/s]
4|3615|Loss: 0.8692267537117004:  56%|█████▌    | 3615/6470 [44:51<36:56,  1.29it/s]
4|3616|Loss: 0.8276029229164124:  56%|█████▌    | 3615/6470 [44:51<36:56,  1.29it/s]
4|3616|Loss: 0.8276029229164124:  56%|█████▌    | 3616/6470 [44:52<36:41,  1.30it/s]
4|3617|Loss: 0.865372896194458:  56%|█████▌    | 3616/6470 [44:52<36:41,  1.30it/s] 
4|3617|Loss: 0.865372896194458:  56%|█████▌    | 3617/6470 [44:52<35:47,  1.33it/s]
4|3618|Loss: 0.7734494805335999:  56%|█████▌    | 3617/6470 [44:53<35:47,  1.33it/s]
4|3618|Loss: 0.7734494805335999:  56%|█████▌    | 3618/6470 [44:53<37:46,  1.26it/s]
4|3619|Loss: 0.7927020192146301:  56%|█████▌    | 3618/6470 [44:54<37:46,  1.26it/s]
4|3619|Loss: 0.7927020192146301:  56%|█████▌    | 3619/6470 [44:54<36:47,  1.29it/s]
4|3620|Loss: 0.7474488019943237:  56%|█████▌    | 3619/6470 [44:55<36:47,  1.29it/s]
4|3620|Loss: 0.7474488019943237:  56%|█████▌    | 3620/6470 [44:55<38:43,  1.23it/s]
4|3621|Loss: 0.6867773532867432:  56%|█████▌    | 3620/6470 [44:56<38:43,  1.23it/s]
4|3621|Loss: 0.6867773532867432:  56%|█████▌    | 3621/6470 [44:56<41:30,  1.14it/s]
4|3622|Loss: 0.7681264877319336:  56%|█████▌    | 3621/6470 [44:57<41:30,  1.14it/s]
4|3622|Loss: 0.7681264877319336:  56%|█████▌    | 3622/6470 [44:57<42:08,  1.13it/s]
4|3623|Loss: 0.6767045259475708:  56%|█████▌    | 3622/6470 [44:57<42:08,  1.13it/s]
4|3623|Loss: 0.6767045259475708:  56%|█████▌    | 3623/6470 [44:58<42:38,  1.11it/s]
4|3624|Loss: 0.8140766024589539:  56%|█████▌    | 3623/6470 [44:58<42:38,  1.11it/s]
4|3624|Loss: 0.8140766024589539:  56%|█████▌    | 3624/6470 [44:59<41:42,  1.14it/s]
4|3625|Loss: 0.7110669016838074:  56%|█████▌    | 3624/6470 [44:59<41:42,  1.14it/s]
4|3625|Loss: 0.7110669016838074:  56%|█████▌    | 3625/6470 [45:00<42:54,  1.10it/s]
4|3626|Loss: 0.9893074631690979:  56%|█████▌    | 3625/6470 [45:00<42:54,  1.10it/s]
4|3626|Loss: 0.9893074631690979:  56%|█████▌    | 3626/6470 [45:00<41:58,  1.13it/s]
4|3627|Loss: 0.7874979972839355:  56%|█████▌    | 3626/6470 [45:01<41:58,  1.13it/s]
4|3627|Loss: 0.7874979972839355:  56%|█████▌    | 3627/6470 [45:01<39:24,  1.20it/s]
4|3628|Loss: 0.7788468599319458:  56%|█████▌    | 3627/6470 [45:02<39:24,  1.20it/s]
4|3628|Loss: 0.7788468599319458:  56%|█████▌    | 3628/6470 [45:02<38:34,  1.23it/s]
4|3629|Loss: 0.8959550261497498:  56%|█████▌    | 3628/6470 [45:02<38:34,  1.23it/s]
4|3629|Loss: 0.8959550261497498:  56%|█████▌    | 3629/6470 [45:03<37:13,  1.27it/s]
4|3630|Loss: 0.7028743624687195:  56%|█████▌    | 3629/6470 [45:03<37:13,  1.27it/s]
4|3630|Loss: 0.7028743624687195:  56%|█████▌    | 3630/6470 [45:03<36:10,  1.31it/s]
4|3631|Loss: 0.8167434930801392:  56%|█████▌    | 3630/6470 [45:04<36:10,  1.31it/s]
4|3631|Loss: 0.8167434930801392:  56%|█████▌    | 3631/6470 [45:04<34:52,  1.36it/s]
4|3632|Loss: 0.8227051496505737:  56%|█████▌    | 3631/6470 [45:04<34:52,  1.36it/s]
4|3632|Loss: 0.8227051496505737:  56%|█████▌    | 3632/6470 [45:05<36:50,  1.28it/s]
4|3633|Loss: 0.7516463398933411:  56%|█████▌    | 3632/6470 [45:05<36:50,  1.28it/s]
4|3633|Loss: 0.7516463398933411:  56%|█████▌    | 3633/6470 [45:06<36:30,  1.30it/s]
4|3634|Loss: 0.7392079830169678:  56%|█████▌    | 3633/6470 [45:06<36:30,  1.30it/s]
4|3634|Loss: 0.7392079830169678:  56%|█████▌    | 3634/6470 [45:06<35:34,  1.33it/s]
4|3635|Loss: 0.9854704737663269:  56%|█████▌    | 3634/6470 [45:07<35:34,  1.33it/s]
4|3635|Loss: 0.9854704737663269:  56%|█████▌    | 3635/6470 [45:07<37:17,  1.27it/s]
4|3636|Loss: 0.9204975962638855:  56%|█████▌    | 3635/6470 [45:08<37:17,  1.27it/s]
4|3636|Loss: 0.9204975962638855:  56%|█████▌    | 3636/6470 [45:08<35:15,  1.34it/s]
4|3637|Loss: 0.8545409440994263:  56%|█████▌    | 3636/6470 [45:08<35:15,  1.34it/s]
4|3637|Loss: 0.8545409440994263:  56%|█████▌    | 3637/6470 [45:09<35:30,  1.33it/s]
4|3638|Loss: 0.8058916330337524:  56%|█████▌    | 3637/6470 [45:09<35:30,  1.33it/s]
4|3638|Loss: 0.8058916330337524:  56%|█████▌    | 3638/6470 [45:09<36:24,  1.30it/s]
4|3639|Loss: 0.7298078536987305:  56%|█████▌    | 3638/6470 [45:10<36:24,  1.30it/s]
4|3639|Loss: 0.7298078536987305:  56%|█████▌    | 3639/6470 [45:10<35:10,  1.34it/s]
4|3640|Loss: 0.6908610463142395:  56%|█████▌    | 3639/6470 [45:10<35:10,  1.34it/s]
4|3640|Loss: 0.6908610463142395:  56%|█████▋    | 3640/6470 [45:11<30:51,  1.53it/s]
4|3641|Loss: 0.757958710193634:  56%|█████▋    | 3640/6470 [45:11<30:51,  1.53it/s] 
4|3641|Loss: 0.757958710193634:  56%|█████▋    | 3641/6470 [45:11<31:27,  1.50it/s]
4|3642|Loss: 0.8224318027496338:  56%|█████▋    | 3641/6470 [45:12<31:27,  1.50it/s]
4|3642|Loss: 0.8224318027496338:  56%|█████▋    | 3642/6470 [45:12<35:00,  1.35it/s]
4|3643|Loss: 0.6723166704177856:  56%|█████▋    | 3642/6470 [45:13<35:00,  1.35it/s]
4|3643|Loss: 0.6723166704177856:  56%|█████▋    | 3643/6470 [45:13<33:59,  1.39it/s]
4|3644|Loss: 0.8031584620475769:  56%|█████▋    | 3643/6470 [45:13<33:59,  1.39it/s]
4|3644|Loss: 0.8031584620475769:  56%|█████▋    | 3644/6470 [45:13<31:33,  1.49it/s]
4|3645|Loss: 0.8419173955917358:  56%|█████▋    | 3644/6470 [45:14<31:33,  1.49it/s]
4|3645|Loss: 0.8419173955917358:  56%|█████▋    | 3645/6470 [45:14<31:21,  1.50it/s]
4|3646|Loss: 0.7671008706092834:  56%|█████▋    | 3645/6470 [45:15<31:21,  1.50it/s]
4|3646|Loss: 0.7671008706092834:  56%|█████▋    | 3646/6470 [45:15<34:40,  1.36it/s]
4|3647|Loss: 0.6653996109962463:  56%|█████▋    | 3646/6470 [45:16<34:40,  1.36it/s]
4|3647|Loss: 0.6653996109962463:  56%|█████▋    | 3647/6470 [45:16<36:55,  1.27it/s]
4|3648|Loss: 0.8165937066078186:  56%|█████▋    | 3647/6470 [45:16<36:55,  1.27it/s]
4|3648|Loss: 0.8165937066078186:  56%|█████▋    | 3648/6470 [45:17<36:48,  1.28it/s]
4|3649|Loss: 0.8869667649269104:  56%|█████▋    | 3648/6470 [45:17<36:48,  1.28it/s]
4|3649|Loss: 0.8869667649269104:  56%|█████▋    | 3649/6470 [45:17<36:33,  1.29it/s]
4|3650|Loss: 0.6896365284919739:  56%|█████▋    | 3649/6470 [45:18<36:33,  1.29it/s]
4|3650|Loss: 0.6896365284919739:  56%|█████▋    | 3650/6470 [45:18<33:07,  1.42it/s]
4|3651|Loss: 0.8778073787689209:  56%|█████▋    | 3650/6470 [45:18<33:07,  1.42it/s]
4|3651|Loss: 0.8778073787689209:  56%|█████▋    | 3651/6470 [45:19<34:38,  1.36it/s]
4|3652|Loss: 0.9014685750007629:  56%|█████▋    | 3651/6470 [45:19<34:38,  1.36it/s]
4|3652|Loss: 0.9014685750007629:  56%|█████▋    | 3652/6470 [45:20<34:27,  1.36it/s]
4|3653|Loss: 1.0714584589004517:  56%|█████▋    | 3652/6470 [45:20<34:27,  1.36it/s]
4|3653|Loss: 1.0714584589004517:  56%|█████▋    | 3653/6470 [45:20<33:34,  1.40it/s]
4|3654|Loss: 0.8835307955741882:  56%|█████▋    | 3653/6470 [45:21<33:34,  1.40it/s]
4|3654|Loss: 0.8835307955741882:  56%|█████▋    | 3654/6470 [45:21<31:58,  1.47it/s]
4|3655|Loss: 0.8443736433982849:  56%|█████▋    | 3654/6470 [45:21<31:58,  1.47it/s]
4|3655|Loss: 0.8443736433982849:  56%|█████▋    | 3655/6470 [45:22<32:38,  1.44it/s]
4|3656|Loss: 0.7226340770721436:  56%|█████▋    | 3655/6470 [45:22<32:38,  1.44it/s]
4|3656|Loss: 0.7226340770721436:  57%|█████▋    | 3656/6470 [45:22<32:53,  1.43it/s]
4|3657|Loss: 0.8516359925270081:  57%|█████▋    | 3656/6470 [45:23<32:53,  1.43it/s]
4|3657|Loss: 0.8516359925270081:  57%|█████▋    | 3657/6470 [45:23<32:41,  1.43it/s]
4|3658|Loss: 0.7091512084007263:  57%|█████▋    | 3657/6470 [45:23<32:41,  1.43it/s]
4|3658|Loss: 0.7091512084007263:  57%|█████▋    | 3658/6470 [45:24<33:49,  1.39it/s]
4|3659|Loss: 0.8604536056518555:  57%|█████▋    | 3658/6470 [45:24<33:49,  1.39it/s]
4|3659|Loss: 0.8604536056518555:  57%|█████▋    | 3659/6470 [45:24<32:07,  1.46it/s]
4|3660|Loss: 0.6648603081703186:  57%|█████▋    | 3659/6470 [45:25<32:07,  1.46it/s]
4|3660|Loss: 0.6648603081703186:  57%|█████▋    | 3660/6470 [45:25<36:09,  1.30it/s]
4|3661|Loss: 0.8315295577049255:  57%|█████▋    | 3660/6470 [45:26<36:09,  1.30it/s]
4|3661|Loss: 0.8315295577049255:  57%|█████▋    | 3661/6470 [45:26<35:59,  1.30it/s]
4|3662|Loss: 0.7306567430496216:  57%|█████▋    | 3661/6470 [45:26<35:59,  1.30it/s]
4|3662|Loss: 0.7306567430496216:  57%|█████▋    | 3662/6470 [45:27<32:24,  1.44it/s]
4|3663|Loss: 0.7569487690925598:  57%|█████▋    | 3662/6470 [45:27<32:24,  1.44it/s]
4|3663|Loss: 0.7569487690925598:  57%|█████▋    | 3663/6470 [45:27<33:04,  1.41it/s]
4|3664|Loss: 0.8634005188941956:  57%|█████▋    | 3663/6470 [45:28<33:04,  1.41it/s]
4|3664|Loss: 0.8634005188941956:  57%|█████▋    | 3664/6470 [45:28<33:59,  1.38it/s]
4|3665|Loss: 0.8664469718933105:  57%|█████▋    | 3664/6470 [45:28<33:59,  1.38it/s]
4|3665|Loss: 0.8664469718933105:  57%|█████▋    | 3665/6470 [45:29<33:26,  1.40it/s]
4|3666|Loss: 0.8531990647315979:  57%|█████▋    | 3665/6470 [45:29<33:26,  1.40it/s]
4|3666|Loss: 0.8531990647315979:  57%|█████▋    | 3666/6470 [45:30<34:57,  1.34it/s]
4|3667|Loss: 0.8269359469413757:  57%|█████▋    | 3666/6470 [45:30<34:57,  1.34it/s]
4|3667|Loss: 0.8269359469413757:  57%|█████▋    | 3667/6470 [45:30<34:27,  1.36it/s]
4|3668|Loss: 0.8209995627403259:  57%|█████▋    | 3667/6470 [45:31<34:27,  1.36it/s]
4|3668|Loss: 0.8209995627403259:  57%|█████▋    | 3668/6470 [45:31<33:49,  1.38it/s]
4|3669|Loss: 0.8973761796951294:  57%|█████▋    | 3668/6470 [45:31<33:49,  1.38it/s]
4|3669|Loss: 0.8973761796951294:  57%|█████▋    | 3669/6470 [45:32<34:27,  1.36it/s]
4|3670|Loss: 0.844096302986145:  57%|█████▋    | 3669/6470 [45:32<34:27,  1.36it/s] 
4|3670|Loss: 0.844096302986145:  57%|█████▋    | 3670/6470 [45:33<35:07,  1.33it/s]
4|3671|Loss: 0.8484200835227966:  57%|█████▋    | 3670/6470 [45:33<35:07,  1.33it/s]
4|3671|Loss: 0.8484200835227966:  57%|█████▋    | 3671/6470 [45:33<35:43,  1.31it/s]
4|3672|Loss: 0.9766167998313904:  57%|█████▋    | 3671/6470 [45:34<35:43,  1.31it/s]
4|3672|Loss: 0.9766167998313904:  57%|█████▋    | 3672/6470 [45:34<36:09,  1.29it/s]
4|3673|Loss: 0.8128756284713745:  57%|█████▋    | 3672/6470 [45:35<36:09,  1.29it/s]
4|3673|Loss: 0.8128756284713745:  57%|█████▋    | 3673/6470 [45:35<38:00,  1.23it/s]
4|3674|Loss: 0.7740999460220337:  57%|█████▋    | 3673/6470 [45:36<38:00,  1.23it/s]
4|3674|Loss: 0.7740999460220337:  57%|█████▋    | 3674/6470 [45:36<39:30,  1.18it/s]
4|3675|Loss: 0.810665488243103:  57%|█████▋    | 3674/6470 [45:36<39:30,  1.18it/s] 
4|3675|Loss: 0.810665488243103:  57%|█████▋    | 3675/6470 [45:37<38:41,  1.20it/s]
4|3676|Loss: 0.7686819434165955:  57%|█████▋    | 3675/6470 [45:37<38:41,  1.20it/s]
4|3676|Loss: 0.7686819434165955:  57%|█████▋    | 3676/6470 [45:37<36:02,  1.29it/s]
4|3677|Loss: 0.9233476519584656:  57%|█████▋    | 3676/6470 [45:38<36:02,  1.29it/s]
4|3677|Loss: 0.9233476519584656:  57%|█████▋    | 3677/6470 [45:38<33:09,  1.40it/s]
4|3678|Loss: 0.9691107869148254:  57%|█████▋    | 3677/6470 [45:38<33:09,  1.40it/s]
4|3678|Loss: 0.9691107869148254:  57%|█████▋    | 3678/6470 [45:39<32:12,  1.44it/s]
4|3679|Loss: 0.7038532495498657:  57%|█████▋    | 3678/6470 [45:39<32:12,  1.44it/s]
4|3679|Loss: 0.7038532495498657:  57%|█████▋    | 3679/6470 [45:39<32:16,  1.44it/s]
4|3680|Loss: 1.0464272499084473:  57%|█████▋    | 3679/6470 [45:40<32:16,  1.44it/s]
4|3680|Loss: 1.0464272499084473:  57%|█████▋    | 3680/6470 [45:40<35:21,  1.32it/s]
4|3681|Loss: 0.8040750622749329:  57%|█████▋    | 3680/6470 [45:41<35:21,  1.32it/s]
4|3681|Loss: 0.8040750622749329:  57%|█████▋    | 3681/6470 [45:41<34:47,  1.34it/s]
4|3682|Loss: 0.7514138221740723:  57%|█████▋    | 3681/6470 [45:41<34:47,  1.34it/s]
4|3682|Loss: 0.7514138221740723:  57%|█████▋    | 3682/6470 [45:42<34:47,  1.34it/s]
4|3683|Loss: 0.9802848696708679:  57%|█████▋    | 3682/6470 [45:42<34:47,  1.34it/s]
4|3683|Loss: 0.9802848696708679:  57%|█████▋    | 3683/6470 [45:43<37:16,  1.25it/s]
4|3684|Loss: 0.8277649283409119:  57%|█████▋    | 3683/6470 [45:43<37:16,  1.25it/s]
4|3684|Loss: 0.8277649283409119:  57%|█████▋    | 3684/6470 [45:43<37:33,  1.24it/s]
4|3685|Loss: 0.8379507660865784:  57%|█████▋    | 3684/6470 [45:44<37:33,  1.24it/s]
4|3685|Loss: 0.8379507660865784:  57%|█████▋    | 3685/6470 [45:44<40:03,  1.16it/s]
4|3686|Loss: 0.9147574305534363:  57%|█████▋    | 3685/6470 [45:45<40:03,  1.16it/s]
4|3686|Loss: 0.9147574305534363:  57%|█████▋    | 3686/6470 [45:45<37:13,  1.25it/s]
4|3687|Loss: 1.0268038511276245:  57%|█████▋    | 3686/6470 [45:45<37:13,  1.25it/s]
4|3687|Loss: 1.0268038511276245:  57%|█████▋    | 3687/6470 [45:46<36:45,  1.26it/s]
4|3688|Loss: 0.7164698839187622:  57%|█████▋    | 3687/6470 [45:46<36:45,  1.26it/s]
4|3688|Loss: 0.7164698839187622:  57%|█████▋    | 3688/6470 [45:47<38:30,  1.20it/s]
4|3689|Loss: 0.8331333994865417:  57%|█████▋    | 3688/6470 [45:47<38:30,  1.20it/s]
4|3689|Loss: 0.8331333994865417:  57%|█████▋    | 3689/6470 [45:48<38:10,  1.21it/s]
4|3690|Loss: 0.9575232267379761:  57%|█████▋    | 3689/6470 [45:48<38:10,  1.21it/s]
4|3690|Loss: 0.9575232267379761:  57%|█████▋    | 3690/6470 [45:49<39:15,  1.18it/s]
4|3691|Loss: 0.757142186164856:  57%|█████▋    | 3690/6470 [45:49<39:15,  1.18it/s] 
4|3691|Loss: 0.757142186164856:  57%|█████▋    | 3691/6470 [45:49<37:54,  1.22it/s]
4|3692|Loss: 0.9521424770355225:  57%|█████▋    | 3691/6470 [45:50<37:54,  1.22it/s]
4|3692|Loss: 0.9521424770355225:  57%|█████▋    | 3692/6470 [45:50<38:36,  1.20it/s]
4|3693|Loss: 0.9767906069755554:  57%|█████▋    | 3692/6470 [45:51<38:36,  1.20it/s]
4|3693|Loss: 0.9767906069755554:  57%|█████▋    | 3693/6470 [45:51<37:22,  1.24it/s]
4|3694|Loss: 0.9808580279350281:  57%|█████▋    | 3693/6470 [45:51<37:22,  1.24it/s]
4|3694|Loss: 0.9808580279350281:  57%|█████▋    | 3694/6470 [45:52<38:34,  1.20it/s]
4|3695|Loss: 0.7734931707382202:  57%|█████▋    | 3694/6470 [45:52<38:34,  1.20it/s]
4|3695|Loss: 0.7734931707382202:  57%|█████▋    | 3695/6470 [45:52<36:12,  1.28it/s]
4|3696|Loss: 0.9511351585388184:  57%|█████▋    | 3695/6470 [45:53<36:12,  1.28it/s]
4|3696|Loss: 0.9511351585388184:  57%|█████▋    | 3696/6470 [45:53<37:14,  1.24it/s]
4|3697|Loss: 0.8637147545814514:  57%|█████▋    | 3696/6470 [45:54<37:14,  1.24it/s]
4|3697|Loss: 0.8637147545814514:  57%|█████▋    | 3697/6470 [45:54<35:19,  1.31it/s]
4|3698|Loss: 0.8983073830604553:  57%|█████▋    | 3697/6470 [45:54<35:19,  1.31it/s]
4|3698|Loss: 0.8983073830604553:  57%|█████▋    | 3698/6470 [45:55<36:07,  1.28it/s]
4|3699|Loss: 1.1121528148651123:  57%|█████▋    | 3698/6470 [45:55<36:07,  1.28it/s]
4|3699|Loss: 1.1121528148651123:  57%|█████▋    | 3699/6470 [45:56<37:33,  1.23it/s]
4|3700|Loss: 0.7844128608703613:  57%|█████▋    | 3699/6470 [45:56<37:33,  1.23it/s]
4|3700|Loss: 0.7844128608703613:  57%|█████▋    | 3700/6470 [45:56<35:41,  1.29it/s]
4|3701|Loss: 0.9099088907241821:  57%|█████▋    | 3700/6470 [45:57<35:41,  1.29it/s]
4|3701|Loss: 0.9099088907241821:  57%|█████▋    | 3701/6470 [45:57<37:44,  1.22it/s]
4|3702|Loss: 0.7826359868049622:  57%|█████▋    | 3701/6470 [45:58<37:44,  1.22it/s]
4|3702|Loss: 0.7826359868049622:  57%|█████▋    | 3702/6470 [45:58<36:57,  1.25it/s]
4|3703|Loss: 0.8132727146148682:  57%|█████▋    | 3702/6470 [45:58<36:57,  1.25it/s]
4|3703|Loss: 0.8132727146148682:  57%|█████▋    | 3703/6470 [45:59<34:43,  1.33it/s]
4|3704|Loss: 0.8796794414520264:  57%|█████▋    | 3703/6470 [45:59<34:43,  1.33it/s]
4|3704|Loss: 0.8796794414520264:  57%|█████▋    | 3704/6470 [45:59<33:14,  1.39it/s]
4|3705|Loss: 0.8370634913444519:  57%|█████▋    | 3704/6470 [46:00<33:14,  1.39it/s]
4|3705|Loss: 0.8370634913444519:  57%|█████▋    | 3705/6470 [46:00<32:42,  1.41it/s]
4|3706|Loss: 0.7719125747680664:  57%|█████▋    | 3705/6470 [46:00<32:42,  1.41it/s]
4|3706|Loss: 0.7719125747680664:  57%|█████▋    | 3706/6470 [46:01<32:17,  1.43it/s]
4|3707|Loss: 0.7267473340034485:  57%|█████▋    | 3706/6470 [46:01<32:17,  1.43it/s]
4|3707|Loss: 0.7267473340034485:  57%|█████▋    | 3707/6470 [46:01<33:30,  1.37it/s]
4|3708|Loss: 0.7981501221656799:  57%|█████▋    | 3707/6470 [46:02<33:30,  1.37it/s]
4|3708|Loss: 0.7981501221656799:  57%|█████▋    | 3708/6470 [46:02<32:16,  1.43it/s]
4|3709|Loss: 0.6704870462417603:  57%|█████▋    | 3708/6470 [46:02<32:16,  1.43it/s]
4|3709|Loss: 0.6704870462417603:  57%|█████▋    | 3709/6470 [46:03<31:47,  1.45it/s]
4|3710|Loss: 0.7623127698898315:  57%|█████▋    | 3709/6470 [46:03<31:47,  1.45it/s]
4|3710|Loss: 0.7623127698898315:  57%|█████▋    | 3710/6470 [46:04<33:02,  1.39it/s]
4|3711|Loss: 0.9013145565986633:  57%|█████▋    | 3710/6470 [46:04<33:02,  1.39it/s]
4|3711|Loss: 0.9013145565986633:  57%|█████▋    | 3711/6470 [46:04<34:08,  1.35it/s]
4|3712|Loss: 0.8628131747245789:  57%|█████▋    | 3711/6470 [46:05<34:08,  1.35it/s]
4|3712|Loss: 0.8628131747245789:  57%|█████▋    | 3712/6470 [46:05<34:50,  1.32it/s]
4|3713|Loss: 0.7836999893188477:  57%|█████▋    | 3712/6470 [46:06<34:50,  1.32it/s]
4|3713|Loss: 0.7836999893188477:  57%|█████▋    | 3713/6470 [46:06<35:17,  1.30it/s]
4|3714|Loss: 0.9263415932655334:  57%|█████▋    | 3713/6470 [46:06<35:17,  1.30it/s]
4|3714|Loss: 0.9263415932655334:  57%|█████▋    | 3714/6470 [46:07<34:10,  1.34it/s]
4|3715|Loss: 0.7756919264793396:  57%|█████▋    | 3714/6470 [46:07<34:10,  1.34it/s]
4|3715|Loss: 0.7756919264793396:  57%|█████▋    | 3715/6470 [46:07<33:03,  1.39it/s]
4|3716|Loss: 1.0330168008804321:  57%|█████▋    | 3715/6470 [46:08<33:03,  1.39it/s]
4|3716|Loss: 1.0330168008804321:  57%|█████▋    | 3716/6470 [46:08<35:27,  1.29it/s]
4|3717|Loss: 0.8134557604789734:  57%|█████▋    | 3716/6470 [46:09<35:27,  1.29it/s]
4|3717|Loss: 0.8134557604789734:  57%|█████▋    | 3717/6470 [46:09<34:52,  1.32it/s]
4|3718|Loss: 0.6233786940574646:  57%|█████▋    | 3717/6470 [46:09<34:52,  1.32it/s]
4|3718|Loss: 0.6233786940574646:  57%|█████▋    | 3718/6470 [46:10<35:07,  1.31it/s]
4|3719|Loss: 0.7843798995018005:  57%|█████▋    | 3718/6470 [46:10<35:07,  1.31it/s]
4|3719|Loss: 0.7843798995018005:  57%|█████▋    | 3719/6470 [46:10<34:12,  1.34it/s]
4|3720|Loss: 0.9043251872062683:  57%|█████▋    | 3719/6470 [46:11<34:12,  1.34it/s]
4|3720|Loss: 0.9043251872062683:  57%|█████▋    | 3720/6470 [46:11<35:16,  1.30it/s]
4|3721|Loss: 0.8576822876930237:  57%|█████▋    | 3720/6470 [46:12<35:16,  1.30it/s]
4|3721|Loss: 0.8576822876930237:  58%|█████▊    | 3721/6470 [46:12<37:26,  1.22it/s]
4|3722|Loss: 0.8446095585823059:  58%|█████▊    | 3721/6470 [46:13<37:26,  1.22it/s]
4|3722|Loss: 0.8446095585823059:  58%|█████▊    | 3722/6470 [46:13<35:24,  1.29it/s]
4|3723|Loss: 0.9710972309112549:  58%|█████▊    | 3722/6470 [46:13<35:24,  1.29it/s]
4|3723|Loss: 0.9710972309112549:  58%|█████▊    | 3723/6470 [46:14<34:11,  1.34it/s]
4|3724|Loss: 0.8125137686729431:  58%|█████▊    | 3723/6470 [46:14<34:11,  1.34it/s]
4|3724|Loss: 0.8125137686729431:  58%|█████▊    | 3724/6470 [46:14<34:51,  1.31it/s]
4|3725|Loss: 0.6993476748466492:  58%|█████▊    | 3724/6470 [46:15<34:51,  1.31it/s]
4|3725|Loss: 0.6993476748466492:  58%|█████▊    | 3725/6470 [46:15<33:21,  1.37it/s]
4|3726|Loss: 0.9555261731147766:  58%|█████▊    | 3725/6470 [46:15<33:21,  1.37it/s]
4|3726|Loss: 0.9555261731147766:  58%|█████▊    | 3726/6470 [46:16<32:22,  1.41it/s]
4|3727|Loss: 0.7345893383026123:  58%|█████▊    | 3726/6470 [46:16<32:22,  1.41it/s]
4|3727|Loss: 0.7345893383026123:  58%|█████▊    | 3727/6470 [46:16<30:53,  1.48it/s]
4|3728|Loss: 0.7637813687324524:  58%|█████▊    | 3727/6470 [46:17<30:53,  1.48it/s]
4|3728|Loss: 0.7637813687324524:  58%|█████▊    | 3728/6470 [46:17<29:16,  1.56it/s]
4|3729|Loss: 0.7186892628669739:  58%|█████▊    | 3728/6470 [46:17<29:16,  1.56it/s]
4|3729|Loss: 0.7186892628669739:  58%|█████▊    | 3729/6470 [46:18<32:20,  1.41it/s]
4|3730|Loss: 0.7478334903717041:  58%|█████▊    | 3729/6470 [46:18<32:20,  1.41it/s]
4|3730|Loss: 0.7478334903717041:  58%|█████▊    | 3730/6470 [46:18<33:30,  1.36it/s]
4|3731|Loss: 0.8194195628166199:  58%|█████▊    | 3730/6470 [46:19<33:30,  1.36it/s]
4|3731|Loss: 0.8194195628166199:  58%|█████▊    | 3731/6470 [46:19<35:18,  1.29it/s]
4|3732|Loss: 0.80588299036026:  58%|█████▊    | 3731/6470 [46:20<35:18,  1.29it/s]  
4|3732|Loss: 0.80588299036026:  58%|█████▊    | 3732/6470 [46:20<34:18,  1.33it/s]
4|3733|Loss: 0.819379448890686:  58%|█████▊    | 3732/6470 [46:20<34:18,  1.33it/s]
4|3733|Loss: 0.819379448890686:  58%|█████▊    | 3733/6470 [46:21<32:15,  1.41it/s]
4|3734|Loss: 0.7492427825927734:  58%|█████▊    | 3733/6470 [46:21<32:15,  1.41it/s]
4|3734|Loss: 0.7492427825927734:  58%|█████▊    | 3734/6470 [46:21<31:49,  1.43it/s]
4|3735|Loss: 0.8941410779953003:  58%|█████▊    | 3734/6470 [46:22<31:49,  1.43it/s]
4|3735|Loss: 0.8941410779953003:  58%|█████▊    | 3735/6470 [46:22<33:33,  1.36it/s]
4|3736|Loss: 0.8355703353881836:  58%|█████▊    | 3735/6470 [46:23<33:33,  1.36it/s]
4|3736|Loss: 0.8355703353881836:  58%|█████▊    | 3736/6470 [46:23<35:38,  1.28it/s]
4|3737|Loss: 0.8094300031661987:  58%|█████▊    | 3736/6470 [46:23<35:38,  1.28it/s]
4|3737|Loss: 0.8094300031661987:  58%|█████▊    | 3737/6470 [46:24<35:31,  1.28it/s]
4|3738|Loss: 0.8010968565940857:  58%|█████▊    | 3737/6470 [46:24<35:31,  1.28it/s]
4|3738|Loss: 0.8010968565940857:  58%|█████▊    | 3738/6470 [46:24<34:08,  1.33it/s]
4|3739|Loss: 0.8353829979896545:  58%|█████▊    | 3738/6470 [46:25<34:08,  1.33it/s]
4|3739|Loss: 0.8353829979896545:  58%|█████▊    | 3739/6470 [46:25<33:18,  1.37it/s]
4|3740|Loss: 0.7089242935180664:  58%|█████▊    | 3739/6470 [46:26<33:18,  1.37it/s]
4|3740|Loss: 0.7089242935180664:  58%|█████▊    | 3740/6470 [46:26<33:26,  1.36it/s]
4|3741|Loss: 0.9574379920959473:  58%|█████▊    | 3740/6470 [46:26<33:26,  1.36it/s]
4|3741|Loss: 0.9574379920959473:  58%|█████▊    | 3741/6470 [46:27<35:05,  1.30it/s]
4|3742|Loss: 0.9139147400856018:  58%|█████▊    | 3741/6470 [46:27<35:05,  1.30it/s]
4|3742|Loss: 0.9139147400856018:  58%|█████▊    | 3742/6470 [46:27<34:38,  1.31it/s]
4|3743|Loss: 0.898507833480835:  58%|█████▊    | 3742/6470 [46:28<34:38,  1.31it/s] 
4|3743|Loss: 0.898507833480835:  58%|█████▊    | 3743/6470 [46:28<33:31,  1.36it/s]
4|3744|Loss: 0.9119267463684082:  58%|█████▊    | 3743/6470 [46:29<33:31,  1.36it/s]
4|3744|Loss: 0.9119267463684082:  58%|█████▊    | 3744/6470 [46:29<35:54,  1.27it/s]
4|3745|Loss: 0.8400612473487854:  58%|█████▊    | 3744/6470 [46:30<35:54,  1.27it/s]
4|3745|Loss: 0.8400612473487854:  58%|█████▊    | 3745/6470 [46:30<34:33,  1.31it/s]
4|3746|Loss: 0.7567538619041443:  58%|█████▊    | 3745/6470 [46:30<34:33,  1.31it/s]
4|3746|Loss: 0.7567538619041443:  58%|█████▊    | 3746/6470 [46:31<35:43,  1.27it/s]
4|3747|Loss: 0.8215184807777405:  58%|█████▊    | 3746/6470 [46:31<35:43,  1.27it/s]
4|3747|Loss: 0.8215184807777405:  58%|█████▊    | 3747/6470 [46:31<35:40,  1.27it/s]
4|3748|Loss: 0.8170291781425476:  58%|█████▊    | 3747/6470 [46:32<35:40,  1.27it/s]
4|3748|Loss: 0.8170291781425476:  58%|█████▊    | 3748/6470 [46:32<34:16,  1.32it/s]
4|3749|Loss: 0.8428157567977905:  58%|█████▊    | 3748/6470 [46:32<34:16,  1.32it/s]
4|3749|Loss: 0.8428157567977905:  58%|█████▊    | 3749/6470 [46:33<32:36,  1.39it/s]
4|3750|Loss: 0.8454521298408508:  58%|█████▊    | 3749/6470 [46:33<32:36,  1.39it/s]
4|3750|Loss: 0.8454521298408508:  58%|█████▊    | 3750/6470 [46:34<33:52,  1.34it/s]
4|3751|Loss: 0.791574239730835:  58%|█████▊    | 3750/6470 [46:34<33:52,  1.34it/s] 
4|3751|Loss: 0.791574239730835:  58%|█████▊    | 3751/6470 [46:34<36:04,  1.26it/s]
4|3752|Loss: 0.8829538822174072:  58%|█████▊    | 3751/6470 [46:35<36:04,  1.26it/s]
4|3752|Loss: 0.8829538822174072:  58%|█████▊    | 3752/6470 [46:35<36:31,  1.24it/s]
4|3753|Loss: 0.7204669713973999:  58%|█████▊    | 3752/6470 [46:36<36:31,  1.24it/s]
4|3753|Loss: 0.7204669713973999:  58%|█████▊    | 3753/6470 [46:36<34:12,  1.32it/s]
4|3754|Loss: 0.8174664378166199:  58%|█████▊    | 3753/6470 [46:36<34:12,  1.32it/s]
4|3754|Loss: 0.8174664378166199:  58%|█████▊    | 3754/6470 [46:36<31:55,  1.42it/s]
4|3755|Loss: 0.8631622791290283:  58%|█████▊    | 3754/6470 [46:37<31:55,  1.42it/s]
4|3755|Loss: 0.8631622791290283:  58%|█████▊    | 3755/6470 [46:37<32:16,  1.40it/s]
4|3756|Loss: 0.7138605713844299:  58%|█████▊    | 3755/6470 [46:38<32:16,  1.40it/s]
4|3756|Loss: 0.7138605713844299:  58%|█████▊    | 3756/6470 [46:38<29:50,  1.52it/s]
4|3757|Loss: 0.8334416747093201:  58%|█████▊    | 3756/6470 [46:38<29:50,  1.52it/s]
4|3757|Loss: 0.8334416747093201:  58%|█████▊    | 3757/6470 [46:39<31:29,  1.44it/s]
4|3758|Loss: 0.6677824854850769:  58%|█████▊    | 3757/6470 [46:39<31:29,  1.44it/s]
4|3758|Loss: 0.6677824854850769:  58%|█████▊    | 3758/6470 [46:39<29:41,  1.52it/s]
4|3759|Loss: 0.9419633150100708:  58%|█████▊    | 3758/6470 [46:39<29:41,  1.52it/s]
4|3759|Loss: 0.9419633150100708:  58%|█████▊    | 3759/6470 [46:40<29:10,  1.55it/s]
4|3760|Loss: 0.8414726257324219:  58%|█████▊    | 3759/6470 [46:40<29:10,  1.55it/s]
4|3760|Loss: 0.8414726257324219:  58%|█████▊    | 3760/6470 [46:40<29:49,  1.51it/s]
4|3761|Loss: 0.5136450529098511:  58%|█████▊    | 3760/6470 [46:41<29:49,  1.51it/s]
4|3761|Loss: 0.5136450529098511:  58%|█████▊    | 3761/6470 [46:41<30:19,  1.49it/s]
4|3762|Loss: 0.895540177822113:  58%|█████▊    | 3761/6470 [46:42<30:19,  1.49it/s] 
4|3762|Loss: 0.895540177822113:  58%|█████▊    | 3762/6470 [46:42<31:46,  1.42it/s]
4|3763|Loss: 0.8077277541160583:  58%|█████▊    | 3762/6470 [46:42<31:46,  1.42it/s]
4|3763|Loss: 0.8077277541160583:  58%|█████▊    | 3763/6470 [46:43<34:24,  1.31it/s]
4|3764|Loss: 1.002244234085083:  58%|█████▊    | 3763/6470 [46:43<34:24,  1.31it/s] 
4|3764|Loss: 1.002244234085083:  58%|█████▊    | 3764/6470 [46:44<35:16,  1.28it/s]
4|3765|Loss: 0.8024752140045166:  58%|█████▊    | 3764/6470 [46:44<35:16,  1.28it/s]
4|3765|Loss: 0.8024752140045166:  58%|█████▊    | 3765/6470 [46:44<35:49,  1.26it/s]
4|3766|Loss: 0.8344050645828247:  58%|█████▊    | 3765/6470 [46:45<35:49,  1.26it/s]
4|3766|Loss: 0.8344050645828247:  58%|█████▊    | 3766/6470 [46:45<31:40,  1.42it/s]
4|3767|Loss: 0.6972100138664246:  58%|█████▊    | 3766/6470 [46:45<31:40,  1.42it/s]
4|3767|Loss: 0.6972100138664246:  58%|█████▊    | 3767/6470 [46:46<32:47,  1.37it/s]
4|3768|Loss: 0.8230909109115601:  58%|█████▊    | 3767/6470 [46:46<32:47,  1.37it/s]
4|3768|Loss: 0.8230909109115601:  58%|█████▊    | 3768/6470 [46:46<32:34,  1.38it/s]
4|3769|Loss: 0.8342837691307068:  58%|█████▊    | 3768/6470 [46:47<32:34,  1.38it/s]
4|3769|Loss: 0.8342837691307068:  58%|█████▊    | 3769/6470 [46:47<31:19,  1.44it/s]
4|3770|Loss: 0.8237167000770569:  58%|█████▊    | 3769/6470 [46:47<31:19,  1.44it/s]
4|3770|Loss: 0.8237167000770569:  58%|█████▊    | 3770/6470 [46:48<31:13,  1.44it/s]
4|3771|Loss: 0.7607661485671997:  58%|█████▊    | 3770/6470 [46:48<31:13,  1.44it/s]
4|3771|Loss: 0.7607661485671997:  58%|█████▊    | 3771/6470 [46:48<31:03,  1.45it/s]
4|3772|Loss: 0.7312141060829163:  58%|█████▊    | 3771/6470 [46:49<31:03,  1.45it/s]
4|3772|Loss: 0.7312141060829163:  58%|█████▊    | 3772/6470 [46:49<32:19,  1.39it/s]
4|3773|Loss: 0.9294295907020569:  58%|█████▊    | 3772/6470 [46:50<32:19,  1.39it/s]
4|3773|Loss: 0.9294295907020569:  58%|█████▊    | 3773/6470 [46:50<35:39,  1.26it/s]
4|3774|Loss: 0.8271973133087158:  58%|█████▊    | 3773/6470 [46:51<35:39,  1.26it/s]
4|3774|Loss: 0.8271973133087158:  58%|█████▊    | 3774/6470 [46:51<37:03,  1.21it/s]
4|3775|Loss: 0.9250896573066711:  58%|█████▊    | 3774/6470 [46:52<37:03,  1.21it/s]
4|3775|Loss: 0.9250896573066711:  58%|█████▊    | 3775/6470 [46:52<37:34,  1.20it/s]
4|3776|Loss: 0.7518339157104492:  58%|█████▊    | 3775/6470 [46:52<37:34,  1.20it/s]
4|3776|Loss: 0.7518339157104492:  58%|█████▊    | 3776/6470 [46:53<38:46,  1.16it/s]
4|3777|Loss: 0.7946662902832031:  58%|█████▊    | 3776/6470 [46:53<38:46,  1.16it/s]
4|3777|Loss: 0.7946662902832031:  58%|█████▊    | 3777/6470 [46:54<39:11,  1.15it/s]
4|3778|Loss: 0.8903780579566956:  58%|█████▊    | 3777/6470 [46:54<39:11,  1.15it/s]
4|3778|Loss: 0.8903780579566956:  58%|█████▊    | 3778/6470 [46:54<36:25,  1.23it/s]
4|3779|Loss: 0.9109821319580078:  58%|█████▊    | 3778/6470 [46:55<36:25,  1.23it/s]
4|3779|Loss: 0.9109821319580078:  58%|█████▊    | 3779/6470 [46:55<35:49,  1.25it/s]
4|3780|Loss: 1.1161248683929443:  58%|█████▊    | 3779/6470 [46:56<35:49,  1.25it/s]
4|3780|Loss: 1.1161248683929443:  58%|█████▊    | 3780/6470 [46:57<42:47,  1.05it/s]
4|3781|Loss: 0.8386085033416748:  58%|█████▊    | 3780/6470 [46:57<42:47,  1.05it/s]
4|3781|Loss: 0.8386085033416748:  58%|█████▊    | 3781/6470 [46:57<41:20,  1.08it/s]
4|3782|Loss: 0.6103474497795105:  58%|█████▊    | 3781/6470 [46:58<41:20,  1.08it/s]
4|3782|Loss: 0.6103474497795105:  58%|█████▊    | 3782/6470 [46:58<38:17,  1.17it/s]
4|3783|Loss: 0.8385561108589172:  58%|█████▊    | 3782/6470 [46:58<38:17,  1.17it/s]
4|3783|Loss: 0.8385561108589172:  58%|█████▊    | 3783/6470 [46:59<36:47,  1.22it/s]
4|3784|Loss: 0.9139142632484436:  58%|█████▊    | 3783/6470 [46:59<36:47,  1.22it/s]
4|3784|Loss: 0.9139142632484436:  58%|█████▊    | 3784/6470 [47:00<37:41,  1.19it/s]
4|3785|Loss: 0.7415633201599121:  58%|█████▊    | 3784/6470 [47:00<37:41,  1.19it/s]
4|3785|Loss: 0.7415633201599121:  59%|█████▊    | 3785/6470 [47:00<36:27,  1.23it/s]
4|3786|Loss: 0.7592563033103943:  59%|█████▊    | 3785/6470 [47:01<36:27,  1.23it/s]
4|3786|Loss: 0.7592563033103943:  59%|█████▊    | 3786/6470 [47:01<35:58,  1.24it/s]
4|3787|Loss: 0.8213813900947571:  59%|█████▊    | 3786/6470 [47:02<35:58,  1.24it/s]
4|3787|Loss: 0.8213813900947571:  59%|█████▊    | 3787/6470 [47:02<31:39,  1.41it/s]
4|3788|Loss: 0.8586527705192566:  59%|█████▊    | 3787/6470 [47:02<31:39,  1.41it/s]
4|3788|Loss: 0.8586527705192566:  59%|█████▊    | 3788/6470 [47:02<31:26,  1.42it/s]
4|3789|Loss: 0.9379834532737732:  59%|█████▊    | 3788/6470 [47:03<31:26,  1.42it/s]
4|3789|Loss: 0.9379834532737732:  59%|█████▊    | 3789/6470 [47:03<34:50,  1.28it/s]
4|3790|Loss: 0.8582542538642883:  59%|█████▊    | 3789/6470 [47:04<34:50,  1.28it/s]
4|3790|Loss: 0.8582542538642883:  59%|█████▊    | 3790/6470 [47:04<35:18,  1.26it/s]
4|3791|Loss: 1.044145941734314:  59%|█████▊    | 3790/6470 [47:05<35:18,  1.26it/s] 
4|3791|Loss: 1.044145941734314:  59%|█████▊    | 3791/6470 [47:05<34:09,  1.31it/s]
4|3792|Loss: 0.7217435836791992:  59%|█████▊    | 3791/6470 [47:05<34:09,  1.31it/s]
4|3792|Loss: 0.7217435836791992:  59%|█████▊    | 3792/6470 [47:05<31:30,  1.42it/s]
4|3793|Loss: 0.9013046026229858:  59%|█████▊    | 3792/6470 [47:06<31:30,  1.42it/s]
4|3793|Loss: 0.9013046026229858:  59%|█████▊    | 3793/6470 [47:06<30:46,  1.45it/s]
4|3794|Loss: 0.8890029191970825:  59%|█████▊    | 3793/6470 [47:07<30:46,  1.45it/s]
4|3794|Loss: 0.8890029191970825:  59%|█████▊    | 3794/6470 [47:07<30:35,  1.46it/s]
4|3795|Loss: 0.8519565463066101:  59%|█████▊    | 3794/6470 [47:07<30:35,  1.46it/s]
4|3795|Loss: 0.8519565463066101:  59%|█████▊    | 3795/6470 [47:08<32:03,  1.39it/s]
4|3796|Loss: 0.8593654036521912:  59%|█████▊    | 3795/6470 [47:08<32:03,  1.39it/s]
4|3796|Loss: 0.8593654036521912:  59%|█████▊    | 3796/6470 [47:08<33:10,  1.34it/s]
4|3797|Loss: 0.6816452741622925:  59%|█████▊    | 3796/6470 [47:09<33:10,  1.34it/s]
4|3797|Loss: 0.6816452741622925:  59%|█████▊    | 3797/6470 [47:09<35:05,  1.27it/s]
4|3798|Loss: 0.9636736512184143:  59%|█████▊    | 3797/6470 [47:10<35:05,  1.27it/s]
4|3798|Loss: 0.9636736512184143:  59%|█████▊    | 3798/6470 [47:10<37:05,  1.20it/s]
4|3799|Loss: 0.7740596532821655:  59%|█████▊    | 3798/6470 [47:11<37:05,  1.20it/s]
4|3799|Loss: 0.7740596532821655:  59%|█████▊    | 3799/6470 [47:11<36:37,  1.22it/s]
4|3800|Loss: 0.8056559562683105:  59%|█████▊    | 3799/6470 [47:11<36:37,  1.22it/s]
4|3800|Loss: 0.8056559562683105:  59%|█████▊    | 3800/6470 [47:12<36:05,  1.23it/s]
4|3801|Loss: 0.8865150213241577:  59%|█████▊    | 3800/6470 [47:12<36:05,  1.23it/s]
4|3801|Loss: 0.8865150213241577:  59%|█████▊    | 3801/6470 [47:12<34:08,  1.30it/s]
4|3802|Loss: 0.9346393346786499:  59%|█████▊    | 3801/6470 [47:13<34:08,  1.30it/s]
4|3802|Loss: 0.9346393346786499:  59%|█████▉    | 3802/6470 [47:13<33:37,  1.32it/s]
4|3803|Loss: 0.7814407348632812:  59%|█████▉    | 3802/6470 [47:14<33:37,  1.32it/s]
4|3803|Loss: 0.7814407348632812:  59%|█████▉    | 3803/6470 [47:14<34:21,  1.29it/s]
4|3804|Loss: 0.8867708444595337:  59%|█████▉    | 3803/6470 [47:14<34:21,  1.29it/s]
4|3804|Loss: 0.8867708444595337:  59%|█████▉    | 3804/6470 [47:15<33:51,  1.31it/s]
4|3805|Loss: 0.7965582609176636:  59%|█████▉    | 3804/6470 [47:15<33:51,  1.31it/s]
4|3805|Loss: 0.7965582609176636:  59%|█████▉    | 3805/6470 [47:16<34:58,  1.27it/s]
4|3806|Loss: 0.8164129853248596:  59%|█████▉    | 3805/6470 [47:16<34:58,  1.27it/s]
4|3806|Loss: 0.8164129853248596:  59%|█████▉    | 3806/6470 [47:17<36:37,  1.21it/s]
4|3807|Loss: 0.872878909111023:  59%|█████▉    | 3806/6470 [47:17<36:37,  1.21it/s] 
4|3807|Loss: 0.872878909111023:  59%|█████▉    | 3807/6470 [47:17<35:33,  1.25it/s]
4|3808|Loss: 0.8479447960853577:  59%|█████▉    | 3807/6470 [47:18<35:33,  1.25it/s]
4|3808|Loss: 0.8479447960853577:  59%|█████▉    | 3808/6470 [47:18<31:48,  1.39it/s]
4|3809|Loss: 0.7140129208564758:  59%|█████▉    | 3808/6470 [47:18<31:48,  1.39it/s]
4|3809|Loss: 0.7140129208564758:  59%|█████▉    | 3809/6470 [47:18<27:29,  1.61it/s]
4|3810|Loss: 0.816286027431488:  59%|█████▉    | 3809/6470 [47:19<27:29,  1.61it/s] 
4|3810|Loss: 0.816286027431488:  59%|█████▉    | 3810/6470 [47:19<27:52,  1.59it/s]
4|3811|Loss: 0.7868040800094604:  59%|█████▉    | 3810/6470 [47:19<27:52,  1.59it/s]
4|3811|Loss: 0.7868040800094604:  59%|█████▉    | 3811/6470 [47:20<28:56,  1.53it/s]
4|3812|Loss: 0.9038827419281006:  59%|█████▉    | 3811/6470 [47:20<28:56,  1.53it/s]
4|3812|Loss: 0.9038827419281006:  59%|█████▉    | 3812/6470 [47:20<28:05,  1.58it/s]
4|3813|Loss: 0.6881397366523743:  59%|█████▉    | 3812/6470 [47:20<28:05,  1.58it/s]
4|3813|Loss: 0.6881397366523743:  59%|█████▉    | 3813/6470 [47:21<27:42,  1.60it/s]
4|3814|Loss: 0.9715453386306763:  59%|█████▉    | 3813/6470 [47:21<27:42,  1.60it/s]
4|3814|Loss: 0.9715453386306763:  59%|█████▉    | 3814/6470 [47:22<30:36,  1.45it/s]
4|3815|Loss: 0.7866983413696289:  59%|█████▉    | 3814/6470 [47:22<30:36,  1.45it/s]
4|3815|Loss: 0.7866983413696289:  59%|█████▉    | 3815/6470 [47:22<30:51,  1.43it/s]
4|3816|Loss: 0.8826981782913208:  59%|█████▉    | 3815/6470 [47:23<30:51,  1.43it/s]
4|3816|Loss: 0.8826981782913208:  59%|█████▉    | 3816/6470 [47:23<30:35,  1.45it/s]
4|3817|Loss: 0.8287513256072998:  59%|█████▉    | 3816/6470 [47:23<30:35,  1.45it/s]
4|3817|Loss: 0.8287513256072998:  59%|█████▉    | 3817/6470 [47:24<31:31,  1.40it/s]
4|3818|Loss: 0.8279087543487549:  59%|█████▉    | 3817/6470 [47:24<31:31,  1.40it/s]
4|3818|Loss: 0.8279087543487549:  59%|█████▉    | 3818/6470 [47:25<34:17,  1.29it/s]
4|3819|Loss: 0.7752609252929688:  59%|█████▉    | 3818/6470 [47:25<34:17,  1.29it/s]
4|3819|Loss: 0.7752609252929688:  59%|█████▉    | 3819/6470 [47:25<34:33,  1.28it/s]
4|3820|Loss: 0.897791862487793:  59%|█████▉    | 3819/6470 [47:26<34:33,  1.28it/s] 
4|3820|Loss: 0.897791862487793:  59%|█████▉    | 3820/6470 [47:26<32:47,  1.35it/s]
4|3821|Loss: 0.8555777072906494:  59%|█████▉    | 3820/6470 [47:26<32:47,  1.35it/s]
4|3821|Loss: 0.8555777072906494:  59%|█████▉    | 3821/6470 [47:27<31:39,  1.39it/s]
4|3822|Loss: 0.7908814549446106:  59%|█████▉    | 3821/6470 [47:27<31:39,  1.39it/s]
4|3822|Loss: 0.7908814549446106:  59%|█████▉    | 3822/6470 [47:28<33:59,  1.30it/s]
4|3823|Loss: 0.7827489972114563:  59%|█████▉    | 3822/6470 [47:28<33:59,  1.30it/s]
4|3823|Loss: 0.7827489972114563:  59%|█████▉    | 3823/6470 [47:28<33:52,  1.30it/s]
4|3824|Loss: 0.8451889753341675:  59%|█████▉    | 3823/6470 [47:29<33:52,  1.30it/s]
4|3824|Loss: 0.8451889753341675:  59%|█████▉    | 3824/6470 [47:29<30:57,  1.42it/s]
4|3825|Loss: 0.8137469291687012:  59%|█████▉    | 3824/6470 [47:29<30:57,  1.42it/s]
4|3825|Loss: 0.8137469291687012:  59%|█████▉    | 3825/6470 [47:30<33:00,  1.34it/s]
4|3826|Loss: 0.8382755517959595:  59%|█████▉    | 3825/6470 [47:30<33:00,  1.34it/s]
4|3826|Loss: 0.8382755517959595:  59%|█████▉    | 3826/6470 [47:31<33:57,  1.30it/s]
4|3827|Loss: 0.957713782787323:  59%|█████▉    | 3826/6470 [47:31<33:57,  1.30it/s] 
4|3827|Loss: 0.957713782787323:  59%|█████▉    | 3827/6470 [47:31<34:20,  1.28it/s]
4|3828|Loss: 0.7547200918197632:  59%|█████▉    | 3827/6470 [47:32<34:20,  1.28it/s]
4|3828|Loss: 0.7547200918197632:  59%|█████▉    | 3828/6470 [47:32<34:26,  1.28it/s]
4|3829|Loss: 0.8018991351127625:  59%|█████▉    | 3828/6470 [47:33<34:26,  1.28it/s]
4|3829|Loss: 0.8018991351127625:  59%|█████▉    | 3829/6470 [47:33<35:43,  1.23it/s]
4|3830|Loss: 0.7315198183059692:  59%|█████▉    | 3829/6470 [47:34<35:43,  1.23it/s]
4|3830|Loss: 0.7315198183059692:  59%|█████▉    | 3830/6470 [47:34<35:41,  1.23it/s]
4|3831|Loss: 0.7428796887397766:  59%|█████▉    | 3830/6470 [47:34<35:41,  1.23it/s]
4|3831|Loss: 0.7428796887397766:  59%|█████▉    | 3831/6470 [47:35<35:25,  1.24it/s]
4|3832|Loss: 0.8363996744155884:  59%|█████▉    | 3831/6470 [47:35<35:25,  1.24it/s]
4|3832|Loss: 0.8363996744155884:  59%|█████▉    | 3832/6470 [47:36<35:59,  1.22it/s]
4|3833|Loss: 0.8601324558258057:  59%|█████▉    | 3832/6470 [47:36<35:59,  1.22it/s]
4|3833|Loss: 0.8601324558258057:  59%|█████▉    | 3833/6470 [47:36<32:30,  1.35it/s]
4|3834|Loss: 0.9844751358032227:  59%|█████▉    | 3833/6470 [47:36<32:30,  1.35it/s]
4|3834|Loss: 0.9844751358032227:  59%|█████▉    | 3834/6470 [47:37<31:14,  1.41it/s]
4|3835|Loss: 0.7417645454406738:  59%|█████▉    | 3834/6470 [47:37<31:14,  1.41it/s]
4|3835|Loss: 0.7417645454406738:  59%|█████▉    | 3835/6470 [47:38<33:37,  1.31it/s]
4|3836|Loss: 0.8022868633270264:  59%|█████▉    | 3835/6470 [47:38<33:37,  1.31it/s]
4|3836|Loss: 0.8022868633270264:  59%|█████▉    | 3836/6470 [47:38<33:12,  1.32it/s]
4|3837|Loss: 0.7525518536567688:  59%|█████▉    | 3836/6470 [47:39<33:12,  1.32it/s]
4|3837|Loss: 0.7525518536567688:  59%|█████▉    | 3837/6470 [47:39<31:12,  1.41it/s]
4|3838|Loss: 0.8566312789916992:  59%|█████▉    | 3837/6470 [47:39<31:12,  1.41it/s]
4|3838|Loss: 0.8566312789916992:  59%|█████▉    | 3838/6470 [47:40<31:54,  1.37it/s]
4|3839|Loss: 0.7283536195755005:  59%|█████▉    | 3838/6470 [47:40<31:54,  1.37it/s]
4|3839|Loss: 0.7283536195755005:  59%|█████▉    | 3839/6470 [47:41<32:44,  1.34it/s]
4|3840|Loss: 0.886292040348053:  59%|█████▉    | 3839/6470 [47:41<32:44,  1.34it/s] 
4|3840|Loss: 0.886292040348053:  59%|█████▉    | 3840/6470 [47:41<33:31,  1.31it/s]
4|3841|Loss: 0.9210246801376343:  59%|█████▉    | 3840/6470 [47:42<33:31,  1.31it/s]
4|3841|Loss: 0.9210246801376343:  59%|█████▉    | 3841/6470 [47:42<35:39,  1.23it/s]
4|3842|Loss: 0.8435428142547607:  59%|█████▉    | 3841/6470 [47:43<35:39,  1.23it/s]
4|3842|Loss: 0.8435428142547607:  59%|█████▉    | 3842/6470 [47:43<32:37,  1.34it/s]
4|3843|Loss: 0.8154456615447998:  59%|█████▉    | 3842/6470 [47:43<32:37,  1.34it/s]
4|3843|Loss: 0.8154456615447998:  59%|█████▉    | 3843/6470 [47:44<33:05,  1.32it/s]
4|3844|Loss: 0.9157504439353943:  59%|█████▉    | 3843/6470 [47:44<33:05,  1.32it/s]
4|3844|Loss: 0.9157504439353943:  59%|█████▉    | 3844/6470 [47:45<35:21,  1.24it/s]
4|3845|Loss: 0.8577173948287964:  59%|█████▉    | 3844/6470 [47:45<35:21,  1.24it/s]
4|3845|Loss: 0.8577173948287964:  59%|█████▉    | 3845/6470 [47:45<35:51,  1.22it/s]
4|3846|Loss: 0.7982394099235535:  59%|█████▉    | 3845/6470 [47:46<35:51,  1.22it/s]
4|3846|Loss: 0.7982394099235535:  59%|█████▉    | 3846/6470 [47:46<35:38,  1.23it/s]
4|3847|Loss: 0.8535578846931458:  59%|█████▉    | 3846/6470 [47:47<35:38,  1.23it/s]
4|3847|Loss: 0.8535578846931458:  59%|█████▉    | 3847/6470 [47:47<36:35,  1.19it/s]
4|3848|Loss: 0.751592218875885:  59%|█████▉    | 3847/6470 [47:48<36:35,  1.19it/s] 
4|3848|Loss: 0.751592218875885:  59%|█████▉    | 3848/6470 [47:48<36:03,  1.21it/s]
4|3849|Loss: 0.7254269123077393:  59%|█████▉    | 3848/6470 [47:48<36:03,  1.21it/s]
4|3849|Loss: 0.7254269123077393:  59%|█████▉    | 3849/6470 [47:49<33:46,  1.29it/s]
4|3850|Loss: 0.7836866974830627:  59%|█████▉    | 3849/6470 [47:49<33:46,  1.29it/s]
4|3850|Loss: 0.7836866974830627:  60%|█████▉    | 3850/6470 [47:49<34:13,  1.28it/s]
4|3851|Loss: 0.833318829536438:  60%|█████▉    | 3850/6470 [47:50<34:13,  1.28it/s] 
4|3851|Loss: 0.833318829536438:  60%|█████▉    | 3851/6470 [47:50<34:04,  1.28it/s]
4|3852|Loss: 0.9652519226074219:  60%|█████▉    | 3851/6470 [47:51<34:04,  1.28it/s]
4|3852|Loss: 0.9652519226074219:  60%|█████▉    | 3852/6470 [47:51<34:23,  1.27it/s]
4|3853|Loss: 0.7579268217086792:  60%|█████▉    | 3852/6470 [47:51<34:23,  1.27it/s]
4|3853|Loss: 0.7579268217086792:  60%|█████▉    | 3853/6470 [47:52<33:19,  1.31it/s]
4|3854|Loss: 0.6578983664512634:  60%|█████▉    | 3853/6470 [47:52<33:19,  1.31it/s]
4|3854|Loss: 0.6578983664512634:  60%|█████▉    | 3854/6470 [47:52<30:39,  1.42it/s]
4|3855|Loss: 0.8456051349639893:  60%|█████▉    | 3854/6470 [47:53<30:39,  1.42it/s]
4|3855|Loss: 0.8456051349639893:  60%|█████▉    | 3855/6470 [47:53<30:08,  1.45it/s]
4|3856|Loss: 1.0584087371826172:  60%|█████▉    | 3855/6470 [47:53<30:08,  1.45it/s]
4|3856|Loss: 1.0584087371826172:  60%|█████▉    | 3856/6470 [47:54<29:10,  1.49it/s]
4|3857|Loss: 0.76109778881073:  60%|█████▉    | 3856/6470 [47:54<29:10,  1.49it/s]  
4|3857|Loss: 0.76109778881073:  60%|█████▉    | 3857/6470 [47:54<30:19,  1.44it/s]
4|3858|Loss: 0.8759708404541016:  60%|█████▉    | 3857/6470 [47:55<30:19,  1.44it/s]
4|3858|Loss: 0.8759708404541016:  60%|█████▉    | 3858/6470 [47:55<31:40,  1.37it/s]
4|3859|Loss: 0.7902349233627319:  60%|█████▉    | 3858/6470 [47:55<31:40,  1.37it/s]
4|3859|Loss: 0.7902349233627319:  60%|█████▉    | 3859/6470 [47:56<30:59,  1.40it/s]
4|3860|Loss: 1.1686220169067383:  60%|█████▉    | 3859/6470 [47:56<30:59,  1.40it/s]
4|3860|Loss: 1.1686220169067383:  60%|█████▉    | 3860/6470 [47:56<31:13,  1.39it/s]
4|3861|Loss: 0.8465636968612671:  60%|█████▉    | 3860/6470 [47:57<31:13,  1.39it/s]
4|3861|Loss: 0.8465636968612671:  60%|█████▉    | 3861/6470 [47:57<32:46,  1.33it/s]
4|3862|Loss: 0.9253987073898315:  60%|█████▉    | 3861/6470 [47:58<32:46,  1.33it/s]
4|3862|Loss: 0.9253987073898315:  60%|█████▉    | 3862/6470 [47:58<33:36,  1.29it/s]
4|3863|Loss: 0.9084299206733704:  60%|█████▉    | 3862/6470 [47:59<33:36,  1.29it/s]
4|3863|Loss: 0.9084299206733704:  60%|█████▉    | 3863/6470 [47:59<35:24,  1.23it/s]
4|3864|Loss: 0.8602613210678101:  60%|█████▉    | 3863/6470 [48:00<35:24,  1.23it/s]
4|3864|Loss: 0.8602613210678101:  60%|█████▉    | 3864/6470 [48:00<35:39,  1.22it/s]
4|3865|Loss: 1.030205488204956:  60%|█████▉    | 3864/6470 [48:00<35:39,  1.22it/s] 
4|3865|Loss: 1.030205488204956:  60%|█████▉    | 3865/6470 [48:01<35:21,  1.23it/s]
4|3866|Loss: 0.6942979693412781:  60%|█████▉    | 3865/6470 [48:01<35:21,  1.23it/s]
4|3866|Loss: 0.6942979693412781:  60%|█████▉    | 3866/6470 [48:01<32:58,  1.32it/s]
4|3867|Loss: 0.9242866039276123:  60%|█████▉    | 3866/6470 [48:02<32:58,  1.32it/s]
4|3867|Loss: 0.9242866039276123:  60%|█████▉    | 3867/6470 [48:02<31:26,  1.38it/s]
4|3868|Loss: 0.8697386384010315:  60%|█████▉    | 3867/6470 [48:02<31:26,  1.38it/s]
4|3868|Loss: 0.8697386384010315:  60%|█████▉    | 3868/6470 [48:03<30:46,  1.41it/s]
4|3869|Loss: 0.8700039982795715:  60%|█████▉    | 3868/6470 [48:03<30:46,  1.41it/s]
4|3869|Loss: 0.8700039982795715:  60%|█████▉    | 3869/6470 [48:03<31:17,  1.39it/s]
4|3870|Loss: 0.9409534335136414:  60%|█████▉    | 3869/6470 [48:04<31:17,  1.39it/s]
4|3870|Loss: 0.9409534335136414:  60%|█████▉    | 3870/6470 [48:04<33:00,  1.31it/s]
4|3871|Loss: 0.8627686500549316:  60%|█████▉    | 3870/6470 [48:05<33:00,  1.31it/s]
4|3871|Loss: 0.8627686500549316:  60%|█████▉    | 3871/6470 [48:05<31:33,  1.37it/s]
4|3872|Loss: 0.9122737050056458:  60%|█████▉    | 3871/6470 [48:05<31:33,  1.37it/s]
4|3872|Loss: 0.9122737050056458:  60%|█████▉    | 3872/6470 [48:05<30:01,  1.44it/s]
4|3873|Loss: 0.8261832594871521:  60%|█████▉    | 3872/6470 [48:06<30:01,  1.44it/s]
4|3873|Loss: 0.8261832594871521:  60%|█████▉    | 3873/6470 [48:06<32:05,  1.35it/s]
4|3874|Loss: 0.7971794009208679:  60%|█████▉    | 3873/6470 [48:07<32:05,  1.35it/s]
4|3874|Loss: 0.7971794009208679:  60%|█████▉    | 3874/6470 [48:07<33:55,  1.28it/s]
4|3875|Loss: 0.6475719809532166:  60%|█████▉    | 3874/6470 [48:08<33:55,  1.28it/s]
4|3875|Loss: 0.6475719809532166:  60%|█████▉    | 3875/6470 [48:08<33:53,  1.28it/s]
4|3876|Loss: 0.814727246761322:  60%|█████▉    | 3875/6470 [48:08<33:53,  1.28it/s] 
4|3876|Loss: 0.814727246761322:  60%|█████▉    | 3876/6470 [48:09<33:29,  1.29it/s]
4|3877|Loss: 0.8235505223274231:  60%|█████▉    | 3876/6470 [48:09<33:29,  1.29it/s]
4|3877|Loss: 0.8235505223274231:  60%|█████▉    | 3877/6470 [48:09<31:32,  1.37it/s]
4|3878|Loss: 0.9231001138687134:  60%|█████▉    | 3877/6470 [48:10<31:32,  1.37it/s]
4|3878|Loss: 0.9231001138687134:  60%|█████▉    | 3878/6470 [48:10<33:27,  1.29it/s]
4|3879|Loss: 0.6867672801017761:  60%|█████▉    | 3878/6470 [48:11<33:27,  1.29it/s]
4|3879|Loss: 0.6867672801017761:  60%|█████▉    | 3879/6470 [48:11<32:56,  1.31it/s]
4|3880|Loss: 0.7491616010665894:  60%|█████▉    | 3879/6470 [48:11<32:56,  1.31it/s]
4|3880|Loss: 0.7491616010665894:  60%|█████▉    | 3880/6470 [48:12<34:13,  1.26it/s]
4|3881|Loss: 0.7979121804237366:  60%|█████▉    | 3880/6470 [48:12<34:13,  1.26it/s]
4|3881|Loss: 0.7979121804237366:  60%|█████▉    | 3881/6470 [48:12<31:48,  1.36it/s]
4|3882|Loss: 0.8025699257850647:  60%|█████▉    | 3881/6470 [48:13<31:48,  1.36it/s]
4|3882|Loss: 0.8025699257850647:  60%|██████    | 3882/6470 [48:13<29:43,  1.45it/s]
4|3883|Loss: 0.907240092754364:  60%|██████    | 3882/6470 [48:13<29:43,  1.45it/s] 
4|3883|Loss: 0.907240092754364:  60%|██████    | 3883/6470 [48:14<30:12,  1.43it/s]
4|3884|Loss: 0.7516575455665588:  60%|██████    | 3883/6470 [48:14<30:12,  1.43it/s]
4|3884|Loss: 0.7516575455665588:  60%|██████    | 3884/6470 [48:15<31:20,  1.37it/s]
4|3885|Loss: 0.8601018190383911:  60%|██████    | 3884/6470 [48:15<31:20,  1.37it/s]
4|3885|Loss: 0.8601018190383911:  60%|██████    | 3885/6470 [48:15<29:58,  1.44it/s]
4|3886|Loss: 0.8566663265228271:  60%|██████    | 3885/6470 [48:16<29:58,  1.44it/s]
4|3886|Loss: 0.8566663265228271:  60%|██████    | 3886/6470 [48:16<30:10,  1.43it/s]
4|3887|Loss: 0.663658857345581:  60%|██████    | 3886/6470 [48:16<30:10,  1.43it/s] 
4|3887|Loss: 0.663658857345581:  60%|██████    | 3887/6470 [48:17<30:14,  1.42it/s]
4|3888|Loss: 0.8642504811286926:  60%|██████    | 3887/6470 [48:17<30:14,  1.42it/s]
4|3888|Loss: 0.8642504811286926:  60%|██████    | 3888/6470 [48:17<30:52,  1.39it/s]
4|3889|Loss: 0.7363739609718323:  60%|██████    | 3888/6470 [48:18<30:52,  1.39it/s]
4|3889|Loss: 0.7363739609718323:  60%|██████    | 3889/6470 [48:18<32:43,  1.31it/s]
4|3890|Loss: 0.7145297527313232:  60%|██████    | 3889/6470 [48:19<32:43,  1.31it/s]
4|3890|Loss: 0.7145297527313232:  60%|██████    | 3890/6470 [48:19<33:16,  1.29it/s]
4|3891|Loss: 0.8128418922424316:  60%|██████    | 3890/6470 [48:19<33:16,  1.29it/s]
4|3891|Loss: 0.8128418922424316:  60%|██████    | 3891/6470 [48:20<33:08,  1.30it/s]
4|3892|Loss: 0.8335921764373779:  60%|██████    | 3891/6470 [48:20<33:08,  1.30it/s]
4|3892|Loss: 0.8335921764373779:  60%|██████    | 3892/6470 [48:21<34:34,  1.24it/s]
4|3893|Loss: 0.7578364014625549:  60%|██████    | 3892/6470 [48:21<34:34,  1.24it/s]
4|3893|Loss: 0.7578364014625549:  60%|██████    | 3893/6470 [48:21<32:58,  1.30it/s]
4|3894|Loss: 0.7724170684814453:  60%|██████    | 3893/6470 [48:22<32:58,  1.30it/s]
4|3894|Loss: 0.7724170684814453:  60%|██████    | 3894/6470 [48:22<34:00,  1.26it/s]
4|3895|Loss: 0.7512745261192322:  60%|██████    | 3894/6470 [48:23<34:00,  1.26it/s]
4|3895|Loss: 0.7512745261192322:  60%|██████    | 3895/6470 [48:23<34:49,  1.23it/s]
4|3896|Loss: 0.7104507684707642:  60%|██████    | 3895/6470 [48:23<34:49,  1.23it/s]
4|3896|Loss: 0.7104507684707642:  60%|██████    | 3896/6470 [48:24<31:57,  1.34it/s]
4|3897|Loss: 0.6782273650169373:  60%|██████    | 3896/6470 [48:24<31:57,  1.34it/s]
4|3897|Loss: 0.6782273650169373:  60%|██████    | 3897/6470 [48:25<33:26,  1.28it/s]
4|3898|Loss: 0.8086217045783997:  60%|██████    | 3897/6470 [48:25<33:26,  1.28it/s]
4|3898|Loss: 0.8086217045783997:  60%|██████    | 3898/6470 [48:25<34:18,  1.25it/s]
4|3899|Loss: 0.9134497046470642:  60%|██████    | 3898/6470 [48:26<34:18,  1.25it/s]
4|3899|Loss: 0.9134497046470642:  60%|██████    | 3899/6470 [48:26<34:22,  1.25it/s]
4|3900|Loss: 0.7944614291191101:  60%|██████    | 3899/6470 [48:27<34:22,  1.25it/s]
4|3900|Loss: 0.7944614291191101:  60%|██████    | 3900/6470 [48:27<35:33,  1.20it/s]
4|3901|Loss: 0.6016258597373962:  60%|██████    | 3900/6470 [48:27<35:33,  1.20it/s]
4|3901|Loss: 0.6016258597373962:  60%|██████    | 3901/6470 [48:28<30:55,  1.38it/s]
4|3902|Loss: 0.7688044905662537:  60%|██████    | 3901/6470 [48:28<30:55,  1.38it/s]
4|3902|Loss: 0.7688044905662537:  60%|██████    | 3902/6470 [48:28<31:14,  1.37it/s]
4|3903|Loss: 0.9944766163825989:  60%|██████    | 3902/6470 [48:29<31:14,  1.37it/s]
4|3903|Loss: 0.9944766163825989:  60%|██████    | 3903/6470 [48:29<32:30,  1.32it/s]
4|3904|Loss: 0.830716609954834:  60%|██████    | 3903/6470 [48:30<32:30,  1.32it/s] 
4|3904|Loss: 0.830716609954834:  60%|██████    | 3904/6470 [48:30<34:52,  1.23it/s]
4|3905|Loss: 0.8630765080451965:  60%|██████    | 3904/6470 [48:31<34:52,  1.23it/s]
4|3905|Loss: 0.8630765080451965:  60%|██████    | 3905/6470 [48:31<35:09,  1.22it/s]
4|3906|Loss: 0.7735833525657654:  60%|██████    | 3905/6470 [48:31<35:09,  1.22it/s]
4|3906|Loss: 0.7735833525657654:  60%|██████    | 3906/6470 [48:32<33:22,  1.28it/s]
4|3907|Loss: 0.8306599259376526:  60%|██████    | 3906/6470 [48:32<33:22,  1.28it/s]
4|3907|Loss: 0.8306599259376526:  60%|██████    | 3907/6470 [48:32<32:41,  1.31it/s]
4|3908|Loss: 0.753873348236084:  60%|██████    | 3907/6470 [48:33<32:41,  1.31it/s] 
4|3908|Loss: 0.753873348236084:  60%|██████    | 3908/6470 [48:33<34:15,  1.25it/s]
4|3909|Loss: 0.6262192726135254:  60%|██████    | 3908/6470 [48:34<34:15,  1.25it/s]
4|3909|Loss: 0.6262192726135254:  60%|██████    | 3909/6470 [48:34<33:33,  1.27it/s]
4|3910|Loss: 0.8060412406921387:  60%|██████    | 3909/6470 [48:34<33:33,  1.27it/s]
4|3910|Loss: 0.8060412406921387:  60%|██████    | 3910/6470 [48:35<33:03,  1.29it/s]
4|3911|Loss: 0.9065669178962708:  60%|██████    | 3910/6470 [48:35<33:03,  1.29it/s]
4|3911|Loss: 0.9065669178962708:  60%|██████    | 3911/6470 [48:36<33:51,  1.26it/s]
4|3912|Loss: 0.8875399827957153:  60%|██████    | 3911/6470 [48:36<33:51,  1.26it/s]
4|3912|Loss: 0.8875399827957153:  60%|██████    | 3912/6470 [48:36<34:45,  1.23it/s]
4|3913|Loss: 0.795650064945221:  60%|██████    | 3912/6470 [48:37<34:45,  1.23it/s] 
4|3913|Loss: 0.795650064945221:  60%|██████    | 3913/6470 [48:37<34:43,  1.23it/s]
4|3914|Loss: 0.8130877614021301:  60%|██████    | 3913/6470 [48:38<34:43,  1.23it/s]
4|3914|Loss: 0.8130877614021301:  60%|██████    | 3914/6470 [48:38<35:54,  1.19it/s]
4|3915|Loss: 0.8990296721458435:  60%|██████    | 3914/6470 [48:39<35:54,  1.19it/s]
4|3915|Loss: 0.8990296721458435:  61%|██████    | 3915/6470 [48:39<33:58,  1.25it/s]
4|3916|Loss: 0.9187295436859131:  61%|██████    | 3915/6470 [48:39<33:58,  1.25it/s]
4|3916|Loss: 0.9187295436859131:  61%|██████    | 3916/6470 [48:39<30:59,  1.37it/s]
4|3917|Loss: 0.7894142270088196:  61%|██████    | 3916/6470 [48:40<30:59,  1.37it/s]
4|3917|Loss: 0.7894142270088196:  61%|██████    | 3917/6470 [48:40<32:14,  1.32it/s]
4|3918|Loss: 0.80924391746521:  61%|██████    | 3917/6470 [48:41<32:14,  1.32it/s]  
4|3918|Loss: 0.80924391746521:  61%|██████    | 3918/6470 [48:41<33:27,  1.27it/s]
4|3919|Loss: 0.8403903245925903:  61%|██████    | 3918/6470 [48:41<33:27,  1.27it/s]
4|3919|Loss: 0.8403903245925903:  61%|██████    | 3919/6470 [48:42<32:03,  1.33it/s]
4|3920|Loss: 0.9224030375480652:  61%|██████    | 3919/6470 [48:42<32:03,  1.33it/s]
4|3920|Loss: 0.9224030375480652:  61%|██████    | 3920/6470 [48:43<33:48,  1.26it/s]
4|3921|Loss: 0.8763935565948486:  61%|██████    | 3920/6470 [48:43<33:48,  1.26it/s]
4|3921|Loss: 0.8763935565948486:  61%|██████    | 3921/6470 [48:43<31:58,  1.33it/s]
4|3922|Loss: 0.8304625749588013:  61%|██████    | 3921/6470 [48:44<31:58,  1.33it/s]
4|3922|Loss: 0.8304625749588013:  61%|██████    | 3922/6470 [48:44<30:51,  1.38it/s]
4|3923|Loss: 0.8626534342765808:  61%|██████    | 3922/6470 [48:44<30:51,  1.38it/s]
4|3923|Loss: 0.8626534342765808:  61%|██████    | 3923/6470 [48:45<31:35,  1.34it/s]
4|3924|Loss: 0.8645190596580505:  61%|██████    | 3923/6470 [48:45<31:35,  1.34it/s]
4|3924|Loss: 0.8645190596580505:  61%|██████    | 3924/6470 [48:46<33:26,  1.27it/s]
4|3925|Loss: 0.8450931906700134:  61%|██████    | 3924/6470 [48:46<33:26,  1.27it/s]
4|3925|Loss: 0.8450931906700134:  61%|██████    | 3925/6470 [48:46<32:28,  1.31it/s]
4|3926|Loss: 0.811979353427887:  61%|██████    | 3925/6470 [48:47<32:28,  1.31it/s] 
4|3926|Loss: 0.811979353427887:  61%|██████    | 3926/6470 [48:47<30:03,  1.41it/s]
4|3927|Loss: 0.8612034916877747:  61%|██████    | 3926/6470 [48:47<30:03,  1.41it/s]
4|3927|Loss: 0.8612034916877747:  61%|██████    | 3927/6470 [48:48<30:15,  1.40it/s]
4|3928|Loss: 0.8481816053390503:  61%|██████    | 3927/6470 [48:48<30:15,  1.40it/s]
4|3928|Loss: 0.8481816053390503:  61%|██████    | 3928/6470 [48:48<29:56,  1.41it/s]
4|3929|Loss: 0.6388954520225525:  61%|██████    | 3928/6470 [48:49<29:56,  1.41it/s]
4|3929|Loss: 0.6388954520225525:  61%|██████    | 3929/6470 [48:49<33:35,  1.26it/s]
4|3930|Loss: 1.0043336153030396:  61%|██████    | 3929/6470 [48:50<33:35,  1.26it/s]
4|3930|Loss: 1.0043336153030396:  61%|██████    | 3930/6470 [48:50<34:25,  1.23it/s]
4|3931|Loss: 0.8931585550308228:  61%|██████    | 3930/6470 [48:51<34:25,  1.23it/s]
4|3931|Loss: 0.8931585550308228:  61%|██████    | 3931/6470 [48:51<32:27,  1.30it/s]
4|3932|Loss: 0.9489890336990356:  61%|██████    | 3931/6470 [48:51<32:27,  1.30it/s]
4|3932|Loss: 0.9489890336990356:  61%|██████    | 3932/6470 [48:52<31:16,  1.35it/s]
4|3933|Loss: 0.8665762543678284:  61%|██████    | 3932/6470 [48:52<31:16,  1.35it/s]
4|3933|Loss: 0.8665762543678284:  61%|██████    | 3933/6470 [48:52<30:39,  1.38it/s]
4|3934|Loss: 0.8546391129493713:  61%|██████    | 3933/6470 [48:53<30:39,  1.38it/s]
4|3934|Loss: 0.8546391129493713:  61%|██████    | 3934/6470 [48:53<30:10,  1.40it/s]
4|3935|Loss: 0.994438648223877:  61%|██████    | 3934/6470 [48:53<30:10,  1.40it/s] 
4|3935|Loss: 0.994438648223877:  61%|██████    | 3935/6470 [48:54<31:03,  1.36it/s]
4|3936|Loss: 0.8785408139228821:  61%|██████    | 3935/6470 [48:54<31:03,  1.36it/s]
4|3936|Loss: 0.8785408139228821:  61%|██████    | 3936/6470 [48:55<32:31,  1.30it/s]
4|3937|Loss: 0.8375274538993835:  61%|██████    | 3936/6470 [48:55<32:31,  1.30it/s]
4|3937|Loss: 0.8375274538993835:  61%|██████    | 3937/6470 [48:55<33:47,  1.25it/s]
4|3938|Loss: 0.6674322485923767:  61%|██████    | 3937/6470 [48:56<33:47,  1.25it/s]
4|3938|Loss: 0.6674322485923767:  61%|██████    | 3938/6470 [48:56<33:38,  1.25it/s]
4|3939|Loss: 0.872028112411499:  61%|██████    | 3938/6470 [48:57<33:38,  1.25it/s] 
4|3939|Loss: 0.872028112411499:  61%|██████    | 3939/6470 [48:57<33:24,  1.26it/s]
4|3940|Loss: 0.7768890261650085:  61%|██████    | 3939/6470 [48:57<33:24,  1.26it/s]
4|3940|Loss: 0.7768890261650085:  61%|██████    | 3940/6470 [48:58<32:54,  1.28it/s]
4|3941|Loss: 0.8094987273216248:  61%|██████    | 3940/6470 [48:58<32:54,  1.28it/s]
4|3941|Loss: 0.8094987273216248:  61%|██████    | 3941/6470 [48:58<31:59,  1.32it/s]
4|3942|Loss: 0.7675095796585083:  61%|██████    | 3941/6470 [48:59<31:59,  1.32it/s]
4|3942|Loss: 0.7675095796585083:  61%|██████    | 3942/6470 [48:59<31:54,  1.32it/s]
4|3943|Loss: 0.8093645572662354:  61%|██████    | 3942/6470 [49:00<31:54,  1.32it/s]
4|3943|Loss: 0.8093645572662354:  61%|██████    | 3943/6470 [49:00<32:29,  1.30it/s]
4|3944|Loss: 1.0266340970993042:  61%|██████    | 3943/6470 [49:00<32:29,  1.30it/s]
4|3944|Loss: 1.0266340970993042:  61%|██████    | 3944/6470 [49:01<32:55,  1.28it/s]
4|3945|Loss: 0.9554904699325562:  61%|██████    | 3944/6470 [49:01<32:55,  1.28it/s]
4|3945|Loss: 0.9554904699325562:  61%|██████    | 3945/6470 [49:02<33:22,  1.26it/s]
4|3946|Loss: 0.6996883153915405:  61%|██████    | 3945/6470 [49:02<33:22,  1.26it/s]
4|3946|Loss: 0.6996883153915405:  61%|██████    | 3946/6470 [49:02<33:38,  1.25it/s]
4|3947|Loss: 0.7487316131591797:  61%|██████    | 3946/6470 [49:03<33:38,  1.25it/s]
4|3947|Loss: 0.7487316131591797:  61%|██████    | 3947/6470 [49:03<35:26,  1.19it/s]
4|3948|Loss: 0.9106476902961731:  61%|██████    | 3947/6470 [49:04<35:26,  1.19it/s]
4|3948|Loss: 0.9106476902961731:  61%|██████    | 3948/6470 [49:04<36:32,  1.15it/s]
4|3949|Loss: 0.7569486498832703:  61%|██████    | 3948/6470 [49:05<36:32,  1.15it/s]
4|3949|Loss: 0.7569486498832703:  61%|██████    | 3949/6470 [49:05<34:58,  1.20it/s]
4|3950|Loss: 0.6657944917678833:  61%|██████    | 3949/6470 [49:05<34:58,  1.20it/s]
4|3950|Loss: 0.6657944917678833:  61%|██████    | 3950/6470 [49:06<33:52,  1.24it/s]
4|3951|Loss: 0.7429534792900085:  61%|██████    | 3950/6470 [49:06<33:52,  1.24it/s]
4|3951|Loss: 0.7429534792900085:  61%|██████    | 3951/6470 [49:06<31:38,  1.33it/s]
4|3952|Loss: 0.8154401779174805:  61%|██████    | 3951/6470 [49:07<31:38,  1.33it/s]
4|3952|Loss: 0.8154401779174805:  61%|██████    | 3952/6470 [49:07<29:52,  1.40it/s]
4|3953|Loss: 0.8890818357467651:  61%|██████    | 3952/6470 [49:07<29:52,  1.40it/s]
4|3953|Loss: 0.8890818357467651:  61%|██████    | 3953/6470 [49:08<28:43,  1.46it/s]
4|3954|Loss: 0.6938456892967224:  61%|██████    | 3953/6470 [49:08<28:43,  1.46it/s]
4|3954|Loss: 0.6938456892967224:  61%|██████    | 3954/6470 [49:09<31:12,  1.34it/s]
4|3955|Loss: 0.8791188597679138:  61%|██████    | 3954/6470 [49:09<31:12,  1.34it/s]
4|3955|Loss: 0.8791188597679138:  61%|██████    | 3955/6470 [49:09<30:28,  1.38it/s]
4|3956|Loss: 0.7584218978881836:  61%|██████    | 3955/6470 [49:10<30:28,  1.38it/s]
4|3956|Loss: 0.7584218978881836:  61%|██████    | 3956/6470 [49:10<31:29,  1.33it/s]
4|3957|Loss: 0.7996145486831665:  61%|██████    | 3956/6470 [49:10<31:29,  1.33it/s]
4|3957|Loss: 0.7996145486831665:  61%|██████    | 3957/6470 [49:11<29:39,  1.41it/s]
4|3958|Loss: 0.7382145524024963:  61%|██████    | 3957/6470 [49:11<29:39,  1.41it/s]
4|3958|Loss: 0.7382145524024963:  61%|██████    | 3958/6470 [49:11<29:34,  1.42it/s]
4|3959|Loss: 0.9917647242546082:  61%|██████    | 3958/6470 [49:12<29:34,  1.42it/s]
4|3959|Loss: 0.9917647242546082:  61%|██████    | 3959/6470 [49:12<31:08,  1.34it/s]
4|3960|Loss: 0.7761125564575195:  61%|██████    | 3959/6470 [49:13<31:08,  1.34it/s]
4|3960|Loss: 0.7761125564575195:  61%|██████    | 3960/6470 [49:13<33:40,  1.24it/s]
4|3961|Loss: 0.6804390549659729:  61%|██████    | 3960/6470 [49:14<33:40,  1.24it/s]
4|3961|Loss: 0.6804390549659729:  61%|██████    | 3961/6470 [49:14<33:43,  1.24it/s]
4|3962|Loss: 0.776342511177063:  61%|██████    | 3961/6470 [49:14<33:43,  1.24it/s] 
4|3962|Loss: 0.776342511177063:  61%|██████    | 3962/6470 [49:15<33:07,  1.26it/s]
4|3963|Loss: 0.8952555656433105:  61%|██████    | 3962/6470 [49:15<33:07,  1.26it/s]
4|3963|Loss: 0.8952555656433105:  61%|██████▏   | 3963/6470 [49:15<30:07,  1.39it/s]
4|3964|Loss: 0.8380594253540039:  61%|██████▏   | 3963/6470 [49:16<30:07,  1.39it/s]
4|3964|Loss: 0.8380594253540039:  61%|██████▏   | 3964/6470 [49:16<26:09,  1.60it/s]
4|3965|Loss: 0.8110452890396118:  61%|██████▏   | 3964/6470 [49:16<26:09,  1.60it/s]
4|3965|Loss: 0.8110452890396118:  61%|██████▏   | 3965/6470 [49:16<25:38,  1.63it/s]
4|3966|Loss: 0.777938961982727:  61%|██████▏   | 3965/6470 [49:17<25:38,  1.63it/s] 
4|3966|Loss: 0.777938961982727:  61%|██████▏   | 3966/6470 [49:17<28:40,  1.46it/s]
4|3967|Loss: 0.9177486896514893:  61%|██████▏   | 3966/6470 [49:18<28:40,  1.46it/s]
4|3967|Loss: 0.9177486896514893:  61%|██████▏   | 3967/6470 [49:18<29:56,  1.39it/s]
4|3968|Loss: 0.8648761510848999:  61%|██████▏   | 3967/6470 [49:18<29:56,  1.39it/s]
4|3968|Loss: 0.8648761510848999:  61%|██████▏   | 3968/6470 [49:19<29:30,  1.41it/s]
4|3969|Loss: 0.6811692118644714:  61%|██████▏   | 3968/6470 [49:19<29:30,  1.41it/s]
4|3969|Loss: 0.6811692118644714:  61%|██████▏   | 3969/6470 [49:19<25:58,  1.61it/s]
4|3970|Loss: 0.8433423638343811:  61%|██████▏   | 3969/6470 [49:20<25:58,  1.61it/s]
4|3970|Loss: 0.8433423638343811:  61%|██████▏   | 3970/6470 [49:20<32:02,  1.30it/s]
4|3971|Loss: 0.8174534440040588:  61%|██████▏   | 3970/6470 [49:21<32:02,  1.30it/s]
4|3971|Loss: 0.8174534440040588:  61%|██████▏   | 3971/6470 [49:21<31:57,  1.30it/s]
4|3972|Loss: 0.8263065218925476:  61%|██████▏   | 3971/6470 [49:21<31:57,  1.30it/s]
4|3972|Loss: 0.8263065218925476:  61%|██████▏   | 3972/6470 [49:21<29:58,  1.39it/s]
4|3973|Loss: 0.6956920027732849:  61%|██████▏   | 3972/6470 [49:22<29:58,  1.39it/s]
4|3973|Loss: 0.6956920027732849:  61%|██████▏   | 3973/6470 [49:22<30:26,  1.37it/s]
4|3974|Loss: 0.6601085662841797:  61%|██████▏   | 3973/6470 [49:23<30:26,  1.37it/s]
4|3974|Loss: 0.6601085662841797:  61%|██████▏   | 3974/6470 [49:23<29:26,  1.41it/s]
4|3975|Loss: 0.8380298614501953:  61%|██████▏   | 3974/6470 [49:23<29:26,  1.41it/s]
4|3975|Loss: 0.8380298614501953:  61%|██████▏   | 3975/6470 [49:24<28:34,  1.46it/s]
4|3976|Loss: 0.8160362839698792:  61%|██████▏   | 3975/6470 [49:24<28:34,  1.46it/s]
4|3976|Loss: 0.8160362839698792:  61%|██████▏   | 3976/6470 [49:24<29:39,  1.40it/s]
4|3977|Loss: 0.9002302885055542:  61%|██████▏   | 3976/6470 [49:25<29:39,  1.40it/s]
4|3977|Loss: 0.9002302885055542:  61%|██████▏   | 3977/6470 [49:25<30:23,  1.37it/s]
4|3978|Loss: 0.7223097085952759:  61%|██████▏   | 3977/6470 [49:25<30:23,  1.37it/s]
4|3978|Loss: 0.7223097085952759:  61%|██████▏   | 3978/6470 [49:26<30:36,  1.36it/s]
4|3979|Loss: 0.8027310371398926:  61%|██████▏   | 3978/6470 [49:26<30:36,  1.36it/s]
4|3979|Loss: 0.8027310371398926:  61%|██████▏   | 3979/6470 [49:27<30:43,  1.35it/s]
4|3980|Loss: 0.7486132979393005:  61%|██████▏   | 3979/6470 [49:27<30:43,  1.35it/s]
4|3980|Loss: 0.7486132979393005:  62%|██████▏   | 3980/6470 [49:27<28:00,  1.48it/s]
4|3981|Loss: 0.7099074125289917:  62%|██████▏   | 3980/6470 [49:27<28:00,  1.48it/s]
4|3981|Loss: 0.7099074125289917:  62%|██████▏   | 3981/6470 [49:28<28:43,  1.44it/s]
4|3982|Loss: 0.7679159045219421:  62%|██████▏   | 3981/6470 [49:28<28:43,  1.44it/s]
4|3982|Loss: 0.7679159045219421:  62%|██████▏   | 3982/6470 [49:28<28:05,  1.48it/s]
4|3983|Loss: 0.9610597491264343:  62%|██████▏   | 3982/6470 [49:29<28:05,  1.48it/s]
4|3983|Loss: 0.9610597491264343:  62%|██████▏   | 3983/6470 [49:29<30:08,  1.37it/s]
4|3984|Loss: 0.9098842740058899:  62%|██████▏   | 3983/6470 [49:30<30:08,  1.37it/s]
4|3984|Loss: 0.9098842740058899:  62%|██████▏   | 3984/6470 [49:30<29:26,  1.41it/s]
4|3985|Loss: 0.6538704037666321:  62%|██████▏   | 3984/6470 [49:30<29:26,  1.41it/s]
4|3985|Loss: 0.6538704037666321:  62%|██████▏   | 3985/6470 [49:31<27:11,  1.52it/s]
4|3986|Loss: 0.7065908312797546:  62%|██████▏   | 3985/6470 [49:31<27:11,  1.52it/s]
4|3986|Loss: 0.7065908312797546:  62%|██████▏   | 3986/6470 [49:31<26:46,  1.55it/s]
4|3987|Loss: 0.9331084489822388:  62%|██████▏   | 3986/6470 [49:32<26:46,  1.55it/s]
4|3987|Loss: 0.9331084489822388:  62%|██████▏   | 3987/6470 [49:32<28:30,  1.45it/s]
4|3988|Loss: 0.9745581150054932:  62%|██████▏   | 3987/6470 [49:32<28:30,  1.45it/s]
4|3988|Loss: 0.9745581150054932:  62%|██████▏   | 3988/6470 [49:33<31:06,  1.33it/s]
4|3989|Loss: 0.9608747363090515:  62%|██████▏   | 3988/6470 [49:33<31:06,  1.33it/s]
4|3989|Loss: 0.9608747363090515:  62%|██████▏   | 3989/6470 [49:34<30:54,  1.34it/s]
4|3990|Loss: 0.9002851247787476:  62%|██████▏   | 3989/6470 [49:34<30:54,  1.34it/s]
4|3990|Loss: 0.9002851247787476:  62%|██████▏   | 3990/6470 [49:34<30:15,  1.37it/s]
4|3991|Loss: 0.9012603163719177:  62%|██████▏   | 3990/6470 [49:35<30:15,  1.37it/s]
4|3991|Loss: 0.9012603163719177:  62%|██████▏   | 3991/6470 [49:35<29:33,  1.40it/s]
4|3992|Loss: 0.8094287514686584:  62%|██████▏   | 3991/6470 [49:35<29:33,  1.40it/s]
4|3992|Loss: 0.8094287514686584:  62%|██████▏   | 3992/6470 [49:36<30:08,  1.37it/s]
4|3993|Loss: 1.1021393537521362:  62%|██████▏   | 3992/6470 [49:36<30:08,  1.37it/s]
4|3993|Loss: 1.1021393537521362:  62%|██████▏   | 3993/6470 [49:36<29:55,  1.38it/s]
4|3994|Loss: 0.7063601016998291:  62%|██████▏   | 3993/6470 [49:37<29:55,  1.38it/s]
4|3994|Loss: 0.7063601016998291:  62%|██████▏   | 3994/6470 [49:37<30:33,  1.35it/s]
4|3995|Loss: 0.8571171760559082:  62%|██████▏   | 3994/6470 [49:38<30:33,  1.35it/s]
4|3995|Loss: 0.8571171760559082:  62%|██████▏   | 3995/6470 [49:38<32:01,  1.29it/s]
4|3996|Loss: 0.8570179343223572:  62%|██████▏   | 3995/6470 [49:38<32:01,  1.29it/s]
4|3996|Loss: 0.8570179343223572:  62%|██████▏   | 3996/6470 [49:39<31:05,  1.33it/s]
4|3997|Loss: 0.7584513425827026:  62%|██████▏   | 3996/6470 [49:39<31:05,  1.33it/s]
4|3997|Loss: 0.7584513425827026:  62%|██████▏   | 3997/6470 [49:39<28:00,  1.47it/s]
4|3998|Loss: 0.9041078686714172:  62%|██████▏   | 3997/6470 [49:40<28:00,  1.47it/s]
4|3998|Loss: 0.9041078686714172:  62%|██████▏   | 3998/6470 [49:40<29:50,  1.38it/s]
4|3999|Loss: 0.8425946235656738:  62%|██████▏   | 3998/6470 [49:41<29:50,  1.38it/s]
4|3999|Loss: 0.8425946235656738:  62%|██████▏   | 3999/6470 [49:41<31:03,  1.33it/s]
4|4000|Loss: 0.8588628172874451:  62%|██████▏   | 3999/6470 [49:41<31:03,  1.33it/s]
4|4000|Loss: 0.8588628172874451:  62%|██████▏   | 4000/6470 [49:42<30:29,  1.35it/s]
4|4001|Loss: 0.8170590996742249:  62%|██████▏   | 4000/6470 [49:42<30:29,  1.35it/s]
4|4001|Loss: 0.8170590996742249:  62%|██████▏   | 4001/6470 [49:42<29:33,  1.39it/s]
4|4002|Loss: 0.7433931231498718:  62%|██████▏   | 4001/6470 [49:43<29:33,  1.39it/s]
4|4002|Loss: 0.7433931231498718:  62%|██████▏   | 4002/6470 [49:43<29:17,  1.40it/s]
4|4003|Loss: 0.6763160824775696:  62%|██████▏   | 4002/6470 [49:43<29:17,  1.40it/s]
4|4003|Loss: 0.6763160824775696:  62%|██████▏   | 4003/6470 [49:44<28:56,  1.42it/s]
4|4004|Loss: 0.7479579448699951:  62%|██████▏   | 4003/6470 [49:44<28:56,  1.42it/s]
4|4004|Loss: 0.7479579448699951:  62%|██████▏   | 4004/6470 [49:44<28:51,  1.42it/s]
4|4005|Loss: 0.7805713415145874:  62%|██████▏   | 4004/6470 [49:45<28:51,  1.42it/s]
4|4005|Loss: 0.7805713415145874:  62%|██████▏   | 4005/6470 [49:45<28:32,  1.44it/s]
4|4006|Loss: 0.8515441417694092:  62%|██████▏   | 4005/6470 [49:46<28:32,  1.44it/s]
4|4006|Loss: 0.8515441417694092:  62%|██████▏   | 4006/6470 [49:46<30:43,  1.34it/s]
4|4007|Loss: 0.829486072063446:  62%|██████▏   | 4006/6470 [49:46<30:43,  1.34it/s] 
4|4007|Loss: 0.829486072063446:  62%|██████▏   | 4007/6470 [49:47<30:20,  1.35it/s]
4|4008|Loss: 0.7606650590896606:  62%|██████▏   | 4007/6470 [49:47<30:20,  1.35it/s]
4|4008|Loss: 0.7606650590896606:  62%|██████▏   | 4008/6470 [49:47<28:22,  1.45it/s]
4|4009|Loss: 0.9251487851142883:  62%|██████▏   | 4008/6470 [49:48<28:22,  1.45it/s]
4|4009|Loss: 0.9251487851142883:  62%|██████▏   | 4009/6470 [49:48<28:58,  1.42it/s]
4|4010|Loss: 0.7540522813796997:  62%|██████▏   | 4009/6470 [49:48<28:58,  1.42it/s]
4|4010|Loss: 0.7540522813796997:  62%|██████▏   | 4010/6470 [49:48<25:54,  1.58it/s]
4|4011|Loss: 0.7113721370697021:  62%|██████▏   | 4010/6470 [49:49<25:54,  1.58it/s]
4|4011|Loss: 0.7113721370697021:  62%|██████▏   | 4011/6470 [49:49<25:09,  1.63it/s]
4|4012|Loss: 0.8047438263893127:  62%|██████▏   | 4011/6470 [49:49<25:09,  1.63it/s]
4|4012|Loss: 0.8047438263893127:  62%|██████▏   | 4012/6470 [49:50<25:43,  1.59it/s]
4|4013|Loss: 0.6952131986618042:  62%|██████▏   | 4012/6470 [49:50<25:43,  1.59it/s]
4|4013|Loss: 0.6952131986618042:  62%|██████▏   | 4013/6470 [49:50<26:09,  1.57it/s]
4|4014|Loss: 0.7136818170547485:  62%|██████▏   | 4013/6470 [49:51<26:09,  1.57it/s]
4|4014|Loss: 0.7136818170547485:  62%|██████▏   | 4014/6470 [49:51<25:59,  1.58it/s]
4|4015|Loss: 0.7896943688392639:  62%|██████▏   | 4014/6470 [49:51<25:59,  1.58it/s]
4|4015|Loss: 0.7896943688392639:  62%|██████▏   | 4015/6470 [49:52<29:03,  1.41it/s]
4|4016|Loss: 0.858221709728241:  62%|██████▏   | 4015/6470 [49:52<29:03,  1.41it/s] 
4|4016|Loss: 0.858221709728241:  62%|██████▏   | 4016/6470 [49:52<28:04,  1.46it/s]
4|4017|Loss: 0.9493374228477478:  62%|██████▏   | 4016/6470 [49:53<28:04,  1.46it/s]
4|4017|Loss: 0.9493374228477478:  62%|██████▏   | 4017/6470 [49:53<27:25,  1.49it/s]
4|4018|Loss: 0.9123862385749817:  62%|██████▏   | 4017/6470 [49:53<27:25,  1.49it/s]
4|4018|Loss: 0.9123862385749817:  62%|██████▏   | 4018/6470 [49:54<27:44,  1.47it/s]
4|4019|Loss: 0.9590885043144226:  62%|██████▏   | 4018/6470 [49:54<27:44,  1.47it/s]
4|4019|Loss: 0.9590885043144226:  62%|██████▏   | 4019/6470 [49:54<27:47,  1.47it/s]
4|4020|Loss: 0.8831441402435303:  62%|██████▏   | 4019/6470 [49:55<27:47,  1.47it/s]
4|4020|Loss: 0.8831441402435303:  62%|██████▏   | 4020/6470 [49:55<28:32,  1.43it/s]
4|4021|Loss: 0.8649084568023682:  62%|██████▏   | 4020/6470 [49:56<28:32,  1.43it/s]
4|4021|Loss: 0.8649084568023682:  62%|██████▏   | 4021/6470 [49:56<30:53,  1.32it/s]
4|4022|Loss: 0.879688560962677:  62%|██████▏   | 4021/6470 [49:57<30:53,  1.32it/s] 
4|4022|Loss: 0.879688560962677:  62%|██████▏   | 4022/6470 [49:57<33:06,  1.23it/s]
4|4023|Loss: 0.9357879757881165:  62%|██████▏   | 4022/6470 [49:58<33:06,  1.23it/s]
4|4023|Loss: 0.9357879757881165:  62%|██████▏   | 4023/6470 [49:58<33:18,  1.22it/s]
4|4024|Loss: 0.704456090927124:  62%|██████▏   | 4023/6470 [49:58<33:18,  1.22it/s] 
4|4024|Loss: 0.704456090927124:  62%|██████▏   | 4024/6470 [49:59<31:30,  1.29it/s]
4|4025|Loss: 0.7181817889213562:  62%|██████▏   | 4024/6470 [49:59<31:30,  1.29it/s]
4|4025|Loss: 0.7181817889213562:  62%|██████▏   | 4025/6470 [49:59<30:53,  1.32it/s]
4|4026|Loss: 0.8028834462165833:  62%|██████▏   | 4025/6470 [50:00<30:53,  1.32it/s]
4|4026|Loss: 0.8028834462165833:  62%|██████▏   | 4026/6470 [50:00<32:16,  1.26it/s]
4|4027|Loss: 0.7464889883995056:  62%|██████▏   | 4026/6470 [50:01<32:16,  1.26it/s]
4|4027|Loss: 0.7464889883995056:  62%|██████▏   | 4027/6470 [50:01<31:44,  1.28it/s]
4|4028|Loss: 0.7997239828109741:  62%|██████▏   | 4027/6470 [50:01<31:44,  1.28it/s]
4|4028|Loss: 0.7997239828109741:  62%|██████▏   | 4028/6470 [50:02<30:49,  1.32it/s]
4|4029|Loss: 0.7342022657394409:  62%|██████▏   | 4028/6470 [50:02<30:49,  1.32it/s]
4|4029|Loss: 0.7342022657394409:  62%|██████▏   | 4029/6470 [50:02<31:57,  1.27it/s]
4|4030|Loss: 0.8554704189300537:  62%|██████▏   | 4029/6470 [50:03<31:57,  1.27it/s]
4|4030|Loss: 0.8554704189300537:  62%|██████▏   | 4030/6470 [50:03<32:15,  1.26it/s]
4|4031|Loss: 0.9089049100875854:  62%|██████▏   | 4030/6470 [50:04<32:15,  1.26it/s]
4|4031|Loss: 0.9089049100875854:  62%|██████▏   | 4031/6470 [50:04<29:07,  1.40it/s]
4|4032|Loss: 0.8315712213516235:  62%|██████▏   | 4031/6470 [50:04<29:07,  1.40it/s]
4|4032|Loss: 0.8315712213516235:  62%|██████▏   | 4032/6470 [50:05<29:10,  1.39it/s]
4|4033|Loss: 0.847000002861023:  62%|██████▏   | 4032/6470 [50:05<29:10,  1.39it/s] 
4|4033|Loss: 0.847000002861023:  62%|██████▏   | 4033/6470 [50:05<28:32,  1.42it/s]
4|4034|Loss: 0.7722543478012085:  62%|██████▏   | 4033/6470 [50:06<28:32,  1.42it/s]
4|4034|Loss: 0.7722543478012085:  62%|██████▏   | 4034/6470 [50:06<28:51,  1.41it/s]
4|4035|Loss: 0.7891963124275208:  62%|██████▏   | 4034/6470 [50:06<28:51,  1.41it/s]
4|4035|Loss: 0.7891963124275208:  62%|██████▏   | 4035/6470 [50:07<28:44,  1.41it/s]
4|4036|Loss: 0.8494980931282043:  62%|██████▏   | 4035/6470 [50:07<28:44,  1.41it/s]
4|4036|Loss: 0.8494980931282043:  62%|██████▏   | 4036/6470 [50:07<27:12,  1.49it/s]
4|4037|Loss: 0.7357797026634216:  62%|██████▏   | 4036/6470 [50:08<27:12,  1.49it/s]
4|4037|Loss: 0.7357797026634216:  62%|██████▏   | 4037/6470 [50:08<25:49,  1.57it/s]
4|4038|Loss: 0.8766962289810181:  62%|██████▏   | 4037/6470 [50:08<25:49,  1.57it/s]
4|4038|Loss: 0.8766962289810181:  62%|██████▏   | 4038/6470 [50:08<26:47,  1.51it/s]
4|4039|Loss: 0.9617393612861633:  62%|██████▏   | 4038/6470 [50:09<26:47,  1.51it/s]
4|4039|Loss: 0.9617393612861633:  62%|██████▏   | 4039/6470 [50:09<26:36,  1.52it/s]
4|4040|Loss: 0.9189774394035339:  62%|██████▏   | 4039/6470 [50:10<26:36,  1.52it/s]
4|4040|Loss: 0.9189774394035339:  62%|██████▏   | 4040/6470 [50:10<27:55,  1.45it/s]
4|4041|Loss: 0.8609161376953125:  62%|██████▏   | 4040/6470 [50:10<27:55,  1.45it/s]
4|4041|Loss: 0.8609161376953125:  62%|██████▏   | 4041/6470 [50:11<28:51,  1.40it/s]
4|4042|Loss: 0.8525660634040833:  62%|██████▏   | 4041/6470 [50:11<28:51,  1.40it/s]
4|4042|Loss: 0.8525660634040833:  62%|██████▏   | 4042/6470 [50:11<29:50,  1.36it/s]
4|4043|Loss: 0.8715840578079224:  62%|██████▏   | 4042/6470 [50:12<29:50,  1.36it/s]
4|4043|Loss: 0.8715840578079224:  62%|██████▏   | 4043/6470 [50:12<30:39,  1.32it/s]
4|4044|Loss: 0.801554262638092:  62%|██████▏   | 4043/6470 [50:13<30:39,  1.32it/s] 
4|4044|Loss: 0.801554262638092:  63%|██████▎   | 4044/6470 [50:13<29:33,  1.37it/s]
4|4045|Loss: 0.9095017910003662:  63%|██████▎   | 4044/6470 [50:13<29:33,  1.37it/s]
4|4045|Loss: 0.9095017910003662:  63%|██████▎   | 4045/6470 [50:14<29:21,  1.38it/s]
4|4046|Loss: 0.7828707695007324:  63%|██████▎   | 4045/6470 [50:14<29:21,  1.38it/s]
4|4046|Loss: 0.7828707695007324:  63%|██████▎   | 4046/6470 [50:15<32:34,  1.24it/s]
4|4047|Loss: 0.9437951445579529:  63%|██████▎   | 4046/6470 [50:15<32:34,  1.24it/s]
4|4047|Loss: 0.9437951445579529:  63%|██████▎   | 4047/6470 [50:16<33:15,  1.21it/s]
4|4048|Loss: 0.7452613115310669:  63%|██████▎   | 4047/6470 [50:16<33:15,  1.21it/s]
4|4048|Loss: 0.7452613115310669:  63%|██████▎   | 4048/6470 [50:16<30:59,  1.30it/s]
4|4049|Loss: 0.8922792077064514:  63%|██████▎   | 4048/6470 [50:16<30:59,  1.30it/s]
4|4049|Loss: 0.8922792077064514:  63%|██████▎   | 4049/6470 [50:17<29:28,  1.37it/s]
4|4050|Loss: 0.7366232872009277:  63%|██████▎   | 4049/6470 [50:17<29:28,  1.37it/s]
4|4050|Loss: 0.7366232872009277:  63%|██████▎   | 4050/6470 [50:18<29:59,  1.34it/s]
4|4051|Loss: 0.8625372648239136:  63%|██████▎   | 4050/6470 [50:18<29:59,  1.34it/s]
4|4051|Loss: 0.8625372648239136:  63%|██████▎   | 4051/6470 [50:18<28:55,  1.39it/s]
4|4052|Loss: 0.6775825023651123:  63%|██████▎   | 4051/6470 [50:19<28:55,  1.39it/s]
4|4052|Loss: 0.6775825023651123:  63%|██████▎   | 4052/6470 [50:19<28:06,  1.43it/s]
4|4053|Loss: 0.9206768274307251:  63%|██████▎   | 4052/6470 [50:19<28:06,  1.43it/s]
4|4053|Loss: 0.9206768274307251:  63%|██████▎   | 4053/6470 [50:20<30:18,  1.33it/s]
4|4054|Loss: 0.6702240109443665:  63%|██████▎   | 4053/6470 [50:20<30:18,  1.33it/s]
4|4054|Loss: 0.6702240109443665:  63%|██████▎   | 4054/6470 [50:21<31:20,  1.28it/s]
4|4055|Loss: 0.806132435798645:  63%|██████▎   | 4054/6470 [50:21<31:20,  1.28it/s] 
4|4055|Loss: 0.806132435798645:  63%|██████▎   | 4055/6470 [50:21<29:24,  1.37it/s]
4|4056|Loss: 0.8132805228233337:  63%|██████▎   | 4055/6470 [50:22<29:24,  1.37it/s]
4|4056|Loss: 0.8132805228233337:  63%|██████▎   | 4056/6470 [50:22<28:13,  1.43it/s]
4|4057|Loss: 0.7777721881866455:  63%|██████▎   | 4056/6470 [50:22<28:13,  1.43it/s]
4|4057|Loss: 0.7777721881866455:  63%|██████▎   | 4057/6470 [50:23<28:40,  1.40it/s]
4|4058|Loss: 0.831452488899231:  63%|██████▎   | 4057/6470 [50:23<28:40,  1.40it/s] 
4|4058|Loss: 0.831452488899231:  63%|██████▎   | 4058/6470 [50:23<29:41,  1.35it/s]
4|4059|Loss: 0.7909817695617676:  63%|██████▎   | 4058/6470 [50:24<29:41,  1.35it/s]
4|4059|Loss: 0.7909817695617676:  63%|██████▎   | 4059/6470 [50:24<31:05,  1.29it/s]
4|4060|Loss: 0.9889165759086609:  63%|██████▎   | 4059/6470 [50:25<31:05,  1.29it/s]
4|4060|Loss: 0.9889165759086609:  63%|██████▎   | 4060/6470 [50:25<29:17,  1.37it/s]
4|4061|Loss: 0.7514427900314331:  63%|██████▎   | 4060/6470 [50:25<29:17,  1.37it/s]
4|4061|Loss: 0.7514427900314331:  63%|██████▎   | 4061/6470 [50:26<29:34,  1.36it/s]
4|4062|Loss: 0.8776658773422241:  63%|██████▎   | 4061/6470 [50:26<29:34,  1.36it/s]
4|4062|Loss: 0.8776658773422241:  63%|██████▎   | 4062/6470 [50:26<29:54,  1.34it/s]
4|4063|Loss: 0.8254438042640686:  63%|██████▎   | 4062/6470 [50:27<29:54,  1.34it/s]
4|4063|Loss: 0.8254438042640686:  63%|██████▎   | 4063/6470 [50:27<28:43,  1.40it/s]
4|4064|Loss: 0.8424025177955627:  63%|██████▎   | 4063/6470 [50:27<28:43,  1.40it/s]
4|4064|Loss: 0.8424025177955627:  63%|██████▎   | 4064/6470 [50:28<30:33,  1.31it/s]
4|4065|Loss: 0.8520347476005554:  63%|██████▎   | 4064/6470 [50:28<30:33,  1.31it/s]
4|4065|Loss: 0.8520347476005554:  63%|██████▎   | 4065/6470 [50:28<27:32,  1.46it/s]
4|4066|Loss: 0.7836306691169739:  63%|██████▎   | 4065/6470 [50:29<27:32,  1.46it/s]
4|4066|Loss: 0.7836306691169739:  63%|██████▎   | 4066/6470 [50:29<26:15,  1.53it/s]
4|4067|Loss: 0.8457613587379456:  63%|██████▎   | 4066/6470 [50:29<26:15,  1.53it/s]
4|4067|Loss: 0.8457613587379456:  63%|██████▎   | 4067/6470 [50:30<27:46,  1.44it/s]
4|4068|Loss: 0.8363508582115173:  63%|██████▎   | 4067/6470 [50:30<27:46,  1.44it/s]
4|4068|Loss: 0.8363508582115173:  63%|██████▎   | 4068/6470 [50:30<26:50,  1.49it/s]
4|4069|Loss: 0.7361796498298645:  63%|██████▎   | 4068/6470 [50:31<26:50,  1.49it/s]
4|4069|Loss: 0.7361796498298645:  63%|██████▎   | 4069/6470 [50:31<26:46,  1.49it/s]
4|4070|Loss: 0.8345375657081604:  63%|██████▎   | 4069/6470 [50:31<26:46,  1.49it/s]
4|4070|Loss: 0.8345375657081604:  63%|██████▎   | 4070/6470 [50:32<26:06,  1.53it/s]
4|4071|Loss: 0.7942910194396973:  63%|██████▎   | 4070/6470 [50:32<26:06,  1.53it/s]
4|4071|Loss: 0.7942910194396973:  63%|██████▎   | 4071/6470 [50:32<25:29,  1.57it/s]
4|4072|Loss: 0.6135625839233398:  63%|██████▎   | 4071/6470 [50:33<25:29,  1.57it/s]
4|4072|Loss: 0.6135625839233398:  63%|██████▎   | 4072/6470 [50:33<25:18,  1.58it/s]
4|4073|Loss: 0.9748677611351013:  63%|██████▎   | 4072/6470 [50:33<25:18,  1.58it/s]
4|4073|Loss: 0.9748677611351013:  63%|██████▎   | 4073/6470 [50:34<26:49,  1.49it/s]
4|4074|Loss: 0.830295979976654:  63%|██████▎   | 4073/6470 [50:34<26:49,  1.49it/s] 
4|4074|Loss: 0.830295979976654:  63%|██████▎   | 4074/6470 [50:34<28:10,  1.42it/s]
4|4075|Loss: 0.7479970455169678:  63%|██████▎   | 4074/6470 [50:35<28:10,  1.42it/s]
4|4075|Loss: 0.7479970455169678:  63%|██████▎   | 4075/6470 [50:35<29:16,  1.36it/s]
4|4076|Loss: 0.6979936361312866:  63%|██████▎   | 4075/6470 [50:36<29:16,  1.36it/s]
4|4076|Loss: 0.6979936361312866:  63%|██████▎   | 4076/6470 [50:36<28:20,  1.41it/s]
4|4077|Loss: 0.8960784077644348:  63%|██████▎   | 4076/6470 [50:36<28:20,  1.41it/s]
4|4077|Loss: 0.8960784077644348:  63%|██████▎   | 4077/6470 [50:37<28:42,  1.39it/s]
4|4078|Loss: 0.8678036332130432:  63%|██████▎   | 4077/6470 [50:37<28:42,  1.39it/s]
4|4078|Loss: 0.8678036332130432:  63%|██████▎   | 4078/6470 [50:37<30:07,  1.32it/s]
4|4079|Loss: 1.0033228397369385:  63%|██████▎   | 4078/6470 [50:38<30:07,  1.32it/s]
4|4079|Loss: 1.0033228397369385:  63%|██████▎   | 4079/6470 [50:38<28:37,  1.39it/s]
4|4080|Loss: 0.8980259895324707:  63%|██████▎   | 4079/6470 [50:38<28:37,  1.39it/s]
4|4080|Loss: 0.8980259895324707:  63%|██████▎   | 4080/6470 [50:39<29:02,  1.37it/s]
4|4081|Loss: 0.9253004193305969:  63%|██████▎   | 4080/6470 [50:39<29:02,  1.37it/s]
4|4081|Loss: 0.9253004193305969:  63%|██████▎   | 4081/6470 [50:40<30:59,  1.28it/s]
4|4082|Loss: 0.8596504330635071:  63%|██████▎   | 4081/6470 [50:40<30:59,  1.28it/s]
4|4082|Loss: 0.8596504330635071:  63%|██████▎   | 4082/6470 [50:40<29:46,  1.34it/s]
4|4083|Loss: 0.904926061630249:  63%|██████▎   | 4082/6470 [50:41<29:46,  1.34it/s] 
4|4083|Loss: 0.904926061630249:  63%|██████▎   | 4083/6470 [50:41<29:56,  1.33it/s]
4|4084|Loss: 0.7689491510391235:  63%|██████▎   | 4083/6470 [50:42<29:56,  1.33it/s]
4|4084|Loss: 0.7689491510391235:  63%|██████▎   | 4084/6470 [50:42<29:12,  1.36it/s]
4|4085|Loss: 0.8736578822135925:  63%|██████▎   | 4084/6470 [50:42<29:12,  1.36it/s]
4|4085|Loss: 0.8736578822135925:  63%|██████▎   | 4085/6470 [50:43<28:39,  1.39it/s]
4|4086|Loss: 0.6485691666603088:  63%|██████▎   | 4085/6470 [50:43<28:39,  1.39it/s]
4|4086|Loss: 0.6485691666603088:  63%|██████▎   | 4086/6470 [50:43<28:07,  1.41it/s]
4|4087|Loss: 0.8244404792785645:  63%|██████▎   | 4086/6470 [50:44<28:07,  1.41it/s]
4|4087|Loss: 0.8244404792785645:  63%|██████▎   | 4087/6470 [50:44<27:52,  1.42it/s]
4|4088|Loss: 0.7244254350662231:  63%|██████▎   | 4087/6470 [50:44<27:52,  1.42it/s]
4|4088|Loss: 0.7244254350662231:  63%|██████▎   | 4088/6470 [50:45<28:52,  1.38it/s]
4|4089|Loss: 0.829418957233429:  63%|██████▎   | 4088/6470 [50:45<28:52,  1.38it/s] 
4|4089|Loss: 0.829418957233429:  63%|██████▎   | 4089/6470 [50:45<29:01,  1.37it/s]
4|4090|Loss: 0.7994163036346436:  63%|██████▎   | 4089/6470 [50:46<29:01,  1.37it/s]
4|4090|Loss: 0.7994163036346436:  63%|██████▎   | 4090/6470 [50:46<30:50,  1.29it/s]
4|4091|Loss: 0.7595471143722534:  63%|██████▎   | 4090/6470 [50:47<30:50,  1.29it/s]
4|4091|Loss: 0.7595471143722534:  63%|██████▎   | 4091/6470 [50:47<30:03,  1.32it/s]
4|4092|Loss: 0.756369411945343:  63%|██████▎   | 4091/6470 [50:47<30:03,  1.32it/s] 
4|4092|Loss: 0.756369411945343:  63%|██████▎   | 4092/6470 [50:48<28:11,  1.41it/s]
4|4093|Loss: 0.7709565758705139:  63%|██████▎   | 4092/6470 [50:48<28:11,  1.41it/s]
4|4093|Loss: 0.7709565758705139:  63%|██████▎   | 4093/6470 [50:48<28:38,  1.38it/s]
4|4094|Loss: 0.8375591039657593:  63%|██████▎   | 4093/6470 [50:49<28:38,  1.38it/s]
4|4094|Loss: 0.8375591039657593:  63%|██████▎   | 4094/6470 [50:49<29:19,  1.35it/s]
4|4095|Loss: 0.8309655785560608:  63%|██████▎   | 4094/6470 [50:50<29:19,  1.35it/s]
4|4095|Loss: 0.8309655785560608:  63%|██████▎   | 4095/6470 [50:50<31:15,  1.27it/s]
4|4096|Loss: 0.8889268636703491:  63%|██████▎   | 4095/6470 [50:51<31:15,  1.27it/s]
4|4096|Loss: 0.8889268636703491:  63%|██████▎   | 4096/6470 [50:51<30:47,  1.28it/s]
4|4097|Loss: 0.8745877742767334:  63%|██████▎   | 4096/6470 [50:51<30:47,  1.28it/s]
4|4097|Loss: 0.8745877742767334:  63%|██████▎   | 4097/6470 [50:52<30:37,  1.29it/s]
4|4098|Loss: 0.6758854389190674:  63%|██████▎   | 4097/6470 [50:52<30:37,  1.29it/s]
4|4098|Loss: 0.6758854389190674:  63%|██████▎   | 4098/6470 [50:52<31:29,  1.26it/s]
4|4099|Loss: 0.827372133731842:  63%|██████▎   | 4098/6470 [50:53<31:29,  1.26it/s] 
4|4099|Loss: 0.827372133731842:  63%|██████▎   | 4099/6470 [50:53<31:15,  1.26it/s]
4|4100|Loss: 0.7704736590385437:  63%|██████▎   | 4099/6470 [50:54<31:15,  1.26it/s]
4|4100|Loss: 0.7704736590385437:  63%|██████▎   | 4100/6470 [50:54<31:08,  1.27it/s]
4|4101|Loss: 0.8542095422744751:  63%|██████▎   | 4100/6470 [50:54<31:08,  1.27it/s]
4|4101|Loss: 0.8542095422744751:  63%|██████▎   | 4101/6470 [50:55<31:06,  1.27it/s]
4|4102|Loss: 0.8727124929428101:  63%|██████▎   | 4101/6470 [50:55<31:06,  1.27it/s]
4|4102|Loss: 0.8727124929428101:  63%|██████▎   | 4102/6470 [50:55<29:41,  1.33it/s]
4|4103|Loss: 0.6473374366760254:  63%|██████▎   | 4102/6470 [50:56<29:41,  1.33it/s]
4|4103|Loss: 0.6473374366760254:  63%|██████▎   | 4103/6470 [50:56<29:15,  1.35it/s]
4|4104|Loss: 0.7558073401451111:  63%|██████▎   | 4103/6470 [50:57<29:15,  1.35it/s]
4|4104|Loss: 0.7558073401451111:  63%|██████▎   | 4104/6470 [50:57<29:48,  1.32it/s]
4|4105|Loss: 0.9308117032051086:  63%|██████▎   | 4104/6470 [50:57<29:48,  1.32it/s]
4|4105|Loss: 0.9308117032051086:  63%|██████▎   | 4105/6470 [50:58<31:24,  1.25it/s]
4|4106|Loss: 0.9292360544204712:  63%|██████▎   | 4105/6470 [50:58<31:24,  1.25it/s]
4|4106|Loss: 0.9292360544204712:  63%|██████▎   | 4106/6470 [50:59<31:37,  1.25it/s]
4|4107|Loss: 0.829707682132721:  63%|██████▎   | 4106/6470 [50:59<31:37,  1.25it/s] 
4|4107|Loss: 0.829707682132721:  63%|██████▎   | 4107/6470 [50:59<30:19,  1.30it/s]
4|4108|Loss: 0.7591215372085571:  63%|██████▎   | 4107/6470 [51:00<30:19,  1.30it/s]
4|4108|Loss: 0.7591215372085571:  63%|██████▎   | 4108/6470 [51:00<31:28,  1.25it/s]
4|4109|Loss: 0.6703635454177856:  63%|██████▎   | 4108/6470 [51:01<31:28,  1.25it/s]
4|4109|Loss: 0.6703635454177856:  64%|██████▎   | 4109/6470 [51:01<29:15,  1.35it/s]
4|4110|Loss: 0.673343300819397:  64%|██████▎   | 4109/6470 [51:01<29:15,  1.35it/s] 
4|4110|Loss: 0.673343300819397:  64%|██████▎   | 4110/6470 [51:02<30:28,  1.29it/s]
4|4111|Loss: 0.8304107189178467:  64%|██████▎   | 4110/6470 [51:02<30:28,  1.29it/s]
4|4111|Loss: 0.8304107189178467:  64%|██████▎   | 4111/6470 [51:03<33:05,  1.19it/s]
4|4112|Loss: 0.7616648077964783:  64%|██████▎   | 4111/6470 [51:03<33:05,  1.19it/s]
4|4112|Loss: 0.7616648077964783:  64%|██████▎   | 4112/6470 [51:03<31:31,  1.25it/s]
4|4113|Loss: 0.862952709197998:  64%|██████▎   | 4112/6470 [51:04<31:31,  1.25it/s] 
4|4113|Loss: 0.862952709197998:  64%|██████▎   | 4113/6470 [51:04<29:23,  1.34it/s]
4|4114|Loss: 0.834106981754303:  64%|██████▎   | 4113/6470 [51:04<29:23,  1.34it/s]
4|4114|Loss: 0.834106981754303:  64%|██████▎   | 4114/6470 [51:05<28:26,  1.38it/s]
4|4115|Loss: 0.8873295783996582:  64%|██████▎   | 4114/6470 [51:05<28:26,  1.38it/s]
4|4115|Loss: 0.8873295783996582:  64%|██████▎   | 4115/6470 [51:05<26:17,  1.49it/s]
4|4116|Loss: 0.9156739711761475:  64%|██████▎   | 4115/6470 [51:06<26:17,  1.49it/s]
4|4116|Loss: 0.9156739711761475:  64%|██████▎   | 4116/6470 [51:06<25:45,  1.52it/s]
4|4117|Loss: 0.8648853302001953:  64%|██████▎   | 4116/6470 [51:06<25:45,  1.52it/s]
4|4117|Loss: 0.8648853302001953:  64%|██████▎   | 4117/6470 [51:07<28:08,  1.39it/s]
4|4118|Loss: 0.8082571029663086:  64%|██████▎   | 4117/6470 [51:07<28:08,  1.39it/s]
4|4118|Loss: 0.8082571029663086:  64%|██████▎   | 4118/6470 [51:08<29:04,  1.35it/s]
4|4119|Loss: 0.761908233165741:  64%|██████▎   | 4118/6470 [51:08<29:04,  1.35it/s] 
4|4119|Loss: 0.761908233165741:  64%|██████▎   | 4119/6470 [51:08<29:17,  1.34it/s]
4|4120|Loss: 0.8010914921760559:  64%|██████▎   | 4119/6470 [51:09<29:17,  1.34it/s]
4|4120|Loss: 0.8010914921760559:  64%|██████▎   | 4120/6470 [51:09<28:10,  1.39it/s]
4|4121|Loss: 0.9005889892578125:  64%|██████▎   | 4120/6470 [51:09<28:10,  1.39it/s]
4|4121|Loss: 0.9005889892578125:  64%|██████▎   | 4121/6470 [51:10<26:25,  1.48it/s]
4|4122|Loss: 0.8504178524017334:  64%|██████▎   | 4121/6470 [51:10<26:25,  1.48it/s]
4|4122|Loss: 0.8504178524017334:  64%|██████▎   | 4122/6470 [51:10<27:22,  1.43it/s]
4|4123|Loss: 0.785606861114502:  64%|██████▎   | 4122/6470 [51:11<27:22,  1.43it/s] 
4|4123|Loss: 0.785606861114502:  64%|██████▎   | 4123/6470 [51:11<28:17,  1.38it/s]
4|4124|Loss: 0.8195399641990662:  64%|██████▎   | 4123/6470 [51:12<28:17,  1.38it/s]
4|4124|Loss: 0.8195399641990662:  64%|██████▎   | 4124/6470 [51:12<31:05,  1.26it/s]
4|4125|Loss: 0.8175355195999146:  64%|██████▎   | 4124/6470 [51:12<31:05,  1.26it/s]
4|4125|Loss: 0.8175355195999146:  64%|██████▍   | 4125/6470 [51:13<30:34,  1.28it/s]
4|4126|Loss: 0.8598775267601013:  64%|██████▍   | 4125/6470 [51:13<30:34,  1.28it/s]
4|4126|Loss: 0.8598775267601013:  64%|██████▍   | 4126/6470 [51:14<31:07,  1.26it/s]
4|4127|Loss: 0.8643736243247986:  64%|██████▍   | 4126/6470 [51:14<31:07,  1.26it/s]
4|4127|Loss: 0.8643736243247986:  64%|██████▍   | 4127/6470 [51:15<32:11,  1.21it/s]
4|4128|Loss: 0.7945165634155273:  64%|██████▍   | 4127/6470 [51:15<32:11,  1.21it/s]
4|4128|Loss: 0.7945165634155273:  64%|██████▍   | 4128/6470 [51:15<31:09,  1.25it/s]
4|4129|Loss: 0.8517022132873535:  64%|██████▍   | 4128/6470 [51:16<31:09,  1.25it/s]
4|4129|Loss: 0.8517022132873535:  64%|██████▍   | 4129/6470 [51:16<30:01,  1.30it/s]
4|4130|Loss: 0.728872537612915:  64%|██████▍   | 4129/6470 [51:16<30:01,  1.30it/s] 
4|4130|Loss: 0.728872537612915:  64%|██████▍   | 4130/6470 [51:16<27:03,  1.44it/s]
4|4131|Loss: 0.7791929841041565:  64%|██████▍   | 4130/6470 [51:17<27:03,  1.44it/s]
4|4131|Loss: 0.7791929841041565:  64%|██████▍   | 4131/6470 [51:17<27:29,  1.42it/s]
4|4132|Loss: 0.749452531337738:  64%|██████▍   | 4131/6470 [51:18<27:29,  1.42it/s] 
4|4132|Loss: 0.749452531337738:  64%|██████▍   | 4132/6470 [51:18<27:50,  1.40it/s]
4|4133|Loss: 0.8697839379310608:  64%|██████▍   | 4132/6470 [51:18<27:50,  1.40it/s]
4|4133|Loss: 0.8697839379310608:  64%|██████▍   | 4133/6470 [51:19<27:47,  1.40it/s]
4|4134|Loss: 0.7221558094024658:  64%|██████▍   | 4133/6470 [51:19<27:47,  1.40it/s]
4|4134|Loss: 0.7221558094024658:  64%|██████▍   | 4134/6470 [51:20<30:58,  1.26it/s]
4|4135|Loss: 0.874334454536438:  64%|██████▍   | 4134/6470 [51:20<30:58,  1.26it/s] 
4|4135|Loss: 0.874334454536438:  64%|██████▍   | 4135/6470 [51:20<29:09,  1.33it/s]
4|4136|Loss: 0.9346910715103149:  64%|██████▍   | 4135/6470 [51:21<29:09,  1.33it/s]
4|4136|Loss: 0.9346910715103149:  64%|██████▍   | 4136/6470 [51:21<27:02,  1.44it/s]
4|4137|Loss: 0.6944543123245239:  64%|██████▍   | 4136/6470 [51:21<27:02,  1.44it/s]
4|4137|Loss: 0.6944543123245239:  64%|██████▍   | 4137/6470 [51:22<27:35,  1.41it/s]
4|4138|Loss: 0.8678752779960632:  64%|██████▍   | 4137/6470 [51:22<27:35,  1.41it/s]
4|4138|Loss: 0.8678752779960632:  64%|██████▍   | 4138/6470 [51:22<26:12,  1.48it/s]
4|4139|Loss: 0.8489423990249634:  64%|██████▍   | 4138/6470 [51:23<26:12,  1.48it/s]
4|4139|Loss: 0.8489423990249634:  64%|██████▍   | 4139/6470 [51:23<26:02,  1.49it/s]
4|4140|Loss: 0.8321946859359741:  64%|██████▍   | 4139/6470 [51:23<26:02,  1.49it/s]
4|4140|Loss: 0.8321946859359741:  64%|██████▍   | 4140/6470 [51:24<26:05,  1.49it/s]
4|4141|Loss: 1.0526862144470215:  64%|██████▍   | 4140/6470 [51:24<26:05,  1.49it/s]
4|4141|Loss: 1.0526862144470215:  64%|██████▍   | 4141/6470 [51:24<24:45,  1.57it/s]
4|4142|Loss: 0.8983970880508423:  64%|██████▍   | 4141/6470 [51:24<24:45,  1.57it/s]
4|4142|Loss: 0.8983970880508423:  64%|██████▍   | 4142/6470 [51:25<25:39,  1.51it/s]
4|4143|Loss: 0.9363088607788086:  64%|██████▍   | 4142/6470 [51:25<25:39,  1.51it/s]
4|4143|Loss: 0.9363088607788086:  64%|██████▍   | 4143/6470 [51:26<27:57,  1.39it/s]
4|4144|Loss: 0.7423205971717834:  64%|██████▍   | 4143/6470 [51:26<27:57,  1.39it/s]
4|4144|Loss: 0.7423205971717834:  64%|██████▍   | 4144/6470 [51:26<27:08,  1.43it/s]
4|4145|Loss: 0.9863219857215881:  64%|██████▍   | 4144/6470 [51:27<27:08,  1.43it/s]
4|4145|Loss: 0.9863219857215881:  64%|██████▍   | 4145/6470 [51:27<29:24,  1.32it/s]
4|4146|Loss: 0.7142287492752075:  64%|██████▍   | 4145/6470 [51:28<29:24,  1.32it/s]
4|4146|Loss: 0.7142287492752075:  64%|██████▍   | 4146/6470 [51:28<26:23,  1.47it/s]
4|4147|Loss: 0.7417353987693787:  64%|██████▍   | 4146/6470 [51:28<26:23,  1.47it/s]
4|4147|Loss: 0.7417353987693787:  64%|██████▍   | 4147/6470 [51:28<25:19,  1.53it/s]
4|4148|Loss: 0.8141971826553345:  64%|██████▍   | 4147/6470 [51:29<25:19,  1.53it/s]
4|4148|Loss: 0.8141971826553345:  64%|██████▍   | 4148/6470 [51:29<27:39,  1.40it/s]
4|4149|Loss: 0.7668736577033997:  64%|██████▍   | 4148/6470 [51:30<27:39,  1.40it/s]
4|4149|Loss: 0.7668736577033997:  64%|██████▍   | 4149/6470 [51:30<27:52,  1.39it/s]
4|4150|Loss: 0.7288780808448792:  64%|██████▍   | 4149/6470 [51:30<27:52,  1.39it/s]
4|4150|Loss: 0.7288780808448792:  64%|██████▍   | 4150/6470 [51:31<28:26,  1.36it/s]
4|4151|Loss: 0.8670670986175537:  64%|██████▍   | 4150/6470 [51:31<28:26,  1.36it/s]
4|4151|Loss: 0.8670670986175537:  64%|██████▍   | 4151/6470 [51:31<27:10,  1.42it/s]
4|4152|Loss: 0.8449008464813232:  64%|██████▍   | 4151/6470 [51:32<27:10,  1.42it/s]
4|4152|Loss: 0.8449008464813232:  64%|██████▍   | 4152/6470 [51:32<27:25,  1.41it/s]
4|4153|Loss: 0.7263515591621399:  64%|██████▍   | 4152/6470 [51:32<27:25,  1.41it/s]
4|4153|Loss: 0.7263515591621399:  64%|██████▍   | 4153/6470 [51:33<26:42,  1.45it/s]
4|4154|Loss: 0.9283879399299622:  64%|██████▍   | 4153/6470 [51:33<26:42,  1.45it/s]
4|4154|Loss: 0.9283879399299622:  64%|██████▍   | 4154/6470 [51:33<24:31,  1.57it/s]
4|4155|Loss: 0.7472493052482605:  64%|██████▍   | 4154/6470 [51:33<24:31,  1.57it/s]
4|4155|Loss: 0.7472493052482605:  64%|██████▍   | 4155/6470 [51:34<23:36,  1.63it/s]
4|4156|Loss: 0.7501063942909241:  64%|██████▍   | 4155/6470 [51:34<23:36,  1.63it/s]
4|4156|Loss: 0.7501063942909241:  64%|██████▍   | 4156/6470 [51:34<24:52,  1.55it/s]
4|4157|Loss: 0.7821642160415649:  64%|██████▍   | 4156/6470 [51:35<24:52,  1.55it/s]
4|4157|Loss: 0.7821642160415649:  64%|██████▍   | 4157/6470 [51:35<26:08,  1.47it/s]
4|4158|Loss: 0.8724663853645325:  64%|██████▍   | 4157/6470 [51:36<26:08,  1.47it/s]
4|4158|Loss: 0.8724663853645325:  64%|██████▍   | 4158/6470 [51:36<25:30,  1.51it/s]
4|4159|Loss: 0.9178579449653625:  64%|██████▍   | 4158/6470 [51:36<25:30,  1.51it/s]
4|4159|Loss: 0.9178579449653625:  64%|██████▍   | 4159/6470 [51:37<27:08,  1.42it/s]
4|4160|Loss: 0.9714818000793457:  64%|██████▍   | 4159/6470 [51:37<27:08,  1.42it/s]
4|4160|Loss: 0.9714818000793457:  64%|██████▍   | 4160/6470 [51:38<29:36,  1.30it/s]
4|4161|Loss: 0.8759973645210266:  64%|██████▍   | 4160/6470 [51:38<29:36,  1.30it/s]
4|4161|Loss: 0.8759973645210266:  64%|██████▍   | 4161/6470 [51:38<29:53,  1.29it/s]
4|4162|Loss: 0.8222975730895996:  64%|██████▍   | 4161/6470 [51:39<29:53,  1.29it/s]
4|4162|Loss: 0.8222975730895996:  64%|██████▍   | 4162/6470 [51:39<31:03,  1.24it/s]
4|4163|Loss: 0.7882956862449646:  64%|██████▍   | 4162/6470 [51:40<31:03,  1.24it/s]
4|4163|Loss: 0.7882956862449646:  64%|██████▍   | 4163/6470 [51:40<32:34,  1.18it/s]
4|4164|Loss: 1.0067311525344849:  64%|██████▍   | 4163/6470 [51:41<32:34,  1.18it/s]
4|4164|Loss: 1.0067311525344849:  64%|██████▍   | 4164/6470 [51:41<32:21,  1.19it/s]
4|4165|Loss: 0.9044023156166077:  64%|██████▍   | 4164/6470 [51:41<32:21,  1.19it/s]
4|4165|Loss: 0.9044023156166077:  64%|██████▍   | 4165/6470 [51:42<31:44,  1.21it/s]
4|4166|Loss: 0.8572095036506653:  64%|██████▍   | 4165/6470 [51:42<31:44,  1.21it/s]
4|4166|Loss: 0.8572095036506653:  64%|██████▍   | 4166/6470 [51:43<31:03,  1.24it/s]
4|4167|Loss: 0.7552028894424438:  64%|██████▍   | 4166/6470 [51:43<31:03,  1.24it/s]
4|4167|Loss: 0.7552028894424438:  64%|██████▍   | 4167/6470 [51:43<30:53,  1.24it/s]
4|4168|Loss: 0.8370148539543152:  64%|██████▍   | 4167/6470 [51:44<30:53,  1.24it/s]
4|4168|Loss: 0.8370148539543152:  64%|██████▍   | 4168/6470 [51:44<32:30,  1.18it/s]
4|4169|Loss: 0.8153613805770874:  64%|██████▍   | 4168/6470 [51:45<32:30,  1.18it/s]
4|4169|Loss: 0.8153613805770874:  64%|██████▍   | 4169/6470 [51:45<33:09,  1.16it/s]
4|4170|Loss: 0.6770879626274109:  64%|██████▍   | 4169/6470 [51:46<33:09,  1.16it/s]
4|4170|Loss: 0.6770879626274109:  64%|██████▍   | 4170/6470 [51:46<32:29,  1.18it/s]
4|4171|Loss: 0.7517795562744141:  64%|██████▍   | 4170/6470 [51:46<32:29,  1.18it/s]
4|4171|Loss: 0.7517795562744141:  64%|██████▍   | 4171/6470 [51:47<31:50,  1.20it/s]
4|4172|Loss: 0.789176881313324:  64%|██████▍   | 4171/6470 [51:47<31:50,  1.20it/s] 
4|4172|Loss: 0.789176881313324:  64%|██████▍   | 4172/6470 [51:47<29:49,  1.28it/s]
4|4173|Loss: 0.9181835651397705:  64%|██████▍   | 4172/6470 [51:48<29:49,  1.28it/s]
4|4173|Loss: 0.9181835651397705:  64%|██████▍   | 4173/6470 [51:48<29:47,  1.28it/s]
4|4174|Loss: 0.792711615562439:  64%|██████▍   | 4173/6470 [51:49<29:47,  1.28it/s] 
4|4174|Loss: 0.792711615562439:  65%|██████▍   | 4174/6470 [51:49<30:55,  1.24it/s]
4|4175|Loss: 0.845449686050415:  65%|██████▍   | 4174/6470 [51:50<30:55,  1.24it/s]
4|4175|Loss: 0.845449686050415:  65%|██████▍   | 4175/6470 [51:50<30:57,  1.24it/s]
4|4176|Loss: 0.8865523338317871:  65%|██████▍   | 4175/6470 [51:50<30:57,  1.24it/s]
4|4176|Loss: 0.8865523338317871:  65%|██████▍   | 4176/6470 [51:51<30:28,  1.25it/s]
4|4177|Loss: 0.7366152405738831:  65%|██████▍   | 4176/6470 [51:51<30:28,  1.25it/s]
4|4177|Loss: 0.7366152405738831:  65%|██████▍   | 4177/6470 [51:51<30:16,  1.26it/s]
4|4178|Loss: 0.7744590640068054:  65%|██████▍   | 4177/6470 [51:52<30:16,  1.26it/s]
4|4178|Loss: 0.7744590640068054:  65%|██████▍   | 4178/6470 [51:52<29:01,  1.32it/s]
4|4179|Loss: 0.6674773693084717:  65%|██████▍   | 4178/6470 [51:53<29:01,  1.32it/s]
4|4179|Loss: 0.6674773693084717:  65%|██████▍   | 4179/6470 [51:53<30:01,  1.27it/s]
4|4180|Loss: 0.9853649735450745:  65%|██████▍   | 4179/6470 [51:53<30:01,  1.27it/s]
4|4180|Loss: 0.9853649735450745:  65%|██████▍   | 4180/6470 [51:54<30:24,  1.25it/s]
4|4181|Loss: 0.7527644634246826:  65%|██████▍   | 4180/6470 [51:54<30:24,  1.25it/s]
4|4181|Loss: 0.7527644634246826:  65%|██████▍   | 4181/6470 [51:55<30:38,  1.25it/s]
4|4182|Loss: 0.8363809585571289:  65%|██████▍   | 4181/6470 [51:55<30:38,  1.25it/s]
4|4182|Loss: 0.8363809585571289:  65%|██████▍   | 4182/6470 [51:55<30:16,  1.26it/s]
4|4183|Loss: 0.8647461533546448:  65%|██████▍   | 4182/6470 [51:56<30:16,  1.26it/s]
4|4183|Loss: 0.8647461533546448:  65%|██████▍   | 4183/6470 [51:56<28:17,  1.35it/s]
4|4184|Loss: 0.8785940408706665:  65%|██████▍   | 4183/6470 [51:56<28:17,  1.35it/s]
4|4184|Loss: 0.8785940408706665:  65%|██████▍   | 4184/6470 [51:57<26:52,  1.42it/s]
4|4185|Loss: 0.7830659747123718:  65%|██████▍   | 4184/6470 [51:57<26:52,  1.42it/s]
4|4185|Loss: 0.7830659747123718:  65%|██████▍   | 4185/6470 [51:58<28:45,  1.32it/s]
4|4186|Loss: 0.8757506012916565:  65%|██████▍   | 4185/6470 [51:58<28:45,  1.32it/s]
4|4186|Loss: 0.8757506012916565:  65%|██████▍   | 4186/6470 [51:58<30:34,  1.25it/s]
4|4187|Loss: 0.9006985425949097:  65%|██████▍   | 4186/6470 [51:59<30:34,  1.25it/s]
4|4187|Loss: 0.9006985425949097:  65%|██████▍   | 4187/6470 [51:59<29:47,  1.28it/s]
4|4188|Loss: 0.8669677376747131:  65%|██████▍   | 4187/6470 [52:00<29:47,  1.28it/s]
4|4188|Loss: 0.8669677376747131:  65%|██████▍   | 4188/6470 [52:00<28:45,  1.32it/s]
4|4189|Loss: 0.7382030487060547:  65%|██████▍   | 4188/6470 [52:00<28:45,  1.32it/s]
4|4189|Loss: 0.7382030487060547:  65%|██████▍   | 4189/6470 [52:01<28:56,  1.31it/s]
4|4190|Loss: 0.8898711204528809:  65%|██████▍   | 4189/6470 [52:01<28:56,  1.31it/s]
4|4190|Loss: 0.8898711204528809:  65%|██████▍   | 4190/6470 [52:01<27:40,  1.37it/s]
4|4191|Loss: 0.7279773354530334:  65%|██████▍   | 4190/6470 [52:02<27:40,  1.37it/s]
4|4191|Loss: 0.7279773354530334:  65%|██████▍   | 4191/6470 [52:02<25:36,  1.48it/s]
4|4192|Loss: 0.8332149982452393:  65%|██████▍   | 4191/6470 [52:02<25:36,  1.48it/s]
4|4192|Loss: 0.8332149982452393:  65%|██████▍   | 4192/6470 [52:02<22:47,  1.67it/s]
4|4193|Loss: 0.8598326444625854:  65%|██████▍   | 4192/6470 [52:03<22:47,  1.67it/s]
4|4193|Loss: 0.8598326444625854:  65%|██████▍   | 4193/6470 [52:03<23:06,  1.64it/s]
4|4194|Loss: 0.7944238185882568:  65%|██████▍   | 4193/6470 [52:03<23:06,  1.64it/s]
4|4194|Loss: 0.7944238185882568:  65%|██████▍   | 4194/6470 [52:04<24:02,  1.58it/s]
4|4195|Loss: 0.8380600214004517:  65%|██████▍   | 4194/6470 [52:04<24:02,  1.58it/s]
4|4195|Loss: 0.8380600214004517:  65%|██████▍   | 4195/6470 [52:04<24:22,  1.56it/s]
4|4196|Loss: 0.8657156229019165:  65%|██████▍   | 4195/6470 [52:05<24:22,  1.56it/s]
4|4196|Loss: 0.8657156229019165:  65%|██████▍   | 4196/6470 [52:05<26:47,  1.41it/s]
4|4197|Loss: 0.8632732033729553:  65%|██████▍   | 4196/6470 [52:06<26:47,  1.41it/s]
4|4197|Loss: 0.8632732033729553:  65%|██████▍   | 4197/6470 [52:06<28:58,  1.31it/s]
4|4198|Loss: 0.6481878161430359:  65%|██████▍   | 4197/6470 [52:07<28:58,  1.31it/s]
4|4198|Loss: 0.6481878161430359:  65%|██████▍   | 4198/6470 [52:07<30:43,  1.23it/s]
4|4199|Loss: 1.029669165611267:  65%|██████▍   | 4198/6470 [52:07<30:43,  1.23it/s] 
4|4199|Loss: 1.029669165611267:  65%|██████▍   | 4199/6470 [52:08<29:44,  1.27it/s]
4|4200|Loss: 0.7833059430122375:  65%|██████▍   | 4199/6470 [52:08<29:44,  1.27it/s]
4|4200|Loss: 0.7833059430122375:  65%|██████▍   | 4200/6470 [52:08<28:18,  1.34it/s]
4|4201|Loss: 1.01604163646698:  65%|██████▍   | 4200/6470 [52:09<28:18,  1.34it/s]  
4|4201|Loss: 1.01604163646698:  65%|██████▍   | 4201/6470 [52:09<28:27,  1.33it/s]
4|4202|Loss: 0.826413094997406:  65%|██████▍   | 4201/6470 [52:10<28:27,  1.33it/s]
4|4202|Loss: 0.826413094997406:  65%|██████▍   | 4202/6470 [52:10<29:52,  1.27it/s]
4|4203|Loss: 0.8273976445198059:  65%|██████▍   | 4202/6470 [52:10<29:52,  1.27it/s]
4|4203|Loss: 0.8273976445198059:  65%|██████▍   | 4203/6470 [52:11<30:26,  1.24it/s]
4|4204|Loss: 0.9636726975440979:  65%|██████▍   | 4203/6470 [52:11<30:26,  1.24it/s]
4|4204|Loss: 0.9636726975440979:  65%|██████▍   | 4204/6470 [52:12<29:27,  1.28it/s]
4|4205|Loss: 0.778079628944397:  65%|██████▍   | 4204/6470 [52:12<29:27,  1.28it/s] 
4|4205|Loss: 0.778079628944397:  65%|██████▍   | 4205/6470 [52:12<29:58,  1.26it/s]
4|4206|Loss: 0.8504541516304016:  65%|██████▍   | 4205/6470 [52:13<29:58,  1.26it/s]
4|4206|Loss: 0.8504541516304016:  65%|██████▌   | 4206/6470 [52:13<31:37,  1.19it/s]
4|4207|Loss: 0.831801176071167:  65%|██████▌   | 4206/6470 [52:14<31:37,  1.19it/s] 
4|4207|Loss: 0.831801176071167:  65%|██████▌   | 4207/6470 [52:14<31:41,  1.19it/s]
4|4208|Loss: 0.8293741345405579:  65%|██████▌   | 4207/6470 [52:15<31:41,  1.19it/s]
4|4208|Loss: 0.8293741345405579:  65%|██████▌   | 4208/6470 [52:15<31:26,  1.20it/s]
4|4209|Loss: 0.7039597630500793:  65%|██████▌   | 4208/6470 [52:15<31:26,  1.20it/s]
4|4209|Loss: 0.7039597630500793:  65%|██████▌   | 4209/6470 [52:16<31:07,  1.21it/s]
4|4210|Loss: 0.9240621328353882:  65%|██████▌   | 4209/6470 [52:16<31:07,  1.21it/s]
4|4210|Loss: 0.9240621328353882:  65%|██████▌   | 4210/6470 [52:17<30:48,  1.22it/s]
4|4211|Loss: 0.7102105617523193:  65%|██████▌   | 4210/6470 [52:17<30:48,  1.22it/s]
4|4211|Loss: 0.7102105617523193:  65%|██████▌   | 4211/6470 [52:17<30:35,  1.23it/s]
4|4212|Loss: 0.8382601141929626:  65%|██████▌   | 4211/6470 [52:18<30:35,  1.23it/s]
4|4212|Loss: 0.8382601141929626:  65%|██████▌   | 4212/6470 [52:18<31:50,  1.18it/s]
4|4213|Loss: 0.8050792813301086:  65%|██████▌   | 4212/6470 [52:19<31:50,  1.18it/s]
4|4213|Loss: 0.8050792813301086:  65%|██████▌   | 4213/6470 [52:19<33:19,  1.13it/s]
4|4214|Loss: 0.7626485824584961:  65%|██████▌   | 4213/6470 [52:20<33:19,  1.13it/s]
4|4214|Loss: 0.7626485824584961:  65%|██████▌   | 4214/6470 [52:20<33:47,  1.11it/s]
4|4215|Loss: 0.7327212691307068:  65%|██████▌   | 4214/6470 [52:21<33:47,  1.11it/s]
4|4215|Loss: 0.7327212691307068:  65%|██████▌   | 4215/6470 [52:21<33:29,  1.12it/s]
4|4216|Loss: 0.7746183276176453:  65%|██████▌   | 4215/6470 [52:22<33:29,  1.12it/s]
4|4216|Loss: 0.7746183276176453:  65%|██████▌   | 4216/6470 [52:22<33:41,  1.11it/s]
4|4217|Loss: 0.7106862664222717:  65%|██████▌   | 4216/6470 [52:22<33:41,  1.11it/s]
4|4217|Loss: 0.7106862664222717:  65%|██████▌   | 4217/6470 [52:23<32:28,  1.16it/s]
4|4218|Loss: 0.7754681706428528:  65%|██████▌   | 4217/6470 [52:23<32:28,  1.16it/s]
4|4218|Loss: 0.7754681706428528:  65%|██████▌   | 4218/6470 [52:23<30:24,  1.23it/s]
4|4219|Loss: 0.8105624318122864:  65%|██████▌   | 4218/6470 [52:24<30:24,  1.23it/s]
4|4219|Loss: 0.8105624318122864:  65%|██████▌   | 4219/6470 [52:24<28:19,  1.32it/s]
4|4220|Loss: 0.8582219481468201:  65%|██████▌   | 4219/6470 [52:24<28:19,  1.32it/s]
4|4220|Loss: 0.8582219481468201:  65%|██████▌   | 4220/6470 [52:25<26:04,  1.44it/s]
4|4221|Loss: 0.8100160360336304:  65%|██████▌   | 4220/6470 [52:25<26:04,  1.44it/s]
4|4221|Loss: 0.8100160360336304:  65%|██████▌   | 4221/6470 [52:26<30:23,  1.23it/s]
4|4222|Loss: 0.8741068243980408:  65%|██████▌   | 4221/6470 [52:26<30:23,  1.23it/s]
4|4222|Loss: 0.8741068243980408:  65%|██████▌   | 4222/6470 [52:27<30:52,  1.21it/s]
4|4223|Loss: 0.8074324131011963:  65%|██████▌   | 4222/6470 [52:27<30:52,  1.21it/s]
4|4223|Loss: 0.8074324131011963:  65%|██████▌   | 4223/6470 [52:27<29:18,  1.28it/s]
4|4224|Loss: 0.6671300530433655:  65%|██████▌   | 4223/6470 [52:28<29:18,  1.28it/s]
4|4224|Loss: 0.6671300530433655:  65%|██████▌   | 4224/6470 [52:28<26:59,  1.39it/s]
4|4225|Loss: 0.6742501854896545:  65%|██████▌   | 4224/6470 [52:28<26:59,  1.39it/s]
4|4225|Loss: 0.6742501854896545:  65%|██████▌   | 4225/6470 [52:28<25:49,  1.45it/s]
4|4226|Loss: 0.8813567161560059:  65%|██████▌   | 4225/6470 [52:29<25:49,  1.45it/s]
4|4226|Loss: 0.8813567161560059:  65%|██████▌   | 4226/6470 [52:29<25:45,  1.45it/s]
4|4227|Loss: 0.8201735019683838:  65%|██████▌   | 4226/6470 [52:29<25:45,  1.45it/s]
4|4227|Loss: 0.8201735019683838:  65%|██████▌   | 4227/6470 [52:30<25:10,  1.49it/s]
4|4228|Loss: 0.8310539722442627:  65%|██████▌   | 4227/6470 [52:30<25:10,  1.49it/s]
4|4228|Loss: 0.8310539722442627:  65%|██████▌   | 4228/6470 [52:30<23:59,  1.56it/s]
4|4229|Loss: 0.891827404499054:  65%|██████▌   | 4228/6470 [52:31<23:59,  1.56it/s] 
4|4229|Loss: 0.891827404499054:  65%|██████▌   | 4229/6470 [52:31<23:51,  1.57it/s]
4|4230|Loss: 0.8595379590988159:  65%|██████▌   | 4229/6470 [52:31<23:51,  1.57it/s]
4|4230|Loss: 0.8595379590988159:  65%|██████▌   | 4230/6470 [52:32<25:07,  1.49it/s]
4|4231|Loss: 0.7478682398796082:  65%|██████▌   | 4230/6470 [52:32<25:07,  1.49it/s]
4|4231|Loss: 0.7478682398796082:  65%|██████▌   | 4231/6470 [52:33<26:29,  1.41it/s]
4|4232|Loss: 0.7788776159286499:  65%|██████▌   | 4231/6470 [52:33<26:29,  1.41it/s]
4|4232|Loss: 0.7788776159286499:  65%|██████▌   | 4232/6470 [52:33<28:47,  1.30it/s]
4|4233|Loss: 0.80153489112854:  65%|██████▌   | 4232/6470 [52:34<28:47,  1.30it/s]  
4|4233|Loss: 0.80153489112854:  65%|██████▌   | 4233/6470 [52:34<26:37,  1.40it/s]
4|4234|Loss: 1.0738197565078735:  65%|██████▌   | 4233/6470 [52:34<26:37,  1.40it/s]
4|4234|Loss: 1.0738197565078735:  65%|██████▌   | 4234/6470 [52:35<27:56,  1.33it/s]
4|4235|Loss: 0.7745985984802246:  65%|██████▌   | 4234/6470 [52:35<27:56,  1.33it/s]
4|4235|Loss: 0.7745985984802246:  65%|██████▌   | 4235/6470 [52:36<28:24,  1.31it/s]
4|4236|Loss: 0.8488754630088806:  65%|██████▌   | 4235/6470 [52:36<28:24,  1.31it/s]
4|4236|Loss: 0.8488754630088806:  65%|██████▌   | 4236/6470 [52:36<29:17,  1.27it/s]
4|4237|Loss: 0.80992192029953:  65%|██████▌   | 4236/6470 [52:37<29:17,  1.27it/s]  
4|4237|Loss: 0.80992192029953:  65%|██████▌   | 4237/6470 [52:37<30:18,  1.23it/s]
4|4238|Loss: 0.7483618855476379:  65%|██████▌   | 4237/6470 [52:38<30:18,  1.23it/s]
4|4238|Loss: 0.7483618855476379:  66%|██████▌   | 4238/6470 [52:38<28:10,  1.32it/s]
4|4239|Loss: 0.7808811664581299:  66%|██████▌   | 4238/6470 [52:38<28:10,  1.32it/s]
4|4239|Loss: 0.7808811664581299:  66%|██████▌   | 4239/6470 [52:39<26:59,  1.38it/s]
4|4240|Loss: 0.7407357692718506:  66%|██████▌   | 4239/6470 [52:39<26:59,  1.38it/s]
4|4240|Loss: 0.7407357692718506:  66%|██████▌   | 4240/6470 [52:39<25:12,  1.47it/s]
4|4241|Loss: 0.8151745200157166:  66%|██████▌   | 4240/6470 [52:40<25:12,  1.47it/s]
4|4241|Loss: 0.8151745200157166:  66%|██████▌   | 4241/6470 [52:40<25:30,  1.46it/s]
4|4242|Loss: 0.9078065752983093:  66%|██████▌   | 4241/6470 [52:40<25:30,  1.46it/s]
4|4242|Loss: 0.9078065752983093:  66%|██████▌   | 4242/6470 [52:41<25:54,  1.43it/s]
4|4243|Loss: 0.9600512385368347:  66%|██████▌   | 4242/6470 [52:41<25:54,  1.43it/s]
4|4243|Loss: 0.9600512385368347:  66%|██████▌   | 4243/6470 [52:42<28:12,  1.32it/s]
4|4244|Loss: 0.8829906582832336:  66%|██████▌   | 4243/6470 [52:42<28:12,  1.32it/s]
4|4244|Loss: 0.8829906582832336:  66%|██████▌   | 4244/6470 [52:42<29:12,  1.27it/s]
4|4245|Loss: 0.9289588928222656:  66%|██████▌   | 4244/6470 [52:43<29:12,  1.27it/s]
4|4245|Loss: 0.9289588928222656:  66%|██████▌   | 4245/6470 [52:43<30:08,  1.23it/s]
4|4246|Loss: 0.7162687182426453:  66%|██████▌   | 4245/6470 [52:44<30:08,  1.23it/s]
4|4246|Loss: 0.7162687182426453:  66%|██████▌   | 4246/6470 [52:44<31:14,  1.19it/s]
4|4247|Loss: 0.7477864623069763:  66%|██████▌   | 4246/6470 [52:45<31:14,  1.19it/s]
4|4247|Loss: 0.7477864623069763:  66%|██████▌   | 4247/6470 [52:45<30:35,  1.21it/s]
4|4248|Loss: 0.8912391662597656:  66%|██████▌   | 4247/6470 [52:45<30:35,  1.21it/s]
4|4248|Loss: 0.8912391662597656:  66%|██████▌   | 4248/6470 [52:46<28:01,  1.32it/s]
4|4249|Loss: 0.734679102897644:  66%|██████▌   | 4248/6470 [52:46<28:01,  1.32it/s] 
4|4249|Loss: 0.734679102897644:  66%|██████▌   | 4249/6470 [52:46<26:39,  1.39it/s]
4|4250|Loss: 0.8114281296730042:  66%|██████▌   | 4249/6470 [52:47<26:39,  1.39it/s]
4|4250|Loss: 0.8114281296730042:  66%|██████▌   | 4250/6470 [52:47<27:20,  1.35it/s]
4|4251|Loss: 0.867504894733429:  66%|██████▌   | 4250/6470 [52:47<27:20,  1.35it/s] 
4|4251|Loss: 0.867504894733429:  66%|██████▌   | 4251/6470 [52:48<26:23,  1.40it/s]
4|4252|Loss: 0.7312275171279907:  66%|██████▌   | 4251/6470 [52:48<26:23,  1.40it/s]
4|4252|Loss: 0.7312275171279907:  66%|██████▌   | 4252/6470 [52:48<25:56,  1.43it/s]
4|4253|Loss: 0.7424562573432922:  66%|██████▌   | 4252/6470 [52:49<25:56,  1.43it/s]
4|4253|Loss: 0.7424562573432922:  66%|██████▌   | 4253/6470 [52:49<25:59,  1.42it/s]
4|4254|Loss: 0.6840143799781799:  66%|██████▌   | 4253/6470 [52:49<25:59,  1.42it/s]
4|4254|Loss: 0.6840143799781799:  66%|██████▌   | 4254/6470 [52:50<25:06,  1.47it/s]
4|4255|Loss: 0.8711048364639282:  66%|██████▌   | 4254/6470 [52:50<25:06,  1.47it/s]
4|4255|Loss: 0.8711048364639282:  66%|██████▌   | 4255/6470 [52:50<25:04,  1.47it/s]
4|4256|Loss: 0.819641649723053:  66%|██████▌   | 4255/6470 [52:51<25:04,  1.47it/s] 
4|4256|Loss: 0.819641649723053:  66%|██████▌   | 4256/6470 [52:51<23:59,  1.54it/s]
4|4257|Loss: 0.7554420828819275:  66%|██████▌   | 4256/6470 [52:51<23:59,  1.54it/s]
4|4257|Loss: 0.7554420828819275:  66%|██████▌   | 4257/6470 [52:51<21:22,  1.73it/s]
4|4258|Loss: 0.8168365359306335:  66%|██████▌   | 4257/6470 [52:52<21:22,  1.73it/s]
4|4258|Loss: 0.8168365359306335:  66%|██████▌   | 4258/6470 [52:52<21:44,  1.70it/s]
4|4259|Loss: 0.7493139505386353:  66%|██████▌   | 4258/6470 [52:52<21:44,  1.70it/s]
4|4259|Loss: 0.7493139505386353:  66%|██████▌   | 4259/6470 [52:53<22:58,  1.60it/s]
4|4260|Loss: 0.7208621501922607:  66%|██████▌   | 4259/6470 [52:53<22:58,  1.60it/s]
4|4260|Loss: 0.7208621501922607:  66%|██████▌   | 4260/6470 [52:53<23:33,  1.56it/s]
4|4261|Loss: 0.7656561732292175:  66%|██████▌   | 4260/6470 [52:54<23:33,  1.56it/s]
4|4261|Loss: 0.7656561732292175:  66%|██████▌   | 4261/6470 [52:54<24:54,  1.48it/s]
4|4262|Loss: 0.8131805062294006:  66%|██████▌   | 4261/6470 [52:54<24:54,  1.48it/s]
4|4262|Loss: 0.8131805062294006:  66%|██████▌   | 4262/6470 [52:55<25:22,  1.45it/s]
4|4263|Loss: 0.8857187628746033:  66%|██████▌   | 4262/6470 [52:55<25:22,  1.45it/s]
4|4263|Loss: 0.8857187628746033:  66%|██████▌   | 4263/6470 [52:56<25:58,  1.42it/s]
4|4264|Loss: 0.8897104263305664:  66%|██████▌   | 4263/6470 [52:56<25:58,  1.42it/s]
4|4264|Loss: 0.8897104263305664:  66%|██████▌   | 4264/6470 [52:56<26:26,  1.39it/s]
4|4265|Loss: 0.6984879374504089:  66%|██████▌   | 4264/6470 [52:57<26:26,  1.39it/s]
4|4265|Loss: 0.6984879374504089:  66%|██████▌   | 4265/6470 [52:57<27:57,  1.31it/s]
4|4266|Loss: 0.9378675818443298:  66%|██████▌   | 4265/6470 [52:58<27:57,  1.31it/s]
4|4266|Loss: 0.9378675818443298:  66%|██████▌   | 4266/6470 [52:58<26:23,  1.39it/s]
4|4267|Loss: 0.808718204498291:  66%|██████▌   | 4266/6470 [52:58<26:23,  1.39it/s] 
4|4267|Loss: 0.808718204498291:  66%|██████▌   | 4267/6470 [52:58<25:47,  1.42it/s]
4|4268|Loss: 0.7091359496116638:  66%|██████▌   | 4267/6470 [52:59<25:47,  1.42it/s]
4|4268|Loss: 0.7091359496116638:  66%|██████▌   | 4268/6470 [52:59<26:40,  1.38it/s]
4|4269|Loss: 0.8034715056419373:  66%|██████▌   | 4268/6470 [53:00<26:40,  1.38it/s]
4|4269|Loss: 0.8034715056419373:  66%|██████▌   | 4269/6470 [53:00<27:14,  1.35it/s]
4|4270|Loss: 0.914085865020752:  66%|██████▌   | 4269/6470 [53:00<27:14,  1.35it/s] 
4|4270|Loss: 0.914085865020752:  66%|██████▌   | 4270/6470 [53:01<27:50,  1.32it/s]
4|4271|Loss: 1.0171024799346924:  66%|██████▌   | 4270/6470 [53:01<27:50,  1.32it/s]
4|4271|Loss: 1.0171024799346924:  66%|██████▌   | 4271/6470 [53:02<29:08,  1.26it/s]
4|4272|Loss: 0.9709709286689758:  66%|██████▌   | 4271/6470 [53:02<29:08,  1.26it/s]
4|4272|Loss: 0.9709709286689758:  66%|██████▌   | 4272/6470 [53:02<29:19,  1.25it/s]
4|4273|Loss: 0.799631655216217:  66%|██████▌   | 4272/6470 [53:03<29:19,  1.25it/s] 
4|4273|Loss: 0.799631655216217:  66%|██████▌   | 4273/6470 [53:03<29:49,  1.23it/s]
4|4274|Loss: 0.8327714204788208:  66%|██████▌   | 4273/6470 [53:04<29:49,  1.23it/s]
4|4274|Loss: 0.8327714204788208:  66%|██████▌   | 4274/6470 [53:04<26:57,  1.36it/s]
4|4275|Loss: 0.8403918743133545:  66%|██████▌   | 4274/6470 [53:04<26:57,  1.36it/s]
4|4275|Loss: 0.8403918743133545:  66%|██████▌   | 4275/6470 [53:04<25:45,  1.42it/s]
4|4276|Loss: 0.7086649537086487:  66%|██████▌   | 4275/6470 [53:05<25:45,  1.42it/s]
4|4276|Loss: 0.7086649537086487:  66%|██████▌   | 4276/6470 [53:05<25:01,  1.46it/s]
4|4277|Loss: 0.8013420701026917:  66%|██████▌   | 4276/6470 [53:06<25:01,  1.46it/s]
4|4277|Loss: 0.8013420701026917:  66%|██████▌   | 4277/6470 [53:06<25:45,  1.42it/s]
4|4278|Loss: 0.7349442839622498:  66%|██████▌   | 4277/6470 [53:06<25:45,  1.42it/s]
4|4278|Loss: 0.7349442839622498:  66%|██████▌   | 4278/6470 [53:07<26:48,  1.36it/s]
4|4279|Loss: 0.8739193677902222:  66%|██████▌   | 4278/6470 [53:07<26:48,  1.36it/s]
4|4279|Loss: 0.8739193677902222:  66%|██████▌   | 4279/6470 [53:07<25:32,  1.43it/s]
4|4280|Loss: 0.7948274612426758:  66%|██████▌   | 4279/6470 [53:08<25:32,  1.43it/s]
4|4280|Loss: 0.7948274612426758:  66%|██████▌   | 4280/6470 [53:08<25:09,  1.45it/s]
4|4281|Loss: 0.8534284830093384:  66%|██████▌   | 4280/6470 [53:08<25:09,  1.45it/s]
4|4281|Loss: 0.8534284830093384:  66%|██████▌   | 4281/6470 [53:09<24:22,  1.50it/s]
4|4282|Loss: 0.6664106845855713:  66%|██████▌   | 4281/6470 [53:09<24:22,  1.50it/s]
4|4282|Loss: 0.6664106845855713:  66%|██████▌   | 4282/6470 [53:09<24:59,  1.46it/s]
4|4283|Loss: 0.9420826435089111:  66%|██████▌   | 4282/6470 [53:10<24:59,  1.46it/s]
4|4283|Loss: 0.9420826435089111:  66%|██████▌   | 4283/6470 [53:10<24:42,  1.48it/s]
4|4284|Loss: 0.8926277756690979:  66%|██████▌   | 4283/6470 [53:10<24:42,  1.48it/s]
4|4284|Loss: 0.8926277756690979:  66%|██████▌   | 4284/6470 [53:11<25:43,  1.42it/s]
4|4285|Loss: 0.8592474460601807:  66%|██████▌   | 4284/6470 [53:11<25:43,  1.42it/s]
4|4285|Loss: 0.8592474460601807:  66%|██████▌   | 4285/6470 [53:12<26:47,  1.36it/s]
4|4286|Loss: 0.7889450192451477:  66%|██████▌   | 4285/6470 [53:12<26:47,  1.36it/s]
4|4286|Loss: 0.7889450192451477:  66%|██████▌   | 4286/6470 [53:12<27:12,  1.34it/s]
4|4287|Loss: 0.8839884996414185:  66%|██████▌   | 4286/6470 [53:13<27:12,  1.34it/s]
4|4287|Loss: 0.8839884996414185:  66%|██████▋   | 4287/6470 [53:13<26:54,  1.35it/s]
4|4288|Loss: 0.7647091746330261:  66%|██████▋   | 4287/6470 [53:14<26:54,  1.35it/s]
4|4288|Loss: 0.7647091746330261:  66%|██████▋   | 4288/6470 [53:14<28:08,  1.29it/s]
4|4289|Loss: 0.908119261264801:  66%|██████▋   | 4288/6470 [53:14<28:08,  1.29it/s] 
4|4289|Loss: 0.908119261264801:  66%|██████▋   | 4289/6470 [53:15<30:21,  1.20it/s]
4|4290|Loss: 0.7281971573829651:  66%|██████▋   | 4289/6470 [53:15<30:21,  1.20it/s]
4|4290|Loss: 0.7281971573829651:  66%|██████▋   | 4290/6470 [53:16<29:30,  1.23it/s]
4|4291|Loss: 0.8621711730957031:  66%|██████▋   | 4290/6470 [53:16<29:30,  1.23it/s]
4|4291|Loss: 0.8621711730957031:  66%|██████▋   | 4291/6470 [53:17<30:56,  1.17it/s]
4|4292|Loss: 0.8372508883476257:  66%|██████▋   | 4291/6470 [53:17<30:56,  1.17it/s]
4|4292|Loss: 0.8372508883476257:  66%|██████▋   | 4292/6470 [53:18<31:34,  1.15it/s]
4|4293|Loss: 0.9591012001037598:  66%|██████▋   | 4292/6470 [53:18<31:34,  1.15it/s]
4|4293|Loss: 0.9591012001037598:  66%|██████▋   | 4293/6470 [53:18<28:13,  1.29it/s]
4|4294|Loss: 0.696058988571167:  66%|██████▋   | 4293/6470 [53:19<28:13,  1.29it/s] 
4|4294|Loss: 0.696058988571167:  66%|██████▋   | 4294/6470 [53:19<29:53,  1.21it/s]
4|4295|Loss: 0.706265389919281:  66%|██████▋   | 4294/6470 [53:19<29:53,  1.21it/s]
4|4295|Loss: 0.706265389919281:  66%|██████▋   | 4295/6470 [53:20<28:36,  1.27it/s]
4|4296|Loss: 0.853817880153656:  66%|██████▋   | 4295/6470 [53:20<28:36,  1.27it/s]
4|4296|Loss: 0.853817880153656:  66%|██████▋   | 4296/6470 [53:21<29:19,  1.24it/s]
4|4297|Loss: 0.800523042678833:  66%|██████▋   | 4296/6470 [53:21<29:19,  1.24it/s]
4|4297|Loss: 0.800523042678833:  66%|██████▋   | 4297/6470 [53:21<29:25,  1.23it/s]
4|4298|Loss: 0.8012287020683289:  66%|██████▋   | 4297/6470 [53:22<29:25,  1.23it/s]
4|4298|Loss: 0.8012287020683289:  66%|██████▋   | 4298/6470 [53:22<29:24,  1.23it/s]
4|4299|Loss: 0.9444029331207275:  66%|██████▋   | 4298/6470 [53:23<29:24,  1.23it/s]
4|4299|Loss: 0.9444029331207275:  66%|██████▋   | 4299/6470 [53:23<27:23,  1.32it/s]
4|4300|Loss: 0.953512966632843:  66%|██████▋   | 4299/6470 [53:23<27:23,  1.32it/s] 
4|4300|Loss: 0.953512966632843:  66%|██████▋   | 4300/6470 [53:24<26:33,  1.36it/s]
4|4301|Loss: 0.9650302529335022:  66%|██████▋   | 4300/6470 [53:24<26:33,  1.36it/s]
4|4301|Loss: 0.9650302529335022:  66%|██████▋   | 4301/6470 [53:24<26:43,  1.35it/s]
4|4302|Loss: 0.7239711284637451:  66%|██████▋   | 4301/6470 [53:25<26:43,  1.35it/s]
4|4302|Loss: 0.7239711284637451:  66%|██████▋   | 4302/6470 [53:25<27:23,  1.32it/s]
4|4303|Loss: 0.7322028875350952:  66%|██████▋   | 4302/6470 [53:26<27:23,  1.32it/s]
4|4303|Loss: 0.7322028875350952:  67%|██████▋   | 4303/6470 [53:26<30:11,  1.20it/s]
4|4304|Loss: 0.9911640286445618:  67%|██████▋   | 4303/6470 [53:27<30:11,  1.20it/s]
4|4304|Loss: 0.9911640286445618:  67%|██████▋   | 4304/6470 [53:27<31:27,  1.15it/s]
4|4305|Loss: 0.8779686093330383:  67%|██████▋   | 4304/6470 [53:27<31:27,  1.15it/s]
4|4305|Loss: 0.8779686093330383:  67%|██████▋   | 4305/6470 [53:28<29:12,  1.24it/s]
4|4306|Loss: 0.8872966766357422:  67%|██████▋   | 4305/6470 [53:28<29:12,  1.24it/s]
4|4306|Loss: 0.8872966766357422:  67%|██████▋   | 4306/6470 [53:28<25:49,  1.40it/s]
4|4307|Loss: 0.715421199798584:  67%|██████▋   | 4306/6470 [53:29<25:49,  1.40it/s] 
4|4307|Loss: 0.715421199798584:  67%|██████▋   | 4307/6470 [53:29<24:34,  1.47it/s]
4|4308|Loss: 0.923490583896637:  67%|██████▋   | 4307/6470 [53:29<24:34,  1.47it/s]
4|4308|Loss: 0.923490583896637:  67%|██████▋   | 4308/6470 [53:30<26:11,  1.38it/s]
4|4309|Loss: 0.661788821220398:  67%|██████▋   | 4308/6470 [53:30<26:11,  1.38it/s]
4|4309|Loss: 0.661788821220398:  67%|██████▋   | 4309/6470 [53:31<28:32,  1.26it/s]
4|4310|Loss: 0.7323022484779358:  67%|██████▋   | 4309/6470 [53:31<28:32,  1.26it/s]
4|4310|Loss: 0.7323022484779358:  67%|██████▋   | 4310/6470 [53:31<27:16,  1.32it/s]
4|4311|Loss: 0.7870035767555237:  67%|██████▋   | 4310/6470 [53:32<27:16,  1.32it/s]
4|4311|Loss: 0.7870035767555237:  67%|██████▋   | 4311/6470 [53:32<26:09,  1.38it/s]
4|4312|Loss: 0.6328088641166687:  67%|██████▋   | 4311/6470 [53:32<26:09,  1.38it/s]
4|4312|Loss: 0.6328088641166687:  67%|██████▋   | 4312/6470 [53:33<25:08,  1.43it/s]
4|4313|Loss: 0.8151506185531616:  67%|██████▋   | 4312/6470 [53:33<25:08,  1.43it/s]
4|4313|Loss: 0.8151506185531616:  67%|██████▋   | 4313/6470 [53:33<24:59,  1.44it/s]
4|4314|Loss: 0.8774005174636841:  67%|██████▋   | 4313/6470 [53:34<24:59,  1.44it/s]
4|4314|Loss: 0.8774005174636841:  67%|██████▋   | 4314/6470 [53:34<25:47,  1.39it/s]
4|4315|Loss: 0.8327237963676453:  67%|██████▋   | 4314/6470 [53:34<25:47,  1.39it/s]
4|4315|Loss: 0.8327237963676453:  67%|██████▋   | 4315/6470 [53:35<26:35,  1.35it/s]
4|4316|Loss: 0.8490939736366272:  67%|██████▋   | 4315/6470 [53:35<26:35,  1.35it/s]
4|4316|Loss: 0.8490939736366272:  67%|██████▋   | 4316/6470 [53:36<27:47,  1.29it/s]
4|4317|Loss: 0.8583608865737915:  67%|██████▋   | 4316/6470 [53:36<27:47,  1.29it/s]
4|4317|Loss: 0.8583608865737915:  67%|██████▋   | 4317/6470 [53:37<29:11,  1.23it/s]
4|4318|Loss: 0.8344889283180237:  67%|██████▋   | 4317/6470 [53:37<29:11,  1.23it/s]
4|4318|Loss: 0.8344889283180237:  67%|██████▋   | 4318/6470 [53:37<29:08,  1.23it/s]
4|4319|Loss: 0.7747777700424194:  67%|██████▋   | 4318/6470 [53:38<29:08,  1.23it/s]
4|4319|Loss: 0.7747777700424194:  67%|██████▋   | 4319/6470 [53:38<28:58,  1.24it/s]
4|4320|Loss: 0.7774291038513184:  67%|██████▋   | 4319/6470 [53:39<28:58,  1.24it/s]
4|4320|Loss: 0.7774291038513184:  67%|██████▋   | 4320/6470 [53:39<28:05,  1.28it/s]
4|4321|Loss: 0.7091662287712097:  67%|██████▋   | 4320/6470 [53:39<28:05,  1.28it/s]
4|4321|Loss: 0.7091662287712097:  67%|██████▋   | 4321/6470 [53:40<26:35,  1.35it/s]
4|4322|Loss: 0.8271065950393677:  67%|██████▋   | 4321/6470 [53:40<26:35,  1.35it/s]
4|4322|Loss: 0.8271065950393677:  67%|██████▋   | 4322/6470 [53:40<27:30,  1.30it/s]
4|4323|Loss: 0.7598943114280701:  67%|██████▋   | 4322/6470 [53:41<27:30,  1.30it/s]
4|4323|Loss: 0.7598943114280701:  67%|██████▋   | 4323/6470 [53:41<27:08,  1.32it/s]
4|4324|Loss: 0.8224668502807617:  67%|██████▋   | 4323/6470 [53:41<27:08,  1.32it/s]
4|4324|Loss: 0.8224668502807617:  67%|██████▋   | 4324/6470 [53:42<26:31,  1.35it/s]
4|4325|Loss: 0.6909672021865845:  67%|██████▋   | 4324/6470 [53:42<26:31,  1.35it/s]
4|4325|Loss: 0.6909672021865845:  67%|██████▋   | 4325/6470 [53:43<27:52,  1.28it/s]
4|4326|Loss: 0.854375422000885:  67%|██████▋   | 4325/6470 [53:43<27:52,  1.28it/s] 
4|4326|Loss: 0.854375422000885:  67%|██████▋   | 4326/6470 [53:43<26:45,  1.34it/s]
4|4327|Loss: 0.7962461113929749:  67%|██████▋   | 4326/6470 [53:44<26:45,  1.34it/s]
4|4327|Loss: 0.7962461113929749:  67%|██████▋   | 4327/6470 [53:44<27:49,  1.28it/s]
4|4328|Loss: 0.7885347604751587:  67%|██████▋   | 4327/6470 [53:45<27:49,  1.28it/s]
4|4328|Loss: 0.7885347604751587:  67%|██████▋   | 4328/6470 [53:45<27:21,  1.30it/s]
4|4329|Loss: 0.7989963889122009:  67%|██████▋   | 4328/6470 [53:45<27:21,  1.30it/s]
4|4329|Loss: 0.7989963889122009:  67%|██████▋   | 4329/6470 [53:46<25:31,  1.40it/s]
4|4330|Loss: 0.8309683203697205:  67%|██████▋   | 4329/6470 [53:46<25:31,  1.40it/s]
4|4330|Loss: 0.8309683203697205:  67%|██████▋   | 4330/6470 [53:46<25:58,  1.37it/s]
4|4331|Loss: 0.7147764563560486:  67%|██████▋   | 4330/6470 [53:47<25:58,  1.37it/s]
4|4331|Loss: 0.7147764563560486:  67%|██████▋   | 4331/6470 [53:48<32:50,  1.09it/s]
4|4332|Loss: 0.8319907784461975:  67%|██████▋   | 4331/6470 [53:48<32:50,  1.09it/s]
4|4332|Loss: 0.8319907784461975:  67%|██████▋   | 4332/6470 [53:49<32:16,  1.10it/s]
4|4333|Loss: 0.840049684047699:  67%|██████▋   | 4332/6470 [53:49<32:16,  1.10it/s] 
4|4333|Loss: 0.840049684047699:  67%|██████▋   | 4333/6470 [53:49<27:56,  1.28it/s]
4|4334|Loss: 0.777683436870575:  67%|██████▋   | 4333/6470 [53:49<27:56,  1.28it/s]
4|4334|Loss: 0.777683436870575:  67%|██████▋   | 4334/6470 [53:50<26:56,  1.32it/s]
4|4335|Loss: 0.8930909633636475:  67%|██████▋   | 4334/6470 [53:50<26:56,  1.32it/s]
4|4335|Loss: 0.8930909633636475:  67%|██████▋   | 4335/6470 [53:50<26:31,  1.34it/s]
4|4336|Loss: 0.7207231521606445:  67%|██████▋   | 4335/6470 [53:51<26:31,  1.34it/s]
4|4336|Loss: 0.7207231521606445:  67%|██████▋   | 4336/6470 [53:51<26:49,  1.33it/s]
4|4337|Loss: 0.9509279727935791:  67%|██████▋   | 4336/6470 [53:52<26:49,  1.33it/s]
4|4337|Loss: 0.9509279727935791:  67%|██████▋   | 4337/6470 [53:52<26:04,  1.36it/s]
4|4338|Loss: 0.8820911645889282:  67%|██████▋   | 4337/6470 [53:52<26:04,  1.36it/s]
4|4338|Loss: 0.8820911645889282:  67%|██████▋   | 4338/6470 [53:53<25:23,  1.40it/s]
4|4339|Loss: 0.7902913689613342:  67%|██████▋   | 4338/6470 [53:53<25:23,  1.40it/s]
4|4339|Loss: 0.7902913689613342:  67%|██████▋   | 4339/6470 [53:53<25:58,  1.37it/s]
4|4340|Loss: 0.8415678143501282:  67%|██████▋   | 4339/6470 [53:54<25:58,  1.37it/s]
4|4340|Loss: 0.8415678143501282:  67%|██████▋   | 4340/6470 [53:54<27:09,  1.31it/s]
4|4341|Loss: 0.7659703493118286:  67%|██████▋   | 4340/6470 [53:55<27:09,  1.31it/s]
4|4341|Loss: 0.7659703493118286:  67%|██████▋   | 4341/6470 [53:55<27:45,  1.28it/s]
4|4342|Loss: 0.9530956745147705:  67%|██████▋   | 4341/6470 [53:55<27:45,  1.28it/s]
4|4342|Loss: 0.9530956745147705:  67%|██████▋   | 4342/6470 [53:56<28:52,  1.23it/s]
4|4343|Loss: 0.9799960255622864:  67%|██████▋   | 4342/6470 [53:56<28:52,  1.23it/s]
4|4343|Loss: 0.9799960255622864:  67%|██████▋   | 4343/6470 [53:57<29:32,  1.20it/s]
4|4344|Loss: 0.7624150514602661:  67%|██████▋   | 4343/6470 [53:57<29:32,  1.20it/s]
4|4344|Loss: 0.7624150514602661:  67%|██████▋   | 4344/6470 [53:58<28:49,  1.23it/s]
4|4345|Loss: 0.7660589814186096:  67%|██████▋   | 4344/6470 [53:58<28:49,  1.23it/s]
4|4345|Loss: 0.7660589814186096:  67%|██████▋   | 4345/6470 [53:58<27:29,  1.29it/s]
4|4346|Loss: 0.7114709615707397:  67%|██████▋   | 4345/6470 [53:59<27:29,  1.29it/s]
4|4346|Loss: 0.7114709615707397:  67%|██████▋   | 4346/6470 [53:59<28:46,  1.23it/s]
4|4347|Loss: 0.8013265132904053:  67%|██████▋   | 4346/6470 [54:00<28:46,  1.23it/s]
4|4347|Loss: 0.8013265132904053:  67%|██████▋   | 4347/6470 [54:00<28:15,  1.25it/s]
4|4348|Loss: 0.8633559346199036:  67%|██████▋   | 4347/6470 [54:00<28:15,  1.25it/s]
4|4348|Loss: 0.8633559346199036:  67%|██████▋   | 4348/6470 [54:00<25:24,  1.39it/s]
4|4349|Loss: 0.7752874493598938:  67%|██████▋   | 4348/6470 [54:01<25:24,  1.39it/s]
4|4349|Loss: 0.7752874493598938:  67%|██████▋   | 4349/6470 [54:01<24:27,  1.44it/s]
4|4350|Loss: 0.6914770007133484:  67%|██████▋   | 4349/6470 [54:01<24:27,  1.44it/s]
4|4350|Loss: 0.6914770007133484:  67%|██████▋   | 4350/6470 [54:02<25:05,  1.41it/s]
4|4351|Loss: 0.8715469241142273:  67%|██████▋   | 4350/6470 [54:02<25:05,  1.41it/s]
4|4351|Loss: 0.8715469241142273:  67%|██████▋   | 4351/6470 [54:03<26:36,  1.33it/s]
4|4352|Loss: 0.8060598969459534:  67%|██████▋   | 4351/6470 [54:03<26:36,  1.33it/s]
4|4352|Loss: 0.8060598969459534:  67%|██████▋   | 4352/6470 [54:03<26:07,  1.35it/s]
4|4353|Loss: 0.9449293613433838:  67%|██████▋   | 4352/6470 [54:04<26:07,  1.35it/s]
4|4353|Loss: 0.9449293613433838:  67%|██████▋   | 4353/6470 [54:04<27:17,  1.29it/s]
4|4354|Loss: 0.8884616494178772:  67%|██████▋   | 4353/6470 [54:05<27:17,  1.29it/s]
4|4354|Loss: 0.8884616494178772:  67%|██████▋   | 4354/6470 [54:05<27:34,  1.28it/s]
4|4355|Loss: 0.9641991257667542:  67%|██████▋   | 4354/6470 [54:05<27:34,  1.28it/s]
4|4355|Loss: 0.9641991257667542:  67%|██████▋   | 4355/6470 [54:06<27:40,  1.27it/s]
4|4356|Loss: 0.782593309879303:  67%|██████▋   | 4355/6470 [54:06<27:40,  1.27it/s] 
4|4356|Loss: 0.782593309879303:  67%|██████▋   | 4356/6470 [54:07<26:58,  1.31it/s]
4|4357|Loss: 0.879692554473877:  67%|██████▋   | 4356/6470 [54:07<26:58,  1.31it/s]
4|4357|Loss: 0.879692554473877:  67%|██████▋   | 4357/6470 [54:07<25:21,  1.39it/s]
4|4358|Loss: 0.8488851189613342:  67%|██████▋   | 4357/6470 [54:07<25:21,  1.39it/s]
4|4358|Loss: 0.8488851189613342:  67%|██████▋   | 4358/6470 [54:08<23:04,  1.53it/s]
4|4359|Loss: 0.9035617709159851:  67%|██████▋   | 4358/6470 [54:08<23:04,  1.53it/s]
4|4359|Loss: 0.9035617709159851:  67%|██████▋   | 4359/6470 [54:08<22:57,  1.53it/s]
4|4360|Loss: 0.8046563863754272:  67%|██████▋   | 4359/6470 [54:09<22:57,  1.53it/s]
4|4360|Loss: 0.8046563863754272:  67%|██████▋   | 4360/6470 [54:09<24:07,  1.46it/s]
4|4361|Loss: 0.7868151068687439:  67%|██████▋   | 4360/6470 [54:10<24:07,  1.46it/s]
4|4361|Loss: 0.7868151068687439:  67%|██████▋   | 4361/6470 [54:10<27:22,  1.28it/s]
4|4362|Loss: 1.0248124599456787:  67%|██████▋   | 4361/6470 [54:11<27:22,  1.28it/s]
4|4362|Loss: 1.0248124599456787:  67%|██████▋   | 4362/6470 [54:11<29:15,  1.20it/s]
4|4363|Loss: 0.8378868699073792:  67%|██████▋   | 4362/6470 [54:11<29:15,  1.20it/s]
4|4363|Loss: 0.8378868699073792:  67%|██████▋   | 4363/6470 [54:12<29:10,  1.20it/s]
4|4364|Loss: 0.7352920174598694:  67%|██████▋   | 4363/6470 [54:12<29:10,  1.20it/s]
4|4364|Loss: 0.7352920174598694:  67%|██████▋   | 4364/6470 [54:13<27:43,  1.27it/s]
4|4365|Loss: 0.8622967004776001:  67%|██████▋   | 4364/6470 [54:13<27:43,  1.27it/s]
4|4365|Loss: 0.8622967004776001:  67%|██████▋   | 4365/6470 [54:13<28:13,  1.24it/s]
4|4366|Loss: 0.8551510572433472:  67%|██████▋   | 4365/6470 [54:14<28:13,  1.24it/s]
4|4366|Loss: 0.8551510572433472:  67%|██████▋   | 4366/6470 [54:14<28:20,  1.24it/s]
4|4367|Loss: 0.7219769954681396:  67%|██████▋   | 4366/6470 [54:15<28:20,  1.24it/s]
4|4367|Loss: 0.7219769954681396:  67%|██████▋   | 4367/6470 [54:15<27:28,  1.28it/s]
4|4368|Loss: 0.8844730854034424:  67%|██████▋   | 4367/6470 [54:15<27:28,  1.28it/s]
4|4368|Loss: 0.8844730854034424:  68%|██████▊   | 4368/6470 [54:16<26:21,  1.33it/s]
4|4369|Loss: 0.7137640118598938:  68%|██████▊   | 4368/6470 [54:16<26:21,  1.33it/s]
4|4369|Loss: 0.7137640118598938:  68%|██████▊   | 4369/6470 [54:16<26:45,  1.31it/s]
4|4370|Loss: 0.8025336861610413:  68%|██████▊   | 4369/6470 [54:17<26:45,  1.31it/s]
4|4370|Loss: 0.8025336861610413:  68%|██████▊   | 4370/6470 [54:17<25:46,  1.36it/s]
4|4371|Loss: 0.7097359895706177:  68%|██████▊   | 4370/6470 [54:18<25:46,  1.36it/s]
4|4371|Loss: 0.7097359895706177:  68%|██████▊   | 4371/6470 [54:18<27:24,  1.28it/s]
4|4372|Loss: 0.8414971828460693:  68%|██████▊   | 4371/6470 [54:18<27:24,  1.28it/s]
4|4372|Loss: 0.8414971828460693:  68%|██████▊   | 4372/6470 [54:19<27:12,  1.29it/s]
4|4373|Loss: 0.9076024293899536:  68%|██████▊   | 4372/6470 [54:19<27:12,  1.29it/s]
4|4373|Loss: 0.9076024293899536:  68%|██████▊   | 4373/6470 [54:19<26:07,  1.34it/s]
4|4374|Loss: 0.861672580242157:  68%|██████▊   | 4373/6470 [54:20<26:07,  1.34it/s] 
4|4374|Loss: 0.861672580242157:  68%|██████▊   | 4374/6470 [54:20<26:28,  1.32it/s]
4|4375|Loss: 0.8666542172431946:  68%|██████▊   | 4374/6470 [54:21<26:28,  1.32it/s]
4|4375|Loss: 0.8666542172431946:  68%|██████▊   | 4375/6470 [54:21<25:32,  1.37it/s]
4|4376|Loss: 1.0296145677566528:  68%|██████▊   | 4375/6470 [54:21<25:32,  1.37it/s]
4|4376|Loss: 1.0296145677566528:  68%|██████▊   | 4376/6470 [54:21<23:12,  1.50it/s]
4|4377|Loss: 0.7320254445075989:  68%|██████▊   | 4376/6470 [54:22<23:12,  1.50it/s]
4|4377|Loss: 0.7320254445075989:  68%|██████▊   | 4377/6470 [54:22<23:33,  1.48it/s]
4|4378|Loss: 0.9126193523406982:  68%|██████▊   | 4377/6470 [54:22<23:33,  1.48it/s]
4|4378|Loss: 0.9126193523406982:  68%|██████▊   | 4378/6470 [54:23<24:05,  1.45it/s]
4|4379|Loss: 0.8584603667259216:  68%|██████▊   | 4378/6470 [54:23<24:05,  1.45it/s]
4|4379|Loss: 0.8584603667259216:  68%|██████▊   | 4379/6470 [54:24<25:57,  1.34it/s]
4|4380|Loss: 0.8604235649108887:  68%|██████▊   | 4379/6470 [54:24<25:57,  1.34it/s]
4|4380|Loss: 0.8604235649108887:  68%|██████▊   | 4380/6470 [54:24<25:09,  1.38it/s]
4|4381|Loss: 0.7717470526695251:  68%|██████▊   | 4380/6470 [54:25<25:09,  1.38it/s]
4|4381|Loss: 0.7717470526695251:  68%|██████▊   | 4381/6470 [54:25<26:08,  1.33it/s]
4|4382|Loss: 0.7893360257148743:  68%|██████▊   | 4381/6470 [54:26<26:08,  1.33it/s]
4|4382|Loss: 0.7893360257148743:  68%|██████▊   | 4382/6470 [54:26<25:20,  1.37it/s]
4|4383|Loss: 0.7125232815742493:  68%|██████▊   | 4382/6470 [54:26<25:20,  1.37it/s]
4|4383|Loss: 0.7125232815742493:  68%|██████▊   | 4383/6470 [54:27<27:10,  1.28it/s]
4|4384|Loss: 0.6863676309585571:  68%|██████▊   | 4383/6470 [54:27<27:10,  1.28it/s]
4|4384|Loss: 0.6863676309585571:  68%|██████▊   | 4384/6470 [54:27<25:09,  1.38it/s]
4|4385|Loss: 0.8537406325340271:  68%|██████▊   | 4384/6470 [54:28<25:09,  1.38it/s]
4|4385|Loss: 0.8537406325340271:  68%|██████▊   | 4385/6470 [54:28<24:02,  1.45it/s]
4|4386|Loss: 0.7516276240348816:  68%|██████▊   | 4385/6470 [54:28<24:02,  1.45it/s]
4|4386|Loss: 0.7516276240348816:  68%|██████▊   | 4386/6470 [54:29<25:05,  1.38it/s]
4|4387|Loss: 0.6345635652542114:  68%|██████▊   | 4386/6470 [54:29<25:05,  1.38it/s]
4|4387|Loss: 0.6345635652542114:  68%|██████▊   | 4387/6470 [54:29<24:53,  1.40it/s]
4|4388|Loss: 0.8906633257865906:  68%|██████▊   | 4387/6470 [54:30<24:53,  1.40it/s]
4|4388|Loss: 0.8906633257865906:  68%|██████▊   | 4388/6470 [54:30<26:36,  1.30it/s]
4|4389|Loss: 0.8739723563194275:  68%|██████▊   | 4388/6470 [54:31<26:36,  1.30it/s]
4|4389|Loss: 0.8739723563194275:  68%|██████▊   | 4389/6470 [54:31<27:20,  1.27it/s]
4|4390|Loss: 0.8776713013648987:  68%|██████▊   | 4389/6470 [54:32<27:20,  1.27it/s]
4|4390|Loss: 0.8776713013648987:  68%|██████▊   | 4390/6470 [54:32<26:02,  1.33it/s]
4|4391|Loss: 0.8198016881942749:  68%|██████▊   | 4390/6470 [54:32<26:02,  1.33it/s]
4|4391|Loss: 0.8198016881942749:  68%|██████▊   | 4391/6470 [54:33<26:22,  1.31it/s]
4|4392|Loss: 0.7615950703620911:  68%|██████▊   | 4391/6470 [54:33<26:22,  1.31it/s]
4|4392|Loss: 0.7615950703620911:  68%|██████▊   | 4392/6470 [54:33<24:56,  1.39it/s]
4|4393|Loss: 0.8057515621185303:  68%|██████▊   | 4392/6470 [54:34<24:56,  1.39it/s]
4|4393|Loss: 0.8057515621185303:  68%|██████▊   | 4393/6470 [54:34<25:28,  1.36it/s]
4|4394|Loss: 0.7175894975662231:  68%|██████▊   | 4393/6470 [54:34<25:28,  1.36it/s]
4|4394|Loss: 0.7175894975662231:  68%|██████▊   | 4394/6470 [54:35<26:04,  1.33it/s]
4|4395|Loss: 0.7819405198097229:  68%|██████▊   | 4394/6470 [54:35<26:04,  1.33it/s]
4|4395|Loss: 0.7819405198097229:  68%|██████▊   | 4395/6470 [54:36<27:42,  1.25it/s]
4|4396|Loss: 0.7637912631034851:  68%|██████▊   | 4395/6470 [54:36<27:42,  1.25it/s]
4|4396|Loss: 0.7637912631034851:  68%|██████▊   | 4396/6470 [54:36<27:21,  1.26it/s]
4|4397|Loss: 0.9261909127235413:  68%|██████▊   | 4396/6470 [54:37<27:21,  1.26it/s]
4|4397|Loss: 0.9261909127235413:  68%|██████▊   | 4397/6470 [54:37<26:16,  1.32it/s]
4|4398|Loss: 0.83744877576828:  68%|██████▊   | 4397/6470 [54:38<26:16,  1.32it/s]  
4|4398|Loss: 0.83744877576828:  68%|██████▊   | 4398/6470 [54:38<26:14,  1.32it/s]
4|4399|Loss: 0.7087584733963013:  68%|██████▊   | 4398/6470 [54:38<26:14,  1.32it/s]
4|4399|Loss: 0.7087584733963013:  68%|██████▊   | 4399/6470 [54:39<27:11,  1.27it/s]
4|4400|Loss: 0.9492576718330383:  68%|██████▊   | 4399/6470 [54:39<27:11,  1.27it/s]
4|4400|Loss: 0.9492576718330383:  68%|██████▊   | 4400/6470 [54:39<26:17,  1.31it/s]
4|4401|Loss: 0.7463572025299072:  68%|██████▊   | 4400/6470 [54:40<26:17,  1.31it/s]
4|4401|Loss: 0.7463572025299072:  68%|██████▊   | 4401/6470 [54:40<26:57,  1.28it/s]
4|4402|Loss: 0.8111140727996826:  68%|██████▊   | 4401/6470 [54:41<26:57,  1.28it/s]
4|4402|Loss: 0.8111140727996826:  68%|██████▊   | 4402/6470 [54:41<28:32,  1.21it/s]
4|4403|Loss: 0.8608412742614746:  68%|██████▊   | 4402/6470 [54:42<28:32,  1.21it/s]
4|4403|Loss: 0.8608412742614746:  68%|██████▊   | 4403/6470 [54:42<27:25,  1.26it/s]
4|4404|Loss: 0.8174670934677124:  68%|██████▊   | 4403/6470 [54:42<27:25,  1.26it/s]
4|4404|Loss: 0.8174670934677124:  68%|██████▊   | 4404/6470 [54:42<24:40,  1.40it/s]
4|4405|Loss: 0.8316671848297119:  68%|██████▊   | 4404/6470 [54:43<24:40,  1.40it/s]
4|4405|Loss: 0.8316671848297119:  68%|██████▊   | 4405/6470 [54:43<24:55,  1.38it/s]
4|4406|Loss: 0.7773349285125732:  68%|██████▊   | 4405/6470 [54:44<24:55,  1.38it/s]
4|4406|Loss: 0.7773349285125732:  68%|██████▊   | 4406/6470 [54:44<24:03,  1.43it/s]
4|4407|Loss: 0.9364826083183289:  68%|██████▊   | 4406/6470 [54:44<24:03,  1.43it/s]
4|4407|Loss: 0.9364826083183289:  68%|██████▊   | 4407/6470 [54:45<23:46,  1.45it/s]
4|4408|Loss: 0.7860915660858154:  68%|██████▊   | 4407/6470 [54:45<23:46,  1.45it/s]
4|4408|Loss: 0.7860915660858154:  68%|██████▊   | 4408/6470 [54:45<24:21,  1.41it/s]
4|4409|Loss: 0.9103344082832336:  68%|██████▊   | 4408/6470 [54:46<24:21,  1.41it/s]
4|4409|Loss: 0.9103344082832336:  68%|██████▊   | 4409/6470 [54:46<26:11,  1.31it/s]
4|4410|Loss: 0.8829951286315918:  68%|██████▊   | 4409/6470 [54:47<26:11,  1.31it/s]
4|4410|Loss: 0.8829951286315918:  68%|██████▊   | 4410/6470 [54:47<25:40,  1.34it/s]
4|4411|Loss: 0.7615828514099121:  68%|██████▊   | 4410/6470 [54:47<25:40,  1.34it/s]
4|4411|Loss: 0.7615828514099121:  68%|██████▊   | 4411/6470 [54:48<24:24,  1.41it/s]
4|4412|Loss: 0.7142295241355896:  68%|██████▊   | 4411/6470 [54:48<24:24,  1.41it/s]
4|4412|Loss: 0.7142295241355896:  68%|██████▊   | 4412/6470 [54:48<26:06,  1.31it/s]
4|4413|Loss: 0.7553466558456421:  68%|██████▊   | 4412/6470 [54:49<26:06,  1.31it/s]
4|4413|Loss: 0.7553466558456421:  68%|██████▊   | 4413/6470 [54:49<26:56,  1.27it/s]
4|4414|Loss: 0.9688672423362732:  68%|██████▊   | 4413/6470 [54:50<26:56,  1.27it/s]
4|4414|Loss: 0.9688672423362732:  68%|██████▊   | 4414/6470 [54:50<27:03,  1.27it/s]
4|4415|Loss: 0.932243287563324:  68%|██████▊   | 4414/6470 [54:50<27:03,  1.27it/s] 
4|4415|Loss: 0.932243287563324:  68%|██████▊   | 4415/6470 [54:51<24:57,  1.37it/s]
4|4416|Loss: 0.9308779239654541:  68%|██████▊   | 4415/6470 [54:51<24:57,  1.37it/s]
4|4416|Loss: 0.9308779239654541:  68%|██████▊   | 4416/6470 [54:51<25:17,  1.35it/s]
4|4417|Loss: 0.6760098338127136:  68%|██████▊   | 4416/6470 [54:52<25:17,  1.35it/s]
4|4417|Loss: 0.6760098338127136:  68%|██████▊   | 4417/6470 [54:52<28:27,  1.20it/s]
4|4418|Loss: 0.747180700302124:  68%|██████▊   | 4417/6470 [54:53<28:27,  1.20it/s] 
4|4418|Loss: 0.747180700302124:  68%|██████▊   | 4418/6470 [54:53<27:11,  1.26it/s]
4|4419|Loss: 0.9768121838569641:  68%|██████▊   | 4418/6470 [54:54<27:11,  1.26it/s]
4|4419|Loss: 0.9768121838569641:  68%|██████▊   | 4419/6470 [54:54<26:00,  1.31it/s]
4|4420|Loss: 0.7899705171585083:  68%|██████▊   | 4419/6470 [54:54<26:00,  1.31it/s]
4|4420|Loss: 0.7899705171585083:  68%|██████▊   | 4420/6470 [54:55<26:47,  1.28it/s]
4|4421|Loss: 0.8614264726638794:  68%|██████▊   | 4420/6470 [54:55<26:47,  1.28it/s]
4|4421|Loss: 0.8614264726638794:  68%|██████▊   | 4421/6470 [54:55<27:08,  1.26it/s]
4|4422|Loss: 0.7720862030982971:  68%|██████▊   | 4421/6470 [54:56<27:08,  1.26it/s]
4|4422|Loss: 0.7720862030982971:  68%|██████▊   | 4422/6470 [54:56<27:19,  1.25it/s]
4|4423|Loss: 0.7778264284133911:  68%|██████▊   | 4422/6470 [54:57<27:19,  1.25it/s]
4|4423|Loss: 0.7778264284133911:  68%|██████▊   | 4423/6470 [54:57<27:22,  1.25it/s]
4|4424|Loss: 0.7235406637191772:  68%|██████▊   | 4423/6470 [54:58<27:22,  1.25it/s]
4|4424|Loss: 0.7235406637191772:  68%|██████▊   | 4424/6470 [54:58<27:57,  1.22it/s]
4|4425|Loss: 0.8429204821586609:  68%|██████▊   | 4424/6470 [54:58<27:57,  1.22it/s]
4|4425|Loss: 0.8429204821586609:  68%|██████▊   | 4425/6470 [54:59<26:56,  1.27it/s]
4|4426|Loss: 1.0632474422454834:  68%|██████▊   | 4425/6470 [54:59<26:56,  1.27it/s]
4|4426|Loss: 1.0632474422454834:  68%|██████▊   | 4426/6470 [54:59<25:32,  1.33it/s]
4|4427|Loss: 0.943875789642334:  68%|██████▊   | 4426/6470 [55:00<25:32,  1.33it/s] 
4|4427|Loss: 0.943875789642334:  68%|██████▊   | 4427/6470 [55:00<25:38,  1.33it/s]
4|4428|Loss: 0.9168223738670349:  68%|██████▊   | 4427/6470 [55:00<25:38,  1.33it/s]
4|4428|Loss: 0.9168223738670349:  68%|██████▊   | 4428/6470 [55:01<24:22,  1.40it/s]
4|4429|Loss: 0.8682217001914978:  68%|██████▊   | 4428/6470 [55:01<24:22,  1.40it/s]
4|4429|Loss: 0.8682217001914978:  68%|██████▊   | 4429/6470 [55:01<24:36,  1.38it/s]
4|4430|Loss: 0.717609167098999:  68%|██████▊   | 4429/6470 [55:02<24:36,  1.38it/s] 
4|4430|Loss: 0.717609167098999:  68%|██████▊   | 4430/6470 [55:02<23:25,  1.45it/s]
4|4431|Loss: 0.8815630674362183:  68%|██████▊   | 4430/6470 [55:02<23:25,  1.45it/s]
4|4431|Loss: 0.8815630674362183:  68%|██████▊   | 4431/6470 [55:03<23:52,  1.42it/s]
4|4432|Loss: 0.7664377689361572:  68%|██████▊   | 4431/6470 [55:03<23:52,  1.42it/s]
4|4432|Loss: 0.7664377689361572:  69%|██████▊   | 4432/6470 [55:04<25:48,  1.32it/s]
4|4433|Loss: 0.8261640071868896:  69%|██████▊   | 4432/6470 [55:04<25:48,  1.32it/s]
4|4433|Loss: 0.8261640071868896:  69%|██████▊   | 4433/6470 [55:05<26:13,  1.29it/s]
4|4434|Loss: 0.8620064854621887:  69%|██████▊   | 4433/6470 [55:05<26:13,  1.29it/s]
4|4434|Loss: 0.8620064854621887:  69%|██████▊   | 4434/6470 [55:05<25:06,  1.35it/s]
4|4435|Loss: 0.7811542749404907:  69%|██████▊   | 4434/6470 [55:06<25:06,  1.35it/s]
4|4435|Loss: 0.7811542749404907:  69%|██████▊   | 4435/6470 [55:06<24:14,  1.40it/s]
4|4436|Loss: 0.8561877012252808:  69%|██████▊   | 4435/6470 [55:06<24:14,  1.40it/s]
4|4436|Loss: 0.8561877012252808:  69%|██████▊   | 4436/6470 [55:06<22:33,  1.50it/s]
4|4437|Loss: 0.9267905354499817:  69%|██████▊   | 4436/6470 [55:07<22:33,  1.50it/s]
4|4437|Loss: 0.9267905354499817:  69%|██████▊   | 4437/6470 [55:07<22:56,  1.48it/s]
4|4438|Loss: 0.7821186184883118:  69%|██████▊   | 4437/6470 [55:07<22:56,  1.48it/s]
4|4438|Loss: 0.7821186184883118:  69%|██████▊   | 4438/6470 [55:08<22:50,  1.48it/s]
4|4439|Loss: 0.8896688222885132:  69%|██████▊   | 4438/6470 [55:08<22:50,  1.48it/s]
4|4439|Loss: 0.8896688222885132:  69%|██████▊   | 4439/6470 [55:08<22:57,  1.47it/s]
4|4440|Loss: 0.8862589597702026:  69%|██████▊   | 4439/6470 [55:09<22:57,  1.47it/s]
4|4440|Loss: 0.8862589597702026:  69%|██████▊   | 4440/6470 [55:09<24:44,  1.37it/s]
4|4441|Loss: 0.7465388178825378:  69%|██████▊   | 4440/6470 [55:10<24:44,  1.37it/s]
4|4441|Loss: 0.7465388178825378:  69%|██████▊   | 4441/6470 [55:10<26:28,  1.28it/s]
4|4442|Loss: 0.9247199892997742:  69%|██████▊   | 4441/6470 [55:11<26:28,  1.28it/s]
4|4442|Loss: 0.9247199892997742:  69%|██████▊   | 4442/6470 [55:11<26:51,  1.26it/s]
4|4443|Loss: 1.0722416639328003:  69%|██████▊   | 4442/6470 [55:11<26:51,  1.26it/s]
4|4443|Loss: 1.0722416639328003:  69%|██████▊   | 4443/6470 [55:12<26:12,  1.29it/s]
4|4444|Loss: 0.7734898924827576:  69%|██████▊   | 4443/6470 [55:12<26:12,  1.29it/s]
4|4444|Loss: 0.7734898924827576:  69%|██████▊   | 4444/6470 [55:12<25:31,  1.32it/s]
4|4445|Loss: 0.7765000462532043:  69%|██████▊   | 4444/6470 [55:13<25:31,  1.32it/s]
4|4445|Loss: 0.7765000462532043:  69%|██████▊   | 4445/6470 [55:13<25:43,  1.31it/s]
4|4446|Loss: 0.8785582184791565:  69%|██████▊   | 4445/6470 [55:14<25:43,  1.31it/s]
4|4446|Loss: 0.8785582184791565:  69%|██████▊   | 4446/6470 [55:14<25:56,  1.30it/s]
4|4447|Loss: 0.7705851793289185:  69%|██████▊   | 4446/6470 [55:14<25:56,  1.30it/s]
4|4447|Loss: 0.7705851793289185:  69%|██████▊   | 4447/6470 [55:15<24:12,  1.39it/s]
4|4448|Loss: 0.7502010464668274:  69%|██████▊   | 4447/6470 [55:15<24:12,  1.39it/s]
4|4448|Loss: 0.7502010464668274:  69%|██████▊   | 4448/6470 [55:15<24:13,  1.39it/s]
4|4449|Loss: 0.8956161141395569:  69%|██████▊   | 4448/6470 [55:16<24:13,  1.39it/s]
4|4449|Loss: 0.8956161141395569:  69%|██████▉   | 4449/6470 [55:16<23:54,  1.41it/s]
4|4450|Loss: 0.8899484872817993:  69%|██████▉   | 4449/6470 [55:16<23:54,  1.41it/s]
4|4450|Loss: 0.8899484872817993:  69%|██████▉   | 4450/6470 [55:17<25:05,  1.34it/s]
4|4451|Loss: 0.8873462080955505:  69%|██████▉   | 4450/6470 [55:17<25:05,  1.34it/s]
4|4451|Loss: 0.8873462080955505:  69%|██████▉   | 4451/6470 [55:18<24:37,  1.37it/s]
4|4452|Loss: 0.9554363489151001:  69%|██████▉   | 4451/6470 [55:18<24:37,  1.37it/s]
4|4452|Loss: 0.9554363489151001:  69%|██████▉   | 4452/6470 [55:18<24:06,  1.40it/s]
4|4453|Loss: 0.9206279516220093:  69%|██████▉   | 4452/6470 [55:19<24:06,  1.40it/s]
4|4453|Loss: 0.9206279516220093:  69%|██████▉   | 4453/6470 [55:19<23:51,  1.41it/s]
4|4454|Loss: 0.7471258640289307:  69%|██████▉   | 4453/6470 [55:19<23:51,  1.41it/s]
4|4454|Loss: 0.7471258640289307:  69%|██████▉   | 4454/6470 [55:20<23:18,  1.44it/s]
4|4455|Loss: 0.7924883365631104:  69%|██████▉   | 4454/6470 [55:20<23:18,  1.44it/s]
4|4455|Loss: 0.7924883365631104:  69%|██████▉   | 4455/6470 [55:20<25:00,  1.34it/s]
4|4456|Loss: 0.7796927094459534:  69%|██████▉   | 4455/6470 [55:21<25:00,  1.34it/s]
4|4456|Loss: 0.7796927094459534:  69%|██████▉   | 4456/6470 [55:21<25:33,  1.31it/s]
4|4457|Loss: 0.7567019462585449:  69%|██████▉   | 4456/6470 [55:22<25:33,  1.31it/s]
4|4457|Loss: 0.7567019462585449:  69%|██████▉   | 4457/6470 [55:22<26:35,  1.26it/s]
4|4458|Loss: 0.8282548189163208:  69%|██████▉   | 4457/6470 [55:23<26:35,  1.26it/s]
4|4458|Loss: 0.8282548189163208:  69%|██████▉   | 4458/6470 [55:23<26:25,  1.27it/s]
4|4459|Loss: 0.7787111401557922:  69%|██████▉   | 4458/6470 [55:23<26:25,  1.27it/s]
4|4459|Loss: 0.7787111401557922:  69%|██████▉   | 4459/6470 [55:24<25:34,  1.31it/s]
4|4460|Loss: 0.8035348653793335:  69%|██████▉   | 4459/6470 [55:24<25:34,  1.31it/s]
4|4460|Loss: 0.8035348653793335:  69%|██████▉   | 4460/6470 [55:24<24:44,  1.35it/s]
4|4461|Loss: 0.8470991849899292:  69%|██████▉   | 4460/6470 [55:25<24:44,  1.35it/s]
4|4461|Loss: 0.8470991849899292:  69%|██████▉   | 4461/6470 [55:25<25:49,  1.30it/s]
4|4462|Loss: 0.860639214515686:  69%|██████▉   | 4461/6470 [55:26<25:49,  1.30it/s] 
4|4462|Loss: 0.860639214515686:  69%|██████▉   | 4462/6470 [55:26<25:17,  1.32it/s]
4|4463|Loss: 0.7951902151107788:  69%|██████▉   | 4462/6470 [55:26<25:17,  1.32it/s]
4|4463|Loss: 0.7951902151107788:  69%|██████▉   | 4463/6470 [55:26<23:10,  1.44it/s]
4|4464|Loss: 0.8650702834129333:  69%|██████▉   | 4463/6470 [55:27<23:10,  1.44it/s]
4|4464|Loss: 0.8650702834129333:  69%|██████▉   | 4464/6470 [55:27<24:59,  1.34it/s]
4|4465|Loss: 0.7590121626853943:  69%|██████▉   | 4464/6470 [55:28<24:59,  1.34it/s]
4|4465|Loss: 0.7590121626853943:  69%|██████▉   | 4465/6470 [55:28<27:38,  1.21it/s]
4|4466|Loss: 0.9882664084434509:  69%|██████▉   | 4465/6470 [55:29<27:38,  1.21it/s]
4|4466|Loss: 0.9882664084434509:  69%|██████▉   | 4466/6470 [55:29<25:18,  1.32it/s]
4|4467|Loss: 0.6847381591796875:  69%|██████▉   | 4466/6470 [55:29<25:18,  1.32it/s]
4|4467|Loss: 0.6847381591796875:  69%|██████▉   | 4467/6470 [55:30<24:47,  1.35it/s]
4|4468|Loss: 0.6869029998779297:  69%|██████▉   | 4467/6470 [55:30<24:47,  1.35it/s]
4|4468|Loss: 0.6869029998779297:  69%|██████▉   | 4468/6470 [55:30<26:08,  1.28it/s]
4|4469|Loss: 0.8115434646606445:  69%|██████▉   | 4468/6470 [55:31<26:08,  1.28it/s]
4|4469|Loss: 0.8115434646606445:  69%|██████▉   | 4469/6470 [55:31<27:46,  1.20it/s]
4|4470|Loss: 0.7818350195884705:  69%|██████▉   | 4469/6470 [55:32<27:46,  1.20it/s]
4|4470|Loss: 0.7818350195884705:  69%|██████▉   | 4470/6470 [55:32<28:47,  1.16it/s]
4|4471|Loss: 0.7560794949531555:  69%|██████▉   | 4470/6470 [55:33<28:47,  1.16it/s]
4|4471|Loss: 0.7560794949531555:  69%|██████▉   | 4471/6470 [55:33<28:16,  1.18it/s]
4|4472|Loss: 0.8952350616455078:  69%|██████▉   | 4471/6470 [55:34<28:16,  1.18it/s]
4|4472|Loss: 0.8952350616455078:  69%|██████▉   | 4472/6470 [55:34<25:44,  1.29it/s]
4|4473|Loss: 0.9385267496109009:  69%|██████▉   | 4472/6470 [55:34<25:44,  1.29it/s]
4|4473|Loss: 0.9385267496109009:  69%|██████▉   | 4473/6470 [55:35<26:16,  1.27it/s]
4|4474|Loss: 0.8250576853752136:  69%|██████▉   | 4473/6470 [55:35<26:16,  1.27it/s]
4|4474|Loss: 0.8250576853752136:  69%|██████▉   | 4474/6470 [55:35<25:31,  1.30it/s]
4|4475|Loss: 0.8919112086296082:  69%|██████▉   | 4474/6470 [55:36<25:31,  1.30it/s]
4|4475|Loss: 0.8919112086296082:  69%|██████▉   | 4475/6470 [55:36<25:33,  1.30it/s]
4|4476|Loss: 0.7588379383087158:  69%|██████▉   | 4475/6470 [55:36<25:33,  1.30it/s]
4|4476|Loss: 0.7588379383087158:  69%|██████▉   | 4476/6470 [55:37<24:25,  1.36it/s]
4|4477|Loss: 0.9270493984222412:  69%|██████▉   | 4476/6470 [55:37<24:25,  1.36it/s]
4|4477|Loss: 0.9270493984222412:  69%|██████▉   | 4477/6470 [55:38<25:12,  1.32it/s]
4|4478|Loss: 0.7466456294059753:  69%|██████▉   | 4477/6470 [55:38<25:12,  1.32it/s]
4|4478|Loss: 0.7466456294059753:  69%|██████▉   | 4478/6470 [55:38<24:54,  1.33it/s]
4|4479|Loss: 0.8036946058273315:  69%|██████▉   | 4478/6470 [55:39<24:54,  1.33it/s]
4|4479|Loss: 0.8036946058273315:  69%|██████▉   | 4479/6470 [55:39<24:11,  1.37it/s]
4|4480|Loss: 0.9312044978141785:  69%|██████▉   | 4479/6470 [55:39<24:11,  1.37it/s]
4|4480|Loss: 0.9312044978141785:  69%|██████▉   | 4480/6470 [55:40<23:31,  1.41it/s]
4|4481|Loss: 0.8569254875183105:  69%|██████▉   | 4480/6470 [55:40<23:31,  1.41it/s]
4|4481|Loss: 0.8569254875183105:  69%|██████▉   | 4481/6470 [55:40<23:23,  1.42it/s]
4|4482|Loss: 0.8139587640762329:  69%|██████▉   | 4481/6470 [55:41<23:23,  1.42it/s]
4|4482|Loss: 0.8139587640762329:  69%|██████▉   | 4482/6470 [55:41<23:44,  1.40it/s]
4|4483|Loss: 0.9309670329093933:  69%|██████▉   | 4482/6470 [55:41<23:44,  1.40it/s]
4|4483|Loss: 0.9309670329093933:  69%|██████▉   | 4483/6470 [55:42<24:34,  1.35it/s]
4|4484|Loss: 0.9896898865699768:  69%|██████▉   | 4483/6470 [55:42<24:34,  1.35it/s]
4|4484|Loss: 0.9896898865699768:  69%|██████▉   | 4484/6470 [55:43<24:30,  1.35it/s]
4|4485|Loss: 0.8457242846488953:  69%|██████▉   | 4484/6470 [55:43<24:30,  1.35it/s]
4|4485|Loss: 0.8457242846488953:  69%|██████▉   | 4485/6470 [55:43<24:10,  1.37it/s]
4|4486|Loss: 0.9675145149230957:  69%|██████▉   | 4485/6470 [55:44<24:10,  1.37it/s]
4|4486|Loss: 0.9675145149230957:  69%|██████▉   | 4486/6470 [55:44<23:52,  1.38it/s]
4|4487|Loss: 0.8358609676361084:  69%|██████▉   | 4486/6470 [55:44<23:52,  1.38it/s]
4|4487|Loss: 0.8358609676361084:  69%|██████▉   | 4487/6470 [55:45<25:16,  1.31it/s]
4|4488|Loss: 0.8139713406562805:  69%|██████▉   | 4487/6470 [55:45<25:16,  1.31it/s]
4|4488|Loss: 0.8139713406562805:  69%|██████▉   | 4488/6470 [55:46<27:19,  1.21it/s]
4|4489|Loss: 0.8709888458251953:  69%|██████▉   | 4488/6470 [55:46<27:19,  1.21it/s]
4|4489|Loss: 0.8709888458251953:  69%|██████▉   | 4489/6470 [55:46<25:34,  1.29it/s]
4|4490|Loss: 0.7289028763771057:  69%|██████▉   | 4489/6470 [55:47<25:34,  1.29it/s]
4|4490|Loss: 0.7289028763771057:  69%|██████▉   | 4490/6470 [55:47<24:25,  1.35it/s]
4|4491|Loss: 0.8493924736976624:  69%|██████▉   | 4490/6470 [55:48<24:25,  1.35it/s]
4|4491|Loss: 0.8493924736976624:  69%|██████▉   | 4491/6470 [55:48<24:50,  1.33it/s]
4|4492|Loss: 0.7608983516693115:  69%|██████▉   | 4491/6470 [55:48<24:50,  1.33it/s]
4|4492|Loss: 0.7608983516693115:  69%|██████▉   | 4492/6470 [55:49<25:12,  1.31it/s]
4|4493|Loss: 0.6877650618553162:  69%|██████▉   | 4492/6470 [55:49<25:12,  1.31it/s]
4|4493|Loss: 0.6877650618553162:  69%|██████▉   | 4493/6470 [55:50<26:04,  1.26it/s]
4|4494|Loss: 0.8662858605384827:  69%|██████▉   | 4493/6470 [55:50<26:04,  1.26it/s]
4|4494|Loss: 0.8662858605384827:  69%|██████▉   | 4494/6470 [55:50<25:49,  1.28it/s]
4|4495|Loss: 0.7768346667289734:  69%|██████▉   | 4494/6470 [55:51<25:49,  1.28it/s]
4|4495|Loss: 0.7768346667289734:  69%|██████▉   | 4495/6470 [55:51<25:49,  1.27it/s]
4|4496|Loss: 0.8210282325744629:  69%|██████▉   | 4495/6470 [55:52<25:49,  1.27it/s]
4|4496|Loss: 0.8210282325744629:  69%|██████▉   | 4496/6470 [55:52<26:08,  1.26it/s]
4|4497|Loss: 0.9201321005821228:  69%|██████▉   | 4496/6470 [55:52<26:08,  1.26it/s]
4|4497|Loss: 0.9201321005821228:  70%|██████▉   | 4497/6470 [55:53<26:20,  1.25it/s]
4|4498|Loss: 0.7832353115081787:  70%|██████▉   | 4497/6470 [55:53<26:20,  1.25it/s]
4|4498|Loss: 0.7832353115081787:  70%|██████▉   | 4498/6470 [55:53<25:17,  1.30it/s]
4|4499|Loss: 0.7434294819831848:  70%|██████▉   | 4498/6470 [55:54<25:17,  1.30it/s]
4|4499|Loss: 0.7434294819831848:  70%|██████▉   | 4499/6470 [55:54<23:55,  1.37it/s]
4|4500|Loss: 0.6587225198745728:  70%|██████▉   | 4499/6470 [55:55<23:55,  1.37it/s]
4|4500|Loss: 0.6587225198745728:  70%|██████▉   | 4500/6470 [55:55<25:46,  1.27it/s]
4|4501|Loss: 0.7435115575790405:  70%|██████▉   | 4500/6470 [55:56<25:46,  1.27it/s]
4|4501|Loss: 0.7435115575790405:  70%|██████▉   | 4501/6470 [55:56<27:08,  1.21it/s]
4|4502|Loss: 0.8836954832077026:  70%|██████▉   | 4501/6470 [55:56<27:08,  1.21it/s]
4|4502|Loss: 0.8836954832077026:  70%|██████▉   | 4502/6470 [55:57<26:54,  1.22it/s]
4|4503|Loss: 0.8102467060089111:  70%|██████▉   | 4502/6470 [55:57<26:54,  1.22it/s]
4|4503|Loss: 0.8102467060089111:  70%|██████▉   | 4503/6470 [55:57<25:22,  1.29it/s]
4|4504|Loss: 0.725034236907959:  70%|██████▉   | 4503/6470 [55:58<25:22,  1.29it/s] 
4|4504|Loss: 0.725034236907959:  70%|██████▉   | 4504/6470 [55:58<24:10,  1.36it/s]
4|4505|Loss: 0.774817705154419:  70%|██████▉   | 4504/6470 [55:58<24:10,  1.36it/s]
4|4505|Loss: 0.774817705154419:  70%|██████▉   | 4505/6470 [55:59<22:24,  1.46it/s]
4|4506|Loss: 0.7859421372413635:  70%|██████▉   | 4505/6470 [55:59<22:24,  1.46it/s]
4|4506|Loss: 0.7859421372413635:  70%|██████▉   | 4506/6470 [55:59<21:36,  1.52it/s]
4|4507|Loss: 0.8996635675430298:  70%|██████▉   | 4506/6470 [56:00<21:36,  1.52it/s]
4|4507|Loss: 0.8996635675430298:  70%|██████▉   | 4507/6470 [56:00<20:43,  1.58it/s]
4|4508|Loss: 0.8289431929588318:  70%|██████▉   | 4507/6470 [56:00<20:43,  1.58it/s]
4|4508|Loss: 0.8289431929588318:  70%|██████▉   | 4508/6470 [56:01<21:47,  1.50it/s]
4|4509|Loss: 0.8248400092124939:  70%|██████▉   | 4508/6470 [56:01<21:47,  1.50it/s]
4|4509|Loss: 0.8248400092124939:  70%|██████▉   | 4509/6470 [56:01<22:14,  1.47it/s]
4|4510|Loss: 0.7126659750938416:  70%|██████▉   | 4509/6470 [56:02<22:14,  1.47it/s]
4|4510|Loss: 0.7126659750938416:  70%|██████▉   | 4510/6470 [56:02<22:52,  1.43it/s]
4|4511|Loss: 0.8287817239761353:  70%|██████▉   | 4510/6470 [56:02<22:52,  1.43it/s]
4|4511|Loss: 0.8287817239761353:  70%|██████▉   | 4511/6470 [56:03<23:35,  1.38it/s]
4|4512|Loss: 0.7374542355537415:  70%|██████▉   | 4511/6470 [56:03<23:35,  1.38it/s]
4|4512|Loss: 0.7374542355537415:  70%|██████▉   | 4512/6470 [56:04<23:56,  1.36it/s]
4|4513|Loss: 0.7281991839408875:  70%|██████▉   | 4512/6470 [56:04<23:56,  1.36it/s]
4|4513|Loss: 0.7281991839408875:  70%|██████▉   | 4513/6470 [56:04<23:38,  1.38it/s]
4|4514|Loss: 0.8221904635429382:  70%|██████▉   | 4513/6470 [56:05<23:38,  1.38it/s]
4|4514|Loss: 0.8221904635429382:  70%|██████▉   | 4514/6470 [56:05<23:06,  1.41it/s]
4|4515|Loss: 0.742343008518219:  70%|██████▉   | 4514/6470 [56:05<23:06,  1.41it/s] 
4|4515|Loss: 0.742343008518219:  70%|██████▉   | 4515/6470 [56:06<22:02,  1.48it/s]
4|4516|Loss: 0.8033615350723267:  70%|██████▉   | 4515/6470 [56:06<22:02,  1.48it/s]
4|4516|Loss: 0.8033615350723267:  70%|██████▉   | 4516/6470 [56:06<22:56,  1.42it/s]
4|4517|Loss: 0.8771359920501709:  70%|██████▉   | 4516/6470 [56:07<22:56,  1.42it/s]
4|4517|Loss: 0.8771359920501709:  70%|██████▉   | 4517/6470 [56:07<22:59,  1.42it/s]
4|4518|Loss: 0.7981215715408325:  70%|██████▉   | 4517/6470 [56:07<22:59,  1.42it/s]
4|4518|Loss: 0.7981215715408325:  70%|██████▉   | 4518/6470 [56:08<21:35,  1.51it/s]
4|4519|Loss: 1.032144546508789:  70%|██████▉   | 4518/6470 [56:08<21:35,  1.51it/s] 
4|4519|Loss: 1.032144546508789:  70%|██████▉   | 4519/6470 [56:08<22:04,  1.47it/s]
4|4520|Loss: 0.919463574886322:  70%|██████▉   | 4519/6470 [56:09<22:04,  1.47it/s]
4|4520|Loss: 0.919463574886322:  70%|██████▉   | 4520/6470 [56:09<22:54,  1.42it/s]
4|4521|Loss: 0.9391325116157532:  70%|██████▉   | 4520/6470 [56:09<22:54,  1.42it/s]
4|4521|Loss: 0.9391325116157532:  70%|██████▉   | 4521/6470 [56:10<23:29,  1.38it/s]
4|4522|Loss: 0.804909884929657:  70%|██████▉   | 4521/6470 [56:10<23:29,  1.38it/s] 
4|4522|Loss: 0.804909884929657:  70%|██████▉   | 4522/6470 [56:10<23:03,  1.41it/s]
4|4523|Loss: 0.8116692304611206:  70%|██████▉   | 4522/6470 [56:11<23:03,  1.41it/s]
4|4523|Loss: 0.8116692304611206:  70%|██████▉   | 4523/6470 [56:11<22:26,  1.45it/s]
4|4524|Loss: 0.8830482959747314:  70%|██████▉   | 4523/6470 [56:12<22:26,  1.45it/s]
4|4524|Loss: 0.8830482959747314:  70%|██████▉   | 4524/6470 [56:12<24:07,  1.34it/s]
4|4525|Loss: 0.9172681570053101:  70%|██████▉   | 4524/6470 [56:12<24:07,  1.34it/s]
4|4525|Loss: 0.9172681570053101:  70%|██████▉   | 4525/6470 [56:13<24:58,  1.30it/s]
4|4526|Loss: 0.8618070483207703:  70%|██████▉   | 4525/6470 [56:13<24:58,  1.30it/s]
4|4526|Loss: 0.8618070483207703:  70%|██████▉   | 4526/6470 [56:13<23:27,  1.38it/s]
4|4527|Loss: 0.6983151435852051:  70%|██████▉   | 4526/6470 [56:14<23:27,  1.38it/s]
4|4527|Loss: 0.6983151435852051:  70%|██████▉   | 4527/6470 [56:14<22:40,  1.43it/s]
4|4528|Loss: 0.7887055277824402:  70%|██████▉   | 4527/6470 [56:14<22:40,  1.43it/s]
4|4528|Loss: 0.7887055277824402:  70%|██████▉   | 4528/6470 [56:15<22:41,  1.43it/s]
4|4529|Loss: 0.8245378136634827:  70%|██████▉   | 4528/6470 [56:15<22:41,  1.43it/s]
4|4529|Loss: 0.8245378136634827:  70%|███████   | 4529/6470 [56:16<24:30,  1.32it/s]
4|4530|Loss: 0.7778640389442444:  70%|███████   | 4529/6470 [56:16<24:30,  1.32it/s]
4|4530|Loss: 0.7778640389442444:  70%|███████   | 4530/6470 [56:16<24:56,  1.30it/s]
4|4531|Loss: 0.7927854657173157:  70%|███████   | 4530/6470 [56:17<24:56,  1.30it/s]
4|4531|Loss: 0.7927854657173157:  70%|███████   | 4531/6470 [56:17<25:03,  1.29it/s]
4|4532|Loss: 0.8775639533996582:  70%|███████   | 4531/6470 [56:18<25:03,  1.29it/s]
4|4532|Loss: 0.8775639533996582:  70%|███████   | 4532/6470 [56:18<24:23,  1.32it/s]
4|4533|Loss: 0.8132709860801697:  70%|███████   | 4532/6470 [56:18<24:23,  1.32it/s]
4|4533|Loss: 0.8132709860801697:  70%|███████   | 4533/6470 [56:19<24:33,  1.31it/s]
4|4534|Loss: 0.7672708630561829:  70%|███████   | 4533/6470 [56:19<24:33,  1.31it/s]
4|4534|Loss: 0.7672708630561829:  70%|███████   | 4534/6470 [56:19<24:04,  1.34it/s]
4|4535|Loss: 0.8913817405700684:  70%|███████   | 4534/6470 [56:20<24:04,  1.34it/s]
4|4535|Loss: 0.8913817405700684:  70%|███████   | 4535/6470 [56:20<23:42,  1.36it/s]
4|4536|Loss: 0.7682546377182007:  70%|███████   | 4535/6470 [56:21<23:42,  1.36it/s]
4|4536|Loss: 0.7682546377182007:  70%|███████   | 4536/6470 [56:21<23:18,  1.38it/s]
4|4537|Loss: 0.7584753036499023:  70%|███████   | 4536/6470 [56:21<23:18,  1.38it/s]
4|4537|Loss: 0.7584753036499023:  70%|███████   | 4537/6470 [56:22<22:27,  1.43it/s]
4|4538|Loss: 0.7485721707344055:  70%|███████   | 4537/6470 [56:22<22:27,  1.43it/s]
4|4538|Loss: 0.7485721707344055:  70%|███████   | 4538/6470 [56:22<21:42,  1.48it/s]
4|4539|Loss: 0.8446349501609802:  70%|███████   | 4538/6470 [56:23<21:42,  1.48it/s]
4|4539|Loss: 0.8446349501609802:  70%|███████   | 4539/6470 [56:23<22:34,  1.43it/s]
4|4540|Loss: 0.765446662902832:  70%|███████   | 4539/6470 [56:23<22:34,  1.43it/s] 
4|4540|Loss: 0.765446662902832:  70%|███████   | 4540/6470 [56:24<23:36,  1.36it/s]
4|4541|Loss: 0.7437039613723755:  70%|███████   | 4540/6470 [56:24<23:36,  1.36it/s]
4|4541|Loss: 0.7437039613723755:  70%|███████   | 4541/6470 [56:25<24:54,  1.29it/s]
4|4542|Loss: 0.7297015190124512:  70%|███████   | 4541/6470 [56:25<24:54,  1.29it/s]
4|4542|Loss: 0.7297015190124512:  70%|███████   | 4542/6470 [56:25<24:55,  1.29it/s]
4|4543|Loss: 0.9361864328384399:  70%|███████   | 4542/6470 [56:26<24:55,  1.29it/s]
4|4543|Loss: 0.9361864328384399:  70%|███████   | 4543/6470 [56:26<24:54,  1.29it/s]
4|4544|Loss: 0.7962331771850586:  70%|███████   | 4543/6470 [56:27<24:54,  1.29it/s]
4|4544|Loss: 0.7962331771850586:  70%|███████   | 4544/6470 [56:27<26:18,  1.22it/s]
4|4545|Loss: 0.8611968159675598:  70%|███████   | 4544/6470 [56:27<26:18,  1.22it/s]
4|4545|Loss: 0.8611968159675598:  70%|███████   | 4545/6470 [56:28<24:42,  1.30it/s]
4|4546|Loss: 0.8712211847305298:  70%|███████   | 4545/6470 [56:28<24:42,  1.30it/s]
4|4546|Loss: 0.8712211847305298:  70%|███████   | 4546/6470 [56:29<25:02,  1.28it/s]
4|4547|Loss: 0.8915242552757263:  70%|███████   | 4546/6470 [56:29<25:02,  1.28it/s]
4|4547|Loss: 0.8915242552757263:  70%|███████   | 4547/6470 [56:29<23:56,  1.34it/s]
4|4548|Loss: 0.8897165656089783:  70%|███████   | 4547/6470 [56:30<23:56,  1.34it/s]
4|4548|Loss: 0.8897165656089783:  70%|███████   | 4548/6470 [56:30<23:42,  1.35it/s]
4|4549|Loss: 0.732333242893219:  70%|███████   | 4548/6470 [56:30<23:42,  1.35it/s] 
4|4549|Loss: 0.732333242893219:  70%|███████   | 4549/6470 [56:31<24:55,  1.28it/s]
4|4550|Loss: 0.9203776121139526:  70%|███████   | 4549/6470 [56:31<24:55,  1.28it/s]
4|4550|Loss: 0.9203776121139526:  70%|███████   | 4550/6470 [56:32<24:54,  1.28it/s]
4|4551|Loss: 0.9250399470329285:  70%|███████   | 4550/6470 [56:32<24:54,  1.28it/s]
4|4551|Loss: 0.9250399470329285:  70%|███████   | 4551/6470 [56:32<25:41,  1.24it/s]
4|4552|Loss: 0.9773623943328857:  70%|███████   | 4551/6470 [56:33<25:41,  1.24it/s]
4|4552|Loss: 0.9773623943328857:  70%|███████   | 4552/6470 [56:33<24:58,  1.28it/s]
4|4553|Loss: 0.9298385977745056:  70%|███████   | 4552/6470 [56:33<24:58,  1.28it/s]
4|4553|Loss: 0.9298385977745056:  70%|███████   | 4553/6470 [56:34<22:28,  1.42it/s]
4|4554|Loss: 0.710074782371521:  70%|███████   | 4553/6470 [56:34<22:28,  1.42it/s] 
4|4554|Loss: 0.710074782371521:  70%|███████   | 4554/6470 [56:34<22:42,  1.41it/s]
4|4555|Loss: 0.8739675879478455:  70%|███████   | 4554/6470 [56:35<22:42,  1.41it/s]
4|4555|Loss: 0.8739675879478455:  70%|███████   | 4555/6470 [56:35<22:25,  1.42it/s]
4|4556|Loss: 0.8745433688163757:  70%|███████   | 4555/6470 [56:35<22:25,  1.42it/s]
4|4556|Loss: 0.8745433688163757:  70%|███████   | 4556/6470 [56:36<22:49,  1.40it/s]
4|4557|Loss: 0.7970741391181946:  70%|███████   | 4556/6470 [56:36<22:49,  1.40it/s]
4|4557|Loss: 0.7970741391181946:  70%|███████   | 4557/6470 [56:37<24:47,  1.29it/s]
4|4558|Loss: 0.8112090826034546:  70%|███████   | 4557/6470 [56:37<24:47,  1.29it/s]
4|4558|Loss: 0.8112090826034546:  70%|███████   | 4558/6470 [56:38<25:10,  1.27it/s]
4|4559|Loss: 0.7363875508308411:  70%|███████   | 4558/6470 [56:38<25:10,  1.27it/s]
4|4559|Loss: 0.7363875508308411:  70%|███████   | 4559/6470 [56:38<23:54,  1.33it/s]
4|4560|Loss: 0.517646312713623:  70%|███████   | 4559/6470 [56:39<23:54,  1.33it/s] 
4|4560|Loss: 0.517646312713623:  70%|███████   | 4560/6470 [56:39<23:11,  1.37it/s]
4|4561|Loss: 0.6741092801094055:  70%|███████   | 4560/6470 [56:39<23:11,  1.37it/s]
4|4561|Loss: 0.6741092801094055:  70%|███████   | 4561/6470 [56:39<21:12,  1.50it/s]
4|4562|Loss: 0.8256609439849854:  70%|███████   | 4561/6470 [56:40<21:12,  1.50it/s]
4|4562|Loss: 0.8256609439849854:  71%|███████   | 4562/6470 [56:40<21:28,  1.48it/s]
4|4563|Loss: 0.9566253423690796:  71%|███████   | 4562/6470 [56:41<21:28,  1.48it/s]
4|4563|Loss: 0.9566253423690796:  71%|███████   | 4563/6470 [56:41<23:19,  1.36it/s]
4|4564|Loss: 0.8129637241363525:  71%|███████   | 4563/6470 [56:41<23:19,  1.36it/s]
4|4564|Loss: 0.8129637241363525:  71%|███████   | 4564/6470 [56:42<24:15,  1.31it/s]
4|4565|Loss: 0.8367475271224976:  71%|███████   | 4564/6470 [56:42<24:15,  1.31it/s]
4|4565|Loss: 0.8367475271224976:  71%|███████   | 4565/6470 [56:42<23:06,  1.37it/s]
4|4566|Loss: 0.7743933200836182:  71%|███████   | 4565/6470 [56:43<23:06,  1.37it/s]
4|4566|Loss: 0.7743933200836182:  71%|███████   | 4566/6470 [56:43<23:58,  1.32it/s]
4|4567|Loss: 0.7943076491355896:  71%|███████   | 4566/6470 [56:44<23:58,  1.32it/s]
4|4567|Loss: 0.7943076491355896:  71%|███████   | 4567/6470 [56:44<24:36,  1.29it/s]
4|4568|Loss: 0.8574144244194031:  71%|███████   | 4567/6470 [56:45<24:36,  1.29it/s]
4|4568|Loss: 0.8574144244194031:  71%|███████   | 4568/6470 [56:45<24:41,  1.28it/s]
4|4569|Loss: 0.7270691990852356:  71%|███████   | 4568/6470 [56:45<24:41,  1.28it/s]
4|4569|Loss: 0.7270691990852356:  71%|███████   | 4569/6470 [56:46<23:11,  1.37it/s]
4|4570|Loss: 0.8199054002761841:  71%|███████   | 4569/6470 [56:46<23:11,  1.37it/s]
4|4570|Loss: 0.8199054002761841:  71%|███████   | 4570/6470 [56:46<22:26,  1.41it/s]
4|4571|Loss: 0.8710330724716187:  71%|███████   | 4570/6470 [56:47<22:26,  1.41it/s]
4|4571|Loss: 0.8710330724716187:  71%|███████   | 4571/6470 [56:47<21:35,  1.47it/s]
4|4572|Loss: 0.7232487201690674:  71%|███████   | 4571/6470 [56:47<21:35,  1.47it/s]
4|4572|Loss: 0.7232487201690674:  71%|███████   | 4572/6470 [56:48<23:19,  1.36it/s]
4|4573|Loss: 0.8211292624473572:  71%|███████   | 4572/6470 [56:48<23:19,  1.36it/s]
4|4573|Loss: 0.8211292624473572:  71%|███████   | 4573/6470 [56:48<23:14,  1.36it/s]
4|4574|Loss: 0.8451860547065735:  71%|███████   | 4573/6470 [56:49<23:14,  1.36it/s]
4|4574|Loss: 0.8451860547065735:  71%|███████   | 4574/6470 [56:49<24:35,  1.29it/s]
4|4575|Loss: 0.8185028433799744:  71%|███████   | 4574/6470 [56:50<24:35,  1.29it/s]
4|4575|Loss: 0.8185028433799744:  71%|███████   | 4575/6470 [56:50<24:55,  1.27it/s]
4|4576|Loss: 0.6892768144607544:  71%|███████   | 4575/6470 [56:50<24:55,  1.27it/s]
4|4576|Loss: 0.6892768144607544:  71%|███████   | 4576/6470 [56:51<24:14,  1.30it/s]
4|4577|Loss: 0.8147962093353271:  71%|███████   | 4576/6470 [56:51<24:14,  1.30it/s]
4|4577|Loss: 0.8147962093353271:  71%|███████   | 4577/6470 [56:52<24:20,  1.30it/s]
4|4578|Loss: 0.7952530384063721:  71%|███████   | 4577/6470 [56:52<24:20,  1.30it/s]
4|4578|Loss: 0.7952530384063721:  71%|███████   | 4578/6470 [56:52<25:35,  1.23it/s]
4|4579|Loss: 0.9435019493103027:  71%|███████   | 4578/6470 [56:53<25:35,  1.23it/s]
4|4579|Loss: 0.9435019493103027:  71%|███████   | 4579/6470 [56:53<25:22,  1.24it/s]
4|4580|Loss: 0.76888507604599:  71%|███████   | 4579/6470 [56:54<25:22,  1.24it/s]  
4|4580|Loss: 0.76888507604599:  71%|███████   | 4580/6470 [56:54<25:21,  1.24it/s]
4|4581|Loss: 0.7933517694473267:  71%|███████   | 4580/6470 [56:55<25:21,  1.24it/s]
4|4581|Loss: 0.7933517694473267:  71%|███████   | 4581/6470 [56:55<25:01,  1.26it/s]
4|4582|Loss: 0.72833651304245:  71%|███████   | 4581/6470 [56:55<25:01,  1.26it/s]  
4|4582|Loss: 0.72833651304245:  71%|███████   | 4582/6470 [56:55<22:31,  1.40it/s]
4|4583|Loss: 0.9033761620521545:  71%|███████   | 4582/6470 [56:56<22:31,  1.40it/s]
4|4583|Loss: 0.9033761620521545:  71%|███████   | 4583/6470 [56:56<21:19,  1.48it/s]
4|4584|Loss: 0.7765184044837952:  71%|███████   | 4583/6470 [56:56<21:19,  1.48it/s]
4|4584|Loss: 0.7765184044837952:  71%|███████   | 4584/6470 [56:57<20:09,  1.56it/s]
4|4585|Loss: 0.8434577584266663:  71%|███████   | 4584/6470 [56:57<20:09,  1.56it/s]
4|4585|Loss: 0.8434577584266663:  71%|███████   | 4585/6470 [56:57<22:15,  1.41it/s]
4|4586|Loss: 0.70708167552948:  71%|███████   | 4585/6470 [56:58<22:15,  1.41it/s]  
4|4586|Loss: 0.70708167552948:  71%|███████   | 4586/6470 [56:58<24:42,  1.27it/s]
4|4587|Loss: 0.8974407911300659:  71%|███████   | 4586/6470 [56:59<24:42,  1.27it/s]
4|4587|Loss: 0.8974407911300659:  71%|███████   | 4587/6470 [56:59<25:17,  1.24it/s]
4|4588|Loss: 0.8970456719398499:  71%|███████   | 4587/6470 [57:00<25:17,  1.24it/s]
4|4588|Loss: 0.8970456719398499:  71%|███████   | 4588/6470 [57:00<25:14,  1.24it/s]
4|4589|Loss: 0.8439551591873169:  71%|███████   | 4588/6470 [57:00<25:14,  1.24it/s]
4|4589|Loss: 0.8439551591873169:  71%|███████   | 4589/6470 [57:01<24:46,  1.26it/s]
4|4590|Loss: 0.8366888165473938:  71%|███████   | 4589/6470 [57:01<24:46,  1.26it/s]
4|4590|Loss: 0.8366888165473938:  71%|███████   | 4590/6470 [57:02<25:23,  1.23it/s]
4|4591|Loss: 0.7573493123054504:  71%|███████   | 4590/6470 [57:02<25:23,  1.23it/s]
4|4591|Loss: 0.7573493123054504:  71%|███████   | 4591/6470 [57:02<24:41,  1.27it/s]
4|4592|Loss: 0.7753767967224121:  71%|███████   | 4591/6470 [57:03<24:41,  1.27it/s]
4|4592|Loss: 0.7753767967224121:  71%|███████   | 4592/6470 [57:03<23:44,  1.32it/s]
4|4593|Loss: 1.007671594619751:  71%|███████   | 4592/6470 [57:03<23:44,  1.32it/s] 
4|4593|Loss: 1.007671594619751:  71%|███████   | 4593/6470 [57:04<24:20,  1.29it/s]
4|4594|Loss: 0.9469798803329468:  71%|███████   | 4593/6470 [57:04<24:20,  1.29it/s]
4|4594|Loss: 0.9469798803329468:  71%|███████   | 4594/6470 [57:04<22:19,  1.40it/s]
4|4595|Loss: 0.7881351709365845:  71%|███████   | 4594/6470 [57:05<22:19,  1.40it/s]
4|4595|Loss: 0.7881351709365845:  71%|███████   | 4595/6470 [57:05<22:30,  1.39it/s]
4|4596|Loss: 0.9296330809593201:  71%|███████   | 4595/6470 [57:06<22:30,  1.39it/s]
4|4596|Loss: 0.9296330809593201:  71%|███████   | 4596/6470 [57:06<22:36,  1.38it/s]
4|4597|Loss: 0.7178525924682617:  71%|███████   | 4596/6470 [57:06<22:36,  1.38it/s]
4|4597|Loss: 0.7178525924682617:  71%|███████   | 4597/6470 [57:07<22:07,  1.41it/s]
4|4598|Loss: 0.8134323358535767:  71%|███████   | 4597/6470 [57:07<22:07,  1.41it/s]
4|4598|Loss: 0.8134323358535767:  71%|███████   | 4598/6470 [57:07<22:17,  1.40it/s]
4|4599|Loss: 0.7854771018028259:  71%|███████   | 4598/6470 [57:08<22:17,  1.40it/s]
4|4599|Loss: 0.7854771018028259:  71%|███████   | 4599/6470 [57:08<21:57,  1.42it/s]
4|4600|Loss: 0.8418577909469604:  71%|███████   | 4599/6470 [57:08<21:57,  1.42it/s]
4|4600|Loss: 0.8418577909469604:  71%|███████   | 4600/6470 [57:09<22:12,  1.40it/s]
4|4601|Loss: 0.9047311544418335:  71%|███████   | 4600/6470 [57:09<22:12,  1.40it/s]
4|4601|Loss: 0.9047311544418335:  71%|███████   | 4601/6470 [57:10<22:55,  1.36it/s]
4|4602|Loss: 0.9496356844902039:  71%|███████   | 4601/6470 [57:10<22:55,  1.36it/s]
4|4602|Loss: 0.9496356844902039:  71%|███████   | 4602/6470 [57:10<24:23,  1.28it/s]
4|4603|Loss: 0.8234843611717224:  71%|███████   | 4602/6470 [57:11<24:23,  1.28it/s]
4|4603|Loss: 0.8234843611717224:  71%|███████   | 4603/6470 [57:11<25:37,  1.21it/s]
4|4604|Loss: 0.9072171449661255:  71%|███████   | 4603/6470 [57:12<25:37,  1.21it/s]
4|4604|Loss: 0.9072171449661255:  71%|███████   | 4604/6470 [57:12<26:31,  1.17it/s]
4|4605|Loss: 0.7717064619064331:  71%|███████   | 4604/6470 [57:13<26:31,  1.17it/s]
4|4605|Loss: 0.7717064619064331:  71%|███████   | 4605/6470 [57:13<26:04,  1.19it/s]
4|4606|Loss: 0.6991734504699707:  71%|███████   | 4605/6470 [57:13<26:04,  1.19it/s]
4|4606|Loss: 0.6991734504699707:  71%|███████   | 4606/6470 [57:14<25:19,  1.23it/s]
4|4607|Loss: 0.8891641497612:  71%|███████   | 4606/6470 [57:14<25:19,  1.23it/s]   
4|4607|Loss: 0.8891641497612:  71%|███████   | 4607/6470 [57:14<23:41,  1.31it/s]
4|4608|Loss: 0.909166693687439:  71%|███████   | 4607/6470 [57:15<23:41,  1.31it/s]
4|4608|Loss: 0.909166693687439:  71%|███████   | 4608/6470 [57:15<23:43,  1.31it/s]
4|4609|Loss: 0.7224449515342712:  71%|███████   | 4608/6470 [57:16<23:43,  1.31it/s]
4|4609|Loss: 0.7224449515342712:  71%|███████   | 4609/6470 [57:16<23:24,  1.33it/s]
4|4610|Loss: 0.8371883630752563:  71%|███████   | 4609/6470 [57:16<23:24,  1.33it/s]
4|4610|Loss: 0.8371883630752563:  71%|███████▏  | 4610/6470 [57:17<22:06,  1.40it/s]
4|4611|Loss: 0.7292441725730896:  71%|███████▏  | 4610/6470 [57:17<22:06,  1.40it/s]
4|4611|Loss: 0.7292441725730896:  71%|███████▏  | 4611/6470 [57:17<21:15,  1.46it/s]
4|4612|Loss: 0.7839862108230591:  71%|███████▏  | 4611/6470 [57:18<21:15,  1.46it/s]
4|4612|Loss: 0.7839862108230591:  71%|███████▏  | 4612/6470 [57:18<21:12,  1.46it/s]
4|4613|Loss: 0.7937199473381042:  71%|███████▏  | 4612/6470 [57:18<21:12,  1.46it/s]
4|4613|Loss: 0.7937199473381042:  71%|███████▏  | 4613/6470 [57:19<21:58,  1.41it/s]
4|4614|Loss: 0.7828588485717773:  71%|███████▏  | 4613/6470 [57:19<21:58,  1.41it/s]
4|4614|Loss: 0.7828588485717773:  71%|███████▏  | 4614/6470 [57:19<22:47,  1.36it/s]
4|4615|Loss: 0.8567168116569519:  71%|███████▏  | 4614/6470 [57:20<22:47,  1.36it/s]
4|4615|Loss: 0.8567168116569519:  71%|███████▏  | 4615/6470 [57:20<24:40,  1.25it/s]
4|4616|Loss: 0.7872654795646667:  71%|███████▏  | 4615/6470 [57:21<24:40,  1.25it/s]
4|4616|Loss: 0.7872654795646667:  71%|███████▏  | 4616/6470 [57:21<24:50,  1.24it/s]
4|4617|Loss: 0.897815465927124:  71%|███████▏  | 4616/6470 [57:22<24:50,  1.24it/s] 
4|4617|Loss: 0.897815465927124:  71%|███████▏  | 4617/6470 [57:22<25:42,  1.20it/s]
4|4618|Loss: 0.7656719088554382:  71%|███████▏  | 4617/6470 [57:23<25:42,  1.20it/s]
4|4618|Loss: 0.7656719088554382:  71%|███████▏  | 4618/6470 [57:23<25:19,  1.22it/s]
4|4619|Loss: 0.8521652817726135:  71%|███████▏  | 4618/6470 [57:23<25:19,  1.22it/s]
4|4619|Loss: 0.8521652817726135:  71%|███████▏  | 4619/6470 [57:24<25:20,  1.22it/s]
4|4620|Loss: 0.6794450283050537:  71%|███████▏  | 4619/6470 [57:24<25:20,  1.22it/s]
4|4620|Loss: 0.6794450283050537:  71%|███████▏  | 4620/6470 [57:24<24:26,  1.26it/s]
4|4621|Loss: 0.7866922616958618:  71%|███████▏  | 4620/6470 [57:25<24:26,  1.26it/s]
4|4621|Loss: 0.7866922616958618:  71%|███████▏  | 4621/6470 [57:25<25:15,  1.22it/s]
4|4622|Loss: 0.8226816058158875:  71%|███████▏  | 4621/6470 [57:26<25:15,  1.22it/s]
4|4622|Loss: 0.8226816058158875:  71%|███████▏  | 4622/6470 [57:26<25:14,  1.22it/s]
4|4623|Loss: 0.8648896813392639:  71%|███████▏  | 4622/6470 [57:27<25:14,  1.22it/s]
4|4623|Loss: 0.8648896813392639:  71%|███████▏  | 4623/6470 [57:27<26:05,  1.18it/s]
4|4624|Loss: 0.6917378306388855:  71%|███████▏  | 4623/6470 [57:27<26:05,  1.18it/s]
4|4624|Loss: 0.6917378306388855:  71%|███████▏  | 4624/6470 [57:28<24:57,  1.23it/s]
4|4625|Loss: 0.9832522869110107:  71%|███████▏  | 4624/6470 [57:28<24:57,  1.23it/s]
4|4625|Loss: 0.9832522869110107:  71%|███████▏  | 4625/6470 [57:29<24:28,  1.26it/s]
4|4626|Loss: 0.7279117703437805:  71%|███████▏  | 4625/6470 [57:29<24:28,  1.26it/s]
4|4626|Loss: 0.7279117703437805:  71%|███████▏  | 4626/6470 [57:29<24:19,  1.26it/s]
4|4627|Loss: 0.7998284101486206:  71%|███████▏  | 4626/6470 [57:30<24:19,  1.26it/s]
4|4627|Loss: 0.7998284101486206:  72%|███████▏  | 4627/6470 [57:30<23:25,  1.31it/s]
4|4628|Loss: 0.9551756381988525:  72%|███████▏  | 4627/6470 [57:30<23:25,  1.31it/s]
4|4628|Loss: 0.9551756381988525:  72%|███████▏  | 4628/6470 [57:31<22:49,  1.35it/s]
4|4629|Loss: 0.7408505082130432:  72%|███████▏  | 4628/6470 [57:31<22:49,  1.35it/s]
4|4629|Loss: 0.7408505082130432:  72%|███████▏  | 4629/6470 [57:31<22:25,  1.37it/s]
4|4630|Loss: 0.8773284554481506:  72%|███████▏  | 4629/6470 [57:32<22:25,  1.37it/s]
4|4630|Loss: 0.8773284554481506:  72%|███████▏  | 4630/6470 [57:32<21:51,  1.40it/s]
4|4631|Loss: 0.8166933655738831:  72%|███████▏  | 4630/6470 [57:32<21:51,  1.40it/s]
4|4631|Loss: 0.8166933655738831:  72%|███████▏  | 4631/6470 [57:33<21:38,  1.42it/s]
4|4632|Loss: 0.7113316655158997:  72%|███████▏  | 4631/6470 [57:33<21:38,  1.42it/s]
4|4632|Loss: 0.7113316655158997:  72%|███████▏  | 4632/6470 [57:34<22:03,  1.39it/s]
4|4633|Loss: 0.7574044466018677:  72%|███████▏  | 4632/6470 [57:34<22:03,  1.39it/s]
4|4633|Loss: 0.7574044466018677:  72%|███████▏  | 4633/6470 [57:34<23:40,  1.29it/s]
4|4634|Loss: 0.8393617868423462:  72%|███████▏  | 4633/6470 [57:35<23:40,  1.29it/s]
4|4634|Loss: 0.8393617868423462:  72%|███████▏  | 4634/6470 [57:35<23:14,  1.32it/s]
4|4635|Loss: 0.6951385140419006:  72%|███████▏  | 4634/6470 [57:36<23:14,  1.32it/s]
4|4635|Loss: 0.6951385140419006:  72%|███████▏  | 4635/6470 [57:36<21:48,  1.40it/s]
4|4636|Loss: 0.960508406162262:  72%|███████▏  | 4635/6470 [57:36<21:48,  1.40it/s] 
4|4636|Loss: 0.960508406162262:  72%|███████▏  | 4636/6470 [57:37<23:03,  1.33it/s]
4|4637|Loss: 0.7253962755203247:  72%|███████▏  | 4636/6470 [57:37<23:03,  1.33it/s]
4|4637|Loss: 0.7253962755203247:  72%|███████▏  | 4637/6470 [57:37<23:41,  1.29it/s]
4|4638|Loss: 0.6879338026046753:  72%|███████▏  | 4637/6470 [57:38<23:41,  1.29it/s]
4|4638|Loss: 0.6879338026046753:  72%|███████▏  | 4638/6470 [57:38<25:21,  1.20it/s]
4|4639|Loss: 0.7648458480834961:  72%|███████▏  | 4638/6470 [57:39<25:21,  1.20it/s]
4|4639|Loss: 0.7648458480834961:  72%|███████▏  | 4639/6470 [57:39<24:26,  1.25it/s]
4|4640|Loss: 0.886844277381897:  72%|███████▏  | 4639/6470 [57:39<24:26,  1.25it/s] 
4|4640|Loss: 0.886844277381897:  72%|███████▏  | 4640/6470 [57:40<22:00,  1.39it/s]
4|4641|Loss: 0.8129475116729736:  72%|███████▏  | 4640/6470 [57:40<22:00,  1.39it/s]
4|4641|Loss: 0.8129475116729736:  72%|███████▏  | 4641/6470 [57:41<23:09,  1.32it/s]
4|4642|Loss: 0.8887536525726318:  72%|███████▏  | 4641/6470 [57:41<23:09,  1.32it/s]
4|4642|Loss: 0.8887536525726318:  72%|███████▏  | 4642/6470 [57:41<24:52,  1.22it/s]
4|4643|Loss: 0.7962411642074585:  72%|███████▏  | 4642/6470 [57:42<24:52,  1.22it/s]
4|4643|Loss: 0.7962411642074585:  72%|███████▏  | 4643/6470 [57:42<23:54,  1.27it/s]
4|4644|Loss: 0.9131633639335632:  72%|███████▏  | 4643/6470 [57:43<23:54,  1.27it/s]
4|4644|Loss: 0.9131633639335632:  72%|███████▏  | 4644/6470 [57:43<24:21,  1.25it/s]
4|4645|Loss: 0.8242162466049194:  72%|███████▏  | 4644/6470 [57:43<24:21,  1.25it/s]
4|4645|Loss: 0.8242162466049194:  72%|███████▏  | 4645/6470 [57:44<24:05,  1.26it/s]
4|4646|Loss: 0.7910351753234863:  72%|███████▏  | 4645/6470 [57:44<24:05,  1.26it/s]
4|4646|Loss: 0.7910351753234863:  72%|███████▏  | 4646/6470 [57:44<23:09,  1.31it/s]
4|4647|Loss: 0.7041289210319519:  72%|███████▏  | 4646/6470 [57:45<23:09,  1.31it/s]
4|4647|Loss: 0.7041289210319519:  72%|███████▏  | 4647/6470 [57:45<24:03,  1.26it/s]
4|4648|Loss: 0.8731129765510559:  72%|███████▏  | 4647/6470 [57:46<24:03,  1.26it/s]
4|4648|Loss: 0.8731129765510559:  72%|███████▏  | 4648/6470 [57:46<22:23,  1.36it/s]
4|4649|Loss: 0.7888808846473694:  72%|███████▏  | 4648/6470 [57:46<22:23,  1.36it/s]
4|4649|Loss: 0.7888808846473694:  72%|███████▏  | 4649/6470 [57:47<21:37,  1.40it/s]
4|4650|Loss: 0.9179146885871887:  72%|███████▏  | 4649/6470 [57:47<21:37,  1.40it/s]
4|4650|Loss: 0.9179146885871887:  72%|███████▏  | 4650/6470 [57:48<23:25,  1.30it/s]
4|4651|Loss: 0.8354016542434692:  72%|███████▏  | 4650/6470 [57:48<23:25,  1.30it/s]
4|4651|Loss: 0.8354016542434692:  72%|███████▏  | 4651/6470 [57:48<22:54,  1.32it/s]
4|4652|Loss: 0.9074569940567017:  72%|███████▏  | 4651/6470 [57:49<22:54,  1.32it/s]
4|4652|Loss: 0.9074569940567017:  72%|███████▏  | 4652/6470 [57:49<22:00,  1.38it/s]
4|4653|Loss: 1.035791277885437:  72%|███████▏  | 4652/6470 [57:49<22:00,  1.38it/s] 
4|4653|Loss: 1.035791277885437:  72%|███████▏  | 4653/6470 [57:50<22:31,  1.34it/s]
4|4654|Loss: 0.8747179508209229:  72%|███████▏  | 4653/6470 [57:50<22:31,  1.34it/s]
4|4654|Loss: 0.8747179508209229:  72%|███████▏  | 4654/6470 [57:50<21:35,  1.40it/s]
4|4655|Loss: 0.8299413323402405:  72%|███████▏  | 4654/6470 [57:51<21:35,  1.40it/s]
4|4655|Loss: 0.8299413323402405:  72%|███████▏  | 4655/6470 [57:51<21:36,  1.40it/s]
4|4656|Loss: 0.8481237888336182:  72%|███████▏  | 4655/6470 [57:51<21:36,  1.40it/s]
4|4656|Loss: 0.8481237888336182:  72%|███████▏  | 4656/6470 [57:52<22:45,  1.33it/s]
4|4657|Loss: 0.9961342215538025:  72%|███████▏  | 4656/6470 [57:52<22:45,  1.33it/s]
4|4657|Loss: 0.9961342215538025:  72%|███████▏  | 4657/6470 [57:52<21:24,  1.41it/s]
4|4658|Loss: 0.7913041710853577:  72%|███████▏  | 4657/6470 [57:53<21:24,  1.41it/s]
4|4658|Loss: 0.7913041710853577:  72%|███████▏  | 4658/6470 [57:53<21:34,  1.40it/s]
4|4659|Loss: 0.7949219942092896:  72%|███████▏  | 4658/6470 [57:54<21:34,  1.40it/s]
4|4659|Loss: 0.7949219942092896:  72%|███████▏  | 4659/6470 [57:54<22:17,  1.35it/s]
4|4660|Loss: 0.6652267575263977:  72%|███████▏  | 4659/6470 [57:54<22:17,  1.35it/s]
4|4660|Loss: 0.6652267575263977:  72%|███████▏  | 4660/6470 [57:55<22:46,  1.32it/s]
4|4661|Loss: 0.5651150941848755:  72%|███████▏  | 4660/6470 [57:55<22:46,  1.32it/s]
4|4661|Loss: 0.5651150941848755:  72%|███████▏  | 4661/6470 [57:55<20:42,  1.46it/s]
4|4662|Loss: 0.7016947269439697:  72%|███████▏  | 4661/6470 [57:56<20:42,  1.46it/s]
4|4662|Loss: 0.7016947269439697:  72%|███████▏  | 4662/6470 [57:56<22:53,  1.32it/s]
4|4663|Loss: 0.6870865225791931:  72%|███████▏  | 4662/6470 [57:57<22:53,  1.32it/s]
4|4663|Loss: 0.6870865225791931:  72%|███████▏  | 4663/6470 [57:57<24:40,  1.22it/s]
4|4664|Loss: 0.7618029713630676:  72%|███████▏  | 4663/6470 [57:58<24:40,  1.22it/s]
4|4664|Loss: 0.7618029713630676:  72%|███████▏  | 4664/6470 [57:58<23:48,  1.26it/s]
4|4665|Loss: 0.7645183801651001:  72%|███████▏  | 4664/6470 [57:58<23:48,  1.26it/s]
4|4665|Loss: 0.7645183801651001:  72%|███████▏  | 4665/6470 [57:59<22:50,  1.32it/s]
4|4666|Loss: 0.865473210811615:  72%|███████▏  | 4665/6470 [57:59<22:50,  1.32it/s] 
4|4666|Loss: 0.865473210811615:  72%|███████▏  | 4666/6470 [57:59<22:59,  1.31it/s]
4|4667|Loss: 0.80781489610672:  72%|███████▏  | 4666/6470 [58:00<22:59,  1.31it/s] 
4|4667|Loss: 0.80781489610672:  72%|███████▏  | 4667/6470 [58:00<23:10,  1.30it/s]
4|4668|Loss: 0.787581741809845:  72%|███████▏  | 4667/6470 [58:01<23:10,  1.30it/s]
4|4668|Loss: 0.787581741809845:  72%|███████▏  | 4668/6470 [58:01<21:33,  1.39it/s]
4|4669|Loss: 0.8170276284217834:  72%|███████▏  | 4668/6470 [58:01<21:33,  1.39it/s]
4|4669|Loss: 0.8170276284217834:  72%|███████▏  | 4669/6470 [58:02<22:44,  1.32it/s]
4|4670|Loss: 0.9143080711364746:  72%|███████▏  | 4669/6470 [58:02<22:44,  1.32it/s]
4|4670|Loss: 0.9143080711364746:  72%|███████▏  | 4670/6470 [58:02<20:31,  1.46it/s]
4|4671|Loss: 0.780041515827179:  72%|███████▏  | 4670/6470 [58:02<20:31,  1.46it/s] 
4|4671|Loss: 0.780041515827179:  72%|███████▏  | 4671/6470 [58:03<20:24,  1.47it/s]
4|4672|Loss: 0.7785645127296448:  72%|███████▏  | 4671/6470 [58:03<20:24,  1.47it/s]
4|4672|Loss: 0.7785645127296448:  72%|███████▏  | 4672/6470 [58:04<21:19,  1.40it/s]
4|4673|Loss: 0.8297988772392273:  72%|███████▏  | 4672/6470 [58:04<21:19,  1.40it/s]
4|4673|Loss: 0.8297988772392273:  72%|███████▏  | 4673/6470 [58:04<20:45,  1.44it/s]
4|4674|Loss: 0.8200765252113342:  72%|███████▏  | 4673/6470 [58:05<20:45,  1.44it/s]
4|4674|Loss: 0.8200765252113342:  72%|███████▏  | 4674/6470 [58:05<20:21,  1.47it/s]
4|4675|Loss: 0.8680441975593567:  72%|███████▏  | 4674/6470 [58:05<20:21,  1.47it/s]
4|4675|Loss: 0.8680441975593567:  72%|███████▏  | 4675/6470 [58:05<19:04,  1.57it/s]
4|4676|Loss: 0.8457056879997253:  72%|███████▏  | 4675/6470 [58:06<19:04,  1.57it/s]
4|4676|Loss: 0.8457056879997253:  72%|███████▏  | 4676/6470 [58:06<20:25,  1.46it/s]
4|4677|Loss: 0.9109699130058289:  72%|███████▏  | 4676/6470 [58:07<20:25,  1.46it/s]
4|4677|Loss: 0.9109699130058289:  72%|███████▏  | 4677/6470 [58:07<21:31,  1.39it/s]
4|4678|Loss: 0.7361673712730408:  72%|███████▏  | 4677/6470 [58:07<21:31,  1.39it/s]
4|4678|Loss: 0.7361673712730408:  72%|███████▏  | 4678/6470 [58:08<21:59,  1.36it/s]
4|4679|Loss: 0.8663651943206787:  72%|███████▏  | 4678/6470 [58:08<21:59,  1.36it/s]
4|4679|Loss: 0.8663651943206787:  72%|███████▏  | 4679/6470 [58:09<22:04,  1.35it/s]
4|4680|Loss: 0.9288409948348999:  72%|███████▏  | 4679/6470 [58:09<22:04,  1.35it/s]
4|4680|Loss: 0.9288409948348999:  72%|███████▏  | 4680/6470 [58:09<23:14,  1.28it/s]
4|4681|Loss: 0.9518659710884094:  72%|███████▏  | 4680/6470 [58:10<23:14,  1.28it/s]
4|4681|Loss: 0.9518659710884094:  72%|███████▏  | 4681/6470 [58:10<22:14,  1.34it/s]
4|4682|Loss: 1.0103918313980103:  72%|███████▏  | 4681/6470 [58:11<22:14,  1.34it/s]
4|4682|Loss: 1.0103918313980103:  72%|███████▏  | 4682/6470 [58:11<23:04,  1.29it/s]
4|4683|Loss: 0.8725290298461914:  72%|███████▏  | 4682/6470 [58:11<23:04,  1.29it/s]
4|4683|Loss: 0.8725290298461914:  72%|███████▏  | 4683/6470 [58:12<23:31,  1.27it/s]
4|4684|Loss: 0.6682112216949463:  72%|███████▏  | 4683/6470 [58:12<23:31,  1.27it/s]
4|4684|Loss: 0.6682112216949463:  72%|███████▏  | 4684/6470 [58:13<24:31,  1.21it/s]
4|4685|Loss: 0.7712222933769226:  72%|███████▏  | 4684/6470 [58:13<24:31,  1.21it/s]
4|4685|Loss: 0.7712222933769226:  72%|███████▏  | 4685/6470 [58:13<24:25,  1.22it/s]
4|4686|Loss: 0.6556837558746338:  72%|███████▏  | 4685/6470 [58:14<24:25,  1.22it/s]
4|4686|Loss: 0.6556837558746338:  72%|███████▏  | 4686/6470 [58:14<24:04,  1.23it/s]
4|4687|Loss: 0.8390556573867798:  72%|███████▏  | 4686/6470 [58:15<24:04,  1.23it/s]
4|4687|Loss: 0.8390556573867798:  72%|███████▏  | 4687/6470 [58:15<21:32,  1.38it/s]
4|4688|Loss: 0.8424966335296631:  72%|███████▏  | 4687/6470 [58:15<21:32,  1.38it/s]
4|4688|Loss: 0.8424966335296631:  72%|███████▏  | 4688/6470 [58:15<20:36,  1.44it/s]
4|4689|Loss: 0.7027543187141418:  72%|███████▏  | 4688/6470 [58:16<20:36,  1.44it/s]
4|4689|Loss: 0.7027543187141418:  72%|███████▏  | 4689/6470 [58:16<19:26,  1.53it/s]
4|4690|Loss: 0.87550288438797:  72%|███████▏  | 4689/6470 [58:16<19:26,  1.53it/s]  
4|4690|Loss: 0.87550288438797:  72%|███████▏  | 4690/6470 [58:17<21:22,  1.39it/s]
4|4691|Loss: 1.0052564144134521:  72%|███████▏  | 4690/6470 [58:17<21:22,  1.39it/s]
4|4691|Loss: 1.0052564144134521:  73%|███████▎  | 4691/6470 [58:18<21:43,  1.36it/s]
4|4692|Loss: 0.8422215580940247:  73%|███████▎  | 4691/6470 [58:18<21:43,  1.36it/s]
4|4692|Loss: 0.8422215580940247:  73%|███████▎  | 4692/6470 [58:19<23:06,  1.28it/s]
4|4693|Loss: 0.8990111351013184:  73%|███████▎  | 4692/6470 [58:19<23:06,  1.28it/s]
4|4693|Loss: 0.8990111351013184:  73%|███████▎  | 4693/6470 [58:19<22:42,  1.30it/s]
4|4694|Loss: 0.7239193916320801:  73%|███████▎  | 4693/6470 [58:20<22:42,  1.30it/s]
4|4694|Loss: 0.7239193916320801:  73%|███████▎  | 4694/6470 [58:20<24:34,  1.20it/s]
4|4695|Loss: 0.8715499639511108:  73%|███████▎  | 4694/6470 [58:21<24:34,  1.20it/s]
4|4695|Loss: 0.8715499639511108:  73%|███████▎  | 4695/6470 [58:21<25:13,  1.17it/s]
4|4696|Loss: 0.8638748526573181:  73%|███████▎  | 4695/6470 [58:22<25:13,  1.17it/s]
4|4696|Loss: 0.8638748526573181:  73%|███████▎  | 4696/6470 [58:22<25:09,  1.17it/s]
4|4697|Loss: 0.8505970239639282:  73%|███████▎  | 4696/6470 [58:22<25:09,  1.17it/s]
4|4697|Loss: 0.8505970239639282:  73%|███████▎  | 4697/6470 [58:23<23:53,  1.24it/s]
4|4698|Loss: 0.8282358646392822:  73%|███████▎  | 4697/6470 [58:23<23:53,  1.24it/s]
4|4698|Loss: 0.8282358646392822:  73%|███████▎  | 4698/6470 [58:23<23:40,  1.25it/s]
4|4699|Loss: 0.7977017164230347:  73%|███████▎  | 4698/6470 [58:24<23:40,  1.25it/s]
4|4699|Loss: 0.7977017164230347:  73%|███████▎  | 4699/6470 [58:24<23:19,  1.27it/s]
4|4700|Loss: 0.7516174912452698:  73%|███████▎  | 4699/6470 [58:25<23:19,  1.27it/s]
4|4700|Loss: 0.7516174912452698:  73%|███████▎  | 4700/6470 [58:25<23:09,  1.27it/s]
4|4701|Loss: 0.8545568585395813:  73%|███████▎  | 4700/6470 [58:25<23:09,  1.27it/s]
4|4701|Loss: 0.8545568585395813:  73%|███████▎  | 4701/6470 [58:26<23:22,  1.26it/s]
4|4702|Loss: 0.9255807995796204:  73%|███████▎  | 4701/6470 [58:26<23:22,  1.26it/s]
4|4702|Loss: 0.9255807995796204:  73%|███████▎  | 4702/6470 [58:26<21:43,  1.36it/s]
4|4703|Loss: 0.8230997920036316:  73%|███████▎  | 4702/6470 [58:27<21:43,  1.36it/s]
4|4703|Loss: 0.8230997920036316:  73%|███████▎  | 4703/6470 [58:27<21:26,  1.37it/s]
4|4704|Loss: 0.8860765695571899:  73%|███████▎  | 4703/6470 [58:28<21:26,  1.37it/s]
4|4704|Loss: 0.8860765695571899:  73%|███████▎  | 4704/6470 [58:28<21:51,  1.35it/s]
4|4705|Loss: 0.8938905000686646:  73%|███████▎  | 4704/6470 [58:28<21:51,  1.35it/s]
4|4705|Loss: 0.8938905000686646:  73%|███████▎  | 4705/6470 [58:29<21:19,  1.38it/s]
4|4706|Loss: 0.8854724168777466:  73%|███████▎  | 4705/6470 [58:29<21:19,  1.38it/s]
4|4706|Loss: 0.8854724168777466:  73%|███████▎  | 4706/6470 [58:29<19:48,  1.48it/s]
4|4707|Loss: 0.7676858901977539:  73%|███████▎  | 4706/6470 [58:30<19:48,  1.48it/s]
4|4707|Loss: 0.7676858901977539:  73%|███████▎  | 4707/6470 [58:30<20:23,  1.44it/s]
4|4708|Loss: 0.7636023163795471:  73%|███████▎  | 4707/6470 [58:30<20:23,  1.44it/s]
4|4708|Loss: 0.7636023163795471:  73%|███████▎  | 4708/6470 [58:31<21:45,  1.35it/s]
4|4709|Loss: 0.73198002576828:  73%|███████▎  | 4708/6470 [58:31<21:45,  1.35it/s]  
4|4709|Loss: 0.73198002576828:  73%|███████▎  | 4709/6470 [58:32<22:48,  1.29it/s]
4|4710|Loss: 0.7129281759262085:  73%|███████▎  | 4709/6470 [58:32<22:48,  1.29it/s]
4|4710|Loss: 0.7129281759262085:  73%|███████▎  | 4710/6470 [58:33<24:20,  1.20it/s]
4|4711|Loss: 0.8642232418060303:  73%|███████▎  | 4710/6470 [58:33<24:20,  1.20it/s]
4|4711|Loss: 0.8642232418060303:  73%|███████▎  | 4711/6470 [58:33<23:12,  1.26it/s]
4|4712|Loss: 0.7639825940132141:  73%|███████▎  | 4711/6470 [58:34<23:12,  1.26it/s]
4|4712|Loss: 0.7639825940132141:  73%|███████▎  | 4712/6470 [58:34<21:45,  1.35it/s]
4|4713|Loss: 0.6874929666519165:  73%|███████▎  | 4712/6470 [58:34<21:45,  1.35it/s]
4|4713|Loss: 0.6874929666519165:  73%|███████▎  | 4713/6470 [58:35<21:15,  1.38it/s]
4|4714|Loss: 0.8002916574478149:  73%|███████▎  | 4713/6470 [58:35<21:15,  1.38it/s]
4|4714|Loss: 0.8002916574478149:  73%|███████▎  | 4714/6470 [58:35<21:26,  1.36it/s]
4|4715|Loss: 0.8171797394752502:  73%|███████▎  | 4714/6470 [58:36<21:26,  1.36it/s]
4|4715|Loss: 0.8171797394752502:  73%|███████▎  | 4715/6470 [58:36<21:05,  1.39it/s]
4|4716|Loss: 0.9102998375892639:  73%|███████▎  | 4715/6470 [58:36<21:05,  1.39it/s]
4|4716|Loss: 0.9102998375892639:  73%|███████▎  | 4716/6470 [58:37<20:53,  1.40it/s]
4|4717|Loss: 0.7647827863693237:  73%|███████▎  | 4716/6470 [58:37<20:53,  1.40it/s]
4|4717|Loss: 0.7647827863693237:  73%|███████▎  | 4717/6470 [58:37<20:28,  1.43it/s]
4|4718|Loss: 0.8590768575668335:  73%|███████▎  | 4717/6470 [58:38<20:28,  1.43it/s]
4|4718|Loss: 0.8590768575668335:  73%|███████▎  | 4718/6470 [58:38<21:33,  1.35it/s]
4|4719|Loss: 0.6695971488952637:  73%|███████▎  | 4718/6470 [58:39<21:33,  1.35it/s]
4|4719|Loss: 0.6695971488952637:  73%|███████▎  | 4719/6470 [58:39<21:16,  1.37it/s]
4|4720|Loss: 0.9230560660362244:  73%|███████▎  | 4719/6470 [58:39<21:16,  1.37it/s]
4|4720|Loss: 0.9230560660362244:  73%|███████▎  | 4720/6470 [58:40<20:30,  1.42it/s]
4|4721|Loss: 0.7894745469093323:  73%|███████▎  | 4720/6470 [58:40<20:30,  1.42it/s]
4|4721|Loss: 0.7894745469093323:  73%|███████▎  | 4721/6470 [58:40<20:56,  1.39it/s]
4|4722|Loss: 0.7509346604347229:  73%|███████▎  | 4721/6470 [58:41<20:56,  1.39it/s]
4|4722|Loss: 0.7509346604347229:  73%|███████▎  | 4722/6470 [58:41<21:16,  1.37it/s]
4|4723|Loss: 0.931175708770752:  73%|███████▎  | 4722/6470 [58:41<21:16,  1.37it/s] 
4|4723|Loss: 0.931175708770752:  73%|███████▎  | 4723/6470 [58:42<18:54,  1.54it/s]
4|4724|Loss: 0.9248124361038208:  73%|███████▎  | 4723/6470 [58:42<18:54,  1.54it/s]
4|4724|Loss: 0.9248124361038208:  73%|███████▎  | 4724/6470 [58:42<20:15,  1.44it/s]
4|4725|Loss: 0.9741753339767456:  73%|███████▎  | 4724/6470 [58:43<20:15,  1.44it/s]
4|4725|Loss: 0.9741753339767456:  73%|███████▎  | 4725/6470 [58:43<21:08,  1.38it/s]
4|4726|Loss: 1.048897624015808:  73%|███████▎  | 4725/6470 [58:44<21:08,  1.38it/s] 
4|4726|Loss: 1.048897624015808:  73%|███████▎  | 4726/6470 [58:44<21:26,  1.36it/s]
4|4727|Loss: 0.6997485756874084:  73%|███████▎  | 4726/6470 [58:44<21:26,  1.36it/s]
4|4727|Loss: 0.6997485756874084:  73%|███████▎  | 4727/6470 [58:45<21:12,  1.37it/s]
4|4728|Loss: 0.8047257661819458:  73%|███████▎  | 4727/6470 [58:45<21:12,  1.37it/s]
4|4728|Loss: 0.8047257661819458:  73%|███████▎  | 4728/6470 [58:45<20:02,  1.45it/s]
4|4729|Loss: 0.9143060445785522:  73%|███████▎  | 4728/6470 [58:46<20:02,  1.45it/s]
4|4729|Loss: 0.9143060445785522:  73%|███████▎  | 4729/6470 [58:46<19:34,  1.48it/s]
4|4730|Loss: 0.6680631637573242:  73%|███████▎  | 4729/6470 [58:46<19:34,  1.48it/s]
4|4730|Loss: 0.6680631637573242:  73%|███████▎  | 4730/6470 [58:47<20:08,  1.44it/s]
4|4731|Loss: 0.8189337849617004:  73%|███████▎  | 4730/6470 [58:47<20:08,  1.44it/s]
4|4731|Loss: 0.8189337849617004:  73%|███████▎  | 4731/6470 [58:47<21:29,  1.35it/s]
4|4732|Loss: 0.6954862475395203:  73%|███████▎  | 4731/6470 [58:48<21:29,  1.35it/s]
4|4732|Loss: 0.6954862475395203:  73%|███████▎  | 4732/6470 [58:48<24:03,  1.20it/s]
4|4733|Loss: 0.8280709981918335:  73%|███████▎  | 4732/6470 [58:49<24:03,  1.20it/s]
4|4733|Loss: 0.8280709981918335:  73%|███████▎  | 4733/6470 [58:49<24:21,  1.19it/s]
4|4734|Loss: 0.7966076731681824:  73%|███████▎  | 4733/6470 [58:50<24:21,  1.19it/s]
4|4734|Loss: 0.7966076731681824:  73%|███████▎  | 4734/6470 [58:50<23:03,  1.25it/s]
4|4735|Loss: 0.8964959383010864:  73%|███████▎  | 4734/6470 [58:50<23:03,  1.25it/s]
4|4735|Loss: 0.8964959383010864:  73%|███████▎  | 4735/6470 [58:51<22:42,  1.27it/s]
4|4736|Loss: 0.7886227965354919:  73%|███████▎  | 4735/6470 [58:51<22:42,  1.27it/s]
4|4736|Loss: 0.7886227965354919:  73%|███████▎  | 4736/6470 [58:51<21:12,  1.36it/s]
4|4737|Loss: 0.7851688861846924:  73%|███████▎  | 4736/6470 [58:52<21:12,  1.36it/s]
4|4737|Loss: 0.7851688861846924:  73%|███████▎  | 4737/6470 [58:52<21:14,  1.36it/s]
4|4738|Loss: 0.7217072248458862:  73%|███████▎  | 4737/6470 [58:53<21:14,  1.36it/s]
4|4738|Loss: 0.7217072248458862:  73%|███████▎  | 4738/6470 [58:53<20:50,  1.39it/s]
4|4739|Loss: 0.6627522706985474:  73%|███████▎  | 4738/6470 [58:53<20:50,  1.39it/s]
4|4739|Loss: 0.6627522706985474:  73%|███████▎  | 4739/6470 [58:54<23:13,  1.24it/s]
4|4740|Loss: 0.8439719676971436:  73%|███████▎  | 4739/6470 [58:54<23:13,  1.24it/s]
4|4740|Loss: 0.8439719676971436:  73%|███████▎  | 4740/6470 [58:55<22:48,  1.26it/s]
4|4741|Loss: 0.8387935757637024:  73%|███████▎  | 4740/6470 [58:55<22:48,  1.26it/s]
4|4741|Loss: 0.8387935757637024:  73%|███████▎  | 4741/6470 [58:55<21:44,  1.33it/s]
4|4742|Loss: 0.8710595965385437:  73%|███████▎  | 4741/6470 [58:56<21:44,  1.33it/s]
4|4742|Loss: 0.8710595965385437:  73%|███████▎  | 4742/6470 [58:56<20:27,  1.41it/s]
4|4743|Loss: 0.8010024428367615:  73%|███████▎  | 4742/6470 [58:56<20:27,  1.41it/s]
4|4743|Loss: 0.8010024428367615:  73%|███████▎  | 4743/6470 [58:57<20:52,  1.38it/s]
4|4744|Loss: 0.8453556895256042:  73%|███████▎  | 4743/6470 [58:57<20:52,  1.38it/s]
4|4744|Loss: 0.8453556895256042:  73%|███████▎  | 4744/6470 [58:57<20:53,  1.38it/s]
4|4745|Loss: 0.759856641292572:  73%|███████▎  | 4744/6470 [58:58<20:53,  1.38it/s] 
4|4745|Loss: 0.759856641292572:  73%|███████▎  | 4745/6470 [58:58<19:36,  1.47it/s]
4|4746|Loss: 0.9102154970169067:  73%|███████▎  | 4745/6470 [58:58<19:36,  1.47it/s]
4|4746|Loss: 0.9102154970169067:  73%|███████▎  | 4746/6470 [58:59<20:10,  1.42it/s]
4|4747|Loss: 0.7967023849487305:  73%|███████▎  | 4746/6470 [58:59<20:10,  1.42it/s]
4|4747|Loss: 0.7967023849487305:  73%|███████▎  | 4747/6470 [59:00<21:30,  1.33it/s]
4|4748|Loss: 0.9649186134338379:  73%|███████▎  | 4747/6470 [59:00<21:30,  1.33it/s]
4|4748|Loss: 0.9649186134338379:  73%|███████▎  | 4748/6470 [59:00<21:40,  1.32it/s]
4|4749|Loss: 0.8141995668411255:  73%|███████▎  | 4748/6470 [59:01<21:40,  1.32it/s]
4|4749|Loss: 0.8141995668411255:  73%|███████▎  | 4749/6470 [59:01<19:54,  1.44it/s]
4|4750|Loss: 0.7506818175315857:  73%|███████▎  | 4749/6470 [59:01<19:54,  1.44it/s]
4|4750|Loss: 0.7506818175315857:  73%|███████▎  | 4750/6470 [59:02<21:11,  1.35it/s]
4|4751|Loss: 0.8096915483474731:  73%|███████▎  | 4750/6470 [59:02<21:11,  1.35it/s]
4|4751|Loss: 0.8096915483474731:  73%|███████▎  | 4751/6470 [59:03<21:47,  1.31it/s]
4|4752|Loss: 0.747520387172699:  73%|███████▎  | 4751/6470 [59:03<21:47,  1.31it/s] 
4|4752|Loss: 0.747520387172699:  73%|███████▎  | 4752/6470 [59:03<21:48,  1.31it/s]
4|4753|Loss: 0.8201014399528503:  73%|███████▎  | 4752/6470 [59:04<21:48,  1.31it/s]
4|4753|Loss: 0.8201014399528503:  73%|███████▎  | 4753/6470 [59:04<21:21,  1.34it/s]
4|4754|Loss: 0.8067145347595215:  73%|███████▎  | 4753/6470 [59:04<21:21,  1.34it/s]
4|4754|Loss: 0.8067145347595215:  73%|███████▎  | 4754/6470 [59:05<21:20,  1.34it/s]
4|4755|Loss: 0.9692203998565674:  73%|███████▎  | 4754/6470 [59:05<21:20,  1.34it/s]
4|4755|Loss: 0.9692203998565674:  73%|███████▎  | 4755/6470 [59:05<20:44,  1.38it/s]
4|4756|Loss: 0.867891788482666:  73%|███████▎  | 4755/6470 [59:06<20:44,  1.38it/s] 
4|4756|Loss: 0.867891788482666:  74%|███████▎  | 4756/6470 [59:06<22:25,  1.27it/s]
4|4757|Loss: 0.9088261127471924:  74%|███████▎  | 4756/6470 [59:07<22:25,  1.27it/s]
4|4757|Loss: 0.9088261127471924:  74%|███████▎  | 4757/6470 [59:07<22:18,  1.28it/s]
4|4758|Loss: 0.7029323577880859:  74%|███████▎  | 4757/6470 [59:07<22:18,  1.28it/s]
4|4758|Loss: 0.7029323577880859:  74%|███████▎  | 4758/6470 [59:08<20:03,  1.42it/s]
4|4759|Loss: 0.9194192886352539:  74%|███████▎  | 4758/6470 [59:08<20:03,  1.42it/s]
4|4759|Loss: 0.9194192886352539:  74%|███████▎  | 4759/6470 [59:08<18:51,  1.51it/s]
4|4760|Loss: 0.7120686769485474:  74%|███████▎  | 4759/6470 [59:09<18:51,  1.51it/s]
4|4760|Loss: 0.7120686769485474:  74%|███████▎  | 4760/6470 [59:09<18:09,  1.57it/s]
4|4761|Loss: 0.746405303478241:  74%|███████▎  | 4760/6470 [59:09<18:09,  1.57it/s] 
4|4761|Loss: 0.746405303478241:  74%|███████▎  | 4761/6470 [59:10<19:35,  1.45it/s]
4|4762|Loss: 0.6694466471672058:  74%|███████▎  | 4761/6470 [59:10<19:35,  1.45it/s]
4|4762|Loss: 0.6694466471672058:  74%|███████▎  | 4762/6470 [59:10<17:43,  1.61it/s]
4|4763|Loss: 0.8284252882003784:  74%|███████▎  | 4762/6470 [59:10<17:43,  1.61it/s]
4|4763|Loss: 0.8284252882003784:  74%|███████▎  | 4763/6470 [59:11<18:27,  1.54it/s]
4|4764|Loss: 0.8522570133209229:  74%|███████▎  | 4763/6470 [59:11<18:27,  1.54it/s]
4|4764|Loss: 0.8522570133209229:  74%|███████▎  | 4764/6470 [59:12<19:46,  1.44it/s]
4|4765|Loss: 0.9348325729370117:  74%|███████▎  | 4764/6470 [59:12<19:46,  1.44it/s]
4|4765|Loss: 0.9348325729370117:  74%|███████▎  | 4765/6470 [59:12<20:34,  1.38it/s]
4|4766|Loss: 0.9701264500617981:  74%|███████▎  | 4765/6470 [59:13<20:34,  1.38it/s]
4|4766|Loss: 0.9701264500617981:  74%|███████▎  | 4766/6470 [59:13<20:20,  1.40it/s]
4|4767|Loss: 0.7467222213745117:  74%|███████▎  | 4766/6470 [59:13<20:20,  1.40it/s]
4|4767|Loss: 0.7467222213745117:  74%|███████▎  | 4767/6470 [59:14<19:02,  1.49it/s]
4|4768|Loss: 0.833535373210907:  74%|███████▎  | 4767/6470 [59:14<19:02,  1.49it/s] 
4|4768|Loss: 0.833535373210907:  74%|███████▎  | 4768/6470 [59:14<19:47,  1.43it/s]
4|4769|Loss: 0.7872053980827332:  74%|███████▎  | 4768/6470 [59:15<19:47,  1.43it/s]
4|4769|Loss: 0.7872053980827332:  74%|███████▎  | 4769/6470 [59:15<17:47,  1.59it/s]
4|4770|Loss: 0.9779971241950989:  74%|███████▎  | 4769/6470 [59:15<17:47,  1.59it/s]
4|4770|Loss: 0.9779971241950989:  74%|███████▎  | 4770/6470 [59:15<17:50,  1.59it/s]
4|4771|Loss: 0.8163131475448608:  74%|███████▎  | 4770/6470 [59:16<17:50,  1.59it/s]
4|4771|Loss: 0.8163131475448608:  74%|███████▎  | 4771/6470 [59:16<19:42,  1.44it/s]
4|4772|Loss: 0.7908761501312256:  74%|███████▎  | 4771/6470 [59:17<19:42,  1.44it/s]
4|4772|Loss: 0.7908761501312256:  74%|███████▍  | 4772/6470 [59:17<19:56,  1.42it/s]
4|4773|Loss: 0.8110848665237427:  74%|███████▍  | 4772/6470 [59:17<19:56,  1.42it/s]
4|4773|Loss: 0.8110848665237427:  74%|███████▍  | 4773/6470 [59:18<20:20,  1.39it/s]
4|4774|Loss: 0.8424031734466553:  74%|███████▍  | 4773/6470 [59:18<20:20,  1.39it/s]
4|4774|Loss: 0.8424031734466553:  74%|███████▍  | 4774/6470 [59:19<20:42,  1.36it/s]
4|4775|Loss: 0.8024653196334839:  74%|███████▍  | 4774/6470 [59:19<20:42,  1.36it/s]
4|4775|Loss: 0.8024653196334839:  74%|███████▍  | 4775/6470 [59:19<20:26,  1.38it/s]
4|4776|Loss: 0.7387781143188477:  74%|███████▍  | 4775/6470 [59:20<20:26,  1.38it/s]
4|4776|Loss: 0.7387781143188477:  74%|███████▍  | 4776/6470 [59:20<21:37,  1.31it/s]
4|4777|Loss: 0.8313339948654175:  74%|███████▍  | 4776/6470 [59:21<21:37,  1.31it/s]
4|4777|Loss: 0.8313339948654175:  74%|███████▍  | 4777/6470 [59:21<21:21,  1.32it/s]
4|4778|Loss: 0.7955211400985718:  74%|███████▍  | 4777/6470 [59:21<21:21,  1.32it/s]
4|4778|Loss: 0.7955211400985718:  74%|███████▍  | 4778/6470 [59:22<20:53,  1.35it/s]
4|4779|Loss: 0.660356342792511:  74%|███████▍  | 4778/6470 [59:22<20:53,  1.35it/s] 
4|4779|Loss: 0.660356342792511:  74%|███████▍  | 4779/6470 [59:22<20:59,  1.34it/s]
4|4780|Loss: 0.6753633618354797:  74%|███████▍  | 4779/6470 [59:23<20:59,  1.34it/s]
4|4780|Loss: 0.6753633618354797:  74%|███████▍  | 4780/6470 [59:23<19:47,  1.42it/s]
4|4781|Loss: 0.8604742884635925:  74%|███████▍  | 4780/6470 [59:23<19:47,  1.42it/s]
4|4781|Loss: 0.8604742884635925:  74%|███████▍  | 4781/6470 [59:24<18:58,  1.48it/s]
4|4782|Loss: 0.9227420687675476:  74%|███████▍  | 4781/6470 [59:24<18:58,  1.48it/s]
4|4782|Loss: 0.9227420687675476:  74%|███████▍  | 4782/6470 [59:24<18:50,  1.49it/s]
4|4783|Loss: 0.7210907340049744:  74%|███████▍  | 4782/6470 [59:25<18:50,  1.49it/s]
4|4783|Loss: 0.7210907340049744:  74%|███████▍  | 4783/6470 [59:25<18:15,  1.54it/s]
4|4784|Loss: 0.7548978924751282:  74%|███████▍  | 4783/6470 [59:25<18:15,  1.54it/s]
4|4784|Loss: 0.7548978924751282:  74%|███████▍  | 4784/6470 [59:26<18:57,  1.48it/s]
4|4785|Loss: 0.7741240859031677:  74%|███████▍  | 4784/6470 [59:26<18:57,  1.48it/s]
4|4785|Loss: 0.7741240859031677:  74%|███████▍  | 4785/6470 [59:26<19:32,  1.44it/s]
4|4786|Loss: 0.8447520732879639:  74%|███████▍  | 4785/6470 [59:27<19:32,  1.44it/s]
4|4786|Loss: 0.8447520732879639:  74%|███████▍  | 4786/6470 [59:27<19:33,  1.43it/s]
4|4787|Loss: 0.8316526412963867:  74%|███████▍  | 4786/6470 [59:27<19:33,  1.43it/s]
4|4787|Loss: 0.8316526412963867:  74%|███████▍  | 4787/6470 [59:28<20:58,  1.34it/s]
4|4788|Loss: 0.7205199003219604:  74%|███████▍  | 4787/6470 [59:28<20:58,  1.34it/s]
4|4788|Loss: 0.7205199003219604:  74%|███████▍  | 4788/6470 [59:29<22:12,  1.26it/s]
4|4789|Loss: 0.7597694396972656:  74%|███████▍  | 4788/6470 [59:29<22:12,  1.26it/s]
4|4789|Loss: 0.7597694396972656:  74%|███████▍  | 4789/6470 [59:30<22:08,  1.27it/s]
4|4790|Loss: 0.8330441117286682:  74%|███████▍  | 4789/6470 [59:30<22:08,  1.27it/s]
4|4790|Loss: 0.8330441117286682:  74%|███████▍  | 4790/6470 [59:30<22:51,  1.22it/s]
4|4791|Loss: 0.6836793422698975:  74%|███████▍  | 4790/6470 [59:31<22:51,  1.22it/s]
4|4791|Loss: 0.6836793422698975:  74%|███████▍  | 4791/6470 [59:31<21:07,  1.32it/s]
4|4792|Loss: 0.673621416091919:  74%|███████▍  | 4791/6470 [59:31<21:07,  1.32it/s] 
4|4792|Loss: 0.673621416091919:  74%|███████▍  | 4792/6470 [59:32<19:44,  1.42it/s]
4|4793|Loss: 0.7874811887741089:  74%|███████▍  | 4792/6470 [59:32<19:44,  1.42it/s]
4|4793|Loss: 0.7874811887741089:  74%|███████▍  | 4793/6470 [59:32<19:32,  1.43it/s]
4|4794|Loss: 0.9095252156257629:  74%|███████▍  | 4793/6470 [59:33<19:32,  1.43it/s]
4|4794|Loss: 0.9095252156257629:  74%|███████▍  | 4794/6470 [59:33<19:18,  1.45it/s]
4|4795|Loss: 0.8275573253631592:  74%|███████▍  | 4794/6470 [59:33<19:18,  1.45it/s]
4|4795|Loss: 0.8275573253631592:  74%|███████▍  | 4795/6470 [59:34<19:15,  1.45it/s]
4|4796|Loss: 0.8590751886367798:  74%|███████▍  | 4795/6470 [59:34<19:15,  1.45it/s]
4|4796|Loss: 0.8590751886367798:  74%|███████▍  | 4796/6470 [59:34<19:04,  1.46it/s]
4|4797|Loss: 0.900771975517273:  74%|███████▍  | 4796/6470 [59:35<19:04,  1.46it/s] 
4|4797|Loss: 0.900771975517273:  74%|███████▍  | 4797/6470 [59:35<19:30,  1.43it/s]
4|4798|Loss: 1.0815292596817017:  74%|███████▍  | 4797/6470 [59:35<19:30,  1.43it/s]
4|4798|Loss: 1.0815292596817017:  74%|███████▍  | 4798/6470 [59:36<19:34,  1.42it/s]
4|4799|Loss: 0.8464255928993225:  74%|███████▍  | 4798/6470 [59:36<19:34,  1.42it/s]
4|4799|Loss: 0.8464255928993225:  74%|███████▍  | 4799/6470 [59:36<19:01,  1.46it/s]
4|4800|Loss: 0.7908419370651245:  74%|███████▍  | 4799/6470 [59:37<19:01,  1.46it/s]
4|4800|Loss: 0.7908419370651245:  74%|███████▍  | 4800/6470 [59:37<18:29,  1.51it/s]
4|4801|Loss: 0.7833489179611206:  74%|███████▍  | 4800/6470 [59:37<18:29,  1.51it/s]
4|4801|Loss: 0.7833489179611206:  74%|███████▍  | 4801/6470 [59:38<19:04,  1.46it/s]
4|4802|Loss: 0.7382996082305908:  74%|███████▍  | 4801/6470 [59:38<19:04,  1.46it/s]
4|4802|Loss: 0.7382996082305908:  74%|███████▍  | 4802/6470 [59:38<19:14,  1.44it/s]
4|4803|Loss: 0.8356167674064636:  74%|███████▍  | 4802/6470 [59:39<19:14,  1.44it/s]
4|4803|Loss: 0.8356167674064636:  74%|███████▍  | 4803/6470 [59:39<19:09,  1.45it/s]
4|4804|Loss: 0.8865200877189636:  74%|███████▍  | 4803/6470 [59:40<19:09,  1.45it/s]
4|4804|Loss: 0.8865200877189636:  74%|███████▍  | 4804/6470 [59:40<19:11,  1.45it/s]
4|4805|Loss: 1.0336874723434448:  74%|███████▍  | 4804/6470 [59:40<19:11,  1.45it/s]
4|4805|Loss: 1.0336874723434448:  74%|███████▍  | 4805/6470 [59:40<18:36,  1.49it/s]
4|4806|Loss: 0.7653913497924805:  74%|███████▍  | 4805/6470 [59:41<18:36,  1.49it/s]
4|4806|Loss: 0.7653913497924805:  74%|███████▍  | 4806/6470 [59:41<18:07,  1.53it/s]
4|4807|Loss: 0.7580732703208923:  74%|███████▍  | 4806/6470 [59:41<18:07,  1.53it/s]
4|4807|Loss: 0.7580732703208923:  74%|███████▍  | 4807/6470 [59:42<19:08,  1.45it/s]
4|4808|Loss: 0.7886319160461426:  74%|███████▍  | 4807/6470 [59:42<19:08,  1.45it/s]
4|4808|Loss: 0.7886319160461426:  74%|███████▍  | 4808/6470 [59:43<20:10,  1.37it/s]
4|4809|Loss: 0.8030629754066467:  74%|███████▍  | 4808/6470 [59:43<20:10,  1.37it/s]
4|4809|Loss: 0.8030629754066467:  74%|███████▍  | 4809/6470 [59:43<19:07,  1.45it/s]
4|4810|Loss: 0.9603458642959595:  74%|███████▍  | 4809/6470 [59:44<19:07,  1.45it/s]
4|4810|Loss: 0.9603458642959595:  74%|███████▍  | 4810/6470 [59:44<17:37,  1.57it/s]
4|4811|Loss: 0.7678675055503845:  74%|███████▍  | 4810/6470 [59:44<17:37,  1.57it/s]
4|4811|Loss: 0.7678675055503845:  74%|███████▍  | 4811/6470 [59:44<16:42,  1.65it/s]
4|4812|Loss: 0.8938788771629333:  74%|███████▍  | 4811/6470 [59:45<16:42,  1.65it/s]
4|4812|Loss: 0.8938788771629333:  74%|███████▍  | 4812/6470 [59:45<15:48,  1.75it/s]
4|4813|Loss: 1.0839424133300781:  74%|███████▍  | 4812/6470 [59:45<15:48,  1.75it/s]
4|4813|Loss: 1.0839424133300781:  74%|███████▍  | 4813/6470 [59:46<17:03,  1.62it/s]
4|4814|Loss: 1.0281567573547363:  74%|███████▍  | 4813/6470 [59:46<17:03,  1.62it/s]
4|4814|Loss: 1.0281567573547363:  74%|███████▍  | 4814/6470 [59:46<17:54,  1.54it/s]
4|4815|Loss: 0.7982330918312073:  74%|███████▍  | 4814/6470 [59:47<17:54,  1.54it/s]
4|4815|Loss: 0.7982330918312073:  74%|███████▍  | 4815/6470 [59:47<19:00,  1.45it/s]
4|4816|Loss: 0.9190336465835571:  74%|███████▍  | 4815/6470 [59:47<19:00,  1.45it/s]
4|4816|Loss: 0.9190336465835571:  74%|███████▍  | 4816/6470 [59:48<19:03,  1.45it/s]
4|4817|Loss: 0.7664994597434998:  74%|███████▍  | 4816/6470 [59:48<19:03,  1.45it/s]
4|4817|Loss: 0.7664994597434998:  74%|███████▍  | 4817/6470 [59:48<18:32,  1.49it/s]
4|4818|Loss: 0.8689965605735779:  74%|███████▍  | 4817/6470 [59:49<18:32,  1.49it/s]
4|4818|Loss: 0.8689965605735779:  74%|███████▍  | 4818/6470 [59:49<19:16,  1.43it/s]
4|4819|Loss: 0.6878778338432312:  74%|███████▍  | 4818/6470 [59:50<19:16,  1.43it/s]
4|4819|Loss: 0.6878778338432312:  74%|███████▍  | 4819/6470 [59:50<20:01,  1.37it/s]
4|4820|Loss: 0.7201767563819885:  74%|███████▍  | 4819/6470 [59:50<20:01,  1.37it/s]
4|4820|Loss: 0.7201767563819885:  74%|███████▍  | 4820/6470 [59:51<21:07,  1.30it/s]
4|4821|Loss: 0.8610062003135681:  74%|███████▍  | 4820/6470 [59:51<21:07,  1.30it/s]
4|4821|Loss: 0.8610062003135681:  75%|███████▍  | 4821/6470 [59:52<21:57,  1.25it/s]
4|4822|Loss: 0.8805544972419739:  75%|███████▍  | 4821/6470 [59:52<21:57,  1.25it/s]
4|4822|Loss: 0.8805544972419739:  75%|███████▍  | 4822/6470 [59:52<21:08,  1.30it/s]
4|4823|Loss: 0.737337589263916:  75%|███████▍  | 4822/6470 [59:53<21:08,  1.30it/s] 
4|4823|Loss: 0.737337589263916:  75%|███████▍  | 4823/6470 [59:53<22:53,  1.20it/s]
4|4824|Loss: 0.8572936058044434:  75%|███████▍  | 4823/6470 [59:54<22:53,  1.20it/s]
4|4824|Loss: 0.8572936058044434:  75%|███████▍  | 4824/6470 [59:54<23:08,  1.19it/s]
4|4825|Loss: 0.7119401693344116:  75%|███████▍  | 4824/6470 [59:55<23:08,  1.19it/s]
4|4825|Loss: 0.7119401693344116:  75%|███████▍  | 4825/6470 [59:55<23:45,  1.15it/s]
4|4826|Loss: 0.5637137293815613:  75%|███████▍  | 4825/6470 [59:56<23:45,  1.15it/s]
4|4826|Loss: 0.5637137293815613:  75%|███████▍  | 4826/6470 [59:56<22:40,  1.21it/s]
4|4827|Loss: 0.8903505206108093:  75%|███████▍  | 4826/6470 [59:56<22:40,  1.21it/s]
4|4827|Loss: 0.8903505206108093:  75%|███████▍  | 4827/6470 [59:57<23:22,  1.17it/s]
4|4828|Loss: 0.6794144511222839:  75%|███████▍  | 4827/6470 [59:57<23:22,  1.17it/s]
4|4828|Loss: 0.6794144511222839:  75%|███████▍  | 4828/6470 [59:58<22:19,  1.23it/s]
4|4829|Loss: 0.8161422610282898:  75%|███████▍  | 4828/6470 [59:58<22:19,  1.23it/s]
4|4829|Loss: 0.8161422610282898:  75%|███████▍  | 4829/6470 [59:58<20:42,  1.32it/s]
4|4830|Loss: 0.7824259996414185:  75%|███████▍  | 4829/6470 [59:59<20:42,  1.32it/s]
4|4830|Loss: 0.7824259996414185:  75%|███████▍  | 4830/6470 [59:59<20:29,  1.33it/s]
4|4831|Loss: 0.7388342618942261:  75%|███████▍  | 4830/6470 [59:59<20:29,  1.33it/s]
4|4831|Loss: 0.7388342618942261:  75%|███████▍  | 4831/6470 [1:00:00<21:25,  1.27it/s]
4|4832|Loss: 0.7904459834098816:  75%|███████▍  | 4831/6470 [1:00:00<21:25,  1.27it/s]
4|4832|Loss: 0.7904459834098816:  75%|███████▍  | 4832/6470 [1:00:00<19:37,  1.39it/s]
4|4833|Loss: 0.7968641519546509:  75%|███████▍  | 4832/6470 [1:00:01<19:37,  1.39it/s]
4|4833|Loss: 0.7968641519546509:  75%|███████▍  | 4833/6470 [1:00:01<19:34,  1.39it/s]
4|4834|Loss: 0.7284638285636902:  75%|███████▍  | 4833/6470 [1:00:01<19:34,  1.39it/s]
4|4834|Loss: 0.7284638285636902:  75%|███████▍  | 4834/6470 [1:00:02<20:18,  1.34it/s]
4|4835|Loss: 0.7427432537078857:  75%|███████▍  | 4834/6470 [1:00:02<20:18,  1.34it/s]
4|4835|Loss: 0.7427432537078857:  75%|███████▍  | 4835/6470 [1:00:02<19:08,  1.42it/s]
4|4836|Loss: 0.7365277409553528:  75%|███████▍  | 4835/6470 [1:00:03<19:08,  1.42it/s]
4|4836|Loss: 0.7365277409553528:  75%|███████▍  | 4836/6470 [1:00:03<20:22,  1.34it/s]
4|4837|Loss: 0.8935844302177429:  75%|███████▍  | 4836/6470 [1:00:04<20:22,  1.34it/s]
4|4837|Loss: 0.8935844302177429:  75%|███████▍  | 4837/6470 [1:00:04<21:55,  1.24it/s]
4|4838|Loss: 0.8185436725616455:  75%|███████▍  | 4837/6470 [1:00:05<21:55,  1.24it/s]
4|4838|Loss: 0.8185436725616455:  75%|███████▍  | 4838/6470 [1:00:05<22:12,  1.22it/s]
4|4839|Loss: 0.7501462697982788:  75%|███████▍  | 4838/6470 [1:00:05<22:12,  1.22it/s]
4|4839|Loss: 0.7501462697982788:  75%|███████▍  | 4839/6470 [1:00:06<20:53,  1.30it/s]
4|4840|Loss: 0.8099901676177979:  75%|███████▍  | 4839/6470 [1:00:06<20:53,  1.30it/s]
4|4840|Loss: 0.8099901676177979:  75%|███████▍  | 4840/6470 [1:00:06<20:41,  1.31it/s]
4|4841|Loss: 0.8490608334541321:  75%|███████▍  | 4840/6470 [1:00:07<20:41,  1.31it/s]
4|4841|Loss: 0.8490608334541321:  75%|███████▍  | 4841/6470 [1:00:07<21:05,  1.29it/s]
4|4842|Loss: 0.8011815547943115:  75%|███████▍  | 4841/6470 [1:00:08<21:05,  1.29it/s]
4|4842|Loss: 0.8011815547943115:  75%|███████▍  | 4842/6470 [1:00:08<19:59,  1.36it/s]
4|4843|Loss: 0.8031876683235168:  75%|███████▍  | 4842/6470 [1:00:08<19:59,  1.36it/s]
4|4843|Loss: 0.8031876683235168:  75%|███████▍  | 4843/6470 [1:00:09<19:14,  1.41it/s]
4|4844|Loss: 0.6748906373977661:  75%|███████▍  | 4843/6470 [1:00:09<19:14,  1.41it/s]
4|4844|Loss: 0.6748906373977661:  75%|███████▍  | 4844/6470 [1:00:09<20:31,  1.32it/s]
4|4845|Loss: 0.8980346322059631:  75%|███████▍  | 4844/6470 [1:00:10<20:31,  1.32it/s]
4|4845|Loss: 0.8980346322059631:  75%|███████▍  | 4845/6470 [1:00:10<20:29,  1.32it/s]
4|4846|Loss: 0.7883548736572266:  75%|███████▍  | 4845/6470 [1:00:11<20:29,  1.32it/s]
4|4846|Loss: 0.7883548736572266:  75%|███████▍  | 4846/6470 [1:00:11<20:46,  1.30it/s]
4|4847|Loss: 0.7385128736495972:  75%|███████▍  | 4846/6470 [1:00:11<20:46,  1.30it/s]
4|4847|Loss: 0.7385128736495972:  75%|███████▍  | 4847/6470 [1:00:12<21:10,  1.28it/s]
4|4848|Loss: 0.8311640620231628:  75%|███████▍  | 4847/6470 [1:00:12<21:10,  1.28it/s]
4|4848|Loss: 0.8311640620231628:  75%|███████▍  | 4848/6470 [1:00:13<22:07,  1.22it/s]
4|4849|Loss: 0.9101243615150452:  75%|███████▍  | 4848/6470 [1:00:13<22:07,  1.22it/s]
4|4849|Loss: 0.9101243615150452:  75%|███████▍  | 4849/6470 [1:00:13<21:38,  1.25it/s]
4|4850|Loss: 0.9596150517463684:  75%|███████▍  | 4849/6470 [1:00:14<21:38,  1.25it/s]
4|4850|Loss: 0.9596150517463684:  75%|███████▍  | 4850/6470 [1:00:14<20:48,  1.30it/s]
4|4851|Loss: 0.8742907643318176:  75%|███████▍  | 4850/6470 [1:00:15<20:48,  1.30it/s]
4|4851|Loss: 0.8742907643318176:  75%|███████▍  | 4851/6470 [1:00:15<22:27,  1.20it/s]
4|4852|Loss: 0.83760666847229:  75%|███████▍  | 4851/6470 [1:00:16<22:27,  1.20it/s]  
4|4852|Loss: 0.83760666847229:  75%|███████▍  | 4852/6470 [1:00:16<21:29,  1.25it/s]
4|4853|Loss: 0.692792534828186:  75%|███████▍  | 4852/6470 [1:00:16<21:29,  1.25it/s]
4|4853|Loss: 0.692792534828186:  75%|███████▌  | 4853/6470 [1:00:17<21:53,  1.23it/s]
4|4854|Loss: 0.7446191906929016:  75%|███████▌  | 4853/6470 [1:00:17<21:53,  1.23it/s]
4|4854|Loss: 0.7446191906929016:  75%|███████▌  | 4854/6470 [1:00:17<21:09,  1.27it/s]
4|4855|Loss: 0.8963444828987122:  75%|███████▌  | 4854/6470 [1:00:18<21:09,  1.27it/s]
4|4855|Loss: 0.8963444828987122:  75%|███████▌  | 4855/6470 [1:00:18<20:15,  1.33it/s]
4|4856|Loss: 0.776835024356842:  75%|███████▌  | 4855/6470 [1:00:18<20:15,  1.33it/s] 
4|4856|Loss: 0.776835024356842:  75%|███████▌  | 4856/6470 [1:00:19<18:20,  1.47it/s]
4|4857|Loss: 0.8278269171714783:  75%|███████▌  | 4856/6470 [1:00:19<18:20,  1.47it/s]
4|4857|Loss: 0.8278269171714783:  75%|███████▌  | 4857/6470 [1:00:19<19:05,  1.41it/s]
4|4858|Loss: 0.6630436778068542:  75%|███████▌  | 4857/6470 [1:00:20<19:05,  1.41it/s]
4|4858|Loss: 0.6630436778068542:  75%|███████▌  | 4858/6470 [1:00:20<19:44,  1.36it/s]
4|4859|Loss: 0.8042675852775574:  75%|███████▌  | 4858/6470 [1:00:21<19:44,  1.36it/s]
4|4859|Loss: 0.8042675852775574:  75%|███████▌  | 4859/6470 [1:00:21<20:09,  1.33it/s]
4|4860|Loss: 0.7228683829307556:  75%|███████▌  | 4859/6470 [1:00:21<20:09,  1.33it/s]
4|4860|Loss: 0.7228683829307556:  75%|███████▌  | 4860/6470 [1:00:22<19:28,  1.38it/s]
4|4861|Loss: 0.7765209078788757:  75%|███████▌  | 4860/6470 [1:00:22<19:28,  1.38it/s]
4|4861|Loss: 0.7765209078788757:  75%|███████▌  | 4861/6470 [1:00:22<18:38,  1.44it/s]
4|4862|Loss: 0.7432775497436523:  75%|███████▌  | 4861/6470 [1:00:23<18:38,  1.44it/s]
4|4862|Loss: 0.7432775497436523:  75%|███████▌  | 4862/6470 [1:00:23<18:09,  1.48it/s]
4|4863|Loss: 0.8914273977279663:  75%|███████▌  | 4862/6470 [1:00:23<18:09,  1.48it/s]
4|4863|Loss: 0.8914273977279663:  75%|███████▌  | 4863/6470 [1:00:24<18:16,  1.47it/s]
4|4864|Loss: 0.8450255393981934:  75%|███████▌  | 4863/6470 [1:00:24<18:16,  1.47it/s]
4|4864|Loss: 0.8450255393981934:  75%|███████▌  | 4864/6470 [1:00:24<18:55,  1.41it/s]
4|4865|Loss: 0.9195742607116699:  75%|███████▌  | 4864/6470 [1:00:25<18:55,  1.41it/s]
4|4865|Loss: 0.9195742607116699:  75%|███████▌  | 4865/6470 [1:00:25<20:18,  1.32it/s]
4|4866|Loss: 0.7502754926681519:  75%|███████▌  | 4865/6470 [1:00:26<20:18,  1.32it/s]
4|4866|Loss: 0.7502754926681519:  75%|███████▌  | 4866/6470 [1:00:26<19:24,  1.38it/s]
4|4867|Loss: 0.9456614255905151:  75%|███████▌  | 4866/6470 [1:00:26<19:24,  1.38it/s]
4|4867|Loss: 0.9456614255905151:  75%|███████▌  | 4867/6470 [1:00:27<20:45,  1.29it/s]
4|4868|Loss: 0.887877345085144:  75%|███████▌  | 4867/6470 [1:00:27<20:45,  1.29it/s] 
4|4868|Loss: 0.887877345085144:  75%|███████▌  | 4868/6470 [1:00:28<20:39,  1.29it/s]
4|4869|Loss: 0.7905339002609253:  75%|███████▌  | 4868/6470 [1:00:28<20:39,  1.29it/s]
4|4869|Loss: 0.7905339002609253:  75%|███████▌  | 4869/6470 [1:00:28<20:47,  1.28it/s]
4|4870|Loss: 0.7727009057998657:  75%|███████▌  | 4869/6470 [1:00:29<20:47,  1.28it/s]
4|4870|Loss: 0.7727009057998657:  75%|███████▌  | 4870/6470 [1:00:29<19:42,  1.35it/s]
4|4871|Loss: 0.7141407132148743:  75%|███████▌  | 4870/6470 [1:00:29<19:42,  1.35it/s]
4|4871|Loss: 0.7141407132148743:  75%|███████▌  | 4871/6470 [1:00:29<17:46,  1.50it/s]
4|4872|Loss: 0.7145436406135559:  75%|███████▌  | 4871/6470 [1:00:30<17:46,  1.50it/s]
4|4872|Loss: 0.7145436406135559:  75%|███████▌  | 4872/6470 [1:00:30<17:03,  1.56it/s]
4|4873|Loss: 0.8473840951919556:  75%|███████▌  | 4872/6470 [1:00:30<17:03,  1.56it/s]
4|4873|Loss: 0.8473840951919556:  75%|███████▌  | 4873/6470 [1:00:31<17:44,  1.50it/s]
4|4874|Loss: 1.0264289379119873:  75%|███████▌  | 4873/6470 [1:00:31<17:44,  1.50it/s]
4|4874|Loss: 1.0264289379119873:  75%|███████▌  | 4874/6470 [1:00:31<17:07,  1.55it/s]
4|4875|Loss: 0.8846625685691833:  75%|███████▌  | 4874/6470 [1:00:32<17:07,  1.55it/s]
4|4875|Loss: 0.8846625685691833:  75%|███████▌  | 4875/6470 [1:00:32<17:16,  1.54it/s]
4|4876|Loss: 0.6119733452796936:  75%|███████▌  | 4875/6470 [1:00:32<17:16,  1.54it/s]
4|4876|Loss: 0.6119733452796936:  75%|███████▌  | 4876/6470 [1:00:33<17:11,  1.55it/s]
4|4877|Loss: 0.8054214715957642:  75%|███████▌  | 4876/6470 [1:00:33<17:11,  1.55it/s]
4|4877|Loss: 0.8054214715957642:  75%|███████▌  | 4877/6470 [1:00:33<16:52,  1.57it/s]
4|4878|Loss: 0.8615037798881531:  75%|███████▌  | 4877/6470 [1:00:34<16:52,  1.57it/s]
4|4878|Loss: 0.8615037798881531:  75%|███████▌  | 4878/6470 [1:00:34<18:31,  1.43it/s]
4|4879|Loss: 0.6936034560203552:  75%|███████▌  | 4878/6470 [1:00:35<18:31,  1.43it/s]
4|4879|Loss: 0.6936034560203552:  75%|███████▌  | 4879/6470 [1:00:35<17:58,  1.47it/s]
4|4880|Loss: 0.8891409635543823:  75%|███████▌  | 4879/6470 [1:00:35<17:58,  1.47it/s]
4|4880|Loss: 0.8891409635543823:  75%|███████▌  | 4880/6470 [1:00:35<18:14,  1.45it/s]
4|4881|Loss: 0.8152326345443726:  75%|███████▌  | 4880/6470 [1:00:36<18:14,  1.45it/s]
4|4881|Loss: 0.8152326345443726:  75%|███████▌  | 4881/6470 [1:00:36<19:24,  1.36it/s]
4|4882|Loss: 0.8639435768127441:  75%|███████▌  | 4881/6470 [1:00:37<19:24,  1.36it/s]
4|4882|Loss: 0.8639435768127441:  75%|███████▌  | 4882/6470 [1:00:37<20:05,  1.32it/s]
4|4883|Loss: 0.9437880516052246:  75%|███████▌  | 4882/6470 [1:00:38<20:05,  1.32it/s]
4|4883|Loss: 0.9437880516052246:  75%|███████▌  | 4883/6470 [1:00:38<20:28,  1.29it/s]
4|4884|Loss: 0.7128493785858154:  75%|███████▌  | 4883/6470 [1:00:38<20:28,  1.29it/s]
4|4884|Loss: 0.7128493785858154:  75%|███████▌  | 4884/6470 [1:00:39<21:15,  1.24it/s]
4|4885|Loss: 0.8030534982681274:  75%|███████▌  | 4884/6470 [1:00:39<21:15,  1.24it/s]
4|4885|Loss: 0.8030534982681274:  76%|███████▌  | 4885/6470 [1:00:40<22:00,  1.20it/s]
4|4886|Loss: 0.8757850527763367:  76%|███████▌  | 4885/6470 [1:00:40<22:00,  1.20it/s]
4|4886|Loss: 0.8757850527763367:  76%|███████▌  | 4886/6470 [1:00:40<21:11,  1.25it/s]
4|4887|Loss: 0.8469219207763672:  76%|███████▌  | 4886/6470 [1:00:41<21:11,  1.25it/s]
4|4887|Loss: 0.8469219207763672:  76%|███████▌  | 4887/6470 [1:00:41<20:57,  1.26it/s]
4|4888|Loss: 0.7525474429130554:  76%|███████▌  | 4887/6470 [1:00:42<20:57,  1.26it/s]
4|4888|Loss: 0.7525474429130554:  76%|███████▌  | 4888/6470 [1:00:42<19:37,  1.34it/s]
4|4889|Loss: 0.8057158589363098:  76%|███████▌  | 4888/6470 [1:00:42<19:37,  1.34it/s]
4|4889|Loss: 0.8057158589363098:  76%|███████▌  | 4889/6470 [1:00:43<19:46,  1.33it/s]
4|4890|Loss: 0.8458635807037354:  76%|███████▌  | 4889/6470 [1:00:43<19:46,  1.33it/s]
4|4890|Loss: 0.8458635807037354:  76%|███████▌  | 4890/6470 [1:00:43<20:09,  1.31it/s]
4|4891|Loss: 0.7021456360816956:  76%|███████▌  | 4890/6470 [1:00:44<20:09,  1.31it/s]
4|4891|Loss: 0.7021456360816956:  76%|███████▌  | 4891/6470 [1:00:44<20:09,  1.31it/s]
4|4892|Loss: 0.875017523765564:  76%|███████▌  | 4891/6470 [1:00:45<20:09,  1.31it/s] 
4|4892|Loss: 0.875017523765564:  76%|███████▌  | 4892/6470 [1:00:45<21:12,  1.24it/s]
4|4893|Loss: 0.80987548828125:  76%|███████▌  | 4892/6470 [1:00:46<21:12,  1.24it/s] 
4|4893|Loss: 0.80987548828125:  76%|███████▌  | 4893/6470 [1:00:46<20:43,  1.27it/s]
4|4894|Loss: 0.7562518119812012:  76%|███████▌  | 4893/6470 [1:00:46<20:43,  1.27it/s]
4|4894|Loss: 0.7562518119812012:  76%|███████▌  | 4894/6470 [1:00:47<20:36,  1.27it/s]
4|4895|Loss: 0.8732785582542419:  76%|███████▌  | 4894/6470 [1:00:47<20:36,  1.27it/s]
4|4895|Loss: 0.8732785582542419:  76%|███████▌  | 4895/6470 [1:00:47<20:43,  1.27it/s]
4|4896|Loss: 1.0070298910140991:  76%|███████▌  | 4895/6470 [1:00:48<20:43,  1.27it/s]
4|4896|Loss: 1.0070298910140991:  76%|███████▌  | 4896/6470 [1:00:48<20:37,  1.27it/s]
4|4897|Loss: 0.8663223385810852:  76%|███████▌  | 4896/6470 [1:00:49<20:37,  1.27it/s]
4|4897|Loss: 0.8663223385810852:  76%|███████▌  | 4897/6470 [1:00:49<20:44,  1.26it/s]
4|4898|Loss: 0.7437373995780945:  76%|███████▌  | 4897/6470 [1:00:49<20:44,  1.26it/s]
4|4898|Loss: 0.7437373995780945:  76%|███████▌  | 4898/6470 [1:00:50<20:12,  1.30it/s]
4|4899|Loss: 0.848146915435791:  76%|███████▌  | 4898/6470 [1:00:50<20:12,  1.30it/s] 
4|4899|Loss: 0.848146915435791:  76%|███████▌  | 4899/6470 [1:00:51<20:57,  1.25it/s]
4|4900|Loss: 0.8616713881492615:  76%|███████▌  | 4899/6470 [1:00:51<20:57,  1.25it/s]
4|4900|Loss: 0.8616713881492615:  76%|███████▌  | 4900/6470 [1:00:51<20:19,  1.29it/s]
4|4901|Loss: 0.8448511362075806:  76%|███████▌  | 4900/6470 [1:00:52<20:19,  1.29it/s]
4|4901|Loss: 0.8448511362075806:  76%|███████▌  | 4901/6470 [1:00:52<18:55,  1.38it/s]
4|4902|Loss: 0.8316687345504761:  76%|███████▌  | 4901/6470 [1:00:52<18:55,  1.38it/s]
4|4902|Loss: 0.8316687345504761:  76%|███████▌  | 4902/6470 [1:00:53<19:07,  1.37it/s]
4|4903|Loss: 0.7684304118156433:  76%|███████▌  | 4902/6470 [1:00:53<19:07,  1.37it/s]
4|4903|Loss: 0.7684304118156433:  76%|███████▌  | 4903/6470 [1:00:54<21:09,  1.23it/s]
4|4904|Loss: 0.8352603316307068:  76%|███████▌  | 4903/6470 [1:00:54<21:09,  1.23it/s]
4|4904|Loss: 0.8352603316307068:  76%|███████▌  | 4904/6470 [1:00:55<21:33,  1.21it/s]
4|4905|Loss: 0.8482285141944885:  76%|███████▌  | 4904/6470 [1:00:55<21:33,  1.21it/s]
4|4905|Loss: 0.8482285141944885:  76%|███████▌  | 4905/6470 [1:00:55<20:43,  1.26it/s]
4|4906|Loss: 0.9676575064659119:  76%|███████▌  | 4905/6470 [1:00:56<20:43,  1.26it/s]
4|4906|Loss: 0.9676575064659119:  76%|███████▌  | 4906/6470 [1:00:56<19:53,  1.31it/s]
4|4907|Loss: 0.8591465950012207:  76%|███████▌  | 4906/6470 [1:00:56<19:53,  1.31it/s]
4|4907|Loss: 0.8591465950012207:  76%|███████▌  | 4907/6470 [1:00:57<19:43,  1.32it/s]
4|4908|Loss: 0.78909832239151:  76%|███████▌  | 4907/6470 [1:00:57<19:43,  1.32it/s]  
4|4908|Loss: 0.78909832239151:  76%|███████▌  | 4908/6470 [1:00:57<19:55,  1.31it/s]
4|4909|Loss: 0.7406176924705505:  76%|███████▌  | 4908/6470 [1:00:58<19:55,  1.31it/s]
4|4909|Loss: 0.7406176924705505:  76%|███████▌  | 4909/6470 [1:00:58<21:01,  1.24it/s]
4|4910|Loss: 0.6705943942070007:  76%|███████▌  | 4909/6470 [1:00:59<21:01,  1.24it/s]
4|4910|Loss: 0.6705943942070007:  76%|███████▌  | 4910/6470 [1:00:59<19:12,  1.35it/s]
4|4911|Loss: 0.6788795590400696:  76%|███████▌  | 4910/6470 [1:00:59<19:12,  1.35it/s]
4|4911|Loss: 0.6788795590400696:  76%|███████▌  | 4911/6470 [1:01:00<19:55,  1.30it/s]
4|4912|Loss: 0.8842219114303589:  76%|███████▌  | 4911/6470 [1:01:00<19:55,  1.30it/s]
4|4912|Loss: 0.8842219114303589:  76%|███████▌  | 4912/6470 [1:01:00<19:22,  1.34it/s]
4|4913|Loss: 0.9855793714523315:  76%|███████▌  | 4912/6470 [1:01:01<19:22,  1.34it/s]
4|4913|Loss: 0.9855793714523315:  76%|███████▌  | 4913/6470 [1:01:01<20:10,  1.29it/s]
4|4914|Loss: 0.8724870681762695:  76%|███████▌  | 4913/6470 [1:01:02<20:10,  1.29it/s]
4|4914|Loss: 0.8724870681762695:  76%|███████▌  | 4914/6470 [1:01:02<20:32,  1.26it/s]
4|4915|Loss: 0.932582437992096:  76%|███████▌  | 4914/6470 [1:01:03<20:32,  1.26it/s] 
4|4915|Loss: 0.932582437992096:  76%|███████▌  | 4915/6470 [1:01:03<19:50,  1.31it/s]
4|4916|Loss: 0.8314715027809143:  76%|███████▌  | 4915/6470 [1:01:03<19:50,  1.31it/s]
4|4916|Loss: 0.8314715027809143:  76%|███████▌  | 4916/6470 [1:01:04<19:04,  1.36it/s]
4|4917|Loss: 0.7692131400108337:  76%|███████▌  | 4916/6470 [1:01:04<19:04,  1.36it/s]
4|4917|Loss: 0.7692131400108337:  76%|███████▌  | 4917/6470 [1:01:04<18:13,  1.42it/s]
4|4918|Loss: 0.7679693698883057:  76%|███████▌  | 4917/6470 [1:01:05<18:13,  1.42it/s]
4|4918|Loss: 0.7679693698883057:  76%|███████▌  | 4918/6470 [1:01:05<19:20,  1.34it/s]
4|4919|Loss: 0.8001305460929871:  76%|███████▌  | 4918/6470 [1:01:05<19:20,  1.34it/s]
4|4919|Loss: 0.8001305460929871:  76%|███████▌  | 4919/6470 [1:01:06<19:15,  1.34it/s]
4|4920|Loss: 0.8735387325286865:  76%|███████▌  | 4919/6470 [1:01:06<19:15,  1.34it/s]
4|4920|Loss: 0.8735387325286865:  76%|███████▌  | 4920/6470 [1:01:06<17:56,  1.44it/s]
4|4921|Loss: 0.8573892116546631:  76%|███████▌  | 4920/6470 [1:01:07<17:56,  1.44it/s]
4|4921|Loss: 0.8573892116546631:  76%|███████▌  | 4921/6470 [1:01:07<18:16,  1.41it/s]
4|4922|Loss: 0.8405920267105103:  76%|███████▌  | 4921/6470 [1:01:07<18:16,  1.41it/s]
4|4922|Loss: 0.8405920267105103:  76%|███████▌  | 4922/6470 [1:01:08<18:39,  1.38it/s]
4|4923|Loss: 0.8176920413970947:  76%|███████▌  | 4922/6470 [1:01:08<18:39,  1.38it/s]
4|4923|Loss: 0.8176920413970947:  76%|███████▌  | 4923/6470 [1:01:09<18:46,  1.37it/s]
4|4924|Loss: 0.9875949025154114:  76%|███████▌  | 4923/6470 [1:01:09<18:46,  1.37it/s]
4|4924|Loss: 0.9875949025154114:  76%|███████▌  | 4924/6470 [1:01:09<19:35,  1.32it/s]
4|4925|Loss: 0.7943099737167358:  76%|███████▌  | 4924/6470 [1:01:10<19:35,  1.32it/s]
4|4925|Loss: 0.7943099737167358:  76%|███████▌  | 4925/6470 [1:01:10<20:33,  1.25it/s]
4|4926|Loss: 0.7631497383117676:  76%|███████▌  | 4925/6470 [1:01:11<20:33,  1.25it/s]
4|4926|Loss: 0.7631497383117676:  76%|███████▌  | 4926/6470 [1:01:11<18:40,  1.38it/s]
4|4927|Loss: 0.7948918342590332:  76%|███████▌  | 4926/6470 [1:01:11<18:40,  1.38it/s]
4|4927|Loss: 0.7948918342590332:  76%|███████▌  | 4927/6470 [1:01:12<18:20,  1.40it/s]
4|4928|Loss: 0.8090634346008301:  76%|███████▌  | 4927/6470 [1:01:12<18:20,  1.40it/s]
4|4928|Loss: 0.8090634346008301:  76%|███████▌  | 4928/6470 [1:01:12<19:40,  1.31it/s]
4|4929|Loss: 0.7248542308807373:  76%|███████▌  | 4928/6470 [1:01:13<19:40,  1.31it/s]
4|4929|Loss: 0.7248542308807373:  76%|███████▌  | 4929/6470 [1:01:13<18:56,  1.36it/s]
4|4930|Loss: 0.8128416538238525:  76%|███████▌  | 4929/6470 [1:01:14<18:56,  1.36it/s]
4|4930|Loss: 0.8128416538238525:  76%|███████▌  | 4930/6470 [1:01:14<19:50,  1.29it/s]
4|4931|Loss: 0.7177962064743042:  76%|███████▌  | 4930/6470 [1:01:14<19:50,  1.29it/s]
4|4931|Loss: 0.7177962064743042:  76%|███████▌  | 4931/6470 [1:01:15<19:34,  1.31it/s]
4|4932|Loss: 0.7166610956192017:  76%|███████▌  | 4931/6470 [1:01:15<19:34,  1.31it/s]
4|4932|Loss: 0.7166610956192017:  76%|███████▌  | 4932/6470 [1:01:15<18:37,  1.38it/s]
4|4933|Loss: 0.8759652376174927:  76%|███████▌  | 4932/6470 [1:01:16<18:37,  1.38it/s]
4|4933|Loss: 0.8759652376174927:  76%|███████▌  | 4933/6470 [1:01:16<19:23,  1.32it/s]
4|4934|Loss: 0.8623287677764893:  76%|███████▌  | 4933/6470 [1:01:17<19:23,  1.32it/s]
4|4934|Loss: 0.8623287677764893:  76%|███████▋  | 4934/6470 [1:01:17<19:42,  1.30it/s]
4|4935|Loss: 0.8668488264083862:  76%|███████▋  | 4934/6470 [1:01:17<19:42,  1.30it/s]
4|4935|Loss: 0.8668488264083862:  76%|███████▋  | 4935/6470 [1:01:18<20:18,  1.26it/s]
4|4936|Loss: 0.7353910803794861:  76%|███████▋  | 4935/6470 [1:01:18<20:18,  1.26it/s]
4|4936|Loss: 0.7353910803794861:  76%|███████▋  | 4936/6470 [1:01:19<20:24,  1.25it/s]
4|4937|Loss: 0.8320454359054565:  76%|███████▋  | 4936/6470 [1:01:19<20:24,  1.25it/s]
4|4937|Loss: 0.8320454359054565:  76%|███████▋  | 4937/6470 [1:01:19<20:18,  1.26it/s]
4|4938|Loss: 0.8595779538154602:  76%|███████▋  | 4937/6470 [1:01:20<20:18,  1.26it/s]
4|4938|Loss: 0.8595779538154602:  76%|███████▋  | 4938/6470 [1:01:20<20:16,  1.26it/s]
4|4939|Loss: 0.8800567984580994:  76%|███████▋  | 4938/6470 [1:01:21<20:16,  1.26it/s]
4|4939|Loss: 0.8800567984580994:  76%|███████▋  | 4939/6470 [1:01:21<19:14,  1.33it/s]
4|4940|Loss: 0.7617155313491821:  76%|███████▋  | 4939/6470 [1:01:21<19:14,  1.33it/s]
4|4940|Loss: 0.7617155313491821:  76%|███████▋  | 4940/6470 [1:01:22<18:38,  1.37it/s]
4|4941|Loss: 0.7161277532577515:  76%|███████▋  | 4940/6470 [1:01:22<18:38,  1.37it/s]
4|4941|Loss: 0.7161277532577515:  76%|███████▋  | 4941/6470 [1:01:22<19:02,  1.34it/s]
4|4942|Loss: 0.8833513855934143:  76%|███████▋  | 4941/6470 [1:01:23<19:02,  1.34it/s]
4|4942|Loss: 0.8833513855934143:  76%|███████▋  | 4942/6470 [1:01:23<20:18,  1.25it/s]
4|4943|Loss: 0.6812323927879333:  76%|███████▋  | 4942/6470 [1:01:24<20:18,  1.25it/s]
4|4943|Loss: 0.6812323927879333:  76%|███████▋  | 4943/6470 [1:01:24<21:11,  1.20it/s]
4|4944|Loss: 0.762084424495697:  76%|███████▋  | 4943/6470 [1:01:25<21:11,  1.20it/s] 
4|4944|Loss: 0.762084424495697:  76%|███████▋  | 4944/6470 [1:01:25<20:28,  1.24it/s]
4|4945|Loss: 0.9247040152549744:  76%|███████▋  | 4944/6470 [1:01:25<20:28,  1.24it/s]
4|4945|Loss: 0.9247040152549744:  76%|███████▋  | 4945/6470 [1:01:26<20:17,  1.25it/s]
4|4946|Loss: 0.7250803709030151:  76%|███████▋  | 4945/6470 [1:01:26<20:17,  1.25it/s]
4|4946|Loss: 0.7250803709030151:  76%|███████▋  | 4946/6470 [1:01:26<17:44,  1.43it/s]
4|4947|Loss: 0.8434232473373413:  76%|███████▋  | 4946/6470 [1:01:26<17:44,  1.43it/s]
4|4947|Loss: 0.8434232473373413:  76%|███████▋  | 4947/6470 [1:01:27<17:04,  1.49it/s]
4|4948|Loss: 0.7742945551872253:  76%|███████▋  | 4947/6470 [1:01:27<17:04,  1.49it/s]
4|4948|Loss: 0.7742945551872253:  76%|███████▋  | 4948/6470 [1:01:28<18:26,  1.38it/s]
4|4949|Loss: 0.6177831888198853:  76%|███████▋  | 4948/6470 [1:01:28<18:26,  1.38it/s]
4|4949|Loss: 0.6177831888198853:  76%|███████▋  | 4949/6470 [1:01:28<17:48,  1.42it/s]
4|4950|Loss: 0.9147418737411499:  76%|███████▋  | 4949/6470 [1:01:29<17:48,  1.42it/s]
4|4950|Loss: 0.9147418737411499:  77%|███████▋  | 4950/6470 [1:01:29<17:56,  1.41it/s]
4|4951|Loss: 0.8939331769943237:  77%|███████▋  | 4950/6470 [1:01:29<17:56,  1.41it/s]
4|4951|Loss: 0.8939331769943237:  77%|███████▋  | 4951/6470 [1:01:30<18:20,  1.38it/s]
4|4952|Loss: 0.8381245136260986:  77%|███████▋  | 4951/6470 [1:01:30<18:20,  1.38it/s]
4|4952|Loss: 0.8381245136260986:  77%|███████▋  | 4952/6470 [1:01:30<17:47,  1.42it/s]
4|4953|Loss: 0.8668361902236938:  77%|███████▋  | 4952/6470 [1:01:31<17:47,  1.42it/s]
4|4953|Loss: 0.8668361902236938:  77%|███████▋  | 4953/6470 [1:01:31<16:44,  1.51it/s]
4|4954|Loss: 0.8270675539970398:  77%|███████▋  | 4953/6470 [1:01:31<16:44,  1.51it/s]
4|4954|Loss: 0.8270675539970398:  77%|███████▋  | 4954/6470 [1:01:32<18:35,  1.36it/s]
4|4955|Loss: 0.7812822461128235:  77%|███████▋  | 4954/6470 [1:01:32<18:35,  1.36it/s]
4|4955|Loss: 0.7812822461128235:  77%|███████▋  | 4955/6470 [1:01:33<18:45,  1.35it/s]
4|4956|Loss: 0.8677403926849365:  77%|███████▋  | 4955/6470 [1:01:33<18:45,  1.35it/s]
4|4956|Loss: 0.8677403926849365:  77%|███████▋  | 4956/6470 [1:01:33<19:00,  1.33it/s]
4|4957|Loss: 0.8167223930358887:  77%|███████▋  | 4956/6470 [1:01:34<19:00,  1.33it/s]
4|4957|Loss: 0.8167223930358887:  77%|███████▋  | 4957/6470 [1:01:34<18:42,  1.35it/s]
4|4958|Loss: 0.7067384123802185:  77%|███████▋  | 4957/6470 [1:01:34<18:42,  1.35it/s]
4|4958|Loss: 0.7067384123802185:  77%|███████▋  | 4958/6470 [1:01:35<18:07,  1.39it/s]
4|4959|Loss: 0.7384754419326782:  77%|███████▋  | 4958/6470 [1:01:35<18:07,  1.39it/s]
4|4959|Loss: 0.7384754419326782:  77%|███████▋  | 4959/6470 [1:01:35<17:40,  1.42it/s]
4|4960|Loss: 0.852935254573822:  77%|███████▋  | 4959/6470 [1:01:36<17:40,  1.42it/s] 
4|4960|Loss: 0.852935254573822:  77%|███████▋  | 4960/6470 [1:01:36<18:17,  1.38it/s]
4|4961|Loss: 0.6528728008270264:  77%|███████▋  | 4960/6470 [1:01:37<18:17,  1.38it/s]
4|4961|Loss: 0.6528728008270264:  77%|███████▋  | 4961/6470 [1:01:37<16:32,  1.52it/s]
4|4962|Loss: 0.972966194152832:  77%|███████▋  | 4961/6470 [1:01:37<16:32,  1.52it/s] 
4|4962|Loss: 0.972966194152832:  77%|███████▋  | 4962/6470 [1:01:37<16:26,  1.53it/s]
4|4963|Loss: 1.09235417842865:  77%|███████▋  | 4962/6470 [1:01:38<16:26,  1.53it/s] 
4|4963|Loss: 1.09235417842865:  77%|███████▋  | 4963/6470 [1:01:38<17:53,  1.40it/s]
4|4964|Loss: 0.9868417978286743:  77%|███████▋  | 4963/6470 [1:01:39<17:53,  1.40it/s]
4|4964|Loss: 0.9868417978286743:  77%|███████▋  | 4964/6470 [1:01:39<18:02,  1.39it/s]
4|4965|Loss: 0.8714062571525574:  77%|███████▋  | 4964/6470 [1:01:39<18:02,  1.39it/s]
4|4965|Loss: 0.8714062571525574:  77%|███████▋  | 4965/6470 [1:01:40<18:26,  1.36it/s]
4|4966|Loss: 1.0024999380111694:  77%|███████▋  | 4965/6470 [1:01:40<18:26,  1.36it/s]
4|4966|Loss: 1.0024999380111694:  77%|███████▋  | 4966/6470 [1:01:41<19:30,  1.28it/s]
4|4967|Loss: 0.9261706471443176:  77%|███████▋  | 4966/6470 [1:01:41<19:30,  1.28it/s]
4|4967|Loss: 0.9261706471443176:  77%|███████▋  | 4967/6470 [1:01:41<19:01,  1.32it/s]
4|4968|Loss: 0.9467185139656067:  77%|███████▋  | 4967/6470 [1:01:42<19:01,  1.32it/s]
4|4968|Loss: 0.9467185139656067:  77%|███████▋  | 4968/6470 [1:01:42<19:04,  1.31it/s]
4|4969|Loss: 0.8494417667388916:  77%|███████▋  | 4968/6470 [1:01:42<19:04,  1.31it/s]
4|4969|Loss: 0.8494417667388916:  77%|███████▋  | 4969/6470 [1:01:43<18:20,  1.36it/s]
4|4970|Loss: 0.7639824151992798:  77%|███████▋  | 4969/6470 [1:01:43<18:20,  1.36it/s]
4|4970|Loss: 0.7639824151992798:  77%|███████▋  | 4970/6470 [1:01:43<17:45,  1.41it/s]
4|4971|Loss: 0.9043092727661133:  77%|███████▋  | 4970/6470 [1:01:44<17:45,  1.41it/s]
4|4971|Loss: 0.9043092727661133:  77%|███████▋  | 4971/6470 [1:01:44<17:08,  1.46it/s]
4|4972|Loss: 0.8078922629356384:  77%|███████▋  | 4971/6470 [1:01:44<17:08,  1.46it/s]
4|4972|Loss: 0.8078922629356384:  77%|███████▋  | 4972/6470 [1:01:45<17:05,  1.46it/s]
4|4973|Loss: 0.8772828578948975:  77%|███████▋  | 4972/6470 [1:01:45<17:05,  1.46it/s]
4|4973|Loss: 0.8772828578948975:  77%|███████▋  | 4973/6470 [1:01:45<17:49,  1.40it/s]
4|4974|Loss: 0.7044409513473511:  77%|███████▋  | 4973/6470 [1:01:46<17:49,  1.40it/s]
4|4974|Loss: 0.7044409513473511:  77%|███████▋  | 4974/6470 [1:01:46<16:31,  1.51it/s]
4|4975|Loss: 0.9413868188858032:  77%|███████▋  | 4974/6470 [1:01:46<16:31,  1.51it/s]
4|4975|Loss: 0.9413868188858032:  77%|███████▋  | 4975/6470 [1:01:47<17:29,  1.42it/s]
4|4976|Loss: 0.8849844336509705:  77%|███████▋  | 4975/6470 [1:01:47<17:29,  1.42it/s]
4|4976|Loss: 0.8849844336509705:  77%|███████▋  | 4976/6470 [1:01:48<18:04,  1.38it/s]
4|4977|Loss: 0.7337430715560913:  77%|███████▋  | 4976/6470 [1:01:48<18:04,  1.38it/s]
4|4977|Loss: 0.7337430715560913:  77%|███████▋  | 4977/6470 [1:01:48<16:53,  1.47it/s]
4|4978|Loss: 0.6925476789474487:  77%|███████▋  | 4977/6470 [1:01:48<16:53,  1.47it/s]
4|4978|Loss: 0.6925476789474487:  77%|███████▋  | 4978/6470 [1:01:49<15:29,  1.61it/s]
4|4979|Loss: 0.8729109764099121:  77%|███████▋  | 4978/6470 [1:01:49<15:29,  1.61it/s]
4|4979|Loss: 0.8729109764099121:  77%|███████▋  | 4979/6470 [1:01:49<16:56,  1.47it/s]
4|4980|Loss: 1.0556020736694336:  77%|███████▋  | 4979/6470 [1:01:50<16:56,  1.47it/s]
4|4980|Loss: 1.0556020736694336:  77%|███████▋  | 4980/6470 [1:01:50<18:28,  1.34it/s]
4|4981|Loss: 0.9313755035400391:  77%|███████▋  | 4980/6470 [1:01:51<18:28,  1.34it/s]
4|4981|Loss: 0.9313755035400391:  77%|███████▋  | 4981/6470 [1:01:51<18:46,  1.32it/s]
4|4982|Loss: 0.8739122152328491:  77%|███████▋  | 4981/6470 [1:01:52<18:46,  1.32it/s]
4|4982|Loss: 0.8739122152328491:  77%|███████▋  | 4982/6470 [1:01:52<18:12,  1.36it/s]
4|4983|Loss: 0.778872013092041:  77%|███████▋  | 4982/6470 [1:01:52<18:12,  1.36it/s] 
4|4983|Loss: 0.778872013092041:  77%|███████▋  | 4983/6470 [1:01:53<18:22,  1.35it/s]
4|4984|Loss: 0.8428284525871277:  77%|███████▋  | 4983/6470 [1:01:53<18:22,  1.35it/s]
4|4984|Loss: 0.8428284525871277:  77%|███████▋  | 4984/6470 [1:01:53<19:25,  1.27it/s]
4|4985|Loss: 0.7693700790405273:  77%|███████▋  | 4984/6470 [1:01:54<19:25,  1.27it/s]
4|4985|Loss: 0.7693700790405273:  77%|███████▋  | 4985/6470 [1:01:54<20:06,  1.23it/s]
4|4986|Loss: 0.6931900382041931:  77%|███████▋  | 4985/6470 [1:01:55<20:06,  1.23it/s]
4|4986|Loss: 0.6931900382041931:  77%|███████▋  | 4986/6470 [1:01:55<18:36,  1.33it/s]
4|4987|Loss: 0.8792011141777039:  77%|███████▋  | 4986/6470 [1:01:55<18:36,  1.33it/s]
4|4987|Loss: 0.8792011141777039:  77%|███████▋  | 4987/6470 [1:01:56<19:18,  1.28it/s]
4|4988|Loss: 0.9377721548080444:  77%|███████▋  | 4987/6470 [1:01:56<19:18,  1.28it/s]
4|4988|Loss: 0.9377721548080444:  77%|███████▋  | 4988/6470 [1:01:57<19:43,  1.25it/s]
4|4989|Loss: 0.750616729259491:  77%|███████▋  | 4988/6470 [1:01:57<19:43,  1.25it/s] 
4|4989|Loss: 0.750616729259491:  77%|███████▋  | 4989/6470 [1:01:57<18:16,  1.35it/s]
4|4990|Loss: 0.7933112382888794:  77%|███████▋  | 4989/6470 [1:01:58<18:16,  1.35it/s]
4|4990|Loss: 0.7933112382888794:  77%|███████▋  | 4990/6470 [1:01:58<17:57,  1.37it/s]
4|4991|Loss: 0.9608190059661865:  77%|███████▋  | 4990/6470 [1:01:58<17:57,  1.37it/s]
4|4991|Loss: 0.9608190059661865:  77%|███████▋  | 4991/6470 [1:01:59<17:52,  1.38it/s]
4|4992|Loss: 0.8452953696250916:  77%|███████▋  | 4991/6470 [1:01:59<17:52,  1.38it/s]
4|4992|Loss: 0.8452953696250916:  77%|███████▋  | 4992/6470 [1:01:59<18:05,  1.36it/s]
4|4993|Loss: 0.9545403718948364:  77%|███████▋  | 4992/6470 [1:02:00<18:05,  1.36it/s]
4|4993|Loss: 0.9545403718948364:  77%|███████▋  | 4993/6470 [1:02:00<19:12,  1.28it/s]
4|4994|Loss: 0.9421738982200623:  77%|███████▋  | 4993/6470 [1:02:01<19:12,  1.28it/s]
4|4994|Loss: 0.9421738982200623:  77%|███████▋  | 4994/6470 [1:02:01<19:33,  1.26it/s]
4|4995|Loss: 0.8090887665748596:  77%|███████▋  | 4994/6470 [1:02:02<19:33,  1.26it/s]
4|4995|Loss: 0.8090887665748596:  77%|███████▋  | 4995/6470 [1:02:02<19:32,  1.26it/s]
4|4996|Loss: 0.8850948214530945:  77%|███████▋  | 4995/6470 [1:02:02<19:32,  1.26it/s]
4|4996|Loss: 0.8850948214530945:  77%|███████▋  | 4996/6470 [1:02:03<20:24,  1.20it/s]
4|4997|Loss: 0.8897507190704346:  77%|███████▋  | 4996/6470 [1:02:03<20:24,  1.20it/s]
4|4997|Loss: 0.8897507190704346:  77%|███████▋  | 4997/6470 [1:02:04<21:21,  1.15it/s]
4|4998|Loss: 0.9516831040382385:  77%|███████▋  | 4997/6470 [1:02:04<21:21,  1.15it/s]
4|4998|Loss: 0.9516831040382385:  77%|███████▋  | 4998/6470 [1:02:05<22:24,  1.09it/s]
4|4999|Loss: 0.8257145881652832:  77%|███████▋  | 4998/6470 [1:02:05<22:24,  1.09it/s]
4|4999|Loss: 0.8257145881652832:  77%|███████▋  | 4999/6470 [1:02:06<21:07,  1.16it/s]
4|5000|Loss: 0.811231255531311:  77%|███████▋  | 4999/6470 [1:02:06<21:07,  1.16it/s] 
4|5000|Loss: 0.811231255531311:  77%|███████▋  | 5000/6470 [1:02:06<21:18,  1.15it/s]
4|5001|Loss: 0.8871459364891052:  77%|███████▋  | 5000/6470 [1:02:07<21:18,  1.15it/s]
4|5001|Loss: 0.8871459364891052:  77%|███████▋  | 5001/6470 [1:02:07<20:48,  1.18it/s]
4|5002|Loss: 0.7078689932823181:  77%|███████▋  | 5001/6470 [1:02:08<20:48,  1.18it/s]
4|5002|Loss: 0.7078689932823181:  77%|███████▋  | 5002/6470 [1:02:08<21:55,  1.12it/s]
4|5003|Loss: 0.7983096241950989:  77%|███████▋  | 5002/6470 [1:02:09<21:55,  1.12it/s]
4|5003|Loss: 0.7983096241950989:  77%|███████▋  | 5003/6470 [1:02:09<21:57,  1.11it/s]
4|5004|Loss: 0.7133920788764954:  77%|███████▋  | 5003/6470 [1:02:10<21:57,  1.11it/s]
4|5004|Loss: 0.7133920788764954:  77%|███████▋  | 5004/6470 [1:02:10<21:06,  1.16it/s]
4|5005|Loss: 1.0038176774978638:  77%|███████▋  | 5004/6470 [1:02:10<21:06,  1.16it/s]
4|5005|Loss: 1.0038176774978638:  77%|███████▋  | 5005/6470 [1:02:11<19:32,  1.25it/s]
4|5006|Loss: 0.7588886618614197:  77%|███████▋  | 5005/6470 [1:02:11<19:32,  1.25it/s]
4|5006|Loss: 0.7588886618614197:  77%|███████▋  | 5006/6470 [1:02:11<19:51,  1.23it/s]
4|5007|Loss: 0.6322508454322815:  77%|███████▋  | 5006/6470 [1:02:12<19:51,  1.23it/s]
4|5007|Loss: 0.6322508454322815:  77%|███████▋  | 5007/6470 [1:02:12<19:34,  1.25it/s]
4|5008|Loss: 0.8648794889450073:  77%|███████▋  | 5007/6470 [1:02:13<19:34,  1.25it/s]
4|5008|Loss: 0.8648794889450073:  77%|███████▋  | 5008/6470 [1:02:13<18:00,  1.35it/s]
4|5009|Loss: 0.9569453001022339:  77%|███████▋  | 5008/6470 [1:02:13<18:00,  1.35it/s]
4|5009|Loss: 0.9569453001022339:  77%|███████▋  | 5009/6470 [1:02:14<18:39,  1.30it/s]
4|5010|Loss: 0.7713256478309631:  77%|███████▋  | 5009/6470 [1:02:14<18:39,  1.30it/s]
4|5010|Loss: 0.7713256478309631:  77%|███████▋  | 5010/6470 [1:02:14<18:51,  1.29it/s]
4|5011|Loss: 0.846808910369873:  77%|███████▋  | 5010/6470 [1:02:15<18:51,  1.29it/s] 
4|5011|Loss: 0.846808910369873:  77%|███████▋  | 5011/6470 [1:02:15<19:42,  1.23it/s]
4|5012|Loss: 0.6893997192382812:  77%|███████▋  | 5011/6470 [1:02:16<19:42,  1.23it/s]
4|5012|Loss: 0.6893997192382812:  77%|███████▋  | 5012/6470 [1:02:16<17:27,  1.39it/s]
4|5013|Loss: 0.8081799149513245:  77%|███████▋  | 5012/6470 [1:02:16<17:27,  1.39it/s]
4|5013|Loss: 0.8081799149513245:  77%|███████▋  | 5013/6470 [1:02:17<17:11,  1.41it/s]
4|5014|Loss: 0.9289942979812622:  77%|███████▋  | 5013/6470 [1:02:17<17:11,  1.41it/s]
4|5014|Loss: 0.9289942979812622:  77%|███████▋  | 5014/6470 [1:02:17<16:48,  1.44it/s]
4|5015|Loss: 0.8556516170501709:  77%|███████▋  | 5014/6470 [1:02:18<16:48,  1.44it/s]
4|5015|Loss: 0.8556516170501709:  78%|███████▊  | 5015/6470 [1:02:18<16:29,  1.47it/s]
4|5016|Loss: 0.7952130436897278:  78%|███████▊  | 5015/6470 [1:02:18<16:29,  1.47it/s]
4|5016|Loss: 0.7952130436897278:  78%|███████▊  | 5016/6470 [1:02:19<16:29,  1.47it/s]
4|5017|Loss: 0.829383909702301:  78%|███████▊  | 5016/6470 [1:02:19<16:29,  1.47it/s] 
4|5017|Loss: 0.829383909702301:  78%|███████▊  | 5017/6470 [1:02:19<17:46,  1.36it/s]
4|5018|Loss: 0.6712279319763184:  78%|███████▊  | 5017/6470 [1:02:20<17:46,  1.36it/s]
4|5018|Loss: 0.6712279319763184:  78%|███████▊  | 5018/6470 [1:02:20<18:24,  1.31it/s]
4|5019|Loss: 0.7963137030601501:  78%|███████▊  | 5018/6470 [1:02:21<18:24,  1.31it/s]
4|5019|Loss: 0.7963137030601501:  78%|███████▊  | 5019/6470 [1:02:21<19:29,  1.24it/s]
4|5020|Loss: 0.8291840553283691:  78%|███████▊  | 5019/6470 [1:02:22<19:29,  1.24it/s]
4|5020|Loss: 0.8291840553283691:  78%|███████▊  | 5020/6470 [1:02:22<18:51,  1.28it/s]
4|5021|Loss: 0.7681640386581421:  78%|███████▊  | 5020/6470 [1:02:22<18:51,  1.28it/s]
4|5021|Loss: 0.7681640386581421:  78%|███████▊  | 5021/6470 [1:02:23<19:59,  1.21it/s]
4|5022|Loss: 0.755613386631012:  78%|███████▊  | 5021/6470 [1:02:23<19:59,  1.21it/s] 
4|5022|Loss: 0.755613386631012:  78%|███████▊  | 5022/6470 [1:02:24<21:19,  1.13it/s]
4|5023|Loss: 1.0496454238891602:  78%|███████▊  | 5022/6470 [1:02:24<21:19,  1.13it/s]
4|5023|Loss: 1.0496454238891602:  78%|███████▊  | 5023/6470 [1:02:25<20:21,  1.18it/s]
4|5024|Loss: 0.6500208973884583:  78%|███████▊  | 5023/6470 [1:02:25<20:21,  1.18it/s]
4|5024|Loss: 0.6500208973884583:  78%|███████▊  | 5024/6470 [1:02:25<17:29,  1.38it/s]
4|5025|Loss: 0.9003371000289917:  78%|███████▊  | 5024/6470 [1:02:25<17:29,  1.38it/s]
4|5025|Loss: 0.9003371000289917:  78%|███████▊  | 5025/6470 [1:02:26<16:48,  1.43it/s]
4|5026|Loss: 0.7223977446556091:  78%|███████▊  | 5025/6470 [1:02:26<16:48,  1.43it/s]
4|5026|Loss: 0.7223977446556091:  78%|███████▊  | 5026/6470 [1:02:26<17:22,  1.39it/s]
4|5027|Loss: 0.9967379570007324:  78%|███████▊  | 5026/6470 [1:02:27<17:22,  1.39it/s]
4|5027|Loss: 0.9967379570007324:  78%|███████▊  | 5027/6470 [1:02:27<17:51,  1.35it/s]
4|5028|Loss: 0.9296732544898987:  78%|███████▊  | 5027/6470 [1:02:28<17:51,  1.35it/s]
4|5028|Loss: 0.9296732544898987:  78%|███████▊  | 5028/6470 [1:02:28<17:31,  1.37it/s]
4|5029|Loss: 0.7921879291534424:  78%|███████▊  | 5028/6470 [1:02:28<17:31,  1.37it/s]
4|5029|Loss: 0.7921879291534424:  78%|███████▊  | 5029/6470 [1:02:29<17:05,  1.41it/s]
4|5030|Loss: 0.7367734313011169:  78%|███████▊  | 5029/6470 [1:02:29<17:05,  1.41it/s]
4|5030|Loss: 0.7367734313011169:  78%|███████▊  | 5030/6470 [1:02:29<16:51,  1.42it/s]
4|5031|Loss: 0.8009992241859436:  78%|███████▊  | 5030/6470 [1:02:30<16:51,  1.42it/s]
4|5031|Loss: 0.8009992241859436:  78%|███████▊  | 5031/6470 [1:02:30<17:07,  1.40it/s]
4|5032|Loss: 0.8360836505889893:  78%|███████▊  | 5031/6470 [1:02:30<17:07,  1.40it/s]
4|5032|Loss: 0.8360836505889893:  78%|███████▊  | 5032/6470 [1:02:31<17:09,  1.40it/s]
4|5033|Loss: 0.8153617978096008:  78%|███████▊  | 5032/6470 [1:02:31<17:09,  1.40it/s]
4|5033|Loss: 0.8153617978096008:  78%|███████▊  | 5033/6470 [1:02:31<17:03,  1.40it/s]
4|5034|Loss: 0.8513880372047424:  78%|███████▊  | 5033/6470 [1:02:32<17:03,  1.40it/s]
4|5034|Loss: 0.8513880372047424:  78%|███████▊  | 5034/6470 [1:02:32<16:58,  1.41it/s]
4|5035|Loss: 0.7654619812965393:  78%|███████▊  | 5034/6470 [1:02:33<16:58,  1.41it/s]
4|5035|Loss: 0.7654619812965393:  78%|███████▊  | 5035/6470 [1:02:33<18:57,  1.26it/s]
4|5036|Loss: 0.7938115000724792:  78%|███████▊  | 5035/6470 [1:02:34<18:57,  1.26it/s]
4|5036|Loss: 0.7938115000724792:  78%|███████▊  | 5036/6470 [1:02:34<19:17,  1.24it/s]
4|5037|Loss: 0.7768133878707886:  78%|███████▊  | 5036/6470 [1:02:34<19:17,  1.24it/s]
4|5037|Loss: 0.7768133878707886:  78%|███████▊  | 5037/6470 [1:02:35<18:57,  1.26it/s]
4|5038|Loss: 0.730975329875946:  78%|███████▊  | 5037/6470 [1:02:35<18:57,  1.26it/s] 
4|5038|Loss: 0.730975329875946:  78%|███████▊  | 5038/6470 [1:02:35<17:45,  1.34it/s]
4|5039|Loss: 0.8838858008384705:  78%|███████▊  | 5038/6470 [1:02:36<17:45,  1.34it/s]
4|5039|Loss: 0.8838858008384705:  78%|███████▊  | 5039/6470 [1:02:36<17:56,  1.33it/s]
4|5040|Loss: 0.8040013909339905:  78%|███████▊  | 5039/6470 [1:02:37<17:56,  1.33it/s]
4|5040|Loss: 0.8040013909339905:  78%|███████▊  | 5040/6470 [1:02:37<17:30,  1.36it/s]
4|5041|Loss: 0.8620506525039673:  78%|███████▊  | 5040/6470 [1:02:37<17:30,  1.36it/s]
4|5041|Loss: 0.8620506525039673:  78%|███████▊  | 5041/6470 [1:02:38<17:49,  1.34it/s]
4|5042|Loss: 0.8258062601089478:  78%|███████▊  | 5041/6470 [1:02:38<17:49,  1.34it/s]
4|5042|Loss: 0.8258062601089478:  78%|███████▊  | 5042/6470 [1:02:38<16:40,  1.43it/s]
4|5043|Loss: 0.8922017812728882:  78%|███████▊  | 5042/6470 [1:02:39<16:40,  1.43it/s]
4|5043|Loss: 0.8922017812728882:  78%|███████▊  | 5043/6470 [1:02:39<17:38,  1.35it/s]
4|5044|Loss: 0.6488954424858093:  78%|███████▊  | 5043/6470 [1:02:39<17:38,  1.35it/s]
4|5044|Loss: 0.6488954424858093:  78%|███████▊  | 5044/6470 [1:02:40<18:02,  1.32it/s]
4|5045|Loss: 0.7820026278495789:  78%|███████▊  | 5044/6470 [1:02:40<18:02,  1.32it/s]
4|5045|Loss: 0.7820026278495789:  78%|███████▊  | 5045/6470 [1:02:41<18:05,  1.31it/s]
4|5046|Loss: 0.8349708318710327:  78%|███████▊  | 5045/6470 [1:02:41<18:05,  1.31it/s]
4|5046|Loss: 0.8349708318710327:  78%|███████▊  | 5046/6470 [1:02:41<18:38,  1.27it/s]
4|5047|Loss: 0.8892081379890442:  78%|███████▊  | 5046/6470 [1:02:42<18:38,  1.27it/s]
4|5047|Loss: 0.8892081379890442:  78%|███████▊  | 5047/6470 [1:02:42<19:09,  1.24it/s]
4|5048|Loss: 0.7919207215309143:  78%|███████▊  | 5047/6470 [1:02:43<19:09,  1.24it/s]
4|5048|Loss: 0.7919207215309143:  78%|███████▊  | 5048/6470 [1:02:43<19:09,  1.24it/s]
4|5049|Loss: 0.8405324220657349:  78%|███████▊  | 5048/6470 [1:02:44<19:09,  1.24it/s]
4|5049|Loss: 0.8405324220657349:  78%|███████▊  | 5049/6470 [1:02:44<19:07,  1.24it/s]
4|5050|Loss: 0.7631649374961853:  78%|███████▊  | 5049/6470 [1:02:44<19:07,  1.24it/s]
4|5050|Loss: 0.7631649374961853:  78%|███████▊  | 5050/6470 [1:02:45<18:53,  1.25it/s]
4|5051|Loss: 0.7733269333839417:  78%|███████▊  | 5050/6470 [1:02:45<18:53,  1.25it/s]
4|5051|Loss: 0.7733269333839417:  78%|███████▊  | 5051/6470 [1:02:46<19:26,  1.22it/s]
4|5052|Loss: 0.7808348536491394:  78%|███████▊  | 5051/6470 [1:02:46<19:26,  1.22it/s]
4|5052|Loss: 0.7808348536491394:  78%|███████▊  | 5052/6470 [1:02:46<20:00,  1.18it/s]
4|5053|Loss: 0.6026824116706848:  78%|███████▊  | 5052/6470 [1:02:47<20:00,  1.18it/s]
4|5053|Loss: 0.6026824116706848:  78%|███████▊  | 5053/6470 [1:02:47<19:13,  1.23it/s]
4|5054|Loss: 0.8278989195823669:  78%|███████▊  | 5053/6470 [1:02:48<19:13,  1.23it/s]
4|5054|Loss: 0.8278989195823669:  78%|███████▊  | 5054/6470 [1:02:48<18:21,  1.29it/s]
4|5055|Loss: 0.8322382569313049:  78%|███████▊  | 5054/6470 [1:02:48<18:21,  1.29it/s]
4|5055|Loss: 0.8322382569313049:  78%|███████▊  | 5055/6470 [1:02:49<18:47,  1.26it/s]
4|5056|Loss: 0.8466154336929321:  78%|███████▊  | 5055/6470 [1:02:49<18:47,  1.26it/s]
4|5056|Loss: 0.8466154336929321:  78%|███████▊  | 5056/6470 [1:02:49<18:40,  1.26it/s]
4|5057|Loss: 0.6711534857749939:  78%|███████▊  | 5056/6470 [1:02:50<18:40,  1.26it/s]
4|5057|Loss: 0.6711534857749939:  78%|███████▊  | 5057/6470 [1:02:50<19:35,  1.20it/s]
4|5058|Loss: 0.9144590497016907:  78%|███████▊  | 5057/6470 [1:02:51<19:35,  1.20it/s]
4|5058|Loss: 0.9144590497016907:  78%|███████▊  | 5058/6470 [1:02:51<18:18,  1.29it/s]
4|5059|Loss: 0.829744279384613:  78%|███████▊  | 5058/6470 [1:02:52<18:18,  1.29it/s] 
4|5059|Loss: 0.829744279384613:  78%|███████▊  | 5059/6470 [1:02:52<18:58,  1.24it/s]
4|5060|Loss: 0.9678661227226257:  78%|███████▊  | 5059/6470 [1:02:52<18:58,  1.24it/s]
4|5060|Loss: 0.9678661227226257:  78%|███████▊  | 5060/6470 [1:02:53<18:39,  1.26it/s]
4|5061|Loss: 0.9038387537002563:  78%|███████▊  | 5060/6470 [1:02:53<18:39,  1.26it/s]
4|5061|Loss: 0.9038387537002563:  78%|███████▊  | 5061/6470 [1:02:53<18:34,  1.26it/s]
4|5062|Loss: 0.7660847902297974:  78%|███████▊  | 5061/6470 [1:02:54<18:34,  1.26it/s]
4|5062|Loss: 0.7660847902297974:  78%|███████▊  | 5062/6470 [1:02:54<17:24,  1.35it/s]
4|5063|Loss: 0.7194488644599915:  78%|███████▊  | 5062/6470 [1:02:54<17:24,  1.35it/s]
4|5063|Loss: 0.7194488644599915:  78%|███████▊  | 5063/6470 [1:02:55<16:40,  1.41it/s]
4|5064|Loss: 0.8150129318237305:  78%|███████▊  | 5063/6470 [1:02:55<16:40,  1.41it/s]
4|5064|Loss: 0.8150129318237305:  78%|███████▊  | 5064/6470 [1:02:55<16:21,  1.43it/s]
4|5065|Loss: 0.7938658595085144:  78%|███████▊  | 5064/6470 [1:02:56<16:21,  1.43it/s]
4|5065|Loss: 0.7938658595085144:  78%|███████▊  | 5065/6470 [1:02:56<15:55,  1.47it/s]
4|5066|Loss: 0.7650062441825867:  78%|███████▊  | 5065/6470 [1:02:56<15:55,  1.47it/s]
4|5066|Loss: 0.7650062441825867:  78%|███████▊  | 5066/6470 [1:02:57<16:56,  1.38it/s]
4|5067|Loss: 0.9460198879241943:  78%|███████▊  | 5066/6470 [1:02:57<16:56,  1.38it/s]
4|5067|Loss: 0.9460198879241943:  78%|███████▊  | 5067/6470 [1:02:58<17:20,  1.35it/s]
4|5068|Loss: 0.9939557909965515:  78%|███████▊  | 5067/6470 [1:02:58<17:20,  1.35it/s]
4|5068|Loss: 0.9939557909965515:  78%|███████▊  | 5068/6470 [1:02:58<17:01,  1.37it/s]
4|5069|Loss: 0.8243504166603088:  78%|███████▊  | 5068/6470 [1:02:59<17:01,  1.37it/s]
4|5069|Loss: 0.8243504166603088:  78%|███████▊  | 5069/6470 [1:02:59<17:16,  1.35it/s]
4|5070|Loss: 0.7301709651947021:  78%|███████▊  | 5069/6470 [1:03:00<17:16,  1.35it/s]
4|5070|Loss: 0.7301709651947021:  78%|███████▊  | 5070/6470 [1:03:00<17:31,  1.33it/s]
4|5071|Loss: 0.8874366283416748:  78%|███████▊  | 5070/6470 [1:03:00<17:31,  1.33it/s]
4|5071|Loss: 0.8874366283416748:  78%|███████▊  | 5071/6470 [1:03:01<16:52,  1.38it/s]
4|5072|Loss: 0.7902145385742188:  78%|███████▊  | 5071/6470 [1:03:01<16:52,  1.38it/s]
4|5072|Loss: 0.7902145385742188:  78%|███████▊  | 5072/6470 [1:03:02<18:21,  1.27it/s]
4|5073|Loss: 0.8122066259384155:  78%|███████▊  | 5072/6470 [1:03:02<18:21,  1.27it/s]
4|5073|Loss: 0.8122066259384155:  78%|███████▊  | 5073/6470 [1:03:02<19:36,  1.19it/s]
4|5074|Loss: 0.8931854367256165:  78%|███████▊  | 5073/6470 [1:03:03<19:36,  1.19it/s]
4|5074|Loss: 0.8931854367256165:  78%|███████▊  | 5074/6470 [1:03:03<19:38,  1.18it/s]
4|5075|Loss: 0.8539193868637085:  78%|███████▊  | 5074/6470 [1:03:04<19:38,  1.18it/s]
4|5075|Loss: 0.8539193868637085:  78%|███████▊  | 5075/6470 [1:03:04<18:54,  1.23it/s]
4|5076|Loss: 0.4794630706310272:  78%|███████▊  | 5075/6470 [1:03:05<18:54,  1.23it/s]
4|5076|Loss: 0.4794630706310272:  78%|███████▊  | 5076/6470 [1:03:05<20:04,  1.16it/s]
4|5077|Loss: 0.8531429767608643:  78%|███████▊  | 5076/6470 [1:03:06<20:04,  1.16it/s]
4|5077|Loss: 0.8531429767608643:  78%|███████▊  | 5077/6470 [1:03:06<19:23,  1.20it/s]
4|5078|Loss: 0.7698354721069336:  78%|███████▊  | 5077/6470 [1:03:06<19:23,  1.20it/s]
4|5078|Loss: 0.7698354721069336:  78%|███████▊  | 5078/6470 [1:03:07<18:42,  1.24it/s]
4|5079|Loss: 0.8570792078971863:  78%|███████▊  | 5078/6470 [1:03:07<18:42,  1.24it/s]
4|5079|Loss: 0.8570792078971863:  79%|███████▊  | 5079/6470 [1:03:07<17:31,  1.32it/s]
4|5080|Loss: 0.8476925492286682:  79%|███████▊  | 5079/6470 [1:03:08<17:31,  1.32it/s]
4|5080|Loss: 0.8476925492286682:  79%|███████▊  | 5080/6470 [1:03:08<15:45,  1.47it/s]
4|5081|Loss: 0.714978814125061:  79%|███████▊  | 5080/6470 [1:03:08<15:45,  1.47it/s] 
4|5081|Loss: 0.714978814125061:  79%|███████▊  | 5081/6470 [1:03:09<16:38,  1.39it/s]
4|5082|Loss: 0.7651627063751221:  79%|███████▊  | 5081/6470 [1:03:09<16:38,  1.39it/s]
4|5082|Loss: 0.7651627063751221:  79%|███████▊  | 5082/6470 [1:03:09<17:01,  1.36it/s]
4|5083|Loss: 0.8125736117362976:  79%|███████▊  | 5082/6470 [1:03:10<17:01,  1.36it/s]
4|5083|Loss: 0.8125736117362976:  79%|███████▊  | 5083/6470 [1:03:10<16:31,  1.40it/s]
4|5084|Loss: 0.8064271807670593:  79%|███████▊  | 5083/6470 [1:03:10<16:31,  1.40it/s]
4|5084|Loss: 0.8064271807670593:  79%|███████▊  | 5084/6470 [1:03:11<16:50,  1.37it/s]
4|5085|Loss: 0.7761038541793823:  79%|███████▊  | 5084/6470 [1:03:11<16:50,  1.37it/s]
4|5085|Loss: 0.7761038541793823:  79%|███████▊  | 5085/6470 [1:03:11<15:53,  1.45it/s]
4|5086|Loss: 0.8666015863418579:  79%|███████▊  | 5085/6470 [1:03:12<15:53,  1.45it/s]
4|5086|Loss: 0.8666015863418579:  79%|███████▊  | 5086/6470 [1:03:12<16:53,  1.37it/s]
4|5087|Loss: 0.9065536856651306:  79%|███████▊  | 5086/6470 [1:03:13<16:53,  1.37it/s]
4|5087|Loss: 0.9065536856651306:  79%|███████▊  | 5087/6470 [1:03:13<18:01,  1.28it/s]
4|5088|Loss: 0.901966392993927:  79%|███████▊  | 5087/6470 [1:03:13<18:01,  1.28it/s] 
4|5088|Loss: 0.901966392993927:  79%|███████▊  | 5088/6470 [1:03:14<16:55,  1.36it/s]
4|5089|Loss: 0.9103968143463135:  79%|███████▊  | 5088/6470 [1:03:14<16:55,  1.36it/s]
4|5089|Loss: 0.9103968143463135:  79%|███████▊  | 5089/6470 [1:03:14<16:24,  1.40it/s]
4|5090|Loss: 0.8295392394065857:  79%|███████▊  | 5089/6470 [1:03:15<16:24,  1.40it/s]
4|5090|Loss: 0.8295392394065857:  79%|███████▊  | 5090/6470 [1:03:15<16:37,  1.38it/s]
4|5091|Loss: 0.9452446699142456:  79%|███████▊  | 5090/6470 [1:03:15<16:37,  1.38it/s]
4|5091|Loss: 0.9452446699142456:  79%|███████▊  | 5091/6470 [1:03:16<16:11,  1.42it/s]
4|5092|Loss: 0.8499879240989685:  79%|███████▊  | 5091/6470 [1:03:16<16:11,  1.42it/s]
4|5092|Loss: 0.8499879240989685:  79%|███████▊  | 5092/6470 [1:03:16<16:16,  1.41it/s]
4|5093|Loss: 1.0959267616271973:  79%|███████▊  | 5092/6470 [1:03:17<16:16,  1.41it/s]
4|5093|Loss: 1.0959267616271973:  79%|███████▊  | 5093/6470 [1:03:17<17:29,  1.31it/s]
4|5094|Loss: 0.8450090885162354:  79%|███████▊  | 5093/6470 [1:03:18<17:29,  1.31it/s]
4|5094|Loss: 0.8450090885162354:  79%|███████▊  | 5094/6470 [1:03:18<18:34,  1.23it/s]
4|5095|Loss: 0.8928622007369995:  79%|███████▊  | 5094/6470 [1:03:19<18:34,  1.23it/s]
4|5095|Loss: 0.8928622007369995:  79%|███████▊  | 5095/6470 [1:03:19<17:22,  1.32it/s]
4|5096|Loss: 0.7803921699523926:  79%|███████▊  | 5095/6470 [1:03:19<17:22,  1.32it/s]
4|5096|Loss: 0.7803921699523926:  79%|███████▉  | 5096/6470 [1:03:20<16:40,  1.37it/s]
4|5097|Loss: 0.7749361991882324:  79%|███████▉  | 5096/6470 [1:03:20<16:40,  1.37it/s]
4|5097|Loss: 0.7749361991882324:  79%|███████▉  | 5097/6470 [1:03:20<17:43,  1.29it/s]
4|5098|Loss: 0.9061298966407776:  79%|███████▉  | 5097/6470 [1:03:21<17:43,  1.29it/s]
4|5098|Loss: 0.9061298966407776:  79%|███████▉  | 5098/6470 [1:03:21<16:52,  1.36it/s]
4|5099|Loss: 0.8480809926986694:  79%|███████▉  | 5098/6470 [1:03:21<16:52,  1.36it/s]
4|5099|Loss: 0.8480809926986694:  79%|███████▉  | 5099/6470 [1:03:22<15:56,  1.43it/s]
4|5100|Loss: 1.1727063655853271:  79%|███████▉  | 5099/6470 [1:03:22<15:56,  1.43it/s]
4|5100|Loss: 1.1727063655853271:  79%|███████▉  | 5100/6470 [1:03:22<15:01,  1.52it/s]
4|5101|Loss: 0.7711873650550842:  79%|███████▉  | 5100/6470 [1:03:23<15:01,  1.52it/s]
4|5101|Loss: 0.7711873650550842:  79%|███████▉  | 5101/6470 [1:03:23<14:51,  1.54it/s]
4|5102|Loss: 0.7809464335441589:  79%|███████▉  | 5101/6470 [1:03:23<14:51,  1.54it/s]
4|5102|Loss: 0.7809464335441589:  79%|███████▉  | 5102/6470 [1:03:24<15:06,  1.51it/s]
4|5103|Loss: 0.7899576425552368:  79%|███████▉  | 5102/6470 [1:03:24<15:06,  1.51it/s]
4|5103|Loss: 0.7899576425552368:  79%|███████▉  | 5103/6470 [1:03:24<15:38,  1.46it/s]
4|5104|Loss: 0.7373939156532288:  79%|███████▉  | 5103/6470 [1:03:25<15:38,  1.46it/s]
4|5104|Loss: 0.7373939156532288:  79%|███████▉  | 5104/6470 [1:03:25<15:35,  1.46it/s]
4|5105|Loss: 0.9432170987129211:  79%|███████▉  | 5104/6470 [1:03:25<15:35,  1.46it/s]
4|5105|Loss: 0.9432170987129211:  79%|███████▉  | 5105/6470 [1:03:26<16:15,  1.40it/s]
4|5106|Loss: 0.893316924571991:  79%|███████▉  | 5105/6470 [1:03:26<16:15,  1.40it/s] 
4|5106|Loss: 0.893316924571991:  79%|███████▉  | 5106/6470 [1:03:26<15:55,  1.43it/s]
4|5107|Loss: 0.9978103637695312:  79%|███████▉  | 5106/6470 [1:03:27<15:55,  1.43it/s]
4|5107|Loss: 0.9978103637695312:  79%|███████▉  | 5107/6470 [1:03:27<16:18,  1.39it/s]
4|5108|Loss: 0.7899820804595947:  79%|███████▉  | 5107/6470 [1:03:28<16:18,  1.39it/s]
4|5108|Loss: 0.7899820804595947:  79%|███████▉  | 5108/6470 [1:03:28<16:37,  1.36it/s]
4|5109|Loss: 0.9331494569778442:  79%|███████▉  | 5108/6470 [1:03:28<16:37,  1.36it/s]
4|5109|Loss: 0.9331494569778442:  79%|███████▉  | 5109/6470 [1:03:29<16:09,  1.40it/s]
4|5110|Loss: 0.9571889638900757:  79%|███████▉  | 5109/6470 [1:03:29<16:09,  1.40it/s]
4|5110|Loss: 0.9571889638900757:  79%|███████▉  | 5110/6470 [1:03:30<17:34,  1.29it/s]
4|5111|Loss: 0.7825422883033752:  79%|███████▉  | 5110/6470 [1:03:30<17:34,  1.29it/s]
4|5111|Loss: 0.7825422883033752:  79%|███████▉  | 5111/6470 [1:03:30<17:04,  1.33it/s]
4|5112|Loss: 0.7811779379844666:  79%|███████▉  | 5111/6470 [1:03:31<17:04,  1.33it/s]
4|5112|Loss: 0.7811779379844666:  79%|███████▉  | 5112/6470 [1:03:31<17:26,  1.30it/s]
4|5113|Loss: 0.9202659130096436:  79%|███████▉  | 5112/6470 [1:03:32<17:26,  1.30it/s]
4|5113|Loss: 0.9202659130096436:  79%|███████▉  | 5113/6470 [1:03:32<18:14,  1.24it/s]
4|5114|Loss: 0.9532159566879272:  79%|███████▉  | 5113/6470 [1:03:32<18:14,  1.24it/s]
4|5114|Loss: 0.9532159566879272:  79%|███████▉  | 5114/6470 [1:03:33<18:53,  1.20it/s]
4|5115|Loss: 0.7840338945388794:  79%|███████▉  | 5114/6470 [1:03:33<18:53,  1.20it/s]
4|5115|Loss: 0.7840338945388794:  79%|███████▉  | 5115/6470 [1:03:34<18:49,  1.20it/s]
4|5116|Loss: 0.8978309035301208:  79%|███████▉  | 5115/6470 [1:03:34<18:49,  1.20it/s]
4|5116|Loss: 0.8978309035301208:  79%|███████▉  | 5116/6470 [1:03:34<18:00,  1.25it/s]
4|5117|Loss: 0.9703390598297119:  79%|███████▉  | 5116/6470 [1:03:35<18:00,  1.25it/s]
4|5117|Loss: 0.9703390598297119:  79%|███████▉  | 5117/6470 [1:03:35<17:19,  1.30it/s]
4|5118|Loss: 0.8481031060218811:  79%|███████▉  | 5117/6470 [1:03:36<17:19,  1.30it/s]
4|5118|Loss: 0.8481031060218811:  79%|███████▉  | 5118/6470 [1:03:36<18:11,  1.24it/s]
4|5119|Loss: 0.7505622506141663:  79%|███████▉  | 5118/6470 [1:03:36<18:11,  1.24it/s]
4|5119|Loss: 0.7505622506141663:  79%|███████▉  | 5119/6470 [1:03:37<17:42,  1.27it/s]
4|5120|Loss: 0.8755756616592407:  79%|███████▉  | 5119/6470 [1:03:37<17:42,  1.27it/s]
4|5120|Loss: 0.8755756616592407:  79%|███████▉  | 5120/6470 [1:03:37<16:57,  1.33it/s]
4|5121|Loss: 0.7714338898658752:  79%|███████▉  | 5120/6470 [1:03:38<16:57,  1.33it/s]
4|5121|Loss: 0.7714338898658752:  79%|███████▉  | 5121/6470 [1:03:38<15:23,  1.46it/s]
4|5122|Loss: 0.734697163105011:  79%|███████▉  | 5121/6470 [1:03:38<15:23,  1.46it/s] 
4|5122|Loss: 0.734697163105011:  79%|███████▉  | 5122/6470 [1:03:39<14:58,  1.50it/s]
4|5123|Loss: 0.7451538443565369:  79%|███████▉  | 5122/6470 [1:03:39<14:58,  1.50it/s]
4|5123|Loss: 0.7451538443565369:  79%|███████▉  | 5123/6470 [1:03:39<15:29,  1.45it/s]
4|5124|Loss: 1.0309606790542603:  79%|███████▉  | 5123/6470 [1:03:40<15:29,  1.45it/s]
4|5124|Loss: 1.0309606790542603:  79%|███████▉  | 5124/6470 [1:03:40<15:11,  1.48it/s]
4|5125|Loss: 0.8593044281005859:  79%|███████▉  | 5124/6470 [1:03:40<15:11,  1.48it/s]
4|5125|Loss: 0.8593044281005859:  79%|███████▉  | 5125/6470 [1:03:41<15:14,  1.47it/s]
4|5126|Loss: 0.6414887309074402:  79%|███████▉  | 5125/6470 [1:03:41<15:14,  1.47it/s]
4|5126|Loss: 0.6414887309074402:  79%|███████▉  | 5126/6470 [1:03:41<15:15,  1.47it/s]
4|5127|Loss: 0.8682495355606079:  79%|███████▉  | 5126/6470 [1:03:42<15:15,  1.47it/s]
4|5127|Loss: 0.8682495355606079:  79%|███████▉  | 5127/6470 [1:03:42<15:00,  1.49it/s]
4|5128|Loss: 0.7808507680892944:  79%|███████▉  | 5127/6470 [1:03:42<15:00,  1.49it/s]
4|5128|Loss: 0.7808507680892944:  79%|███████▉  | 5128/6470 [1:03:43<14:28,  1.55it/s]
4|5129|Loss: 0.7572801113128662:  79%|███████▉  | 5128/6470 [1:03:43<14:28,  1.55it/s]
4|5129|Loss: 0.7572801113128662:  79%|███████▉  | 5129/6470 [1:03:43<13:54,  1.61it/s]
4|5130|Loss: 0.9408909678459167:  79%|███████▉  | 5129/6470 [1:03:43<13:54,  1.61it/s]
4|5130|Loss: 0.9408909678459167:  79%|███████▉  | 5130/6470 [1:03:44<13:47,  1.62it/s]
4|5131|Loss: 0.9351698756217957:  79%|███████▉  | 5130/6470 [1:03:44<13:47,  1.62it/s]
4|5131|Loss: 0.9351698756217957:  79%|███████▉  | 5131/6470 [1:03:45<15:14,  1.46it/s]
4|5132|Loss: 0.8462275862693787:  79%|███████▉  | 5131/6470 [1:03:45<15:14,  1.46it/s]
4|5132|Loss: 0.8462275862693787:  79%|███████▉  | 5132/6470 [1:03:45<15:33,  1.43it/s]
4|5133|Loss: 0.6943899989128113:  79%|███████▉  | 5132/6470 [1:03:46<15:33,  1.43it/s]
4|5133|Loss: 0.6943899989128113:  79%|███████▉  | 5133/6470 [1:03:46<15:14,  1.46it/s]
4|5134|Loss: 0.9346374273300171:  79%|███████▉  | 5133/6470 [1:03:46<15:14,  1.46it/s]
4|5134|Loss: 0.9346374273300171:  79%|███████▉  | 5134/6470 [1:03:47<15:53,  1.40it/s]
4|5135|Loss: 0.6919316053390503:  79%|███████▉  | 5134/6470 [1:03:47<15:53,  1.40it/s]
4|5135|Loss: 0.6919316053390503:  79%|███████▉  | 5135/6470 [1:03:47<15:25,  1.44it/s]
4|5136|Loss: 0.9212390780448914:  79%|███████▉  | 5135/6470 [1:03:48<15:25,  1.44it/s]
4|5136|Loss: 0.9212390780448914:  79%|███████▉  | 5136/6470 [1:03:48<15:18,  1.45it/s]
4|5137|Loss: 0.8546934127807617:  79%|███████▉  | 5136/6470 [1:03:48<15:18,  1.45it/s]
4|5137|Loss: 0.8546934127807617:  79%|███████▉  | 5137/6470 [1:03:49<15:24,  1.44it/s]
4|5138|Loss: 0.7090439200401306:  79%|███████▉  | 5137/6470 [1:03:49<15:24,  1.44it/s]
4|5138|Loss: 0.7090439200401306:  79%|███████▉  | 5138/6470 [1:03:49<14:59,  1.48it/s]
4|5139|Loss: 0.8795627951622009:  79%|███████▉  | 5138/6470 [1:03:50<14:59,  1.48it/s]
4|5139|Loss: 0.8795627951622009:  79%|███████▉  | 5139/6470 [1:03:50<15:37,  1.42it/s]
4|5140|Loss: 0.7803944945335388:  79%|███████▉  | 5139/6470 [1:03:51<15:37,  1.42it/s]
4|5140|Loss: 0.7803944945335388:  79%|███████▉  | 5140/6470 [1:03:51<15:38,  1.42it/s]
4|5141|Loss: 0.8559253215789795:  79%|███████▉  | 5140/6470 [1:03:51<15:38,  1.42it/s]
4|5141|Loss: 0.8559253215789795:  79%|███████▉  | 5141/6470 [1:03:52<15:26,  1.43it/s]
4|5142|Loss: 0.9262655973434448:  79%|███████▉  | 5141/6470 [1:03:52<15:26,  1.43it/s]
4|5142|Loss: 0.9262655973434448:  79%|███████▉  | 5142/6470 [1:03:52<15:55,  1.39it/s]
4|5143|Loss: 0.83561110496521:  79%|███████▉  | 5142/6470 [1:03:53<15:55,  1.39it/s]  
4|5143|Loss: 0.83561110496521:  79%|███████▉  | 5143/6470 [1:03:53<14:58,  1.48it/s]
4|5144|Loss: 0.8607199192047119:  79%|███████▉  | 5143/6470 [1:03:53<14:58,  1.48it/s]
4|5144|Loss: 0.8607199192047119:  80%|███████▉  | 5144/6470 [1:03:54<15:23,  1.44it/s]
4|5145|Loss: 0.8215826749801636:  80%|███████▉  | 5144/6470 [1:03:54<15:23,  1.44it/s]
4|5145|Loss: 0.8215826749801636:  80%|███████▉  | 5145/6470 [1:03:54<15:51,  1.39it/s]
4|5146|Loss: 0.7296913862228394:  80%|███████▉  | 5145/6470 [1:03:55<15:51,  1.39it/s]
4|5146|Loss: 0.7296913862228394:  80%|███████▉  | 5146/6470 [1:03:55<17:08,  1.29it/s]
4|5147|Loss: 0.7603088617324829:  80%|███████▉  | 5146/6470 [1:03:56<17:08,  1.29it/s]
4|5147|Loss: 0.7603088617324829:  80%|███████▉  | 5147/6470 [1:03:56<16:11,  1.36it/s]
4|5148|Loss: 0.6977121233940125:  80%|███████▉  | 5147/6470 [1:03:56<16:11,  1.36it/s]
4|5148|Loss: 0.6977121233940125:  80%|███████▉  | 5148/6470 [1:03:57<16:10,  1.36it/s]
4|5149|Loss: 0.8830223679542542:  80%|███████▉  | 5148/6470 [1:03:57<16:10,  1.36it/s]
4|5149|Loss: 0.8830223679542542:  80%|███████▉  | 5149/6470 [1:03:58<16:28,  1.34it/s]
4|5150|Loss: 0.7975200414657593:  80%|███████▉  | 5149/6470 [1:03:58<16:28,  1.34it/s]
4|5150|Loss: 0.7975200414657593:  80%|███████▉  | 5150/6470 [1:03:58<16:48,  1.31it/s]
4|5151|Loss: 0.7753534317016602:  80%|███████▉  | 5150/6470 [1:03:59<16:48,  1.31it/s]
4|5151|Loss: 0.7753534317016602:  80%|███████▉  | 5151/6470 [1:03:59<17:05,  1.29it/s]
4|5152|Loss: 1.1291993856430054:  80%|███████▉  | 5151/6470 [1:04:00<17:05,  1.29it/s]
4|5152|Loss: 1.1291993856430054:  80%|███████▉  | 5152/6470 [1:04:00<17:45,  1.24it/s]
4|5153|Loss: 0.8434788584709167:  80%|███████▉  | 5152/6470 [1:04:00<17:45,  1.24it/s]
4|5153|Loss: 0.8434788584709167:  80%|███████▉  | 5153/6470 [1:04:01<18:29,  1.19it/s]
4|5154|Loss: 0.8403024077415466:  80%|███████▉  | 5153/6470 [1:04:01<18:29,  1.19it/s]
4|5154|Loss: 0.8403024077415466:  80%|███████▉  | 5154/6470 [1:04:02<18:22,  1.19it/s]
4|5155|Loss: 0.8366397023200989:  80%|███████▉  | 5154/6470 [1:04:02<18:22,  1.19it/s]
4|5155|Loss: 0.8366397023200989:  80%|███████▉  | 5155/6470 [1:04:03<18:33,  1.18it/s]
4|5156|Loss: 0.8535355925559998:  80%|███████▉  | 5155/6470 [1:04:03<18:33,  1.18it/s]
4|5156|Loss: 0.8535355925559998:  80%|███████▉  | 5156/6470 [1:04:03<17:48,  1.23it/s]
4|5157|Loss: 0.9068389534950256:  80%|███████▉  | 5156/6470 [1:04:04<17:48,  1.23it/s]
4|5157|Loss: 0.9068389534950256:  80%|███████▉  | 5157/6470 [1:04:04<18:01,  1.21it/s]
4|5158|Loss: 0.8951118588447571:  80%|███████▉  | 5157/6470 [1:04:05<18:01,  1.21it/s]
4|5158|Loss: 0.8951118588447571:  80%|███████▉  | 5158/6470 [1:04:05<17:24,  1.26it/s]
4|5159|Loss: 0.8610711097717285:  80%|███████▉  | 5158/6470 [1:04:05<17:24,  1.26it/s]
4|5159|Loss: 0.8610711097717285:  80%|███████▉  | 5159/6470 [1:04:06<18:04,  1.21it/s]
4|5160|Loss: 0.7162541151046753:  80%|███████▉  | 5159/6470 [1:04:06<18:04,  1.21it/s]
4|5160|Loss: 0.7162541151046753:  80%|███████▉  | 5160/6470 [1:04:07<17:25,  1.25it/s]
4|5161|Loss: 0.8376672267913818:  80%|███████▉  | 5160/6470 [1:04:07<17:25,  1.25it/s]
4|5161|Loss: 0.8376672267913818:  80%|███████▉  | 5161/6470 [1:04:07<16:42,  1.31it/s]
4|5162|Loss: 0.9692678451538086:  80%|███████▉  | 5161/6470 [1:04:08<16:42,  1.31it/s]
4|5162|Loss: 0.9692678451538086:  80%|███████▉  | 5162/6470 [1:04:08<16:16,  1.34it/s]
4|5163|Loss: 0.9121827483177185:  80%|███████▉  | 5162/6470 [1:04:08<16:16,  1.34it/s]
4|5163|Loss: 0.9121827483177185:  80%|███████▉  | 5163/6470 [1:04:09<17:13,  1.27it/s]
4|5164|Loss: 0.8887109160423279:  80%|███████▉  | 5163/6470 [1:04:09<17:13,  1.27it/s]
4|5164|Loss: 0.8887109160423279:  80%|███████▉  | 5164/6470 [1:04:10<16:35,  1.31it/s]
4|5165|Loss: 0.7340488433837891:  80%|███████▉  | 5164/6470 [1:04:10<16:35,  1.31it/s]
4|5165|Loss: 0.7340488433837891:  80%|███████▉  | 5165/6470 [1:04:10<16:40,  1.30it/s]
4|5166|Loss: 1.0138779878616333:  80%|███████▉  | 5165/6470 [1:04:11<16:40,  1.30it/s]
4|5166|Loss: 1.0138779878616333:  80%|███████▉  | 5166/6470 [1:04:11<15:44,  1.38it/s]
4|5167|Loss: 0.9416978359222412:  80%|███████▉  | 5166/6470 [1:04:11<15:44,  1.38it/s]
4|5167|Loss: 0.9416978359222412:  80%|███████▉  | 5167/6470 [1:04:11<14:36,  1.49it/s]
4|5168|Loss: 0.8862991333007812:  80%|███████▉  | 5167/6470 [1:04:12<14:36,  1.49it/s]
4|5168|Loss: 0.8862991333007812:  80%|███████▉  | 5168/6470 [1:04:12<16:00,  1.36it/s]
4|5169|Loss: 0.7501358389854431:  80%|███████▉  | 5168/6470 [1:04:13<16:00,  1.36it/s]
4|5169|Loss: 0.7501358389854431:  80%|███████▉  | 5169/6470 [1:04:13<16:41,  1.30it/s]
4|5170|Loss: 0.7296246290206909:  80%|███████▉  | 5169/6470 [1:04:14<16:41,  1.30it/s]
4|5170|Loss: 0.7296246290206909:  80%|███████▉  | 5170/6470 [1:04:14<15:50,  1.37it/s]
4|5171|Loss: 0.8560183644294739:  80%|███████▉  | 5170/6470 [1:04:14<15:50,  1.37it/s]
4|5171|Loss: 0.8560183644294739:  80%|███████▉  | 5171/6470 [1:04:15<15:33,  1.39it/s]
4|5172|Loss: 0.8283414840698242:  80%|███████▉  | 5171/6470 [1:04:15<15:33,  1.39it/s]
4|5172|Loss: 0.8283414840698242:  80%|███████▉  | 5172/6470 [1:04:15<15:29,  1.40it/s]
4|5173|Loss: 0.79795902967453:  80%|███████▉  | 5172/6470 [1:04:16<15:29,  1.40it/s]  
4|5173|Loss: 0.79795902967453:  80%|███████▉  | 5173/6470 [1:04:16<15:57,  1.35it/s]
4|5174|Loss: 0.8159583806991577:  80%|███████▉  | 5173/6470 [1:04:16<15:57,  1.35it/s]
4|5174|Loss: 0.8159583806991577:  80%|███████▉  | 5174/6470 [1:04:17<15:28,  1.40it/s]
4|5175|Loss: 0.8294554352760315:  80%|███████▉  | 5174/6470 [1:04:17<15:28,  1.40it/s]
4|5175|Loss: 0.8294554352760315:  80%|███████▉  | 5175/6470 [1:04:17<14:16,  1.51it/s]
4|5176|Loss: 1.0671533346176147:  80%|███████▉  | 5175/6470 [1:04:18<14:16,  1.51it/s]
4|5176|Loss: 1.0671533346176147:  80%|████████  | 5176/6470 [1:04:18<14:03,  1.53it/s]
4|5177|Loss: 0.7144942879676819:  80%|████████  | 5176/6470 [1:04:18<14:03,  1.53it/s]
4|5177|Loss: 0.7144942879676819:  80%|████████  | 5177/6470 [1:04:19<14:54,  1.45it/s]
4|5178|Loss: 0.8440316319465637:  80%|████████  | 5177/6470 [1:04:19<14:54,  1.45it/s]
4|5178|Loss: 0.8440316319465637:  80%|████████  | 5178/6470 [1:04:19<14:27,  1.49it/s]
4|5179|Loss: 0.7148271203041077:  80%|████████  | 5178/6470 [1:04:20<14:27,  1.49it/s]
4|5179|Loss: 0.7148271203041077:  80%|████████  | 5179/6470 [1:04:20<13:56,  1.54it/s]
4|5180|Loss: 0.8744321465492249:  80%|████████  | 5179/6470 [1:04:20<13:56,  1.54it/s]
4|5180|Loss: 0.8744321465492249:  80%|████████  | 5180/6470 [1:04:21<14:38,  1.47it/s]
4|5181|Loss: 0.860730767250061:  80%|████████  | 5180/6470 [1:04:21<14:38,  1.47it/s] 
4|5181|Loss: 0.860730767250061:  80%|████████  | 5181/6470 [1:04:21<15:19,  1.40it/s]
4|5182|Loss: 0.821839451789856:  80%|████████  | 5181/6470 [1:04:22<15:19,  1.40it/s]
4|5182|Loss: 0.821839451789856:  80%|████████  | 5182/6470 [1:04:22<13:51,  1.55it/s]
4|5183|Loss: 0.9209325313568115:  80%|████████  | 5182/6470 [1:04:22<13:51,  1.55it/s]
4|5183|Loss: 0.9209325313568115:  80%|████████  | 5183/6470 [1:04:23<13:50,  1.55it/s]
4|5184|Loss: 0.893899142742157:  80%|████████  | 5183/6470 [1:04:23<13:50,  1.55it/s] 
4|5184|Loss: 0.893899142742157:  80%|████████  | 5184/6470 [1:04:23<13:45,  1.56it/s]
4|5185|Loss: 0.7534645199775696:  80%|████████  | 5184/6470 [1:04:24<13:45,  1.56it/s]
4|5185|Loss: 0.7534645199775696:  80%|████████  | 5185/6470 [1:04:24<15:24,  1.39it/s]
4|5186|Loss: 0.7948338985443115:  80%|████████  | 5185/6470 [1:04:25<15:24,  1.39it/s]
4|5186|Loss: 0.7948338985443115:  80%|████████  | 5186/6470 [1:04:25<16:37,  1.29it/s]
4|5187|Loss: 0.7256457805633545:  80%|████████  | 5186/6470 [1:04:25<16:37,  1.29it/s]
4|5187|Loss: 0.7256457805633545:  80%|████████  | 5187/6470 [1:04:26<14:52,  1.44it/s]
4|5188|Loss: 0.8415580987930298:  80%|████████  | 5187/6470 [1:04:26<14:52,  1.44it/s]
4|5188|Loss: 0.8415580987930298:  80%|████████  | 5188/6470 [1:04:26<14:57,  1.43it/s]
4|5189|Loss: 0.7556160092353821:  80%|████████  | 5188/6470 [1:04:27<14:57,  1.43it/s]
4|5189|Loss: 0.7556160092353821:  80%|████████  | 5189/6470 [1:04:27<14:58,  1.43it/s]
4|5190|Loss: 0.8822064399719238:  80%|████████  | 5189/6470 [1:04:27<14:58,  1.43it/s]
4|5190|Loss: 0.8822064399719238:  80%|████████  | 5190/6470 [1:04:28<15:14,  1.40it/s]
4|5191|Loss: 0.8351085782051086:  80%|████████  | 5190/6470 [1:04:28<15:14,  1.40it/s]
4|5191|Loss: 0.8351085782051086:  80%|████████  | 5191/6470 [1:04:28<15:09,  1.41it/s]
4|5192|Loss: 0.8415685892105103:  80%|████████  | 5191/6470 [1:04:29<15:09,  1.41it/s]
4|5192|Loss: 0.8415685892105103:  80%|████████  | 5192/6470 [1:04:29<14:33,  1.46it/s]
4|5193|Loss: 0.8697230815887451:  80%|████████  | 5192/6470 [1:04:29<14:33,  1.46it/s]
4|5193|Loss: 0.8697230815887451:  80%|████████  | 5193/6470 [1:04:30<14:31,  1.47it/s]
4|5194|Loss: 0.9207730889320374:  80%|████████  | 5193/6470 [1:04:30<14:31,  1.47it/s]
4|5194|Loss: 0.9207730889320374:  80%|████████  | 5194/6470 [1:04:30<15:14,  1.40it/s]
4|5195|Loss: 0.941749632358551:  80%|████████  | 5194/6470 [1:04:31<15:14,  1.40it/s] 
4|5195|Loss: 0.941749632358551:  80%|████████  | 5195/6470 [1:04:31<16:39,  1.28it/s]
4|5196|Loss: 0.8024024963378906:  80%|████████  | 5195/6470 [1:04:32<16:39,  1.28it/s]
4|5196|Loss: 0.8024024963378906:  80%|████████  | 5196/6470 [1:04:32<17:19,  1.23it/s]
4|5197|Loss: 0.8586003184318542:  80%|████████  | 5196/6470 [1:04:33<17:19,  1.23it/s]
4|5197|Loss: 0.8586003184318542:  80%|████████  | 5197/6470 [1:04:33<18:00,  1.18it/s]
4|5198|Loss: 0.894620418548584:  80%|████████  | 5197/6470 [1:04:34<18:00,  1.18it/s] 
4|5198|Loss: 0.894620418548584:  80%|████████  | 5198/6470 [1:04:34<18:56,  1.12it/s]
4|5199|Loss: 0.6808125376701355:  80%|████████  | 5198/6470 [1:04:35<18:56,  1.12it/s]
4|5199|Loss: 0.6808125376701355:  80%|████████  | 5199/6470 [1:04:35<19:10,  1.11it/s]
4|5200|Loss: 0.8007080554962158:  80%|████████  | 5199/6470 [1:04:36<19:10,  1.11it/s]
4|5200|Loss: 0.8007080554962158:  80%|████████  | 5200/6470 [1:04:36<18:21,  1.15it/s]
4|5201|Loss: 0.796568751335144:  80%|████████  | 5200/6470 [1:04:36<18:21,  1.15it/s] 
4|5201|Loss: 0.796568751335144:  80%|████████  | 5201/6470 [1:04:37<17:37,  1.20it/s]
4|5202|Loss: 0.757781445980072:  80%|████████  | 5201/6470 [1:04:37<17:37,  1.20it/s]
4|5202|Loss: 0.757781445980072:  80%|████████  | 5202/6470 [1:04:38<17:55,  1.18it/s]
4|5203|Loss: 0.8820027709007263:  80%|████████  | 5202/6470 [1:04:38<17:55,  1.18it/s]
4|5203|Loss: 0.8820027709007263:  80%|████████  | 5203/6470 [1:04:39<18:31,  1.14it/s]
4|5204|Loss: 0.8029611706733704:  80%|████████  | 5203/6470 [1:04:39<18:31,  1.14it/s]
4|5204|Loss: 0.8029611706733704:  80%|████████  | 5204/6470 [1:04:39<18:35,  1.14it/s]
4|5205|Loss: 0.8806170225143433:  80%|████████  | 5204/6470 [1:04:40<18:35,  1.14it/s]
4|5205|Loss: 0.8806170225143433:  80%|████████  | 5205/6470 [1:04:40<18:52,  1.12it/s]
4|5206|Loss: 0.7815998196601868:  80%|████████  | 5205/6470 [1:04:41<18:52,  1.12it/s]
4|5206|Loss: 0.7815998196601868:  80%|████████  | 5206/6470 [1:04:41<19:09,  1.10it/s]
4|5207|Loss: 0.9377126693725586:  80%|████████  | 5206/6470 [1:04:42<19:09,  1.10it/s]
4|5207|Loss: 0.9377126693725586:  80%|████████  | 5207/6470 [1:04:42<17:06,  1.23it/s]
4|5208|Loss: 0.8485127687454224:  80%|████████  | 5207/6470 [1:04:42<17:06,  1.23it/s]
4|5208|Loss: 0.8485127687454224:  80%|████████  | 5208/6470 [1:04:43<16:36,  1.27it/s]
4|5209|Loss: 1.0306349992752075:  80%|████████  | 5208/6470 [1:04:43<16:36,  1.27it/s]
4|5209|Loss: 1.0306349992752075:  81%|████████  | 5209/6470 [1:04:43<16:28,  1.28it/s]
4|5210|Loss: 0.8719833493232727:  81%|████████  | 5209/6470 [1:04:44<16:28,  1.28it/s]
4|5210|Loss: 0.8719833493232727:  81%|████████  | 5210/6470 [1:04:44<15:03,  1.39it/s]
4|5211|Loss: 0.8028943538665771:  81%|████████  | 5210/6470 [1:04:44<15:03,  1.39it/s]
4|5211|Loss: 0.8028943538665771:  81%|████████  | 5211/6470 [1:04:44<13:45,  1.52it/s]
4|5212|Loss: 0.8087594509124756:  81%|████████  | 5211/6470 [1:04:45<13:45,  1.52it/s]
4|5212|Loss: 0.8087594509124756:  81%|████████  | 5212/6470 [1:04:45<13:21,  1.57it/s]
4|5213|Loss: 0.7640337347984314:  81%|████████  | 5212/6470 [1:04:46<13:21,  1.57it/s]
4|5213|Loss: 0.7640337347984314:  81%|████████  | 5213/6470 [1:04:46<14:50,  1.41it/s]
4|5214|Loss: 1.0342947244644165:  81%|████████  | 5213/6470 [1:04:46<14:50,  1.41it/s]
4|5214|Loss: 1.0342947244644165:  81%|████████  | 5214/6470 [1:04:47<16:18,  1.28it/s]
4|5215|Loss: 0.931299090385437:  81%|████████  | 5214/6470 [1:04:47<16:18,  1.28it/s] 
4|5215|Loss: 0.931299090385437:  81%|████████  | 5215/6470 [1:04:48<17:24,  1.20it/s]
4|5216|Loss: 0.8743709921836853:  81%|████████  | 5215/6470 [1:04:48<17:24,  1.20it/s]
4|5216|Loss: 0.8743709921836853:  81%|████████  | 5216/6470 [1:04:49<16:46,  1.25it/s]
4|5217|Loss: 0.7610055208206177:  81%|████████  | 5216/6470 [1:04:49<16:46,  1.25it/s]
4|5217|Loss: 0.7610055208206177:  81%|████████  | 5217/6470 [1:04:49<15:43,  1.33it/s]
4|5218|Loss: 0.858241617679596:  81%|████████  | 5217/6470 [1:04:50<15:43,  1.33it/s] 
4|5218|Loss: 0.858241617679596:  81%|████████  | 5218/6470 [1:04:50<16:25,  1.27it/s]
4|5219|Loss: 0.8700206875801086:  81%|████████  | 5218/6470 [1:04:51<16:25,  1.27it/s]
4|5219|Loss: 0.8700206875801086:  81%|████████  | 5219/6470 [1:04:51<16:42,  1.25it/s]
4|5220|Loss: 0.8981689214706421:  81%|████████  | 5219/6470 [1:04:51<16:42,  1.25it/s]
4|5220|Loss: 0.8981689214706421:  81%|████████  | 5220/6470 [1:04:52<16:40,  1.25it/s]
4|5221|Loss: 0.8753595948219299:  81%|████████  | 5220/6470 [1:04:52<16:40,  1.25it/s]
4|5221|Loss: 0.8753595948219299:  81%|████████  | 5221/6470 [1:04:52<16:27,  1.26it/s]
4|5222|Loss: 0.8407802581787109:  81%|████████  | 5221/6470 [1:04:53<16:27,  1.26it/s]
4|5222|Loss: 0.8407802581787109:  81%|████████  | 5222/6470 [1:04:53<17:03,  1.22it/s]
4|5223|Loss: 0.7279479503631592:  81%|████████  | 5222/6470 [1:04:54<17:03,  1.22it/s]
4|5223|Loss: 0.7279479503631592:  81%|████████  | 5223/6470 [1:04:54<16:11,  1.28it/s]
4|5224|Loss: 0.8371413946151733:  81%|████████  | 5223/6470 [1:04:54<16:11,  1.28it/s]
4|5224|Loss: 0.8371413946151733:  81%|████████  | 5224/6470 [1:04:55<16:21,  1.27it/s]
4|5225|Loss: 0.9910486936569214:  81%|████████  | 5224/6470 [1:04:55<16:21,  1.27it/s]
4|5225|Loss: 0.9910486936569214:  81%|████████  | 5225/6470 [1:04:56<17:35,  1.18it/s]
4|5226|Loss: 0.7833986282348633:  81%|████████  | 5225/6470 [1:04:56<17:35,  1.18it/s]
4|5226|Loss: 0.7833986282348633:  81%|████████  | 5226/6470 [1:04:57<17:52,  1.16it/s]
4|5227|Loss: 0.8659901022911072:  81%|████████  | 5226/6470 [1:04:57<17:52,  1.16it/s]
4|5227|Loss: 0.8659901022911072:  81%|████████  | 5227/6470 [1:04:58<17:37,  1.18it/s]
4|5228|Loss: 0.8673354983329773:  81%|████████  | 5227/6470 [1:04:58<17:37,  1.18it/s]
4|5228|Loss: 0.8673354983329773:  81%|████████  | 5228/6470 [1:04:58<18:04,  1.14it/s]
4|5229|Loss: 0.8264505863189697:  81%|████████  | 5228/6470 [1:04:59<18:04,  1.14it/s]
4|5229|Loss: 0.8264505863189697:  81%|████████  | 5229/6470 [1:04:59<17:02,  1.21it/s]
4|5230|Loss: 0.5475777387619019:  81%|████████  | 5229/6470 [1:05:00<17:02,  1.21it/s]
4|5230|Loss: 0.5475777387619019:  81%|████████  | 5230/6470 [1:05:00<17:51,  1.16it/s]
4|5231|Loss: 0.9762805104255676:  81%|████████  | 5230/6470 [1:05:01<17:51,  1.16it/s]
4|5231|Loss: 0.9762805104255676:  81%|████████  | 5231/6470 [1:05:01<18:22,  1.12it/s]
4|5232|Loss: 0.9372536540031433:  81%|████████  | 5231/6470 [1:05:02<18:22,  1.12it/s]
4|5232|Loss: 0.9372536540031433:  81%|████████  | 5232/6470 [1:05:02<17:50,  1.16it/s]
4|5233|Loss: 0.8876583576202393:  81%|████████  | 5232/6470 [1:05:02<17:50,  1.16it/s]
4|5233|Loss: 0.8876583576202393:  81%|████████  | 5233/6470 [1:05:03<17:31,  1.18it/s]
4|5234|Loss: 0.9774618744850159:  81%|████████  | 5233/6470 [1:05:03<17:31,  1.18it/s]
4|5234|Loss: 0.9774618744850159:  81%|████████  | 5234/6470 [1:05:04<17:10,  1.20it/s]
4|5235|Loss: 0.7736386060714722:  81%|████████  | 5234/6470 [1:05:04<17:10,  1.20it/s]
4|5235|Loss: 0.7736386060714722:  81%|████████  | 5235/6470 [1:05:04<16:42,  1.23it/s]
4|5236|Loss: 0.7531301379203796:  81%|████████  | 5235/6470 [1:05:05<16:42,  1.23it/s]
4|5236|Loss: 0.7531301379203796:  81%|████████  | 5236/6470 [1:05:05<14:58,  1.37it/s]
4|5237|Loss: 0.9799785017967224:  81%|████████  | 5236/6470 [1:05:05<14:58,  1.37it/s]
4|5237|Loss: 0.9799785017967224:  81%|████████  | 5237/6470 [1:05:05<14:20,  1.43it/s]
4|5238|Loss: 0.9315244555473328:  81%|████████  | 5237/6470 [1:05:06<14:20,  1.43it/s]
4|5238|Loss: 0.9315244555473328:  81%|████████  | 5238/6470 [1:05:06<15:27,  1.33it/s]
4|5239|Loss: 0.787429928779602:  81%|████████  | 5238/6470 [1:05:07<15:27,  1.33it/s] 
4|5239|Loss: 0.787429928779602:  81%|████████  | 5239/6470 [1:05:07<15:10,  1.35it/s]
4|5240|Loss: 0.865012526512146:  81%|████████  | 5239/6470 [1:05:07<15:10,  1.35it/s]
4|5240|Loss: 0.865012526512146:  81%|████████  | 5240/6470 [1:05:08<14:38,  1.40it/s]
4|5241|Loss: 0.9398989081382751:  81%|████████  | 5240/6470 [1:05:08<14:38,  1.40it/s]
4|5241|Loss: 0.9398989081382751:  81%|████████  | 5241/6470 [1:05:08<14:50,  1.38it/s]
4|5242|Loss: 0.748605489730835:  81%|████████  | 5241/6470 [1:05:09<14:50,  1.38it/s] 
4|5242|Loss: 0.748605489730835:  81%|████████  | 5242/6470 [1:05:09<14:37,  1.40it/s]
4|5243|Loss: 0.7542603611946106:  81%|████████  | 5242/6470 [1:05:09<14:37,  1.40it/s]
4|5243|Loss: 0.7542603611946106:  81%|████████  | 5243/6470 [1:05:10<14:28,  1.41it/s]
4|5244|Loss: 0.7915414571762085:  81%|████████  | 5243/6470 [1:05:10<14:28,  1.41it/s]
4|5244|Loss: 0.7915414571762085:  81%|████████  | 5244/6470 [1:05:11<14:51,  1.38it/s]
4|5245|Loss: 0.7423226237297058:  81%|████████  | 5244/6470 [1:05:11<14:51,  1.38it/s]
4|5245|Loss: 0.7423226237297058:  81%|████████  | 5245/6470 [1:05:11<14:37,  1.40it/s]
4|5246|Loss: 0.8074844479560852:  81%|████████  | 5245/6470 [1:05:12<14:37,  1.40it/s]
4|5246|Loss: 0.8074844479560852:  81%|████████  | 5246/6470 [1:05:12<14:55,  1.37it/s]
4|5247|Loss: 1.010219693183899:  81%|████████  | 5246/6470 [1:05:12<14:55,  1.37it/s] 
4|5247|Loss: 1.010219693183899:  81%|████████  | 5247/6470 [1:05:13<15:03,  1.35it/s]
4|5248|Loss: 0.8525009155273438:  81%|████████  | 5247/6470 [1:05:13<15:03,  1.35it/s]
4|5248|Loss: 0.8525009155273438:  81%|████████  | 5248/6470 [1:05:13<14:41,  1.39it/s]
4|5249|Loss: 0.8609590530395508:  81%|████████  | 5248/6470 [1:05:14<14:41,  1.39it/s]
4|5249|Loss: 0.8609590530395508:  81%|████████  | 5249/6470 [1:05:14<15:31,  1.31it/s]
4|5250|Loss: 0.8870842456817627:  81%|████████  | 5249/6470 [1:05:15<15:31,  1.31it/s]
4|5250|Loss: 0.8870842456817627:  81%|████████  | 5250/6470 [1:05:15<16:20,  1.24it/s]
4|5251|Loss: 0.8467007279396057:  81%|████████  | 5250/6470 [1:05:16<16:20,  1.24it/s]
4|5251|Loss: 0.8467007279396057:  81%|████████  | 5251/6470 [1:05:16<15:31,  1.31it/s]
4|5252|Loss: 0.8429222702980042:  81%|████████  | 5251/6470 [1:05:16<15:31,  1.31it/s]
4|5252|Loss: 0.8429222702980042:  81%|████████  | 5252/6470 [1:05:17<15:47,  1.29it/s]
4|5253|Loss: 0.7956751585006714:  81%|████████  | 5252/6470 [1:05:17<15:47,  1.29it/s]
4|5253|Loss: 0.7956751585006714:  81%|████████  | 5253/6470 [1:05:18<16:33,  1.23it/s]
4|5254|Loss: 0.7997755408287048:  81%|████████  | 5253/6470 [1:05:18<16:33,  1.23it/s]
4|5254|Loss: 0.7997755408287048:  81%|████████  | 5254/6470 [1:05:18<16:31,  1.23it/s]
4|5255|Loss: 0.7738680243492126:  81%|████████  | 5254/6470 [1:05:19<16:31,  1.23it/s]
4|5255|Loss: 0.7738680243492126:  81%|████████  | 5255/6470 [1:05:19<17:05,  1.18it/s]
4|5256|Loss: 0.9640972018241882:  81%|████████  | 5255/6470 [1:05:20<17:05,  1.18it/s]
4|5256|Loss: 0.9640972018241882:  81%|████████  | 5256/6470 [1:05:20<16:37,  1.22it/s]
4|5257|Loss: 0.7809441089630127:  81%|████████  | 5256/6470 [1:05:21<16:37,  1.22it/s]
4|5257|Loss: 0.7809441089630127:  81%|████████▏ | 5257/6470 [1:05:21<17:07,  1.18it/s]
4|5258|Loss: 0.877651035785675:  81%|████████▏ | 5257/6470 [1:05:22<17:07,  1.18it/s] 
4|5258|Loss: 0.877651035785675:  81%|████████▏ | 5258/6470 [1:05:22<17:37,  1.15it/s]
4|5259|Loss: 0.835259735584259:  81%|████████▏ | 5258/6470 [1:05:22<17:37,  1.15it/s]
4|5259|Loss: 0.835259735584259:  81%|████████▏ | 5259/6470 [1:05:23<17:52,  1.13it/s]
4|5260|Loss: 0.7956736087799072:  81%|████████▏ | 5259/6470 [1:05:23<17:52,  1.13it/s]
4|5260|Loss: 0.7956736087799072:  81%|████████▏ | 5260/6470 [1:05:24<16:46,  1.20it/s]
4|5261|Loss: 0.8952717185020447:  81%|████████▏ | 5260/6470 [1:05:24<16:46,  1.20it/s]
4|5261|Loss: 0.8952717185020447:  81%|████████▏ | 5261/6470 [1:05:24<16:54,  1.19it/s]
4|5262|Loss: 0.812899112701416:  81%|████████▏ | 5261/6470 [1:05:25<16:54,  1.19it/s] 
4|5262|Loss: 0.812899112701416:  81%|████████▏ | 5262/6470 [1:05:25<16:33,  1.22it/s]
4|5263|Loss: 0.8391401767730713:  81%|████████▏ | 5262/6470 [1:05:26<16:33,  1.22it/s]
4|5263|Loss: 0.8391401767730713:  81%|████████▏ | 5263/6470 [1:05:26<17:09,  1.17it/s]
4|5264|Loss: 0.7936611175537109:  81%|████████▏ | 5263/6470 [1:05:27<17:09,  1.17it/s]
4|5264|Loss: 0.7936611175537109:  81%|████████▏ | 5264/6470 [1:05:27<16:28,  1.22it/s]
4|5265|Loss: 0.9473899006843567:  81%|████████▏ | 5264/6470 [1:05:27<16:28,  1.22it/s]
4|5265|Loss: 0.9473899006843567:  81%|████████▏ | 5265/6470 [1:05:28<16:08,  1.24it/s]
4|5266|Loss: 0.9649438858032227:  81%|████████▏ | 5265/6470 [1:05:28<16:08,  1.24it/s]
4|5266|Loss: 0.9649438858032227:  81%|████████▏ | 5266/6470 [1:05:28<14:48,  1.36it/s]
4|5267|Loss: 0.8310964107513428:  81%|████████▏ | 5266/6470 [1:05:29<14:48,  1.36it/s]
4|5267|Loss: 0.8310964107513428:  81%|████████▏ | 5267/6470 [1:05:29<14:22,  1.40it/s]
4|5268|Loss: 0.9227036237716675:  81%|████████▏ | 5267/6470 [1:05:29<14:22,  1.40it/s]
4|5268|Loss: 0.9227036237716675:  81%|████████▏ | 5268/6470 [1:05:30<15:26,  1.30it/s]
4|5269|Loss: 0.8180726170539856:  81%|████████▏ | 5268/6470 [1:05:30<15:26,  1.30it/s]
4|5269|Loss: 0.8180726170539856:  81%|████████▏ | 5269/6470 [1:05:31<15:26,  1.30it/s]
4|5270|Loss: 1.0411845445632935:  81%|████████▏ | 5269/6470 [1:05:31<15:26,  1.30it/s]
4|5270|Loss: 1.0411845445632935:  81%|████████▏ | 5270/6470 [1:05:31<14:39,  1.36it/s]
4|5271|Loss: 0.6066954135894775:  81%|████████▏ | 5270/6470 [1:05:32<14:39,  1.36it/s]
4|5271|Loss: 0.6066954135894775:  81%|████████▏ | 5271/6470 [1:05:32<14:36,  1.37it/s]
4|5272|Loss: 0.8609909415245056:  81%|████████▏ | 5271/6470 [1:05:32<14:36,  1.37it/s]
4|5272|Loss: 0.8609909415245056:  81%|████████▏ | 5272/6470 [1:05:33<14:45,  1.35it/s]
4|5273|Loss: 0.8204529285430908:  81%|████████▏ | 5272/6470 [1:05:33<14:45,  1.35it/s]
4|5273|Loss: 0.8204529285430908:  81%|████████▏ | 5273/6470 [1:05:33<14:34,  1.37it/s]
4|5274|Loss: 0.7556465268135071:  81%|████████▏ | 5273/6470 [1:05:34<14:34,  1.37it/s]
4|5274|Loss: 0.7556465268135071:  82%|████████▏ | 5274/6470 [1:05:34<13:36,  1.47it/s]
4|5275|Loss: 1.014475703239441:  82%|████████▏ | 5274/6470 [1:05:34<13:36,  1.47it/s] 
4|5275|Loss: 1.014475703239441:  82%|████████▏ | 5275/6470 [1:05:35<14:04,  1.42it/s]
4|5276|Loss: 0.8795689344406128:  82%|████████▏ | 5275/6470 [1:05:35<14:04,  1.42it/s]
4|5276|Loss: 0.8795689344406128:  82%|████████▏ | 5276/6470 [1:05:36<15:27,  1.29it/s]
4|5277|Loss: 0.7891702055931091:  82%|████████▏ | 5276/6470 [1:05:36<15:27,  1.29it/s]
4|5277|Loss: 0.7891702055931091:  82%|████████▏ | 5277/6470 [1:05:37<15:48,  1.26it/s]
4|5278|Loss: 0.8915230631828308:  82%|████████▏ | 5277/6470 [1:05:37<15:48,  1.26it/s]
4|5278|Loss: 0.8915230631828308:  82%|████████▏ | 5278/6470 [1:05:37<15:35,  1.27it/s]
4|5279|Loss: 0.7603464126586914:  82%|████████▏ | 5278/6470 [1:05:38<15:35,  1.27it/s]
4|5279|Loss: 0.7603464126586914:  82%|████████▏ | 5279/6470 [1:05:38<15:56,  1.25it/s]
4|5280|Loss: 0.7627381682395935:  82%|████████▏ | 5279/6470 [1:05:38<15:56,  1.25it/s]
4|5280|Loss: 0.7627381682395935:  82%|████████▏ | 5280/6470 [1:05:39<14:44,  1.35it/s]
4|5281|Loss: 0.8624155521392822:  82%|████████▏ | 5280/6470 [1:05:39<14:44,  1.35it/s]
4|5281|Loss: 0.8624155521392822:  82%|████████▏ | 5281/6470 [1:05:39<14:47,  1.34it/s]
4|5282|Loss: 0.8231588006019592:  82%|████████▏ | 5281/6470 [1:05:40<14:47,  1.34it/s]
4|5282|Loss: 0.8231588006019592:  82%|████████▏ | 5282/6470 [1:05:40<14:27,  1.37it/s]
4|5283|Loss: 0.8568848371505737:  82%|████████▏ | 5282/6470 [1:05:41<14:27,  1.37it/s]
4|5283|Loss: 0.8568848371505737:  82%|████████▏ | 5283/6470 [1:05:41<14:02,  1.41it/s]
4|5284|Loss: 0.8312103152275085:  82%|████████▏ | 5283/6470 [1:05:41<14:02,  1.41it/s]
4|5284|Loss: 0.8312103152275085:  82%|████████▏ | 5284/6470 [1:05:41<12:52,  1.54it/s]
4|5285|Loss: 0.8748480677604675:  82%|████████▏ | 5284/6470 [1:05:42<12:52,  1.54it/s]
4|5285|Loss: 0.8748480677604675:  82%|████████▏ | 5285/6470 [1:05:42<13:21,  1.48it/s]
4|5286|Loss: 0.970917284488678:  82%|████████▏ | 5285/6470 [1:05:43<13:21,  1.48it/s] 
4|5286|Loss: 0.970917284488678:  82%|████████▏ | 5286/6470 [1:05:43<14:10,  1.39it/s]
4|5287|Loss: 0.892356812953949:  82%|████████▏ | 5286/6470 [1:05:43<14:10,  1.39it/s]
4|5287|Loss: 0.892356812953949:  82%|████████▏ | 5287/6470 [1:05:44<14:32,  1.36it/s]
4|5288|Loss: 1.0401726961135864:  82%|████████▏ | 5287/6470 [1:05:44<14:32,  1.36it/s]
4|5288|Loss: 1.0401726961135864:  82%|████████▏ | 5288/6470 [1:05:44<12:56,  1.52it/s]
4|5289|Loss: 0.8256210088729858:  82%|████████▏ | 5288/6470 [1:05:45<12:56,  1.52it/s]
4|5289|Loss: 0.8256210088729858:  82%|████████▏ | 5289/6470 [1:05:45<13:54,  1.42it/s]
4|5290|Loss: 0.8942205309867859:  82%|████████▏ | 5289/6470 [1:05:45<13:54,  1.42it/s]
4|5290|Loss: 0.8942205309867859:  82%|████████▏ | 5290/6470 [1:05:46<14:21,  1.37it/s]
4|5291|Loss: 0.7680361866950989:  82%|████████▏ | 5290/6470 [1:05:46<14:21,  1.37it/s]
4|5291|Loss: 0.7680361866950989:  82%|████████▏ | 5291/6470 [1:05:47<15:19,  1.28it/s]
4|5292|Loss: 0.634977400302887:  82%|████████▏ | 5291/6470 [1:05:47<15:19,  1.28it/s] 
4|5292|Loss: 0.634977400302887:  82%|████████▏ | 5292/6470 [1:05:48<16:04,  1.22it/s]
4|5293|Loss: 0.8272590637207031:  82%|████████▏ | 5292/6470 [1:05:48<16:04,  1.22it/s]
4|5293|Loss: 0.8272590637207031:  82%|████████▏ | 5293/6470 [1:05:48<16:31,  1.19it/s]
4|5294|Loss: 0.7426815032958984:  82%|████████▏ | 5293/6470 [1:05:49<16:31,  1.19it/s]
4|5294|Loss: 0.7426815032958984:  82%|████████▏ | 5294/6470 [1:05:49<15:55,  1.23it/s]
4|5295|Loss: 0.8155978918075562:  82%|████████▏ | 5294/6470 [1:05:50<15:55,  1.23it/s]
4|5295|Loss: 0.8155978918075562:  82%|████████▏ | 5295/6470 [1:05:50<15:44,  1.24it/s]
4|5296|Loss: 0.7825657725334167:  82%|████████▏ | 5295/6470 [1:05:50<15:44,  1.24it/s]
4|5296|Loss: 0.7825657725334167:  82%|████████▏ | 5296/6470 [1:05:51<15:09,  1.29it/s]
4|5297|Loss: 1.1149754524230957:  82%|████████▏ | 5296/6470 [1:05:51<15:09,  1.29it/s]
4|5297|Loss: 1.1149754524230957:  82%|████████▏ | 5297/6470 [1:05:52<16:05,  1.22it/s]
4|5298|Loss: 0.8359467387199402:  82%|████████▏ | 5297/6470 [1:05:52<16:05,  1.22it/s]
4|5298|Loss: 0.8359467387199402:  82%|████████▏ | 5298/6470 [1:05:52<15:59,  1.22it/s]
4|5299|Loss: 0.8173924684524536:  82%|████████▏ | 5298/6470 [1:05:53<15:59,  1.22it/s]
4|5299|Loss: 0.8173924684524536:  82%|████████▏ | 5299/6470 [1:05:53<14:52,  1.31it/s]
4|5300|Loss: 0.8315479159355164:  82%|████████▏ | 5299/6470 [1:05:53<14:52,  1.31it/s]
4|5300|Loss: 0.8315479159355164:  82%|████████▏ | 5300/6470 [1:05:54<14:36,  1.33it/s]
4|5301|Loss: 0.9521883130073547:  82%|████████▏ | 5300/6470 [1:05:54<14:36,  1.33it/s]
4|5301|Loss: 0.9521883130073547:  82%|████████▏ | 5301/6470 [1:05:55<14:46,  1.32it/s]
4|5302|Loss: 0.8640075922012329:  82%|████████▏ | 5301/6470 [1:05:55<14:46,  1.32it/s]
4|5302|Loss: 0.8640075922012329:  82%|████████▏ | 5302/6470 [1:05:55<15:07,  1.29it/s]
4|5303|Loss: 0.7246481776237488:  82%|████████▏ | 5302/6470 [1:05:56<15:07,  1.29it/s]
4|5303|Loss: 0.7246481776237488:  82%|████████▏ | 5303/6470 [1:05:56<14:18,  1.36it/s]
4|5304|Loss: 0.8005737066268921:  82%|████████▏ | 5303/6470 [1:05:56<14:18,  1.36it/s]
4|5304|Loss: 0.8005737066268921:  82%|████████▏ | 5304/6470 [1:05:57<13:23,  1.45it/s]
4|5305|Loss: 0.9078247547149658:  82%|████████▏ | 5304/6470 [1:05:57<13:23,  1.45it/s]
4|5305|Loss: 0.9078247547149658:  82%|████████▏ | 5305/6470 [1:05:57<13:15,  1.46it/s]
4|5306|Loss: 0.857490062713623:  82%|████████▏ | 5305/6470 [1:05:58<13:15,  1.46it/s] 
4|5306|Loss: 0.857490062713623:  82%|████████▏ | 5306/6470 [1:05:58<13:04,  1.48it/s]
4|5307|Loss: 0.7802907228469849:  82%|████████▏ | 5306/6470 [1:05:58<13:04,  1.48it/s]
4|5307|Loss: 0.7802907228469849:  82%|████████▏ | 5307/6470 [1:05:59<13:20,  1.45it/s]
4|5308|Loss: 0.931734561920166:  82%|████████▏ | 5307/6470 [1:05:59<13:20,  1.45it/s] 
4|5308|Loss: 0.931734561920166:  82%|████████▏ | 5308/6470 [1:05:59<13:42,  1.41it/s]
4|5309|Loss: 0.8145721554756165:  82%|████████▏ | 5308/6470 [1:06:00<13:42,  1.41it/s]
4|5309|Loss: 0.8145721554756165:  82%|████████▏ | 5309/6470 [1:06:00<14:54,  1.30it/s]
4|5310|Loss: 0.8093902468681335:  82%|████████▏ | 5309/6470 [1:06:01<14:54,  1.30it/s]
4|5310|Loss: 0.8093902468681335:  82%|████████▏ | 5310/6470 [1:06:01<15:45,  1.23it/s]
4|5311|Loss: 0.8413982391357422:  82%|████████▏ | 5310/6470 [1:06:02<15:45,  1.23it/s]
4|5311|Loss: 0.8413982391357422:  82%|████████▏ | 5311/6470 [1:06:02<14:33,  1.33it/s]
4|5312|Loss: 0.7021909356117249:  82%|████████▏ | 5311/6470 [1:06:02<14:33,  1.33it/s]
4|5312|Loss: 0.7021909356117249:  82%|████████▏ | 5312/6470 [1:06:02<13:41,  1.41it/s]
4|5313|Loss: 0.8599197864532471:  82%|████████▏ | 5312/6470 [1:06:03<13:41,  1.41it/s]
4|5313|Loss: 0.8599197864532471:  82%|████████▏ | 5313/6470 [1:06:03<13:31,  1.43it/s]
4|5314|Loss: 0.7959567308425903:  82%|████████▏ | 5313/6470 [1:06:04<13:31,  1.43it/s]
4|5314|Loss: 0.7959567308425903:  82%|████████▏ | 5314/6470 [1:06:04<13:32,  1.42it/s]
4|5315|Loss: 0.7548539042472839:  82%|████████▏ | 5314/6470 [1:06:04<13:32,  1.42it/s]
4|5315|Loss: 0.7548539042472839:  82%|████████▏ | 5315/6470 [1:06:05<14:19,  1.34it/s]
4|5316|Loss: 0.9109388589859009:  82%|████████▏ | 5315/6470 [1:06:05<14:19,  1.34it/s]
4|5316|Loss: 0.9109388589859009:  82%|████████▏ | 5316/6470 [1:06:06<14:46,  1.30it/s]
4|5317|Loss: 0.9002556204795837:  82%|████████▏ | 5316/6470 [1:06:06<14:46,  1.30it/s]
4|5317|Loss: 0.9002556204795837:  82%|████████▏ | 5317/6470 [1:06:06<14:57,  1.28it/s]
4|5318|Loss: 0.7422604560852051:  82%|████████▏ | 5317/6470 [1:06:07<14:57,  1.28it/s]
4|5318|Loss: 0.7422604560852051:  82%|████████▏ | 5318/6470 [1:06:07<15:04,  1.27it/s]
4|5319|Loss: 0.7590775489807129:  82%|████████▏ | 5318/6470 [1:06:08<15:04,  1.27it/s]
4|5319|Loss: 0.7590775489807129:  82%|████████▏ | 5319/6470 [1:06:08<15:38,  1.23it/s]
4|5320|Loss: 0.759626030921936:  82%|████████▏ | 5319/6470 [1:06:08<15:38,  1.23it/s] 
4|5320|Loss: 0.759626030921936:  82%|████████▏ | 5320/6470 [1:06:09<15:37,  1.23it/s]
4|5321|Loss: 0.839268684387207:  82%|████████▏ | 5320/6470 [1:06:09<15:37,  1.23it/s]
4|5321|Loss: 0.839268684387207:  82%|████████▏ | 5321/6470 [1:06:10<15:29,  1.24it/s]
4|5322|Loss: 0.6471627950668335:  82%|████████▏ | 5321/6470 [1:06:10<15:29,  1.24it/s]
4|5322|Loss: 0.6471627950668335:  82%|████████▏ | 5322/6470 [1:06:10<14:17,  1.34it/s]
4|5323|Loss: 0.9540798664093018:  82%|████████▏ | 5322/6470 [1:06:11<14:17,  1.34it/s]
4|5323|Loss: 0.9540798664093018:  82%|████████▏ | 5323/6470 [1:06:11<14:53,  1.28it/s]
4|5324|Loss: 0.779445469379425:  82%|████████▏ | 5323/6470 [1:06:12<14:53,  1.28it/s] 
4|5324|Loss: 0.779445469379425:  82%|████████▏ | 5324/6470 [1:06:12<14:38,  1.30it/s]
4|5325|Loss: 0.7602537870407104:  82%|████████▏ | 5324/6470 [1:06:12<14:38,  1.30it/s]
4|5325|Loss: 0.7602537870407104:  82%|████████▏ | 5325/6470 [1:06:12<14:04,  1.36it/s]
4|5326|Loss: 0.8453139066696167:  82%|████████▏ | 5325/6470 [1:06:13<14:04,  1.36it/s]
4|5326|Loss: 0.8453139066696167:  82%|████████▏ | 5326/6470 [1:06:13<12:26,  1.53it/s]
4|5327|Loss: 0.956166684627533:  82%|████████▏ | 5326/6470 [1:06:13<12:26,  1.53it/s] 
4|5327|Loss: 0.956166684627533:  82%|████████▏ | 5327/6470 [1:06:14<11:59,  1.59it/s]
4|5328|Loss: 0.8183079361915588:  82%|████████▏ | 5327/6470 [1:06:14<11:59,  1.59it/s]
4|5328|Loss: 0.8183079361915588:  82%|████████▏ | 5328/6470 [1:06:14<13:43,  1.39it/s]
4|5329|Loss: 0.8492110967636108:  82%|████████▏ | 5328/6470 [1:06:15<13:43,  1.39it/s]
4|5329|Loss: 0.8492110967636108:  82%|████████▏ | 5329/6470 [1:06:15<13:58,  1.36it/s]
4|5330|Loss: 0.8664358258247375:  82%|████████▏ | 5329/6470 [1:06:16<13:58,  1.36it/s]
4|5330|Loss: 0.8664358258247375:  82%|████████▏ | 5330/6470 [1:06:16<14:15,  1.33it/s]
4|5331|Loss: 0.9269520044326782:  82%|████████▏ | 5330/6470 [1:06:16<14:15,  1.33it/s]
4|5331|Loss: 0.9269520044326782:  82%|████████▏ | 5331/6470 [1:06:17<14:17,  1.33it/s]
4|5332|Loss: 1.0151647329330444:  82%|████████▏ | 5331/6470 [1:06:17<14:17,  1.33it/s]
4|5332|Loss: 1.0151647329330444:  82%|████████▏ | 5332/6470 [1:06:18<14:28,  1.31it/s]
4|5333|Loss: 0.7910001277923584:  82%|████████▏ | 5332/6470 [1:06:18<14:28,  1.31it/s]
4|5333|Loss: 0.7910001277923584:  82%|████████▏ | 5333/6470 [1:06:18<15:08,  1.25it/s]
4|5334|Loss: 0.8772997856140137:  82%|████████▏ | 5333/6470 [1:06:19<15:08,  1.25it/s]
4|5334|Loss: 0.8772997856140137:  82%|████████▏ | 5334/6470 [1:06:19<15:00,  1.26it/s]
4|5335|Loss: 0.8107132315635681:  82%|████████▏ | 5334/6470 [1:06:20<15:00,  1.26it/s]
4|5335|Loss: 0.8107132315635681:  82%|████████▏ | 5335/6470 [1:06:20<13:39,  1.39it/s]
4|5336|Loss: 0.8790163397789001:  82%|████████▏ | 5335/6470 [1:06:20<13:39,  1.39it/s]
4|5336|Loss: 0.8790163397789001:  82%|████████▏ | 5336/6470 [1:06:21<14:27,  1.31it/s]
4|5337|Loss: 0.8751193284988403:  82%|████████▏ | 5336/6470 [1:06:21<14:27,  1.31it/s]
4|5337|Loss: 0.8751193284988403:  82%|████████▏ | 5337/6470 [1:06:21<14:43,  1.28it/s]
4|5338|Loss: 0.8745737075805664:  82%|████████▏ | 5337/6470 [1:06:22<14:43,  1.28it/s]
4|5338|Loss: 0.8745737075805664:  83%|████████▎ | 5338/6470 [1:06:22<14:40,  1.29it/s]
4|5339|Loss: 0.9005876183509827:  83%|████████▎ | 5338/6470 [1:06:23<14:40,  1.29it/s]
4|5339|Loss: 0.9005876183509827:  83%|████████▎ | 5339/6470 [1:06:23<13:17,  1.42it/s]
4|5340|Loss: 0.810931921005249:  83%|████████▎ | 5339/6470 [1:06:23<13:17,  1.42it/s] 
4|5340|Loss: 0.810931921005249:  83%|████████▎ | 5340/6470 [1:06:23<13:33,  1.39it/s]
4|5341|Loss: 0.8290882110595703:  83%|████████▎ | 5340/6470 [1:06:24<13:33,  1.39it/s]
4|5341|Loss: 0.8290882110595703:  83%|████████▎ | 5341/6470 [1:06:24<13:17,  1.42it/s]
4|5342|Loss: 0.8421241044998169:  83%|████████▎ | 5341/6470 [1:06:25<13:17,  1.42it/s]
4|5342|Loss: 0.8421241044998169:  83%|████████▎ | 5342/6470 [1:06:25<13:26,  1.40it/s]
4|5343|Loss: 0.6713221669197083:  83%|████████▎ | 5342/6470 [1:06:25<13:26,  1.40it/s]
4|5343|Loss: 0.6713221669197083:  83%|████████▎ | 5343/6470 [1:06:26<14:06,  1.33it/s]
4|5344|Loss: 1.0146069526672363:  83%|████████▎ | 5343/6470 [1:06:26<14:06,  1.33it/s]
4|5344|Loss: 1.0146069526672363:  83%|████████▎ | 5344/6470 [1:06:27<14:26,  1.30it/s]
4|5345|Loss: 0.7886633276939392:  83%|████████▎ | 5344/6470 [1:06:27<14:26,  1.30it/s]
4|5345|Loss: 0.7886633276939392:  83%|████████▎ | 5345/6470 [1:06:27<14:34,  1.29it/s]
4|5346|Loss: 0.95201575756073:  83%|████████▎ | 5345/6470 [1:06:28<14:34,  1.29it/s]  
4|5346|Loss: 0.95201575756073:  83%|████████▎ | 5346/6470 [1:06:28<14:32,  1.29it/s]
4|5347|Loss: 0.7731538414955139:  83%|████████▎ | 5346/6470 [1:06:28<14:32,  1.29it/s]
4|5347|Loss: 0.7731538414955139:  83%|████████▎ | 5347/6470 [1:06:29<13:09,  1.42it/s]
4|5348|Loss: 0.814514696598053:  83%|████████▎ | 5347/6470 [1:06:29<13:09,  1.42it/s] 
4|5348|Loss: 0.814514696598053:  83%|████████▎ | 5348/6470 [1:06:29<13:18,  1.41it/s]
4|5349|Loss: 0.9789707660675049:  83%|████████▎ | 5348/6470 [1:06:30<13:18,  1.41it/s]
4|5349|Loss: 0.9789707660675049:  83%|████████▎ | 5349/6470 [1:06:30<12:53,  1.45it/s]
4|5350|Loss: 0.81285160779953:  83%|████████▎ | 5349/6470 [1:06:30<12:53,  1.45it/s]  
4|5350|Loss: 0.81285160779953:  83%|████████▎ | 5350/6470 [1:06:31<13:40,  1.37it/s]
4|5351|Loss: 0.8747670650482178:  83%|████████▎ | 5350/6470 [1:06:31<13:40,  1.37it/s]
4|5351|Loss: 0.8747670650482178:  83%|████████▎ | 5351/6470 [1:06:32<14:27,  1.29it/s]
4|5352|Loss: 0.8459896445274353:  83%|████████▎ | 5351/6470 [1:06:32<14:27,  1.29it/s]
4|5352|Loss: 0.8459896445274353:  83%|████████▎ | 5352/6470 [1:06:32<13:34,  1.37it/s]
4|5353|Loss: 0.9109178781509399:  83%|████████▎ | 5352/6470 [1:06:33<13:34,  1.37it/s]
4|5353|Loss: 0.9109178781509399:  83%|████████▎ | 5353/6470 [1:06:33<14:21,  1.30it/s]
4|5354|Loss: 0.8441758751869202:  83%|████████▎ | 5353/6470 [1:06:34<14:21,  1.30it/s]
4|5354|Loss: 0.8441758751869202:  83%|████████▎ | 5354/6470 [1:06:34<14:32,  1.28it/s]
4|5355|Loss: 0.8000578284263611:  83%|████████▎ | 5354/6470 [1:06:34<14:32,  1.28it/s]
4|5355|Loss: 0.8000578284263611:  83%|████████▎ | 5355/6470 [1:06:35<14:24,  1.29it/s]
4|5356|Loss: 0.762590765953064:  83%|████████▎ | 5355/6470 [1:06:35<14:24,  1.29it/s] 
4|5356|Loss: 0.762590765953064:  83%|████████▎ | 5356/6470 [1:06:36<14:19,  1.30it/s]
4|5357|Loss: 0.8411740660667419:  83%|████████▎ | 5356/6470 [1:06:36<14:19,  1.30it/s]
4|5357|Loss: 0.8411740660667419:  83%|████████▎ | 5357/6470 [1:06:36<15:02,  1.23it/s]
4|5358|Loss: 0.836394190788269:  83%|████████▎ | 5357/6470 [1:06:37<15:02,  1.23it/s] 
4|5358|Loss: 0.836394190788269:  83%|████████▎ | 5358/6470 [1:06:37<14:35,  1.27it/s]
4|5359|Loss: 1.077168345451355:  83%|████████▎ | 5358/6470 [1:06:38<14:35,  1.27it/s]
4|5359|Loss: 1.077168345451355:  83%|████████▎ | 5359/6470 [1:06:38<13:54,  1.33it/s]
4|5360|Loss: 0.9573633670806885:  83%|████████▎ | 5359/6470 [1:06:38<13:54,  1.33it/s]
4|5360|Loss: 0.9573633670806885:  83%|████████▎ | 5360/6470 [1:06:38<12:57,  1.43it/s]
4|5361|Loss: 0.8153976798057556:  83%|████████▎ | 5360/6470 [1:06:39<12:57,  1.43it/s]
4|5361|Loss: 0.8153976798057556:  83%|████████▎ | 5361/6470 [1:06:39<13:41,  1.35it/s]
4|5362|Loss: 0.7815220952033997:  83%|████████▎ | 5361/6470 [1:06:40<13:41,  1.35it/s]
4|5362|Loss: 0.7815220952033997:  83%|████████▎ | 5362/6470 [1:06:40<14:03,  1.31it/s]
4|5363|Loss: 0.7147858142852783:  83%|████████▎ | 5362/6470 [1:06:40<14:03,  1.31it/s]
4|5363|Loss: 0.7147858142852783:  83%|████████▎ | 5363/6470 [1:06:41<13:51,  1.33it/s]
4|5364|Loss: 0.7244096398353577:  83%|████████▎ | 5363/6470 [1:06:41<13:51,  1.33it/s]
4|5364|Loss: 0.7244096398353577:  83%|████████▎ | 5364/6470 [1:06:42<13:50,  1.33it/s]
4|5365|Loss: 0.9772974848747253:  83%|████████▎ | 5364/6470 [1:06:42<13:50,  1.33it/s]
4|5365|Loss: 0.9772974848747253:  83%|████████▎ | 5365/6470 [1:06:42<13:50,  1.33it/s]
4|5366|Loss: 0.7573818564414978:  83%|████████▎ | 5365/6470 [1:06:43<13:50,  1.33it/s]
4|5366|Loss: 0.7573818564414978:  83%|████████▎ | 5366/6470 [1:06:43<13:15,  1.39it/s]
4|5367|Loss: 0.7451063394546509:  83%|████████▎ | 5366/6470 [1:06:43<13:15,  1.39it/s]
4|5367|Loss: 0.7451063394546509:  83%|████████▎ | 5367/6470 [1:06:43<12:11,  1.51it/s]
4|5368|Loss: 0.6857958436012268:  83%|████████▎ | 5367/6470 [1:06:44<12:11,  1.51it/s]
4|5368|Loss: 0.6857958436012268:  83%|████████▎ | 5368/6470 [1:06:44<12:00,  1.53it/s]
4|5369|Loss: 0.8828911781311035:  83%|████████▎ | 5368/6470 [1:06:45<12:00,  1.53it/s]
4|5369|Loss: 0.8828911781311035:  83%|████████▎ | 5369/6470 [1:06:45<12:40,  1.45it/s]
4|5370|Loss: 0.877692699432373:  83%|████████▎ | 5369/6470 [1:06:45<12:40,  1.45it/s] 
4|5370|Loss: 0.877692699432373:  83%|████████▎ | 5370/6470 [1:06:45<11:48,  1.55it/s]
4|5371|Loss: 0.5974927544593811:  83%|████████▎ | 5370/6470 [1:06:46<11:48,  1.55it/s]
4|5371|Loss: 0.5974927544593811:  83%|████████▎ | 5371/6470 [1:06:46<12:57,  1.41it/s]
4|5372|Loss: 0.8618854880332947:  83%|████████▎ | 5371/6470 [1:06:47<12:57,  1.41it/s]
4|5372|Loss: 0.8618854880332947:  83%|████████▎ | 5372/6470 [1:06:47<14:10,  1.29it/s]
4|5373|Loss: 0.8516563177108765:  83%|████████▎ | 5372/6470 [1:06:48<14:10,  1.29it/s]
4|5373|Loss: 0.8516563177108765:  83%|████████▎ | 5373/6470 [1:06:48<13:49,  1.32it/s]
4|5374|Loss: 0.7480544447898865:  83%|████████▎ | 5373/6470 [1:06:48<13:49,  1.32it/s]
4|5374|Loss: 0.7480544447898865:  83%|████████▎ | 5374/6470 [1:06:48<12:28,  1.46it/s]
4|5375|Loss: 0.8457485437393188:  83%|████████▎ | 5374/6470 [1:06:49<12:28,  1.46it/s]
4|5375|Loss: 0.8457485437393188:  83%|████████▎ | 5375/6470 [1:06:49<12:45,  1.43it/s]
4|5376|Loss: 0.8651635050773621:  83%|████████▎ | 5375/6470 [1:06:50<12:45,  1.43it/s]
4|5376|Loss: 0.8651635050773621:  83%|████████▎ | 5376/6470 [1:06:50<12:46,  1.43it/s]
4|5377|Loss: 0.7452440857887268:  83%|████████▎ | 5376/6470 [1:06:50<12:46,  1.43it/s]
4|5377|Loss: 0.7452440857887268:  83%|████████▎ | 5377/6470 [1:06:51<12:43,  1.43it/s]
4|5378|Loss: 0.9667699337005615:  83%|████████▎ | 5377/6470 [1:06:51<12:43,  1.43it/s]
4|5378|Loss: 0.9667699337005615:  83%|████████▎ | 5378/6470 [1:06:51<12:43,  1.43it/s]
4|5379|Loss: 0.8185580968856812:  83%|████████▎ | 5378/6470 [1:06:52<12:43,  1.43it/s]
4|5379|Loss: 0.8185580968856812:  83%|████████▎ | 5379/6470 [1:06:52<13:27,  1.35it/s]
4|5380|Loss: 0.8877047896385193:  83%|████████▎ | 5379/6470 [1:06:53<13:27,  1.35it/s]
4|5380|Loss: 0.8877047896385193:  83%|████████▎ | 5380/6470 [1:06:53<13:54,  1.31it/s]
4|5381|Loss: 0.8555289506912231:  83%|████████▎ | 5380/6470 [1:06:53<13:54,  1.31it/s]
4|5381|Loss: 0.8555289506912231:  83%|████████▎ | 5381/6470 [1:06:54<14:33,  1.25it/s]
4|5382|Loss: 0.8818389177322388:  83%|████████▎ | 5381/6470 [1:06:54<14:33,  1.25it/s]
4|5382|Loss: 0.8818389177322388:  83%|████████▎ | 5382/6470 [1:06:55<14:13,  1.27it/s]
4|5383|Loss: 0.8317791223526001:  83%|████████▎ | 5382/6470 [1:06:55<14:13,  1.27it/s]
4|5383|Loss: 0.8317791223526001:  83%|████████▎ | 5383/6470 [1:06:55<14:09,  1.28it/s]
4|5384|Loss: 0.701941192150116:  83%|████████▎ | 5383/6470 [1:06:56<14:09,  1.28it/s] 
4|5384|Loss: 0.701941192150116:  83%|████████▎ | 5384/6470 [1:06:56<13:37,  1.33it/s]
4|5385|Loss: 0.8455845713615417:  83%|████████▎ | 5384/6470 [1:06:56<13:37,  1.33it/s]
4|5385|Loss: 0.8455845713615417:  83%|████████▎ | 5385/6470 [1:06:57<13:51,  1.31it/s]
4|5386|Loss: 0.8550572991371155:  83%|████████▎ | 5385/6470 [1:06:57<13:51,  1.31it/s]
4|5386|Loss: 0.8550572991371155:  83%|████████▎ | 5386/6470 [1:06:58<14:30,  1.24it/s]
4|5387|Loss: 0.8861111998558044:  83%|████████▎ | 5386/6470 [1:06:58<14:30,  1.24it/s]
4|5387|Loss: 0.8861111998558044:  83%|████████▎ | 5387/6470 [1:06:58<14:08,  1.28it/s]
4|5388|Loss: 0.6987813711166382:  83%|████████▎ | 5387/6470 [1:06:59<14:08,  1.28it/s]
4|5388|Loss: 0.6987813711166382:  83%|████████▎ | 5388/6470 [1:06:59<13:35,  1.33it/s]
4|5389|Loss: 0.6255160570144653:  83%|████████▎ | 5388/6470 [1:07:00<13:35,  1.33it/s]
4|5389|Loss: 0.6255160570144653:  83%|████████▎ | 5389/6470 [1:07:00<12:55,  1.39it/s]
4|5390|Loss: 0.8014756441116333:  83%|████████▎ | 5389/6470 [1:07:00<12:55,  1.39it/s]
4|5390|Loss: 0.8014756441116333:  83%|████████▎ | 5390/6470 [1:07:01<13:06,  1.37it/s]
4|5391|Loss: 0.7810614109039307:  83%|████████▎ | 5390/6470 [1:07:01<13:06,  1.37it/s]
4|5391|Loss: 0.7810614109039307:  83%|████████▎ | 5391/6470 [1:07:01<13:03,  1.38it/s]
4|5392|Loss: 0.8021180033683777:  83%|████████▎ | 5391/6470 [1:07:02<13:03,  1.38it/s]
4|5392|Loss: 0.8021180033683777:  83%|████████▎ | 5392/6470 [1:07:02<13:19,  1.35it/s]
4|5393|Loss: 0.9290035367012024:  83%|████████▎ | 5392/6470 [1:07:02<13:19,  1.35it/s]
4|5393|Loss: 0.9290035367012024:  83%|████████▎ | 5393/6470 [1:07:03<12:41,  1.41it/s]
4|5394|Loss: 0.7166452407836914:  83%|████████▎ | 5393/6470 [1:07:03<12:41,  1.41it/s]
4|5394|Loss: 0.7166452407836914:  83%|████████▎ | 5394/6470 [1:07:03<12:42,  1.41it/s]
4|5395|Loss: 0.9071330428123474:  83%|████████▎ | 5394/6470 [1:07:04<12:42,  1.41it/s]
4|5395|Loss: 0.9071330428123474:  83%|████████▎ | 5395/6470 [1:07:04<13:14,  1.35it/s]
4|5396|Loss: 0.8660010695457458:  83%|████████▎ | 5395/6470 [1:07:05<13:14,  1.35it/s]
4|5396|Loss: 0.8660010695457458:  83%|████████▎ | 5396/6470 [1:07:05<12:23,  1.44it/s]
4|5397|Loss: 0.783776581287384:  83%|████████▎ | 5396/6470 [1:07:05<12:23,  1.44it/s] 
4|5397|Loss: 0.783776581287384:  83%|████████▎ | 5397/6470 [1:07:05<12:33,  1.42it/s]
4|5398|Loss: 0.950406014919281:  83%|████████▎ | 5397/6470 [1:07:06<12:33,  1.42it/s]
4|5398|Loss: 0.950406014919281:  83%|████████▎ | 5398/6470 [1:07:06<13:03,  1.37it/s]
4|5399|Loss: 0.8567673563957214:  83%|████████▎ | 5398/6470 [1:07:07<13:03,  1.37it/s]
4|5399|Loss: 0.8567673563957214:  83%|████████▎ | 5399/6470 [1:07:07<13:02,  1.37it/s]
4|5400|Loss: 0.8057593703269958:  83%|████████▎ | 5399/6470 [1:07:07<13:02,  1.37it/s]
4|5400|Loss: 0.8057593703269958:  83%|████████▎ | 5400/6470 [1:07:08<12:48,  1.39it/s]
4|5401|Loss: 0.9190889000892639:  83%|████████▎ | 5400/6470 [1:07:08<12:48,  1.39it/s]
4|5401|Loss: 0.9190889000892639:  83%|████████▎ | 5401/6470 [1:07:08<12:38,  1.41it/s]
4|5402|Loss: 0.9033163785934448:  83%|████████▎ | 5401/6470 [1:07:09<12:38,  1.41it/s]
4|5402|Loss: 0.9033163785934448:  83%|████████▎ | 5402/6470 [1:07:09<11:58,  1.49it/s]
4|5403|Loss: 0.9085000157356262:  83%|████████▎ | 5402/6470 [1:07:09<11:58,  1.49it/s]
4|5403|Loss: 0.9085000157356262:  84%|████████▎ | 5403/6470 [1:07:10<13:06,  1.36it/s]
4|5404|Loss: 0.8355003595352173:  84%|████████▎ | 5403/6470 [1:07:10<13:06,  1.36it/s]
4|5404|Loss: 0.8355003595352173:  84%|████████▎ | 5404/6470 [1:07:11<14:16,  1.24it/s]
4|5405|Loss: 0.7790074944496155:  84%|████████▎ | 5404/6470 [1:07:11<14:16,  1.24it/s]
4|5405|Loss: 0.7790074944496155:  84%|████████▎ | 5405/6470 [1:07:12<14:29,  1.22it/s]
4|5406|Loss: 0.8817887306213379:  84%|████████▎ | 5405/6470 [1:07:12<14:29,  1.22it/s]
4|5406|Loss: 0.8817887306213379:  84%|████████▎ | 5406/6470 [1:07:12<13:54,  1.28it/s]
4|5407|Loss: 0.8929441571235657:  84%|████████▎ | 5406/6470 [1:07:13<13:54,  1.28it/s]
4|5407|Loss: 0.8929441571235657:  84%|████████▎ | 5407/6470 [1:07:13<14:23,  1.23it/s]
4|5408|Loss: 0.9662250876426697:  84%|████████▎ | 5407/6470 [1:07:14<14:23,  1.23it/s]
4|5408|Loss: 0.9662250876426697:  84%|████████▎ | 5408/6470 [1:07:14<15:13,  1.16it/s]
4|5409|Loss: 0.9380049705505371:  84%|████████▎ | 5408/6470 [1:07:15<15:13,  1.16it/s]
4|5409|Loss: 0.9380049705505371:  84%|████████▎ | 5409/6470 [1:07:15<15:46,  1.12it/s]
4|5410|Loss: 0.8682123422622681:  84%|████████▎ | 5409/6470 [1:07:16<15:46,  1.12it/s]
4|5410|Loss: 0.8682123422622681:  84%|████████▎ | 5410/6470 [1:07:16<15:31,  1.14it/s]
4|5411|Loss: 0.999794065952301:  84%|████████▎ | 5410/6470 [1:07:17<15:31,  1.14it/s] 
4|5411|Loss: 0.999794065952301:  84%|████████▎ | 5411/6470 [1:07:17<15:45,  1.12it/s]
4|5412|Loss: 0.7328628301620483:  84%|████████▎ | 5411/6470 [1:07:17<15:45,  1.12it/s]
4|5412|Loss: 0.7328628301620483:  84%|████████▎ | 5412/6470 [1:07:18<14:31,  1.21it/s]
4|5413|Loss: 0.8723528981208801:  84%|████████▎ | 5412/6470 [1:07:18<14:31,  1.21it/s]
4|5413|Loss: 0.8723528981208801:  84%|████████▎ | 5413/6470 [1:07:18<13:22,  1.32it/s]
4|5414|Loss: 0.9258161783218384:  84%|████████▎ | 5413/6470 [1:07:19<13:22,  1.32it/s]
4|5414|Loss: 0.9258161783218384:  84%|████████▎ | 5414/6470 [1:07:19<14:01,  1.25it/s]
4|5415|Loss: 0.7847257256507874:  84%|████████▎ | 5414/6470 [1:07:20<14:01,  1.25it/s]
4|5415|Loss: 0.7847257256507874:  84%|████████▎ | 5415/6470 [1:07:20<14:12,  1.24it/s]
4|5416|Loss: 0.783997654914856:  84%|████████▎ | 5415/6470 [1:07:20<14:12,  1.24it/s] 
4|5416|Loss: 0.783997654914856:  84%|████████▎ | 5416/6470 [1:07:21<14:08,  1.24it/s]
4|5417|Loss: 0.7168379426002502:  84%|████████▎ | 5416/6470 [1:07:21<14:08,  1.24it/s]
4|5417|Loss: 0.7168379426002502:  84%|████████▎ | 5417/6470 [1:07:22<14:19,  1.23it/s]
4|5418|Loss: 0.7577638626098633:  84%|████████▎ | 5417/6470 [1:07:22<14:19,  1.23it/s]
4|5418|Loss: 0.7577638626098633:  84%|████████▎ | 5418/6470 [1:07:22<14:16,  1.23it/s]
4|5419|Loss: 0.7959526181221008:  84%|████████▎ | 5418/6470 [1:07:23<14:16,  1.23it/s]
4|5419|Loss: 0.7959526181221008:  84%|████████▍ | 5419/6470 [1:07:23<13:18,  1.32it/s]
4|5420|Loss: 0.8754709362983704:  84%|████████▍ | 5419/6470 [1:07:24<13:18,  1.32it/s]
4|5420|Loss: 0.8754709362983704:  84%|████████▍ | 5420/6470 [1:07:24<14:21,  1.22it/s]
4|5421|Loss: 0.7973761558532715:  84%|████████▍ | 5420/6470 [1:07:24<14:21,  1.22it/s]
4|5421|Loss: 0.7973761558532715:  84%|████████▍ | 5421/6470 [1:07:25<13:24,  1.30it/s]
4|5422|Loss: 0.6986022591590881:  84%|████████▍ | 5421/6470 [1:07:25<13:24,  1.30it/s]
4|5422|Loss: 0.6986022591590881:  84%|████████▍ | 5422/6470 [1:07:25<12:36,  1.39it/s]
4|5423|Loss: 0.7950934767723083:  84%|████████▍ | 5422/6470 [1:07:26<12:36,  1.39it/s]
4|5423|Loss: 0.7950934767723083:  84%|████████▍ | 5423/6470 [1:07:26<12:54,  1.35it/s]
4|5424|Loss: 0.8187838792800903:  84%|████████▍ | 5423/6470 [1:07:26<12:54,  1.35it/s]
4|5424|Loss: 0.8187838792800903:  84%|████████▍ | 5424/6470 [1:07:27<13:14,  1.32it/s]
4|5425|Loss: 0.7664777636528015:  84%|████████▍ | 5424/6470 [1:07:27<13:14,  1.32it/s]
4|5425|Loss: 0.7664777636528015:  84%|████████▍ | 5425/6470 [1:07:27<12:24,  1.40it/s]
4|5426|Loss: 0.8054476380348206:  84%|████████▍ | 5425/6470 [1:07:28<12:24,  1.40it/s]
4|5426|Loss: 0.8054476380348206:  84%|████████▍ | 5426/6470 [1:07:28<11:58,  1.45it/s]
4|5427|Loss: 0.764400839805603:  84%|████████▍ | 5426/6470 [1:07:29<11:58,  1.45it/s] 
4|5427|Loss: 0.764400839805603:  84%|████████▍ | 5427/6470 [1:07:29<13:00,  1.34it/s]
4|5428|Loss: 0.7679486870765686:  84%|████████▍ | 5427/6470 [1:07:29<13:00,  1.34it/s]
4|5428|Loss: 0.7679486870765686:  84%|████████▍ | 5428/6470 [1:07:30<13:27,  1.29it/s]
4|5429|Loss: 0.9083170890808105:  84%|████████▍ | 5428/6470 [1:07:30<13:27,  1.29it/s]
4|5429|Loss: 0.9083170890808105:  84%|████████▍ | 5429/6470 [1:07:31<13:25,  1.29it/s]
4|5430|Loss: 0.8261473178863525:  84%|████████▍ | 5429/6470 [1:07:31<13:25,  1.29it/s]
4|5430|Loss: 0.8261473178863525:  84%|████████▍ | 5430/6470 [1:07:31<12:44,  1.36it/s]
4|5431|Loss: 0.772247850894928:  84%|████████▍ | 5430/6470 [1:07:32<12:44,  1.36it/s] 
4|5431|Loss: 0.772247850894928:  84%|████████▍ | 5431/6470 [1:07:32<12:29,  1.39it/s]
4|5432|Loss: 0.7711755037307739:  84%|████████▍ | 5431/6470 [1:07:32<12:29,  1.39it/s]
4|5432|Loss: 0.7711755037307739:  84%|████████▍ | 5432/6470 [1:07:33<12:45,  1.36it/s]
4|5433|Loss: 0.7426272034645081:  84%|████████▍ | 5432/6470 [1:07:33<12:45,  1.36it/s]
4|5433|Loss: 0.7426272034645081:  84%|████████▍ | 5433/6470 [1:07:33<11:45,  1.47it/s]
4|5434|Loss: 0.8867316246032715:  84%|████████▍ | 5433/6470 [1:07:34<11:45,  1.47it/s]
4|5434|Loss: 0.8867316246032715:  84%|████████▍ | 5434/6470 [1:07:34<11:57,  1.44it/s]
4|5435|Loss: 0.7387391924858093:  84%|████████▍ | 5434/6470 [1:07:34<11:57,  1.44it/s]
4|5435|Loss: 0.7387391924858093:  84%|████████▍ | 5435/6470 [1:07:35<13:02,  1.32it/s]
4|5436|Loss: 0.8957064747810364:  84%|████████▍ | 5435/6470 [1:07:35<13:02,  1.32it/s]
4|5436|Loss: 0.8957064747810364:  84%|████████▍ | 5436/6470 [1:07:36<13:59,  1.23it/s]
4|5437|Loss: 0.7218167185783386:  84%|████████▍ | 5436/6470 [1:07:36<13:59,  1.23it/s]
4|5437|Loss: 0.7218167185783386:  84%|████████▍ | 5437/6470 [1:07:37<13:24,  1.28it/s]
4|5438|Loss: 0.6650733947753906:  84%|████████▍ | 5437/6470 [1:07:37<13:24,  1.28it/s]
4|5438|Loss: 0.6650733947753906:  84%|████████▍ | 5438/6470 [1:07:37<11:33,  1.49it/s]
4|5439|Loss: 0.9444735050201416:  84%|████████▍ | 5438/6470 [1:07:37<11:33,  1.49it/s]
4|5439|Loss: 0.9444735050201416:  84%|████████▍ | 5439/6470 [1:07:38<12:19,  1.39it/s]
4|5440|Loss: 0.6468738317489624:  84%|████████▍ | 5439/6470 [1:07:38<12:19,  1.39it/s]
4|5440|Loss: 0.6468738317489624:  84%|████████▍ | 5440/6470 [1:07:39<13:28,  1.27it/s]
4|5441|Loss: 0.8102381825447083:  84%|████████▍ | 5440/6470 [1:07:39<13:28,  1.27it/s]
4|5441|Loss: 0.8102381825447083:  84%|████████▍ | 5441/6470 [1:07:39<13:12,  1.30it/s]
4|5442|Loss: 0.8356527090072632:  84%|████████▍ | 5441/6470 [1:07:40<13:12,  1.30it/s]
4|5442|Loss: 0.8356527090072632:  84%|████████▍ | 5442/6470 [1:07:40<13:13,  1.30it/s]
4|5443|Loss: 0.8520877957344055:  84%|████████▍ | 5442/6470 [1:07:41<13:13,  1.30it/s]
4|5443|Loss: 0.8520877957344055:  84%|████████▍ | 5443/6470 [1:07:41<12:48,  1.34it/s]
4|5444|Loss: 0.7978411912918091:  84%|████████▍ | 5443/6470 [1:07:41<12:48,  1.34it/s]
4|5444|Loss: 0.7978411912918091:  84%|████████▍ | 5444/6470 [1:07:42<12:30,  1.37it/s]
4|5445|Loss: 0.7686352729797363:  84%|████████▍ | 5444/6470 [1:07:42<12:30,  1.37it/s]
4|5445|Loss: 0.7686352729797363:  84%|████████▍ | 5445/6470 [1:07:43<13:35,  1.26it/s]
4|5446|Loss: 0.8079063296318054:  84%|████████▍ | 5445/6470 [1:07:43<13:35,  1.26it/s]
4|5446|Loss: 0.8079063296318054:  84%|████████▍ | 5446/6470 [1:07:43<14:10,  1.20it/s]
4|5447|Loss: 0.8261395692825317:  84%|████████▍ | 5446/6470 [1:07:44<14:10,  1.20it/s]
4|5447|Loss: 0.8261395692825317:  84%|████████▍ | 5447/6470 [1:07:44<12:21,  1.38it/s]
4|5448|Loss: 0.847562849521637:  84%|████████▍ | 5447/6470 [1:07:44<12:21,  1.38it/s] 
4|5448|Loss: 0.847562849521637:  84%|████████▍ | 5448/6470 [1:07:45<13:00,  1.31it/s]
4|5449|Loss: 0.7916896939277649:  84%|████████▍ | 5448/6470 [1:07:45<13:00,  1.31it/s]
4|5449|Loss: 0.7916896939277649:  84%|████████▍ | 5449/6470 [1:07:46<13:00,  1.31it/s]
4|5450|Loss: 0.764052152633667:  84%|████████▍ | 5449/6470 [1:07:46<13:00,  1.31it/s] 
4|5450|Loss: 0.764052152633667:  84%|████████▍ | 5450/6470 [1:07:46<11:44,  1.45it/s]
4|5451|Loss: 0.738628089427948:  84%|████████▍ | 5450/6470 [1:07:46<11:44,  1.45it/s]
4|5451|Loss: 0.738628089427948:  84%|████████▍ | 5451/6470 [1:07:47<11:22,  1.49it/s]
4|5452|Loss: 0.8430389761924744:  84%|████████▍ | 5451/6470 [1:07:47<11:22,  1.49it/s]
4|5452|Loss: 0.8430389761924744:  84%|████████▍ | 5452/6470 [1:07:48<13:59,  1.21it/s]
4|5453|Loss: 0.8040362000465393:  84%|████████▍ | 5452/6470 [1:07:48<13:59,  1.21it/s]
4|5453|Loss: 0.8040362000465393:  84%|████████▍ | 5453/6470 [1:07:49<14:00,  1.21it/s]
4|5454|Loss: 0.7788317203521729:  84%|████████▍ | 5453/6470 [1:07:49<14:00,  1.21it/s]
4|5454|Loss: 0.7788317203521729:  84%|████████▍ | 5454/6470 [1:07:49<12:54,  1.31it/s]
4|5455|Loss: 0.9013314843177795:  84%|████████▍ | 5454/6470 [1:07:50<12:54,  1.31it/s]
4|5455|Loss: 0.9013314843177795:  84%|████████▍ | 5455/6470 [1:07:50<12:29,  1.35it/s]
4|5456|Loss: 0.8561137318611145:  84%|████████▍ | 5455/6470 [1:07:50<12:29,  1.35it/s]
4|5456|Loss: 0.8561137318611145:  84%|████████▍ | 5456/6470 [1:07:51<13:09,  1.28it/s]
4|5457|Loss: 0.834148645401001:  84%|████████▍ | 5456/6470 [1:07:51<13:09,  1.28it/s] 
4|5457|Loss: 0.834148645401001:  84%|████████▍ | 5457/6470 [1:07:52<13:52,  1.22it/s]
4|5458|Loss: 0.7523314952850342:  84%|████████▍ | 5457/6470 [1:07:52<13:52,  1.22it/s]
4|5458|Loss: 0.7523314952850342:  84%|████████▍ | 5458/6470 [1:07:52<12:22,  1.36it/s]
4|5459|Loss: 0.8358216285705566:  84%|████████▍ | 5458/6470 [1:07:53<12:22,  1.36it/s]
4|5459|Loss: 0.8358216285705566:  84%|████████▍ | 5459/6470 [1:07:53<12:56,  1.30it/s]
4|5460|Loss: 0.8827007412910461:  84%|████████▍ | 5459/6470 [1:07:54<12:56,  1.30it/s]
4|5460|Loss: 0.8827007412910461:  84%|████████▍ | 5460/6470 [1:07:54<12:53,  1.31it/s]
4|5461|Loss: 0.7423716187477112:  84%|████████▍ | 5460/6470 [1:07:54<12:53,  1.31it/s]
4|5461|Loss: 0.7423716187477112:  84%|████████▍ | 5461/6470 [1:07:55<12:31,  1.34it/s]
4|5462|Loss: 0.7312573790550232:  84%|████████▍ | 5461/6470 [1:07:55<12:31,  1.34it/s]
4|5462|Loss: 0.7312573790550232:  84%|████████▍ | 5462/6470 [1:07:55<12:55,  1.30it/s]
4|5463|Loss: 0.8862813115119934:  84%|████████▍ | 5462/6470 [1:07:56<12:55,  1.30it/s]
4|5463|Loss: 0.8862813115119934:  84%|████████▍ | 5463/6470 [1:07:56<12:29,  1.34it/s]
4|5464|Loss: 0.7693924903869629:  84%|████████▍ | 5463/6470 [1:07:57<12:29,  1.34it/s]
4|5464|Loss: 0.7693924903869629:  84%|████████▍ | 5464/6470 [1:07:57<12:30,  1.34it/s]
4|5465|Loss: 0.703540027141571:  84%|████████▍ | 5464/6470 [1:07:57<12:30,  1.34it/s] 
4|5465|Loss: 0.703540027141571:  84%|████████▍ | 5465/6470 [1:07:58<13:45,  1.22it/s]
4|5466|Loss: 0.8661662936210632:  84%|████████▍ | 5465/6470 [1:07:58<13:45,  1.22it/s]
4|5466|Loss: 0.8661662936210632:  84%|████████▍ | 5466/6470 [1:07:59<13:50,  1.21it/s]
4|5467|Loss: 1.035525918006897:  84%|████████▍ | 5466/6470 [1:07:59<13:50,  1.21it/s] 
4|5467|Loss: 1.035525918006897:  84%|████████▍ | 5467/6470 [1:07:59<13:32,  1.23it/s]
4|5468|Loss: 0.9237803816795349:  84%|████████▍ | 5467/6470 [1:08:00<13:32,  1.23it/s]
4|5468|Loss: 0.9237803816795349:  85%|████████▍ | 5468/6470 [1:08:00<13:03,  1.28it/s]
4|5469|Loss: 0.6960442066192627:  85%|████████▍ | 5468/6470 [1:08:01<13:03,  1.28it/s]
4|5469|Loss: 0.6960442066192627:  85%|████████▍ | 5469/6470 [1:08:01<12:53,  1.29it/s]
4|5470|Loss: 0.7373873591423035:  85%|████████▍ | 5469/6470 [1:08:01<12:53,  1.29it/s]
4|5470|Loss: 0.7373873591423035:  85%|████████▍ | 5470/6470 [1:08:02<12:21,  1.35it/s]
4|5471|Loss: 0.8282622694969177:  85%|████████▍ | 5470/6470 [1:08:02<12:21,  1.35it/s]
4|5471|Loss: 0.8282622694969177:  85%|████████▍ | 5471/6470 [1:08:02<12:22,  1.34it/s]
4|5472|Loss: 0.7813912034034729:  85%|████████▍ | 5471/6470 [1:08:03<12:22,  1.34it/s]
4|5472|Loss: 0.7813912034034729:  85%|████████▍ | 5472/6470 [1:08:03<12:32,  1.33it/s]
4|5473|Loss: 0.8052786588668823:  85%|████████▍ | 5472/6470 [1:08:04<12:32,  1.33it/s]
4|5473|Loss: 0.8052786588668823:  85%|████████▍ | 5473/6470 [1:08:04<12:14,  1.36it/s]
4|5474|Loss: 0.8120337724685669:  85%|████████▍ | 5473/6470 [1:08:04<12:14,  1.36it/s]
4|5474|Loss: 0.8120337724685669:  85%|████████▍ | 5474/6470 [1:08:05<12:50,  1.29it/s]
4|5475|Loss: 0.7134266495704651:  85%|████████▍ | 5474/6470 [1:08:05<12:50,  1.29it/s]
4|5475|Loss: 0.7134266495704651:  85%|████████▍ | 5475/6470 [1:08:05<12:49,  1.29it/s]
4|5476|Loss: 0.7341341972351074:  85%|████████▍ | 5475/6470 [1:08:06<12:49,  1.29it/s]
4|5476|Loss: 0.7341341972351074:  85%|████████▍ | 5476/6470 [1:08:06<12:31,  1.32it/s]
4|5477|Loss: 0.9327973127365112:  85%|████████▍ | 5476/6470 [1:08:07<12:31,  1.32it/s]
4|5477|Loss: 0.9327973127365112:  85%|████████▍ | 5477/6470 [1:08:07<12:06,  1.37it/s]
4|5478|Loss: 0.859521746635437:  85%|████████▍ | 5477/6470 [1:08:07<12:06,  1.37it/s] 
4|5478|Loss: 0.859521746635437:  85%|████████▍ | 5478/6470 [1:08:08<12:08,  1.36it/s]
4|5479|Loss: 0.8081902861595154:  85%|████████▍ | 5478/6470 [1:08:08<12:08,  1.36it/s]
4|5479|Loss: 0.8081902861595154:  85%|████████▍ | 5479/6470 [1:08:08<11:54,  1.39it/s]
4|5480|Loss: 0.8016286492347717:  85%|████████▍ | 5479/6470 [1:08:09<11:54,  1.39it/s]
4|5480|Loss: 0.8016286492347717:  85%|████████▍ | 5480/6470 [1:08:09<12:08,  1.36it/s]
4|5481|Loss: 0.7920829653739929:  85%|████████▍ | 5480/6470 [1:08:10<12:08,  1.36it/s]
4|5481|Loss: 0.7920829653739929:  85%|████████▍ | 5481/6470 [1:08:10<12:28,  1.32it/s]
4|5482|Loss: 0.7831389307975769:  85%|████████▍ | 5481/6470 [1:08:10<12:28,  1.32it/s]
4|5482|Loss: 0.7831389307975769:  85%|████████▍ | 5482/6470 [1:08:11<12:31,  1.32it/s]
4|5483|Loss: 0.79842609167099:  85%|████████▍ | 5482/6470 [1:08:11<12:31,  1.32it/s]  
4|5483|Loss: 0.79842609167099:  85%|████████▍ | 5483/6470 [1:08:11<12:15,  1.34it/s]
4|5484|Loss: 0.7858884334564209:  85%|████████▍ | 5483/6470 [1:08:12<12:15,  1.34it/s]
4|5484|Loss: 0.7858884334564209:  85%|████████▍ | 5484/6470 [1:08:12<13:14,  1.24it/s]
4|5485|Loss: 0.6802628040313721:  85%|████████▍ | 5484/6470 [1:08:13<13:14,  1.24it/s]
4|5485|Loss: 0.6802628040313721:  85%|████████▍ | 5485/6470 [1:08:13<12:59,  1.26it/s]
4|5486|Loss: 0.93746417760849:  85%|████████▍ | 5485/6470 [1:08:14<12:59,  1.26it/s]  
4|5486|Loss: 0.93746417760849:  85%|████████▍ | 5486/6470 [1:08:14<13:24,  1.22it/s]
4|5487|Loss: 0.6994345784187317:  85%|████████▍ | 5486/6470 [1:08:14<13:24,  1.22it/s]
4|5487|Loss: 0.6994345784187317:  85%|████████▍ | 5487/6470 [1:08:15<13:39,  1.20it/s]
4|5488|Loss: 0.7837640643119812:  85%|████████▍ | 5487/6470 [1:08:15<13:39,  1.20it/s]
4|5488|Loss: 0.7837640643119812:  85%|████████▍ | 5488/6470 [1:08:16<13:24,  1.22it/s]
4|5489|Loss: 0.7492749094963074:  85%|████████▍ | 5488/6470 [1:08:16<13:24,  1.22it/s]
4|5489|Loss: 0.7492749094963074:  85%|████████▍ | 5489/6470 [1:08:16<12:42,  1.29it/s]
4|5490|Loss: 0.8230088949203491:  85%|████████▍ | 5489/6470 [1:08:17<12:42,  1.29it/s]
4|5490|Loss: 0.8230088949203491:  85%|████████▍ | 5490/6470 [1:08:17<12:16,  1.33it/s]
4|5491|Loss: 0.8276323676109314:  85%|████████▍ | 5490/6470 [1:08:17<12:16,  1.33it/s]
4|5491|Loss: 0.8276323676109314:  85%|████████▍ | 5491/6470 [1:08:18<12:22,  1.32it/s]
4|5492|Loss: 0.817152202129364:  85%|████████▍ | 5491/6470 [1:08:18<12:22,  1.32it/s] 
4|5492|Loss: 0.817152202129364:  85%|████████▍ | 5492/6470 [1:08:18<12:02,  1.35it/s]
4|5493|Loss: 0.6482400298118591:  85%|████████▍ | 5492/6470 [1:08:19<12:02,  1.35it/s]
4|5493|Loss: 0.6482400298118591:  85%|████████▍ | 5493/6470 [1:08:19<12:05,  1.35it/s]
4|5494|Loss: 0.8777981996536255:  85%|████████▍ | 5493/6470 [1:08:20<12:05,  1.35it/s]
4|5494|Loss: 0.8777981996536255:  85%|████████▍ | 5494/6470 [1:08:20<11:56,  1.36it/s]
4|5495|Loss: 0.9277936220169067:  85%|████████▍ | 5494/6470 [1:08:20<11:56,  1.36it/s]
4|5495|Loss: 0.9277936220169067:  85%|████████▍ | 5495/6470 [1:08:21<12:16,  1.32it/s]
4|5496|Loss: 0.8007994890213013:  85%|████████▍ | 5495/6470 [1:08:21<12:16,  1.32it/s]
4|5496|Loss: 0.8007994890213013:  85%|████████▍ | 5496/6470 [1:08:22<12:55,  1.26it/s]
4|5497|Loss: 0.7466092109680176:  85%|████████▍ | 5496/6470 [1:08:22<12:55,  1.26it/s]
4|5497|Loss: 0.7466092109680176:  85%|████████▍ | 5497/6470 [1:08:22<13:02,  1.24it/s]
4|5498|Loss: 0.8175876140594482:  85%|████████▍ | 5497/6470 [1:08:23<13:02,  1.24it/s]
4|5498|Loss: 0.8175876140594482:  85%|████████▍ | 5498/6470 [1:08:23<12:41,  1.28it/s]
4|5499|Loss: 0.9885677695274353:  85%|████████▍ | 5498/6470 [1:08:24<12:41,  1.28it/s]
4|5499|Loss: 0.9885677695274353:  85%|████████▍ | 5499/6470 [1:08:24<12:38,  1.28it/s]
4|5500|Loss: 0.7081378698348999:  85%|████████▍ | 5499/6470 [1:08:24<12:38,  1.28it/s]
4|5500|Loss: 0.7081378698348999:  85%|████████▌ | 5500/6470 [1:08:25<12:41,  1.27it/s]
4|5501|Loss: 0.7766029834747314:  85%|████████▌ | 5500/6470 [1:08:25<12:41,  1.27it/s]
4|5501|Loss: 0.7766029834747314:  85%|████████▌ | 5501/6470 [1:08:25<12:06,  1.33it/s]
4|5502|Loss: 0.7293947339057922:  85%|████████▌ | 5501/6470 [1:08:26<12:06,  1.33it/s]
4|5502|Loss: 0.7293947339057922:  85%|████████▌ | 5502/6470 [1:08:26<11:09,  1.45it/s]
4|5503|Loss: 0.7698067426681519:  85%|████████▌ | 5502/6470 [1:08:26<11:09,  1.45it/s]
4|5503|Loss: 0.7698067426681519:  85%|████████▌ | 5503/6470 [1:08:27<11:24,  1.41it/s]
4|5504|Loss: 0.9511520266532898:  85%|████████▌ | 5503/6470 [1:08:27<11:24,  1.41it/s]
4|5504|Loss: 0.9511520266532898:  85%|████████▌ | 5504/6470 [1:08:27<11:07,  1.45it/s]
4|5505|Loss: 0.8219475746154785:  85%|████████▌ | 5504/6470 [1:08:28<11:07,  1.45it/s]
4|5505|Loss: 0.8219475746154785:  85%|████████▌ | 5505/6470 [1:08:28<11:03,  1.45it/s]
4|5506|Loss: 0.7027808427810669:  85%|████████▌ | 5505/6470 [1:08:28<11:03,  1.45it/s]
4|5506|Loss: 0.7027808427810669:  85%|████████▌ | 5506/6470 [1:08:29<11:21,  1.41it/s]
4|5507|Loss: 0.7730477452278137:  85%|████████▌ | 5506/6470 [1:08:29<11:21,  1.41it/s]
4|5507|Loss: 0.7730477452278137:  85%|████████▌ | 5507/6470 [1:08:30<11:37,  1.38it/s]
4|5508|Loss: 0.8181992173194885:  85%|████████▌ | 5507/6470 [1:08:30<11:37,  1.38it/s]
4|5508|Loss: 0.8181992173194885:  85%|████████▌ | 5508/6470 [1:08:30<10:58,  1.46it/s]
4|5509|Loss: 0.7494764924049377:  85%|████████▌ | 5508/6470 [1:08:30<10:58,  1.46it/s]
4|5509|Loss: 0.7494764924049377:  85%|████████▌ | 5509/6470 [1:08:31<10:26,  1.53it/s]
4|5510|Loss: 0.8688252568244934:  85%|████████▌ | 5509/6470 [1:08:31<10:26,  1.53it/s]
4|5510|Loss: 0.8688252568244934:  85%|████████▌ | 5510/6470 [1:08:31<10:31,  1.52it/s]
4|5511|Loss: 0.7570645809173584:  85%|████████▌ | 5510/6470 [1:08:32<10:31,  1.52it/s]
4|5511|Loss: 0.7570645809173584:  85%|████████▌ | 5511/6470 [1:08:32<10:28,  1.53it/s]
4|5512|Loss: 0.9298638701438904:  85%|████████▌ | 5511/6470 [1:08:32<10:28,  1.53it/s]
4|5512|Loss: 0.9298638701438904:  85%|████████▌ | 5512/6470 [1:08:33<11:29,  1.39it/s]
4|5513|Loss: 0.8630444407463074:  85%|████████▌ | 5512/6470 [1:08:33<11:29,  1.39it/s]
4|5513|Loss: 0.8630444407463074:  85%|████████▌ | 5513/6470 [1:08:33<10:48,  1.48it/s]
4|5514|Loss: 0.8342378735542297:  85%|████████▌ | 5513/6470 [1:08:34<10:48,  1.48it/s]
4|5514|Loss: 0.8342378735542297:  85%|████████▌ | 5514/6470 [1:08:34<10:37,  1.50it/s]
4|5515|Loss: 0.6296966671943665:  85%|████████▌ | 5514/6470 [1:08:35<10:37,  1.50it/s]
4|5515|Loss: 0.6296966671943665:  85%|████████▌ | 5515/6470 [1:08:35<12:10,  1.31it/s]
4|5516|Loss: 0.9290587902069092:  85%|████████▌ | 5515/6470 [1:08:36<12:10,  1.31it/s]
4|5516|Loss: 0.9290587902069092:  85%|████████▌ | 5516/6470 [1:08:36<12:35,  1.26it/s]
4|5517|Loss: 0.825233519077301:  85%|████████▌ | 5516/6470 [1:08:36<12:35,  1.26it/s] 
4|5517|Loss: 0.825233519077301:  85%|████████▌ | 5517/6470 [1:08:37<13:25,  1.18it/s]
4|5518|Loss: 0.9424281120300293:  85%|████████▌ | 5517/6470 [1:08:37<13:25,  1.18it/s]
4|5518|Loss: 0.9424281120300293:  85%|████████▌ | 5518/6470 [1:08:38<13:39,  1.16it/s]
4|5519|Loss: 0.7741697430610657:  85%|████████▌ | 5518/6470 [1:08:38<13:39,  1.16it/s]
4|5519|Loss: 0.7741697430610657:  85%|████████▌ | 5519/6470 [1:08:38<12:28,  1.27it/s]
4|5520|Loss: 0.8001919388771057:  85%|████████▌ | 5519/6470 [1:08:39<12:28,  1.27it/s]
4|5520|Loss: 0.8001919388771057:  85%|████████▌ | 5520/6470 [1:08:39<11:42,  1.35it/s]
4|5521|Loss: 0.7578048706054688:  85%|████████▌ | 5520/6470 [1:08:40<11:42,  1.35it/s]
4|5521|Loss: 0.7578048706054688:  85%|████████▌ | 5521/6470 [1:08:40<11:52,  1.33it/s]
4|5522|Loss: 0.901111900806427:  85%|████████▌ | 5521/6470 [1:08:40<11:52,  1.33it/s] 
4|5522|Loss: 0.901111900806427:  85%|████████▌ | 5522/6470 [1:08:41<12:00,  1.32it/s]
4|5523|Loss: 0.6978499889373779:  85%|████████▌ | 5522/6470 [1:08:41<12:00,  1.32it/s]
4|5523|Loss: 0.6978499889373779:  85%|████████▌ | 5523/6470 [1:08:41<11:16,  1.40it/s]
4|5524|Loss: 0.8327435851097107:  85%|████████▌ | 5523/6470 [1:08:42<11:16,  1.40it/s]
4|5524|Loss: 0.8327435851097107:  85%|████████▌ | 5524/6470 [1:08:42<10:26,  1.51it/s]
4|5525|Loss: 0.8204407095909119:  85%|████████▌ | 5524/6470 [1:08:42<10:26,  1.51it/s]
4|5525|Loss: 0.8204407095909119:  85%|████████▌ | 5525/6470 [1:08:42<09:53,  1.59it/s]
4|5526|Loss: 0.7458819150924683:  85%|████████▌ | 5525/6470 [1:08:43<09:53,  1.59it/s]
4|5526|Loss: 0.7458819150924683:  85%|████████▌ | 5526/6470 [1:08:43<10:42,  1.47it/s]
4|5527|Loss: 0.9173508882522583:  85%|████████▌ | 5526/6470 [1:08:44<10:42,  1.47it/s]
4|5527|Loss: 0.9173508882522583:  85%|████████▌ | 5527/6470 [1:08:44<11:36,  1.35it/s]
4|5528|Loss: 0.9422471523284912:  85%|████████▌ | 5527/6470 [1:08:44<11:36,  1.35it/s]
4|5528|Loss: 0.9422471523284912:  85%|████████▌ | 5528/6470 [1:08:45<11:55,  1.32it/s]
4|5529|Loss: 0.809782862663269:  85%|████████▌ | 5528/6470 [1:08:45<11:55,  1.32it/s] 
4|5529|Loss: 0.809782862663269:  85%|████████▌ | 5529/6470 [1:08:46<12:19,  1.27it/s]
4|5530|Loss: 0.9574514031410217:  85%|████████▌ | 5529/6470 [1:08:46<12:19,  1.27it/s]
4|5530|Loss: 0.9574514031410217:  85%|████████▌ | 5530/6470 [1:08:47<13:04,  1.20it/s]
4|5531|Loss: 0.8150813579559326:  85%|████████▌ | 5530/6470 [1:08:47<13:04,  1.20it/s]
4|5531|Loss: 0.8150813579559326:  85%|████████▌ | 5531/6470 [1:08:48<13:52,  1.13it/s]
4|5532|Loss: 0.8981515169143677:  85%|████████▌ | 5531/6470 [1:08:48<13:52,  1.13it/s]
4|5532|Loss: 0.8981515169143677:  86%|████████▌ | 5532/6470 [1:08:48<13:11,  1.19it/s]
4|5533|Loss: 0.7556027770042419:  86%|████████▌ | 5532/6470 [1:08:49<13:11,  1.19it/s]
4|5533|Loss: 0.7556027770042419:  86%|████████▌ | 5533/6470 [1:08:49<12:50,  1.22it/s]
4|5534|Loss: 0.7836810350418091:  86%|████████▌ | 5533/6470 [1:08:50<12:50,  1.22it/s]
4|5534|Loss: 0.7836810350418091:  86%|████████▌ | 5534/6470 [1:08:50<13:05,  1.19it/s]
4|5535|Loss: 0.8365108370780945:  86%|████████▌ | 5534/6470 [1:08:50<13:05,  1.19it/s]
4|5535|Loss: 0.8365108370780945:  86%|████████▌ | 5535/6470 [1:08:51<12:48,  1.22it/s]
4|5536|Loss: 0.7074583768844604:  86%|████████▌ | 5535/6470 [1:08:51<12:48,  1.22it/s]
4|5536|Loss: 0.7074583768844604:  86%|████████▌ | 5536/6470 [1:08:51<10:59,  1.42it/s]
4|5537|Loss: 0.8086240291595459:  86%|████████▌ | 5536/6470 [1:08:52<10:59,  1.42it/s]
4|5537|Loss: 0.8086240291595459:  86%|████████▌ | 5537/6470 [1:08:52<10:02,  1.55it/s]
4|5538|Loss: 0.8002914786338806:  86%|████████▌ | 5537/6470 [1:08:52<10:02,  1.55it/s]
4|5538|Loss: 0.8002914786338806:  86%|████████▌ | 5538/6470 [1:08:53<10:59,  1.41it/s]
4|5539|Loss: 0.7651305794715881:  86%|████████▌ | 5538/6470 [1:08:53<10:59,  1.41it/s]
4|5539|Loss: 0.7651305794715881:  86%|████████▌ | 5539/6470 [1:08:54<11:59,  1.29it/s]
4|5540|Loss: 0.7103407979011536:  86%|████████▌ | 5539/6470 [1:08:54<11:59,  1.29it/s]
4|5540|Loss: 0.7103407979011536:  86%|████████▌ | 5540/6470 [1:08:54<11:38,  1.33it/s]
4|5541|Loss: 0.8043514490127563:  86%|████████▌ | 5540/6470 [1:08:55<11:38,  1.33it/s]
4|5541|Loss: 0.8043514490127563:  86%|████████▌ | 5541/6470 [1:08:55<11:32,  1.34it/s]
4|5542|Loss: 0.7467239499092102:  86%|████████▌ | 5541/6470 [1:08:55<11:32,  1.34it/s]
4|5542|Loss: 0.7467239499092102:  86%|████████▌ | 5542/6470 [1:08:56<11:06,  1.39it/s]
4|5543|Loss: 0.8251529335975647:  86%|████████▌ | 5542/6470 [1:08:56<11:06,  1.39it/s]
4|5543|Loss: 0.8251529335975647:  86%|████████▌ | 5543/6470 [1:08:56<11:21,  1.36it/s]
4|5544|Loss: 0.9266731142997742:  86%|████████▌ | 5543/6470 [1:08:57<11:21,  1.36it/s]
4|5544|Loss: 0.9266731142997742:  86%|████████▌ | 5544/6470 [1:08:57<11:39,  1.32it/s]
4|5545|Loss: 0.6566008925437927:  86%|████████▌ | 5544/6470 [1:08:58<11:39,  1.32it/s]
4|5545|Loss: 0.6566008925437927:  86%|████████▌ | 5545/6470 [1:08:58<12:04,  1.28it/s]
4|5546|Loss: 0.7919100522994995:  86%|████████▌ | 5545/6470 [1:08:58<12:04,  1.28it/s]
4|5546|Loss: 0.7919100522994995:  86%|████████▌ | 5546/6470 [1:08:59<12:12,  1.26it/s]
4|5547|Loss: 0.8258674144744873:  86%|████████▌ | 5546/6470 [1:08:59<12:12,  1.26it/s]
4|5547|Loss: 0.8258674144744873:  86%|████████▌ | 5547/6470 [1:08:59<11:02,  1.39it/s]
4|5548|Loss: 0.7510812878608704:  86%|████████▌ | 5547/6470 [1:09:00<11:02,  1.39it/s]
4|5548|Loss: 0.7510812878608704:  86%|████████▌ | 5548/6470 [1:09:00<09:53,  1.55it/s]
4|5549|Loss: 0.8333283066749573:  86%|████████▌ | 5548/6470 [1:09:00<09:53,  1.55it/s]
4|5549|Loss: 0.8333283066749573:  86%|████████▌ | 5549/6470 [1:09:00<08:49,  1.74it/s]
4|5550|Loss: 0.8131709694862366:  86%|████████▌ | 5549/6470 [1:09:01<08:49,  1.74it/s]
4|5550|Loss: 0.8131709694862366:  86%|████████▌ | 5550/6470 [1:09:01<09:53,  1.55it/s]
4|5551|Loss: 0.9197149276733398:  86%|████████▌ | 5550/6470 [1:09:02<09:53,  1.55it/s]
4|5551|Loss: 0.9197149276733398:  86%|████████▌ | 5551/6470 [1:09:02<10:19,  1.48it/s]
4|5552|Loss: 0.8213657140731812:  86%|████████▌ | 5551/6470 [1:09:02<10:19,  1.48it/s]
4|5552|Loss: 0.8213657140731812:  86%|████████▌ | 5552/6470 [1:09:02<10:07,  1.51it/s]
4|5553|Loss: 0.7732376456260681:  86%|████████▌ | 5552/6470 [1:09:03<10:07,  1.51it/s]
4|5553|Loss: 0.7732376456260681:  86%|████████▌ | 5553/6470 [1:09:03<10:30,  1.45it/s]
4|5554|Loss: 0.8138972520828247:  86%|████████▌ | 5553/6470 [1:09:04<10:30,  1.45it/s]
4|5554|Loss: 0.8138972520828247:  86%|████████▌ | 5554/6470 [1:09:04<10:37,  1.44it/s]
4|5555|Loss: 1.1561870574951172:  86%|████████▌ | 5554/6470 [1:09:04<10:37,  1.44it/s]
4|5555|Loss: 1.1561870574951172:  86%|████████▌ | 5555/6470 [1:09:05<12:28,  1.22it/s]
4|5556|Loss: 1.0373785495758057:  86%|████████▌ | 5555/6470 [1:09:06<12:28,  1.22it/s]
4|5556|Loss: 1.0373785495758057:  86%|████████▌ | 5556/6470 [1:09:06<12:18,  1.24it/s]
4|5557|Loss: 0.6878514885902405:  86%|████████▌ | 5556/6470 [1:09:06<12:18,  1.24it/s]
4|5557|Loss: 0.6878514885902405:  86%|████████▌ | 5557/6470 [1:09:06<11:10,  1.36it/s]
4|5558|Loss: 0.8987414836883545:  86%|████████▌ | 5557/6470 [1:09:07<11:10,  1.36it/s]
4|5558|Loss: 0.8987414836883545:  86%|████████▌ | 5558/6470 [1:09:07<11:10,  1.36it/s]
4|5559|Loss: 0.8439139127731323:  86%|████████▌ | 5558/6470 [1:09:08<11:10,  1.36it/s]
4|5559|Loss: 0.8439139127731323:  86%|████████▌ | 5559/6470 [1:09:08<10:58,  1.38it/s]
4|5560|Loss: 0.8118802905082703:  86%|████████▌ | 5559/6470 [1:09:08<10:58,  1.38it/s]
4|5560|Loss: 0.8118802905082703:  86%|████████▌ | 5560/6470 [1:09:08<10:40,  1.42it/s]
4|5561|Loss: 0.8201304078102112:  86%|████████▌ | 5560/6470 [1:09:09<10:40,  1.42it/s]
4|5561|Loss: 0.8201304078102112:  86%|████████▌ | 5561/6470 [1:09:09<11:19,  1.34it/s]
4|5562|Loss: 0.7320410013198853:  86%|████████▌ | 5561/6470 [1:09:10<11:19,  1.34it/s]
4|5562|Loss: 0.7320410013198853:  86%|████████▌ | 5562/6470 [1:09:10<11:25,  1.33it/s]
4|5563|Loss: 0.8033828139305115:  86%|████████▌ | 5562/6470 [1:09:11<11:25,  1.33it/s]
4|5563|Loss: 0.8033828139305115:  86%|████████▌ | 5563/6470 [1:09:11<11:36,  1.30it/s]
4|5564|Loss: 0.8417249321937561:  86%|████████▌ | 5563/6470 [1:09:11<11:36,  1.30it/s]
4|5564|Loss: 0.8417249321937561:  86%|████████▌ | 5564/6470 [1:09:12<11:46,  1.28it/s]
4|5565|Loss: 0.8306196331977844:  86%|████████▌ | 5564/6470 [1:09:12<11:46,  1.28it/s]
4|5565|Loss: 0.8306196331977844:  86%|████████▌ | 5565/6470 [1:09:12<10:59,  1.37it/s]
4|5566|Loss: 0.9081991314888:  86%|████████▌ | 5565/6470 [1:09:13<10:59,  1.37it/s]   
4|5566|Loss: 0.9081991314888:  86%|████████▌ | 5566/6470 [1:09:13<11:25,  1.32it/s]
4|5567|Loss: 0.938689112663269:  86%|████████▌ | 5566/6470 [1:09:14<11:25,  1.32it/s]
4|5567|Loss: 0.938689112663269:  86%|████████▌ | 5567/6470 [1:09:14<11:28,  1.31it/s]
4|5568|Loss: 0.7840356826782227:  86%|████████▌ | 5567/6470 [1:09:14<11:28,  1.31it/s]
4|5568|Loss: 0.7840356826782227:  86%|████████▌ | 5568/6470 [1:09:15<11:35,  1.30it/s]
4|5569|Loss: 0.7985441088676453:  86%|████████▌ | 5568/6470 [1:09:15<11:35,  1.30it/s]
4|5569|Loss: 0.7985441088676453:  86%|████████▌ | 5569/6470 [1:09:15<11:25,  1.31it/s]
4|5570|Loss: 0.7668164372444153:  86%|████████▌ | 5569/6470 [1:09:16<11:25,  1.31it/s]
4|5570|Loss: 0.7668164372444153:  86%|████████▌ | 5570/6470 [1:09:16<11:28,  1.31it/s]
4|5571|Loss: 0.7206723690032959:  86%|████████▌ | 5570/6470 [1:09:17<11:28,  1.31it/s]
4|5571|Loss: 0.7206723690032959:  86%|████████▌ | 5571/6470 [1:09:17<11:26,  1.31it/s]
4|5572|Loss: 0.7451475262641907:  86%|████████▌ | 5571/6470 [1:09:17<11:26,  1.31it/s]
4|5572|Loss: 0.7451475262641907:  86%|████████▌ | 5572/6470 [1:09:18<11:29,  1.30it/s]
4|5573|Loss: 0.7919204235076904:  86%|████████▌ | 5572/6470 [1:09:18<11:29,  1.30it/s]
4|5573|Loss: 0.7919204235076904:  86%|████████▌ | 5573/6470 [1:09:19<11:58,  1.25it/s]
4|5574|Loss: 0.8299907445907593:  86%|████████▌ | 5573/6470 [1:09:19<11:58,  1.25it/s]
4|5574|Loss: 0.8299907445907593:  86%|████████▌ | 5574/6470 [1:09:19<11:38,  1.28it/s]
4|5575|Loss: 0.8370535969734192:  86%|████████▌ | 5574/6470 [1:09:20<11:38,  1.28it/s]
4|5575|Loss: 0.8370535969734192:  86%|████████▌ | 5575/6470 [1:09:20<12:11,  1.22it/s]
4|5576|Loss: 0.7430035471916199:  86%|████████▌ | 5575/6470 [1:09:21<12:11,  1.22it/s]
4|5576|Loss: 0.7430035471916199:  86%|████████▌ | 5576/6470 [1:09:21<11:08,  1.34it/s]
4|5577|Loss: 0.7280005812644958:  86%|████████▌ | 5576/6470 [1:09:21<11:08,  1.34it/s]
4|5577|Loss: 0.7280005812644958:  86%|████████▌ | 5577/6470 [1:09:22<11:05,  1.34it/s]
4|5578|Loss: 1.00261652469635:  86%|████████▌ | 5577/6470 [1:09:22<11:05,  1.34it/s]  
4|5578|Loss: 1.00261652469635:  86%|████████▌ | 5578/6470 [1:09:22<10:16,  1.45it/s]
4|5579|Loss: 0.8262248039245605:  86%|████████▌ | 5578/6470 [1:09:23<10:16,  1.45it/s]
4|5579|Loss: 0.8262248039245605:  86%|████████▌ | 5579/6470 [1:09:23<10:48,  1.37it/s]
4|5580|Loss: 0.9664155840873718:  86%|████████▌ | 5579/6470 [1:09:23<10:48,  1.37it/s]
4|5580|Loss: 0.9664155840873718:  86%|████████▌ | 5580/6470 [1:09:24<11:07,  1.33it/s]
4|5581|Loss: 0.6672475934028625:  86%|████████▌ | 5580/6470 [1:09:24<11:07,  1.33it/s]
4|5581|Loss: 0.6672475934028625:  86%|████████▋ | 5581/6470 [1:09:25<11:18,  1.31it/s]
4|5582|Loss: 0.9879061579704285:  86%|████████▋ | 5581/6470 [1:09:25<11:18,  1.31it/s]
4|5582|Loss: 0.9879061579704285:  86%|████████▋ | 5582/6470 [1:09:25<11:16,  1.31it/s]
4|5583|Loss: 0.9492776989936829:  86%|████████▋ | 5582/6470 [1:09:26<11:16,  1.31it/s]
4|5583|Loss: 0.9492776989936829:  86%|████████▋ | 5583/6470 [1:09:26<10:54,  1.35it/s]
4|5584|Loss: 0.7614049315452576:  86%|████████▋ | 5583/6470 [1:09:26<10:54,  1.35it/s]
4|5584|Loss: 0.7614049315452576:  86%|████████▋ | 5584/6470 [1:09:27<11:02,  1.34it/s]
4|5585|Loss: 0.9170103669166565:  86%|████████▋ | 5584/6470 [1:09:27<11:02,  1.34it/s]
4|5585|Loss: 0.9170103669166565:  86%|████████▋ | 5585/6470 [1:09:27<10:25,  1.41it/s]
4|5586|Loss: 0.7725529074668884:  86%|████████▋ | 5585/6470 [1:09:28<10:25,  1.41it/s]
4|5586|Loss: 0.7725529074668884:  86%|████████▋ | 5586/6470 [1:09:28<09:59,  1.47it/s]
4|5587|Loss: 0.8082982301712036:  86%|████████▋ | 5586/6470 [1:09:28<09:59,  1.47it/s]
4|5587|Loss: 0.8082982301712036:  86%|████████▋ | 5587/6470 [1:09:29<10:45,  1.37it/s]
4|5588|Loss: 0.8367928862571716:  86%|████████▋ | 5587/6470 [1:09:29<10:45,  1.37it/s]
4|5588|Loss: 0.8367928862571716:  86%|████████▋ | 5588/6470 [1:09:29<10:00,  1.47it/s]
4|5589|Loss: 0.7914624214172363:  86%|████████▋ | 5588/6470 [1:09:30<10:00,  1.47it/s]
4|5589|Loss: 0.7914624214172363:  86%|████████▋ | 5589/6470 [1:09:30<10:10,  1.44it/s]
4|5590|Loss: 0.9417619705200195:  86%|████████▋ | 5589/6470 [1:09:31<10:10,  1.44it/s]
4|5590|Loss: 0.9417619705200195:  86%|████████▋ | 5590/6470 [1:09:31<10:16,  1.43it/s]
4|5591|Loss: 0.6702438592910767:  86%|████████▋ | 5590/6470 [1:09:31<10:16,  1.43it/s]
4|5591|Loss: 0.6702438592910767:  86%|████████▋ | 5591/6470 [1:09:32<10:50,  1.35it/s]
4|5592|Loss: 0.9285740852355957:  86%|████████▋ | 5591/6470 [1:09:32<10:50,  1.35it/s]
4|5592|Loss: 0.9285740852355957:  86%|████████▋ | 5592/6470 [1:09:33<11:10,  1.31it/s]
4|5593|Loss: 0.8020941615104675:  86%|████████▋ | 5592/6470 [1:09:33<11:10,  1.31it/s]
4|5593|Loss: 0.8020941615104675:  86%|████████▋ | 5593/6470 [1:09:33<10:59,  1.33it/s]
4|5594|Loss: 0.7414629459381104:  86%|████████▋ | 5593/6470 [1:09:34<10:59,  1.33it/s]
4|5594|Loss: 0.7414629459381104:  86%|████████▋ | 5594/6470 [1:09:34<10:59,  1.33it/s]
4|5595|Loss: 0.8074913620948792:  86%|████████▋ | 5594/6470 [1:09:34<10:59,  1.33it/s]
4|5595|Loss: 0.8074913620948792:  86%|████████▋ | 5595/6470 [1:09:35<11:11,  1.30it/s]
4|5596|Loss: 1.0289428234100342:  86%|████████▋ | 5595/6470 [1:09:35<11:11,  1.30it/s]
4|5596|Loss: 1.0289428234100342:  86%|████████▋ | 5596/6470 [1:09:36<11:55,  1.22it/s]
4|5597|Loss: 0.7328173518180847:  86%|████████▋ | 5596/6470 [1:09:36<11:55,  1.22it/s]
4|5597|Loss: 0.7328173518180847:  87%|████████▋ | 5597/6470 [1:09:36<11:33,  1.26it/s]
4|5598|Loss: 0.7126732468605042:  87%|████████▋ | 5597/6470 [1:09:37<11:33,  1.26it/s]
4|5598|Loss: 0.7126732468605042:  87%|████████▋ | 5598/6470 [1:09:37<11:39,  1.25it/s]
4|5599|Loss: 0.7871361970901489:  87%|████████▋ | 5598/6470 [1:09:38<11:39,  1.25it/s]
4|5599|Loss: 0.7871361970901489:  87%|████████▋ | 5599/6470 [1:09:38<11:19,  1.28it/s]
4|5600|Loss: 0.6624464392662048:  87%|████████▋ | 5599/6470 [1:09:38<11:19,  1.28it/s]
4|5600|Loss: 0.6624464392662048:  87%|████████▋ | 5600/6470 [1:09:39<11:49,  1.23it/s]
4|5601|Loss: 0.7505031228065491:  87%|████████▋ | 5600/6470 [1:09:39<11:49,  1.23it/s]
4|5601|Loss: 0.7505031228065491:  87%|████████▋ | 5601/6470 [1:09:40<12:25,  1.16it/s]
4|5602|Loss: 0.8250235319137573:  87%|████████▋ | 5601/6470 [1:09:40<12:25,  1.16it/s]
4|5602|Loss: 0.8250235319137573:  87%|████████▋ | 5602/6470 [1:09:41<12:17,  1.18it/s]
4|5603|Loss: 0.8534958362579346:  87%|████████▋ | 5602/6470 [1:09:41<12:17,  1.18it/s]
4|5603|Loss: 0.8534958362579346:  87%|████████▋ | 5603/6470 [1:09:41<11:35,  1.25it/s]
4|5604|Loss: 0.880100667476654:  87%|████████▋ | 5603/6470 [1:09:42<11:35,  1.25it/s] 
4|5604|Loss: 0.880100667476654:  87%|████████▋ | 5604/6470 [1:09:42<10:59,  1.31it/s]
4|5605|Loss: 0.9530209302902222:  87%|████████▋ | 5604/6470 [1:09:43<10:59,  1.31it/s]
4|5605|Loss: 0.9530209302902222:  87%|████████▋ | 5605/6470 [1:09:43<11:23,  1.27it/s]
4|5606|Loss: 0.9780732989311218:  87%|████████▋ | 5605/6470 [1:09:43<11:23,  1.27it/s]
4|5606|Loss: 0.9780732989311218:  87%|████████▋ | 5606/6470 [1:09:44<12:03,  1.19it/s]
4|5607|Loss: 0.8328631520271301:  87%|████████▋ | 5606/6470 [1:09:44<12:03,  1.19it/s]
4|5607|Loss: 0.8328631520271301:  87%|████████▋ | 5607/6470 [1:09:45<11:36,  1.24it/s]
4|5608|Loss: 0.7302870750427246:  87%|████████▋ | 5607/6470 [1:09:45<11:36,  1.24it/s]
4|5608|Loss: 0.7302870750427246:  87%|████████▋ | 5608/6470 [1:09:45<10:51,  1.32it/s]
4|5609|Loss: 0.8361730575561523:  87%|████████▋ | 5608/6470 [1:09:46<10:51,  1.32it/s]
4|5609|Loss: 0.8361730575561523:  87%|████████▋ | 5609/6470 [1:09:46<10:24,  1.38it/s]
4|5610|Loss: 0.7920737862586975:  87%|████████▋ | 5609/6470 [1:09:46<10:24,  1.38it/s]
4|5610|Loss: 0.7920737862586975:  87%|████████▋ | 5610/6470 [1:09:46<09:09,  1.56it/s]
4|5611|Loss: 0.7429050207138062:  87%|████████▋ | 5610/6470 [1:09:47<09:09,  1.56it/s]
4|5611|Loss: 0.7429050207138062:  87%|████████▋ | 5611/6470 [1:09:47<09:32,  1.50it/s]
4|5612|Loss: 0.8016541600227356:  87%|████████▋ | 5611/6470 [1:09:48<09:32,  1.50it/s]
4|5612|Loss: 0.8016541600227356:  87%|████████▋ | 5612/6470 [1:09:48<10:26,  1.37it/s]
4|5613|Loss: 0.7371032238006592:  87%|████████▋ | 5612/6470 [1:09:48<10:26,  1.37it/s]
4|5613|Loss: 0.7371032238006592:  87%|████████▋ | 5613/6470 [1:09:49<10:52,  1.31it/s]
4|5614|Loss: 0.7103366255760193:  87%|████████▋ | 5613/6470 [1:09:49<10:52,  1.31it/s]
4|5614|Loss: 0.7103366255760193:  87%|████████▋ | 5614/6470 [1:09:49<10:36,  1.34it/s]
4|5615|Loss: 0.9471204876899719:  87%|████████▋ | 5614/6470 [1:09:50<10:36,  1.34it/s]
4|5615|Loss: 0.9471204876899719:  87%|████████▋ | 5615/6470 [1:09:50<10:16,  1.39it/s]
4|5616|Loss: 0.7718650102615356:  87%|████████▋ | 5615/6470 [1:09:50<10:16,  1.39it/s]
4|5616|Loss: 0.7718650102615356:  87%|████████▋ | 5616/6470 [1:09:51<09:35,  1.48it/s]
4|5617|Loss: 0.7912624478340149:  87%|████████▋ | 5616/6470 [1:09:51<09:35,  1.48it/s]
4|5617|Loss: 0.7912624478340149:  87%|████████▋ | 5617/6470 [1:09:51<09:08,  1.55it/s]
4|5618|Loss: 0.865739643573761:  87%|████████▋ | 5617/6470 [1:09:52<09:08,  1.55it/s] 
4|5618|Loss: 0.865739643573761:  87%|████████▋ | 5618/6470 [1:09:52<10:18,  1.38it/s]
4|5619|Loss: 0.9313384294509888:  87%|████████▋ | 5618/6470 [1:09:53<10:18,  1.38it/s]
4|5619|Loss: 0.9313384294509888:  87%|████████▋ | 5619/6470 [1:09:53<10:39,  1.33it/s]
4|5620|Loss: 0.8287070989608765:  87%|████████▋ | 5619/6470 [1:09:53<10:39,  1.33it/s]
4|5620|Loss: 0.8287070989608765:  87%|████████▋ | 5620/6470 [1:09:54<10:41,  1.33it/s]
4|5621|Loss: 0.7815518379211426:  87%|████████▋ | 5620/6470 [1:09:54<10:41,  1.33it/s]
4|5621|Loss: 0.7815518379211426:  87%|████████▋ | 5621/6470 [1:09:55<10:50,  1.31it/s]
4|5622|Loss: 0.8362544775009155:  87%|████████▋ | 5621/6470 [1:09:55<10:50,  1.31it/s]
4|5622|Loss: 0.8362544775009155:  87%|████████▋ | 5622/6470 [1:09:55<10:21,  1.36it/s]
4|5623|Loss: 0.8809518218040466:  87%|████████▋ | 5622/6470 [1:09:56<10:21,  1.36it/s]
4|5623|Loss: 0.8809518218040466:  87%|████████▋ | 5623/6470 [1:09:56<10:56,  1.29it/s]
4|5624|Loss: 0.8741554617881775:  87%|████████▋ | 5623/6470 [1:09:57<10:56,  1.29it/s]
4|5624|Loss: 0.8741554617881775:  87%|████████▋ | 5624/6470 [1:09:57<11:13,  1.26it/s]
4|5625|Loss: 0.7379553914070129:  87%|████████▋ | 5624/6470 [1:09:57<11:13,  1.26it/s]
4|5625|Loss: 0.7379553914070129:  87%|████████▋ | 5625/6470 [1:09:58<11:07,  1.27it/s]
4|5626|Loss: 0.8645372986793518:  87%|████████▋ | 5625/6470 [1:09:58<11:07,  1.27it/s]
4|5626|Loss: 0.8645372986793518:  87%|████████▋ | 5626/6470 [1:09:59<11:50,  1.19it/s]
4|5627|Loss: 0.9613920450210571:  87%|████████▋ | 5626/6470 [1:09:59<11:50,  1.19it/s]
4|5627|Loss: 0.9613920450210571:  87%|████████▋ | 5627/6470 [1:09:59<11:30,  1.22it/s]
4|5628|Loss: 0.8663823008537292:  87%|████████▋ | 5627/6470 [1:10:00<11:30,  1.22it/s]
4|5628|Loss: 0.8663823008537292:  87%|████████▋ | 5628/6470 [1:10:00<11:02,  1.27it/s]
4|5629|Loss: 0.7516841292381287:  87%|████████▋ | 5628/6470 [1:10:01<11:02,  1.27it/s]
4|5629|Loss: 0.7516841292381287:  87%|████████▋ | 5629/6470 [1:10:01<10:17,  1.36it/s]
4|5630|Loss: 1.0765864849090576:  87%|████████▋ | 5629/6470 [1:10:01<10:17,  1.36it/s]
4|5630|Loss: 1.0765864849090576:  87%|████████▋ | 5630/6470 [1:10:02<10:48,  1.30it/s]
4|5631|Loss: 0.6799571514129639:  87%|████████▋ | 5630/6470 [1:10:02<10:48,  1.30it/s]
4|5631|Loss: 0.6799571514129639:  87%|████████▋ | 5631/6470 [1:10:02<10:12,  1.37it/s]
4|5632|Loss: 0.836632251739502:  87%|████████▋ | 5631/6470 [1:10:03<10:12,  1.37it/s] 
4|5632|Loss: 0.836632251739502:  87%|████████▋ | 5632/6470 [1:10:03<09:57,  1.40it/s]
4|5633|Loss: 0.7639469504356384:  87%|████████▋ | 5632/6470 [1:10:03<09:57,  1.40it/s]
4|5633|Loss: 0.7639469504356384:  87%|████████▋ | 5633/6470 [1:10:04<09:38,  1.45it/s]
4|5634|Loss: 0.7605087161064148:  87%|████████▋ | 5633/6470 [1:10:04<09:38,  1.45it/s]
4|5634|Loss: 0.7605087161064148:  87%|████████▋ | 5634/6470 [1:10:04<09:17,  1.50it/s]
4|5635|Loss: 0.8288517594337463:  87%|████████▋ | 5634/6470 [1:10:05<09:17,  1.50it/s]
4|5635|Loss: 0.8288517594337463:  87%|████████▋ | 5635/6470 [1:10:05<08:51,  1.57it/s]
4|5636|Loss: 0.8674404621124268:  87%|████████▋ | 5635/6470 [1:10:05<08:51,  1.57it/s]
4|5636|Loss: 0.8674404621124268:  87%|████████▋ | 5636/6470 [1:10:05<08:13,  1.69it/s]
4|5637|Loss: 0.8252190947532654:  87%|████████▋ | 5636/6470 [1:10:06<08:13,  1.69it/s]
4|5637|Loss: 0.8252190947532654:  87%|████████▋ | 5637/6470 [1:10:06<08:27,  1.64it/s]
4|5638|Loss: 0.8459733128547668:  87%|████████▋ | 5637/6470 [1:10:06<08:27,  1.64it/s]
4|5638|Loss: 0.8459733128547668:  87%|████████▋ | 5638/6470 [1:10:07<08:51,  1.56it/s]
4|5639|Loss: 0.7977898716926575:  87%|████████▋ | 5638/6470 [1:10:07<08:51,  1.56it/s]
4|5639|Loss: 0.7977898716926575:  87%|████████▋ | 5639/6470 [1:10:07<09:18,  1.49it/s]
4|5640|Loss: 0.7830139994621277:  87%|████████▋ | 5639/6470 [1:10:08<09:18,  1.49it/s]
4|5640|Loss: 0.7830139994621277:  87%|████████▋ | 5640/6470 [1:10:08<09:37,  1.44it/s]
4|5641|Loss: 1.0358822345733643:  87%|████████▋ | 5640/6470 [1:10:09<09:37,  1.44it/s]
4|5641|Loss: 1.0358822345733643:  87%|████████▋ | 5641/6470 [1:10:09<09:49,  1.41it/s]
4|5642|Loss: 0.7781412601470947:  87%|████████▋ | 5641/6470 [1:10:09<09:49,  1.41it/s]
4|5642|Loss: 0.7781412601470947:  87%|████████▋ | 5642/6470 [1:10:09<09:15,  1.49it/s]
4|5643|Loss: 0.8148689866065979:  87%|████████▋ | 5642/6470 [1:10:10<09:15,  1.49it/s]
4|5643|Loss: 0.8148689866065979:  87%|████████▋ | 5643/6470 [1:10:10<09:31,  1.45it/s]
4|5644|Loss: 0.7971904873847961:  87%|████████▋ | 5643/6470 [1:10:11<09:31,  1.45it/s]
4|5644|Loss: 0.7971904873847961:  87%|████████▋ | 5644/6470 [1:10:11<10:01,  1.37it/s]
4|5645|Loss: 0.898106575012207:  87%|████████▋ | 5644/6470 [1:10:11<10:01,  1.37it/s] 
4|5645|Loss: 0.898106575012207:  87%|████████▋ | 5645/6470 [1:10:12<09:47,  1.41it/s]
4|5646|Loss: 0.8654744029045105:  87%|████████▋ | 5645/6470 [1:10:12<09:47,  1.41it/s]
4|5646|Loss: 0.8654744029045105:  87%|████████▋ | 5646/6470 [1:10:12<09:22,  1.46it/s]
4|5647|Loss: 0.7445447444915771:  87%|████████▋ | 5646/6470 [1:10:13<09:22,  1.46it/s]
4|5647|Loss: 0.7445447444915771:  87%|████████▋ | 5647/6470 [1:10:13<08:37,  1.59it/s]
4|5648|Loss: 0.9727853536605835:  87%|████████▋ | 5647/6470 [1:10:13<08:37,  1.59it/s]
4|5648|Loss: 0.9727853536605835:  87%|████████▋ | 5648/6470 [1:10:13<08:33,  1.60it/s]
4|5649|Loss: 0.8442879915237427:  87%|████████▋ | 5648/6470 [1:10:14<08:33,  1.60it/s]
4|5649|Loss: 0.8442879915237427:  87%|████████▋ | 5649/6470 [1:10:14<09:38,  1.42it/s]
4|5650|Loss: 0.9046532511711121:  87%|████████▋ | 5649/6470 [1:10:15<09:38,  1.42it/s]
4|5650|Loss: 0.9046532511711121:  87%|████████▋ | 5650/6470 [1:10:15<10:32,  1.30it/s]
4|5651|Loss: 0.8079367280006409:  87%|████████▋ | 5650/6470 [1:10:16<10:32,  1.30it/s]
4|5651|Loss: 0.8079367280006409:  87%|████████▋ | 5651/6470 [1:10:16<10:40,  1.28it/s]
4|5652|Loss: 0.8587077260017395:  87%|████████▋ | 5651/6470 [1:10:16<10:40,  1.28it/s]
4|5652|Loss: 0.8587077260017395:  87%|████████▋ | 5652/6470 [1:10:17<11:02,  1.24it/s]
4|5653|Loss: 0.8796829581260681:  87%|████████▋ | 5652/6470 [1:10:17<11:02,  1.24it/s]
4|5653|Loss: 0.8796829581260681:  87%|████████▋ | 5653/6470 [1:10:18<10:42,  1.27it/s]
4|5654|Loss: 0.9057698249816895:  87%|████████▋ | 5653/6470 [1:10:18<10:42,  1.27it/s]
4|5654|Loss: 0.9057698249816895:  87%|████████▋ | 5654/6470 [1:10:18<10:13,  1.33it/s]
4|5655|Loss: 0.7628111839294434:  87%|████████▋ | 5654/6470 [1:10:19<10:13,  1.33it/s]
4|5655|Loss: 0.7628111839294434:  87%|████████▋ | 5655/6470 [1:10:19<09:42,  1.40it/s]
4|5656|Loss: 0.8575776219367981:  87%|████████▋ | 5655/6470 [1:10:19<09:42,  1.40it/s]
4|5656|Loss: 0.8575776219367981:  87%|████████▋ | 5656/6470 [1:10:20<09:52,  1.37it/s]
4|5657|Loss: 0.8463365435600281:  87%|████████▋ | 5656/6470 [1:10:20<09:52,  1.37it/s]
4|5657|Loss: 0.8463365435600281:  87%|████████▋ | 5657/6470 [1:10:20<10:08,  1.34it/s]
4|5658|Loss: 0.8019351959228516:  87%|████████▋ | 5657/6470 [1:10:21<10:08,  1.34it/s]
4|5658|Loss: 0.8019351959228516:  87%|████████▋ | 5658/6470 [1:10:21<10:17,  1.32it/s]
4|5659|Loss: 0.7407527565956116:  87%|████████▋ | 5658/6470 [1:10:22<10:17,  1.32it/s]
4|5659|Loss: 0.7407527565956116:  87%|████████▋ | 5659/6470 [1:10:22<10:40,  1.27it/s]
4|5660|Loss: 0.5833304524421692:  87%|████████▋ | 5659/6470 [1:10:22<10:40,  1.27it/s]
4|5660|Loss: 0.5833304524421692:  87%|████████▋ | 5660/6470 [1:10:23<09:41,  1.39it/s]
4|5661|Loss: 0.8884169459342957:  87%|████████▋ | 5660/6470 [1:10:23<09:41,  1.39it/s]
4|5661|Loss: 0.8884169459342957:  87%|████████▋ | 5661/6470 [1:10:23<09:24,  1.43it/s]
4|5662|Loss: 0.742692232131958:  87%|████████▋ | 5661/6470 [1:10:24<09:24,  1.43it/s] 
4|5662|Loss: 0.742692232131958:  88%|████████▊ | 5662/6470 [1:10:24<10:35,  1.27it/s]
4|5663|Loss: 0.8205251097679138:  88%|████████▊ | 5662/6470 [1:10:25<10:35,  1.27it/s]
4|5663|Loss: 0.8205251097679138:  88%|████████▊ | 5663/6470 [1:10:25<11:03,  1.22it/s]
4|5664|Loss: 0.7088860869407654:  88%|████████▊ | 5663/6470 [1:10:26<11:03,  1.22it/s]
4|5664|Loss: 0.7088860869407654:  88%|████████▊ | 5664/6470 [1:10:26<11:19,  1.19it/s]
4|5665|Loss: 0.82654869556427:  88%|████████▊ | 5664/6470 [1:10:27<11:19,  1.19it/s]  
4|5665|Loss: 0.82654869556427:  88%|████████▊ | 5665/6470 [1:10:27<10:54,  1.23it/s]
4|5666|Loss: 0.8797541856765747:  88%|████████▊ | 5665/6470 [1:10:27<10:54,  1.23it/s]
4|5666|Loss: 0.8797541856765747:  88%|████████▊ | 5666/6470 [1:10:28<11:21,  1.18it/s]
4|5667|Loss: 0.9822791814804077:  88%|████████▊ | 5666/6470 [1:10:28<11:21,  1.18it/s]
4|5667|Loss: 0.9822791814804077:  88%|████████▊ | 5667/6470 [1:10:28<10:20,  1.29it/s]
4|5668|Loss: 0.8466050624847412:  88%|████████▊ | 5667/6470 [1:10:29<10:20,  1.29it/s]
4|5668|Loss: 0.8466050624847412:  88%|████████▊ | 5668/6470 [1:10:29<09:49,  1.36it/s]
4|5669|Loss: 0.8344284892082214:  88%|████████▊ | 5668/6470 [1:10:29<09:49,  1.36it/s]
4|5669|Loss: 0.8344284892082214:  88%|████████▊ | 5669/6470 [1:10:30<09:17,  1.44it/s]
4|5670|Loss: 0.8818840384483337:  88%|████████▊ | 5669/6470 [1:10:30<09:17,  1.44it/s]
4|5670|Loss: 0.8818840384483337:  88%|████████▊ | 5670/6470 [1:10:30<09:48,  1.36it/s]
4|5671|Loss: 0.8475111722946167:  88%|████████▊ | 5670/6470 [1:10:31<09:48,  1.36it/s]
4|5671|Loss: 0.8475111722946167:  88%|████████▊ | 5671/6470 [1:10:31<09:44,  1.37it/s]
4|5672|Loss: 0.7636752128601074:  88%|████████▊ | 5671/6470 [1:10:32<09:44,  1.37it/s]
4|5672|Loss: 0.7636752128601074:  88%|████████▊ | 5672/6470 [1:10:32<10:20,  1.29it/s]
4|5673|Loss: 0.8355305194854736:  88%|████████▊ | 5672/6470 [1:10:33<10:20,  1.29it/s]
4|5673|Loss: 0.8355305194854736:  88%|████████▊ | 5673/6470 [1:10:33<10:32,  1.26it/s]
4|5674|Loss: 0.8608932495117188:  88%|████████▊ | 5673/6470 [1:10:33<10:32,  1.26it/s]
4|5674|Loss: 0.8608932495117188:  88%|████████▊ | 5674/6470 [1:10:34<10:54,  1.22it/s]
4|5675|Loss: 0.7891194820404053:  88%|████████▊ | 5674/6470 [1:10:34<10:54,  1.22it/s]
4|5675|Loss: 0.7891194820404053:  88%|████████▊ | 5675/6470 [1:10:35<11:17,  1.17it/s]
4|5676|Loss: 0.709653377532959:  88%|████████▊ | 5675/6470 [1:10:35<11:17,  1.17it/s] 
4|5676|Loss: 0.709653377532959:  88%|████████▊ | 5676/6470 [1:10:35<10:48,  1.23it/s]
4|5677|Loss: 0.8488756418228149:  88%|████████▊ | 5676/6470 [1:10:36<10:48,  1.23it/s]
4|5677|Loss: 0.8488756418228149:  88%|████████▊ | 5677/6470 [1:10:36<10:03,  1.31it/s]
4|5678|Loss: 0.8303653001785278:  88%|████████▊ | 5677/6470 [1:10:36<10:03,  1.31it/s]
4|5678|Loss: 0.8303653001785278:  88%|████████▊ | 5678/6470 [1:10:37<09:38,  1.37it/s]
4|5679|Loss: 0.9162141680717468:  88%|████████▊ | 5678/6470 [1:10:37<09:38,  1.37it/s]
4|5679|Loss: 0.9162141680717468:  88%|████████▊ | 5679/6470 [1:10:37<09:13,  1.43it/s]
4|5680|Loss: 0.7840874791145325:  88%|████████▊ | 5679/6470 [1:10:38<09:13,  1.43it/s]
4|5680|Loss: 0.7840874791145325:  88%|████████▊ | 5680/6470 [1:10:38<08:49,  1.49it/s]
4|5681|Loss: 0.7298719882965088:  88%|████████▊ | 5680/6470 [1:10:38<08:49,  1.49it/s]
4|5681|Loss: 0.7298719882965088:  88%|████████▊ | 5681/6470 [1:10:39<09:28,  1.39it/s]
4|5682|Loss: 0.6898278594017029:  88%|████████▊ | 5681/6470 [1:10:39<09:28,  1.39it/s]
4|5682|Loss: 0.6898278594017029:  88%|████████▊ | 5682/6470 [1:10:40<09:44,  1.35it/s]
4|5683|Loss: 0.8855971693992615:  88%|████████▊ | 5682/6470 [1:10:40<09:44,  1.35it/s]
4|5683|Loss: 0.8855971693992615:  88%|████████▊ | 5683/6470 [1:10:40<10:12,  1.29it/s]
4|5684|Loss: 0.7509954571723938:  88%|████████▊ | 5683/6470 [1:10:41<10:12,  1.29it/s]
4|5684|Loss: 0.7509954571723938:  88%|████████▊ | 5684/6470 [1:10:41<10:42,  1.22it/s]
4|5685|Loss: 0.7988806366920471:  88%|████████▊ | 5684/6470 [1:10:42<10:42,  1.22it/s]
4|5685|Loss: 0.7988806366920471:  88%|████████▊ | 5685/6470 [1:10:42<09:50,  1.33it/s]
4|5686|Loss: 0.8962279558181763:  88%|████████▊ | 5685/6470 [1:10:42<09:50,  1.33it/s]
4|5686|Loss: 0.8962279558181763:  88%|████████▊ | 5686/6470 [1:10:43<09:12,  1.42it/s]
4|5687|Loss: 0.7281184792518616:  88%|████████▊ | 5686/6470 [1:10:43<09:12,  1.42it/s]
4|5687|Loss: 0.7281184792518616:  88%|████████▊ | 5687/6470 [1:10:43<09:57,  1.31it/s]
4|5688|Loss: 0.8774843811988831:  88%|████████▊ | 5687/6470 [1:10:44<09:57,  1.31it/s]
4|5688|Loss: 0.8774843811988831:  88%|████████▊ | 5688/6470 [1:10:44<09:49,  1.33it/s]
4|5689|Loss: 0.8610011339187622:  88%|████████▊ | 5688/6470 [1:10:45<09:49,  1.33it/s]
4|5689|Loss: 0.8610011339187622:  88%|████████▊ | 5689/6470 [1:10:45<09:47,  1.33it/s]
4|5690|Loss: 0.8076896667480469:  88%|████████▊ | 5689/6470 [1:10:45<09:47,  1.33it/s]
4|5690|Loss: 0.8076896667480469:  88%|████████▊ | 5690/6470 [1:10:46<09:08,  1.42it/s]
4|5691|Loss: 0.8933044075965881:  88%|████████▊ | 5690/6470 [1:10:46<09:08,  1.42it/s]
4|5691|Loss: 0.8933044075965881:  88%|████████▊ | 5691/6470 [1:10:46<09:17,  1.40it/s]
4|5692|Loss: 0.7871668338775635:  88%|████████▊ | 5691/6470 [1:10:47<09:17,  1.40it/s]
4|5692|Loss: 0.7871668338775635:  88%|████████▊ | 5692/6470 [1:10:47<09:45,  1.33it/s]
4|5693|Loss: 0.7601872682571411:  88%|████████▊ | 5692/6470 [1:10:48<09:45,  1.33it/s]
4|5693|Loss: 0.7601872682571411:  88%|████████▊ | 5693/6470 [1:10:48<09:21,  1.38it/s]
4|5694|Loss: 0.8558130264282227:  88%|████████▊ | 5693/6470 [1:10:48<09:21,  1.38it/s]
4|5694|Loss: 0.8558130264282227:  88%|████████▊ | 5694/6470 [1:10:49<09:42,  1.33it/s]
4|5695|Loss: 0.7986019253730774:  88%|████████▊ | 5694/6470 [1:10:49<09:42,  1.33it/s]
4|5695|Loss: 0.7986019253730774:  88%|████████▊ | 5695/6470 [1:10:49<09:47,  1.32it/s]
4|5696|Loss: 0.7931368947029114:  88%|████████▊ | 5695/6470 [1:10:50<09:47,  1.32it/s]
4|5696|Loss: 0.7931368947029114:  88%|████████▊ | 5696/6470 [1:10:50<08:56,  1.44it/s]
4|5697|Loss: 0.8156647682189941:  88%|████████▊ | 5696/6470 [1:10:50<08:56,  1.44it/s]
4|5697|Loss: 0.8156647682189941:  88%|████████▊ | 5697/6470 [1:10:50<08:31,  1.51it/s]
4|5698|Loss: 0.7841905951499939:  88%|████████▊ | 5697/6470 [1:10:51<08:31,  1.51it/s]
4|5698|Loss: 0.7841905951499939:  88%|████████▊ | 5698/6470 [1:10:51<08:31,  1.51it/s]
4|5699|Loss: 0.7292867302894592:  88%|████████▊ | 5698/6470 [1:10:52<08:31,  1.51it/s]
4|5699|Loss: 0.7292867302894592:  88%|████████▊ | 5699/6470 [1:10:52<09:11,  1.40it/s]
4|5700|Loss: 0.7506226897239685:  88%|████████▊ | 5699/6470 [1:10:52<09:11,  1.40it/s]
4|5700|Loss: 0.7506226897239685:  88%|████████▊ | 5700/6470 [1:10:53<09:08,  1.40it/s]
4|5701|Loss: 0.8363733887672424:  88%|████████▊ | 5700/6470 [1:10:53<09:08,  1.40it/s]
4|5701|Loss: 0.8363733887672424:  88%|████████▊ | 5701/6470 [1:10:54<09:53,  1.30it/s]
4|5702|Loss: 0.7476841807365417:  88%|████████▊ | 5701/6470 [1:10:54<09:53,  1.30it/s]
4|5702|Loss: 0.7476841807365417:  88%|████████▊ | 5702/6470 [1:10:54<09:59,  1.28it/s]
4|5703|Loss: 0.8656264543533325:  88%|████████▊ | 5702/6470 [1:10:55<09:59,  1.28it/s]
4|5703|Loss: 0.8656264543533325:  88%|████████▊ | 5703/6470 [1:10:55<09:44,  1.31it/s]
4|5704|Loss: 0.9423975944519043:  88%|████████▊ | 5703/6470 [1:10:56<09:44,  1.31it/s]
4|5704|Loss: 0.9423975944519043:  88%|████████▊ | 5704/6470 [1:10:56<09:40,  1.32it/s]
4|5705|Loss: 0.7679128050804138:  88%|████████▊ | 5704/6470 [1:10:56<09:40,  1.32it/s]
4|5705|Loss: 0.7679128050804138:  88%|████████▊ | 5705/6470 [1:10:57<09:49,  1.30it/s]
4|5706|Loss: 0.6953256130218506:  88%|████████▊ | 5705/6470 [1:10:57<09:49,  1.30it/s]
4|5706|Loss: 0.6953256130218506:  88%|████████▊ | 5706/6470 [1:10:57<09:49,  1.30it/s]
4|5707|Loss: 0.7195391654968262:  88%|████████▊ | 5706/6470 [1:10:58<09:49,  1.30it/s]
4|5707|Loss: 0.7195391654968262:  88%|████████▊ | 5707/6470 [1:10:58<09:44,  1.31it/s]
4|5708|Loss: 0.7032546401023865:  88%|████████▊ | 5707/6470 [1:10:59<09:44,  1.31it/s]
4|5708|Loss: 0.7032546401023865:  88%|████████▊ | 5708/6470 [1:10:59<09:22,  1.35it/s]
4|5709|Loss: 0.7652589678764343:  88%|████████▊ | 5708/6470 [1:10:59<09:22,  1.35it/s]
4|5709|Loss: 0.7652589678764343:  88%|████████▊ | 5709/6470 [1:11:00<09:02,  1.40it/s]
4|5710|Loss: 0.8490732312202454:  88%|████████▊ | 5709/6470 [1:11:00<09:02,  1.40it/s]
4|5710|Loss: 0.8490732312202454:  88%|████████▊ | 5710/6470 [1:11:00<09:36,  1.32it/s]
4|5711|Loss: 0.8555896878242493:  88%|████████▊ | 5710/6470 [1:11:01<09:36,  1.32it/s]
4|5711|Loss: 0.8555896878242493:  88%|████████▊ | 5711/6470 [1:11:01<10:07,  1.25it/s]
4|5712|Loss: 0.828446090221405:  88%|████████▊ | 5711/6470 [1:11:02<10:07,  1.25it/s] 
4|5712|Loss: 0.828446090221405:  88%|████████▊ | 5712/6470 [1:11:02<10:06,  1.25it/s]
4|5713|Loss: 0.8289732336997986:  88%|████████▊ | 5712/6470 [1:11:02<10:06,  1.25it/s]
4|5713|Loss: 0.8289732336997986:  88%|████████▊ | 5713/6470 [1:11:03<09:31,  1.32it/s]
4|5714|Loss: 0.9559482336044312:  88%|████████▊ | 5713/6470 [1:11:03<09:31,  1.32it/s]
4|5714|Loss: 0.9559482336044312:  88%|████████▊ | 5714/6470 [1:11:04<09:45,  1.29it/s]
4|5715|Loss: 0.8480305671691895:  88%|████████▊ | 5714/6470 [1:11:04<09:45,  1.29it/s]
4|5715|Loss: 0.8480305671691895:  88%|████████▊ | 5715/6470 [1:11:04<09:08,  1.38it/s]
4|5716|Loss: 0.7250223755836487:  88%|████████▊ | 5715/6470 [1:11:05<09:08,  1.38it/s]
4|5716|Loss: 0.7250223755836487:  88%|████████▊ | 5716/6470 [1:11:05<09:55,  1.27it/s]
4|5717|Loss: 0.9294865131378174:  88%|████████▊ | 5716/6470 [1:11:06<09:55,  1.27it/s]
4|5717|Loss: 0.9294865131378174:  88%|████████▊ | 5717/6470 [1:11:06<09:45,  1.29it/s]
4|5718|Loss: 0.9503160715103149:  88%|████████▊ | 5717/6470 [1:11:06<09:45,  1.29it/s]
4|5718|Loss: 0.9503160715103149:  88%|████████▊ | 5718/6470 [1:11:07<09:36,  1.30it/s]
4|5719|Loss: 0.8696702718734741:  88%|████████▊ | 5718/6470 [1:11:07<09:36,  1.30it/s]
4|5719|Loss: 0.8696702718734741:  88%|████████▊ | 5719/6470 [1:11:07<09:42,  1.29it/s]
4|5720|Loss: 0.8424990177154541:  88%|████████▊ | 5719/6470 [1:11:08<09:42,  1.29it/s]
4|5720|Loss: 0.8424990177154541:  88%|████████▊ | 5720/6470 [1:11:08<09:22,  1.33it/s]
4|5721|Loss: 0.7166300415992737:  88%|████████▊ | 5720/6470 [1:11:08<09:22,  1.33it/s]
4|5721|Loss: 0.7166300415992737:  88%|████████▊ | 5721/6470 [1:11:09<08:45,  1.42it/s]
4|5722|Loss: 0.7631111741065979:  88%|████████▊ | 5721/6470 [1:11:09<08:45,  1.42it/s]
4|5722|Loss: 0.7631111741065979:  88%|████████▊ | 5722/6470 [1:11:09<08:58,  1.39it/s]
4|5723|Loss: 0.8669730424880981:  88%|████████▊ | 5722/6470 [1:11:10<08:58,  1.39it/s]
4|5723|Loss: 0.8669730424880981:  88%|████████▊ | 5723/6470 [1:11:10<09:16,  1.34it/s]
4|5724|Loss: 0.7677584886550903:  88%|████████▊ | 5723/6470 [1:11:11<09:16,  1.34it/s]
4|5724|Loss: 0.7677584886550903:  88%|████████▊ | 5724/6470 [1:11:11<09:46,  1.27it/s]
4|5725|Loss: 0.6623975038528442:  88%|████████▊ | 5724/6470 [1:11:12<09:46,  1.27it/s]
4|5725|Loss: 0.6623975038528442:  88%|████████▊ | 5725/6470 [1:11:12<09:46,  1.27it/s]
4|5726|Loss: 0.6178774237632751:  88%|████████▊ | 5725/6470 [1:11:12<09:46,  1.27it/s]
4|5726|Loss: 0.6178774237632751:  89%|████████▊ | 5726/6470 [1:11:13<09:37,  1.29it/s]
4|5727|Loss: 0.7495059370994568:  89%|████████▊ | 5726/6470 [1:11:13<09:37,  1.29it/s]
4|5727|Loss: 0.7495059370994568:  89%|████████▊ | 5727/6470 [1:11:13<09:08,  1.35it/s]
4|5728|Loss: 0.8043646812438965:  89%|████████▊ | 5727/6470 [1:11:14<09:08,  1.35it/s]
4|5728|Loss: 0.8043646812438965:  89%|████████▊ | 5728/6470 [1:11:14<09:06,  1.36it/s]
4|5729|Loss: 0.7983930110931396:  89%|████████▊ | 5728/6470 [1:11:14<09:06,  1.36it/s]
4|5729|Loss: 0.7983930110931396:  89%|████████▊ | 5729/6470 [1:11:15<08:58,  1.38it/s]
4|5730|Loss: 0.9684417843818665:  89%|████████▊ | 5729/6470 [1:11:15<08:58,  1.38it/s]
4|5730|Loss: 0.9684417843818665:  89%|████████▊ | 5730/6470 [1:11:16<09:10,  1.34it/s]
4|5731|Loss: 0.8530306816101074:  89%|████████▊ | 5730/6470 [1:11:16<09:10,  1.34it/s]
4|5731|Loss: 0.8530306816101074:  89%|████████▊ | 5731/6470 [1:11:16<09:21,  1.32it/s]
4|5732|Loss: 0.8258465528488159:  89%|████████▊ | 5731/6470 [1:11:17<09:21,  1.32it/s]
4|5732|Loss: 0.8258465528488159:  89%|████████▊ | 5732/6470 [1:11:17<09:05,  1.35it/s]
4|5733|Loss: 0.6841427683830261:  89%|████████▊ | 5732/6470 [1:11:17<09:05,  1.35it/s]
4|5733|Loss: 0.6841427683830261:  89%|████████▊ | 5733/6470 [1:11:18<08:48,  1.39it/s]
4|5734|Loss: 0.6976594924926758:  89%|████████▊ | 5733/6470 [1:11:18<08:48,  1.39it/s]
4|5734|Loss: 0.6976594924926758:  89%|████████▊ | 5734/6470 [1:11:18<08:04,  1.52it/s]
4|5735|Loss: 0.855034351348877:  89%|████████▊ | 5734/6470 [1:11:19<08:04,  1.52it/s] 
4|5735|Loss: 0.855034351348877:  89%|████████▊ | 5735/6470 [1:11:19<08:10,  1.50it/s]
4|5736|Loss: 0.6425696015357971:  89%|████████▊ | 5735/6470 [1:11:19<08:10,  1.50it/s]
4|5736|Loss: 0.6425696015357971:  89%|████████▊ | 5736/6470 [1:11:20<08:15,  1.48it/s]
4|5737|Loss: 0.7522536516189575:  89%|████████▊ | 5736/6470 [1:11:20<08:15,  1.48it/s]
4|5737|Loss: 0.7522536516189575:  89%|████████▊ | 5737/6470 [1:11:20<07:54,  1.55it/s]
4|5738|Loss: 0.7088374495506287:  89%|████████▊ | 5737/6470 [1:11:21<07:54,  1.55it/s]
4|5738|Loss: 0.7088374495506287:  89%|████████▊ | 5738/6470 [1:11:21<07:52,  1.55it/s]
4|5739|Loss: 0.7687121629714966:  89%|████████▊ | 5738/6470 [1:11:21<07:52,  1.55it/s]
4|5739|Loss: 0.7687121629714966:  89%|████████▊ | 5739/6470 [1:11:22<08:24,  1.45it/s]
4|5740|Loss: 0.795742392539978:  89%|████████▊ | 5739/6470 [1:11:22<08:24,  1.45it/s] 
4|5740|Loss: 0.795742392539978:  89%|████████▊ | 5740/6470 [1:11:22<08:52,  1.37it/s]
4|5741|Loss: 0.8164312839508057:  89%|████████▊ | 5740/6470 [1:11:23<08:52,  1.37it/s]
4|5741|Loss: 0.8164312839508057:  89%|████████▊ | 5741/6470 [1:11:23<08:52,  1.37it/s]
4|5742|Loss: 0.8231061697006226:  89%|████████▊ | 5741/6470 [1:11:24<08:52,  1.37it/s]
4|5742|Loss: 0.8231061697006226:  89%|████████▊ | 5742/6470 [1:11:24<08:54,  1.36it/s]
4|5743|Loss: 0.8623424768447876:  89%|████████▊ | 5742/6470 [1:11:24<08:54,  1.36it/s]
4|5743|Loss: 0.8623424768447876:  89%|████████▉ | 5743/6470 [1:11:25<08:59,  1.35it/s]
4|5744|Loss: 0.8867573738098145:  89%|████████▉ | 5743/6470 [1:11:25<08:59,  1.35it/s]
4|5744|Loss: 0.8867573738098145:  89%|████████▉ | 5744/6470 [1:11:25<08:33,  1.41it/s]
4|5745|Loss: 0.9223626852035522:  89%|████████▉ | 5744/6470 [1:11:26<08:33,  1.41it/s]
4|5745|Loss: 0.9223626852035522:  89%|████████▉ | 5745/6470 [1:11:26<09:18,  1.30it/s]
4|5746|Loss: 0.7302640676498413:  89%|████████▉ | 5745/6470 [1:11:27<09:18,  1.30it/s]
4|5746|Loss: 0.7302640676498413:  89%|████████▉ | 5746/6470 [1:11:27<09:56,  1.21it/s]
4|5747|Loss: 0.7206820845603943:  89%|████████▉ | 5746/6470 [1:11:28<09:56,  1.21it/s]
4|5747|Loss: 0.7206820845603943:  89%|████████▉ | 5747/6470 [1:11:28<09:41,  1.24it/s]
4|5748|Loss: 0.8662504553794861:  89%|████████▉ | 5747/6470 [1:11:28<09:41,  1.24it/s]
4|5748|Loss: 0.8662504553794861:  89%|████████▉ | 5748/6470 [1:11:29<09:29,  1.27it/s]
4|5749|Loss: 0.9513999223709106:  89%|████████▉ | 5748/6470 [1:11:29<09:29,  1.27it/s]
4|5749|Loss: 0.9513999223709106:  89%|████████▉ | 5749/6470 [1:11:29<08:48,  1.36it/s]
4|5750|Loss: 0.7586883902549744:  89%|████████▉ | 5749/6470 [1:11:30<08:48,  1.36it/s]
4|5750|Loss: 0.7586883902549744:  89%|████████▉ | 5750/6470 [1:11:30<08:52,  1.35it/s]
4|5751|Loss: 0.7866928577423096:  89%|████████▉ | 5750/6470 [1:11:31<08:52,  1.35it/s]
4|5751|Loss: 0.7866928577423096:  89%|████████▉ | 5751/6470 [1:11:31<09:27,  1.27it/s]
4|5752|Loss: 0.7798295021057129:  89%|████████▉ | 5751/6470 [1:11:31<09:27,  1.27it/s]
4|5752|Loss: 0.7798295021057129:  89%|████████▉ | 5752/6470 [1:11:32<09:19,  1.28it/s]
4|5753|Loss: 0.9012492299079895:  89%|████████▉ | 5752/6470 [1:11:32<09:19,  1.28it/s]
4|5753|Loss: 0.9012492299079895:  89%|████████▉ | 5753/6470 [1:11:32<08:46,  1.36it/s]
4|5754|Loss: 0.8371601700782776:  89%|████████▉ | 5753/6470 [1:11:33<08:46,  1.36it/s]
4|5754|Loss: 0.8371601700782776:  89%|████████▉ | 5754/6470 [1:11:33<08:21,  1.43it/s]
4|5755|Loss: 0.9012274146080017:  89%|████████▉ | 5754/6470 [1:11:33<08:21,  1.43it/s]
4|5755|Loss: 0.9012274146080017:  89%|████████▉ | 5755/6470 [1:11:34<07:58,  1.49it/s]
4|5756|Loss: 0.9602925181388855:  89%|████████▉ | 5755/6470 [1:11:34<07:58,  1.49it/s]
4|5756|Loss: 0.9602925181388855:  89%|████████▉ | 5756/6470 [1:11:34<08:00,  1.49it/s]
4|5757|Loss: 0.7707197666168213:  89%|████████▉ | 5756/6470 [1:11:35<08:00,  1.49it/s]
4|5757|Loss: 0.7707197666168213:  89%|████████▉ | 5757/6470 [1:11:35<08:01,  1.48it/s]
4|5758|Loss: 1.0171949863433838:  89%|████████▉ | 5757/6470 [1:11:35<08:01,  1.48it/s]
4|5758|Loss: 1.0171949863433838:  89%|████████▉ | 5758/6470 [1:11:36<08:22,  1.42it/s]
4|5759|Loss: 0.8176923990249634:  89%|████████▉ | 5758/6470 [1:11:36<08:22,  1.42it/s]
4|5759|Loss: 0.8176923990249634:  89%|████████▉ | 5759/6470 [1:11:36<07:47,  1.52it/s]
4|5760|Loss: 0.6639717817306519:  89%|████████▉ | 5759/6470 [1:11:37<07:47,  1.52it/s]
4|5760|Loss: 0.6639717817306519:  89%|████████▉ | 5760/6470 [1:11:37<08:25,  1.41it/s]
4|5761|Loss: 0.8256393074989319:  89%|████████▉ | 5760/6470 [1:11:38<08:25,  1.41it/s]
4|5761|Loss: 0.8256393074989319:  89%|████████▉ | 5761/6470 [1:11:38<09:14,  1.28it/s]
4|5762|Loss: 0.7783937454223633:  89%|████████▉ | 5761/6470 [1:11:38<09:14,  1.28it/s]
4|5762|Loss: 0.7783937454223633:  89%|████████▉ | 5762/6470 [1:11:39<09:09,  1.29it/s]
4|5763|Loss: 0.6348385810852051:  89%|████████▉ | 5762/6470 [1:11:39<09:09,  1.29it/s]
4|5763|Loss: 0.6348385810852051:  89%|████████▉ | 5763/6470 [1:11:40<09:27,  1.25it/s]
4|5764|Loss: 0.7139025926589966:  89%|████████▉ | 5763/6470 [1:11:40<09:27,  1.25it/s]
4|5764|Loss: 0.7139025926589966:  89%|████████▉ | 5764/6470 [1:11:40<08:52,  1.33it/s]
4|5765|Loss: 0.8503333926200867:  89%|████████▉ | 5764/6470 [1:11:41<08:52,  1.33it/s]
4|5765|Loss: 0.8503333926200867:  89%|████████▉ | 5765/6470 [1:11:41<08:11,  1.44it/s]
4|5766|Loss: 0.6997339129447937:  89%|████████▉ | 5765/6470 [1:11:41<08:11,  1.44it/s]
4|5766|Loss: 0.6997339129447937:  89%|████████▉ | 5766/6470 [1:11:41<07:40,  1.53it/s]
4|5767|Loss: 0.7572148442268372:  89%|████████▉ | 5766/6470 [1:11:42<07:40,  1.53it/s]
4|5767|Loss: 0.7572148442268372:  89%|████████▉ | 5767/6470 [1:11:42<07:15,  1.61it/s]
4|5768|Loss: 0.8730795979499817:  89%|████████▉ | 5767/6470 [1:11:42<07:15,  1.61it/s]
4|5768|Loss: 0.8730795979499817:  89%|████████▉ | 5768/6470 [1:11:43<07:35,  1.54it/s]
4|5769|Loss: 0.8503521680831909:  89%|████████▉ | 5768/6470 [1:11:43<07:35,  1.54it/s]
4|5769|Loss: 0.8503521680831909:  89%|████████▉ | 5769/6470 [1:11:43<07:36,  1.53it/s]
4|5770|Loss: 0.945781409740448:  89%|████████▉ | 5769/6470 [1:11:44<07:36,  1.53it/s] 
4|5770|Loss: 0.945781409740448:  89%|████████▉ | 5770/6470 [1:11:44<07:56,  1.47it/s]
4|5771|Loss: 0.8051736354827881:  89%|████████▉ | 5770/6470 [1:11:44<07:56,  1.47it/s]
4|5771|Loss: 0.8051736354827881:  89%|████████▉ | 5771/6470 [1:11:45<08:12,  1.42it/s]
4|5772|Loss: 0.6474587321281433:  89%|████████▉ | 5771/6470 [1:11:45<08:12,  1.42it/s]
4|5772|Loss: 0.6474587321281433:  89%|████████▉ | 5772/6470 [1:11:46<08:27,  1.37it/s]
4|5773|Loss: 0.7998083829879761:  89%|████████▉ | 5772/6470 [1:11:46<08:27,  1.37it/s]
4|5773|Loss: 0.7998083829879761:  89%|████████▉ | 5773/6470 [1:11:46<09:02,  1.28it/s]
4|5774|Loss: 0.8371223211288452:  89%|████████▉ | 5773/6470 [1:11:47<09:02,  1.28it/s]
4|5774|Loss: 0.8371223211288452:  89%|████████▉ | 5774/6470 [1:11:47<08:44,  1.33it/s]
4|5775|Loss: 0.8127555847167969:  89%|████████▉ | 5774/6470 [1:11:48<08:44,  1.33it/s]
4|5775|Loss: 0.8127555847167969:  89%|████████▉ | 5775/6470 [1:11:48<08:26,  1.37it/s]
4|5776|Loss: 0.8960063457489014:  89%|████████▉ | 5775/6470 [1:11:48<08:26,  1.37it/s]
4|5776|Loss: 0.8960063457489014:  89%|████████▉ | 5776/6470 [1:11:48<08:01,  1.44it/s]
4|5777|Loss: 0.8263393640518188:  89%|████████▉ | 5776/6470 [1:11:49<08:01,  1.44it/s]
4|5777|Loss: 0.8263393640518188:  89%|████████▉ | 5777/6470 [1:11:49<08:11,  1.41it/s]
4|5778|Loss: 0.8177998661994934:  89%|████████▉ | 5777/6470 [1:11:50<08:11,  1.41it/s]
4|5778|Loss: 0.8177998661994934:  89%|████████▉ | 5778/6470 [1:11:50<08:05,  1.42it/s]
4|5779|Loss: 0.8773528337478638:  89%|████████▉ | 5778/6470 [1:11:50<08:05,  1.42it/s]
4|5779|Loss: 0.8773528337478638:  89%|████████▉ | 5779/6470 [1:11:51<08:22,  1.38it/s]
4|5780|Loss: 0.8985002636909485:  89%|████████▉ | 5779/6470 [1:11:51<08:22,  1.38it/s]
4|5780|Loss: 0.8985002636909485:  89%|████████▉ | 5780/6470 [1:11:51<08:22,  1.37it/s]
4|5781|Loss: 0.976433277130127:  89%|████████▉ | 5780/6470 [1:11:52<08:22,  1.37it/s] 
4|5781|Loss: 0.976433277130127:  89%|████████▉ | 5781/6470 [1:11:52<08:24,  1.37it/s]
4|5782|Loss: 0.9548245072364807:  89%|████████▉ | 5781/6470 [1:11:53<08:24,  1.37it/s]
4|5782|Loss: 0.9548245072364807:  89%|████████▉ | 5782/6470 [1:11:53<08:45,  1.31it/s]
4|5783|Loss: 0.8891211748123169:  89%|████████▉ | 5782/6470 [1:11:53<08:45,  1.31it/s]
4|5783|Loss: 0.8891211748123169:  89%|████████▉ | 5783/6470 [1:11:54<08:09,  1.40it/s]
4|5784|Loss: 0.7061852216720581:  89%|████████▉ | 5783/6470 [1:11:54<08:09,  1.40it/s]
4|5784|Loss: 0.7061852216720581:  89%|████████▉ | 5784/6470 [1:11:54<07:56,  1.44it/s]
4|5785|Loss: 0.7377903461456299:  89%|████████▉ | 5784/6470 [1:11:55<07:56,  1.44it/s]
4|5785|Loss: 0.7377903461456299:  89%|████████▉ | 5785/6470 [1:11:55<07:45,  1.47it/s]
4|5786|Loss: 0.9009496569633484:  89%|████████▉ | 5785/6470 [1:11:55<07:45,  1.47it/s]
4|5786|Loss: 0.9009496569633484:  89%|████████▉ | 5786/6470 [1:11:56<07:44,  1.47it/s]
4|5787|Loss: 0.8960571885108948:  89%|████████▉ | 5786/6470 [1:11:56<07:44,  1.47it/s]
4|5787|Loss: 0.8960571885108948:  89%|████████▉ | 5787/6470 [1:11:56<08:02,  1.42it/s]
4|5788|Loss: 0.8803778886795044:  89%|████████▉ | 5787/6470 [1:11:57<08:02,  1.42it/s]
4|5788|Loss: 0.8803778886795044:  89%|████████▉ | 5788/6470 [1:11:57<07:49,  1.45it/s]
4|5789|Loss: 0.8657935261726379:  89%|████████▉ | 5788/6470 [1:11:57<07:49,  1.45it/s]
4|5789|Loss: 0.8657935261726379:  89%|████████▉ | 5789/6470 [1:11:58<07:42,  1.47it/s]
4|5790|Loss: 0.692256510257721:  89%|████████▉ | 5789/6470 [1:11:58<07:42,  1.47it/s] 
4|5790|Loss: 0.692256510257721:  89%|████████▉ | 5790/6470 [1:11:59<08:29,  1.33it/s]
4|5791|Loss: 0.7276848554611206:  89%|████████▉ | 5790/6470 [1:11:59<08:29,  1.33it/s]
4|5791|Loss: 0.7276848554611206:  90%|████████▉ | 5791/6470 [1:12:00<09:11,  1.23it/s]
4|5792|Loss: 0.8853058218955994:  90%|████████▉ | 5791/6470 [1:12:00<09:11,  1.23it/s]
4|5792|Loss: 0.8853058218955994:  90%|████████▉ | 5792/6470 [1:12:00<08:57,  1.26it/s]
4|5793|Loss: 0.9729424118995667:  90%|████████▉ | 5792/6470 [1:12:01<08:57,  1.26it/s]
4|5793|Loss: 0.9729424118995667:  90%|████████▉ | 5793/6470 [1:12:01<08:11,  1.38it/s]
4|5794|Loss: 0.8843664526939392:  90%|████████▉ | 5793/6470 [1:12:01<08:11,  1.38it/s]
4|5794|Loss: 0.8843664526939392:  90%|████████▉ | 5794/6470 [1:12:02<08:09,  1.38it/s]
4|5795|Loss: 0.9554914236068726:  90%|████████▉ | 5794/6470 [1:12:02<08:09,  1.38it/s]
4|5795|Loss: 0.9554914236068726:  90%|████████▉ | 5795/6470 [1:12:02<08:31,  1.32it/s]
4|5796|Loss: 0.8173210024833679:  90%|████████▉ | 5795/6470 [1:12:03<08:31,  1.32it/s]
4|5796|Loss: 0.8173210024833679:  90%|████████▉ | 5796/6470 [1:12:03<08:36,  1.30it/s]
4|5797|Loss: 0.6857200860977173:  90%|████████▉ | 5796/6470 [1:12:04<08:36,  1.30it/s]
4|5797|Loss: 0.6857200860977173:  90%|████████▉ | 5797/6470 [1:12:04<08:21,  1.34it/s]
4|5798|Loss: 0.679412305355072:  90%|████████▉ | 5797/6470 [1:12:04<08:21,  1.34it/s] 
4|5798|Loss: 0.679412305355072:  90%|████████▉ | 5798/6470 [1:12:05<08:43,  1.28it/s]
4|5799|Loss: 0.8665761351585388:  90%|████████▉ | 5798/6470 [1:12:05<08:43,  1.28it/s]
4|5799|Loss: 0.8665761351585388:  90%|████████▉ | 5799/6470 [1:12:05<08:16,  1.35it/s]
4|5800|Loss: 0.8740134239196777:  90%|████████▉ | 5799/6470 [1:12:06<08:16,  1.35it/s]
4|5800|Loss: 0.8740134239196777:  90%|████████▉ | 5800/6470 [1:12:06<08:18,  1.35it/s]
4|5801|Loss: 0.8681374192237854:  90%|████████▉ | 5800/6470 [1:12:07<08:18,  1.35it/s]
4|5801|Loss: 0.8681374192237854:  90%|████████▉ | 5801/6470 [1:12:07<08:11,  1.36it/s]
4|5802|Loss: 0.9191558361053467:  90%|████████▉ | 5801/6470 [1:12:07<08:11,  1.36it/s]
4|5802|Loss: 0.9191558361053467:  90%|████████▉ | 5802/6470 [1:12:08<08:20,  1.33it/s]
4|5803|Loss: 0.8540175557136536:  90%|████████▉ | 5802/6470 [1:12:08<08:20,  1.33it/s]
4|5803|Loss: 0.8540175557136536:  90%|████████▉ | 5803/6470 [1:12:08<08:46,  1.27it/s]
4|5804|Loss: 0.821951687335968:  90%|████████▉ | 5803/6470 [1:12:09<08:46,  1.27it/s] 
4|5804|Loss: 0.821951687335968:  90%|████████▉ | 5804/6470 [1:12:09<08:28,  1.31it/s]
4|5805|Loss: 0.7685372233390808:  90%|████████▉ | 5804/6470 [1:12:10<08:28,  1.31it/s]
4|5805|Loss: 0.7685372233390808:  90%|████████▉ | 5805/6470 [1:12:10<07:36,  1.46it/s]
4|5806|Loss: 0.9250633120536804:  90%|████████▉ | 5805/6470 [1:12:10<07:36,  1.46it/s]
4|5806|Loss: 0.9250633120536804:  90%|████████▉ | 5806/6470 [1:12:10<07:10,  1.54it/s]
4|5807|Loss: 0.5274143218994141:  90%|████████▉ | 5806/6470 [1:12:11<07:10,  1.54it/s]
4|5807|Loss: 0.5274143218994141:  90%|████████▉ | 5807/6470 [1:12:11<07:05,  1.56it/s]
4|5808|Loss: 0.9089521169662476:  90%|████████▉ | 5807/6470 [1:12:11<07:05,  1.56it/s]
4|5808|Loss: 0.9089521169662476:  90%|████████▉ | 5808/6470 [1:12:12<07:23,  1.49it/s]
4|5809|Loss: 0.8190553188323975:  90%|████████▉ | 5808/6470 [1:12:12<07:23,  1.49it/s]
4|5809|Loss: 0.8190553188323975:  90%|████████▉ | 5809/6470 [1:12:12<07:39,  1.44it/s]
4|5810|Loss: 0.9784548878669739:  90%|████████▉ | 5809/6470 [1:12:13<07:39,  1.44it/s]
4|5810|Loss: 0.9784548878669739:  90%|████████▉ | 5810/6470 [1:12:13<08:08,  1.35it/s]
4|5811|Loss: 0.6823349595069885:  90%|████████▉ | 5810/6470 [1:12:14<08:08,  1.35it/s]
4|5811|Loss: 0.6823349595069885:  90%|████████▉ | 5811/6470 [1:12:14<08:14,  1.33it/s]
4|5812|Loss: 0.9443146586418152:  90%|████████▉ | 5811/6470 [1:12:14<08:14,  1.33it/s]
4|5812|Loss: 0.9443146586418152:  90%|████████▉ | 5812/6470 [1:12:15<08:13,  1.33it/s]
4|5813|Loss: 0.8378729820251465:  90%|████████▉ | 5812/6470 [1:12:15<08:13,  1.33it/s]
4|5813|Loss: 0.8378729820251465:  90%|████████▉ | 5813/6470 [1:12:16<08:19,  1.32it/s]
4|5814|Loss: 0.8690433502197266:  90%|████████▉ | 5813/6470 [1:12:16<08:19,  1.32it/s]
4|5814|Loss: 0.8690433502197266:  90%|████████▉ | 5814/6470 [1:12:16<08:24,  1.30it/s]
4|5815|Loss: 0.8596857190132141:  90%|████████▉ | 5814/6470 [1:12:17<08:24,  1.30it/s]
4|5815|Loss: 0.8596857190132141:  90%|████████▉ | 5815/6470 [1:12:17<07:21,  1.48it/s]
4|5816|Loss: 0.8448393940925598:  90%|████████▉ | 5815/6470 [1:12:17<07:21,  1.48it/s]
4|5816|Loss: 0.8448393940925598:  90%|████████▉ | 5816/6470 [1:12:17<06:57,  1.57it/s]
4|5817|Loss: 0.8165584206581116:  90%|████████▉ | 5816/6470 [1:12:18<06:57,  1.57it/s]
4|5817|Loss: 0.8165584206581116:  90%|████████▉ | 5817/6470 [1:12:18<06:58,  1.56it/s]
4|5818|Loss: 0.771205723285675:  90%|████████▉ | 5817/6470 [1:12:18<06:58,  1.56it/s] 
4|5818|Loss: 0.771205723285675:  90%|████████▉ | 5818/6470 [1:12:19<07:44,  1.40it/s]
4|5819|Loss: 0.8374664187431335:  90%|████████▉ | 5818/6470 [1:12:19<07:44,  1.40it/s]
4|5819|Loss: 0.8374664187431335:  90%|████████▉ | 5819/6470 [1:12:19<07:17,  1.49it/s]
4|5820|Loss: 0.8389970064163208:  90%|████████▉ | 5819/6470 [1:12:20<07:17,  1.49it/s]
4|5820|Loss: 0.8389970064163208:  90%|████████▉ | 5820/6470 [1:12:20<06:59,  1.55it/s]
4|5821|Loss: 0.6920725107192993:  90%|████████▉ | 5820/6470 [1:12:20<06:59,  1.55it/s]
4|5821|Loss: 0.6920725107192993:  90%|████████▉ | 5821/6470 [1:12:21<07:07,  1.52it/s]
4|5822|Loss: 0.7247840762138367:  90%|████████▉ | 5821/6470 [1:12:21<07:07,  1.52it/s]
4|5822|Loss: 0.7247840762138367:  90%|████████▉ | 5822/6470 [1:12:21<07:23,  1.46it/s]
4|5823|Loss: 0.7599465847015381:  90%|████████▉ | 5822/6470 [1:12:22<07:23,  1.46it/s]
4|5823|Loss: 0.7599465847015381:  90%|█████████ | 5823/6470 [1:12:22<06:56,  1.55it/s]
4|5824|Loss: 0.7641088366508484:  90%|█████████ | 5823/6470 [1:12:22<06:56,  1.55it/s]
4|5824|Loss: 0.7641088366508484:  90%|█████████ | 5824/6470 [1:12:23<06:53,  1.56it/s]
4|5825|Loss: 0.8897026181221008:  90%|█████████ | 5824/6470 [1:12:23<06:53,  1.56it/s]
4|5825|Loss: 0.8897026181221008:  90%|█████████ | 5825/6470 [1:12:23<07:35,  1.42it/s]
4|5826|Loss: 0.8877840042114258:  90%|█████████ | 5825/6470 [1:12:24<07:35,  1.42it/s]
4|5826|Loss: 0.8877840042114258:  90%|█████████ | 5826/6470 [1:12:24<07:51,  1.37it/s]
4|5827|Loss: 0.7477098703384399:  90%|█████████ | 5826/6470 [1:12:25<07:51,  1.37it/s]
4|5827|Loss: 0.7477098703384399:  90%|█████████ | 5827/6470 [1:12:25<07:35,  1.41it/s]
4|5828|Loss: 0.8344214558601379:  90%|█████████ | 5827/6470 [1:12:25<07:35,  1.41it/s]
4|5828|Loss: 0.8344214558601379:  90%|█████████ | 5828/6470 [1:12:26<07:44,  1.38it/s]
4|5829|Loss: 0.9094727039337158:  90%|█████████ | 5828/6470 [1:12:26<07:44,  1.38it/s]
4|5829|Loss: 0.9094727039337158:  90%|█████████ | 5829/6470 [1:12:26<07:53,  1.35it/s]
4|5830|Loss: 0.7780153751373291:  90%|█████████ | 5829/6470 [1:12:27<07:53,  1.35it/s]
4|5830|Loss: 0.7780153751373291:  90%|█████████ | 5830/6470 [1:12:27<06:55,  1.54it/s]
4|5831|Loss: 0.9378131628036499:  90%|█████████ | 5830/6470 [1:12:27<06:55,  1.54it/s]
4|5831|Loss: 0.9378131628036499:  90%|█████████ | 5831/6470 [1:12:28<07:02,  1.51it/s]
4|5832|Loss: 0.6381180286407471:  90%|█████████ | 5831/6470 [1:12:28<07:02,  1.51it/s]
4|5832|Loss: 0.6381180286407471:  90%|█████████ | 5832/6470 [1:12:28<06:55,  1.54it/s]
4|5833|Loss: 0.7927901148796082:  90%|█████████ | 5832/6470 [1:12:29<06:55,  1.54it/s]
4|5833|Loss: 0.7927901148796082:  90%|█████████ | 5833/6470 [1:12:29<07:12,  1.47it/s]
4|5834|Loss: 0.7565951347351074:  90%|█████████ | 5833/6470 [1:12:29<07:12,  1.47it/s]
4|5834|Loss: 0.7565951347351074:  90%|█████████ | 5834/6470 [1:12:30<07:33,  1.40it/s]
4|5835|Loss: 0.8567479848861694:  90%|█████████ | 5834/6470 [1:12:30<07:33,  1.40it/s]
4|5835|Loss: 0.8567479848861694:  90%|█████████ | 5835/6470 [1:12:30<07:31,  1.41it/s]
4|5836|Loss: 0.8142457604408264:  90%|█████████ | 5835/6470 [1:12:31<07:31,  1.41it/s]
4|5836|Loss: 0.8142457604408264:  90%|█████████ | 5836/6470 [1:12:31<07:35,  1.39it/s]
4|5837|Loss: 0.7381671667098999:  90%|█████████ | 5836/6470 [1:12:32<07:35,  1.39it/s]
4|5837|Loss: 0.7381671667098999:  90%|█████████ | 5837/6470 [1:12:32<06:58,  1.51it/s]
4|5838|Loss: 0.8061519861221313:  90%|█████████ | 5837/6470 [1:12:32<06:58,  1.51it/s]
4|5838|Loss: 0.8061519861221313:  90%|█████████ | 5838/6470 [1:12:32<06:45,  1.56it/s]
4|5839|Loss: 0.8203849792480469:  90%|█████████ | 5838/6470 [1:12:33<06:45,  1.56it/s]
4|5839|Loss: 0.8203849792480469:  90%|█████████ | 5839/6470 [1:12:33<07:07,  1.48it/s]
4|5840|Loss: 0.8687729835510254:  90%|█████████ | 5839/6470 [1:12:34<07:07,  1.48it/s]
4|5840|Loss: 0.8687729835510254:  90%|█████████ | 5840/6470 [1:12:34<07:28,  1.40it/s]
4|5841|Loss: 0.8832314014434814:  90%|█████████ | 5840/6470 [1:12:34<07:28,  1.40it/s]
4|5841|Loss: 0.8832314014434814:  90%|█████████ | 5841/6470 [1:12:35<07:45,  1.35it/s]
4|5842|Loss: 0.8039506673812866:  90%|█████████ | 5841/6470 [1:12:35<07:45,  1.35it/s]
4|5842|Loss: 0.8039506673812866:  90%|█████████ | 5842/6470 [1:12:35<07:55,  1.32it/s]
4|5843|Loss: 0.9217215180397034:  90%|█████████ | 5842/6470 [1:12:36<07:55,  1.32it/s]
4|5843|Loss: 0.9217215180397034:  90%|█████████ | 5843/6470 [1:12:36<07:15,  1.44it/s]
4|5844|Loss: 0.7432146668434143:  90%|█████████ | 5843/6470 [1:12:36<07:15,  1.44it/s]
4|5844|Loss: 0.7432146668434143:  90%|█████████ | 5844/6470 [1:12:37<07:44,  1.35it/s]
4|5845|Loss: 0.7264885902404785:  90%|█████████ | 5844/6470 [1:12:37<07:44,  1.35it/s]
4|5845|Loss: 0.7264885902404785:  90%|█████████ | 5845/6470 [1:12:38<08:14,  1.26it/s]
4|5846|Loss: 0.9731863141059875:  90%|█████████ | 5845/6470 [1:12:38<08:14,  1.26it/s]
4|5846|Loss: 0.9731863141059875:  90%|█████████ | 5846/6470 [1:12:39<08:28,  1.23it/s]
4|5847|Loss: 0.8712219595909119:  90%|█████████ | 5846/6470 [1:12:39<08:28,  1.23it/s]
4|5847|Loss: 0.8712219595909119:  90%|█████████ | 5847/6470 [1:12:39<08:04,  1.29it/s]
4|5848|Loss: 0.7230959534645081:  90%|█████████ | 5847/6470 [1:12:40<08:04,  1.29it/s]
4|5848|Loss: 0.7230959534645081:  90%|█████████ | 5848/6470 [1:12:40<07:32,  1.37it/s]
4|5849|Loss: 0.8599286079406738:  90%|█████████ | 5848/6470 [1:12:40<07:32,  1.37it/s]
4|5849|Loss: 0.8599286079406738:  90%|█████████ | 5849/6470 [1:12:41<08:03,  1.28it/s]
4|5850|Loss: 0.8015095591545105:  90%|█████████ | 5849/6470 [1:12:41<08:03,  1.28it/s]
4|5850|Loss: 0.8015095591545105:  90%|█████████ | 5850/6470 [1:12:42<07:43,  1.34it/s]
4|5851|Loss: 0.8274760842323303:  90%|█████████ | 5850/6470 [1:12:42<07:43,  1.34it/s]
4|5851|Loss: 0.8274760842323303:  90%|█████████ | 5851/6470 [1:12:42<07:01,  1.47it/s]
4|5852|Loss: 0.7655625939369202:  90%|█████████ | 5851/6470 [1:12:42<07:01,  1.47it/s]
4|5852|Loss: 0.7655625939369202:  90%|█████████ | 5852/6470 [1:12:43<07:03,  1.46it/s]
4|5853|Loss: 0.8647329211235046:  90%|█████████ | 5852/6470 [1:12:43<07:03,  1.46it/s]
4|5853|Loss: 0.8647329211235046:  90%|█████████ | 5853/6470 [1:12:43<07:02,  1.46it/s]
4|5854|Loss: 0.8514404296875:  90%|█████████ | 5853/6470 [1:12:44<07:02,  1.46it/s]   
4|5854|Loss: 0.8514404296875:  90%|█████████ | 5854/6470 [1:12:44<07:18,  1.40it/s]
4|5855|Loss: 0.6954836249351501:  90%|█████████ | 5854/6470 [1:12:45<07:18,  1.40it/s]
4|5855|Loss: 0.6954836249351501:  90%|█████████ | 5855/6470 [1:12:45<07:51,  1.31it/s]
4|5856|Loss: 0.8267632126808167:  90%|█████████ | 5855/6470 [1:12:46<07:51,  1.31it/s]
4|5856|Loss: 0.8267632126808167:  91%|█████████ | 5856/6470 [1:12:46<08:09,  1.25it/s]
4|5857|Loss: 0.8612043857574463:  91%|█████████ | 5856/6470 [1:12:46<08:09,  1.25it/s]
4|5857|Loss: 0.8612043857574463:  91%|█████████ | 5857/6470 [1:12:47<08:32,  1.20it/s]
4|5858|Loss: 0.8704017400741577:  91%|█████████ | 5857/6470 [1:12:47<08:32,  1.20it/s]
4|5858|Loss: 0.8704017400741577:  91%|█████████ | 5858/6470 [1:12:48<08:26,  1.21it/s]
4|5859|Loss: 0.8189040422439575:  91%|█████████ | 5858/6470 [1:12:48<08:26,  1.21it/s]
4|5859|Loss: 0.8189040422439575:  91%|█████████ | 5859/6470 [1:12:48<08:16,  1.23it/s]
4|5860|Loss: 0.7627878189086914:  91%|█████████ | 5859/6470 [1:12:49<08:16,  1.23it/s]
4|5860|Loss: 0.7627878189086914:  91%|█████████ | 5860/6470 [1:12:49<07:31,  1.35it/s]
4|5861|Loss: 0.6392260789871216:  91%|█████████ | 5860/6470 [1:12:49<07:31,  1.35it/s]
4|5861|Loss: 0.6392260789871216:  91%|█████████ | 5861/6470 [1:12:50<06:42,  1.51it/s]
4|5862|Loss: 0.7760398387908936:  91%|█████████ | 5861/6470 [1:12:50<06:42,  1.51it/s]
4|5862|Loss: 0.7760398387908936:  91%|█████████ | 5862/6470 [1:12:50<06:52,  1.47it/s]
4|5863|Loss: 0.9948066473007202:  91%|█████████ | 5862/6470 [1:12:51<06:52,  1.47it/s]
4|5863|Loss: 0.9948066473007202:  91%|█████████ | 5863/6470 [1:12:51<07:07,  1.42it/s]
4|5864|Loss: 0.8695026636123657:  91%|█████████ | 5863/6470 [1:12:51<07:07,  1.42it/s]
4|5864|Loss: 0.8695026636123657:  91%|█████████ | 5864/6470 [1:12:52<06:59,  1.44it/s]
4|5865|Loss: 0.8532246947288513:  91%|█████████ | 5864/6470 [1:12:52<06:59,  1.44it/s]
4|5865|Loss: 0.8532246947288513:  91%|█████████ | 5865/6470 [1:12:52<06:55,  1.46it/s]
4|5866|Loss: 0.8597545027732849:  91%|█████████ | 5865/6470 [1:12:53<06:55,  1.46it/s]
4|5866|Loss: 0.8597545027732849:  91%|█████████ | 5866/6470 [1:12:53<07:27,  1.35it/s]
4|5867|Loss: 0.766456663608551:  91%|█████████ | 5866/6470 [1:12:54<07:27,  1.35it/s] 
4|5867|Loss: 0.766456663608551:  91%|█████████ | 5867/6470 [1:12:54<07:25,  1.35it/s]
4|5868|Loss: 0.8529341816902161:  91%|█████████ | 5867/6470 [1:12:54<07:25,  1.35it/s]
4|5868|Loss: 0.8529341816902161:  91%|█████████ | 5868/6470 [1:12:55<07:18,  1.37it/s]
4|5869|Loss: 0.7479137182235718:  91%|█████████ | 5868/6470 [1:12:55<07:18,  1.37it/s]
4|5869|Loss: 0.7479137182235718:  91%|█████████ | 5869/6470 [1:12:55<07:20,  1.37it/s]
4|5870|Loss: 0.8335624933242798:  91%|█████████ | 5869/6470 [1:12:56<07:20,  1.37it/s]
4|5870|Loss: 0.8335624933242798:  91%|█████████ | 5870/6470 [1:12:56<06:52,  1.46it/s]
4|5871|Loss: 0.85296231508255:  91%|█████████ | 5870/6470 [1:12:56<06:52,  1.46it/s]  
4|5871|Loss: 0.85296231508255:  91%|█████████ | 5871/6470 [1:12:57<06:55,  1.44it/s]
4|5872|Loss: 0.7455224990844727:  91%|█████████ | 5871/6470 [1:12:57<06:55,  1.44it/s]
4|5872|Loss: 0.7455224990844727:  91%|█████████ | 5872/6470 [1:12:57<07:05,  1.41it/s]
4|5873|Loss: 0.8141642808914185:  91%|█████████ | 5872/6470 [1:12:58<07:05,  1.41it/s]
4|5873|Loss: 0.8141642808914185:  91%|█████████ | 5873/6470 [1:12:58<07:27,  1.33it/s]
4|5874|Loss: 0.7355473041534424:  91%|█████████ | 5873/6470 [1:12:59<07:27,  1.33it/s]
4|5874|Loss: 0.7355473041534424:  91%|█████████ | 5874/6470 [1:12:59<07:39,  1.30it/s]
4|5875|Loss: 0.7925228476524353:  91%|█████████ | 5874/6470 [1:13:00<07:39,  1.30it/s]
4|5875|Loss: 0.7925228476524353:  91%|█████████ | 5875/6470 [1:13:00<07:25,  1.34it/s]
4|5876|Loss: 0.8214051127433777:  91%|█████████ | 5875/6470 [1:13:00<07:25,  1.34it/s]
4|5876|Loss: 0.8214051127433777:  91%|█████████ | 5876/6470 [1:13:01<07:40,  1.29it/s]
4|5877|Loss: 0.8701011538505554:  91%|█████████ | 5876/6470 [1:13:01<07:40,  1.29it/s]
4|5877|Loss: 0.8701011538505554:  91%|█████████ | 5877/6470 [1:13:01<07:39,  1.29it/s]
4|5878|Loss: 0.782538115978241:  91%|█████████ | 5877/6470 [1:13:02<07:39,  1.29it/s] 
4|5878|Loss: 0.782538115978241:  91%|█████████ | 5878/6470 [1:13:02<07:04,  1.39it/s]
4|5879|Loss: 0.9752406477928162:  91%|█████████ | 5878/6470 [1:13:02<07:04,  1.39it/s]
4|5879|Loss: 0.9752406477928162:  91%|█████████ | 5879/6470 [1:13:03<07:02,  1.40it/s]
4|5880|Loss: 0.7952650189399719:  91%|█████████ | 5879/6470 [1:13:03<07:02,  1.40it/s]
4|5880|Loss: 0.7952650189399719:  91%|█████████ | 5880/6470 [1:13:04<07:22,  1.33it/s]
4|5881|Loss: 0.8851078152656555:  91%|█████████ | 5880/6470 [1:13:04<07:22,  1.33it/s]
4|5881|Loss: 0.8851078152656555:  91%|█████████ | 5881/6470 [1:13:04<07:27,  1.32it/s]
4|5882|Loss: 0.7981323003768921:  91%|█████████ | 5881/6470 [1:13:05<07:27,  1.32it/s]
4|5882|Loss: 0.7981323003768921:  91%|█████████ | 5882/6470 [1:13:05<07:50,  1.25it/s]
4|5883|Loss: 0.7992489337921143:  91%|█████████ | 5882/6470 [1:13:06<07:50,  1.25it/s]
4|5883|Loss: 0.7992489337921143:  91%|█████████ | 5883/6470 [1:13:06<07:54,  1.24it/s]
4|5884|Loss: 0.8068690299987793:  91%|█████████ | 5883/6470 [1:13:07<07:54,  1.24it/s]
4|5884|Loss: 0.8068690299987793:  91%|█████████ | 5884/6470 [1:13:07<08:12,  1.19it/s]
4|5885|Loss: 0.927223265171051:  91%|█████████ | 5884/6470 [1:13:07<08:12,  1.19it/s] 
4|5885|Loss: 0.927223265171051:  91%|█████████ | 5885/6470 [1:13:08<07:45,  1.26it/s]
4|5886|Loss: 0.7841777205467224:  91%|█████████ | 5885/6470 [1:13:08<07:45,  1.26it/s]
4|5886|Loss: 0.7841777205467224:  91%|█████████ | 5886/6470 [1:13:08<07:18,  1.33it/s]
4|5887|Loss: 0.7488058805465698:  91%|█████████ | 5886/6470 [1:13:09<07:18,  1.33it/s]
4|5887|Loss: 0.7488058805465698:  91%|█████████ | 5887/6470 [1:13:09<06:56,  1.40it/s]
4|5888|Loss: 0.9357554912567139:  91%|█████████ | 5887/6470 [1:13:09<06:56,  1.40it/s]
4|5888|Loss: 0.9357554912567139:  91%|█████████ | 5888/6470 [1:13:10<07:06,  1.36it/s]
4|5889|Loss: 0.7342310547828674:  91%|█████████ | 5888/6470 [1:13:10<07:06,  1.36it/s]
4|5889|Loss: 0.7342310547828674:  91%|█████████ | 5889/6470 [1:13:10<07:11,  1.35it/s]
4|5890|Loss: 0.8620079755783081:  91%|█████████ | 5889/6470 [1:13:11<07:11,  1.35it/s]
4|5890|Loss: 0.8620079755783081:  91%|█████████ | 5890/6470 [1:13:11<06:51,  1.41it/s]
4|5891|Loss: 0.9664644002914429:  91%|█████████ | 5890/6470 [1:13:11<06:51,  1.41it/s]
4|5891|Loss: 0.9664644002914429:  91%|█████████ | 5891/6470 [1:13:12<06:39,  1.45it/s]
4|5892|Loss: 0.8080400228500366:  91%|█████████ | 5891/6470 [1:13:12<06:39,  1.45it/s]
4|5892|Loss: 0.8080400228500366:  91%|█████████ | 5892/6470 [1:13:12<06:27,  1.49it/s]
4|5893|Loss: 0.7738482356071472:  91%|█████████ | 5892/6470 [1:13:13<06:27,  1.49it/s]
4|5893|Loss: 0.7738482356071472:  91%|█████████ | 5893/6470 [1:13:13<07:06,  1.35it/s]
4|5894|Loss: 0.7218061089515686:  91%|█████████ | 5893/6470 [1:13:14<07:06,  1.35it/s]
4|5894|Loss: 0.7218061089515686:  91%|█████████ | 5894/6470 [1:13:14<07:38,  1.26it/s]
4|5895|Loss: 0.8136839270591736:  91%|█████████ | 5894/6470 [1:13:15<07:38,  1.26it/s]
4|5895|Loss: 0.8136839270591736:  91%|█████████ | 5895/6470 [1:13:15<06:47,  1.41it/s]
4|5896|Loss: 1.004120945930481:  91%|█████████ | 5895/6470 [1:13:15<06:47,  1.41it/s] 
4|5896|Loss: 1.004120945930481:  91%|█████████ | 5896/6470 [1:13:16<07:12,  1.33it/s]
4|5897|Loss: 0.809550404548645:  91%|█████████ | 5896/6470 [1:13:16<07:12,  1.33it/s]
4|5897|Loss: 0.809550404548645:  91%|█████████ | 5897/6470 [1:13:16<07:14,  1.32it/s]
4|5898|Loss: 0.7253256440162659:  91%|█████████ | 5897/6470 [1:13:17<07:14,  1.32it/s]
4|5898|Loss: 0.7253256440162659:  91%|█████████ | 5898/6470 [1:13:17<06:32,  1.46it/s]
4|5899|Loss: 0.835391640663147:  91%|█████████ | 5898/6470 [1:13:17<06:32,  1.46it/s] 
4|5899|Loss: 0.835391640663147:  91%|█████████ | 5899/6470 [1:13:17<06:09,  1.55it/s]
4|5900|Loss: 0.9454666376113892:  91%|█████████ | 5899/6470 [1:13:18<06:09,  1.55it/s]
4|5900|Loss: 0.9454666376113892:  91%|█████████ | 5900/6470 [1:13:18<06:12,  1.53it/s]
4|5901|Loss: 0.8196344375610352:  91%|█████████ | 5900/6470 [1:13:18<06:12,  1.53it/s]
4|5901|Loss: 0.8196344375610352:  91%|█████████ | 5901/6470 [1:13:19<06:20,  1.50it/s]
4|5902|Loss: 0.8719754219055176:  91%|█████████ | 5901/6470 [1:13:19<06:20,  1.50it/s]
4|5902|Loss: 0.8719754219055176:  91%|█████████ | 5902/6470 [1:13:20<06:55,  1.37it/s]
4|5903|Loss: 0.780997097492218:  91%|█████████ | 5902/6470 [1:13:20<06:55,  1.37it/s] 
4|5903|Loss: 0.780997097492218:  91%|█████████ | 5903/6470 [1:13:20<07:04,  1.33it/s]
4|5904|Loss: 0.7734454274177551:  91%|█████████ | 5903/6470 [1:13:21<07:04,  1.33it/s]
4|5904|Loss: 0.7734454274177551:  91%|█████████▏| 5904/6470 [1:13:21<07:10,  1.32it/s]
4|5905|Loss: 0.8179950714111328:  91%|█████████▏| 5904/6470 [1:13:22<07:10,  1.32it/s]
4|5905|Loss: 0.8179950714111328:  91%|█████████▏| 5905/6470 [1:13:22<07:09,  1.32it/s]
4|5906|Loss: 0.8697202205657959:  91%|█████████▏| 5905/6470 [1:13:22<07:09,  1.32it/s]
4|5906|Loss: 0.8697202205657959:  91%|█████████▏| 5906/6470 [1:13:23<07:28,  1.26it/s]
4|5907|Loss: 0.7427344918251038:  91%|█████████▏| 5906/6470 [1:13:23<07:28,  1.26it/s]
4|5907|Loss: 0.7427344918251038:  91%|█████████▏| 5907/6470 [1:13:24<07:54,  1.19it/s]
4|5908|Loss: 0.9745029211044312:  91%|█████████▏| 5907/6470 [1:13:24<07:54,  1.19it/s]
4|5908|Loss: 0.9745029211044312:  91%|█████████▏| 5908/6470 [1:13:25<08:08,  1.15it/s]
4|5909|Loss: 0.7978736162185669:  91%|█████████▏| 5908/6470 [1:13:25<08:08,  1.15it/s]
4|5909|Loss: 0.7978736162185669:  91%|█████████▏| 5909/6470 [1:13:26<08:00,  1.17it/s]
4|5910|Loss: 0.7162290811538696:  91%|█████████▏| 5909/6470 [1:13:26<08:00,  1.17it/s]
4|5910|Loss: 0.7162290811538696:  91%|█████████▏| 5910/6470 [1:13:26<07:37,  1.22it/s]
4|5911|Loss: 0.8653292655944824:  91%|█████████▏| 5910/6470 [1:13:27<07:37,  1.22it/s]
4|5911|Loss: 0.8653292655944824:  91%|█████████▏| 5911/6470 [1:13:27<07:38,  1.22it/s]
4|5912|Loss: 0.7086613774299622:  91%|█████████▏| 5911/6470 [1:13:28<07:38,  1.22it/s]
4|5912|Loss: 0.7086613774299622:  91%|█████████▏| 5912/6470 [1:13:28<07:02,  1.32it/s]
4|5913|Loss: 0.7631127238273621:  91%|█████████▏| 5912/6470 [1:13:28<07:02,  1.32it/s]
4|5913|Loss: 0.7631127238273621:  91%|█████████▏| 5913/6470 [1:13:28<06:59,  1.33it/s]
4|5914|Loss: 0.7818527221679688:  91%|█████████▏| 5913/6470 [1:13:29<06:59,  1.33it/s]
4|5914|Loss: 0.7818527221679688:  91%|█████████▏| 5914/6470 [1:13:29<07:06,  1.30it/s]
4|5915|Loss: 0.7895187139511108:  91%|█████████▏| 5914/6470 [1:13:30<07:06,  1.30it/s]
4|5915|Loss: 0.7895187139511108:  91%|█████████▏| 5915/6470 [1:13:30<07:08,  1.30it/s]
4|5916|Loss: 0.8431194424629211:  91%|█████████▏| 5915/6470 [1:13:31<07:08,  1.30it/s]
4|5916|Loss: 0.8431194424629211:  91%|█████████▏| 5916/6470 [1:13:31<07:12,  1.28it/s]
4|5917|Loss: 0.8294731974601746:  91%|█████████▏| 5916/6470 [1:13:31<07:12,  1.28it/s]
4|5917|Loss: 0.8294731974601746:  91%|█████████▏| 5917/6470 [1:13:31<06:41,  1.38it/s]
4|5918|Loss: 0.7653750777244568:  91%|█████████▏| 5917/6470 [1:13:32<06:41,  1.38it/s]
4|5918|Loss: 0.7653750777244568:  91%|█████████▏| 5918/6470 [1:13:32<06:45,  1.36it/s]
4|5919|Loss: 0.8727108240127563:  91%|█████████▏| 5918/6470 [1:13:33<06:45,  1.36it/s]
4|5919|Loss: 0.8727108240127563:  91%|█████████▏| 5919/6470 [1:13:33<07:18,  1.26it/s]
4|5920|Loss: 0.8619108200073242:  91%|█████████▏| 5919/6470 [1:13:34<07:18,  1.26it/s]
4|5920|Loss: 0.8619108200073242:  91%|█████████▏| 5920/6470 [1:13:34<06:58,  1.31it/s]
4|5921|Loss: 0.838659942150116:  91%|█████████▏| 5920/6470 [1:13:34<06:58,  1.31it/s] 
4|5921|Loss: 0.838659942150116:  92%|█████████▏| 5921/6470 [1:13:35<06:51,  1.33it/s]
4|5922|Loss: 0.6287156343460083:  92%|█████████▏| 5921/6470 [1:13:35<06:51,  1.33it/s]
4|5922|Loss: 0.6287156343460083:  92%|█████████▏| 5922/6470 [1:13:35<06:51,  1.33it/s]
4|5923|Loss: 1.0365537405014038:  92%|█████████▏| 5922/6470 [1:13:36<06:51,  1.33it/s]
4|5923|Loss: 1.0365537405014038:  92%|█████████▏| 5923/6470 [1:13:36<07:32,  1.21it/s]
4|5924|Loss: 0.773140549659729:  92%|█████████▏| 5923/6470 [1:13:37<07:32,  1.21it/s] 
4|5924|Loss: 0.773140549659729:  92%|█████████▏| 5924/6470 [1:13:37<07:24,  1.23it/s]
4|5925|Loss: 0.8003134727478027:  92%|█████████▏| 5924/6470 [1:13:38<07:24,  1.23it/s]
4|5925|Loss: 0.8003134727478027:  92%|█████████▏| 5925/6470 [1:13:38<07:01,  1.29it/s]
4|5926|Loss: 0.9182302951812744:  92%|█████████▏| 5925/6470 [1:13:38<07:01,  1.29it/s]
4|5926|Loss: 0.9182302951812744:  92%|█████████▏| 5926/6470 [1:13:38<06:38,  1.37it/s]
4|5927|Loss: 0.7628774642944336:  92%|█████████▏| 5926/6470 [1:13:39<06:38,  1.37it/s]
4|5927|Loss: 0.7628774642944336:  92%|█████████▏| 5927/6470 [1:13:39<06:41,  1.35it/s]
4|5928|Loss: 0.7600073218345642:  92%|█████████▏| 5927/6470 [1:13:40<06:41,  1.35it/s]
4|5928|Loss: 0.7600073218345642:  92%|█████████▏| 5928/6470 [1:13:40<06:36,  1.37it/s]
4|5929|Loss: 0.7620199918746948:  92%|█████████▏| 5928/6470 [1:13:40<06:36,  1.37it/s]
4|5929|Loss: 0.7620199918746948:  92%|█████████▏| 5929/6470 [1:13:41<07:01,  1.28it/s]
4|5930|Loss: 0.7641991972923279:  92%|█████████▏| 5929/6470 [1:13:41<07:01,  1.28it/s]
4|5930|Loss: 0.7641991972923279:  92%|█████████▏| 5930/6470 [1:13:42<07:08,  1.26it/s]
4|5931|Loss: 0.8095928430557251:  92%|█████████▏| 5930/6470 [1:13:42<07:08,  1.26it/s]
4|5931|Loss: 0.8095928430557251:  92%|█████████▏| 5931/6470 [1:13:43<07:24,  1.21it/s]
4|5932|Loss: 0.8864419460296631:  92%|█████████▏| 5931/6470 [1:13:43<07:24,  1.21it/s]
4|5932|Loss: 0.8864419460296631:  92%|█████████▏| 5932/6470 [1:13:43<07:34,  1.18it/s]
4|5933|Loss: 0.6638977527618408:  92%|█████████▏| 5932/6470 [1:13:44<07:34,  1.18it/s]
4|5933|Loss: 0.6638977527618408:  92%|█████████▏| 5933/6470 [1:13:44<07:26,  1.20it/s]
4|5934|Loss: 0.8790635466575623:  92%|█████████▏| 5933/6470 [1:13:45<07:26,  1.20it/s]
4|5934|Loss: 0.8790635466575623:  92%|█████████▏| 5934/6470 [1:13:45<07:21,  1.21it/s]
4|5935|Loss: 0.8824419975280762:  92%|█████████▏| 5934/6470 [1:13:46<07:21,  1.21it/s]
4|5935|Loss: 0.8824419975280762:  92%|█████████▏| 5935/6470 [1:13:46<07:43,  1.15it/s]
4|5936|Loss: 0.8295090198516846:  92%|█████████▏| 5935/6470 [1:13:47<07:43,  1.15it/s]
4|5936|Loss: 0.8295090198516846:  92%|█████████▏| 5936/6470 [1:13:47<07:57,  1.12it/s]
4|5937|Loss: 0.8144276738166809:  92%|█████████▏| 5936/6470 [1:13:47<07:57,  1.12it/s]
4|5937|Loss: 0.8144276738166809:  92%|█████████▏| 5937/6470 [1:13:48<07:33,  1.18it/s]
4|5938|Loss: 0.8955249190330505:  92%|█████████▏| 5937/6470 [1:13:48<07:33,  1.18it/s]
4|5938|Loss: 0.8955249190330505:  92%|█████████▏| 5938/6470 [1:13:49<07:36,  1.16it/s]
4|5939|Loss: 0.8403177857398987:  92%|█████████▏| 5938/6470 [1:13:49<07:36,  1.16it/s]
4|5939|Loss: 0.8403177857398987:  92%|█████████▏| 5939/6470 [1:13:49<07:29,  1.18it/s]
4|5940|Loss: 0.7702800631523132:  92%|█████████▏| 5939/6470 [1:13:50<07:29,  1.18it/s]
4|5940|Loss: 0.7702800631523132:  92%|█████████▏| 5940/6470 [1:13:50<07:23,  1.19it/s]
4|5941|Loss: 0.7847896218299866:  92%|█████████▏| 5940/6470 [1:13:51<07:23,  1.19it/s]
4|5941|Loss: 0.7847896218299866:  92%|█████████▏| 5941/6470 [1:13:51<07:02,  1.25it/s]
4|5942|Loss: 0.8598335385322571:  92%|█████████▏| 5941/6470 [1:13:51<07:02,  1.25it/s]
4|5942|Loss: 0.8598335385322571:  92%|█████████▏| 5942/6470 [1:13:51<06:04,  1.45it/s]
4|5943|Loss: 0.8115113377571106:  92%|█████████▏| 5942/6470 [1:13:52<06:04,  1.45it/s]
4|5943|Loss: 0.8115113377571106:  92%|█████████▏| 5943/6470 [1:13:52<06:03,  1.45it/s]
4|5944|Loss: 0.7299867868423462:  92%|█████████▏| 5943/6470 [1:13:52<06:03,  1.45it/s]
4|5944|Loss: 0.7299867868423462:  92%|█████████▏| 5944/6470 [1:13:53<05:29,  1.60it/s]
4|5945|Loss: 0.9297136664390564:  92%|█████████▏| 5944/6470 [1:13:53<05:29,  1.60it/s]
4|5945|Loss: 0.9297136664390564:  92%|█████████▏| 5945/6470 [1:13:53<05:43,  1.53it/s]
4|5946|Loss: 0.8194964528083801:  92%|█████████▏| 5945/6470 [1:13:54<05:43,  1.53it/s]
4|5946|Loss: 0.8194964528083801:  92%|█████████▏| 5946/6470 [1:13:54<06:16,  1.39it/s]
4|5947|Loss: 0.8907166123390198:  92%|█████████▏| 5946/6470 [1:13:55<06:16,  1.39it/s]
4|5947|Loss: 0.8907166123390198:  92%|█████████▏| 5947/6470 [1:13:55<06:43,  1.30it/s]
4|5948|Loss: 0.7979735136032104:  92%|█████████▏| 5947/6470 [1:13:55<06:43,  1.30it/s]
4|5948|Loss: 0.7979735136032104:  92%|█████████▏| 5948/6470 [1:13:56<06:44,  1.29it/s]
4|5949|Loss: 0.8729206919670105:  92%|█████████▏| 5948/6470 [1:13:56<06:44,  1.29it/s]
4|5949|Loss: 0.8729206919670105:  92%|█████████▏| 5949/6470 [1:13:56<06:24,  1.35it/s]
4|5950|Loss: 0.8626247048377991:  92%|█████████▏| 5949/6470 [1:13:57<06:24,  1.35it/s]
4|5950|Loss: 0.8626247048377991:  92%|█████████▏| 5950/6470 [1:13:57<06:29,  1.33it/s]
4|5951|Loss: 0.7159050703048706:  92%|█████████▏| 5950/6470 [1:13:58<06:29,  1.33it/s]
4|5951|Loss: 0.7159050703048706:  92%|█████████▏| 5951/6470 [1:13:58<06:33,  1.32it/s]
4|5952|Loss: 0.8141496777534485:  92%|█████████▏| 5951/6470 [1:13:58<06:33,  1.32it/s]
4|5952|Loss: 0.8141496777534485:  92%|█████████▏| 5952/6470 [1:13:59<06:33,  1.32it/s]
4|5953|Loss: 0.8764150142669678:  92%|█████████▏| 5952/6470 [1:13:59<06:33,  1.32it/s]
4|5953|Loss: 0.8764150142669678:  92%|█████████▏| 5953/6470 [1:14:00<06:32,  1.32it/s]
4|5954|Loss: 0.9067140817642212:  92%|█████████▏| 5953/6470 [1:14:00<06:32,  1.32it/s]
4|5954|Loss: 0.9067140817642212:  92%|█████████▏| 5954/6470 [1:14:00<06:36,  1.30it/s]
4|5955|Loss: 0.7166802287101746:  92%|█████████▏| 5954/6470 [1:14:01<06:36,  1.30it/s]
4|5955|Loss: 0.7166802287101746:  92%|█████████▏| 5955/6470 [1:14:01<06:41,  1.28it/s]
4|5956|Loss: 0.6622233986854553:  92%|█████████▏| 5955/6470 [1:14:01<06:41,  1.28it/s]
4|5956|Loss: 0.6622233986854553:  92%|█████████▏| 5956/6470 [1:14:02<06:12,  1.38it/s]
4|5957|Loss: 0.7693891525268555:  92%|█████████▏| 5956/6470 [1:14:02<06:12,  1.38it/s]
4|5957|Loss: 0.7693891525268555:  92%|█████████▏| 5957/6470 [1:14:02<06:15,  1.37it/s]
4|5958|Loss: 0.7743074297904968:  92%|█████████▏| 5957/6470 [1:14:03<06:15,  1.37it/s]
4|5958|Loss: 0.7743074297904968:  92%|█████████▏| 5958/6470 [1:14:03<05:49,  1.46it/s]
4|5959|Loss: 0.9365300536155701:  92%|█████████▏| 5958/6470 [1:14:03<05:49,  1.46it/s]
4|5959|Loss: 0.9365300536155701:  92%|█████████▏| 5959/6470 [1:14:04<05:54,  1.44it/s]
4|5960|Loss: 0.9416252374649048:  92%|█████████▏| 5959/6470 [1:14:04<05:54,  1.44it/s]
4|5960|Loss: 0.9416252374649048:  92%|█████████▏| 5960/6470 [1:14:04<06:04,  1.40it/s]
4|5961|Loss: 0.8984962105751038:  92%|█████████▏| 5960/6470 [1:14:05<06:04,  1.40it/s]
4|5961|Loss: 0.8984962105751038:  92%|█████████▏| 5961/6470 [1:14:05<06:16,  1.35it/s]
4|5962|Loss: 0.8020521998405457:  92%|█████████▏| 5961/6470 [1:14:06<06:16,  1.35it/s]
4|5962|Loss: 0.8020521998405457:  92%|█████████▏| 5962/6470 [1:14:06<06:27,  1.31it/s]
4|5963|Loss: 0.7469568252563477:  92%|█████████▏| 5962/6470 [1:14:07<06:27,  1.31it/s]
4|5963|Loss: 0.7469568252563477:  92%|█████████▏| 5963/6470 [1:14:07<06:41,  1.26it/s]
4|5964|Loss: 0.7435086369514465:  92%|█████████▏| 5963/6470 [1:14:07<06:41,  1.26it/s]
4|5964|Loss: 0.7435086369514465:  92%|█████████▏| 5964/6470 [1:14:08<06:44,  1.25it/s]
4|5965|Loss: 0.6887815594673157:  92%|█████████▏| 5964/6470 [1:14:08<06:44,  1.25it/s]
4|5965|Loss: 0.6887815594673157:  92%|█████████▏| 5965/6470 [1:14:09<06:39,  1.26it/s]
4|5966|Loss: 0.8181280493736267:  92%|█████████▏| 5965/6470 [1:14:09<06:39,  1.26it/s]
4|5966|Loss: 0.8181280493736267:  92%|█████████▏| 5966/6470 [1:14:09<06:14,  1.35it/s]
4|5967|Loss: 0.8433200716972351:  92%|█████████▏| 5966/6470 [1:14:10<06:14,  1.35it/s]
4|5967|Loss: 0.8433200716972351:  92%|█████████▏| 5967/6470 [1:14:10<06:04,  1.38it/s]
4|5968|Loss: 0.7873702645301819:  92%|█████████▏| 5967/6470 [1:14:10<06:04,  1.38it/s]
4|5968|Loss: 0.7873702645301819:  92%|█████████▏| 5968/6470 [1:14:11<06:00,  1.39it/s]
4|5969|Loss: 0.8222489953041077:  92%|█████████▏| 5968/6470 [1:14:11<06:00,  1.39it/s]
4|5969|Loss: 0.8222489953041077:  92%|█████████▏| 5969/6470 [1:14:11<06:03,  1.38it/s]
4|5970|Loss: 0.8102777004241943:  92%|█████████▏| 5969/6470 [1:14:12<06:03,  1.38it/s]
4|5970|Loss: 0.8102777004241943:  92%|█████████▏| 5970/6470 [1:14:12<05:57,  1.40it/s]
4|5971|Loss: 0.8508160710334778:  92%|█████████▏| 5970/6470 [1:14:12<05:57,  1.40it/s]
4|5971|Loss: 0.8508160710334778:  92%|█████████▏| 5971/6470 [1:14:13<05:50,  1.43it/s]
4|5972|Loss: 0.9600330591201782:  92%|█████████▏| 5971/6470 [1:14:13<05:50,  1.43it/s]
4|5972|Loss: 0.9600330591201782:  92%|█████████▏| 5972/6470 [1:14:13<05:53,  1.41it/s]
4|5973|Loss: 0.8669688105583191:  92%|█████████▏| 5972/6470 [1:14:14<05:53,  1.41it/s]
4|5973|Loss: 0.8669688105583191:  92%|█████████▏| 5973/6470 [1:14:14<06:02,  1.37it/s]
4|5974|Loss: 0.7784584164619446:  92%|█████████▏| 5973/6470 [1:14:15<06:02,  1.37it/s]
4|5974|Loss: 0.7784584164619446:  92%|█████████▏| 5974/6470 [1:14:15<06:06,  1.35it/s]
4|5975|Loss: 0.9818070530891418:  92%|█████████▏| 5974/6470 [1:14:15<06:06,  1.35it/s]
4|5975|Loss: 0.9818070530891418:  92%|█████████▏| 5975/6470 [1:14:16<05:52,  1.40it/s]
4|5976|Loss: 0.7329924702644348:  92%|█████████▏| 5975/6470 [1:14:16<05:52,  1.40it/s]
4|5976|Loss: 0.7329924702644348:  92%|█████████▏| 5976/6470 [1:14:16<05:56,  1.38it/s]
4|5977|Loss: 0.9223049879074097:  92%|█████████▏| 5976/6470 [1:14:17<05:56,  1.38it/s]
4|5977|Loss: 0.9223049879074097:  92%|█████████▏| 5977/6470 [1:14:17<06:23,  1.29it/s]
4|5978|Loss: 0.8289693593978882:  92%|█████████▏| 5977/6470 [1:14:18<06:23,  1.29it/s]
4|5978|Loss: 0.8289693593978882:  92%|█████████▏| 5978/6470 [1:14:18<05:58,  1.37it/s]
4|5979|Loss: 0.8798498511314392:  92%|█████████▏| 5978/6470 [1:14:18<05:58,  1.37it/s]
4|5979|Loss: 0.8798498511314392:  92%|█████████▏| 5979/6470 [1:14:18<05:42,  1.43it/s]
4|5980|Loss: 0.8450003266334534:  92%|█████████▏| 5979/6470 [1:14:19<05:42,  1.43it/s]
4|5980|Loss: 0.8450003266334534:  92%|█████████▏| 5980/6470 [1:14:19<06:13,  1.31it/s]
4|5981|Loss: 0.8624788522720337:  92%|█████████▏| 5980/6470 [1:14:20<06:13,  1.31it/s]
4|5981|Loss: 0.8624788522720337:  92%|█████████▏| 5981/6470 [1:14:20<05:57,  1.37it/s]
4|5982|Loss: 0.7850346565246582:  92%|█████████▏| 5981/6470 [1:14:20<05:57,  1.37it/s]
4|5982|Loss: 0.7850346565246582:  92%|█████████▏| 5982/6470 [1:14:21<05:42,  1.43it/s]
4|5983|Loss: 0.7058939337730408:  92%|█████████▏| 5982/6470 [1:14:21<05:42,  1.43it/s]
4|5983|Loss: 0.7058939337730408:  92%|█████████▏| 5983/6470 [1:14:21<05:48,  1.40it/s]
4|5984|Loss: 0.7486557960510254:  92%|█████████▏| 5983/6470 [1:14:22<05:48,  1.40it/s]
4|5984|Loss: 0.7486557960510254:  92%|█████████▏| 5984/6470 [1:14:22<05:30,  1.47it/s]
4|5985|Loss: 0.7146169543266296:  92%|█████████▏| 5984/6470 [1:14:22<05:30,  1.47it/s]
4|5985|Loss: 0.7146169543266296:  93%|█████████▎| 5985/6470 [1:14:23<05:48,  1.39it/s]
4|5986|Loss: 0.880725622177124:  93%|█████████▎| 5985/6470 [1:14:23<05:48,  1.39it/s] 
4|5986|Loss: 0.880725622177124:  93%|█████████▎| 5986/6470 [1:14:24<05:54,  1.37it/s]
4|5987|Loss: 0.6605271100997925:  93%|█████████▎| 5986/6470 [1:14:24<05:54,  1.37it/s]
4|5987|Loss: 0.6605271100997925:  93%|█████████▎| 5987/6470 [1:14:24<05:31,  1.46it/s]
4|5988|Loss: 0.7902820706367493:  93%|█████████▎| 5987/6470 [1:14:25<05:31,  1.46it/s]
4|5988|Loss: 0.7902820706367493:  93%|█████████▎| 5988/6470 [1:14:25<05:26,  1.48it/s]
4|5989|Loss: 0.9757190942764282:  93%|█████████▎| 5988/6470 [1:14:25<05:26,  1.48it/s]
4|5989|Loss: 0.9757190942764282:  93%|█████████▎| 5989/6470 [1:14:26<05:41,  1.41it/s]
4|5990|Loss: 0.7771661281585693:  93%|█████████▎| 5989/6470 [1:14:26<05:41,  1.41it/s]
4|5990|Loss: 0.7771661281585693:  93%|█████████▎| 5990/6470 [1:14:26<05:49,  1.37it/s]
4|5991|Loss: 0.7139108180999756:  93%|█████████▎| 5990/6470 [1:14:27<05:49,  1.37it/s]
4|5991|Loss: 0.7139108180999756:  93%|█████████▎| 5991/6470 [1:14:27<05:54,  1.35it/s]
4|5992|Loss: 0.6788257360458374:  93%|█████████▎| 5991/6470 [1:14:28<05:54,  1.35it/s]
4|5992|Loss: 0.6788257360458374:  93%|█████████▎| 5992/6470 [1:14:28<06:07,  1.30it/s]
4|5993|Loss: 0.9208943247795105:  93%|█████████▎| 5992/6470 [1:14:28<06:07,  1.30it/s]
4|5993|Loss: 0.9208943247795105:  93%|█████████▎| 5993/6470 [1:14:29<06:11,  1.28it/s]
4|5994|Loss: 0.8868855237960815:  93%|█████████▎| 5993/6470 [1:14:29<06:11,  1.28it/s]
4|5994|Loss: 0.8868855237960815:  93%|█████████▎| 5994/6470 [1:14:30<06:11,  1.28it/s]
4|5995|Loss: 0.8334982395172119:  93%|█████████▎| 5994/6470 [1:14:30<06:11,  1.28it/s]
4|5995|Loss: 0.8334982395172119:  93%|█████████▎| 5995/6470 [1:14:30<06:13,  1.27it/s]
4|5996|Loss: 0.7779265642166138:  93%|█████████▎| 5995/6470 [1:14:31<06:13,  1.27it/s]
4|5996|Loss: 0.7779265642166138:  93%|█████████▎| 5996/6470 [1:14:31<06:09,  1.28it/s]
4|5997|Loss: 0.968654990196228:  93%|█████████▎| 5996/6470 [1:14:32<06:09,  1.28it/s] 
4|5997|Loss: 0.968654990196228:  93%|█████████▎| 5997/6470 [1:14:32<05:51,  1.35it/s]
4|5998|Loss: 0.9615574479103088:  93%|█████████▎| 5997/6470 [1:14:32<05:51,  1.35it/s]
4|5998|Loss: 0.9615574479103088:  93%|█████████▎| 5998/6470 [1:14:32<05:41,  1.38it/s]
4|5999|Loss: 0.7470594048500061:  93%|█████████▎| 5998/6470 [1:14:33<05:41,  1.38it/s]
4|5999|Loss: 0.7470594048500061:  93%|█████████▎| 5999/6470 [1:14:33<05:37,  1.40it/s]
4|6000|Loss: 0.8900275826454163:  93%|█████████▎| 5999/6470 [1:14:34<05:37,  1.40it/s]
4|6000|Loss: 0.8900275826454163:  93%|█████████▎| 6000/6470 [1:14:34<05:55,  1.32it/s]
4|6001|Loss: 0.7683919668197632:  93%|█████████▎| 6000/6470 [1:14:34<05:55,  1.32it/s]
4|6001|Loss: 0.7683919668197632:  93%|█████████▎| 6001/6470 [1:14:35<05:33,  1.41it/s]
4|6002|Loss: 0.7121186852455139:  93%|█████████▎| 6001/6470 [1:14:35<05:33,  1.41it/s]
4|6002|Loss: 0.7121186852455139:  93%|█████████▎| 6002/6470 [1:14:35<05:49,  1.34it/s]
4|6003|Loss: 0.6870651245117188:  93%|█████████▎| 6002/6470 [1:14:36<05:49,  1.34it/s]
4|6003|Loss: 0.6870651245117188:  93%|█████████▎| 6003/6470 [1:14:36<05:58,  1.30it/s]
4|6004|Loss: 1.1161264181137085:  93%|█████████▎| 6003/6470 [1:14:37<05:58,  1.30it/s]
4|6004|Loss: 1.1161264181137085:  93%|█████████▎| 6004/6470 [1:14:37<05:43,  1.36it/s]
4|6005|Loss: 0.7007532119750977:  93%|█████████▎| 6004/6470 [1:14:37<05:43,  1.36it/s]
4|6005|Loss: 0.7007532119750977:  93%|█████████▎| 6005/6470 [1:14:38<05:46,  1.34it/s]
4|6006|Loss: 0.824476420879364:  93%|█████████▎| 6005/6470 [1:14:38<05:46,  1.34it/s] 
4|6006|Loss: 0.824476420879364:  93%|█████████▎| 6006/6470 [1:14:38<05:36,  1.38it/s]
4|6007|Loss: 0.8159631490707397:  93%|█████████▎| 6006/6470 [1:14:39<05:36,  1.38it/s]
4|6007|Loss: 0.8159631490707397:  93%|█████████▎| 6007/6470 [1:14:39<05:41,  1.35it/s]
4|6008|Loss: 0.7854810357093811:  93%|█████████▎| 6007/6470 [1:14:40<05:41,  1.35it/s]
4|6008|Loss: 0.7854810357093811:  93%|█████████▎| 6008/6470 [1:14:40<05:31,  1.40it/s]
4|6009|Loss: 0.8777773976325989:  93%|█████████▎| 6008/6470 [1:14:40<05:31,  1.40it/s]
4|6009|Loss: 0.8777773976325989:  93%|█████████▎| 6009/6470 [1:14:41<05:36,  1.37it/s]
4|6010|Loss: 1.0295802354812622:  93%|█████████▎| 6009/6470 [1:14:41<05:36,  1.37it/s]
4|6010|Loss: 1.0295802354812622:  93%|█████████▎| 6010/6470 [1:14:41<05:34,  1.37it/s]
4|6011|Loss: 0.8100711107254028:  93%|█████████▎| 6010/6470 [1:14:42<05:34,  1.37it/s]
4|6011|Loss: 0.8100711107254028:  93%|█████████▎| 6011/6470 [1:14:42<05:41,  1.35it/s]
4|6012|Loss: 0.8589375019073486:  93%|█████████▎| 6011/6470 [1:14:43<05:41,  1.35it/s]
4|6012|Loss: 0.8589375019073486:  93%|█████████▎| 6012/6470 [1:14:43<06:02,  1.26it/s]
4|6013|Loss: 0.7394523620605469:  93%|█████████▎| 6012/6470 [1:14:43<06:02,  1.26it/s]
4|6013|Loss: 0.7394523620605469:  93%|█████████▎| 6013/6470 [1:14:44<06:03,  1.26it/s]
4|6014|Loss: 0.8528447151184082:  93%|█████████▎| 6013/6470 [1:14:44<06:03,  1.26it/s]
4|6014|Loss: 0.8528447151184082:  93%|█████████▎| 6014/6470 [1:14:44<05:36,  1.36it/s]
4|6015|Loss: 0.849719226360321:  93%|█████████▎| 6014/6470 [1:14:45<05:36,  1.36it/s] 
4|6015|Loss: 0.849719226360321:  93%|█████████▎| 6015/6470 [1:14:45<05:21,  1.41it/s]
4|6016|Loss: 0.8182994723320007:  93%|█████████▎| 6015/6470 [1:14:45<05:21,  1.41it/s]
4|6016|Loss: 0.8182994723320007:  93%|█████████▎| 6016/6470 [1:14:46<05:17,  1.43it/s]
4|6017|Loss: 0.8186072707176208:  93%|█████████▎| 6016/6470 [1:14:46<05:17,  1.43it/s]
4|6017|Loss: 0.8186072707176208:  93%|█████████▎| 6017/6470 [1:14:46<04:52,  1.55it/s]
4|6018|Loss: 0.796977162361145:  93%|█████████▎| 6017/6470 [1:14:47<04:52,  1.55it/s] 
4|6018|Loss: 0.796977162361145:  93%|█████████▎| 6018/6470 [1:14:47<04:35,  1.64it/s]
4|6019|Loss: 0.8332134485244751:  93%|█████████▎| 6018/6470 [1:14:47<04:35,  1.64it/s]
4|6019|Loss: 0.8332134485244751:  93%|█████████▎| 6019/6470 [1:14:48<05:11,  1.45it/s]
4|6020|Loss: 0.9085922241210938:  93%|█████████▎| 6019/6470 [1:14:48<05:11,  1.45it/s]
4|6020|Loss: 0.9085922241210938:  93%|█████████▎| 6020/6470 [1:14:48<05:25,  1.38it/s]
4|6021|Loss: 0.777082085609436:  93%|█████████▎| 6020/6470 [1:14:49<05:25,  1.38it/s] 
4|6021|Loss: 0.777082085609436:  93%|█████████▎| 6021/6470 [1:14:49<05:20,  1.40it/s]
4|6022|Loss: 0.8376312255859375:  93%|█████████▎| 6021/6470 [1:14:50<05:20,  1.40it/s]
4|6022|Loss: 0.8376312255859375:  93%|█████████▎| 6022/6470 [1:14:50<05:27,  1.37it/s]
4|6023|Loss: 0.727451741695404:  93%|█████████▎| 6022/6470 [1:14:50<05:27,  1.37it/s] 
4|6023|Loss: 0.727451741695404:  93%|█████████▎| 6023/6470 [1:14:50<05:04,  1.47it/s]
4|6024|Loss: 0.8125239610671997:  93%|█████████▎| 6023/6470 [1:14:51<05:04,  1.47it/s]
4|6024|Loss: 0.8125239610671997:  93%|█████████▎| 6024/6470 [1:14:51<05:24,  1.37it/s]
4|6025|Loss: 0.7975368499755859:  93%|█████████▎| 6024/6470 [1:14:52<05:24,  1.37it/s]
4|6025|Loss: 0.7975368499755859:  93%|█████████▎| 6025/6470 [1:14:52<05:36,  1.32it/s]
4|6026|Loss: 0.9524903893470764:  93%|█████████▎| 6025/6470 [1:14:53<05:36,  1.32it/s]
4|6026|Loss: 0.9524903893470764:  93%|█████████▎| 6026/6470 [1:14:53<05:29,  1.35it/s]
4|6027|Loss: 0.7979528903961182:  93%|█████████▎| 6026/6470 [1:14:53<05:29,  1.35it/s]
4|6027|Loss: 0.7979528903961182:  93%|█████████▎| 6027/6470 [1:14:53<05:15,  1.40it/s]
4|6028|Loss: 0.7526968717575073:  93%|█████████▎| 6027/6470 [1:14:54<05:15,  1.40it/s]
4|6028|Loss: 0.7526968717575073:  93%|█████████▎| 6028/6470 [1:14:54<04:58,  1.48it/s]
4|6029|Loss: 0.6957095861434937:  93%|█████████▎| 6028/6470 [1:14:54<04:58,  1.48it/s]
4|6029|Loss: 0.6957095861434937:  93%|█████████▎| 6029/6470 [1:14:55<04:34,  1.60it/s]
4|6030|Loss: 0.8637545108795166:  93%|█████████▎| 6029/6470 [1:14:55<04:34,  1.60it/s]
4|6030|Loss: 0.8637545108795166:  93%|█████████▎| 6030/6470 [1:14:55<04:59,  1.47it/s]
4|6031|Loss: 0.8915944695472717:  93%|█████████▎| 6030/6470 [1:14:56<04:59,  1.47it/s]
4|6031|Loss: 0.8915944695472717:  93%|█████████▎| 6031/6470 [1:14:56<05:07,  1.43it/s]
4|6032|Loss: 1.0199263095855713:  93%|█████████▎| 6031/6470 [1:14:57<05:07,  1.43it/s]
4|6032|Loss: 1.0199263095855713:  93%|█████████▎| 6032/6470 [1:14:57<05:27,  1.34it/s]
4|6033|Loss: 0.7225398421287537:  93%|█████████▎| 6032/6470 [1:14:57<05:27,  1.34it/s]
4|6033|Loss: 0.7225398421287537:  93%|█████████▎| 6033/6470 [1:14:58<05:16,  1.38it/s]
4|6034|Loss: 0.9350610375404358:  93%|█████████▎| 6033/6470 [1:14:58<05:16,  1.38it/s]
4|6034|Loss: 0.9350610375404358:  93%|█████████▎| 6034/6470 [1:14:58<05:10,  1.40it/s]
4|6035|Loss: 0.8084772825241089:  93%|█████████▎| 6034/6470 [1:14:59<05:10,  1.40it/s]
4|6035|Loss: 0.8084772825241089:  93%|█████████▎| 6035/6470 [1:14:59<05:19,  1.36it/s]
4|6036|Loss: 0.6296350359916687:  93%|█████████▎| 6035/6470 [1:15:00<05:19,  1.36it/s]
4|6036|Loss: 0.6296350359916687:  93%|█████████▎| 6036/6470 [1:15:00<05:28,  1.32it/s]
4|6037|Loss: 0.8526716828346252:  93%|█████████▎| 6036/6470 [1:15:00<05:28,  1.32it/s]
4|6037|Loss: 0.8526716828346252:  93%|█████████▎| 6037/6470 [1:15:01<05:53,  1.22it/s]
4|6038|Loss: 0.7586251497268677:  93%|█████████▎| 6037/6470 [1:15:01<05:53,  1.22it/s]
4|6038|Loss: 0.7586251497268677:  93%|█████████▎| 6038/6470 [1:15:02<05:56,  1.21it/s]
4|6039|Loss: 0.8757706880569458:  93%|█████████▎| 6038/6470 [1:15:02<05:56,  1.21it/s]
4|6039|Loss: 0.8757706880569458:  93%|█████████▎| 6039/6470 [1:15:03<05:51,  1.23it/s]
4|6040|Loss: 0.7941890358924866:  93%|█████████▎| 6039/6470 [1:15:03<05:51,  1.23it/s]
4|6040|Loss: 0.7941890358924866:  93%|█████████▎| 6040/6470 [1:15:03<05:52,  1.22it/s]
4|6041|Loss: 0.7782980799674988:  93%|█████████▎| 6040/6470 [1:15:04<05:52,  1.22it/s]
4|6041|Loss: 0.7782980799674988:  93%|█████████▎| 6041/6470 [1:15:04<05:40,  1.26it/s]
4|6042|Loss: 0.8232359290122986:  93%|█████████▎| 6041/6470 [1:15:05<05:40,  1.26it/s]
4|6042|Loss: 0.8232359290122986:  93%|█████████▎| 6042/6470 [1:15:05<05:40,  1.26it/s]
4|6043|Loss: 0.9448030591011047:  93%|█████████▎| 6042/6470 [1:15:05<05:40,  1.26it/s]
4|6043|Loss: 0.9448030591011047:  93%|█████████▎| 6043/6470 [1:15:06<05:29,  1.30it/s]
4|6044|Loss: 0.7783554196357727:  93%|█████████▎| 6043/6470 [1:15:06<05:29,  1.30it/s]
4|6044|Loss: 0.7783554196357727:  93%|█████████▎| 6044/6470 [1:15:06<05:32,  1.28it/s]
4|6045|Loss: 0.917076587677002:  93%|█████████▎| 6044/6470 [1:15:07<05:32,  1.28it/s] 
4|6045|Loss: 0.917076587677002:  93%|█████████▎| 6045/6470 [1:15:07<05:25,  1.31it/s]
4|6046|Loss: 0.8080240488052368:  93%|█████████▎| 6045/6470 [1:15:08<05:25,  1.31it/s]
4|6046|Loss: 0.8080240488052368:  93%|█████████▎| 6046/6470 [1:15:08<05:35,  1.26it/s]
4|6047|Loss: 0.8182174563407898:  93%|█████████▎| 6046/6470 [1:15:08<05:35,  1.26it/s]
4|6047|Loss: 0.8182174563407898:  93%|█████████▎| 6047/6470 [1:15:09<05:32,  1.27it/s]
4|6048|Loss: 0.9205372929573059:  93%|█████████▎| 6047/6470 [1:15:09<05:32,  1.27it/s]
4|6048|Loss: 0.9205372929573059:  93%|█████████▎| 6048/6470 [1:15:10<05:32,  1.27it/s]
4|6049|Loss: 0.884374737739563:  93%|█████████▎| 6048/6470 [1:15:10<05:32,  1.27it/s] 
4|6049|Loss: 0.884374737739563:  93%|█████████▎| 6049/6470 [1:15:10<05:23,  1.30it/s]
4|6050|Loss: 0.8717520236968994:  93%|█████████▎| 6049/6470 [1:15:11<05:23,  1.30it/s]
4|6050|Loss: 0.8717520236968994:  94%|█████████▎| 6050/6470 [1:15:11<05:30,  1.27it/s]
4|6051|Loss: 0.7357304692268372:  94%|█████████▎| 6050/6470 [1:15:12<05:30,  1.27it/s]
4|6051|Loss: 0.7357304692268372:  94%|█████████▎| 6051/6470 [1:15:12<05:50,  1.20it/s]
4|6052|Loss: 0.7208090424537659:  94%|█████████▎| 6051/6470 [1:15:13<05:50,  1.20it/s]
4|6052|Loss: 0.7208090424537659:  94%|█████████▎| 6052/6470 [1:15:13<05:41,  1.23it/s]
4|6053|Loss: 0.8278765678405762:  94%|█████████▎| 6052/6470 [1:15:13<05:41,  1.23it/s]
4|6053|Loss: 0.8278765678405762:  94%|█████████▎| 6053/6470 [1:15:14<05:27,  1.27it/s]
4|6054|Loss: 0.7478881478309631:  94%|█████████▎| 6053/6470 [1:15:14<05:27,  1.27it/s]
4|6054|Loss: 0.7478881478309631:  94%|█████████▎| 6054/6470 [1:15:15<05:53,  1.18it/s]
4|6055|Loss: 0.9634667038917542:  94%|█████████▎| 6054/6470 [1:15:15<05:53,  1.18it/s]
4|6055|Loss: 0.9634667038917542:  94%|█████████▎| 6055/6470 [1:15:15<05:42,  1.21it/s]
4|6056|Loss: 0.7863205671310425:  94%|█████████▎| 6055/6470 [1:15:16<05:42,  1.21it/s]
4|6056|Loss: 0.7863205671310425:  94%|█████████▎| 6056/6470 [1:15:16<05:37,  1.23it/s]
4|6057|Loss: 0.8044189810752869:  94%|█████████▎| 6056/6470 [1:15:17<05:37,  1.23it/s]
4|6057|Loss: 0.8044189810752869:  94%|█████████▎| 6057/6470 [1:15:17<05:33,  1.24it/s]
4|6058|Loss: 0.8264737725257874:  94%|█████████▎| 6057/6470 [1:15:17<05:33,  1.24it/s]
4|6058|Loss: 0.8264737725257874:  94%|█████████▎| 6058/6470 [1:15:17<05:00,  1.37it/s]
4|6059|Loss: 0.8952142000198364:  94%|█████████▎| 6058/6470 [1:15:18<05:00,  1.37it/s]
4|6059|Loss: 0.8952142000198364:  94%|█████████▎| 6059/6470 [1:15:18<04:41,  1.46it/s]
4|6060|Loss: 0.7745593786239624:  94%|█████████▎| 6059/6470 [1:15:18<04:41,  1.46it/s]
4|6060|Loss: 0.7745593786239624:  94%|█████████▎| 6060/6470 [1:15:19<04:32,  1.51it/s]
4|6061|Loss: 0.7467649579048157:  94%|█████████▎| 6060/6470 [1:15:19<04:32,  1.51it/s]
4|6061|Loss: 0.7467649579048157:  94%|█████████▎| 6061/6470 [1:15:19<04:36,  1.48it/s]
4|6062|Loss: 0.7394089102745056:  94%|█████████▎| 6061/6470 [1:15:20<04:36,  1.48it/s]
4|6062|Loss: 0.7394089102745056:  94%|█████████▎| 6062/6470 [1:15:20<05:00,  1.36it/s]
4|6063|Loss: 0.9349038004875183:  94%|█████████▎| 6062/6470 [1:15:21<05:00,  1.36it/s]
4|6063|Loss: 0.9349038004875183:  94%|█████████▎| 6063/6470 [1:15:21<05:10,  1.31it/s]
4|6064|Loss: 0.8142805099487305:  94%|█████████▎| 6063/6470 [1:15:21<05:10,  1.31it/s]
4|6064|Loss: 0.8142805099487305:  94%|█████████▎| 6064/6470 [1:15:22<05:08,  1.31it/s]
4|6065|Loss: 0.7415974736213684:  94%|█████████▎| 6064/6470 [1:15:22<05:08,  1.31it/s]
4|6065|Loss: 0.7415974736213684:  94%|█████████▎| 6065/6470 [1:15:23<05:21,  1.26it/s]
4|6066|Loss: 0.8651083707809448:  94%|█████████▎| 6065/6470 [1:15:23<05:21,  1.26it/s]
4|6066|Loss: 0.8651083707809448:  94%|█████████▍| 6066/6470 [1:15:23<05:22,  1.25it/s]
4|6067|Loss: 0.811759352684021:  94%|█████████▍| 6066/6470 [1:15:24<05:22,  1.25it/s] 
4|6067|Loss: 0.811759352684021:  94%|█████████▍| 6067/6470 [1:15:24<05:18,  1.27it/s]
4|6068|Loss: 0.885540783405304:  94%|█████████▍| 6067/6470 [1:15:25<05:18,  1.27it/s]
4|6068|Loss: 0.885540783405304:  94%|█████████▍| 6068/6470 [1:15:25<05:24,  1.24it/s]
4|6069|Loss: 0.9153726100921631:  94%|█████████▍| 6068/6470 [1:15:25<05:24,  1.24it/s]
4|6069|Loss: 0.9153726100921631:  94%|█████████▍| 6069/6470 [1:15:26<04:59,  1.34it/s]
4|6070|Loss: 0.868076503276825:  94%|█████████▍| 6069/6470 [1:15:26<04:59,  1.34it/s] 
4|6070|Loss: 0.868076503276825:  94%|█████████▍| 6070/6470 [1:15:26<04:47,  1.39it/s]
4|6071|Loss: 1.0643659830093384:  94%|█████████▍| 6070/6470 [1:15:27<04:47,  1.39it/s]
4|6071|Loss: 1.0643659830093384:  94%|█████████▍| 6071/6470 [1:15:27<04:45,  1.40it/s]
4|6072|Loss: 0.6952871680259705:  94%|█████████▍| 6071/6470 [1:15:28<04:45,  1.40it/s]
4|6072|Loss: 0.6952871680259705:  94%|█████████▍| 6072/6470 [1:15:28<04:53,  1.36it/s]
4|6073|Loss: 0.8465486168861389:  94%|█████████▍| 6072/6470 [1:15:28<04:53,  1.36it/s]
4|6073|Loss: 0.8465486168861389:  94%|█████████▍| 6073/6470 [1:15:29<04:48,  1.37it/s]
4|6074|Loss: 0.7077277302742004:  94%|█████████▍| 6073/6470 [1:15:29<04:48,  1.37it/s]
4|6074|Loss: 0.7077277302742004:  94%|█████████▍| 6074/6470 [1:15:29<04:55,  1.34it/s]
4|6075|Loss: 0.7950347661972046:  94%|█████████▍| 6074/6470 [1:15:30<04:55,  1.34it/s]
4|6075|Loss: 0.7950347661972046:  94%|█████████▍| 6075/6470 [1:15:30<05:01,  1.31it/s]
4|6076|Loss: 0.8374969959259033:  94%|█████████▍| 6075/6470 [1:15:31<05:01,  1.31it/s]
4|6076|Loss: 0.8374969959259033:  94%|█████████▍| 6076/6470 [1:15:31<04:56,  1.33it/s]
4|6077|Loss: 0.7354258298873901:  94%|█████████▍| 6076/6470 [1:15:31<04:56,  1.33it/s]
4|6077|Loss: 0.7354258298873901:  94%|█████████▍| 6077/6470 [1:15:32<04:53,  1.34it/s]
4|6078|Loss: 0.7552215456962585:  94%|█████████▍| 6077/6470 [1:15:32<04:53,  1.34it/s]
4|6078|Loss: 0.7552215456962585:  94%|█████████▍| 6078/6470 [1:15:33<05:13,  1.25it/s]
4|6079|Loss: 0.6580426692962646:  94%|█████████▍| 6078/6470 [1:15:33<05:13,  1.25it/s]
4|6079|Loss: 0.6580426692962646:  94%|█████████▍| 6079/6470 [1:15:34<05:34,  1.17it/s]
4|6080|Loss: 0.7221421599388123:  94%|█████████▍| 6079/6470 [1:15:34<05:34,  1.17it/s]
4|6080|Loss: 0.7221421599388123:  94%|█████████▍| 6080/6470 [1:15:34<05:10,  1.26it/s]
4|6081|Loss: 0.9048263430595398:  94%|█████████▍| 6080/6470 [1:15:35<05:10,  1.26it/s]
4|6081|Loss: 0.9048263430595398:  94%|█████████▍| 6081/6470 [1:15:35<04:51,  1.34it/s]
4|6082|Loss: 0.774422824382782:  94%|█████████▍| 6081/6470 [1:15:35<04:51,  1.34it/s] 
4|6082|Loss: 0.774422824382782:  94%|█████████▍| 6082/6470 [1:15:36<05:09,  1.25it/s]
4|6083|Loss: 0.8455649018287659:  94%|█████████▍| 6082/6470 [1:15:36<05:09,  1.25it/s]
4|6083|Loss: 0.8455649018287659:  94%|█████████▍| 6083/6470 [1:15:37<05:14,  1.23it/s]
4|6084|Loss: 0.7681528329849243:  94%|█████████▍| 6083/6470 [1:15:37<05:14,  1.23it/s]
4|6084|Loss: 0.7681528329849243:  94%|█████████▍| 6084/6470 [1:15:37<05:25,  1.19it/s]
4|6085|Loss: 0.9149094820022583:  94%|█████████▍| 6084/6470 [1:15:38<05:25,  1.19it/s]
4|6085|Loss: 0.9149094820022583:  94%|█████████▍| 6085/6470 [1:15:38<04:58,  1.29it/s]
4|6086|Loss: 0.8757609128952026:  94%|█████████▍| 6085/6470 [1:15:39<04:58,  1.29it/s]
4|6086|Loss: 0.8757609128952026:  94%|█████████▍| 6086/6470 [1:15:39<05:21,  1.19it/s]
4|6087|Loss: 0.7339859008789062:  94%|█████████▍| 6086/6470 [1:15:40<05:21,  1.19it/s]
4|6087|Loss: 0.7339859008789062:  94%|█████████▍| 6087/6470 [1:15:40<05:23,  1.19it/s]
4|6088|Loss: 0.781410276889801:  94%|█████████▍| 6087/6470 [1:15:40<05:23,  1.19it/s] 
4|6088|Loss: 0.781410276889801:  94%|█████████▍| 6088/6470 [1:15:41<05:18,  1.20it/s]
4|6089|Loss: 0.752831757068634:  94%|█████████▍| 6088/6470 [1:15:41<05:18,  1.20it/s]
4|6089|Loss: 0.752831757068634:  94%|█████████▍| 6089/6470 [1:15:42<05:13,  1.22it/s]
4|6090|Loss: 0.8790251612663269:  94%|█████████▍| 6089/6470 [1:15:42<05:13,  1.22it/s]
4|6090|Loss: 0.8790251612663269:  94%|█████████▍| 6090/6470 [1:15:42<05:04,  1.25it/s]
4|6091|Loss: 0.7032958269119263:  94%|█████████▍| 6090/6470 [1:15:43<05:04,  1.25it/s]
4|6091|Loss: 0.7032958269119263:  94%|█████████▍| 6091/6470 [1:15:43<04:32,  1.39it/s]
4|6092|Loss: 0.6517594456672668:  94%|█████████▍| 6091/6470 [1:15:43<04:32,  1.39it/s]
4|6092|Loss: 0.6517594456672668:  94%|█████████▍| 6092/6470 [1:15:44<04:27,  1.42it/s]
4|6093|Loss: 0.7667456865310669:  94%|█████████▍| 6092/6470 [1:15:44<04:27,  1.42it/s]
4|6093|Loss: 0.7667456865310669:  94%|█████████▍| 6093/6470 [1:15:44<04:27,  1.41it/s]
4|6094|Loss: 0.7002953886985779:  94%|█████████▍| 6093/6470 [1:15:45<04:27,  1.41it/s]
4|6094|Loss: 0.7002953886985779:  94%|█████████▍| 6094/6470 [1:15:45<04:26,  1.41it/s]
4|6095|Loss: 0.9644074440002441:  94%|█████████▍| 6094/6470 [1:15:45<04:26,  1.41it/s]
4|6095|Loss: 0.9644074440002441:  94%|█████████▍| 6095/6470 [1:15:46<04:45,  1.31it/s]
4|6096|Loss: 0.7559841871261597:  94%|█████████▍| 6095/6470 [1:15:46<04:45,  1.31it/s]
4|6096|Loss: 0.7559841871261597:  94%|█████████▍| 6096/6470 [1:15:47<04:42,  1.32it/s]
4|6097|Loss: 0.7641138434410095:  94%|█████████▍| 6096/6470 [1:15:47<04:42,  1.32it/s]
4|6097|Loss: 0.7641138434410095:  94%|█████████▍| 6097/6470 [1:15:47<04:44,  1.31it/s]
4|6098|Loss: 0.868695080280304:  94%|█████████▍| 6097/6470 [1:15:48<04:44,  1.31it/s] 
4|6098|Loss: 0.868695080280304:  94%|█████████▍| 6098/6470 [1:15:48<04:38,  1.34it/s]
4|6099|Loss: 0.7993493676185608:  94%|█████████▍| 6098/6470 [1:15:48<04:38,  1.34it/s]
4|6099|Loss: 0.7993493676185608:  94%|█████████▍| 6099/6470 [1:15:49<04:24,  1.40it/s]
4|6100|Loss: 0.8591319918632507:  94%|█████████▍| 6099/6470 [1:15:49<04:24,  1.40it/s]
4|6100|Loss: 0.8591319918632507:  94%|█████████▍| 6100/6470 [1:15:49<04:13,  1.46it/s]
4|6101|Loss: 0.8811322450637817:  94%|█████████▍| 6100/6470 [1:15:50<04:13,  1.46it/s]
4|6101|Loss: 0.8811322450637817:  94%|█████████▍| 6101/6470 [1:15:50<04:22,  1.41it/s]
4|6102|Loss: 0.8713635206222534:  94%|█████████▍| 6101/6470 [1:15:51<04:22,  1.41it/s]
4|6102|Loss: 0.8713635206222534:  94%|█████████▍| 6102/6470 [1:15:51<04:41,  1.31it/s]
4|6103|Loss: 0.8376642465591431:  94%|█████████▍| 6102/6470 [1:15:51<04:41,  1.31it/s]
4|6103|Loss: 0.8376642465591431:  94%|█████████▍| 6103/6470 [1:15:52<04:35,  1.33it/s]
4|6104|Loss: 0.7244600057601929:  94%|█████████▍| 6103/6470 [1:15:52<04:35,  1.33it/s]
4|6104|Loss: 0.7244600057601929:  94%|█████████▍| 6104/6470 [1:15:52<04:23,  1.39it/s]
4|6105|Loss: 0.7758494019508362:  94%|█████████▍| 6104/6470 [1:15:53<04:23,  1.39it/s]
4|6105|Loss: 0.7758494019508362:  94%|█████████▍| 6105/6470 [1:15:53<04:18,  1.41it/s]
4|6106|Loss: 0.9937190413475037:  94%|█████████▍| 6105/6470 [1:15:53<04:18,  1.41it/s]
4|6106|Loss: 0.9937190413475037:  94%|█████████▍| 6106/6470 [1:15:54<04:10,  1.45it/s]
4|6107|Loss: 0.8787239193916321:  94%|█████████▍| 6106/6470 [1:15:54<04:10,  1.45it/s]
4|6107|Loss: 0.8787239193916321:  94%|█████████▍| 6107/6470 [1:15:54<04:18,  1.41it/s]
4|6108|Loss: 0.9086787700653076:  94%|█████████▍| 6107/6470 [1:15:55<04:18,  1.41it/s]
4|6108|Loss: 0.9086787700653076:  94%|█████████▍| 6108/6470 [1:15:55<04:09,  1.45it/s]
4|6109|Loss: 0.7803277373313904:  94%|█████████▍| 6108/6470 [1:15:55<04:09,  1.45it/s]
4|6109|Loss: 0.7803277373313904:  94%|█████████▍| 6109/6470 [1:15:56<04:12,  1.43it/s]
4|6110|Loss: 0.7553391456604004:  94%|█████████▍| 6109/6470 [1:15:56<04:12,  1.43it/s]
4|6110|Loss: 0.7553391456604004:  94%|█████████▍| 6110/6470 [1:15:56<04:06,  1.46it/s]
4|6111|Loss: 0.8566866517066956:  94%|█████████▍| 6110/6470 [1:15:57<04:06,  1.46it/s]
4|6111|Loss: 0.8566866517066956:  94%|█████████▍| 6111/6470 [1:15:57<04:05,  1.46it/s]
4|6112|Loss: 0.994769275188446:  94%|█████████▍| 6111/6470 [1:15:58<04:05,  1.46it/s] 
4|6112|Loss: 0.994769275188446:  94%|█████████▍| 6112/6470 [1:15:58<04:06,  1.45it/s]
4|6113|Loss: 0.7673205137252808:  94%|█████████▍| 6112/6470 [1:15:58<04:06,  1.45it/s]
4|6113|Loss: 0.7673205137252808:  94%|█████████▍| 6113/6470 [1:15:59<04:21,  1.37it/s]
4|6114|Loss: 0.9410778880119324:  94%|█████████▍| 6113/6470 [1:15:59<04:21,  1.37it/s]
4|6114|Loss: 0.9410778880119324:  94%|█████████▍| 6114/6470 [1:15:59<04:11,  1.41it/s]
4|6115|Loss: 0.7294927835464478:  94%|█████████▍| 6114/6470 [1:16:00<04:11,  1.41it/s]
4|6115|Loss: 0.7294927835464478:  95%|█████████▍| 6115/6470 [1:16:00<04:23,  1.35it/s]
4|6116|Loss: 0.730364978313446:  95%|█████████▍| 6115/6470 [1:16:01<04:23,  1.35it/s] 
4|6116|Loss: 0.730364978313446:  95%|█████████▍| 6116/6470 [1:16:01<04:16,  1.38it/s]
4|6117|Loss: 0.6862426400184631:  95%|█████████▍| 6116/6470 [1:16:01<04:16,  1.38it/s]
4|6117|Loss: 0.6862426400184631:  95%|█████████▍| 6117/6470 [1:16:02<04:18,  1.36it/s]
4|6118|Loss: 0.7868554592132568:  95%|█████████▍| 6117/6470 [1:16:02<04:18,  1.36it/s]
4|6118|Loss: 0.7868554592132568:  95%|█████████▍| 6118/6470 [1:16:02<04:15,  1.38it/s]
4|6119|Loss: 0.898975670337677:  95%|█████████▍| 6118/6470 [1:16:03<04:15,  1.38it/s] 
4|6119|Loss: 0.898975670337677:  95%|█████████▍| 6119/6470 [1:16:03<04:20,  1.35it/s]
4|6120|Loss: 0.8105999231338501:  95%|█████████▍| 6119/6470 [1:16:03<04:20,  1.35it/s]
4|6120|Loss: 0.8105999231338501:  95%|█████████▍| 6120/6470 [1:16:04<04:22,  1.33it/s]
4|6121|Loss: 0.8563924431800842:  95%|█████████▍| 6120/6470 [1:16:04<04:22,  1.33it/s]
4|6121|Loss: 0.8563924431800842:  95%|█████████▍| 6121/6470 [1:16:04<04:12,  1.38it/s]
4|6122|Loss: 0.747188150882721:  95%|█████████▍| 6121/6470 [1:16:05<04:12,  1.38it/s] 
4|6122|Loss: 0.747188150882721:  95%|█████████▍| 6122/6470 [1:16:05<04:15,  1.36it/s]
4|6123|Loss: 0.8211479783058167:  95%|█████████▍| 6122/6470 [1:16:06<04:15,  1.36it/s]
4|6123|Loss: 0.8211479783058167:  95%|█████████▍| 6123/6470 [1:16:06<04:10,  1.39it/s]
4|6124|Loss: 0.7957231998443604:  95%|█████████▍| 6123/6470 [1:16:06<04:10,  1.39it/s]
4|6124|Loss: 0.7957231998443604:  95%|█████████▍| 6124/6470 [1:16:06<03:52,  1.49it/s]
4|6125|Loss: 0.7796419858932495:  95%|█████████▍| 6124/6470 [1:16:07<03:52,  1.49it/s]
4|6125|Loss: 0.7796419858932495:  95%|█████████▍| 6125/6470 [1:16:07<04:10,  1.38it/s]
4|6126|Loss: 0.7852454781532288:  95%|█████████▍| 6125/6470 [1:16:08<04:10,  1.38it/s]
4|6126|Loss: 0.7852454781532288:  95%|█████████▍| 6126/6470 [1:16:08<04:21,  1.32it/s]
4|6127|Loss: 0.8219723701477051:  95%|█████████▍| 6126/6470 [1:16:09<04:21,  1.32it/s]
4|6127|Loss: 0.8219723701477051:  95%|█████████▍| 6127/6470 [1:16:09<04:25,  1.29it/s]
4|6128|Loss: 0.8353914618492126:  95%|█████████▍| 6127/6470 [1:16:09<04:25,  1.29it/s]
4|6128|Loss: 0.8353914618492126:  95%|█████████▍| 6128/6470 [1:16:10<04:44,  1.20it/s]
4|6129|Loss: 0.7780950665473938:  95%|█████████▍| 6128/6470 [1:16:10<04:44,  1.20it/s]
4|6129|Loss: 0.7780950665473938:  95%|█████████▍| 6129/6470 [1:16:11<04:31,  1.26it/s]
4|6130|Loss: 0.8332496881484985:  95%|█████████▍| 6129/6470 [1:16:11<04:31,  1.26it/s]
4|6130|Loss: 0.8332496881484985:  95%|█████████▍| 6130/6470 [1:16:12<04:37,  1.23it/s]
4|6131|Loss: 0.6308789849281311:  95%|█████████▍| 6130/6470 [1:16:12<04:37,  1.23it/s]
4|6131|Loss: 0.6308789849281311:  95%|█████████▍| 6131/6470 [1:16:12<04:44,  1.19it/s]
4|6132|Loss: 0.6044759750366211:  95%|█████████▍| 6131/6470 [1:16:13<04:44,  1.19it/s]
4|6132|Loss: 0.6044759750366211:  95%|█████████▍| 6132/6470 [1:16:13<04:07,  1.37it/s]
4|6133|Loss: 0.7686235904693604:  95%|█████████▍| 6132/6470 [1:16:13<04:07,  1.37it/s]
4|6133|Loss: 0.7686235904693604:  95%|█████████▍| 6133/6470 [1:16:14<03:53,  1.44it/s]
4|6134|Loss: 0.6960447430610657:  95%|█████████▍| 6133/6470 [1:16:14<03:53,  1.44it/s]
4|6134|Loss: 0.6960447430610657:  95%|█████████▍| 6134/6470 [1:16:14<03:43,  1.50it/s]
4|6135|Loss: 0.7693140506744385:  95%|█████████▍| 6134/6470 [1:16:14<03:43,  1.50it/s]
4|6135|Loss: 0.7693140506744385:  95%|█████████▍| 6135/6470 [1:16:15<03:44,  1.49it/s]
4|6136|Loss: 0.7539608478546143:  95%|█████████▍| 6135/6470 [1:16:15<03:44,  1.49it/s]
4|6136|Loss: 0.7539608478546143:  95%|█████████▍| 6136/6470 [1:16:15<03:44,  1.49it/s]
4|6137|Loss: 0.7973363399505615:  95%|█████████▍| 6136/6470 [1:16:16<03:44,  1.49it/s]
4|6137|Loss: 0.7973363399505615:  95%|█████████▍| 6137/6470 [1:16:16<03:58,  1.40it/s]
4|6138|Loss: 0.7802804708480835:  95%|█████████▍| 6137/6470 [1:16:17<03:58,  1.40it/s]
4|6138|Loss: 0.7802804708480835:  95%|█████████▍| 6138/6470 [1:16:17<04:00,  1.38it/s]
4|6139|Loss: 0.8286920785903931:  95%|█████████▍| 6138/6470 [1:16:17<04:00,  1.38it/s]
4|6139|Loss: 0.8286920785903931:  95%|█████████▍| 6139/6470 [1:16:18<03:56,  1.40it/s]
4|6140|Loss: 0.7023388743400574:  95%|█████████▍| 6139/6470 [1:16:18<03:56,  1.40it/s]
4|6140|Loss: 0.7023388743400574:  95%|█████████▍| 6140/6470 [1:16:18<03:53,  1.41it/s]
4|6141|Loss: 0.7792021632194519:  95%|█████████▍| 6140/6470 [1:16:19<03:53,  1.41it/s]
4|6141|Loss: 0.7792021632194519:  95%|█████████▍| 6141/6470 [1:16:19<03:56,  1.39it/s]
4|6142|Loss: 0.8885128498077393:  95%|█████████▍| 6141/6470 [1:16:20<03:56,  1.39it/s]
4|6142|Loss: 0.8885128498077393:  95%|█████████▍| 6142/6470 [1:16:20<03:49,  1.43it/s]
4|6143|Loss: 0.7801629900932312:  95%|█████████▍| 6142/6470 [1:16:20<03:49,  1.43it/s]
4|6143|Loss: 0.7801629900932312:  95%|█████████▍| 6143/6470 [1:16:21<04:05,  1.33it/s]
4|6144|Loss: 0.8129847645759583:  95%|█████████▍| 6143/6470 [1:16:21<04:05,  1.33it/s]
4|6144|Loss: 0.8129847645759583:  95%|█████████▍| 6144/6470 [1:16:22<04:16,  1.27it/s]
4|6145|Loss: 0.8823632597923279:  95%|█████████▍| 6144/6470 [1:16:22<04:16,  1.27it/s]
4|6145|Loss: 0.8823632597923279:  95%|█████████▍| 6145/6470 [1:16:22<04:18,  1.26it/s]
4|6146|Loss: 0.7403267621994019:  95%|█████████▍| 6145/6470 [1:16:23<04:18,  1.26it/s]
4|6146|Loss: 0.7403267621994019:  95%|█████████▍| 6146/6470 [1:16:23<04:16,  1.26it/s]
4|6147|Loss: 0.8221349120140076:  95%|█████████▍| 6146/6470 [1:16:24<04:16,  1.26it/s]
4|6147|Loss: 0.8221349120140076:  95%|█████████▌| 6147/6470 [1:16:24<04:17,  1.26it/s]
4|6148|Loss: 0.8783125877380371:  95%|█████████▌| 6147/6470 [1:16:24<04:17,  1.26it/s]
4|6148|Loss: 0.8783125877380371:  95%|█████████▌| 6148/6470 [1:16:25<04:32,  1.18it/s]
4|6149|Loss: 0.8154522776603699:  95%|█████████▌| 6148/6470 [1:16:25<04:32,  1.18it/s]
4|6149|Loss: 0.8154522776603699:  95%|█████████▌| 6149/6470 [1:16:26<04:21,  1.23it/s]
4|6150|Loss: 0.7820046544075012:  95%|█████████▌| 6149/6470 [1:16:26<04:21,  1.23it/s]
4|6150|Loss: 0.7820046544075012:  95%|█████████▌| 6150/6470 [1:16:26<04:06,  1.30it/s]
4|6151|Loss: 0.9380196928977966:  95%|█████████▌| 6150/6470 [1:16:27<04:06,  1.30it/s]
4|6151|Loss: 0.9380196928977966:  95%|█████████▌| 6151/6470 [1:16:27<04:05,  1.30it/s]
4|6152|Loss: 0.9107454419136047:  95%|█████████▌| 6151/6470 [1:16:28<04:05,  1.30it/s]
4|6152|Loss: 0.9107454419136047:  95%|█████████▌| 6152/6470 [1:16:28<04:09,  1.27it/s]
4|6153|Loss: 0.850335955619812:  95%|█████████▌| 6152/6470 [1:16:28<04:09,  1.27it/s] 
4|6153|Loss: 0.850335955619812:  95%|█████████▌| 6153/6470 [1:16:29<04:27,  1.19it/s]
4|6154|Loss: 0.833236813545227:  95%|█████████▌| 6153/6470 [1:16:29<04:27,  1.19it/s]
4|6154|Loss: 0.833236813545227:  95%|█████████▌| 6154/6470 [1:16:30<04:25,  1.19it/s]
4|6155|Loss: 0.8263424038887024:  95%|█████████▌| 6154/6470 [1:16:30<04:25,  1.19it/s]
4|6155|Loss: 0.8263424038887024:  95%|█████████▌| 6155/6470 [1:16:31<04:25,  1.19it/s]
4|6156|Loss: 0.7395117878913879:  95%|█████████▌| 6155/6470 [1:16:31<04:25,  1.19it/s]
4|6156|Loss: 0.7395117878913879:  95%|█████████▌| 6156/6470 [1:16:31<04:18,  1.22it/s]
4|6157|Loss: 0.9034982919692993:  95%|█████████▌| 6156/6470 [1:16:32<04:18,  1.22it/s]
4|6157|Loss: 0.9034982919692993:  95%|█████████▌| 6157/6470 [1:16:32<03:56,  1.33it/s]
4|6158|Loss: 0.8118323683738708:  95%|█████████▌| 6157/6470 [1:16:32<03:56,  1.33it/s]
4|6158|Loss: 0.8118323683738708:  95%|█████████▌| 6158/6470 [1:16:33<03:54,  1.33it/s]
4|6159|Loss: 0.9635632634162903:  95%|█████████▌| 6158/6470 [1:16:33<03:54,  1.33it/s]
4|6159|Loss: 0.9635632634162903:  95%|█████████▌| 6159/6470 [1:16:33<03:43,  1.39it/s]
4|6160|Loss: 0.8189325332641602:  95%|█████████▌| 6159/6470 [1:16:34<03:43,  1.39it/s]
4|6160|Loss: 0.8189325332641602:  95%|█████████▌| 6160/6470 [1:16:34<03:46,  1.37it/s]
4|6161|Loss: 0.7415153384208679:  95%|█████████▌| 6160/6470 [1:16:35<03:46,  1.37it/s]
4|6161|Loss: 0.7415153384208679:  95%|█████████▌| 6161/6470 [1:16:35<03:50,  1.34it/s]
4|6162|Loss: 0.7660258412361145:  95%|█████████▌| 6161/6470 [1:16:35<03:50,  1.34it/s]
4|6162|Loss: 0.7660258412361145:  95%|█████████▌| 6162/6470 [1:16:36<04:01,  1.27it/s]
4|6163|Loss: 0.8444658517837524:  95%|█████████▌| 6162/6470 [1:16:36<04:01,  1.27it/s]
4|6163|Loss: 0.8444658517837524:  95%|█████████▌| 6163/6470 [1:16:37<04:12,  1.21it/s]
4|6164|Loss: 0.8038421273231506:  95%|█████████▌| 6163/6470 [1:16:37<04:12,  1.21it/s]
4|6164|Loss: 0.8038421273231506:  95%|█████████▌| 6164/6470 [1:16:37<04:02,  1.26it/s]
4|6165|Loss: 0.8528417944908142:  95%|█████████▌| 6164/6470 [1:16:38<04:02,  1.26it/s]
4|6165|Loss: 0.8528417944908142:  95%|█████████▌| 6165/6470 [1:16:38<04:06,  1.24it/s]
4|6166|Loss: 0.7725104093551636:  95%|█████████▌| 6165/6470 [1:16:39<04:06,  1.24it/s]
4|6166|Loss: 0.7725104093551636:  95%|█████████▌| 6166/6470 [1:16:39<03:53,  1.30it/s]
4|6167|Loss: 0.9134188890457153:  95%|█████████▌| 6166/6470 [1:16:39<03:53,  1.30it/s]
4|6167|Loss: 0.9134188890457153:  95%|█████████▌| 6167/6470 [1:16:40<03:38,  1.39it/s]
4|6168|Loss: 0.8532626628875732:  95%|█████████▌| 6167/6470 [1:16:40<03:38,  1.39it/s]
4|6168|Loss: 0.8532626628875732:  95%|█████████▌| 6168/6470 [1:16:40<03:48,  1.32it/s]
4|6169|Loss: 0.7854154706001282:  95%|█████████▌| 6168/6470 [1:16:41<03:48,  1.32it/s]
4|6169|Loss: 0.7854154706001282:  95%|█████████▌| 6169/6470 [1:16:41<03:53,  1.29it/s]
4|6170|Loss: 0.7557613253593445:  95%|█████████▌| 6169/6470 [1:16:42<03:53,  1.29it/s]
4|6170|Loss: 0.7557613253593445:  95%|█████████▌| 6170/6470 [1:16:42<03:43,  1.34it/s]
4|6171|Loss: 0.8145753145217896:  95%|█████████▌| 6170/6470 [1:16:42<03:43,  1.34it/s]
4|6171|Loss: 0.8145753145217896:  95%|█████████▌| 6171/6470 [1:16:42<03:34,  1.39it/s]
4|6172|Loss: 0.8410805463790894:  95%|█████████▌| 6171/6470 [1:16:43<03:34,  1.39it/s]
4|6172|Loss: 0.8410805463790894:  95%|█████████▌| 6172/6470 [1:16:43<03:31,  1.41it/s]
4|6173|Loss: 0.7556130290031433:  95%|█████████▌| 6172/6470 [1:16:44<03:31,  1.41it/s]
4|6173|Loss: 0.7556130290031433:  95%|█████████▌| 6173/6470 [1:16:44<03:34,  1.39it/s]
4|6174|Loss: 0.7717398405075073:  95%|█████████▌| 6173/6470 [1:16:44<03:34,  1.39it/s]
4|6174|Loss: 0.7717398405075073:  95%|█████████▌| 6174/6470 [1:16:45<03:20,  1.48it/s]
4|6175|Loss: 0.8773084282875061:  95%|█████████▌| 6174/6470 [1:16:45<03:20,  1.48it/s]
4|6175|Loss: 0.8773084282875061:  95%|█████████▌| 6175/6470 [1:16:45<03:18,  1.49it/s]
4|6176|Loss: 0.8603039383888245:  95%|█████████▌| 6175/6470 [1:16:46<03:18,  1.49it/s]
4|6176|Loss: 0.8603039383888245:  95%|█████████▌| 6176/6470 [1:16:46<03:14,  1.51it/s]
4|6177|Loss: 0.9195724725723267:  95%|█████████▌| 6176/6470 [1:16:46<03:14,  1.51it/s]
4|6177|Loss: 0.9195724725723267:  95%|█████████▌| 6177/6470 [1:16:46<03:13,  1.51it/s]
4|6178|Loss: 0.5618218183517456:  95%|█████████▌| 6177/6470 [1:16:47<03:13,  1.51it/s]
4|6178|Loss: 0.5618218183517456:  95%|█████████▌| 6178/6470 [1:16:47<03:19,  1.47it/s]
4|6179|Loss: 0.930088222026825:  95%|█████████▌| 6178/6470 [1:16:48<03:19,  1.47it/s] 
4|6179|Loss: 0.930088222026825:  96%|█████████▌| 6179/6470 [1:16:48<03:24,  1.42it/s]
4|6180|Loss: 0.7863285541534424:  96%|█████████▌| 6179/6470 [1:16:48<03:24,  1.42it/s]
4|6180|Loss: 0.7863285541534424:  96%|█████████▌| 6180/6470 [1:16:49<03:25,  1.41it/s]
4|6181|Loss: 0.8789127469062805:  96%|█████████▌| 6180/6470 [1:16:49<03:25,  1.41it/s]
4|6181|Loss: 0.8789127469062805:  96%|█████████▌| 6181/6470 [1:16:49<03:28,  1.39it/s]
4|6182|Loss: 0.8854063153266907:  96%|█████████▌| 6181/6470 [1:16:50<03:28,  1.39it/s]
4|6182|Loss: 0.8854063153266907:  96%|█████████▌| 6182/6470 [1:16:50<03:39,  1.31it/s]
4|6183|Loss: 0.8113254904747009:  96%|█████████▌| 6182/6470 [1:16:51<03:39,  1.31it/s]
4|6183|Loss: 0.8113254904747009:  96%|█████████▌| 6183/6470 [1:16:51<03:41,  1.29it/s]
4|6184|Loss: 0.8922097086906433:  96%|█████████▌| 6183/6470 [1:16:52<03:41,  1.29it/s]
4|6184|Loss: 0.8922097086906433:  96%|█████████▌| 6184/6470 [1:16:52<03:52,  1.23it/s]
4|6185|Loss: 0.9011362791061401:  96%|█████████▌| 6184/6470 [1:16:52<03:52,  1.23it/s]
4|6185|Loss: 0.9011362791061401:  96%|█████████▌| 6185/6470 [1:16:53<03:49,  1.24it/s]
4|6186|Loss: 0.7737038731575012:  96%|█████████▌| 6185/6470 [1:16:53<03:49,  1.24it/s]
4|6186|Loss: 0.7737038731575012:  96%|█████████▌| 6186/6470 [1:16:54<03:43,  1.27it/s]
4|6187|Loss: 0.8906382918357849:  96%|█████████▌| 6186/6470 [1:16:54<03:43,  1.27it/s]
4|6187|Loss: 0.8906382918357849:  96%|█████████▌| 6187/6470 [1:16:54<03:42,  1.27it/s]
4|6188|Loss: 0.7448496222496033:  96%|█████████▌| 6187/6470 [1:16:55<03:42,  1.27it/s]
4|6188|Loss: 0.7448496222496033:  96%|█████████▌| 6188/6470 [1:16:55<03:41,  1.27it/s]
4|6189|Loss: 0.6895580291748047:  96%|█████████▌| 6188/6470 [1:16:56<03:41,  1.27it/s]
4|6189|Loss: 0.6895580291748047:  96%|█████████▌| 6189/6470 [1:16:56<03:46,  1.24it/s]
4|6190|Loss: 0.8138673901557922:  96%|█████████▌| 6189/6470 [1:16:56<03:46,  1.24it/s]
4|6190|Loss: 0.8138673901557922:  96%|█████████▌| 6190/6470 [1:16:57<03:39,  1.27it/s]
4|6191|Loss: 0.7101722359657288:  96%|█████████▌| 6190/6470 [1:16:57<03:39,  1.27it/s]
4|6191|Loss: 0.7101722359657288:  96%|█████████▌| 6191/6470 [1:16:57<03:35,  1.30it/s]
4|6192|Loss: 0.8420901298522949:  96%|█████████▌| 6191/6470 [1:16:58<03:35,  1.30it/s]
4|6192|Loss: 0.8420901298522949:  96%|█████████▌| 6192/6470 [1:16:58<03:33,  1.30it/s]
4|6193|Loss: 0.936547040939331:  96%|█████████▌| 6192/6470 [1:16:59<03:33,  1.30it/s] 
4|6193|Loss: 0.936547040939331:  96%|█████████▌| 6193/6470 [1:16:59<03:44,  1.23it/s]
4|6194|Loss: 0.6822462677955627:  96%|█████████▌| 6193/6470 [1:17:00<03:44,  1.23it/s]
4|6194|Loss: 0.6822462677955627:  96%|█████████▌| 6194/6470 [1:17:00<03:42,  1.24it/s]
4|6195|Loss: 0.8925655484199524:  96%|█████████▌| 6194/6470 [1:17:00<03:42,  1.24it/s]
4|6195|Loss: 0.8925655484199524:  96%|█████████▌| 6195/6470 [1:17:01<03:37,  1.26it/s]
4|6196|Loss: 0.7551001906394958:  96%|█████████▌| 6195/6470 [1:17:01<03:37,  1.26it/s]
4|6196|Loss: 0.7551001906394958:  96%|█████████▌| 6196/6470 [1:17:01<03:30,  1.30it/s]
4|6197|Loss: 0.8389176726341248:  96%|█████████▌| 6196/6470 [1:17:02<03:30,  1.30it/s]
4|6197|Loss: 0.8389176726341248:  96%|█████████▌| 6197/6470 [1:17:02<03:30,  1.29it/s]
4|6198|Loss: 0.8038472533226013:  96%|█████████▌| 6197/6470 [1:17:03<03:30,  1.29it/s]
4|6198|Loss: 0.8038472533226013:  96%|█████████▌| 6198/6470 [1:17:03<03:24,  1.33it/s]
4|6199|Loss: 0.6801703572273254:  96%|█████████▌| 6198/6470 [1:17:03<03:24,  1.33it/s]
4|6199|Loss: 0.6801703572273254:  96%|█████████▌| 6199/6470 [1:17:04<03:26,  1.31it/s]
4|6200|Loss: 0.5826090574264526:  96%|█████████▌| 6199/6470 [1:17:04<03:26,  1.31it/s]
4|6200|Loss: 0.5826090574264526:  96%|█████████▌| 6200/6470 [1:17:04<03:02,  1.48it/s]
4|6201|Loss: 0.7901199460029602:  96%|█████████▌| 6200/6470 [1:17:04<03:02,  1.48it/s]
4|6201|Loss: 0.7901199460029602:  96%|█████████▌| 6201/6470 [1:17:05<02:53,  1.55it/s]
4|6202|Loss: 0.9411208629608154:  96%|█████████▌| 6201/6470 [1:17:05<02:53,  1.55it/s]
4|6202|Loss: 0.9411208629608154:  96%|█████████▌| 6202/6470 [1:17:05<03:00,  1.49it/s]
4|6203|Loss: 1.0087734460830688:  96%|█████████▌| 6202/6470 [1:17:06<03:00,  1.49it/s]
4|6203|Loss: 1.0087734460830688:  96%|█████████▌| 6203/6470 [1:17:06<03:08,  1.42it/s]
4|6204|Loss: 0.9782992005348206:  96%|█████████▌| 6203/6470 [1:17:07<03:08,  1.42it/s]
4|6204|Loss: 0.9782992005348206:  96%|█████████▌| 6204/6470 [1:17:07<03:13,  1.38it/s]
4|6205|Loss: 0.8182320594787598:  96%|█████████▌| 6204/6470 [1:17:07<03:13,  1.38it/s]
4|6205|Loss: 0.8182320594787598:  96%|█████████▌| 6205/6470 [1:17:08<03:18,  1.34it/s]
4|6206|Loss: 0.726348340511322:  96%|█████████▌| 6205/6470 [1:17:08<03:18,  1.34it/s] 
4|6206|Loss: 0.726348340511322:  96%|█████████▌| 6206/6470 [1:17:09<03:26,  1.28it/s]
4|6207|Loss: 1.1006673574447632:  96%|█████████▌| 6206/6470 [1:17:09<03:26,  1.28it/s]
4|6207|Loss: 1.1006673574447632:  96%|█████████▌| 6207/6470 [1:17:09<03:28,  1.26it/s]
4|6208|Loss: 0.7596994042396545:  96%|█████████▌| 6207/6470 [1:17:10<03:28,  1.26it/s]
4|6208|Loss: 0.7596994042396545:  96%|█████████▌| 6208/6470 [1:17:10<03:22,  1.30it/s]
4|6209|Loss: 0.7619224786758423:  96%|█████████▌| 6208/6470 [1:17:11<03:22,  1.30it/s]
4|6209|Loss: 0.7619224786758423:  96%|█████████▌| 6209/6470 [1:17:11<03:13,  1.35it/s]
4|6210|Loss: 0.6735592484474182:  96%|█████████▌| 6209/6470 [1:17:11<03:13,  1.35it/s]
4|6210|Loss: 0.6735592484474182:  96%|█████████▌| 6210/6470 [1:17:12<03:12,  1.35it/s]
4|6211|Loss: 0.9229506254196167:  96%|█████████▌| 6210/6470 [1:17:12<03:12,  1.35it/s]
4|6211|Loss: 0.9229506254196167:  96%|█████████▌| 6211/6470 [1:17:12<03:15,  1.33it/s]
4|6212|Loss: 0.750373125076294:  96%|█████████▌| 6211/6470 [1:17:13<03:15,  1.33it/s] 
4|6212|Loss: 0.750373125076294:  96%|█████████▌| 6212/6470 [1:17:13<03:11,  1.35it/s]
4|6213|Loss: 0.8505721688270569:  96%|█████████▌| 6212/6470 [1:17:13<03:11,  1.35it/s]
4|6213|Loss: 0.8505721688270569:  96%|█████████▌| 6213/6470 [1:17:14<02:47,  1.53it/s]
4|6214|Loss: 0.9937288165092468:  96%|█████████▌| 6213/6470 [1:17:14<02:47,  1.53it/s]
4|6214|Loss: 0.9937288165092468:  96%|█████████▌| 6214/6470 [1:17:14<02:32,  1.67it/s]
4|6215|Loss: 0.7563866972923279:  96%|█████████▌| 6214/6470 [1:17:14<02:32,  1.67it/s]
4|6215|Loss: 0.7563866972923279:  96%|█████████▌| 6215/6470 [1:17:15<02:40,  1.59it/s]
4|6216|Loss: 0.8959559202194214:  96%|█████████▌| 6215/6470 [1:17:15<02:40,  1.59it/s]
4|6216|Loss: 0.8959559202194214:  96%|█████████▌| 6216/6470 [1:17:16<02:58,  1.42it/s]
4|6217|Loss: 0.8584129214286804:  96%|█████████▌| 6216/6470 [1:17:16<02:58,  1.42it/s]
4|6217|Loss: 0.8584129214286804:  96%|█████████▌| 6217/6470 [1:17:16<03:10,  1.33it/s]
4|6218|Loss: 0.829316258430481:  96%|█████████▌| 6217/6470 [1:17:17<03:10,  1.33it/s] 
4|6218|Loss: 0.829316258430481:  96%|█████████▌| 6218/6470 [1:17:17<03:05,  1.36it/s]
4|6219|Loss: 0.711823046207428:  96%|█████████▌| 6218/6470 [1:17:17<03:05,  1.36it/s]
4|6219|Loss: 0.711823046207428:  96%|█████████▌| 6219/6470 [1:17:18<02:50,  1.47it/s]
4|6220|Loss: 0.836556613445282:  96%|█████████▌| 6219/6470 [1:17:18<02:50,  1.47it/s]
4|6220|Loss: 0.836556613445282:  96%|█████████▌| 6220/6470 [1:17:19<03:01,  1.38it/s]
4|6221|Loss: 0.6884713172912598:  96%|█████████▌| 6220/6470 [1:17:19<03:01,  1.38it/s]
4|6221|Loss: 0.6884713172912598:  96%|█████████▌| 6221/6470 [1:17:19<03:08,  1.32it/s]
4|6222|Loss: 0.8983066082000732:  96%|█████████▌| 6221/6470 [1:17:20<03:08,  1.32it/s]
4|6222|Loss: 0.8983066082000732:  96%|█████████▌| 6222/6470 [1:17:20<03:11,  1.30it/s]
4|6223|Loss: 0.8419058918952942:  96%|█████████▌| 6222/6470 [1:17:21<03:11,  1.30it/s]
4|6223|Loss: 0.8419058918952942:  96%|█████████▌| 6223/6470 [1:17:21<03:18,  1.25it/s]
4|6224|Loss: 0.7593298554420471:  96%|█████████▌| 6223/6470 [1:17:21<03:18,  1.25it/s]
4|6224|Loss: 0.7593298554420471:  96%|█████████▌| 6224/6470 [1:17:22<03:19,  1.24it/s]
4|6225|Loss: 0.7397356033325195:  96%|█████████▌| 6224/6470 [1:17:22<03:19,  1.24it/s]
4|6225|Loss: 0.7397356033325195:  96%|█████████▌| 6225/6470 [1:17:23<03:10,  1.28it/s]
4|6226|Loss: 0.7218415141105652:  96%|█████████▌| 6225/6470 [1:17:23<03:10,  1.28it/s]
4|6226|Loss: 0.7218415141105652:  96%|█████████▌| 6226/6470 [1:17:23<02:57,  1.38it/s]
4|6227|Loss: 0.7283335328102112:  96%|█████████▌| 6226/6470 [1:17:24<02:57,  1.38it/s]
4|6227|Loss: 0.7283335328102112:  96%|█████████▌| 6227/6470 [1:17:24<02:46,  1.46it/s]
4|6228|Loss: 0.9332330822944641:  96%|█████████▌| 6227/6470 [1:17:24<02:46,  1.46it/s]
4|6228|Loss: 0.9332330822944641:  96%|█████████▋| 6228/6470 [1:17:25<02:50,  1.42it/s]
4|6229|Loss: 0.7570764422416687:  96%|█████████▋| 6228/6470 [1:17:25<02:50,  1.42it/s]
4|6229|Loss: 0.7570764422416687:  96%|█████████▋| 6229/6470 [1:17:25<03:02,  1.32it/s]
4|6230|Loss: 0.6967266201972961:  96%|█████████▋| 6229/6470 [1:17:26<03:02,  1.32it/s]
4|6230|Loss: 0.6967266201972961:  96%|█████████▋| 6230/6470 [1:17:26<03:00,  1.33it/s]
4|6231|Loss: 0.7133364081382751:  96%|█████████▋| 6230/6470 [1:17:27<03:00,  1.33it/s]
4|6231|Loss: 0.7133364081382751:  96%|█████████▋| 6231/6470 [1:17:27<03:02,  1.31it/s]
4|6232|Loss: 0.8602926731109619:  96%|█████████▋| 6231/6470 [1:17:27<03:02,  1.31it/s]
4|6232|Loss: 0.8602926731109619:  96%|█████████▋| 6232/6470 [1:17:28<03:01,  1.31it/s]
4|6233|Loss: 0.8679590821266174:  96%|█████████▋| 6232/6470 [1:17:28<03:01,  1.31it/s]
4|6233|Loss: 0.8679590821266174:  96%|█████████▋| 6233/6470 [1:17:28<02:55,  1.35it/s]
4|6234|Loss: 0.9737624526023865:  96%|█████████▋| 6233/6470 [1:17:29<02:55,  1.35it/s]
4|6234|Loss: 0.9737624526023865:  96%|█████████▋| 6234/6470 [1:17:29<03:03,  1.29it/s]
4|6235|Loss: 0.9563233256340027:  96%|█████████▋| 6234/6470 [1:17:30<03:03,  1.29it/s]
4|6235|Loss: 0.9563233256340027:  96%|█████████▋| 6235/6470 [1:17:30<03:03,  1.28it/s]
4|6236|Loss: 0.8389567136764526:  96%|█████████▋| 6235/6470 [1:17:30<03:03,  1.28it/s]
4|6236|Loss: 0.8389567136764526:  96%|█████████▋| 6236/6470 [1:17:31<03:04,  1.27it/s]
4|6237|Loss: 0.8658332824707031:  96%|█████████▋| 6236/6470 [1:17:31<03:04,  1.27it/s]
4|6237|Loss: 0.8658332824707031:  96%|█████████▋| 6237/6470 [1:17:32<03:00,  1.29it/s]
4|6238|Loss: 0.6656870245933533:  96%|█████████▋| 6237/6470 [1:17:32<03:00,  1.29it/s]
4|6238|Loss: 0.6656870245933533:  96%|█████████▋| 6238/6470 [1:17:32<03:07,  1.23it/s]
4|6239|Loss: 0.7725018262863159:  96%|█████████▋| 6238/6470 [1:17:33<03:07,  1.23it/s]
4|6239|Loss: 0.7725018262863159:  96%|█████████▋| 6239/6470 [1:17:33<02:59,  1.29it/s]
4|6240|Loss: 0.7505065202713013:  96%|█████████▋| 6239/6470 [1:17:34<02:59,  1.29it/s]
4|6240|Loss: 0.7505065202713013:  96%|█████████▋| 6240/6470 [1:17:34<02:50,  1.35it/s]
4|6241|Loss: 0.76664799451828:  96%|█████████▋| 6240/6470 [1:17:34<02:50,  1.35it/s]  
4|6241|Loss: 0.76664799451828:  96%|█████████▋| 6241/6470 [1:17:35<02:50,  1.34it/s]
4|6242|Loss: 0.9683362245559692:  96%|█████████▋| 6241/6470 [1:17:35<02:50,  1.34it/s]
4|6242|Loss: 0.9683362245559692:  96%|█████████▋| 6242/6470 [1:17:35<02:52,  1.32it/s]
4|6243|Loss: 0.8641121983528137:  96%|█████████▋| 6242/6470 [1:17:36<02:52,  1.32it/s]
4|6243|Loss: 0.8641121983528137:  96%|█████████▋| 6243/6470 [1:17:36<02:54,  1.30it/s]
4|6244|Loss: 0.8389108777046204:  96%|█████████▋| 6243/6470 [1:17:37<02:54,  1.30it/s]
4|6244|Loss: 0.8389108777046204:  97%|█████████▋| 6244/6470 [1:17:37<02:53,  1.30it/s]
4|6245|Loss: 0.9092336893081665:  97%|█████████▋| 6244/6470 [1:17:37<02:53,  1.30it/s]
4|6245|Loss: 0.9092336893081665:  97%|█████████▋| 6245/6470 [1:17:38<02:48,  1.33it/s]
4|6246|Loss: 0.9355000257492065:  97%|█████████▋| 6245/6470 [1:17:38<02:48,  1.33it/s]
4|6246|Loss: 0.9355000257492065:  97%|█████████▋| 6246/6470 [1:17:38<02:51,  1.31it/s]
4|6247|Loss: 0.9341095685958862:  97%|█████████▋| 6246/6470 [1:17:39<02:51,  1.31it/s]
4|6247|Loss: 0.9341095685958862:  97%|█████████▋| 6247/6470 [1:17:39<02:52,  1.29it/s]
4|6248|Loss: 0.8479635715484619:  97%|█████████▋| 6247/6470 [1:17:40<02:52,  1.29it/s]
4|6248|Loss: 0.8479635715484619:  97%|█████████▋| 6248/6470 [1:17:40<02:48,  1.32it/s]
4|6249|Loss: 0.8183949589729309:  97%|█████████▋| 6248/6470 [1:17:40<02:48,  1.32it/s]
4|6249|Loss: 0.8183949589729309:  97%|█████████▋| 6249/6470 [1:17:41<02:43,  1.35it/s]
4|6250|Loss: 0.8330731987953186:  97%|█████████▋| 6249/6470 [1:17:41<02:43,  1.35it/s]
4|6250|Loss: 0.8330731987953186:  97%|█████████▋| 6250/6470 [1:17:41<02:45,  1.33it/s]
4|6251|Loss: 0.706264853477478:  97%|█████████▋| 6250/6470 [1:17:42<02:45,  1.33it/s] 
4|6251|Loss: 0.706264853477478:  97%|█████████▋| 6251/6470 [1:17:42<02:33,  1.42it/s]
4|6252|Loss: 0.9650158286094666:  97%|█████████▋| 6251/6470 [1:17:42<02:33,  1.42it/s]
4|6252|Loss: 0.9650158286094666:  97%|█████████▋| 6252/6470 [1:17:43<02:33,  1.42it/s]
4|6253|Loss: 0.8032417893409729:  97%|█████████▋| 6252/6470 [1:17:43<02:33,  1.42it/s]
4|6253|Loss: 0.8032417893409729:  97%|█████████▋| 6253/6470 [1:17:44<02:44,  1.32it/s]
4|6254|Loss: 0.83140629529953:  97%|█████████▋| 6253/6470 [1:17:44<02:44,  1.32it/s]  
4|6254|Loss: 0.83140629529953:  97%|█████████▋| 6254/6470 [1:17:44<02:41,  1.34it/s]
4|6255|Loss: 0.8847340941429138:  97%|█████████▋| 6254/6470 [1:17:45<02:41,  1.34it/s]
4|6255|Loss: 0.8847340941429138:  97%|█████████▋| 6255/6470 [1:17:45<02:43,  1.31it/s]
4|6256|Loss: 0.7773427367210388:  97%|█████████▋| 6255/6470 [1:17:46<02:43,  1.31it/s]
4|6256|Loss: 0.7773427367210388:  97%|█████████▋| 6256/6470 [1:17:46<02:49,  1.26it/s]
4|6257|Loss: 0.7748624682426453:  97%|█████████▋| 6256/6470 [1:17:46<02:49,  1.26it/s]
4|6257|Loss: 0.7748624682426453:  97%|█████████▋| 6257/6470 [1:17:47<02:43,  1.31it/s]
4|6258|Loss: 0.8450191617012024:  97%|█████████▋| 6257/6470 [1:17:47<02:43,  1.31it/s]
4|6258|Loss: 0.8450191617012024:  97%|█████████▋| 6258/6470 [1:17:47<02:40,  1.32it/s]
4|6259|Loss: 0.8013622760772705:  97%|█████████▋| 6258/6470 [1:17:48<02:40,  1.32it/s]
4|6259|Loss: 0.8013622760772705:  97%|█████████▋| 6259/6470 [1:17:48<02:36,  1.35it/s]
4|6260|Loss: 0.9177607893943787:  97%|█████████▋| 6259/6470 [1:17:48<02:36,  1.35it/s]
4|6260|Loss: 0.9177607893943787:  97%|█████████▋| 6260/6470 [1:17:49<02:25,  1.45it/s]
4|6261|Loss: 0.8321887254714966:  97%|█████████▋| 6260/6470 [1:17:49<02:25,  1.45it/s]
4|6261|Loss: 0.8321887254714966:  97%|█████████▋| 6261/6470 [1:17:49<02:21,  1.48it/s]
4|6262|Loss: 0.8357023000717163:  97%|█████████▋| 6261/6470 [1:17:50<02:21,  1.48it/s]
4|6262|Loss: 0.8357023000717163:  97%|█████████▋| 6262/6470 [1:17:50<02:27,  1.41it/s]
4|6263|Loss: 0.9773910641670227:  97%|█████████▋| 6262/6470 [1:17:51<02:27,  1.41it/s]
4|6263|Loss: 0.9773910641670227:  97%|█████████▋| 6263/6470 [1:17:51<02:36,  1.33it/s]
4|6264|Loss: 0.8245437145233154:  97%|█████████▋| 6263/6470 [1:17:51<02:36,  1.33it/s]
4|6264|Loss: 0.8245437145233154:  97%|█████████▋| 6264/6470 [1:17:52<02:22,  1.44it/s]
4|6265|Loss: 0.7689756155014038:  97%|█████████▋| 6264/6470 [1:17:52<02:22,  1.44it/s]
4|6265|Loss: 0.7689756155014038:  97%|█████████▋| 6265/6470 [1:17:52<02:15,  1.52it/s]
4|6266|Loss: 0.7622120976448059:  97%|█████████▋| 6265/6470 [1:17:52<02:15,  1.52it/s]
4|6266|Loss: 0.7622120976448059:  97%|█████████▋| 6266/6470 [1:17:53<02:13,  1.53it/s]
4|6267|Loss: 0.7185054421424866:  97%|█████████▋| 6266/6470 [1:17:53<02:13,  1.53it/s]
4|6267|Loss: 0.7185054421424866:  97%|█████████▋| 6267/6470 [1:17:54<02:19,  1.46it/s]
4|6268|Loss: 0.7840183973312378:  97%|█████████▋| 6267/6470 [1:17:54<02:19,  1.46it/s]
4|6268|Loss: 0.7840183973312378:  97%|█████████▋| 6268/6470 [1:17:54<02:24,  1.40it/s]
4|6269|Loss: 1.0024065971374512:  97%|█████████▋| 6268/6470 [1:17:55<02:24,  1.40it/s]
4|6269|Loss: 1.0024065971374512:  97%|█████████▋| 6269/6470 [1:17:55<02:14,  1.50it/s]
4|6270|Loss: 0.8171950578689575:  97%|█████████▋| 6269/6470 [1:17:55<02:14,  1.50it/s]
4|6270|Loss: 0.8171950578689575:  97%|█████████▋| 6270/6470 [1:17:56<02:23,  1.39it/s]
4|6271|Loss: 0.7630230188369751:  97%|█████████▋| 6270/6470 [1:17:56<02:23,  1.39it/s]
4|6271|Loss: 0.7630230188369751:  97%|█████████▋| 6271/6470 [1:17:56<02:26,  1.36it/s]
4|6272|Loss: 0.8041372895240784:  97%|█████████▋| 6271/6470 [1:17:57<02:26,  1.36it/s]
4|6272|Loss: 0.8041372895240784:  97%|█████████▋| 6272/6470 [1:17:57<02:26,  1.35it/s]
4|6273|Loss: 0.8494600057601929:  97%|█████████▋| 6272/6470 [1:17:58<02:26,  1.35it/s]
4|6273|Loss: 0.8494600057601929:  97%|█████████▋| 6273/6470 [1:17:58<02:27,  1.34it/s]
4|6274|Loss: 0.7783412933349609:  97%|█████████▋| 6273/6470 [1:17:58<02:27,  1.34it/s]
4|6274|Loss: 0.7783412933349609:  97%|█████████▋| 6274/6470 [1:17:59<02:29,  1.31it/s]
4|6275|Loss: 0.795912504196167:  97%|█████████▋| 6274/6470 [1:17:59<02:29,  1.31it/s] 
4|6275|Loss: 0.795912504196167:  97%|█████████▋| 6275/6470 [1:17:59<02:25,  1.34it/s]
4|6276|Loss: 0.7281739115715027:  97%|█████████▋| 6275/6470 [1:18:00<02:25,  1.34it/s]
4|6276|Loss: 0.7281739115715027:  97%|█████████▋| 6276/6470 [1:18:00<02:28,  1.30it/s]
4|6277|Loss: 0.75870680809021:  97%|█████████▋| 6276/6470 [1:18:01<02:28,  1.30it/s]  
4|6277|Loss: 0.75870680809021:  97%|█████████▋| 6277/6470 [1:18:01<02:33,  1.26it/s]
4|6278|Loss: 0.7256362438201904:  97%|█████████▋| 6277/6470 [1:18:02<02:33,  1.26it/s]
4|6278|Loss: 0.7256362438201904:  97%|█████████▋| 6278/6470 [1:18:02<02:29,  1.29it/s]
4|6279|Loss: 0.806530773639679:  97%|█████████▋| 6278/6470 [1:18:02<02:29,  1.29it/s] 
4|6279|Loss: 0.806530773639679:  97%|█████████▋| 6279/6470 [1:18:03<02:26,  1.30it/s]
4|6280|Loss: 0.8387697339057922:  97%|█████████▋| 6279/6470 [1:18:03<02:26,  1.30it/s]
4|6280|Loss: 0.8387697339057922:  97%|█████████▋| 6280/6470 [1:18:03<02:26,  1.30it/s]
4|6281|Loss: 0.8612107634544373:  97%|█████████▋| 6280/6470 [1:18:04<02:26,  1.30it/s]
4|6281|Loss: 0.8612107634544373:  97%|█████████▋| 6281/6470 [1:18:04<02:25,  1.30it/s]
4|6282|Loss: 0.7965933680534363:  97%|█████████▋| 6281/6470 [1:18:05<02:25,  1.30it/s]
4|6282|Loss: 0.7965933680534363:  97%|█████████▋| 6282/6470 [1:18:05<02:24,  1.31it/s]
4|6283|Loss: 0.9461662173271179:  97%|█████████▋| 6282/6470 [1:18:05<02:24,  1.31it/s]
4|6283|Loss: 0.9461662173271179:  97%|█████████▋| 6283/6470 [1:18:06<02:30,  1.25it/s]
4|6284|Loss: 0.7950646877288818:  97%|█████████▋| 6283/6470 [1:18:06<02:30,  1.25it/s]
4|6284|Loss: 0.7950646877288818:  97%|█████████▋| 6284/6470 [1:18:07<02:25,  1.27it/s]
4|6285|Loss: 0.7002714276313782:  97%|█████████▋| 6284/6470 [1:18:07<02:25,  1.27it/s]
4|6285|Loss: 0.7002714276313782:  97%|█████████▋| 6285/6470 [1:18:07<02:28,  1.25it/s]
4|6286|Loss: 0.8230298161506653:  97%|█████████▋| 6285/6470 [1:18:08<02:28,  1.25it/s]
4|6286|Loss: 0.8230298161506653:  97%|█████████▋| 6286/6470 [1:18:08<02:27,  1.25it/s]
4|6287|Loss: 0.7801074981689453:  97%|█████████▋| 6286/6470 [1:18:09<02:27,  1.25it/s]
4|6287|Loss: 0.7801074981689453:  97%|█████████▋| 6287/6470 [1:18:09<02:25,  1.26it/s]
4|6288|Loss: 0.8158711194992065:  97%|█████████▋| 6287/6470 [1:18:09<02:25,  1.26it/s]
4|6288|Loss: 0.8158711194992065:  97%|█████████▋| 6288/6470 [1:18:10<02:17,  1.32it/s]
4|6289|Loss: 1.0303360223770142:  97%|█████████▋| 6288/6470 [1:18:10<02:17,  1.32it/s]
4|6289|Loss: 1.0303360223770142:  97%|█████████▋| 6289/6470 [1:18:10<02:06,  1.43it/s]
4|6290|Loss: 0.9263975620269775:  97%|█████████▋| 6289/6470 [1:18:11<02:06,  1.43it/s]
4|6290|Loss: 0.9263975620269775:  97%|█████████▋| 6290/6470 [1:18:11<02:03,  1.45it/s]
4|6291|Loss: 0.809762716293335:  97%|█████████▋| 6290/6470 [1:18:11<02:03,  1.45it/s] 
4|6291|Loss: 0.809762716293335:  97%|█████████▋| 6291/6470 [1:18:12<02:07,  1.40it/s]
4|6292|Loss: 0.8605675101280212:  97%|█████████▋| 6291/6470 [1:18:12<02:07,  1.40it/s]
4|6292|Loss: 0.8605675101280212:  97%|█████████▋| 6292/6470 [1:18:12<02:11,  1.35it/s]
4|6293|Loss: 0.7684713006019592:  97%|█████████▋| 6292/6470 [1:18:13<02:11,  1.35it/s]
4|6293|Loss: 0.7684713006019592:  97%|█████████▋| 6293/6470 [1:18:13<02:09,  1.36it/s]
4|6294|Loss: 0.8440288305282593:  97%|█████████▋| 6293/6470 [1:18:14<02:09,  1.36it/s]
4|6294|Loss: 0.8440288305282593:  97%|█████████▋| 6294/6470 [1:18:14<02:10,  1.35it/s]
4|6295|Loss: 0.7787261605262756:  97%|█████████▋| 6294/6470 [1:18:14<02:10,  1.35it/s]
4|6295|Loss: 0.7787261605262756:  97%|█████████▋| 6295/6470 [1:18:15<02:07,  1.37it/s]
4|6296|Loss: 0.8499950170516968:  97%|█████████▋| 6295/6470 [1:18:15<02:07,  1.37it/s]
4|6296|Loss: 0.8499950170516968:  97%|█████████▋| 6296/6470 [1:18:15<02:08,  1.36it/s]
4|6297|Loss: 0.7535974383354187:  97%|█████████▋| 6296/6470 [1:18:16<02:08,  1.36it/s]
4|6297|Loss: 0.7535974383354187:  97%|█████████▋| 6297/6470 [1:18:16<02:05,  1.38it/s]
4|6298|Loss: 0.9218860864639282:  97%|█████████▋| 6297/6470 [1:18:16<02:05,  1.38it/s]
4|6298|Loss: 0.9218860864639282:  97%|█████████▋| 6298/6470 [1:18:17<02:02,  1.40it/s]
4|6299|Loss: 0.9322330355644226:  97%|█████████▋| 6298/6470 [1:18:17<02:02,  1.40it/s]
4|6299|Loss: 0.9322330355644226:  97%|█████████▋| 6299/6470 [1:18:17<01:59,  1.44it/s]
4|6300|Loss: 0.815401017665863:  97%|█████████▋| 6299/6470 [1:18:18<01:59,  1.44it/s] 
4|6300|Loss: 0.815401017665863:  97%|█████████▋| 6300/6470 [1:18:18<01:57,  1.44it/s]
4|6301|Loss: 0.7689301371574402:  97%|█████████▋| 6300/6470 [1:18:19<01:57,  1.44it/s]
4|6301|Loss: 0.7689301371574402:  97%|█████████▋| 6301/6470 [1:18:19<02:00,  1.41it/s]
4|6302|Loss: 0.8775014877319336:  97%|█████████▋| 6301/6470 [1:18:19<02:00,  1.41it/s]
4|6302|Loss: 0.8775014877319336:  97%|█████████▋| 6302/6470 [1:18:19<01:50,  1.52it/s]
4|6303|Loss: 0.8490496277809143:  97%|█████████▋| 6302/6470 [1:18:20<01:50,  1.52it/s]
4|6303|Loss: 0.8490496277809143:  97%|█████████▋| 6303/6470 [1:18:20<01:56,  1.43it/s]
4|6304|Loss: 0.9568933248519897:  97%|█████████▋| 6303/6470 [1:18:21<01:56,  1.43it/s]
4|6304|Loss: 0.9568933248519897:  97%|█████████▋| 6304/6470 [1:18:21<01:54,  1.45it/s]
4|6305|Loss: 0.9041008949279785:  97%|█████████▋| 6304/6470 [1:18:21<01:54,  1.45it/s]
4|6305|Loss: 0.9041008949279785:  97%|█████████▋| 6305/6470 [1:18:22<01:57,  1.40it/s]
4|6306|Loss: 0.9712384939193726:  97%|█████████▋| 6305/6470 [1:18:22<01:57,  1.40it/s]
4|6306|Loss: 0.9712384939193726:  97%|█████████▋| 6306/6470 [1:18:22<01:56,  1.41it/s]
4|6307|Loss: 0.7401285767555237:  97%|█████████▋| 6306/6470 [1:18:23<01:56,  1.41it/s]
4|6307|Loss: 0.7401285767555237:  97%|█████████▋| 6307/6470 [1:18:23<01:48,  1.50it/s]
4|6308|Loss: 0.8305803537368774:  97%|█████████▋| 6307/6470 [1:18:23<01:48,  1.50it/s]
4|6308|Loss: 0.8305803537368774:  97%|█████████▋| 6308/6470 [1:18:24<01:48,  1.49it/s]
4|6309|Loss: 0.8043190240859985:  97%|█████████▋| 6308/6470 [1:18:24<01:48,  1.49it/s]
4|6309|Loss: 0.8043190240859985:  98%|█████████▊| 6309/6470 [1:18:24<01:56,  1.39it/s]
4|6310|Loss: 0.811589241027832:  98%|█████████▊| 6309/6470 [1:18:25<01:56,  1.39it/s] 
4|6310|Loss: 0.811589241027832:  98%|█████████▊| 6310/6470 [1:18:25<01:56,  1.38it/s]
4|6311|Loss: 0.8888834714889526:  98%|█████████▊| 6310/6470 [1:18:25<01:56,  1.38it/s]
4|6311|Loss: 0.8888834714889526:  98%|█████████▊| 6311/6470 [1:18:26<01:42,  1.56it/s]
4|6312|Loss: 1.0091696977615356:  98%|█████████▊| 6311/6470 [1:18:26<01:42,  1.56it/s]
4|6312|Loss: 1.0091696977615356:  98%|█████████▊| 6312/6470 [1:18:26<01:50,  1.43it/s]
4|6313|Loss: 0.7865358591079712:  98%|█████████▊| 6312/6470 [1:18:27<01:50,  1.43it/s]
4|6313|Loss: 0.7865358591079712:  98%|█████████▊| 6313/6470 [1:18:27<01:48,  1.45it/s]
4|6314|Loss: 0.7726058959960938:  98%|█████████▊| 6313/6470 [1:18:27<01:48,  1.45it/s]
4|6314|Loss: 0.7726058959960938:  98%|█████████▊| 6314/6470 [1:18:28<01:40,  1.55it/s]
4|6315|Loss: 0.7231014370918274:  98%|█████████▊| 6314/6470 [1:18:28<01:40,  1.55it/s]
4|6315|Loss: 0.7231014370918274:  98%|█████████▊| 6315/6470 [1:18:28<01:40,  1.54it/s]
4|6316|Loss: 0.9068545699119568:  98%|█████████▊| 6315/6470 [1:18:29<01:40,  1.54it/s]
4|6316|Loss: 0.9068545699119568:  98%|█████████▊| 6316/6470 [1:18:29<01:44,  1.47it/s]
4|6317|Loss: 0.7649308443069458:  98%|█████████▊| 6316/6470 [1:18:29<01:44,  1.47it/s]
4|6317|Loss: 0.7649308443069458:  98%|█████████▊| 6317/6470 [1:18:30<01:44,  1.47it/s]
4|6318|Loss: 0.9595118165016174:  98%|█████████▊| 6317/6470 [1:18:30<01:44,  1.47it/s]
4|6318|Loss: 0.9595118165016174:  98%|█████████▊| 6318/6470 [1:18:31<01:50,  1.38it/s]
4|6319|Loss: 0.7829753756523132:  98%|█████████▊| 6318/6470 [1:18:31<01:50,  1.38it/s]
4|6319|Loss: 0.7829753756523132:  98%|█████████▊| 6319/6470 [1:18:31<01:52,  1.35it/s]
4|6320|Loss: 0.858173668384552:  98%|█████████▊| 6319/6470 [1:18:32<01:52,  1.35it/s] 
4|6320|Loss: 0.858173668384552:  98%|█████████▊| 6320/6470 [1:18:32<01:52,  1.33it/s]
4|6321|Loss: 0.8586184978485107:  98%|█████████▊| 6320/6470 [1:18:33<01:52,  1.33it/s]
4|6321|Loss: 0.8586184978485107:  98%|█████████▊| 6321/6470 [1:18:33<01:53,  1.31it/s]
4|6322|Loss: 0.7509834170341492:  98%|█████████▊| 6321/6470 [1:18:33<01:53,  1.31it/s]
4|6322|Loss: 0.7509834170341492:  98%|█████████▊| 6322/6470 [1:18:34<01:47,  1.38it/s]
4|6323|Loss: 0.7354969382286072:  98%|█████████▊| 6322/6470 [1:18:34<01:47,  1.38it/s]
4|6323|Loss: 0.7354969382286072:  98%|█████████▊| 6323/6470 [1:18:34<01:52,  1.30it/s]
4|6324|Loss: 0.7237650752067566:  98%|█████████▊| 6323/6470 [1:18:35<01:52,  1.30it/s]
4|6324|Loss: 0.7237650752067566:  98%|█████████▊| 6324/6470 [1:18:35<01:51,  1.31it/s]
4|6325|Loss: 0.7929909825325012:  98%|█████████▊| 6324/6470 [1:18:36<01:51,  1.31it/s]
4|6325|Loss: 0.7929909825325012:  98%|█████████▊| 6325/6470 [1:18:36<01:47,  1.35it/s]
4|6326|Loss: 0.8895752429962158:  98%|█████████▊| 6325/6470 [1:18:36<01:47,  1.35it/s]
4|6326|Loss: 0.8895752429962158:  98%|█████████▊| 6326/6470 [1:18:37<01:48,  1.33it/s]
4|6327|Loss: 0.832884669303894:  98%|█████████▊| 6326/6470 [1:18:37<01:48,  1.33it/s] 
4|6327|Loss: 0.832884669303894:  98%|█████████▊| 6327/6470 [1:18:37<01:43,  1.38it/s]
4|6328|Loss: 0.7937110066413879:  98%|█████████▊| 6327/6470 [1:18:38<01:43,  1.38it/s]
4|6328|Loss: 0.7937110066413879:  98%|█████████▊| 6328/6470 [1:18:38<01:41,  1.40it/s]
4|6329|Loss: 0.9978969693183899:  98%|█████████▊| 6328/6470 [1:18:38<01:41,  1.40it/s]
4|6329|Loss: 0.9978969693183899:  98%|█████████▊| 6329/6470 [1:18:39<01:46,  1.32it/s]
4|6330|Loss: 0.5448453426361084:  98%|█████████▊| 6329/6470 [1:18:39<01:46,  1.32it/s]
4|6330|Loss: 0.5448453426361084:  98%|█████████▊| 6330/6470 [1:18:40<02:00,  1.16it/s]
4|6331|Loss: 0.8609684705734253:  98%|█████████▊| 6330/6470 [1:18:40<02:00,  1.16it/s]
4|6331|Loss: 0.8609684705734253:  98%|█████████▊| 6331/6470 [1:18:41<02:02,  1.13it/s]
4|6332|Loss: 0.7379546165466309:  98%|█████████▊| 6331/6470 [1:18:41<02:02,  1.13it/s]
4|6332|Loss: 0.7379546165466309:  98%|█████████▊| 6332/6470 [1:18:42<01:58,  1.16it/s]
4|6333|Loss: 0.8156420588493347:  98%|█████████▊| 6332/6470 [1:18:42<01:58,  1.16it/s]
4|6333|Loss: 0.8156420588493347:  98%|█████████▊| 6333/6470 [1:18:42<01:51,  1.23it/s]
4|6334|Loss: 0.7881696224212646:  98%|█████████▊| 6333/6470 [1:18:43<01:51,  1.23it/s]
4|6334|Loss: 0.7881696224212646:  98%|█████████▊| 6334/6470 [1:18:43<01:48,  1.26it/s]
4|6335|Loss: 0.8138486742973328:  98%|█████████▊| 6334/6470 [1:18:44<01:48,  1.26it/s]
4|6335|Loss: 0.8138486742973328:  98%|█████████▊| 6335/6470 [1:18:44<01:46,  1.27it/s]
4|6336|Loss: 0.8905835151672363:  98%|█████████▊| 6335/6470 [1:18:44<01:46,  1.27it/s]
4|6336|Loss: 0.8905835151672363:  98%|█████████▊| 6336/6470 [1:18:45<01:42,  1.30it/s]
4|6337|Loss: 0.933064341545105:  98%|█████████▊| 6336/6470 [1:18:45<01:42,  1.30it/s] 
4|6337|Loss: 0.933064341545105:  98%|█████████▊| 6337/6470 [1:18:45<01:42,  1.30it/s]
4|6338|Loss: 0.7858811616897583:  98%|█████████▊| 6337/6470 [1:18:46<01:42,  1.30it/s]
4|6338|Loss: 0.7858811616897583:  98%|█████████▊| 6338/6470 [1:18:46<01:45,  1.25it/s]
4|6339|Loss: 0.8446775078773499:  98%|█████████▊| 6338/6470 [1:18:47<01:45,  1.25it/s]
4|6339|Loss: 0.8446775078773499:  98%|█████████▊| 6339/6470 [1:18:47<01:41,  1.29it/s]
4|6340|Loss: 0.7492033839225769:  98%|█████████▊| 6339/6470 [1:18:47<01:41,  1.29it/s]
4|6340|Loss: 0.7492033839225769:  98%|█████████▊| 6340/6470 [1:18:48<01:39,  1.30it/s]
4|6341|Loss: 0.8091811537742615:  98%|█████████▊| 6340/6470 [1:18:48<01:39,  1.30it/s]
4|6341|Loss: 0.8091811537742615:  98%|█████████▊| 6341/6470 [1:18:48<01:33,  1.38it/s]
4|6342|Loss: 0.8200753331184387:  98%|█████████▊| 6341/6470 [1:18:49<01:33,  1.38it/s]
4|6342|Loss: 0.8200753331184387:  98%|█████████▊| 6342/6470 [1:18:49<01:32,  1.38it/s]
4|6343|Loss: 0.7713056802749634:  98%|█████████▊| 6342/6470 [1:18:49<01:32,  1.38it/s]
4|6343|Loss: 0.7713056802749634:  98%|█████████▊| 6343/6470 [1:18:50<01:27,  1.45it/s]
4|6344|Loss: 0.9259371161460876:  98%|█████████▊| 6343/6470 [1:18:50<01:27,  1.45it/s]
4|6344|Loss: 0.9259371161460876:  98%|█████████▊| 6344/6470 [1:18:50<01:28,  1.43it/s]
4|6345|Loss: 0.8756745457649231:  98%|█████████▊| 6344/6470 [1:18:51<01:28,  1.43it/s]
4|6345|Loss: 0.8756745457649231:  98%|█████████▊| 6345/6470 [1:18:51<01:21,  1.54it/s]
4|6346|Loss: 0.796731173992157:  98%|█████████▊| 6345/6470 [1:18:51<01:21,  1.54it/s] 
4|6346|Loss: 0.796731173992157:  98%|█████████▊| 6346/6470 [1:18:52<01:20,  1.54it/s]
4|6347|Loss: 0.7353724837303162:  98%|█████████▊| 6346/6470 [1:18:52<01:20,  1.54it/s]
4|6347|Loss: 0.7353724837303162:  98%|█████████▊| 6347/6470 [1:18:52<01:19,  1.54it/s]
4|6348|Loss: 0.9125176668167114:  98%|█████████▊| 6347/6470 [1:18:53<01:19,  1.54it/s]
4|6348|Loss: 0.9125176668167114:  98%|█████████▊| 6348/6470 [1:18:53<01:22,  1.47it/s]
4|6349|Loss: 0.8668244481086731:  98%|█████████▊| 6348/6470 [1:18:53<01:22,  1.47it/s]
4|6349|Loss: 0.8668244481086731:  98%|█████████▊| 6349/6470 [1:18:54<01:25,  1.41it/s]
4|6350|Loss: 0.8078187704086304:  98%|█████████▊| 6349/6470 [1:18:54<01:25,  1.41it/s]
4|6350|Loss: 0.8078187704086304:  98%|█████████▊| 6350/6470 [1:18:55<01:32,  1.30it/s]
4|6351|Loss: 0.9246214032173157:  98%|█████████▊| 6350/6470 [1:18:55<01:32,  1.30it/s]
4|6351|Loss: 0.9246214032173157:  98%|█████████▊| 6351/6470 [1:18:56<01:37,  1.22it/s]
4|6352|Loss: 0.8481756448745728:  98%|█████████▊| 6351/6470 [1:18:56<01:37,  1.22it/s]
4|6352|Loss: 0.8481756448745728:  98%|█████████▊| 6352/6470 [1:18:56<01:36,  1.23it/s]
4|6353|Loss: 0.6655343770980835:  98%|█████████▊| 6352/6470 [1:18:57<01:36,  1.23it/s]
4|6353|Loss: 0.6655343770980835:  98%|█████████▊| 6353/6470 [1:18:57<01:26,  1.35it/s]
4|6354|Loss: 0.8484066128730774:  98%|█████████▊| 6353/6470 [1:18:57<01:26,  1.35it/s]
4|6354|Loss: 0.8484066128730774:  98%|█████████▊| 6354/6470 [1:18:58<01:28,  1.30it/s]
4|6355|Loss: 0.8091123104095459:  98%|█████████▊| 6354/6470 [1:18:58<01:28,  1.30it/s]
4|6355|Loss: 0.8091123104095459:  98%|█████████▊| 6355/6470 [1:18:58<01:24,  1.36it/s]
4|6356|Loss: 0.7270299196243286:  98%|█████████▊| 6355/6470 [1:18:59<01:24,  1.36it/s]
4|6356|Loss: 0.7270299196243286:  98%|█████████▊| 6356/6470 [1:18:59<01:16,  1.49it/s]
4|6357|Loss: 0.789665937423706:  98%|█████████▊| 6356/6470 [1:18:59<01:16,  1.49it/s] 
4|6357|Loss: 0.789665937423706:  98%|█████████▊| 6357/6470 [1:19:00<01:14,  1.51it/s]
4|6358|Loss: 0.6871551871299744:  98%|█████████▊| 6357/6470 [1:19:00<01:14,  1.51it/s]
4|6358|Loss: 0.6871551871299744:  98%|█████████▊| 6358/6470 [1:19:01<01:23,  1.35it/s]
4|6359|Loss: 0.8126068711280823:  98%|█████████▊| 6358/6470 [1:19:01<01:23,  1.35it/s]
4|6359|Loss: 0.8126068711280823:  98%|█████████▊| 6359/6470 [1:19:01<01:27,  1.27it/s]
4|6360|Loss: 0.7778754830360413:  98%|█████████▊| 6359/6470 [1:19:02<01:27,  1.27it/s]
4|6360|Loss: 0.7778754830360413:  98%|█████████▊| 6360/6470 [1:19:02<01:24,  1.30it/s]
4|6361|Loss: 0.8442259430885315:  98%|█████████▊| 6360/6470 [1:19:03<01:24,  1.30it/s]
4|6361|Loss: 0.8442259430885315:  98%|█████████▊| 6361/6470 [1:19:03<01:23,  1.31it/s]
4|6362|Loss: 0.8468457460403442:  98%|█████████▊| 6361/6470 [1:19:03<01:23,  1.31it/s]
4|6362|Loss: 0.8468457460403442:  98%|█████████▊| 6362/6470 [1:19:04<01:18,  1.38it/s]
4|6363|Loss: 0.8604773283004761:  98%|█████████▊| 6362/6470 [1:19:04<01:18,  1.38it/s]
4|6363|Loss: 0.8604773283004761:  98%|█████████▊| 6363/6470 [1:19:04<01:14,  1.44it/s]
4|6364|Loss: 0.8867194652557373:  98%|█████████▊| 6363/6470 [1:19:05<01:14,  1.44it/s]
4|6364|Loss: 0.8867194652557373:  98%|█████████▊| 6364/6470 [1:19:05<01:10,  1.51it/s]
4|6365|Loss: 0.8783899545669556:  98%|█████████▊| 6364/6470 [1:19:05<01:10,  1.51it/s]
4|6365|Loss: 0.8783899545669556:  98%|█████████▊| 6365/6470 [1:19:05<01:09,  1.51it/s]
4|6366|Loss: 0.7951861023902893:  98%|█████████▊| 6365/6470 [1:19:06<01:09,  1.51it/s]
4|6366|Loss: 0.7951861023902893:  98%|█████████▊| 6366/6470 [1:19:06<01:08,  1.51it/s]
4|6367|Loss: 0.7091786861419678:  98%|█████████▊| 6366/6470 [1:19:07<01:08,  1.51it/s]
4|6367|Loss: 0.7091786861419678:  98%|█████████▊| 6367/6470 [1:19:07<01:15,  1.36it/s]
4|6368|Loss: 0.8470039367675781:  98%|█████████▊| 6367/6470 [1:19:08<01:15,  1.36it/s]
4|6368|Loss: 0.8470039367675781:  98%|█████████▊| 6368/6470 [1:19:08<01:23,  1.23it/s]
4|6369|Loss: 0.7063090205192566:  98%|█████████▊| 6368/6470 [1:19:09<01:23,  1.23it/s]
4|6369|Loss: 0.7063090205192566:  98%|█████████▊| 6369/6470 [1:19:09<01:35,  1.05it/s]
4|6370|Loss: 0.7613673210144043:  98%|█████████▊| 6369/6470 [1:19:10<01:35,  1.05it/s]
4|6370|Loss: 0.7613673210144043:  98%|█████████▊| 6370/6470 [1:19:10<01:26,  1.15it/s]
4|6371|Loss: 0.9275727868080139:  98%|█████████▊| 6370/6470 [1:19:10<01:26,  1.15it/s]
4|6371|Loss: 0.9275727868080139:  98%|█████████▊| 6371/6470 [1:19:11<01:21,  1.21it/s]
4|6372|Loss: 0.807954728603363:  98%|█████████▊| 6371/6470 [1:19:11<01:21,  1.21it/s] 
4|6372|Loss: 0.807954728603363:  98%|█████████▊| 6372/6470 [1:19:11<01:17,  1.26it/s]
4|6373|Loss: 0.7558597922325134:  98%|█████████▊| 6372/6470 [1:19:12<01:17,  1.26it/s]
4|6373|Loss: 0.7558597922325134:  99%|█████████▊| 6373/6470 [1:19:12<01:19,  1.22it/s]
4|6374|Loss: 0.8154498934745789:  99%|█████████▊| 6373/6470 [1:19:13<01:19,  1.22it/s]
4|6374|Loss: 0.8154498934745789:  99%|█████████▊| 6374/6470 [1:19:13<01:20,  1.20it/s]
4|6375|Loss: 0.814428448677063:  99%|█████████▊| 6374/6470 [1:19:14<01:20,  1.20it/s] 
4|6375|Loss: 0.814428448677063:  99%|█████████▊| 6375/6470 [1:19:14<01:14,  1.27it/s]
4|6376|Loss: 0.892228364944458:  99%|█████████▊| 6375/6470 [1:19:14<01:14,  1.27it/s]
4|6376|Loss: 0.892228364944458:  99%|█████████▊| 6376/6470 [1:19:15<01:25,  1.10it/s]
4|6377|Loss: 0.6995440721511841:  99%|█████████▊| 6376/6470 [1:19:16<01:25,  1.10it/s]
4|6377|Loss: 0.6995440721511841:  99%|█████████▊| 6377/6470 [1:19:16<01:24,  1.10it/s]
4|6378|Loss: 0.7095888257026672:  99%|█████████▊| 6377/6470 [1:19:16<01:24,  1.10it/s]
4|6378|Loss: 0.7095888257026672:  99%|█████████▊| 6378/6470 [1:19:17<01:21,  1.13it/s]
4|6379|Loss: 0.9450149536132812:  99%|█████████▊| 6378/6470 [1:19:17<01:21,  1.13it/s]
4|6379|Loss: 0.9450149536132812:  99%|█████████▊| 6379/6470 [1:19:18<01:17,  1.17it/s]
4|6380|Loss: 0.7284283638000488:  99%|█████████▊| 6379/6470 [1:19:18<01:17,  1.17it/s]
4|6380|Loss: 0.7284283638000488:  99%|█████████▊| 6380/6470 [1:19:18<01:05,  1.37it/s]
4|6381|Loss: 0.7433781623840332:  99%|█████████▊| 6380/6470 [1:19:18<01:05,  1.37it/s]
4|6381|Loss: 0.7433781623840332:  99%|█████████▊| 6381/6470 [1:19:19<01:08,  1.30it/s]
4|6382|Loss: 0.7960224747657776:  99%|█████████▊| 6381/6470 [1:19:19<01:08,  1.30it/s]
4|6382|Loss: 0.7960224747657776:  99%|█████████▊| 6382/6470 [1:19:20<01:09,  1.26it/s]
4|6383|Loss: 0.7323453426361084:  99%|█████████▊| 6382/6470 [1:19:20<01:09,  1.26it/s]
4|6383|Loss: 0.7323453426361084:  99%|█████████▊| 6383/6470 [1:19:20<01:04,  1.34it/s]
4|6384|Loss: 0.9249680638313293:  99%|█████████▊| 6383/6470 [1:19:21<01:04,  1.34it/s]
4|6384|Loss: 0.9249680638313293:  99%|█████████▊| 6384/6470 [1:19:21<01:01,  1.41it/s]
4|6385|Loss: 0.8225183486938477:  99%|█████████▊| 6384/6470 [1:19:21<01:01,  1.41it/s]
4|6385|Loss: 0.8225183486938477:  99%|█████████▊| 6385/6470 [1:19:22<00:58,  1.44it/s]
4|6386|Loss: 1.2505625486373901:  99%|█████████▊| 6385/6470 [1:19:22<00:58,  1.44it/s]
4|6386|Loss: 1.2505625486373901:  99%|█████████▊| 6386/6470 [1:19:22<00:59,  1.41it/s]
4|6387|Loss: 0.8294406533241272:  99%|█████████▊| 6386/6470 [1:19:23<00:59,  1.41it/s]
4|6387|Loss: 0.8294406533241272:  99%|█████████▊| 6387/6470 [1:19:23<00:57,  1.43it/s]
4|6388|Loss: 0.8245141506195068:  99%|█████████▊| 6387/6470 [1:19:23<00:57,  1.43it/s]
4|6388|Loss: 0.8245141506195068:  99%|█████████▊| 6388/6470 [1:19:24<00:58,  1.41it/s]
4|6389|Loss: 0.8627122640609741:  99%|█████████▊| 6388/6470 [1:19:24<00:58,  1.41it/s]
4|6389|Loss: 0.8627122640609741:  99%|█████████▊| 6389/6470 [1:19:24<00:57,  1.42it/s]
4|6390|Loss: 0.8224540948867798:  99%|█████████▊| 6389/6470 [1:19:25<00:57,  1.42it/s]
4|6390|Loss: 0.8224540948867798:  99%|█████████▉| 6390/6470 [1:19:25<00:53,  1.51it/s]
4|6391|Loss: 0.9542524218559265:  99%|█████████▉| 6390/6470 [1:19:25<00:53,  1.51it/s]
4|6391|Loss: 0.9542524218559265:  99%|█████████▉| 6391/6470 [1:19:26<00:50,  1.57it/s]
4|6392|Loss: 1.01012122631073:  99%|█████████▉| 6391/6470 [1:19:26<00:50,  1.57it/s]  
4|6392|Loss: 1.01012122631073:  99%|█████████▉| 6392/6470 [1:19:26<00:54,  1.43it/s]
4|6393|Loss: 0.8111633062362671:  99%|█████████▉| 6392/6470 [1:19:27<00:54,  1.43it/s]
4|6393|Loss: 0.8111633062362671:  99%|█████████▉| 6393/6470 [1:19:27<00:55,  1.38it/s]
4|6394|Loss: 0.8261735439300537:  99%|█████████▉| 6393/6470 [1:19:28<00:55,  1.38it/s]
4|6394|Loss: 0.8261735439300537:  99%|█████████▉| 6394/6470 [1:19:28<00:53,  1.42it/s]
4|6395|Loss: 0.8983485102653503:  99%|█████████▉| 6394/6470 [1:19:28<00:53,  1.42it/s]
4|6395|Loss: 0.8983485102653503:  99%|█████████▉| 6395/6470 [1:19:29<00:54,  1.38it/s]
4|6396|Loss: 0.8194251656532288:  99%|█████████▉| 6395/6470 [1:19:29<00:54,  1.38it/s]
4|6396|Loss: 0.8194251656532288:  99%|█████████▉| 6396/6470 [1:19:29<00:54,  1.35it/s]
4|6397|Loss: 0.786118745803833:  99%|█████████▉| 6396/6470 [1:19:30<00:54,  1.35it/s] 
4|6397|Loss: 0.786118745803833:  99%|█████████▉| 6397/6470 [1:19:30<00:52,  1.40it/s]
4|6398|Loss: 0.8691589832305908:  99%|█████████▉| 6397/6470 [1:19:30<00:52,  1.40it/s]
4|6398|Loss: 0.8691589832305908:  99%|█████████▉| 6398/6470 [1:19:31<00:52,  1.37it/s]
4|6399|Loss: 0.7265341281890869:  99%|█████████▉| 6398/6470 [1:19:31<00:52,  1.37it/s]
4|6399|Loss: 0.7265341281890869:  99%|█████████▉| 6399/6470 [1:19:32<00:53,  1.34it/s]
4|6400|Loss: 0.6643664240837097:  99%|█████████▉| 6399/6470 [1:19:32<00:53,  1.34it/s]
4|6400|Loss: 0.6643664240837097:  99%|█████████▉| 6400/6470 [1:19:32<00:54,  1.30it/s]
4|6401|Loss: 0.735783576965332:  99%|█████████▉| 6400/6470 [1:19:33<00:54,  1.30it/s] 
4|6401|Loss: 0.735783576965332:  99%|█████████▉| 6401/6470 [1:19:33<00:55,  1.25it/s]
4|6402|Loss: 0.8217317461967468:  99%|█████████▉| 6401/6470 [1:19:34<00:55,  1.25it/s]
4|6402|Loss: 0.8217317461967468:  99%|█████████▉| 6402/6470 [1:19:34<00:51,  1.33it/s]
4|6403|Loss: 0.7397685050964355:  99%|█████████▉| 6402/6470 [1:19:34<00:51,  1.33it/s]
4|6403|Loss: 0.7397685050964355:  99%|█████████▉| 6403/6470 [1:19:35<00:49,  1.35it/s]
4|6404|Loss: 0.8899758458137512:  99%|█████████▉| 6403/6470 [1:19:35<00:49,  1.35it/s]
4|6404|Loss: 0.8899758458137512:  99%|█████████▉| 6404/6470 [1:19:35<00:49,  1.34it/s]
4|6405|Loss: 0.825894296169281:  99%|█████████▉| 6404/6470 [1:19:36<00:49,  1.34it/s] 
4|6405|Loss: 0.825894296169281:  99%|█████████▉| 6405/6470 [1:19:36<00:48,  1.33it/s]
4|6406|Loss: 0.7357394099235535:  99%|█████████▉| 6405/6470 [1:19:37<00:48,  1.33it/s]
4|6406|Loss: 0.7357394099235535:  99%|█████████▉| 6406/6470 [1:19:37<00:48,  1.32it/s]
4|6407|Loss: 0.7738483548164368:  99%|█████████▉| 6406/6470 [1:19:37<00:48,  1.32it/s]
4|6407|Loss: 0.7738483548164368:  99%|█████████▉| 6407/6470 [1:19:38<00:48,  1.30it/s]
4|6408|Loss: 0.8699111342430115:  99%|█████████▉| 6407/6470 [1:19:38<00:48,  1.30it/s]
4|6408|Loss: 0.8699111342430115:  99%|█████████▉| 6408/6470 [1:19:39<00:48,  1.28it/s]
4|6409|Loss: 0.7914555072784424:  99%|█████████▉| 6408/6470 [1:19:39<00:48,  1.28it/s]
4|6409|Loss: 0.7914555072784424:  99%|█████████▉| 6409/6470 [1:19:39<00:49,  1.24it/s]
4|6410|Loss: 0.758540689945221:  99%|█████████▉| 6409/6470 [1:19:40<00:49,  1.24it/s] 
4|6410|Loss: 0.758540689945221:  99%|█████████▉| 6410/6470 [1:19:40<00:46,  1.29it/s]
4|6411|Loss: 0.5375667810440063:  99%|█████████▉| 6410/6470 [1:19:41<00:46,  1.29it/s]
4|6411|Loss: 0.5375667810440063:  99%|█████████▉| 6411/6470 [1:19:41<00:48,  1.22it/s]
4|6412|Loss: 0.7946810126304626:  99%|█████████▉| 6411/6470 [1:19:42<00:48,  1.22it/s]
4|6412|Loss: 0.7946810126304626:  99%|█████████▉| 6412/6470 [1:19:42<00:45,  1.27it/s]
4|6413|Loss: 0.8648643493652344:  99%|█████████▉| 6412/6470 [1:19:42<00:45,  1.27it/s]
4|6413|Loss: 0.8648643493652344:  99%|█████████▉| 6413/6470 [1:19:43<00:44,  1.29it/s]
4|6414|Loss: 0.9662947654724121:  99%|█████████▉| 6413/6470 [1:19:43<00:44,  1.29it/s]
4|6414|Loss: 0.9662947654724121:  99%|█████████▉| 6414/6470 [1:19:43<00:42,  1.33it/s]
4|6415|Loss: 0.8003415465354919:  99%|█████████▉| 6414/6470 [1:19:44<00:42,  1.33it/s]
4|6415|Loss: 0.8003415465354919:  99%|█████████▉| 6415/6470 [1:19:44<00:41,  1.33it/s]
4|6416|Loss: 0.7607876062393188:  99%|█████████▉| 6415/6470 [1:19:44<00:41,  1.33it/s]
4|6416|Loss: 0.7607876062393188:  99%|█████████▉| 6416/6470 [1:19:45<00:41,  1.32it/s]
4|6417|Loss: 0.8478549122810364:  99%|█████████▉| 6416/6470 [1:19:45<00:41,  1.32it/s]
4|6417|Loss: 0.8478549122810364:  99%|█████████▉| 6417/6470 [1:19:45<00:39,  1.34it/s]
4|6418|Loss: 0.8039243817329407:  99%|█████████▉| 6417/6470 [1:19:46<00:39,  1.34it/s]
4|6418|Loss: 0.8039243817329407:  99%|█████████▉| 6418/6470 [1:19:46<00:37,  1.40it/s]
4|6419|Loss: 0.829517662525177:  99%|█████████▉| 6418/6470 [1:19:47<00:37,  1.40it/s] 
4|6419|Loss: 0.829517662525177:  99%|█████████▉| 6419/6470 [1:19:47<00:38,  1.31it/s]
4|6420|Loss: 0.6567061543464661:  99%|█████████▉| 6419/6470 [1:19:47<00:38,  1.31it/s]
4|6420|Loss: 0.6567061543464661:  99%|█████████▉| 6420/6470 [1:19:48<00:37,  1.33it/s]
4|6421|Loss: 0.7214348912239075:  99%|█████████▉| 6420/6470 [1:19:48<00:37,  1.33it/s]
4|6421|Loss: 0.7214348912239075:  99%|█████████▉| 6421/6470 [1:19:48<00:35,  1.37it/s]
4|6422|Loss: 0.9443973302841187:  99%|█████████▉| 6421/6470 [1:19:49<00:35,  1.37it/s]
4|6422|Loss: 0.9443973302841187:  99%|█████████▉| 6422/6470 [1:19:49<00:34,  1.41it/s]
4|6423|Loss: 0.7398841977119446:  99%|█████████▉| 6422/6470 [1:19:49<00:34,  1.41it/s]
4|6423|Loss: 0.7398841977119446:  99%|█████████▉| 6423/6470 [1:19:50<00:33,  1.42it/s]
4|6424|Loss: 0.8230293393135071:  99%|█████████▉| 6423/6470 [1:19:50<00:33,  1.42it/s]
4|6424|Loss: 0.8230293393135071:  99%|█████████▉| 6424/6470 [1:19:51<00:33,  1.39it/s]
4|6425|Loss: 0.846480667591095:  99%|█████████▉| 6424/6470 [1:19:51<00:33,  1.39it/s] 
4|6425|Loss: 0.846480667591095:  99%|█████████▉| 6425/6470 [1:19:51<00:32,  1.39it/s]
4|6426|Loss: 0.8082270622253418:  99%|█████████▉| 6425/6470 [1:19:52<00:32,  1.39it/s]
4|6426|Loss: 0.8082270622253418:  99%|█████████▉| 6426/6470 [1:19:52<00:31,  1.41it/s]
4|6427|Loss: 0.8909947872161865:  99%|█████████▉| 6426/6470 [1:19:52<00:31,  1.41it/s]
4|6427|Loss: 0.8909947872161865:  99%|█████████▉| 6427/6470 [1:19:53<00:30,  1.40it/s]
4|6428|Loss: 0.8990122675895691:  99%|█████████▉| 6427/6470 [1:19:53<00:30,  1.40it/s]
4|6428|Loss: 0.8990122675895691:  99%|█████████▉| 6428/6470 [1:19:53<00:27,  1.51it/s]
4|6429|Loss: 0.8440934419631958:  99%|█████████▉| 6428/6470 [1:19:54<00:27,  1.51it/s]
4|6429|Loss: 0.8440934419631958:  99%|█████████▉| 6429/6470 [1:19:54<00:27,  1.50it/s]
4|6430|Loss: 0.8556429147720337:  99%|█████████▉| 6429/6470 [1:19:54<00:27,  1.50it/s]
4|6430|Loss: 0.8556429147720337:  99%|█████████▉| 6430/6470 [1:19:55<00:28,  1.41it/s]
4|6431|Loss: 0.8890581130981445:  99%|█████████▉| 6430/6470 [1:19:55<00:28,  1.41it/s]
4|6431|Loss: 0.8890581130981445:  99%|█████████▉| 6431/6470 [1:19:55<00:28,  1.35it/s]
4|6432|Loss: 0.7728426456451416:  99%|█████████▉| 6431/6470 [1:19:56<00:28,  1.35it/s]
4|6432|Loss: 0.7728426456451416:  99%|█████████▉| 6432/6470 [1:19:56<00:28,  1.32it/s]
4|6433|Loss: 0.8218353986740112:  99%|█████████▉| 6432/6470 [1:19:57<00:28,  1.32it/s]
4|6433|Loss: 0.8218353986740112:  99%|█████████▉| 6433/6470 [1:19:57<00:28,  1.32it/s]
4|6434|Loss: 0.8120101690292358:  99%|█████████▉| 6433/6470 [1:19:57<00:28,  1.32it/s]
4|6434|Loss: 0.8120101690292358:  99%|█████████▉| 6434/6470 [1:19:58<00:24,  1.49it/s]
4|6435|Loss: 0.9594606757164001:  99%|█████████▉| 6434/6470 [1:19:58<00:24,  1.49it/s]
4|6435|Loss: 0.9594606757164001:  99%|█████████▉| 6435/6470 [1:19:58<00:23,  1.48it/s]
4|6436|Loss: 0.9177315831184387:  99%|█████████▉| 6435/6470 [1:19:59<00:23,  1.48it/s]
4|6436|Loss: 0.9177315831184387:  99%|█████████▉| 6436/6470 [1:19:59<00:21,  1.57it/s]
4|6437|Loss: 0.8910326361656189:  99%|█████████▉| 6436/6470 [1:19:59<00:21,  1.57it/s]
4|6437|Loss: 0.8910326361656189:  99%|█████████▉| 6437/6470 [1:20:00<00:22,  1.47it/s]
4|6438|Loss: 0.8618184924125671:  99%|█████████▉| 6437/6470 [1:20:00<00:22,  1.47it/s]
4|6438|Loss: 0.8618184924125671: 100%|█████████▉| 6438/6470 [1:20:00<00:23,  1.37it/s]
4|6439|Loss: 0.7945727109909058: 100%|█████████▉| 6438/6470 [1:20:01<00:23,  1.37it/s]
4|6439|Loss: 0.7945727109909058: 100%|█████████▉| 6439/6470 [1:20:01<00:23,  1.32it/s]
4|6440|Loss: 0.7571349740028381: 100%|█████████▉| 6439/6470 [1:20:02<00:23,  1.32it/s]
4|6440|Loss: 0.7571349740028381: 100%|█████████▉| 6440/6470 [1:20:02<00:23,  1.27it/s]
4|6441|Loss: 0.8154001235961914: 100%|█████████▉| 6440/6470 [1:20:02<00:23,  1.27it/s]
4|6441|Loss: 0.8154001235961914: 100%|█████████▉| 6441/6470 [1:20:03<00:22,  1.27it/s]
4|6442|Loss: 0.8047418594360352: 100%|█████████▉| 6441/6470 [1:20:03<00:22,  1.27it/s]
4|6442|Loss: 0.8047418594360352: 100%|█████████▉| 6442/6470 [1:20:04<00:22,  1.27it/s]
4|6443|Loss: 1.0564706325531006: 100%|█████████▉| 6442/6470 [1:20:04<00:22,  1.27it/s]
4|6443|Loss: 1.0564706325531006: 100%|█████████▉| 6443/6470 [1:20:04<00:21,  1.27it/s]
4|6444|Loss: 0.8457797169685364: 100%|█████████▉| 6443/6470 [1:20:05<00:21,  1.27it/s]
4|6444|Loss: 0.8457797169685364: 100%|█████████▉| 6444/6470 [1:20:05<00:20,  1.26it/s]
4|6445|Loss: 0.7274003028869629: 100%|█████████▉| 6444/6470 [1:20:06<00:20,  1.26it/s]
4|6445|Loss: 0.7274003028869629: 100%|█████████▉| 6445/6470 [1:20:06<00:18,  1.35it/s]
4|6446|Loss: 0.8841381072998047: 100%|█████████▉| 6445/6470 [1:20:06<00:18,  1.35it/s]
4|6446|Loss: 0.8841381072998047: 100%|█████████▉| 6446/6470 [1:20:06<00:16,  1.50it/s]
4|6447|Loss: 0.7650144696235657: 100%|█████████▉| 6446/6470 [1:20:07<00:16,  1.50it/s]
4|6447|Loss: 0.7650144696235657: 100%|█████████▉| 6447/6470 [1:20:07<00:14,  1.60it/s]
4|6448|Loss: 0.790671169757843: 100%|█████████▉| 6447/6470 [1:20:07<00:14,  1.60it/s] 
4|6448|Loss: 0.790671169757843: 100%|█████████▉| 6448/6470 [1:20:08<00:13,  1.59it/s]
4|6449|Loss: 0.7745415568351746: 100%|█████████▉| 6448/6470 [1:20:08<00:13,  1.59it/s]
4|6449|Loss: 0.7745415568351746: 100%|█████████▉| 6449/6470 [1:20:08<00:13,  1.54it/s]
4|6450|Loss: 0.8048815727233887: 100%|█████████▉| 6449/6470 [1:20:09<00:13,  1.54it/s]
4|6450|Loss: 0.8048815727233887: 100%|█████████▉| 6450/6470 [1:20:09<00:13,  1.50it/s]
4|6451|Loss: 0.9619048237800598: 100%|█████████▉| 6450/6470 [1:20:09<00:13,  1.50it/s]
4|6451|Loss: 0.9619048237800598: 100%|█████████▉| 6451/6470 [1:20:10<00:12,  1.49it/s]
4|6452|Loss: 0.96256422996521: 100%|█████████▉| 6451/6470 [1:20:10<00:12,  1.49it/s]  
4|6452|Loss: 0.96256422996521: 100%|█████████▉| 6452/6470 [1:20:10<00:11,  1.56it/s]
4|6453|Loss: 0.890123724937439: 100%|█████████▉| 6452/6470 [1:20:11<00:11,  1.56it/s]
4|6453|Loss: 0.890123724937439: 100%|█████████▉| 6453/6470 [1:20:11<00:10,  1.55it/s]
4|6454|Loss: 0.974662721157074: 100%|█████████▉| 6453/6470 [1:20:11<00:10,  1.55it/s]
4|6454|Loss: 0.974662721157074: 100%|█████████▉| 6454/6470 [1:20:11<00:09,  1.60it/s]
4|6455|Loss: 0.8002532124519348: 100%|█████████▉| 6454/6470 [1:20:12<00:09,  1.60it/s]
4|6455|Loss: 0.8002532124519348: 100%|█████████▉| 6455/6470 [1:20:12<00:08,  1.76it/s]
4|6456|Loss: 0.8846045136451721: 100%|█████████▉| 6455/6470 [1:20:12<00:08,  1.76it/s]
4|6456|Loss: 0.8846045136451721: 100%|█████████▉| 6456/6470 [1:20:13<00:08,  1.66it/s]
4|6457|Loss: 0.8254805207252502: 100%|█████████▉| 6456/6470 [1:20:13<00:08,  1.66it/s]
4|6457|Loss: 0.8254805207252502: 100%|█████████▉| 6457/6470 [1:20:13<00:08,  1.47it/s]
4|6458|Loss: 0.7363565564155579: 100%|█████████▉| 6457/6470 [1:20:14<00:08,  1.47it/s]
4|6458|Loss: 0.7363565564155579: 100%|█████████▉| 6458/6470 [1:20:14<00:08,  1.44it/s]
4|6459|Loss: 0.731057345867157: 100%|█████████▉| 6458/6470 [1:20:15<00:08,  1.44it/s] 
4|6459|Loss: 0.731057345867157: 100%|█████████▉| 6459/6470 [1:20:15<00:08,  1.32it/s]
4|6460|Loss: 0.7037480473518372: 100%|█████████▉| 6459/6470 [1:20:15<00:08,  1.32it/s]
4|6460|Loss: 0.7037480473518372: 100%|█████████▉| 6460/6470 [1:20:16<00:07,  1.32it/s]
4|6461|Loss: 0.8631097674369812: 100%|█████████▉| 6460/6470 [1:20:16<00:07,  1.32it/s]
4|6461|Loss: 0.8631097674369812: 100%|█████████▉| 6461/6470 [1:20:16<00:06,  1.35it/s]
4|6462|Loss: 0.8996862769126892: 100%|█████████▉| 6461/6470 [1:20:17<00:06,  1.35it/s]
4|6462|Loss: 0.8996862769126892: 100%|█████████▉| 6462/6470 [1:20:17<00:06,  1.33it/s]
4|6463|Loss: 0.780410647392273: 100%|█████████▉| 6462/6470 [1:20:18<00:06,  1.33it/s] 
4|6463|Loss: 0.780410647392273: 100%|█████████▉| 6463/6470 [1:20:18<00:05,  1.36it/s]
4|6464|Loss: 0.9296358227729797: 100%|█████████▉| 6463/6470 [1:20:18<00:05,  1.36it/s]
4|6464|Loss: 0.9296358227729797: 100%|█████████▉| 6464/6470 [1:20:19<00:04,  1.38it/s]
4|6465|Loss: 0.8624957203865051: 100%|█████████▉| 6464/6470 [1:20:19<00:04,  1.38it/s]
4|6465|Loss: 0.8624957203865051: 100%|█████████▉| 6465/6470 [1:20:19<00:03,  1.30it/s]
4|6466|Loss: 0.7619339823722839: 100%|█████████▉| 6465/6470 [1:20:20<00:03,  1.30it/s]
4|6466|Loss: 0.7619339823722839: 100%|█████████▉| 6466/6470 [1:20:20<00:03,  1.22it/s]
4|6467|Loss: 0.7814288139343262: 100%|█████████▉| 6466/6470 [1:20:21<00:03,  1.22it/s]
4|6467|Loss: 0.7814288139343262: 100%|█████████▉| 6467/6470 [1:20:21<00:02,  1.28it/s]
4|6468|Loss: 0.8050540089607239: 100%|█████████▉| 6467/6470 [1:20:21<00:02,  1.28it/s]
4|6468|Loss: 0.8050540089607239: 100%|█████████▉| 6468/6470 [1:20:22<00:01,  1.37it/s]
4|6469|Loss: 0.7486886382102966: 100%|█████████▉| 6468/6470 [1:20:22<00:01,  1.37it/s]
4|6469|Loss: 0.7486886382102966: 100%|█████████▉| 6469/6470 [1:20:22<00:00,  1.36it/s]
4|6470|Loss: 1.0032275915145874: 100%|█████████▉| 6469/6470 [1:20:23<00:00,  1.36it/s]
4|6470|Loss: 1.0032275915145874: 100%|██████████| 6470/6470 [1:20:23<00:00,  1.28it/s]
4|6470|Loss: 1.0032275915145874: 100%|██████████| 6470/6470 [1:20:23<00:00,  1.34it/s]
Model checkpoint of size 25961 MB saved to output/alpaca-colorful-llama2-finetune/model_3_25880.ckpt
Skipping uploading to HuggingFace Hub (no repo id specified)
wandb: - 0.004 MB of 0.004 MB uploaded
wandb: \ 0.004 MB of 0.020 MB uploaded
wandb: | 0.004 MB of 0.020 MB uploaded
wandb: / 0.020 MB of 0.020 MB uploaded
wandb:                                                                                
wandb: 
wandb: Run history:
wandb: gpu_resources ▄▇▅▇▄▃▃▅▃▅▆▅▅▄▅▄▆▆▄▇▆▃▆▃▂▇▅▇▄▇▂▂▄█▅▄▅▁▄▁
wandb:          loss ██▇▇▆▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁
wandb:            lr ▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁
wandb: 
wandb: Run summary:
wandb: gpu_resources 44811305472
wandb:          loss 1.00323
wandb:            lr 2e-05
wandb: 
wandb: 🚀 View run lucky-envelope-8 at: https://wandb.ai/laurence_r/colorful-llama/runs/qsv24pfw
wandb: ️⚡ View job at https://wandb.ai/laurence_r/colorful-llama/jobs/QXJ0aWZhY3RDb2xsZWN0aW9uOjE0MDkwNjI2OA==/version_details/v0
wandb: Synced 5 W&B file(s), 0 media file(s), 0 artifact file(s) and 0 other file(s)
wandb: Find logs at: ./wandb/run-20240218_171649-qsv24pfw/logs